Skip to main content
. 2007 Sep 5;2(9):e837. doi: 10.1371/journal.pone.0000837

Figure 4. Inactivation through phosphorylation of Runx2 by GSK-3β.

Figure 4

(A) Subcellular nuclear (N) and cytoplasmic (C) localizations of Runx2 by immunoblot analysis (top) and Runx2 mRNA level determined by real-time RT-PCR (bottom) in Gsk-3β+/+ and Gsk-3β+/– calvarial osteoblasts overexpressing CA-GSK-3β or treated with LiCl, and cultured for 3 days. The mRNA levels are mean (bars)±SEM (error bars) of the relative amount compared to the control culture of 6 wells per group. (B) EMSA for specific binding (arrowheads) of a labeled OSE2 oligonucleotide probe with the nuclear extracts (N.E.) from Gsk-3β+/+ or Gsk-3β+/– osteoblasts overexpressing Runx2. Cold competition (Comp.) was performed with 50-fold excess of unlabeled wild-type OSE2 probe (wt) and the mutated probe lacking the Runx2 binding sequence (mut). For controls, incubations without the probe (the 1st lane) and using nuclear extracts from osteoblasts without Runx2 transfection (the last lane) were performed. (C, D) Co-immunoprecipitation (co-IP) analysis of GSK-3b and Runx2. (C) Whole cell lysate (CL) and co-IP precipitant by anti-FLAG antibody-immobilized beads were immunoblotted with either anti-HA tag or anti FLAG tag antibodies. Filled arrowhead indicates non-specific band, and blank arrowhead indicates specific band. (D) Whole cell lysate (CL) and co-IP precipitant by anti-HA tag antibody or IgG (as a negative control) were immunoblotted with either anti-FLAG tag or anti-HA tag antibodies. (E) Luciferase reporter analysis of the effects of Runx2 mutations at the five consensus sites for the phosphorylation by GSK-3β on the Runx2 transcriptional activity. Mutations were created by three to four amino acid replacements as follows; S92A-S96A-S100A [M(96)3], S369A-S373A-S377A [M(373)3], S389A-T393A-S397A [M(393)3], T394A-S398A-T402A [M(398)3], and T476A-T480A-S484A-S488A [M(480)4]. HuH-7 cells were transfected with 1,050 OC-Luc alone or in combination with the plasmids expressing wild-type Runx2 (WT) or the mutants above, then cultured for 2 days. Data are mean (bars)±SEM (error bars) of the relative activity compared to control of 6 wells per group. *P<0.01 vs. WT-Runx2. (F) In vitro kinase assay. WT-Runx2 and M(373)3-Runx2 proteins were extracted by immunoprecipitation of the overexpresssing HeLa cells, and were incubated with recombinant GSK-3β. Reaction products were analyzed by immunoblotting using an antibody to phosphoserine. (G) EMSA for specific binding (arrowheads) of a labeled OSE2 probe with the nuclear extracts (N.E.) from HeLa cells transfected with wild-type Runx2 (WT) and M(373)3 Runx2 (M). Cold competition (Comp.) was performed as above. (H) Luciferase reporter analysis of the effects of GSK-3β signaling on the Runx2 transcriptional activity induced by WT-Runx2 and M(373)3-Runx2. HuH-7 cells were transfected with 1,050 OC-Luc alone or in combination with the plasmid expressing WT-Runx2 or M(373)3-Runx2 in the presence or absence of CA-GSK-3β overexpression or LiCl, then cultured for 2 days. Data are mean (bars)±SEM (error bars) of the relative activity compared to control of 6 wells per group. *P<0.01, significant effect of CA-GSK-3β overexpression or LiCl.