Akt inhibition blocks the effect of insulin on glucose transport uptake but not lipolysis. (A) 3T3L1 adipocytes were subjected to a glycerol release assay with increasing doses of isoproterenol in the absence or presence of insulin (1 nM) or Akt inhibitor (5 μM) for 2 h. Data are expressed as means ± standard deviations (SD) from two experiments performed in duplicate. Immunoblots for phospho-Akt Thr308 and phospho-AS160/TBC1D4 using phospho-Akt substrate antibody were performed on cell lysates treated with the indicated conditions, including isoproterenol (2 nM), insulin (25 nM), and Akt inhibitor (20 μM). (B) 3T3-L1 adipocytes were used to generate an insulin dose-response curve of glycerol release (left) and fatty acid release (right) at a low concentration of isoproterenol (2 nM) in the presence and absence of Akt inhibitor (10 μM). Data are expressed as means ± SD from two experiments. (C) Simultaneous glycerol release and glucose uptake assays were performed on cells plated on the same day and cultured side by side with the indicated additions at the following concentrations: isoproterenol (2 nM), insulin (25 nM), and Akt inhibitor (20 μM). Data are expressed as means ± standard errors of the means from three experiments for glycerol release and means ± SD from two experiments for glucose uptake. *, P < 0.05 versus basal levels alone; **, P < 0.05 versus isoproterenol alone; ns, not significant. (D) Glycerol release assay using 1 μM forskolin was performed as described for panel A. Data are expressed as means ± SD from two experiments.

Comparison of Akt and PI3K inhibition demonstrates that insulin regulates lipolysis independently of Akt kinase activity. Glycerol release was performed on 3T3-L1 adipocytes using the indicated additions: isoproterenol (Iso; 2 nM), insulin (Ins; 25 nM), Akt inhibitor (Akti; 10 μM), and LY294002 (LY; 50 and 100 μM). Western blotting was performed on whole-cell lysates as shown. Immunoprecipitation was performed using pan-Akt antibody. The immunoprecipitates were subjected to Western blot analysis, and Akt kinase activity was measured using GSK fusion protein. The degree of phosphorylation was assessed using the Licor Odyssey system and graphed as indicated.

Phosphorylation of HSL is partially dependent on Akt and does not correlate with glycerol release. (A) Following a glycerol release assay as illustrated in Fig. 2 and 3, cells were lysed and lysates were immunoblotted for phospho-HSL Ser660, HSL, and tubulin using the Licor Odyssey system. The doublet represents the two isoforms of HSL expressed in adipocytes (34). (B) The intensities of the lower bands in each lane were quantified and normalized to tubulin and then further normalized to basal values for each experiment. Data are expressed as means ± standard errors of the means from three experiments for Akt inhibitor experiments and means ± standard deviations from two experiments for W/M experiments.

Lack of correlation between total cellular PKA activity and glycerol release. Duplicate sets of cells were assessed in parallel, one for glycerol release and another for kinase activity under the indicated conditions: isoproterenol (Iso; 6 nM), insulin (Ins; 25 nM), Akt inhibitor (Akti; 10 μM), and wortmannin (W/M; 100 nM) for 5 min. PKA activity was assessed from lysates incubated with and without Kemptide in the presence and absence of PKI. Data are expressed as means ± standard errors of the means from three experiments. *, P < 0.05 versus isoproterenol alone. ns, no significance versus isoproterenol alone.

Differential effects of Akt inhibition at low and high concentrations of isoproterenol. (A) Floxed Akt2 fibroblasts stably expressing PPARγ were infected with either Adeno-GFP or Adeno-Cre and then differentiated into adipocytes. These cells were serum starved and treated with 1 nM insulin and analyzed by immunoblotting using Licor Odyssey for the excision of Akt2 and the loss of phospho-Akt Ser473 signal (left panel; representative of two experiments). The quantification of immunoblot analysis of phospho-Akt Ser473 of Akt2 ^lox/lox adipocytes infected with Ad-GFP or Ad-Cre is shown (middle panel). A glycerol release assay (right panel) was performed with increasing doses of isoproterenol in the presence and absence of 1 nM insulin for 2 h. Data are expressed as means ± standard deviations (SD) from two experiments performed in duplicate. (B) RNA inteference-mediated reduction in Akt2 does not affect insulin inhibition of glycerol release. 3T3-L1 preadipocytes were infected with an shRNA lentiviral construct that targets Akt2 and sorted for the low and high expression of GFP. Adipocytes were treated with the indicated concentrations of insulin and subjected to the immunoblot analysis of Akt2 and phospho-Akt Thr308, confirming the efficiency of knockdown (left). Highly expressing cells were differentiated into adipocytes, and a glycerol release assay was performed using 2 nM isoproterenol with increasing doses of insulin (upper right). Data are expressed as means ± SD from two experiments performed in duplicate. A glucose uptake assay also was performed on differentiated control, highly, and lowly expressing cells (bottom right).

PI3K dependence of insulin action on lipolysis and glucose transport. (A) 3T3-L1 adipocytes were serum starved for 2 h and pretreated for 30 min with 100 nM wortmannin (W/M) where indicated. Insulin dose-response curves of glycerol release (left) and fatty acid release (right) were generated in the absence and presence of wortmannin using 2 nM isoproterenol stimulation. Data are expressed as means ± standard errors of the means (SEM) from four experiments for glycerol release and means ± standard deviations (SD) from two experiments for fatty acid release. *, P < 0.05 versus the corresponding point on the isoproterenol-alone graph. (B) Parallel glycerol release and glucose uptake assays were performed on cells plated identically with the indicated additions: isoproterenol (Iso; 2 nM), insulin (Ins; 25 nM), and wortmannin (W/M; 200 nM). Data are expressed as means ± SD from two experiments. Immunoblotting was performed using phospho-Akt Thr308 antibody on cell lysates after a glycerol release assay to confirm the efficacy of wortmannin treatment. (C) Glycerol release assay using 1 μM forskolin stimulation was performed as described for panel A. Data are expressed as means ± SEM from three experiments. (D) Glycerol release assay was performed using the indicated additions: isoproterenol (Iso; 2 nM), insulin (Ins; 25 nM), Akt inhibitor (Akti; 10 μM), wortmannin (W/M; 200 nM), and LY294002 (LY; 25 μM). Data are expressed as means ± SD from two experiments performed in duplicate.

Primary adipocyte analysis demonstrates Akt-independent and PI3K-dependent regulation of lipolysis by insulin. Primary rat adipocytes were isolated and a glycerol release assay was performed using the following conditions: isoproterenol (Iso; 10 nM), insulin (Ins; 20 nM), Akt inhibitor (Akti; 10 μM), and wortmannin (W/M; 100 nM). *, P < 0.05 versus basal values.

Insulin regulates perilipin phosphorylation independently of Akt and correlates with glycerol release. (A) Following a glycerol release assay as shown in Fig. 2 and 3, lysates were immunoblotted for perilipin, tubulin, and the PKA phospho-substrate antibody; the dominant ∼60-kDa protein (identified as perilipin) is shown. Immunoblots were quantified using Licor Odyssey and normalized to tubulin. Data are expressed as means ± standard errors of the means from three experiments for Akt inhibitor experiments and means ± standard deviations from two experiments for W/M experiments. (B) Perilipin immunoprecipitates from 3T3-L1 adipocytes treated as shown (top) were blotted with phospho-PKA substrate or perilipin antibody as indicated. Additions were isoproterenol (Iso; 2 nM), insulin (Ins; 25 nM), Akt inhibitor (Akti; 10 μM), and wortmannin (W/M; 200 nM).