Adenovirus Protein E4-ORF1 Activation of PI3 Kinase Reveals Differential Regulation of Downstream Effector Pathways in Adipocytes

Abstract

Insulin activation of phosphatidylinositol 3-kinase (PI3K) regulates metabolism, including the translocation of the Glut4 glucose transporter to the plasma membrane and inactivation of the FoxO1 transcription factor. Adenoviral protein E4-ORF1 stimulates cellular glucose metabolism by mimicking growth-factor activation of PI3K. We have used E4-ORF1 as a tool to dissect PI3K-mediated signaling in adipocytes. E4-ORF1 activation of PI3K in adipocytes recapitulates insulin regulation of FoxO1 but not regulation of Glut4. This uncoupling of PI3K effects occurs despite E4-ORF1 activating PI3K and downstream signaling to levels achieved by insulin. Although E4-ORF1 does not fully recapitulate insulin's effects on Glut4, it enhances insulin-stimulated insertion of Glut4-containing vesicles to the plasma membrane independent of Rab10, a key regulator of Glut4 trafficking. E4-ORF1 also stimulates plasma membrane translocation of ubiquitously expressed Glut1 glucose transporter, an effect that is likely essential for E4-ORF1 to promote an anabolic metabolism in a broad range of cell types.

Keywords: E4ORF1; GLUT1; GLUT4; PI3-kinase; PI3K-AKT signaling; RAB10.

Figure 1. Ad5 E4-ORF1 expression enhances growth factor-stimulated Glut4 translocation without an effect on general endocytic trafficking

(A) Quantitative RT-PCR showing relative mRNA expression of E4-ORF1 in control and stably E4-ORF1 expressing adipocyt. (B) Representative Western blot of whole cell lysates from control and stably FLAG-E4-ORF1 expressing adipocytes probed for FLAG. (C) Anti-Flag IF for transiently expressing (Left) and stably FLAG-E4-ORF1 (Right) expressing adipocyte. Epifluorescence (EPI) images, anti-Flag staining in green and nuclei (DAPI) in red are shown. Arrows indicate cells with prominent plasma membrane anti-Flag staining. Scale bar 20 μ. (D) Schematic of Glut4 reporter depicting the location of the tag. (E, F) Quantification of surface (Cy3 anti-HA fluorescence)-to-total (GFP fluorescence) ratio of HA-Glut4-GFP expression in transient (E) and stable (F) E4-ORF1-expressing adipocytes in basal and insulin treated condition. Data is normalized to the HA-Glut4-GFP surface-to-total value for control cells with 1nM insulin treatment for each experiment. N = 6 assays. (G) Representative Airyscan confocal images showing GFP expression for HA-Glut4-GFP and endogenous Syntaxin 6 staining (Cy3) in basal and 0.1nM insulin treated control and E4-ORF1 expressing adipocytes. Scale bar 5 μm. (H) Quantification of surface-to-total ratio of HA-Glut4-GFP in control, control adipocytes expressing FLAG-E4-ORF6, E4-ORF1 stable adipocytes and E4-ORF1 stable adipocytes co-expressing FLAG-E4-ORF6 respectively, under basal and insulin treated conditions. N = 3 assay. (I) Quantification of surface-to-total ratio of HA-Glut4-GFP expression in control and stable E4-ORF1 adipocytes under basal, insulin and IGF treated conditions. Data are normalized to the HA-Glut4-GFP surface-to-total value for insulin treated control cells for each experiment. N = 3 assay. (J) Quantification of surface-to-total ratio of transferrin receptor in control and E4-ORF1 expressing adipocytes under basal and insulin treated conditions. Data normalized to the HA-Glut-GFP surface-to-total value for 1nM insulin treated control cells for each experiment. N = 3 assays. Data represents mean normalized values ± SEM. *p<0.05, **p<0.005. See also Figure S1.

Figure 2. E4-ORF1 activates Akt

Control and stably expressing E4-ORF1 adipocytes were treated with indicated insulin concentrations, and the lysates were subjected to Western blot analysis. Representative blot showing (A) phospho-Akt S473, (B) phospho-Akt T308 and total Akt expression (upper) and respective densitometric analysis of immunoblots (lower), (C) phospho-AS160 T642 and total AS160 expression (upper) and densitometric analysis of immunoblots (lower), (D) phospho-FoxO1 S256 and total FoxO1 expression (upper) and densitometric analysis of immunoblots (lower). For densitometric analysis N = 3 assays. Mean values ± SEM, *p<0.05.

