Characterization of Dkk1 Gene Regulation by the Osteoblast-Specific Transcription Factor Osx

Abstract

Bone formation is a developmental process involving the differentiation of mesenchymal stem cells to osteoblasts. Osterix (Osx) is an osteoblast-specific transcription factor required for bone formation and osteoblast differentiation. Previous observation that Osx inhibits Wnt signaling pathway provides a novel concept of feedback control mechanisms involved in bone formation. Wnt antagonist Dickkopf- 1(Dkk1) plays an important role on skeletal development and bone remodeling. Osx has been shown to activate the Dkk1 promoter; however, the detailed mechanism of Osx regulation on Dkk1 expression is not fully understood. In this study, quantitative real-time RT-PCR results demonstrated that Dkk1 expression was downregulated in Osx-null calvaria at two different points of E15.5 and E18.5 in mice embryos. Overexpression of Osx resulted in upregulation of Dkk1 expression in Tet-off stable C2C12 cell line. Inhibition of Osx expression by siRNA led to downregulation of Dkk1 in osteoblasts. These data suggest that Osx may target Dkk1 directly. To define minimal region of Dkk1 promoter activated by Osx, we made a series of deletion mutants of Dkk1 promoter constructs, and narrowed down the minimal region to the proximal 250bp by transient transfection assay. It was shown that two GC-rich binding sites within this minimal region of Dkk1 promoter were required for the Dkk1 promoter activation by Osx. Importantly, quantitative Chromatin immunoprecipitation (ChIP) assays were performed to show that endogenous Osx associated with native Dkk1 promoter in primary osteoblasts. Taken together, these findings support our hypothesis that Dkk1 is a direct target of Osx.

Introduction

Bone formation takes place through two distinct processes: endochondral ossification involving a cartilage model and intramembranous ossification by which bones form directly from condensations of mesenchymal cells without a cartilage intermediate. Bone formation is a highly regulated process involving the differentiation from mesenchymal progenitor cells into preosteoblast, then into osteoblast lineage, and finally into osteocytes [1; 2]. Osteoblast differentiation is regulated by different transcription factors and signaling proteins including Indian hedgehog, Runx2, Osterix (Osx) and Wnt signaling pathway. Ihh is required for endochondral but not for intramembranous bone formation [3] and is needed for the establishment of the osteogenic portion of the perichondrium/periosteum and for the initial activation of the gene for Runx2. Runx2 is needed for bone formation since no endochondral and no membranous bones are formed in Runx2-null mice [4]. Runx2 is required for the differentiation of mesenchymal cells into preosteoblasts. As a downstream gene of Runx2, Osx is required for the differentiation of preosteoblasts into mature osteoblasts. Osx is specifically expressed in all osteoblasts. In Osx-null embryos, cartilage is formed normally, but the embryos completely lack bone formation [2]. Osx expression pattern in mice indicates that the presence of Osx transcript begins as early as the commitment time for mesenchymal cells to enter osteoblast lineage and its signal becomes stronger as osteoblast differentiation occurs. The C terminal region of Osx contains the DNA-binding domain which can bind to specific GC-rich sequences to control target gene expression, such as osteoblast differentiation markers type 1 collagen, bone sialoprotein, and osteocalcin (OC).

Wnt signaling has been studied for its broad range of activities in cell proliferation, differentiation and cell death during both embryonic development and the adult stage in a variety of tissue types including bone [5]. Wnts are secreted glycoproteins that bind to Frizzled family receptors and low-density lipoprotein receptor-related proteins (LRP) 5/6 coreceptors. In the absence of Wnt, β-catenin forms a complex with the APC, Axin and the kinases glycogen synthase kinase 3, which facilitates phosphorylation and proteosomal degradation of β-catenin. Stimulation of these receptors by Wnts leads to the intracellular molecule β-catenin to accumulate and translocate into the nucleus, where it interacts with TCF/Lef1 transcription factor to activate transcription of target genes. Wnt/β-catenin pathway has been known to play a crucial role in bone formation and bone metabolism [6]. Gain-of-function mutants of Lrp5 cause high bone mass syndrome in patients [7] and in mice [8]. Conditional inactivation of β-catenin in either skeletal progenitor cells or at a later stage of osteoblast development in mouse embryos blocks osteoblast differentiation [9; 10; 11; 12].

