Development of a new model system to dissect isoform specific Akt signalling in adipocytes

Abstract

Protein kinase B (Akt) kinases are critical signal transducers mediating insulin action. Genetic studies revealed that Akt1 and Akt2 signalling differentially contribute to sustain lipid and glucose homoeostasis; however Akt isoform-specific effectors remain elusive due to the lack of a suitable model system to mechanistically interrogate Akt isoform-specific signalling. To overcome those technical limitations we developed a novel model system that provides acute and specific control of signalling by Akt isoforms. We generated mutants of Akt1 and Akt2 resistant to the allosteric Akt inhibitor MK-2206. We then developed adipocyte cell lines, in which endogenous Akt1 or Akt2 has been replaced by their corresponding drug-resistant Akt mutant. Treatment of those cells with MK-2206 allowed for acute and specific control of either Akt1 or Akt2 function. Our data showed that Akt1(W80A) and Akt2(W80A) mutants are resistant to MK-2206, dynamically regulated by insulin and able to signal to Akt downstream effectors. Analyses of insulin action in this cellular system showed that Akt1 and Akt2 are both able to mediate insulin regulation of the transcription factor forkhead box O1 (FoxO1) and the glucose transporter 4 (GLUT4), revealing a redundant role for these Akt kinases in the control of glucose transport into fat cells. In contrast, Akt1 signalling is uniquely required for adipogenesis, by controlling the mitotic clonal expansion (MCE) of pre-adipocytes that precedes white adipose cell differentiation. Our data provide new insights into the role of Akt kinases in glucose transport and adipogenesis and support our model system as a valuable tool for the biochemical characterization of signalling by specific Akt isoforms.

Keywords: Akt isoform signalling; MK-2206; adipocyte; adipogenesis; glucose transporter 4 (GLUT4); insulin action.

Figure 1. Development of adipocytes stably expressing Akt mutants resistant to MK-2206

(A) Schematic representation of the protocol used to develop 3T3-L1 adipocytes stably expressing Flag-tagged Akt1^WT, Akt1^W80A, Akt2^WT, Akt2^W80A. (B) Western blot analyses of protein extracts from adipocytes stably expressing Flag-tagged WT or W80A Akt1 and Akt2. (C) Immunoblot analyses of protein extracts from adipocytes expressing WT or W80A Akt1 or Akt2. Cells were treated with 1 nM insulin for 15 min. Some cells were treated with 1 μM MK-2206 for 1 h prior to insulin addition as noted. IRS1 was immunoprecipitated from cell lysates using anti-IRS1 specific antibody. (D) Immunofluorescence analyses of Flag-tagged Akt2^WT, Akt1^W80A and Akt2^W80A using an anti-Flag antibody. Epifluorescence and TIRF microscopy were used to detect total and plasma membrane-targeted Akt constructs respectively. Adipocytes were starved for 2 h followed by vehicle (Basal) or 10 nM insulin stimulation for 15 min (Insulin). Some cells were pre-treated with 10 μM MK-2206 for 1 h prior to insulin addition (Insulin, MK-2206). (E) Quantification of insulin-induced Flag-tagged Akt2^WT, Akt1^W80A and Akt2^W80A redistribution to the plasma membrane using TIRF microscopy. Indirect immunofluorescence of the Flag epitope in basal or 10 nM insulin-stimulated adipocytes was measured in the epifluorescence mode and in the TIRF mode. The anti-Flag TIRF is normalized to the anti-Flag fluorescence in the epifluorescence mode. Data are normalized to that of insulin-treated Akt2^W80A expressing cells. Each data point represents the mean ± S.E.M., n=28–70 cells; *P<0.01 (paired t test).

Figure 2. Akt1^W80A and Akt2^W80A mediate insulin signal to Akt downstream effectors

(A) Immunoblot analyses of protein extracts from adipocytes expressing WT or W80A Akt mutants. Adipocytes were pre-treated with MK-2206 for 1 h as noted followed by stimulation with insulin for 15 min. (B) Densitometric analyses of immunoblots described in (A). For each cell line data were normalized to the insulin-treated, MK-2206 untreated condition. Each data point is the mean of 3–5 independent experiments ± S.E.M.

