Functional characterisation of the regulation of CAAT enhancer binding protein alpha by GSK-3 phosphorylation of Threonines 222/226 - PubMed (original) (raw)
Functional characterisation of the regulation of CAAT enhancer binding protein alpha by GSK-3 phosphorylation of Threonines 222/226
H-K Liu et al. BMC Mol Biol. 2006.
Abstract
Background: Glycogen Synthase Kinase-3 (GSK3) activity is repressed following insulin treatment of cells. Pharmacological inhibition of GSK3 mimics the effect of insulin on Phosphoenolpyruvate Carboxykinase (PEPCK), Glucose-6 Phosphatase (G6Pase) and IGF binding protein-1 (IGFBP1) gene expression. CAAT/enhancer binding protein alpha (C/EBPalpha) regulates these gene promoters in liver and is phosphorylated on two residues (T222/T226) by GSK3, although the functional outcome of the phosphorylation has not been established. We aimed to establish whether CEBPalpha is a link between GSK3 and these gene promoters.
Results: C/EBPalpha represses the IGFBP1 thymine-rich insulin response element (TIRE), but mutation of T222 or T226 of C/EBPalpha to non-phosphorylatable alanines has no effect on C/EBPalpha activity in liver cells (towards the TIRE or a consensus C/EBP binding sequence). Phosphorylation of T222/T226 is decreased by GSK3 inhibition, suggesting GSK3 does phosphorylate T222/226 in intact cells. However, phosphorylation was not altered by treatment of liver cells with insulin. Meanwhile C/EBPalpha activity in 3T3 L1 preadipocytes was enhanced by mutation of T222/T226 and/or S230 to alanine residues. Finally, we demonstrate that C/EBPalpha is a very poor substrate for GSK3 in vitro and in cells.
Conclusion: The work demonstrates an important role for this domain in the regulation of C/EBPalpha activity in adipocytes but not hepatocytes, however GSK3 phosphorylation of these residues does not mediate regulation of this C/EBP activity. In short, we find no evidence that C/EBPalpha activity is regulated by direct phosphorylation by GSK3.
Figures
Figure 1
C/EBPα expression suppresses the BPl-TIRE-Luc reporter activity in H4IIE cells, independently of the phosphorylation status of T222/T226. (A) H4IIE cells were transfected with BPl-TIRE-Luc (10 μg) or DM5-Luc (10 μg) along with either pcDNA6 vector (1 μg), Flag-C/EBPα (1 μg) or Flag-C/EBPα-AAA (1 μg). Subsequently, cells were incubated with or without 10 nM insulin for 20 h prior to lysis and luciferase assay. Results are presented as relative luciferase activity to the control (serum free) and are average ± SEM (n = 3) from two independent experiments. (B) H4IIE cells were infected with adenovirus expressing either β-Galactosidase (AD-βgal) or active GSK3β (Ad-GSK3 S9A) for 20 h prior to transfection as described in (A). Results are presented as relative luciferase activity to the control (serum free) and are average ± SEM (n = 6) from two independent experiments. *** p < 0.001 compared to control (ad-b-gal).
Figure 2
C/EBPα phosphorylation in intact cells. AD293 cells were transiently transfected with Flag-C/EBPα (10 μg) prior to incubation with CHIR99021 as indicated (A). Cell lysates were then subjected to Western Blotting with the antibodies as labelled. A representative immunoblot (of 2 two separate experiments) is given. (B) AD293 cells were transfected with Flag-C/EBPα (10 μg) plus GSK3α-V5 (10 μg) or pcDNA6 (10 μg) prior to 24 h incubation with CHIR99021 (2 μM), lysis and Western Blot analysis. A representative immunoblot (of three separate experiments) is given. (C) H4IIE cells were infected with adenovirus expressing Flag-C/EBPα. After 16 h the infected cells were serum starved for 3 h prior to incubation with or without insulin (10 nM) or CHIR99021 (2 μM) for the times indicated. Cell lysates were immunoblotted with the antibodies indicated and a representative experiment is provided.
Figure 3
Quantification of three experiments performed as in Fig 2C. Results are the average ± SEM.
Figure 4
C/EBPα phosphorylation in vitro. (A) Recombinant Flag-C/EBPα (6 pmol) was generated by scission of the GST-C/EBPα fusion protein and incubated with either recombinant GSK3β (2 U/ml) or p42 MAPK (10 U/ml) and [γ-32P]-ATP for 1, 10, 30, and 60 min. The reaction was stopped and separated by SDS-PAGE. (B) The stoichiometry of Flag-C/EBPα phosphorylation in (A) by either MAPK or GSK3β is quantified and presented. (C) GST-C/EBPα (7.5 pmol) was incubated with recombinant GSK3β (2 U/ml) and ATP for 1, 10, 30, and 60 min. The reaction was stopped and separated by SDS-PAGE and subjected to Western Blot analysis for phospho-T222/226. The equivalent stoichiometry of phosphorylation (mol/mol) calculated in A is indicated above each band.
