TIEG1/KLF10 modulates Runx2 expression and activity in osteoblasts

Abstract

Deletion of TIEG1/KLF10 in mice results in a gender specific osteopenic skeletal phenotype with significant defects in both cortical and trabecular bone, which are observed only in female animals. Calvarial osteoblasts isolated from TIEG1 knockout (KO) mice display reduced expression levels of multiple bone related genes, including Runx2, and exhibit significant delays in their mineralization rates relative to wildtype controls. These data suggest that TIEG1 plays an important role in regulating Runx2 expression in bone and that decreased Runx2 expression in TIEG1 KO mice is in part responsible for the observed osteopenic phenotype. In this manuscript, data is presented demonstrating that over-expression of TIEG1 results in increased expression of Runx2 while repression of TIEG1 results in suppression of Runx2. Transient transfection and chromatin immunoprecipitation assays reveal that TIEG1 directly binds to and activates the Runx2 promoter. The zinc finger containing domain of TIEG1 is necessary for this regulation supporting that activation occurs through direct DNA binding. A role for the ubiquitin/proteasome pathway in fine tuning the regulation of Runx2 expression by TIEG1 is also implicated in this study. Additionally, the regulation of Runx2 expression by cytokines such as TGFβ1 and BMP2 is shown to be inhibited in the absence of TIEG1. Co-immunoprecipitation and co-localization assays indicate that TIEG1 protein associates with Runx2 protein resulting in co-activation of Runx2 transcriptional activity. Lastly, Runx2 adenoviral infection of TIEG1 KO calvarial osteoblasts leads to increased expression of Runx2 and enhancement of their ability to differentiate and mineralize in culture. Taken together, these data implicate an important role for TIEG1 in regulating the expression and activity of Runx2 in osteoblasts and suggest that decreased expression of Runx2 in TIEG1 KO mice contributes to the observed osteopenic bone phenotype.

Figure 1. Runx2 expression levels are decreased in osteoblasts isolated from TIEG1 KO mice relative to WT controls.

Calvarial osteoblasts were isolated from three wild-type (WT) and three TIEG1 knockout (KO) neonatal pups and cultured in vitro. (A) Total RNA was isolated from proliferating cells and real-time PCR analysis was performed to measure Runx2 mRNA levels. The results are depicted as relative expression levels compared to WT cells and represent average Runx2 expression across three distinct cell lines. (B) Representative confocal microscopy image depicting TIEG1 and Runx2 protein levels in WT and TIEG1 KO calvarial osteoblasts. (C) Quantitation of TIEG1 and Runx2 protein levels in WT and TIEG1 KO calvarial osteoblasts. Asterisks denote significance at the p<0.05 level (ANOVA) compared with WT cells.

Figure 2. Suppression of TIEG1 in osteoblasts results in decreased expression of Runx2 and other osteoblast-related genes.

Calvarial osteoblasts isolated from three wild-type neonatal pups were transfected with indicated siRNA constructs for 48 hours. Total RNA was isolated and TIEG1 (A) and Runx2 (B) expression levels were determined using real-time PCR. Data is reported as relative expression levels compared to untransfected cells. (C) The expression levels of osteocalcin (OC), osteopontin (OP), osterix (OSX) and bone sialoprotein (BSP) were also determined in WT calvarial osteoblasts transfected with the TIEG1 specific siRNA relative to cells transfected with the scrambled siRNA. Asterisks denote significance at the p<0.05 level (ANOVA) compared with indicated controls.

Figure 3. Over-expression of TIEG1 in osteoblasts results in increased expression of Runx2 and other osteoblast-related genes.

(A) Inducible U2OS-TIEG1 cells (U2OS Tet-TIEG1) were treated with doxycycline for indicated times. Total RNA was harvested and TIEG1 and Runx2 expression levels were monitored by real-time PCR. (B–C) Calvarial osteoblasts isolated from three TIEG1 knockout (KO) neonatal pups were infected with either control or TIEG1 adenovirus for 24 hours. Total RNA was isolated and TIEG1 and Runx2 expression levels (B), as well as osteocalcin (OC), osteopontin (OP), osterix (OSX) and bone sialoprotein (BSP) (C), were determined using real-time PCR. The results are expressed as relative fold change compared to no doxycycline treatment (A) or to control adenoviral infected cells (B–C). Asterisks denote significance at the p<0.05 level (ANOVA) compared with controls.

Figure 4. TIEG1 mediates Runx2 expression in osteoblasts following TGFβ1 and BMP2 stimulation.

(A) Calvarial osteoblasts isolated from three wild-type (WT) and three TIEG1 knockout (KO) neonatal pups were treated with vehicle, TGFβ1 or BMP2 for 2 hours. Total RNA was isolated and Runx2 expression levels were monitored by real-time PCR. The results are expressed as relative fold change compared to vehicle treated cells and represent average Runx2 expression across three distinct cell lines. (B) WT calvarial osteoblasts were transfected with a scrambled (NT) or TIEG1 specific siRNA for 24 hours and subsequently treated with vehicle, TGFβ1 or BMP2 for 2 hours. Total RNA was isolated and Runx2 expression levels were determined by real-time PCR relative to vehicle controls. Asterisks denote significance at the p<0.05 level (ANOVA) compared to vehicle controls. δ denotes significance at the p<0.05 level (ANOVA) between WT and KO cells (A) or between WT cells transfected with either a scrambled (NT) or TIEG1 specific siRNA (B).

