Sulforaphane induces lipophagy through the activation of AMPK-mTOR-ULK1 pathway signaling in adipocytes

Abstract

Lipophagy, a form of selective autophagy, degrades lipid droplet (LD) in adipose tissue and the liver. The chemotherapeutic isothiocyanate sulforaphane (SFN) contributes to lipolysis through the activation of hormone-sensitive lipase and the browning of white adipocytes. However, the details concerning the regulation of lipolysis in adipocytes by SFN-mediated autophagy remain unclear. In this study, we investigated the effects of SFN on autophagy in the epididymal fat of mice fed a high-fat diet (HFD) or control-fat diet and on the molecular mechanisms of autophagy in differentiated 3T3-L1 cells. Western blotting revealed that the protein expression of lipidated LC3 (LC3-II), an autophagic substrate, was induced after 3T3-L1 adipocytes treatment with SFN. In addition, SFN increased the LC3-II protein expression in the epididymal fat of mice fed an HFD. Immunofluorescence showed that the SFN-induced LC3 expression was co-localized with LDs in 3T3-L1 adipocytes and with perilipin, the most abundant adipocyte-specific protein, in adipocytes of mice fed an HFD. Next, we confirmed that SFN activates autophagy flux in differentiated 3T3-L1 cells using the mCherry-EGFP-LC3 and GFP-LC3-RFP-LC3ΔG probe. Furthermore, we examined the induction mechanisms of autophagy by SFN in 3T3-L1 adipocytes using western blotting. ATG5 knockdown partially blocked the SFN-induced release of fatty acids from LDs in mature 3T3-L1 adipocytes. SFN time-dependently elicited the phosphorylation of AMPK, the dephosphorylation of mTOR, and the phosphorylation of ULK1 in differentiated 3T3-L1 cells. Taken together, these results suggest that SFN may provoke lipophagy through AMPK-mTOR-ULK1 pathway signaling, resulting in partial lipolysis of adipocytes.

Keywords: Adipose; Autophagy; Cell biology; Dietary factors; Lipid droplets; Obesity.

Fig. 1.

Effects of SFN treatment on the expression of LC3 expression in 3T3-L1 adipocytes. (A, B) Differentiated 3T3-L1 cells were treated with SFN (100 μM), 500 nM rapamycin, or 0.1% DMSO (Control) for 10 d. Cells were then fixed, stained with oil red-O staining, and analyzed with a BZ-9000 fluorescence microscope. Lipid accumulation was quantified using the ImageJ imaging software program. Scale bar=50 μm. Values are the mean±S.E.M. (n=9). *P<.001 versus Control (one-way ANOVA with a Student-Newman post hoc test). (C, D) Western blotting of LC3 and p62 in the 3T3-L1 adipocytes. Differentiated 3T3-L1 cells were treated with 10 μM SFN or 0.1% DMSO (Control) for the indicated periods (0, 0.5, 1, 3, 6, and 9 h). β-actin was used as an internal control. Values are the mean±S.E.M. (n=3). *P<.05 versus 0 h (one-way ANOVA with a Student-Newman post hoc test). (E, F) After differentiated 3T3-L1 cells were treated with 10 μM SFN or 0.1% DMSO (Control) for 3 h, cells were fixed and stained for LC3 Ab conjugated to Alexa Fluor 568 (red). LDs were stained with BODIPY (493/503). Images were taken with a confocal laser-scanning microscope. Percentage colocalization of LC3 with BODIPY. Scale bar=10 μm. Values are the mean±S.E.M. (n=3). *P<.05 (two-tailed unpaired Student’s t test).

Fig. 2.

Effects of SFN treatment on the expression of LC3 in epididymal fat of obese mice. Seven-week-old male mice were fed a high-fat diet (HFD) or control-fat diet (CFD) for 8 weeks. After fasting for 18 h, the mice were randomly divided into two groups and treated with DMSO (Control) or SFN for 3 h. (A, B) Western blotting of LC3 and p62 in epididymal fat. β-actin was used as an internal control. (C, D) LC3 Ab and a secondary Ab conjugated to Alexa fluor 555 (red), and Perilipin Ab and a secondary Ab conjugated to Alexa Fluor 488 (green). Nuclear staining with DAPI is shown in blue. Images were taken with a fluorescence microscope. Percentage colocalization of LC3 with perilipin. Scale bar=100 μm. Values are the mean±S.E.M. (n=4). *P<.05 (two-tailed unpaired Student’s t test).