Figure 3. E4-ORF1 signals through p110a and Akt and mimics the effects of insulin on FoxO1

(A) Both control and stable E4-ORF1 adipocytes were transfected with FoxO1-GFP and treated with indicated insulin concentration. Data shows percentage of cells with cytosolic FoxO1-GFP under basal and insulin treated conditions. N = 4 assays. (B) Control and stably expressing E4-ORF1were pre-treated with PIK75 (0.1μM), TGX221 (0.1μM) or with DMSO (vehicle) for 1 h followed by insulin stimulation. Quantification of surface-to-total ratios of HA-Glut4-GFP are shown. Data are normalized to the HA-GLUT4-GFP surface-to-total value for insulin treated control cells for each experiment. N = 4 assays. (C) Quantification of surface-to-total ratio of HA-Glut4-GFP measured in control, transiently expressing Flag-E4-ORF1 wild type and Flag-E4-ORF1-V128A adipocytes under basal and insulin treated conditions. Data is normalized to the HA-Glut-GFP surface-to-total value for basal control cells for each experiment. N = 4 assays. (D) Control, transiently expressing Flag-E4-ORF1 wild type and Flag-E4-ORF1-V128A adipocytes were transfected with FoxO1-GFP. Data shows percentage of cells with nuclear FoxO1-GFP under basal condition. N = 3 assays. (E) Control and adipocytes stably expressing E4-ORF1 were pre-treated with DMSO (vehicle) or MK2206 (1μM) for 1 h followed by insulin stimulation. Quantification of surface-to-total ratios of HA-Glut4-GFP are shown. Data are normalized to the HA-Glut4-GFP surface-to-total value for insulin treated control cells for each experiment, N = 4 assays. (F) FoxO1-GFP expressing control and E4-ORF1-expressing adipocytes were treated with DMSO (vehicle), PIK75 (0.1μM), TGX221 (0.1μM) or with MK2206 (1μM) respectively for 1h. Data shows percentage of cells with nuclear FoxO1-GFP under basal condition. N = 4 assays. Data represents mean normalized values ± SEM, *p<0.05, **p<0.005, ***p<0.001. See also Figure S2 and S3.

Figure 4. E4-ORF1 colocalize with p110a and activates PI3K activity

(A) Both control and stable E4-ORF1-expressing adipocytes were transfected with GFP-PH1. Quantification of plasma membrane (TIRF GFP measurements) to total (EPI GFP measurements) ratio of GFP-PH1 expression in basal and insulin-stimulated conditions is shown. Data normalized to GFP-PH1 TIRF-to-EPI value for control cells under basal condition in each experiment. N = 4 assays. (B) GFP-PH1 expressing control and stable E4-ORF1 adipocytes were treated with PIK75 (0.1μM) for 1 h. Quantification of TIRF to EPI ratio of GFP-PH1 expression under basal condition is shown. Data normalized to GFP-PH1 TIRF-to-EPI value for control untreated cells in each experiment. N = 3 assays. (C) Representative immunoblots for p110α, p85 and actin expression in control and stable E4-ORF1-exprerssing adipocytes in basal conditions. (D) Representative TIRF images for FLAG-E4-ORF1 in basal and 0.1 nM insulin stimulated conditions. Scale bar 10 μm. (E) Quantification of TIRF imaging shown in (D). The fluorescence power of anti-Flag staining in the TIRF zone divided by anti-Flag fluorescence staining in EPI for individual cells in basal and 0.1 nM insulin stimulated condition. Individual points are data from single cells. N=40 cells. (F) Representative TIRF image showing colocalization between E4-ORF1 (anti-FLAG staining) and endogenous p110α stained puncta in basal adipocytes. Scale bar 10 μm. (G) Quantification of the endogenous p110α puncta that also contain E4-ORF1 (anti-FLAG staining). N= 2 assays. Mean values ± SD. (H) Quantification of the colocalization between E4-ORF1 (anti-FLAG staining) and endogenous insulin receptor. N= 2 assays. Mean values ± SD. Data represents mean normalized values ± SEM unless otherwise stated, *p<0.05, **p<0.005. See also Figure S4.

