The G(0)/G(1) switch gene 2 regulates adipose lipolysis through association with adipose triglyceride lipase

Abstract

Adipose triglyceride lipase (ATGL) is the rate-limiting enzyme for triacylglycerol (TAG) hydrolysis in adipocytes. The precise mechanisms whereby ATGL is regulated remain uncertain. Here, we demonstrate that a protein encoded by G(0)/G(1) switch gene 2 (G0S2) is a selective regulator of ATGL. G0S2 is highly expressed in adipose tissue and differentiated adipocytes. When overexpressed in HeLa cells, G0S2 localizes to lipid droplets and prevents their degradation mediated by ATGL. Moreover, G0S2 specifically interacts with ATGL through the hydrophobic domain of G0S2 and the patatin-like domain of ATGL. More importantly, interaction with G0S2 inhibits ATGL TAG hydrolase activity. Knockdown of endogenous G0S2 accelerates basal and stimulated lipolysis in adipocytes, whereas overexpression of G0S2 diminishes the rate of lipolysis in both adipocytes and adipose tissue explants. Thus, G0S2 functions to attenuate ATGL action both in vitro and in vivo and by this mechanism regulates TAG hydrolysis.

Figure 1. Regulation of G0S2 protein expression

A. Tissue distribution of G0S2 protein by immunoblotting analysis. 1, epididymal white adipose tissue; 2, interscapular brown adipose tissue; 3, heart; 4, skeletal muscle (hindlimb, upper leg muscle); 5, liver; 6, lung; 7, spleen; 8, pancreas; 9, kidney. β-tubulin was used as a loading control. B. Expression of G0S2 protein was analyzed by immunoblotting in lysates from differentiated 3T3-L1 and T37i adipocytes. Day 1 represents undifferentiated preadipocytes. aP2 was used as an adipocytes differentiation marker. UCP1 was used as a brown adipocyte specific marker. β-actin was used as a loading control. C. Immunoprecipitation was performed in extracts (equal amount of total protein) of mouse epididymal adipose using either anti-G0S2 serum (+) or preimmune serum (−). G0S2 in the immunoprecipitates were analyzed by immunoblotting. Mice used are wild type mice (wt) and db/db mice on normal chow, and wild type mice fed with high fat diet for 8 weeks (wt/HFD). D. Following 4 hr serum deprivation, 3T3-L1 and T37i adipocytes were treated for 8 h with 50 nM insulin (In), 10 nM triiodothyronine (T3), 10 µM GW501516 (GW), 1 µM rosiglitazone (Rosi), 1 µM isoproterenol/0.25 mM IBMX (Iso) or 20 ng/ml TNFα in serum-free medium. Immunoblotting was performed to detect the levels of G0S2 in the cell lysates, using β-actin as a loading control.

Figure 2. Overexpression of G0S2 inhibits lipid droplet degradation mediated by nutrient withdrawal

A. Immunofluorescence staining with anti-G0S2 antibodies was performed to reveal localization of overexpressed G0S2 in HeLa cells. Lipid droplets were co-stained with BODIPY 493/503 fluorescence dye. B. HeLa cells transiently expressing G0S2 were incubated under normal growth conditions with 400 µM of oleic acid complexed to albumin at a molar ratio of 8:1 for 3 h (upper panel) or 24 h (middle panel). A separate set of cells were incubated in serum-free and glucose-free medium for 4 h following 24 h of incubation with 400 µM of oleic acid (lower panel). Immunofluorescence staining with anti-G0S2 antibodies was performed to reveal the transfected cells. Lipid droplets were co-stained with BODIPY 493/503 fluorescence dye. C. Quantification of lipid droplet diameter. An average of 80 lipid droplets was measured for each point. Data are shown as mean ± SD, **p < 0.01, t-test.

Figure 3. G0S2 inhibits lipid droplet degradation mediated by ATGL

A. HeLa cells transfected with ATGL in the absence (upper panel) or presence (lower panel) of G0S2-FLAG were incubated under normal growth conditions with 400 µM of oleic acid complexed to albumin for 3 h. Immunofluorescence staining was performed by using anti-ATGL (red) and anti-FLAG (blue) antibodies. Lipid droplets were co-stained with BODIPY 493/503 fluorescence dye. B & C. Stable HeLa cell clones with or without untagged G0S2 expression were treated with ATGL siRNA (ATGL KD) or control mismatch siRNA (Ctrl KD). Protein expression was analyzed by immunoblotting with anti-ATGL and anti-G0S2 antibodies (B). The cells were incubated in 400 µM of oleic acid for 3 h and the intracellular TAG content was determined as described in Materials and Methods. The data were normalized with the total protein amounts in the cell extracts (C) (data are shown as mean ± SD and represent three independent experiments, **p < 0.01, t-test).

Figure 4. G0S2 selectively inhibits the TAG hydrolase activity of ATGL

A. Schematic structures of generated mutants of murine G0S2 and ATGL. The location of the hydrophobic domain (HD) in both proteins is indicated as gray bars, and the patatin domain (PT) in ATGL is indicated by a black region. B. HeLa cells from control clone or G0S2-expressing clone were transfected with vector alone, ATGL, ATGLΔHD. 24 h after transfection, ATGL and G0S2 proteins were detected with anti-ATGL and anti-G0S2 antibodies in immunoblotting using cell extracts. TAG hydrolase activity in cell extracts was measured using ³H-labeled triolein as substrate, and was normalized with the total protein levels of the cell extracts. C. Extracts of HeLa cells singly expressing ATGL, CGI-58 or G0S2 were mixed in various combinations, and TAG hydrolase activity was determined. Proteins in parallel mixtures were revealed in immunoblotting. D. ATGL and CGI-58 produced by using in vitro translation system were mixed with increasing amounts of purified recombinant G0S2, and were subjected to TAG hydrolase activity assays. E. Increasing amounts of recombinant purified G0S2 was added to extracts (100 µg total protein) of epididymal (WAT) and interscapular (BAT) fat tissue, and TAG hydrolase activity was determined. All data are shown as mean ± SD and represent three independent experiments, *p < 0.05, **p < 0.01, ***p < 0.001, t-test.