Figure 5. Regulation of Runx2 promoter activity by TIEG1 in osteoblasts.

(A) Putative krüppel-like transcription factor binding sites located within the −600 bp fragment of the Runx2 P1 promoter as identified using the Genomatix software suite. Krüppel-like transcription factor (KLTF), mouse krüppel like factor (MKLF), krüppel like zinc finger (KLZF) and C2H2 zinc finger containing transcription factor (C2H2) binding sites are indicated. (B) U2OS cells were transiently transfected with indicated control, full-length TIEG1 (1–480) or truncated TIEG1 (1–370) expression vectors and the full-length Runx2 promoter (−600) fused to a luciferase reporter. (C) U2OS cells were transiently transfected with indicated control or Runx2 promoter constructs fused to a luciferase reporter along with the full-length (1–480) TIEG1 expression construct. Twenty four hours post-transfection, luciferase activity was monitored and values are reported as relative fold change compared to controls following normalization to total protein levels. Asterisks denote significance at the p<0.05 level (ANOVA) compared with control values. δ denotes significance at the p<0.05 level (ANOVA) between indicated promoter constructs.

Figure 6. Stabilization of TIEG1 protein levels result in enhancement of Runx2 promoter activity.

(A) Western blot analysis depicting wild-type TIEG1 and TIEG1-NxN protein levels following co-transfection with increasing amounts of a SIAH1 expression construct in U2OS cells using a Flag-specific primary antibody. (B) Indicated TIEG1 expression vectors and the full-length Runx2 promoter (−600) reporter construct were transiently transfected into U2OS cells. Twenty four hours post-transfection, luciferase activity was monitored and values are reported as relative fold change compared to controls. Asterisks denote significance at the p<0.05 level (ANOVA) compared with control values. δ denotes significance at the p<0.05 level (ANOVA) between wild-type and NxN TIEG1 expression constructs.

Figure 7. TIEG1 associates with the Runx2 promoter in a DNA binding dependent manner.

(A–D) Transient chromatin immunoprecipitation (ChIP) assays were performed in U2OS cells transfected with indicated Flag-tagged expression vectors and promoter constructs for 24 hours. Chromatin was prepared, immunoprecipitated with a Flag specific antibody and amplified by both real-time PCR (A and C) and semi-quantitative PCR (B and D). Real-time PCR analysis was utilized for quantitation purposes and the data are expressed as the abundance of the Runx2 promoter relative to cells transfected with a control expression construct. All data were normalized using input samples. Asterisks denote significance at the p<0.05 level (ANOVA) compared with controls. δ denotes significance at the p<0.05 level (ANOVA) between indicated promoter constructs. The products obtained by semi-quantitative PCR were separated using agarose gel electrophoresis.

Figure 8. TIEG1 interacts with and co-activates Runx2 transcriptional activity in osteoblasts.

(A) U2OS cells were co-transfected with TIEG1 and Runx2 expression constructs. Cells were lysed and equal amounts of protein were immunoprecipitated with either a Runx2 or TIEG1 specific antibody as well as an IgG control antibody. Immunoprecipitated protein complexes were separated by SDS-PAGE and western blotting was performed using either a TIEG1 or Runx2 specific antibody. Whole cell extracts (WCE) were loaded as positive controls. Arrows indicate TIEG1 and Runx2 protein following immunoprecipitation. (B) Representative confocal microscopy image depicting co-localization of TIEG1 and Runx2 protein in wild-type (WT), but not TIEG1 knockout (KO), calvarial osteoblasts. (C) U2OS cells were transiently transfected with a p6OSE₂ luciferase reporter construct and TIEG1 or Runx2 expression vectors as indicated. Twenty four hours post-transfection, luciferase activity was monitored and values are reported as relative fold change compared to empty vector controls. Asterisks denote significance at the p<0.05 level (ANOVA) compared with control values. δ denotes significance at the p<0.05 level (ANOVA) between cells transfected with Runx2 alone and cells transfected with both Runx2 and TIEG1.

Figure 9. Restoration of Runx2 expression in TIEG1 KO osteoblasts partially rescues their differentiation and mineralization defects.

(A) Calvarial osteoblasts isolated from three wild-type (WT) and three TIEG1 knockout (KO) neonatal pups were infected with either a control or Runx2 adenovirus for 24 hours. Total RNA was isolated and Runx2 expression levels were determined using real-time PCR. The results are expressed as relative fold change compared to control adenoviral infected cells and represent average Runx2 expression across three distinct cell lines. Asterisks denote significance at the p<0.05 level (ANOVA) compared with WT control. (B) Three WT and three TIEG1 KO calvarial osteoblast cell lines were plated in 6-well plates and allowed to proliferate to confluence. Once confluent, cells were cultured in osteoblastic differentiation medium in the presence of a control or Runx2 adenovirus as indicated for 21 days. Following differentiation, cells were stained with alizarin red and a representative wild-type and TIEG1 KO cell line is shown.

Figure 10. Model depicting the mechanisms by which TIEG1 mediates Runx2 expression and activity in osteoblasts.