Adiponectin translation is increased by the PPARgamma agonists pioglitazone and omega-3 fatty acids

Abstract

Adiponectin, made exclusively by adipocytes, is a 30-kDa secretory protein assembled posttranslationally into low-molecular weight, middle-molecular weight, and high-molecular weight homo-oligomers. PPARgamma ligand thiozolidinediones, which are widely used in the treatment of type II diabetes, increase adiponectin levels. PPARgamma also has several putative ligands that include fatty acid derivatives. Overnight treatment of rat adipocytes with pioglitazone, docosahexaenoic acid (DHA), or eicosapentaenoic acid (EPA) triggered a twofold increase in the synthesis and secretion of HMW adiponectin, and this increase was blocked by the addition of PPARgamma inhibitor GW-9662. Inhibition of glycosylation using 2,2'-dipyridyl decreased the synthesis of high-molecular weight adiponectin by pioglitazone, EPA, and DHA, but there was increased secretion of trimeric adiponectin resulting from increased translation. Although pioglitazone, DHA, and EPA increased adiponectin synthesis by more than 60%, there was no increase in total protein synthesis and no corresponding change in adiponectin mRNA expression, indicating the upregulation of translation. We examined the possibility of transacting factors in the cytoplasmic extracts from adipocytes treated with pioglitazone or DHA. In vitro translation of adiponectin mRNA was inhibited by S-100 fraction of control adipocytes and increased by S-100 extracts from adipocytes treated with pioglitazone or DHA. Consistent with this observation, both pioglitazone and DHA treatments increased the association of adiponectin mRNA with the heavier polysome fractions. Together, these data suggest that pioglitazone and the fish oils DHA or EPA are PPARgamma agonists in adipocytes with regard to adiponectin expression, and the predominant mode of adiponectin stimulation is via an increase in translation.

Fig. 1.

Effect of pioglitazone and ω-3 fatty acids on the expression and secretion of adiponectin isoforms: A: rat adipocytes were treated with or without 25 μM palmitate, 3 μM pioglitazone, 25 μM eicosapentaenoic acid (EPA), or 25 μM docsahexaenoic acid (DHA) as described in methods. Cells and medium were collected from each treatment at 4 and 24 h and separated on a 4–20% nondenaturing and nonreducing gel and Western blotted as described in methods. Cellular adiponectin tends to increase during this time as the cells adapt to the tissue culture environment. The figure represents 1 of 4 similar experiments. B: analysis of total adiponectin and high-molecular weight (HMW) adiponectin at 4 h in cells and at 24 h in medium quantitated by densitometry and expressed in relation to total adiponectin with no treatment. C: adiponectin mRNA in rat adipocytes determined by real-time RT-PCR expressed in relation to 18S RNA. The results are the mean ± SE of 4 different experiments. *P < 0.02 vs. control HMW. **P < 0.05 vs. control total adiponectin. MMW, middle molecular weight; LMW, low molecular weight.

Fig. 2.

Pioglitazone (PIO) increases the synthesis and secretion of adiponectin. Rat adipocytes were treated with and without 3 μM PIO overnight (ON) and labeled with 100 μCi/ml [³⁵S]methionine ([³⁵S]met) for 15 min the next day. Time 0 represents the end of the pulse and beginning of the chase. The labeled cells were chased for the indicated time periods. At each point, cells (A) and medium (B) were immunoprecipitated and separated as a 30-kDa monomer by SDS-PAGE followed by autoradiography and densitometry. Radiolabeled adiponectin at different time points is represented as a percent of 0 min time of control. The results are the mean ± SE of 4 different experiments; *P < 0.02 vs. control.

Fig. 3.