Inhibitors of Wnt signaling can bind to frizzled (serum frizzled-related proteins), Wnt (Wnt inhibitory factors) or LRP 5/6 (sclerostin and Dickkopf-1). These agonist proteins prevent Wnt from activating the frizzled LRP 5/6 receptor signaling pathway, leading to a decrease in signaling. The canonical Wnt pathway is regulated by a large number of antagonists, including the Dkk family and secreted frizzled-related proteins. Dickkopf (Dkk) is a Wnt antagonist. It binds to LRP5/6 receptor to form a complex with Kremen1 and 2 and inhibits Wnt signaling by reducing the availability of LRP5/6 [5]. To date, four Dkk proteins have been identified in mammals [13], among which Dkk1 and Dkk2 have been well characterized and found to act as antagonists to the canonical Wnt pathway by binding to LRP5/6 in combination with a second coreceptor designated as Kremen [14; 15]. Transgenic overexpression of Dkk1 under the control of the ColA1 promoter leads to decreased bone mass [16]. In agreement with this observation, deletion of Dkk1 expression in osteoblasts results in an increase in bone formation and mass [17]. It has been discovered that in addition to its essential role in osteoblast differentiation, the osteoblast-specific transcription factor Osx also inhibits osteoblast proliferation and negatively regulates Wnt/β-catenin signaling [18]. Further data have indicated that Osx controls Wnt signaling by two different mechanisms (i) stimulates Wnt antagonist DKK1 expression (ii) disrupts Tcf1 binding to DNA to inhibit the transcriptional activity of β-catenin/Tcf. Osx inhibition of Wnt signaling provides a feedback control mechanism involved in bone formation. Osx has been shown to activate the Dkk1 promoter; however, the detailed mechanism of Osx regulation on Dkk1 expression is not fully understood.

In this study, our results from quantitative real time RT-PCR revealed that Dkk1 expression was downregulated at both E15.5 and E18.5 in the calvaria of Osx-null mouse embryos, suggesting Osx is essential for Dkk1 expression. Dkk1 gene regulation by Osx was further characterized. We provide evidences to demonstrate that Dkk1 is a direct target of Osx.

Materials and Methods

Plasmid constructs and subcloning

The fragments of Dkk1 promoter region were generated by PCR using mouse genomic DNA as a template and subcloned into the XhoI and MluI sites of pGL-3 vector. Primers were obtained from Integrated DNA Technologies (IDT) (Coralville, IA), and the sequences were as follows: 1) Dkk1-Xho-3 5′TGG TGG AGT CTC TGG CTG CCA, 2) Dkk1-Mlu-2k 5′GGC ATC TAT GCA AGG TTC AG, 3) Dkk1-Mlu-1k 5′TAT TAA CCC ACC GCT GGG AAC, 4) Dkk1-Mlu-500 5′TTG ATG AAT GGC TGC TCG CA, 5) Dkk1-Mlu-250 5′TAG TGC TCT AGT GAC CCA CAC. A. Dkk1 point mutants were made with the QuickChange site-directed mutagenesis kit (Stratagene) using Dkk1-250 as a template by the following primers: 1) Dkk1-M1-1 5′GGG ACC ACA GTG CAA TTT ATT TTC GAG GGG AGA GTG TC, 2) Dkk1-M1-2 5′GA CAC TCT CCC CTC GAA AAT AAA TTG CAC TGT GGT CCC, 3) Dkk1-M2-1 5′CGA CAC ACA AAC ACT AAA AAT AAA AGC TCC TCC CAA AGC, 4) Dkk1-M2-2 5′GCT TTG GGA GGA GCT TTT ATT TTT AGT GTT TGT GTG TCG. All constructs including mutants were verified by DNA sequencing.

