AICAR positively regulate glycogen synthase activity and LDL receptor expression through Raf-1/MEK/p42/44MAPK/p90RSK/GSK-3 signaling cascade

Abstract

5-Aminoimidazole-4-carboxamide-1-beta-D-ribofuranoside (AICAR) is a commonly used pharmacological agent to study physiological effects which are similar to those of exercise. However, signal transduction pathways by which AICAR elicits downstream effects in liver are poorly understood. We report here that AICAR not only activated AMPK but also phosphorylated/deactivated glycogen synthase kinase-3 alpha/beta (GSK-3alpha/beta) and dephophorylated/activated glycogen synthase (GS) in a time-dependent manner in human hepatoma HepG2 cells. The signal connection between AICAR and GSK-3 is indirect and involves activation of Raf-1/MEK/p42/44(MAPK)/p90(RSK) signaling cascade as pharmacologic inhibition of MEK significantly reduced phosphorylation/deactivation of GSK-3 and consequent dephosphorylation/activation of GS. Moreover, silencing the expression of p90(RSK), a substrate of p42/44(MAPK), attenuated AICAR-dependent GSK-3 phosphorylation, implicating this kinase as a key mediator of AICAR signaling to GSK-3. Furthermore, consistent with the involvement of Raf-1 kinase cascade, AICAR-induced low-density lipoprotein (LDL) receptor expression in a p42/44(MAPK)-dependent manner. Finally, AICAR requires AMPK-alpha2-dependent and -independent pathways to activate Raf-1 kinase cascade as suppression of AMPKalpha2 activity, and not of AMPKalpha1, partially blocked AICAR-dependent p42/44(MAPK) activation and GSK-3 phosphorylation/deactivation. Collectively, these results highlight Raf-1 signaling cascade as the critical mediator of AICAR action on glucose and lipid metabolism in HepG2 cells.

FIG. 1

Treatment with AICAR activated AMPK and phosphorylated ACC and GSK-3 with simultaneous dephosphorylation of GS in a time- and dose-dependent manner in HepG2 cells. A: Kinetics of changes in phosphorylation of AMPK, ACC, GSK-3 and GS by AICAR. HepG2 cells were treated with 2 mM AICAR for the indicated time periods, and cell lysates were immunoblotted for phospho-AMPK, total AMPK, phospho-ACC, phospho-GSK-3, total GSK-3, and phospho-GS. The phospho data are presented as relative to the untreated cells. Values shown are means ± SE of three separate experiments. B: AICAR induced phosphorylation of GSK-3 in a dose-dependent manner. HepG2 cells were treated with indicated concentrations of AICAR for 3h to probe phosphorylation status of GSK-3. The phospho data are presented as relative to the untreated cells. Values shown are means ± SE of three separate experiments. C: AICAR increases glycogen content in HepG2 cells. Cells were treated with AICAR for 15 min prior to the addition of [¹⁴C-(U)]glucose. Following 1h incubation in the presence of AICAR, total glycogen contents were determined by scintillation counting. *, p<0.05 relative to the value in the absence of AICAR, n=5. Immunoblots shown are representative of three experiments which produced similar results. Values obtained from the control cells grown in the absence of AICAR were arbitrarily set at 1.

FIG. 2

AICAR-induced GSK-3 phosphorylation is independent of Akt pathway. Cells incubated overnight in media with 0.5% serum were treated with 2 mM AICAR for 3h, 1 ng/ml insulin for 10 min, or both. Cell lysates were immunoblotted for phospho-ACC, phospho-Akt, total Akt, phospho-GSK-3 and total GSK-3. Immunoblot shown are representative of three separate experiments which produced similar results. The phospho data is representative of three separate experiments and values shown are mean±SE.

FIG. 3

Activation of Raf-1/MEK/p42/44^MAPK/p90^RSK cascade by AICAR in HepG2 cells. A: Kinetics of activation of MEK, p42/44^MAPK, p90^RSK, and p38^MAPK by AICAR. Cells treated with 2 mM AICAR for different time periods were immunoblotted for the indicated antibodies. Immunoblots shown are representative of three experiments, which produced similar results. B: Kinetics of Raf-1 phosphorylation by AICAR. HepG2 cells treated with AICAR were probed for two different phospho-Raf-1 antibodies and total Raf-1. C: Kinetics of increase in Raf-1 activity by AICAR treatment. HepG2 cells starved for 16 h were treated with 2 mM AICAR as indicated. Cells were lysed, and the supernatant of the lysates were used for Raf-1 immunoprecipitation and subsequent kinase assay. Phosphorylation of the Raf-1 substrate was analyzed by phosphoimaging and results are expressed as fold stimulation by AICAR as compared with untreated cells. *, <0.05 relative to the value in the absence of AICAR, n=3.