Effect of PIO and ω-3 fatty acids on the synthesis and secretion of adiponectin in rat adipocytes. A: adiponectin synthesis from cells. Rat adipocytes were treated with and without 3 μM PIO, 25 μM EPA, and 25 μM DHA ON in presence and absence of the peroxisome proliferator-activated receptor-γ inhibitor GW-9662 (1.5 μM) and labeled with 100 μCi/ml [³⁵S]met for 15 min. Equal TCA-precipitable counts were immunoprecipitated and analyzed by SDS-PAGE followed by autoradiography. Adiponectin is synthesized and immunoprecipitated as a 30-kDa monomer. Bar graph represents adiponectin synthesis expressed as %control. B: adiponectin secreted into the medium. Rat adipocytes were treated as described above. After labeling with [³⁵S]met, cells were chased overnight with unlabeled medium, which was collected, immunoprecipitated, and analyzed. The results are the mean ± SE of 4 different experiments. *P < 0.05 vs. control adipocytes and treated adipocytes in the presence of GW-9662.

Fig. 4.

Adiponectin processing in presence of 2,2′-dipyridyl. A: rat adipocytes (150 μl) were pulse labeled for 15 min with 100 μCi/ml [³⁵S]met and chased with unlabeled methionine for the indicated time in presence or absence of 2,2′-dipyridyl (1 mM). Cells and medium were collected at each time point and immunoprecipitated and analyzed on a denaturing SDS-PAGE. Arrows indicate 3 discrete bands separating in both the cells and medium. The figure represents 1 of 3 similar experiments. B: effect of glycosylation inhibition on HMW adiponectin formation. Equal amount of rat adipocytes (150 μl) were treated with or without 3 μM PIO, 25 μM each of EPA or DHA in presence or absence of 2,2′-dipyridyl (1 mM), and cellular adiponectin was analyzed on a 4–20% nondenaturing and nonreducing gel and blotted with anti-adiponectin antibody, as described in methods. The figure represents 1 of 3 similar experiments.

Fig. 5.

Effect of PIO and ω-3 fatty acids on the synthesis and mRNA expression of adiponectin in 3T3-F442A cells. A: 3T3-F442A cells were treated with or without 3 μM PIO, 25 μM EPA, and 25 μM DHA ON and pulse labeled with 100 μCi/ml of [³⁵S]met for 15 min. Equal TCA-precipitable counts were immunoprecipitated with anti-adiponectin antibody and analyzed by SDS-PAGE followed by autoradiography. Figure represents 1 of 3 similar experiments. The results are means ± SE (P < 0.03). B: expression of adiponectin mRNA in 3T3-F442A cells following treatment with or without PIO, EPA, DHA, or palmitate expressed as a ratio of 18S RNA. The results are the mean ± SE of 4 different experiments. *P < 0.05 vs. control.

Fig. 6.

Effect of PIO and DHA on adiponectin translation in vitro. Cytoplasmic S-100 extracts were prepared from adipocytes treated with or without PIO or DHA, as described in methods. The specified extract was added to the in vitro translation assay containing [³⁵S]met and adiponectin mRNA transcript and analyzed by SDS-PAGE. The control cell extract was from cells that were not treated with PIO or ω-3 fatty acid. The autoradiograms were quantified by densitometry and expressed as a percent of blank, which represents autoradiographic image obtained with no cytoplasmic extract added to the in vitro translation system. A: extracts were prepared from 3T3-F442A adipocytes and added to the in vitro translation system containing the adiponectin mRNA. The results are the mean ± SE of 4 different experiments (P < 0.05). B: cytoplasmic S-100 extracts were prepared from rat adipocytes and added to the in vitro translation system containing the adiponectin mRNA. The results are the mean ± SE of 6 different experiments (P < 0.05). C: cytoplasmic S-100 extracts were prepared from rat adipocytes and were added to the in vitro translation assay containing an irrelevant (luciferase) mRNA transcript. *P < 0.05 vs. control cell extract.

Fig. 7.

Effect of PIO and ω-3 fatty acids on the distribution of adiponectin mRNA on the polysomes. Postmitochondrial supernatants from control and PIO- and DHA-treated adipocytes were fractionated on a 10–50% sucrose gradient. Fractions are numbered from the top of the gradient. The 260-nm absorption profile is represented by the line graph. Fractions 1–4 represent the monosomes, and fractions 7–17 contain the various sizes of polysomes. Fractions 4–17 contain the 18- and 28-s ribosomal subunits (data not shown). Adiponectin mRNA in each fraction was estimated using RT-PCR and expressed as %total adiponectin mRNA, as described in methods. The data shown are representative of 3 similar experiments.