On). The observed production, m/z 269, generated by CID of the parent ion [M-CH2C6F5]-, is consistent with fragmentation of Title Loaded From File 15d-PGJ3 ([MCH2C6F5-CO2]-) (Fig. 6B). To further substantiate the formation 15d-PGJ3 from EPA in 3T3-L1, cells were incubated with d5-EPA. As described above, products were separated by HPLC and the HPLC fraction containing the putative d5-15d-PGJ3 was collected and then analyzed by GC-MS/MS. The predicted molecular ionEPA-Derived Prostaglandin and AdiponectinFigure 4. GC-MS (A) and GC-MS/MS (B) spectra of 15d-PGJ3 generated during incubation of PGD3 in phosphate-buffered saline. The anion at m/z 313 [M-CH2C6F5]- generated by NICI was subjected to CID. The structure of the derivative and the proposed structure of the product ion produced by CID of m/z 313 are shown in the figure. doi:10.1371/journal.pone.0063997.gfor [M-CH2C6F5]- of d5-15d-PGJ3 is m/z 318. CID of m/z 318 generates the product ion at m/z 274 ([M-CH2C6F5-CO2]-) (Fig 7).ion [M-CH2C6F5]-, is consistent with fragmentation of 15d-PGJ3 ([M-CH2C6F5-CO2]-) (Fig. 8B).Formation of 15d-PGJ3 in vivoWe then undertook experiments to determine whether 15dPGJ3 is formed in adipose tissue in vivo. Because baseline levels of EPA in animals are very low, we determined if 15d-PGJ3 is generated in vivo by examining this compound in mice fed with the EPA-enriched diet previously mentioned. A representative GCMS/MS ion current chromatogram obtained from these analyses is shown in Fig. 8. The chromatogram based on m/z 313 detection represents endogenous 15d-PGJ3 (Fig. 8A). The observed production, m/z 269, generated by CID of the parentPPAR-c Antagonism Decreases PGD3- or 15d-PGJ3induced Increase in AdiponectinNext, we determined whether PGD3 or 15d-PGJ3 increased adiponectin via a PPAR-c-dependent mechanism. 3T3-L1 adipocytes were incubated with a PPAR-c antagonist, GW9662, alone or in combination with PGD3 or 15d-PGJ3. GW9662 alone, which has been previously reported to induce irreversible loss of ligand binding activity [37], had no effect on adiponectin secretion. As shown in Fig. 9, GW9662 significantly attenuated the increase of adiponectin level in cell medium observed in response to PGD3 or 15d-PGJ3. However, the adiponectin concentration was stillEPA-Derived Prostaglandin and AdiponectinFigure 5. Effects of PGD3 and 15d-PGJ3 on adiponectin secretion by 3T3-L1 adipocytes. Cells were incubated for 2 h with 1 mM PGD3 or 0.1 mM 15d-PGJ3. Adiponectin in the medium was determined by ELISA. Results are means 6 sem (n = 4 in triplicate), expressed as percentage of the control. Statistical significance is represented as **P,0.01 vs control. doi:10.1371/journal.pone.0063997.gsignificantly increased when GW9662 was added with PGD3 or 15d-PGJ3 compared with control condition, suggesting that part of the prostaglandin effect was not attributable to PPAR-c stimulation.15d-PGJ3 Mediates Induction of PPAR-c Title Loaded From File Target Gene Expression in 3T3-L1 AdipocytesTo go further in the demonstration of the involvement of PPAR-c, the effects of 15d-PGJ3 on the expression of characterized PPAR-c target genes, FAS, FABP4 and adiponectin were examined. 