PPARgamma regulates adipose triglyceride lipase in adipocytes in vitro and in vivo

Abstract

Peroxisome proliferator-activated receptor-gamma (PPARgamma) regulates adipocyte genes involved in adipogenesis and lipid metabolism and is the molecular target for thiazolidinedione (TZD) antidiabetic agents. Adipose triglyceride lipase (ATGL) is a recently described triglyceride-specific lipase that is induced during adipogenesis and remains highly expressed in mature adipocytes. This study evaluates the ability of PPARgamma to directly regulate ATGL expression in adipocytes in vitro and in vivo. In fully differentiated 3T3-L1 adipocytes, ATGL mRNA and protein are increased by TZD and non-TZD PPARgamma agonists in a dose- and time-dependent manner. Rosiglitazone-mediated induction of ATGL mRNA is rapid and is not inhibited by the protein synthesis inhibitor cycloheximide, indicating that intervening protein synthesis is not required for this effect. Rosiglitazone-mediated induction of ATGL mRNA and protein is inhibited by the PPARgamma-specific antagonist GW-9662 and is also significantly reduced following siRNA-mediated knockdown of PPARgamma, supporting the direct transcriptional regulation of ATGL by PPARgamma. In vivo, ATGL mRNA and protein are increased by rosiglitazone treatment in white and brown adipose tissue of mice with and without obesity due to high-fat diet or leptin deficiency. Thus, PPARgamma positively regulates ATGL mRNA and protein expression in mature adipocytes in vitro and in adipose tissue in vivo, suggesting a role for ATGL in mediating PPARgamma's effects on lipid metabolism.

Fig. 1

Effect of thiazolidinediones (TZD) and non-TZD peroxisome proliferator-activated receptor-γ (PPARγ) agonists on adipose triglyceride lipase (ATGL) mRNA and protein expression in 3T3-L1 adipocytes. Fully differentiated 3T3-L1 adipocytes were treated with DMSO (Cont), rosiglitazone (Rosi), troglitazone (Trog), ciglitazone (Cig), pioglitazone (Pio), MCC-555 (MCC), N-(9-fluorenylmethoxycarbonyl)-l-leucine (FMOC), or 15-deoxy-Δ^12,14-PGJ₂ (PGJ₂) at 10 µM for 24 h (n = 6–7/group for both mRNA and protein). A: ATGL mRNA expression was determined by quantitative PCR (qPCR), normalized to 18S ribosomal RNA, and expressed relative to the control group. *P < 0.05 for comparisons between treatment and control groups. B: ATGL protein expression was determined by Western blot analysis using the Ran GTPase (RAN) as a loading control (representative blot shown). ATGL protein expression was highly correlated with mRNA expression (r = 0.862, P < 0.05) and was increased compared with the control group for all treatments (P < 0.05).

Fig. 2

Dose response and time course for ATGL and adiponutrin expression in 3T3-L1 adipocytes treated with rosiglitazone. Dose response for ATGL mRNA (A), ATGL protein (C), and adiponutrin mRNA (E): fully differentiated 3T3-L1 adipocytes were treated with increasing doses of Rosi from 0 to 10,000 nM for 24 h (n = 6/group for both mRNA and protein). Time course for ATGL mRNA (B), ATGL protein (D), and adiponutrin mRNA (F): fully differentiated 3T3-L1 adipocytes were treated with 100 nM Rosi for increasing time intervals from 0 to 24 h (n = 6/group). ATGL and adiponutrin mRNA expression were determined by qPCR, normalized to 18S ribosomal RNA, and expressed relative to the control group. P < 0.05 for effect of dose and time on gene expression for both ATGL and adiponutrin. ATGL protein expression was determined by Western blot analysis, with RAN as a loading control (representative blots shown). P < 0.05 for effect of dose and time on ATGL protein expression. ATGL protein expression correlated with mRNA expression (r = 0.894 and 0.864 for dose and time, respectively, P < 0.05).