Cell culture and transient transfection assay

HEK293 cells (ATCC) were cultured in Dulbecco’s modified Eagle’s medium (Invitrogen, CA) with 10% fetal bovine serum and 100 units/ml penicillin plus 100 μg/ml streptomycin at 37 °C. HEK293 cells were plated in 12-well plates, cultured to 60%–80% confluence and transfected with FuGENE 6 (Roche) according to the manufacture’s instruction. Cells were cotransfected with 300ng of Dkk1 promoter reporter, Osx expression plasmid as indicated and 25ng of pSV2-beta-gal. After transfection, cells were incubated for 24h before harvest. The reporter assays were analyzed with BD Monolight system (BD Biosciences). Luciferase activity was normalized by β-galactosidase activity. Every transfection experiment was done at least three times. Values were presented as the mean ±S.D. MC3T3 cells (ATCC) were cultured in Alpha Minimum Essential Medium with ribonucleosides, deoxyribonucleosides, 2 mM L-glutamine and 1 mM sodium pyruvate, but without ascorbic acid (GIBCO), and with 10% FBS and penicillin plus streptomycin. Stable C2C12 mesenchymal cells expressing Osx were generated with pTet-off Advanced Inducible Gene Expression System (Clontech) as previously used [18]. Osx expression was induced in the absence of tetracycline. C2C12 cells were cultured in ATCC described medium with additives G418, Hygromycine, and with or without Doxycycline (Dox), a member of the tetracycline antibiotics group.

siRNA interference

MC3T3 cells were transfected by siRNA against mouse Osx with Lipofectamine 2000. siRNA oligos were purchased from Thermo Scientific Dharmacon, and siGENOME Lamin A/C Control siRNA was used as a non-specific control. Cells were cultured in 6-well plates. One day before transfection, cells were plated in 1ml of growth medium without antibiotics. Cells were 30–50% confluent at the time of transfection. For each sample, siRNA:Lipofectamine 2000 transfection complex was prepared as follows: (1) Dilute 2μl of 50μM siRNA in 50μl of Opti-MEM I Reduced Serum Medium without serum; (2) Mix Lipofectamine 2000 gently, then dilute 3μl in 50μl of Opti-MEM I Medium; (3) Combine the diluted siRNA with the diluted Lipofectamine 2000; (4) Add 100μl of siRNA:Lipofectamine 2000 complex to each well. After 4hr incubation, the growth medium was replaced. Cells were cultured at 37°C in a CO₂ incubator for 24hr before harvest.

Chromatin immunoprecipitation (ChIP) assay

ChIP Assay Kit was from Millipore. ChIP assays were performed according to previously described protocol [19], provided in supplementary materials.

Results

Dkk1 expression is down-regulated in the absence of Osx

Our recent studies have demonstrated that Osx can inhibit Wnt signaling, a possible mechanism for Osx to inhibit osteoblast proliferation [18], and that Osx activates Dkk1 expression. In this study, we further characterized Dkk1 regulation by Osx. We carried out quantitative real time RT-PCR to compare RNA levels of Dkk1 between Osx wild type and knockout mice at two different points of E15.5 and E18.5 in mice embryos. RNA was isolated from calvaria of both E15.5 and E18.5 mouse embryos. As shown in Figure 1A, Osx expression was abolished at E15.5 in Osx-null calvaria. Osteoblast marker gene OC was at undetected level as expected. Interestingly, Wnt signaling antagonist Dkk1 expression was found downregulated by about 16 folds in Osx-null calvaria compared with that in wild type calvaria. As shown in Figure 1B, both Osx and OC expression were abolished at E18.5 in Osx-null calvaria. Dkk1 expression was found downregulated by about 64 folds in Osx-null calvaria compared with that in wild type calvaria. The marked decrease in Dkk1 RNA level in Osx knockout mice suggests that Osx controls Dkk1 gene expression.

Figure 1.

Effect of Osx on Dkk1 expression. (A) Fold change in RNA levels from E15.5 wild-type and Osx-null mice embryos. Calvaria RNAs were isolated from E15.5 Osx wild-type and Osx-null embryos. (B) Fold change in RNA levels from E18.5 wild-type and Osx-null mice embryos. Calvaria RNAs were isolated from E18.5 Osx wild-type and Osx-null embryos. RNA levels for Osx, OC and Dkk1 were analyzed by quantitative real-time RT-PCR. The level of each RNA from Osx-null calvaria was normalized to a value of 1. *: A star indicates statistical significance compared to Osx wild type group.