FIG. 4

AICAR-dependent p42/44^MAPK activation is mediated through Raf-1/MEK cascade. A: Correlation of the loss of phosphorylation of p42/44^MAPK, p90^RSK, P-GSK-3 with an increase in phosphorylation of GS upon removal of AICAR. HepG2 cells were incubated for 3h with 2 mM AICAR and then switched to a medium lacking AICAR for the indicated time periods. Phosphorylation levels of the indicated proteins were detected by immunoblotting. Results shown are representative of three separate experiments. B: Effect of inhibition of MEK on AICAR-induced p42/44^MAPK activation in HepG2 cells. Cells pretreated for 30 min with the indicated concentrations of U0126 were either untreated or treated with 2 mM AICAR for 3h. Cell lysates were subjected to immunoblotting with the indicated antibodies. Immunoblot shown are representative of three experiments which produced similar results. C: AICAR-induced GSK-3 phosphorylation requires p90^RSK. siRNA directed against p90^RSK significantly reduced endogenous levels of p90^RSK and also significantly reduced AICAR-induced GSK-3 phosphorylation. HepG2 cells cultured in 6 well dishes were transfected with 50 or 100 nM of either control siRNA or p90^RSK siRNA. After two days, transfected cells starved for 16h were subsequently treated for 3h with 2 mM AICAR and processed for immunoblotting with the indicated antibodies. *, p<0.05 relative to the value in the absence of AICAR and siRNA, n=3; **<0.01 relative to the value in the absence of AICAR and siRNA, n=3.

FIG. 5

AICAR-induced LDL receptor expression in a p42/44^MAPK-dependent manner in HepG2 cells. A: 2×10⁵ cells were plated on day 0, and were refed with fresh medium on day 2. On day 3, cells were treated with 2 mM AICAR for the indicated time periods. Total RNA was subjected to RT-PCR to determine amounts of LDL receptor, HMG-CoA reductase, SS, and β-actin mRNAs. A representative autoradiogram is shown showing changes in expression of the above gene. B: AICAR-induced LDL receptor is mediated through MEK-1/2 pathway. Cells pretreated for 30 min with the indicated concentrations of PD98059 were treated with 2 mM AICAR for 6h. Total RNA from each treatment was subjected to RT-PCR analysis. Results shown are representative of three independent experiments. Autoradiograms were quantified densitometrically and RNA levels were normalized to β-actin levels. Densitometric analysis of each autoradiogram is expressed as means ± SE of three experiments performed in duplicate. Values obtained from control cells grown in the absence of AICAR or inhibitor were arbitrarily set at 1.

FIG. 6

Partial requirement of AMPKα2, and not AMPKαl, for AICAR-induced p42/44^MAPK activation in HepG2 cells. A: siRNA directed against AMPKαl significantly reduced endogenous levels of this isoform, without significantly affecting p42/44^MAPK activation following treatment with 2 mM AICAR. HepG2 cells cultured in 6-well dishes were transfected with 20 or 40 nM AMPKαl siRNA. After two days, transfected cells starved for 16 h were treated for 1 h with 2 mM AICAR and processed for immunoblotting with the indicated antibodies. B: Effect of adenovirus-mediated overexpression of AMPKα2^DN on AICAR-induced phosphorylation of ACC and p42/44MAPK. HepG2 cells transduced with indicated doses of either Ad-Null or Ad-AMPKα2^DN virus for two days were starved for 16 h and then treated for 3h with 2 mM AICAR to examine phosphorylation levels of ACC and p42/44^MAPK. Anti-c-myc was used to demonstrate overexpression of c-myc-tagged-AMPKα2^DN. The results shown are representative of three separate experiments. Signals were quantified with a phosphoimager and the results are expressed as fold stimulation by AICAR as compared to untreated cells (set at 1). Quantitative data are means ± SE from three separate experiments. *, <0.05 relative to the value in the absence of AICAR and siRNA, n=4; **, <0.01 relative to the value in the absence of AICAR and siRNA, n=4.