Fig. 3

Effect of cycloheximide (CHX) on Rosi-mediated regulation of ATGL mRNA expression in 3T3-L1 adipocytes. Fully differentiated 3T3-L1 adipocytes were pretreated with either DMSO or 5 µg/ml CHX for 30 min, followed by addition of either DMSO or 100 nM Rosi for an additional 12 h (n = 7/group). ATGL mRNA expression was determined by qPCR, normalized to 18S ribosomal RNA, and expressed relative to the control group. P < 0.05 for overall effect of Rosi treatment on AGTL expression. *P < 0.05 for the effect of Rosi within the same CHX treatment; †P < 0.05 for the effect of CHX within the same Rosi treatment. ns, Not significant.

Fig. 4

Effect of the PPARγ antagonist GW-9662 (GW) on Rosi-mediated regulation of ATGL mRNA and protein expression in 3T3-L1 adipocytes. Fully differentiated 3T3-L1 adipocytes were treated with DMSO, 10 µM GW alone, 100 nM Rosi alone, or 100 nM Rosi plus either 0.1 or 10 µM GW (n = 3/group for both mRNA and protein). A: ATGL mRNA expression was determined by qPCR, normalized to 18S ribosomal RNA, and expressed relative to the control group. P < 0.05 for overall effect of Rosi treatment on ATGL expression. *P < 0.05 for the effect of Rosi within the same GW treatment; †P < 0.05 for the effect of GW within the same Rosi treatment. P < 0.05 for dose effect of GW on inhibition of Rosi-mediated induction of ATGL expression. B: ATGL protein expression was determined by Western blot analysis, with RAN as a loading control (representative blot shown). ATGL protein expression correlated with mRNA expression (r = 0.887, P < 0.05). Significance of statistical comparisons between groups were the same as stated for mRNA.

Fig. 5

Effect of small interfering RNA (siRNA)-mediated knockdown of PPARγ ATGL mRNA and protein expression in 3T3-L1 adipocytes. 3T3-L1 adipocytes were electroporated in the presence of either control or PPARγ-specific siRNAs on day 7 of differentiation. Cells were collected 48 h later (day 9 of differentiation) for determination of mRNA (3 experiments with n = 3 each) and protein expression (n = 2). A: PPARγ and ATGL protein expression were determined by Western blot analysis, with RAN as a loading control (representative blot shown). B: ATGL mRNA expression was determined by qPCR, normalized to 36B4 RNA, and expressed relative to ATGL expression in the control group. *P < 0.05.

Fig. 6

In vivo effect of Rosi on ATGL mRNA and protein expression in white (WAT) and brown adipose tissue (BAT). Lep^?ob/+ and Lep^ob/ob mice fed chow or high-fat diet (HFD) for 18 wk were treated with either carboxymethyl cellulose vehicle (control) or 4 mg·kg⁻¹ ·day⁻¹ Rosi for 10 days by oral gavage (n = 6–8/group for mRNA and protein). A and B: ATGL mRNA expression was determined in perigonadal white adipose tissue (PGAT; A) or BAT (B) by qPCR, normalized to 18S ribosomal RNA, and expressed relative to the vehicle-treated chow-fed Lep^?ob/+ control group. P < 0.05 for overall effect of Rosi treatment and genotype on ATGL expression in both PGAT and BAT. *P < 0.05 for comparisons between vehicle- and Rosi-treated animals within each genotype/diet group; †P < 0.05 for comparisons between Lep^?ob/+ vs. Lep^ob/ob mice or between chow-fed vs. HFD-fed mice within the same Rosi treatment group. C and D: ATGL protein expression in PGAT (C) or BAT (D) by Western blot analysis, with RAN as a loading control (representative blots shown). ATGL protein expression correlated with mRNA expression (r = 0.888 and 0.886 for PGAT and BAT, respectively, P < 0.05). Significance of statistical comparisons between groups was the same as stated for mRNA.