Overexpression of Osx activates Dkk1 expression

Next we asked whether Osx could positively regulate Dkk1 expression. To address this question, a C2C12 stable cell line was used in which the expression of Osx could be induced by using the tetracycline (Tet) system as previously described [18]. Osx expression was tuned on in the absence of Dox (Figure 2A). Total RNA was isolated from C2C12 stable cell line in the presence or absence of Dox and measured by real time RT-PCR for Dkk1 expression. As shown in Figure 2B, in the absence of Dox when Osx expression was induced, Dkk1 expression was upregulated by about 4.8 folds. This observation indicates that Osx activates Dkk1 expression.

Figure 2.

Osx controls Dkk1 expression. (A) Western immunoblot analysis of the Dox-regulated Osx-expressing C2C12 cells. Osx expression was turned on in the absence of Dox. Beta-actin was used as a loading control. (B) Overexpression of Osx activates Dkk1 gene expression in C2C12 mesenchymal cells. Dkk1 mRNA levels in a stable, Tet-off C2C12 mesenchymal cell line. RNA was obtained from cultures treated with or without Dox. Osx expression was induced in the absence of Dox. Dkk1 mRNA levels were quantitated by real-time RT-PCR. The Dkk1 RNA level obtained from the cells cultured with Dox was normalized to a value of 1. Values are presented as the mean ±S.D. A paired t-test was performed comparing Dox (−) and Dox (+) groups. *: A star indicates statistical significance compared to Dox (+) group. (C) siRNA-directed knockdown of Osx impairs Dkk1 gene expression in MC3T3 osteoblasts. MC3T3 osteoblasts were transfected with siRNA targeting mouse Osx. RNA was isolated 24 hr posttransfection. RNA expression levels as determined by quantitative real-time RT-PCR. The RNA level from the control siRNA group was normalized to a value of 1. Values were presented as the mean ±S.D. A paired t-test was performed comparing si-control group and si-Osx group. *: A star indicates statistical significance compared to control group.

Inhibition of Osx by siRNA led to repression of Dkk1 expression in osteoblasts

To further confirm the effect of Osx on Dkk1 expression, we used siRNA transfection to knockdown Osx expression in MC3T3 osteoblast cells to determine the possible change of Dkk1 expression. Real-time RT-PCR was performed to analyze mRNA expression level. As shown in Figure 2C, when Osx RNA expression was decreased by 80% by siRNA against Osx, Dkk1 RNA expression was repressed by 60%. Therefore, these data suggest that Osx regulates Dkk1 positively in osteoblasts.

Identification of the Osx binding site in the promoter of Dkk1 gene

We have shown Osx can stimulate Dkk1 promoter activity, however it is still not clear which region within Dkk1 promoter is responsible for Osx activation. To address this question, the deletion analysis and transient transfection assay were carried out to narrow down responsible region within 2kb Dkk1 promoter for Osx activation. Dkk1 luciferase reporter constructs driven by different lengths of Dkk1 promoter region were generated. As shown in Figure 3A, Osx was able to activate Dkk1 promoter reporters of Dkk1-2kb, Dkk1-1kb, Dkk1-500bp and Dkk1-250bp in transient transfection assay. Previous study has shown that Osx is a zinc finger-containing transcription factor and belongs to the Sp/XKLF family [2]. This family of transcription factors tends to bind to GC-rich sequence in target gene promoter region to control target gene expression. According to the sequence analysis of the region within 250bp of Dkk1 promoter, two tentative binding sites for Osx were identified as GC rich elements. To test which binding site is responsible for Dkk1 promoter activation by Osx, we generated two point mutants of Dkk1-250 promoter reporter by the QuickChange site-directed mutagenesis kit (Stratagene) using Dkk1-250bp as a template. The two mutants were designated Dkk1-250-M1 and Dkk1-250-M2, in which the guanines (Gs) and the cytosine (Cs) in each element were replaced with thymine (Ts) and adenines (As) respectively. As shown in Figure 3B, Osx activation of the Dkk1-250-M1 and Dkk1-250-M2 promoter mutants was only 55% and 58% respectively, that of the wild-type Dkk1-250bp promoter. Finally, we generated a double mutant in which both GC-rich elements were disrupted (Dkk1-250-M12). As shown in Figure 3B, Dkk1-250-M12 mutant almost abolished Osx activation of the Dkk1-250bp promoter reporter. Thus, these results revealed that two GC-rich sequences in Dkk1-250bp region were responsible for DKk1 promoter activation by Osx.