Figure 5. G0S2 specifically interacts with ATGL

A. HeLa cells were co-transfected with vector alone or G0S2-FLAG together with different ATGL constructs (1, ATGL+vector; 2, ATGL+G0S2-FLAG; 3, ATGLΔHD+vector; 4, ATGLΔHD+G0S2-FLAG; 5, ATGLΔPT+vector; 6, ATGLΔPT+G0S2-FLAG). G0S2-FLAG proteins were immunoprecipitated with anti-FLAG antibodies. G0S2 and ATGL proteins in immunoprecipitates and lysates were detected by immunoblotting with FLAG and ATGL antibodies. B. HeLa cells were co-transfected with wild type ATGL along with vector alone or various G0S2-FLAG constructs. Immunoprecipitation and immunoblotting analysis were performed as described in A. C. 3T3-L1 adipocytes were pretreated with or without 1 µM isoproterenol or 100 nM insulin for 30 min. Immunoprecipitation of endogenous ATGL was performed in the lysates in the presence or absence of an ATGL epitope-blocking peptide. Nonspecific control and HSL antibodies were used as controls. G0S2, ATGL and HSL in precipitates and lysates were analyzed by immunoblotting with respective antibodies. D. Anti-FLAG immunoprecipitation was performed following the in vitro transcription/translation reactions. ATGL and FLAG-tagged G0S2 proteins in precipitates were analyzed by immunoblotting with anti-ATGL and anti-FLAG antibodies. E. The TAG hydrolase activity in the cell extracts was measured using ³H-labeled triolein as substrate. The activity was normalized with the total protein levels of the cell extracts, and is shown in relation to basal activity detected in vector-transfected cells from the control clone. Data are shown as mean ± SD and represent three independent experiments, **p < 0.01, t-test.

Figure 6. Lipid droplet localization of ATGL and G0S2 in adipocytes

A. 3T3 L1 adipocytes were treated with or without 1µm isoproterenol or insulin for 30 min. Lipid droplets were isolated by ultracentrifugation. Total and lipid droplet-associated proteins were subjected to immunoblotting using antibodies against ATGL, perilipin, HSL, G0S2 and aP2. B. Immunofluorescence staining with anti-ATGL antibodies was performed to reveal localization of endogenous ATGL in 3T3-L1 adipocytes pretreated with or without 1 µM isoproterenol/0.25 mM IBMX for 30 min. Lipid droplets were co-stained with BODIPY 493/503. C. siRNA-mediated knockdown was performed by electroporating 3T3-L1 adipocytes with either control siRNA (Ctrl) or ATGL-specific siRNA (KD). 3 days later, cells were treated with or without 1 µM isoproterenol for 30 min followed by lipid droplets isolation. Total and lipid droplet-associated levels of ATGL, HSL, G0S2 and perilipin were analyzed by immunoblotting.

Figure 7. G0S2 attenuates basal and stimulated lipolysis in adipocytes

A. siRNA-mediated knockdown of G0S2 in 3T3-L1 adipocytes was achieved by electroporating cells with either control siRNA or G0S2-specific siRNA. Cells were treated 3 days later with (stimulated) or without (basal) 1 µm isoproterenol for 30 min, 60 min and 2 h. Basal and stimulated FFA and glycerol release were measured and normalized with the total protein levels in the cell extracts. Expression of G0S2 protein was analyzed by immunoblotting, using β-actin as a loading control. Data are shown as mean ± SD and represent three independent experiments, *P < 0.05, **P < 0.01, one-way ANOVA. B. Overexpession of G0S2 in 3T3-L1 adipocytes was achieved by infecting cells with recombinant adenovirus encoding murine G0S2 (Ad-G0S2). A null virus was used as a control. G0S2 expression was analyzed by immunoblotting 3 days after infection: 1, null virus at high dosage; 2, null virus at low dosage; 3, Ad-G0S2 at high dosage; 4, Ad-G0S2 at low dosage. 3 days following infection with either null or Ad-G0S2 at high dosage, lipolysis of 3T3-L1 adipocytes was measured as described in A. Data are shown as mean ± SD and represent three independent experiments, *P < 0.05, **P < 0.01, ***P < 0.001, one-way ANOVA. C. Overexpession of G0S2 in gonadal fat explants was achieved by infection with Ad-G0S2 using null virus as a control. 2 days post infection explants were treated with (stimulated) or without (basal) 1 µm isoproterenol for 2 h. Basal and stimulated FFA release was measured and normalized with the total mass of explants. Data are shown as mean ± SD and represent three independent experiments, **P < 0.01, t-test.

Figure 8. Model for the regulation of adipose lipolysis by G0S2

Under basal conditions when the lipolytic rate is very low, ATGL is mostly localized in ER-related membranes in the cytoplasm. At least a small fraction of ATGL is in complex with G0S2. Upon β-adrenergic stimulation, PKA activation results in phosphorylation of perilipin A at multiple sites. Consequently, ATGL, both G0S2-bound and –unbound form, undergoes translocation onto lipid droplets. Translocation of unbound ATGL along with HSL leads to acute activation of TAG hydrolysis. Prolonged β-adrenergic stimulation subsequently downregulates protein level of G0S2, thereby releasing more ATGL for sustained lipolysis.