Pyruvate dehydrogenase kinase 4: regulation by thiazolidinediones and implication in glyceroneogenesis in adipose tissue

Abstract

Objective: Pyruvate dehydrogenase complex (PDC) serves as the metabolic switch between glucose and fatty acid utilization. PDC activity is inhibited by PDC kinase (PDK). PDC shares the same substrate, i.e., pyruvate, as glyceroneogenesis, a pathway controlling fatty acid release from white adipose tissue (WAT). Thiazolidinediones activate glyceroneogenesis. We studied the regulation by rosiglitazone of PDK2 and PDK4 isoforms and tested the hypothesis that glyceroneogenesis could be controlled by PDK.

Research design and methods: Rosiglitazone was administered to Zucker fa/fa rats, and then PDK4 and PDK2 mRNAs were examined in subcutaneous, periepididymal, and retroperitoneal WAT, liver, and muscle by real-time RT-PCR. Cultured WAT explants from humans and rats and 3T3-F442A adipocytes were rosiglitazone-treated before analyses of PDK2 and PDK4 mRNA and protein. Small interfering RNA (siRNA) was transfected by electroporation. Glyceroneogenesis was determined using [1-(14)C]pyruvate incorporation into lipids.

Results: Rosiglitazone increased PDK4 mRNA in all WAT depots but not in liver and muscle. PDK2 transcript was not affected. This isoform selectivity was also found in ex vivo-treated explants. In 3T3-F442A adipocytes, Pdk4 expression was strongly and selectively induced by rosiglitazone in a direct and transcriptional manner, with a concentration required for half-maximal effect at 1 nmol/l. The use of dichloroacetic acid or leelamine, two PDK inhibitors, or a specific PDK4 siRNA demonstrated that PDK4 participated in glyceroneogenesis, therefore altering nonesterified fatty acid release in both basal and rosiglitazone-activated conditions.

Conclusions: These data show that PDK4 upregulation in adipocytes participates in the hypolipidemic effect of thiazolidinediones through modulation of glyceroneogenesis.

FIG. 1.

Rosiglitazone induces PDK4 gene expression in fa/fa rat adipose tissue in vivo. Male Zucker fa/fa rats were treated for 4 days with rosiglitazone (5 mg · kg⁻¹ · day ⁻¹). Rats were killed, and tissues were removed. Total RNA was extracted from subcutaneous, periepididymal, and retroperitoneal WAT, liver, and soleus muscle. PDK4 (A) and PDK2 (B) mRNA were analyzed by real-time RT-PCR. Values were normalized to 18S rRNA and expressed as percentage of untreated rats. Results are mean ± SE from four rats in each group. *P < 0.05; **P < 0.01; ***P < 0.001.

FIG. 2.

Rosiglitazone induces Pdk4 expression in rat and human adipose tissue ex vivo and in 3T3–F442A adipocytes. Rat periepididymal (A) and human subcutaneous abdominal WAT (B) were cultured in DMEM containing 12.5 mmol/l glucose and treated or not with 1 μmol/l rosiglitazone for 5 h. C: 3T3–F442A adipocytes at day 6 of differentiation were cultured in DMEM containing 25 mmol/l glucose and treated or not for 2 h with 1 μmol/l rosiglitazone. Total RNA was extracted, and PDK4, PDK2, and PC mRNA were analyzed by real-time RT-PCR with normalization to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA. D: Protein extracts were obtained from differentiated 3T3–F442A adipocytes and subjected to Western blotting using antibodies against PDK4, PDK2, and the E2 subunit of PDC complex. Data are means ± SE from three to five independent experiments, each performed in triplicate, and are expressed as percent of control. **P < 0.01.

FIG. 3.

Mechanism of rosiglitazone stimulation of Pdk4 expression in 3T3–F442A adipocytes. Quantitative changes in the abundance of PDK4 mRNA in 3T3–F442A adipocytes either cultured for the indicated times with 1 μmol/l rosiglitazone in the absence or in the presence of 80 μmol/l DRB (A) or treated for 2 h with the indicated concentrations of rosiglitazone (B). In C, 3T3–F442A adipocytes were treated or not with 1 μmol/l rosiglitazone for 2 h before DRB addition, and then mRNA for PDK4 was estimated over time. Values were normalized to GAPDH and are expressed as percentage of control. In D, 40 μg mitochondrial proteins from adipocytes treated or not with 1 μmol/l rosiglitazone for 18 h were subjected to Western blotting. Densitometry values were normalized to PDC-E2 and expressed as percentage of control. Results are means ± SE of three to five independent experiments, each performed in triplicate. *P < 0.05; **P < 0.01; ***P < 0.001.

FIG. 4.

PDK4 inhibition alters basal and rosiglitazone-activated glyceroneogenesis. A: Using [1-¹⁴C]pyruvate incorporation into neutral lipids, glyceroneogenic flux was monitored in isolated adipocytes from rat periepididymal fat pads for 2 h in the presence or not of PDK inhibitors, leelamine, or DCA at the indicated concentrations. 3T3–F442A adipocytes were treated or not with 1 μmol/l rosiglitazone for 18 h, and then glyceroneogenic flux (B) and medium NEFA concentration (C) were monitored in the presence or not of leelamine for 2 h. Results are means ± SE of three independent experiments, each performed in triplicate. **P < 0.01; ***P < 0.001 vs. control; §§§ P < 0.001 vs. rosiglitazone treatment.

FIG. 5.

Influence of PDK4 siRNA on gene expression and glyceroneogenesis in 3T3–F442A adipocytes. 3T3–F442A adipocytes were transfected with a control or a PDK4-specific siRNA. Forty-eight hours after transfection, cells were used for mRNA studies (A) or [1-¹⁴C]pyruvate incorporation experiments (B). A: Total RNA was extracted, and concentration of PDK4, PDK2, PDP1, PDP2, pyruvate carboxylase, PEPCK-C, and IFN-β mRNA was analyzed by real-time RT-PCR with normalization to 18S rRNA. Control or PDK4 siRNA-transfected adipocytes were treated or not with 0.1 μmol/l rosiglitazone for 18 h, and glyceroneogenic flux was monitored. Results are means ± SE of three independent experiments, each performed in triplicate. *P < 0.05, **P < 0.01; ***P < 0.001 vs. control; §§P < 0.01 vs. rosiglitazone treatment.

FIG. 6.

Pyruvate routing for fatty acid reesterification: influence of thiazolidinediones. Pyruvate is either decarboxylated by PDC into acetyl-CoA or carboxylated to oxaloacetate (OA) by PC in mitochondria. Oxaloacetate is the precursor of G3P, and PEPCK-C is involved in this process, allowing for reesterification into triglyceride (TG) of NEFA arising from lipolysis. This could lead to decreased NEFA release in the blood flow, without affecting glycerol release. PDC is mainly regulated by a phosphorylation/dephosphorylation cycle. PDK4 inactivates PDC, thereby regulating pyruvate availability for glyceroneogenesis in adipocytes. Rosiglitazone induces both PDK4 and PEPCK-C, thereby reducing pyruvate flux toward acetyl-CoA and favoring G3P production.