Figure 3.

Identification of the Osx binding site in the promoter of Dkk1 gene. (A) Deletion analysis of Dkk1 promoter reporter activated by Osx. Dkk1-2k, Dkk1-1k, Dkk1-500bp and Dkk1-250bp Dkk1 promoter reporter plasmids were constructed. Each plasmid was cotransfected with Osx expression plasmid in HEK293 cells. Luciferase activity was normalized by β-galactosidase activity. (B) The two GC-rich elements are responsible for Dkk1 promoter reporter activation by Osx. Dkk1-250-M1, Dkk1-250-M2 point mutants and double mutant Dkk1-250-M12 were constructed. Each plasmid was cotransfected with Osx expression plasmid in HEK293 cells. Luciferase activity was normalized by β-galactosidase activity.

Osx associates with native Dkk1 promoter in vivo

The studies above indicate that Osx can positively regulate Dkk1 expression and activate the Dkk1 promoter in vitro through two GC-rich sequences in Dkk1 promoter. Next we tested whether endogenous Osx can bind to the native Dkk1 promoter in vivo. Therefore, ChIP assay was performed to determine whether Osx could associate with Dkk1 promoter in primary calvarial cells isolated from new born wild-type mice. Two sets of primers were designed. Primer set 1 covered GC-rich element identified within 250bp Dkk1 promoter in Figure 3. Primer set 2 covered a sequence without GC-rich region at the distal 3kb of Dkk1 promoter. Anti-Osx antibody was used as previously described for ChIP analysis [18]. The precipitated chromatin was analyzed by quantitative real-time PCR using Primer set 1 and 2. As shown in Figure 4, Osx was found to associate with the Dkk1 promoter region containing GC rich sequence (Primer set 1) compared with IgG control group. Osx did not associate with the distal Dkk1 promoter region without GC rich sequence (Primer set 2), demonstrating that the observed Osx-DNA association was specific. The data suggest that Osx associates with Dkk1 promoter in vivo.

Figure 4.

Endogenous Osx in primary osteoblasts is associated with the native Dkk1 promoter in vivo. Chromatin Immunoprecipitation (ChIP) assays were conducted using primary calvarial osteoblasts isolated from new born wild-type mice. Anti-Osx antibody (a-Osx) was used for ChIP analysis, and IgG was used as a negative control. The precipitated chromatin was analyzed by quantitative real-time PCR. As described in the Methods, primer Set 1 corresponds to a segment covering two GC-rich elements within 250bp the Dkk1 promoter. As a negative control, Primer Set 2 covers a distal 3kb region of the Dkk1 promoter, which does not contain GC-rich sequences. A paired t-test was performed comparing IgG and a-Osx group. *: A star indicates statistical significance compared to IgG control group.

Discussion

Osx is an osteoblast-specific transcription factor that controls the expression of essential genes needed for appropriate osteoblast differentiation and function. Osx inhibits Wnt pathway. In this study, we further characterized Osx regulation on Wnt antagonist Dkk1 expression. The findings presented here support our hypothesis that Dkk1 is a direct target of Osx.