Figure 3. Akt1^W80A and Akt2^W80A regulate insulin-mediated FoxO1 nuclear exclusion and GLUT4 translocation

(A) Percentage of cells expressing Akt1^WT, Akt1^W80A, Akt2^WT or Akt2^W80A that display cytosolic FoxO1–GFP in basal (starved) and following 1 nM insulin stimulation for 30 min. (B) Surface to total HA–GLUT4–GFP in basal (starved) and 1 nM insulin-stimulated adipocytes expressing Akt1^WT, Akt1^W80A, Akt2^WT and Akt2^W80A. For each experiment surface to total GLUT4 values were normalized to that of insulin-treated Akt1^WT cells. Dose–response analyses of MK-2206 inhibition of insulin-mediated FoxO1–GFP nuclear exclusion (C) and HA–GLUT4–GFP translocation (D) in adipocytes. Adipocytes expressing WT or W80A Akt mutants were pre-treated with MK-2206 for 1 h as noted followed by stimulation with 1 nM insulin for 30 min. (E) Insulin-dose response analyses of HA–GLUT4–GFP translocation in Akt1^WT and Akt2^WT adipocytes in the presence or absence of 1 μM MK-2206. For each experiment, surface to total GLUT4 measurements were normalized to that of 10 nM insulin-treated Akt1^WT cells. (F), Insulin dose–response analyses of HA–GLUT4–GFP plasma membrane translocation in Akt1^W80A and Akt2^W80A adipocytes in the presence or absence of 1 μM MK-2206. For each experiment surface to total GLUT4 measurements were normalized to that of 10 nM insulin-treated Akt1^W80A cells. (G) Time course analyses of insulin-mediated GLUT4 translocation in adipocytes. Cells were treated with vehicle or 1 μM MK-2206 for 1 h prior to 1 nM insulin stimulation. For each experiment surface to total GLUT4 values were normalized to that of 15 min insulin-treated Akt1^W80A cells. All data are the average of at least three independent experiments ± S.E.M.

Figure 4. Akt1-specific signalling controls adipocyte differentiation

(A) Oil–Red-O staining of adipocytes expressing WT or W80A Akt mutants differentiated in the presence or absence of MK-2206. Neutral lipid staining (B) and quantification (C) using LipidTox green stain in adipocytes expressing WT or W80A Akt mutants differentiated in the presence or absence of MK-2206. Bars are the mean of four independent experiments ± S.E.M. *P<0.01 (paired t test) (D) RT-qPCR analyses of adipocyte markers in cells expressing WT or W80A Akt constructs differentiated into adipocytes in the presence or absence of MK-2206. Gene expression was normalized to the housekeeping gene RPLP0. Data represent the log₂-fold change in gene expression in MK-2206-treated compared with -untreated conditions. Data are the mean of 3–6 independent experiments ± S.E.M. *P<0.05 (ANOVA). Statistical significance for the data comparisons Akt1^W80A compared with Akt1^WT and Akt2^W80A compared with Akt2^WT is shown. (E) Immunoblot analyses of protein extracts derived from pre-adipocytes expressing WT and W80A Akt isoforms. Confluent pre-adipocytes were starved for 2 days and then treated with induction medium for 15 min with or without 2 μM MK-2206. (F) EdU incorporation in pre-adipocytes expressing WT or W80A Akt1 and Akt2 constructs. Cells were kept for 16 h in the presence of growth media (no induction), differentiation media (induction) or differentiation media plus MK-2206 (induction, MK-2206), followed by a 2-h incubation with EdU. Nuclei were counterstained with Hoechst. (G) Percentage of cells with EdU incorporation determined by fluorescence microscopy. Data are the mean of four independent experiments ± S.E.M. *P<0.01 (paired t test induction compared with induction/MK-2206).