Figure 5. E4-ORF1 enhances Glut4 translocation to the plasma membrane by increasing Glut4 exocytosis independent of Rab10

(A) Quantitative RT-PCR showing relative mRNA expression of Rab10 in control adipocytes and stable Rab10 knockdown (KD) adipocytes. (B) Quantification of surface-to-total ratio of HA-Glut4-GFP in basal and insulin-stimulated conditions performed in Rab10 KD adipocytes ectopically re-expressing Rab10 (control), Rab10 KD adipocytes ectopically co-expressing Rab10 and Flag-E4-ORF1 (E4-ORF1), Rab10 KD adipocytes and Rab10 KD adipocytes expressing Flag-E4-ORF1 respectively. Data normalized to the HA-Glut4-GFP surface-to-total value for insulin treated control cells for each experiment. N = 4 assays. (C) Average insulin-stimulated exocytic rate constant for HA-Glut4-GFP measured in control and stable E4-ORF1 adipocytes. N = 5 assays. (D) Schematic of method to measure HA-Glut4-GFP accumulation in the TIRF zone (evanescent field) (TIRF-GFP/EPI-GFP fluorescence ratio) and the fraction of Glut4 that is inserted into the PM (anti-HA TIRF-Cy3/TIRF-GFP fluorescence ratio). (E) Representative TIRF images for HA-Glut4-GFP, showing anti-HA TIRF-Cy3 fluorescence and TIRF-GFP fluorescence in control and stable E4-ORF1 adipocytes under basal and 0.1 nM insulin stimulated conditions. Scale bar 10 μm. (F) Quantification of the fraction of Glut4 inserted into the PM by measuring anti-HA TIRF-Cy3 fluorescence to TIRF-GFP fluorescence ratio, in control and stable E4-ORF1 adipocytes under basal and insulin stimulated condition. Data normalized to the TIRF-Cy3-to-TIRF-GFP value for insulin treated control cells for each experiment. N = 3 assays. Data represents mean normalized values ± SEM, *p<0.05.

Figure 6. E4-ORF1 increases glucose uptake by increasing Glut1 in the plasma membrane of adipocytes

(A) ³H-2-Deoxyglucose uptake, normalized to amount of protein in each sample, in control and stable E4-ORF1 adipocytes in basal and insulin-stimulated conditions. Data normalized to uptake for insulin-treated control adipocytes for each experiment. N = 5 assays. (B) ³H-2-Deoxyglucose uptake, normalized to amount of protein in each sample, in control and stable E4-ORF1 adipocytes either treated with MK2006 or vehicle control (DMSO), under unstimulated conditions. N = 3 assays. (C) Representative immunoblots for Glut4, Glut1 and respective actin expression in control and stable E4-ORF1 adipocytes cell lysate. (D) Quantification of surface-to-total ratio of HA-Glut1 expression in control and E4-ORF1 expressing adipocytes under basal alone, pre-treatment with PIK75 and MK2206 respectively. N = 4 assays. (E) Quantification of surface-to-total ratio of HA-Glut1 expression in Rab10 KD adipocytes ectopically re-expressing Rab10 (control), Rab10 KD adipocytes ectopically co-expressing Rab10 and Flag-E4-ORF1 (E4-ORF1), Rab10 KD adipocytes and Rab10 KD adipocytes expressing Flag-E4-ORF1 respectively under basal conditions. E4-ORF1 data normalized to Glut1 surface-to-total amount in control adipocytes and Rab10 KD expressing E4-ORF1 data was normalized to Glut1 surface-to-total amount in Rab10 KD adipocytes for each experiment. N = 4 assays. (F) ELISA based quantification of Lactate in media of control, stable E4-ORF1 adipocytes and stable E4-ORF1 adipocytes with Glut1 KD. Data normalized to lactate concentration in control adipocytes for each experiment. N = 3 assays. Mean normalized values ± SD. (G) Mean-difference plot representing differential gene expression, 2136 genes. Genes colored in red represent differentially expressed genes (adj. p < 0.05) in stable E4-ORF1 adipocytes compared to control cells. N=3 per group. (H) Heat map of differentially expressed genes in control and stable E4-ORF1 adipocytes with more than 2 fold change in expression. Color-coded on basis of fold change relative to control adipocytes (290 total genes, Green-153 down regulated genes, Red-137 up regulated genes), N=3 per group, adj. p-value <0.05. (I) Quantification of relative mRNA expression levels of listed genes from differentially expressed gene set of E4-ORF1 adipocytes. Data represents mean normalized values ± SEM unless otherwise stated, *p<0.05, **p<0.005, ***p<0.001, ****p<0.0001. See also Figure S5, Table S1 and Table S2.