First, we identified Dkk1 as an Osx target gene. This is supported by the Osx-directed gene expression studies in several in vivo and in vitro model systems. For example, Osx-null calvaria displayed defective in vivo expression of Dkk1 at E15.5 and E18.5 compared to wild-type calvaria in mice (Figure 1). A Tet-off inducible C2C12 stable cell system revealed that ectopic expression of Osx resulted in an increase in Dkk1 transcript level (Figure 2B) and MC3T3 osteoblast cultures also had markedly impaired Dkk1 gene expression when Osx expression was knocked-down using siRNA targeting strategies (Figure 2C). Moreover, a direct regulation of Dkk1 gene transcription by Osx was evident in the ability of recombinant Osx to activate Dkk1 promoter-reporter constructs (Figure 3A), thus indicating that the RNA expression studies were likely due to the effects of Osx expression on Dkk1 gene transcription.

In terms of mechanism, Osx is an SP/KLF family member that presumably functions by binding directly to DNA promoter elements via an SP1-like DNA-binding domain consisting of three C2H2-type zinc fingers located within its C-terminus [2]. They likely are GC-rich in nature based on the similarities of DNA-binding domains and elements among this family of transcription factors. Indeed, two GC-rich regions exist in the most proximal regions of the murine Dkk1 promoter. Importantly, our studies define these two elements as critical for mediating the transcriptional activation of the Dkk1 promoter by Osx. Mutation of each element (Dkk1-250-M1 or Dkk1-250-M2) reduced Osx-directed activation of the Dkk1 promoter by approximately 50% and mutation of both elements Dkk1-250-M12 nearly abolished the response (Figure 3B). Of course, limitations of this particular approach include the heterologous expression and somewhat artificial nature of the plasmid DNA constructs used as a measure of a biologically relevant transcriptional event. Thus, the chromatin immunoprecipitation approaches (Figure 4) strongly support the reporter gene expression data. These studies clearly demonstrated that endogenous Osx was associated with the GC-rich proximal region of the native Dkk1 promoter in primary osteoblastic cells. These studies help establish Osx mechanisms involving direct binding of Osx to sequence specific, GC-rich promoter elements to activate the expression of genes in osteoblastic cells.

Wnt signaling is known to have the major impact at different stages of bone formation and bone metabolism [5; 6]. Wnt signaling-mediated gene expression can promote osteoblast proliferation and differentiation. Some studies investigated the role of Wnt/β-catenin signaling in nonunion and osteoporosis, suggesting Wnt signaling could possibly have potential to become a target of pharmacological intervention to increase bone formation [20; 21]. Dkk1 is one of the Wnt antagonists. Our observation that Dkk1 expression is absent in Osx-null calvaria of at both E15.5 and E18.5 embryos parallels the previously reported finding that Dkk1 expression was absent in tibia of E18.5 embryos in which β-catenin was inactivated using an Osx-Cre transgene [12]. This suggests that both Osx and β-catenin are required for Dkk1 expression in osteoblasts. Thus, the Dkk1 gene might be a common target of these transcription factors in these cells. Due to the important role of Wnt signaling during bone formation, it will be interesting to investigate regulations of Wnt signaling. Osx is required for osteoblast differentiation and bone formation [2]. Previous study has demonstrated that Osx inhibits Wnt signaling activity during bone formation, a possible mechanism for the inhibition by Osx of osteoblast proliferation [18]. The findings in this study that Osx controls Dkk1 expression should further reinforce the feedback control effect that Dkk1 exerts on Wnt/β-catenin signaling.

In summary, we performed in vitro and in vivo experiments to demonstrate that the Wnt antagonist Dkk1 is a direct target of osteoblast-specific transcription factor Osx in osteoblasts.

Research highlights.

• Dkk1 expression is down-regulated in the absence of Osx

• Overexpression of Osx activates Dkk1 expression

• Inhibition of Osx by siRNA results in repression of Dkk1 expression

• GC-rich sequences of Dkk1 promoter are responsible for Osx activation

• Osx associates with native Dkk1 promoter in vivo

Abbreviations

Osx

Osterix

Dkk1

Dickkopf 1

OC

osteocalcin

E15.5

embryonic day 15.5

E18.5

embryonic day 18.5

ChIP

Chromatin immunoprecipitation

Dox

Doxycycline

LRP

low-density lipoprotein receptor-related protein