Fig. 3.

Effects of SFN on the autophagy flux in 3T3-L1 adipocytes. (A) Western blotting of LC3 in 3T3-L1 adipocytes. Differentiated 3T3-L1 cells were treated with 10 μM SFN with or without 100 nM Bafilomycin (BAF), or 0.1% DMSO (Control) for 6 h. β-actin was used as an internal control. (B) Schematic illustration of the tandem mCherry-EGFP-LC3 plasmid. (C) Differentiated 3T3-L1 cells stably expressing mCherry-EGFP-LC3 were treated with 10 μM SFN, 100 nM Bafilomycin (BAF), or 0.1% DMSO for 3 h. Scale bar=50 μm. (D) Schematic illustration of the measurement of autophagic flux with a GFP-LC3-RFP-ΔG probe. (E) Differentiated 3T3-L1 cells stably expressing GFP-LC3-RFP-ΔG were treated with SFN (10 μM) or 0.1% DMSO (Control) for 3 h. GFP/RFP ratio data were expressed as the fold-value against controls. Scale bar=50 μm. Values are the mean±S.E.M. (n=4). *P<.05, **P<.01 (one-way ANOVA with a Student-Newman post hoc test).

Fig. 4.

Effects of ATG5-knockdown on the SFN-induced cutdown of LDs in 3T3-L1 adipocytes. (A) Endogenous ATG5 mRNA and (B) LC3 protein were detected by real-time PCR and western blotting 48 h after transfection with 100 pmpl ATG5 siRNA (siATG5) or negative control (siCont.) in differentiated 3T3-L1 cells. Values are the mean±S.E.M. (n=3). *P<.05 (two-tailed unpaired Student’s t test). (C, D) Differentiated 3T3-L1 cells were treated with 10 μM SFN or DMSO (control) for 10 d after transfection with 100 pmol ATG5 siRNA or negative control. Cells were then fixed, stained with oil red-O staining, and analyzed with a BZ-9000 fluorescence microscope. Lipid accumulation was quantified using the ImageJ imaging software program. Scale bar=50 μm. Values are the mean±S.E.M. (n=10). (E) Differentiated 3T3-L1 cells were treated with 10 μM SFN for DMSO (control) 24 h after transfection with 100 pmol ATG5 siRNA or negative control. The culture medium was collected and assayed for NEFA content (mEq/L/mg protein). Values are the mean±S.E.M. (n=9–12). *P<.05 (one-way ANOVA with a Student-Newman post hoc test).

Fig. 5.

Induction of autophagy by SFN through the AMPK pathway in 3T3-L1 adipocytes. Differentiated 3T3-L1 cells were treated with 10 μM SFN at the indicated time (0, 0.5, 1, 3, 6, and 9 h). (A, B) The expression of phosphorylated AMPKα (p-AMPKα), total AMPKα (AMPKα), phosphorylated mTOR (p-mTOR), total mTOR (mTOR), phosphorylated ULK1 (p-ULK1), total ULK1 (ULK1), phosphorylated 4E-BP1 (p-4E-BP1), total 4E-BP1 (4E-BP1), (C, D) phosphorylated ERK1/2 (p-ERK1/2), total ERK1/2 (ERK1/2), Rubicon protein, and nuclear Nrf2 protein were determined by western blotting with specific antibodies. β-actin or Histone H3 was used as an internal control. Values are the mean±S.E.M. (n=3). *P<.05 versus 0 h (one-way ANOVA with a Student-Newman post hoc test). (E, F) Differentiated 3T3-L1 cells were treated with 10 μM SFN for DMSO (control) 24 h after transfection with 100 pmol AMPKα1/2 siRNA (siAMPKα) or negative control (siCont.) for 24 h. (E) The expression of p-AMPKα and AMPKα protein was determined by western blotting with specific antibodies. (F) The culture medium was collected and assayed for NEFA content (mEq/L/mg protein). Values are the mean±S.E.M. (n=4). *P<.05 (one-way ANOVA with a Student-Newman post hoc test). (G) Schematic illustration of the induction of lipophagy by SFN in 3T3-L1 adipocytes.