Figure 5
S230 phosphorylation does not prime for T226 phosphorylation by GSK3. (A) AD293 cells were transfected with 10 μg of Flag-C/EBPα, Flag-C/EBPα-AA, Flag-C/EBPα-AAA or Flag-C/EBPα-S230A prior to 24 h incubation with 2 μM CHIR99021 (CT), lysis and Western Blot analysis. A representative immunoblot of two such experiments is given. (B) AD293 cells were transfected with 10 μg of Flag-C/EBPα or Flag-C/EBPα-S230A, then serum starved overnight prior to a 2 h incubation with CHIR99021 (2 μM) or 10%(v/v) FBS. Cell lysates were then subjected to Western Blotting with the antibodies as labelled. A representative immunoblot of two such experiments is given.
Figure 6
MAPK does not influence CEBPα phosphorylation by GSK3. (A) Recombinant Flag-C/EBPα (6 pmol) was incubated for 60 min with or without p42 MAPK (10 U/ml) and/or GSK3β (2 U/ml) and Mg ATP, as indicated in figure. The reaction was stopped and separated by SDS-PAGE and subjected to Western Blot. (B) AD293 cells were transfected with 10 μg of Flag-C/EBPα, Flag-C/EBPα-AA, or Flag-C/EBPα-S230A prior to 24 h incubation with 10 μM PD98059, SB203580 or DMSO, lysis and Western Blot analysis. A representative immunoblot of two such experiments is given.
Figure 7
C/EBPα activity in H4IIE cells. (A) H4IIE cells were transfected with the CBE-Luc (10 μg) along with 1 μg of pcDNA6, Flag-C/EBPα, Flag-C/EBPα-AA, Flag-C/EBPα-AAA or Flag-C/EBPα-S230A prior to a 20 h incubation with insulin (10 nM) or CHIR99021 (CT, 2 μM). Results are presented as relative luciferase activity to control (serum free) and are average ± SEM (n = 6) from two independent experiments. * p = 0.0683, ** p = 0.0523. (B) H4IIE cells were co-transfected with 10 μg of either CBE-Luc reporter or pGL3-Luc reporter plus 1 μg of pcDNA6 or Flag-C/EBPα prior to 20 h incubation with dexamethasone (500 nM), insulin (10 nM), or CHIR99021 (CT, 2 μM). Results are presented as relative luciferase activity to the control (serum free) and are average ± SEM (n = 6) from two independent experiments. ***p < 0.001
Figure 8
Regulation of endogenous IGFBP1 gene expression in H4IIE cells following C/EBPα overexpression. H4IIE cells were infected with adenovirus expressing either β-Galactosidase (AD-βgal) or Flag-C/EBPα (Ad-C/EBPα). Infected cells were serum starved overnight prior to a 3 h incubation with insulin (10 nM), dexamethasone (500 nM), CHIR99021 (CT, 2 μM), or a combination of each as indicated. RNA was extracted and IGFBP1 expression was measured by RPA. Expression was calculated relative to cyclophilin mRNA levels. Data are presented as the average ± SEM of three experiments, with a representative autoradiograph shown in the upper panels.
Figure 9
Regulation of endogenous G6Pase gene expression in H4IIE cells following C/EBPα overexpression. RNA produced as in Fig. 8 was analysed for G6Pase expression by RPA. Expression was calculated relative to cyclophilin mRNA levels. Data are presented as the average ± SEM of three experiments, with a representative autoradiograph shown in the upper panels.
Figure 10
Regulation of the isolated TIRE in H4IIE cells following C/EBPα overexpression. H4IIE cells, infected as in Fig. 8, were also transfected with 10 μg of BPl-TIRE-Luc or DM5-Luc and luciferase activity measured 20 h later. Results are presented as relative luciferase activity to the control (AD-βgal, serum free) and are average ± SEM of three experiments. ** p < 0.01 AD-βgal versus Ad-C/EBPα.
Figure 11
Mutation of T222/226 and S230 alters C/EBPα activity in preadipocytes. A) 3T3L1 preadipocytes were transfected with 2 μg of C/EBP-Luc reporter along with 200 ng of pcDNA6, Flag-C/EBPα, AA, AAA, or S230A as described in the material and methods. Cells were incubated with insulin (l00 nM) or CHIR99021 (CT, l0 μM) for 20 hr prior to harvest for luciferase assay. Results are relative luciferase activity and are average ± SEM (n = 6) from two independent experiments. * p < 0.05, ** p < 0.01, *** p < 0.005 compared with control. B) 3T3L1 preadipocytes were transfected with 200 ng of pcDNA6, Flag-C/EBPα wild type, or Flag-C/EBPα S230A, ± 2 μM CHIR99021 (CT). Cells were lysed, the recombinant protein immunoprecipitated using anti-FLAG, and visualised using anti-FLAG or anti-phospho222/226. A representative experiment is shown in the upper panel, while a quantification of three such experiments is given in the lower panel.
References
- Embi N, Rylatt DB, Cohen P. Glycogen Synthase Kinase-3 from Rabbit Skeletal Muscle. Eur J Biochem. 1980;107:519–527. - PubMed
- Woodgett JR. cDNA cloning and properties of GSK3. Meth Enzymol. 1991;200:564–577. - PubMed
Publication types
MeSH terms
Substances
LinkOut - more resources
Full Text Sources
Other Literature Sources
Molecular Biology Databases
Research Materials