3T3-L1 cells were incubated with or without 100 nM of 15d-PGJ3 for 2 h. As depicted in Fig. 10, 15d-PGJ3 induced increase in mRNA level of all PPAR-c target genes tested with a significant increase in the level of FABP4 and FAS mRNA. However, treatment with 15d-PGJ3 had no effect on PDK4 mRNA transcript abundance.DiscussionThis study demonstrates that.On). The observed production, m/z 269, generated by CID of the parent ion [M-CH2C6F5]-, is consistent with fragmentation of 15d-PGJ3 ([MCH2C6F5-CO2]-) (Fig. 6B). To further substantiate the formation 15d-PGJ3 from EPA in 3T3-L1, cells were incubated with d5-EPA. As described above, products were separated by HPLC and the HPLC fraction containing the putative d5-15d-PGJ3 was collected and then analyzed by GC-MS/MS. The predicted molecular ionEPA-Derived Prostaglandin and AdiponectinFigure 4. GC-MS (A) and GC-MS/MS (B) spectra of 15d-PGJ3 generated during incubation of PGD3 in phosphate-buffered saline. The anion at m/z 313 [M-CH2C6F5]- generated by NICI was subjected to CID. The structure of the derivative and the proposed structure of the product ion produced by CID of m/z 313 are shown in the figure. doi:10.1371/journal.pone.0063997.gfor [M-CH2C6F5]- of d5-15d-PGJ3 is m/z 318. CID of m/z 318 generates the product ion at m/z 274 ([M-CH2C6F5-CO2]-) (Fig 7).ion [M-CH2C6F5]-, is consistent with fragmentation of 15d-PGJ3 ([M-CH2C6F5-CO2]-) (Fig. 8B).Formation of 15d-PGJ3 in vivoWe then undertook experiments to determine whether 15dPGJ3 is formed in adipose tissue in vivo. Because baseline levels of EPA in animals are very low, we determined if 15d-PGJ3 is generated in vivo by examining this compound in mice fed with the EPA-enriched diet previously mentioned. A representative GCMS/MS ion current chromatogram obtained from these analyses is shown in Fig. 8. The chromatogram based on m/z 313 detection represents endogenous 15d-PGJ3 (Fig. 8A). The observed production, m/z 269, generated by CID of the parentPPAR-c Antagonism Decreases PGD3- or 15d-PGJ3induced Increase in AdiponectinNext, we determined whether PGD3 or 15d-PGJ3 increased adiponectin via a PPAR-c-dependent mechanism. 3T3-L1 adipocytes were incubated with a PPAR-c antagonist, GW9662, alone or in combination with PGD3 or 15d-PGJ3. GW9662 alone, which has been previously reported to induce irreversible loss of ligand binding activity [37], had no effect on adiponectin secretion. As shown in Fig. 9, GW9662 significantly attenuated the increase of adiponectin level in cell medium observed in response to PGD3 or 15d-PGJ3. However, the adiponectin concentration was stillEPA-Derived Prostaglandin and AdiponectinFigure 5. Effects of PGD3 and 15d-PGJ3 on adiponectin secretion by 3T3-L1 adipocytes. Cells were incubated for 2 h with 1 mM PGD3 or 0.1 mM 15d-PGJ3. Adiponectin in the medium was determined by ELISA. Results are means 6 sem (n = 4 in triplicate), expressed as percentage of the control. Statistical significance is represented as **P,0.01 vs control. doi:10.1371/journal.pone.0063997.gsignificantly increased when GW9662 was added with PGD3 or 15d-PGJ3 compared with control condition, suggesting that part of the prostaglandin effect was not attributable to PPAR-c stimulation.15d-PGJ3 Mediates Induction of PPAR-c Target Gene Expression in 3T3-L1 AdipocytesTo go further in the demonstration of the involvement of PPAR-c, the effects of 15d-PGJ3 on the expression of characterized PPAR-c target genes, FAS, FABP4 and adiponectin were examined. 3T3-L1 cells were incubated with or without 100 nM of 15d-PGJ3 for 2 h. As depicted in Fig. 10, 15d-PGJ3 induced increase in mRNA level of all PPAR-c target genes tested with a significant increase in the level of FABP4 and FAS mRNA. However, treatment with 15d-PGJ3 had no effect on PDK4 mRNA transcript abundance.DiscussionThis study demonstrates that.