Nduced a significant increase in adiponectin secretion.Synthesis of T 4uC with 5 nonfat milk in Tris-buffered saline (25 mM Tris, 137 mM 15d-PGJ3 from PGD3 in Phosphate-buffered SalinePreviously, it has been shown that PGD2, a prostaglandin derived from 10781694 AA, is converted sequentially to J2 prostaglandins in vitro [14, 15]. Shibata et al [15] showed that PGD2 is initially converted to the dehydration products 15d-PGD2 and PGJ2, the latter being converted to 15d-PGJ2. A recent study showed that J3 prostanoids are also formed in vitro from PGD3 [35]. PGD2 or PGD3 was incubated in PBS at 37uC for 72h and the products were analyzed by HPLC. By comparison with PGs of the 2-series (formed from PGD2, Title Loaded From File Figure 3A) or commercial (not shown), peaks I, II and III were assumed to be PGJ3, 15d-PGD3 and 15d-PGJ3 (Fig. 3B, peaks I, II and III, respectively). To further substantiate the structural identity of peak III as 15d-PGJ3, the HPLC fraction containing this peak was collected and analyzed by GC-MS and GC-MS/MS. The predicted unique [M-CH2C6F5]- ion for the pentafluorobenzyl ester derivative 15d-PGJ3 is m/z 313. The GC/ NICI/MS analysis of the product chromatogram only showed a m/z 313 peak compatible with the formation of 15d-PGJ3 (Fig. 4A). This peak was analyzed by CID and the CID spectrum is shown in figure 4B. A predicted fragment, m/z 269 [M-CH2C6F5-CO2]-, was obtained.Figure 2. (A) Plasma levels 16985061 of adiponectin from mice fed a standard diet or an EPA-rich diet. Plasma levels are expressed in mg/mL. Results are means 6 sem (n = 8). *P,0.05 as compared to the control group. (B) Body weight gain (g) of mice fed a standard diet or an EPA-rich diet. Mice were killed on days 0 and 4. Results are means 6 sem (n = 8). *P,0.05 as compared to the control group. (C) Effects of eicosapentaenoic acid on adiponectin secretion by 3T3-L1 adipocytes. Cells were incubated for 2 h (gray) or 4 h (black)EPA-Derived Prostaglandin and AdiponectinFigure 3. HPLC profile of metabolites formed from PGD2 (A) or PGD3 (B). 1 mM of PGD2 or PGD3 was incubated in phosphate-buffered saline at 37uC for 24 h. PGD2/PGD3 and their metabolites were chromatographed on a Waters Xbridge C18 column (4.66250 mm, 3.5 mm) at a flow rate of 1 ml/min starting at 100 solvent A (acetonitrile/water acidified to pH 3, 2/8 v/v) to 100 solvent B (acetonitrile) from 1 to 30 min. The elution profiles were monitored by UV absorbance at 195 nm. doi:10.1371/journal.pone.0063997.gExposure of 3T3-L1 Adipocytes to PGD3 and 15d-PGJ3 Leads to Increased Adiponectin LevelsWe then examined whether PGD3 and 15d-PGJ3 could increase adiponectin secretion by 3T3-L1 adipocytes. Exposure of cells to 1 mM PGD3 and 100 nM 15d-PGJ3 resulted in increased adiponectin levels in 3T3-L1 adipocytes medium by 55 and 28 , respectively compared to control cells (Fig. 5). These concentrations were chosen since the increase in adiponectin secretion was observed from 1 mM EPA. We also observed an increased adiponectin secretion (+25 ) after incubation of cells for 24 h with 100 nM 15d-PGJ3 compared to control cells (not shown).Formation of 15d-PGJ3 in Cell Medium after Incubation of Cells with EPAWe then sought to determine whether 15d-PGJ3 could be detected in the culture medium of cells incubated with EPA. 3T3L1 were incubated with 10 mM EPA. Products were separated by HPLC; the HPLC fraction supposed to contain 15d-PGJ3 was collected and 15d-PGJ3 was analyzed by GC-MS and GC-MS/ MS. As shown in Fig. 6A, a significant amount of 15d-PGJ3 was detected in the culture medium (nanomolar concentrati.Nduced a significant increase in adiponectin secretion.Synthesis of 15d-PGJ3 from PGD3 in Phosphate-buffered SalinePreviously, it has been shown that PGD2, a prostaglandin derived from 10781694 AA, is converted sequentially to J2 prostaglandins in vitro [14, 15]. Shibata et al [15] showed that PGD2 is initially converted to the dehydration products 15d-PGD2 and PGJ2, the latter being converted to 15d-PGJ2. A recent study showed that J3 prostanoids are also formed in vitro from PGD3 [35]. PGD2 or PGD3 was incubated in PBS at 37uC for 72h and the products were analyzed by HPLC. By comparison with PGs of the 2-series (formed from PGD2, Figure 3A) or commercial (not shown), peaks I, II and III were assumed to be PGJ3, 15d-PGD3 and 15d-PGJ3 (Fig. 3B, peaks I, II and III, respectively). To further substantiate the structural identity of peak III as 15d-PGJ3, the HPLC fraction containing this peak was collected and analyzed by GC-MS and GC-MS/MS. The predicted unique [M-CH2C6F5]- ion for the pentafluorobenzyl ester derivative 15d-PGJ3 is m/z 313. The GC/ NICI/MS analysis of the product chromatogram only showed a m/z 313 peak compatible with the formation of 15d-PGJ3 (Fig. 4A). This peak was analyzed by CID and the CID spectrum is shown in figure 4B. A predicted fragment, m/z 269 [M-CH2C6F5-CO2]-, was obtained.Figure 2. (A) Plasma levels 16985061 of adiponectin from mice fed a standard diet or an EPA-rich diet. Plasma levels are expressed in mg/mL. Results are means 6 sem (n = 8). *P,0.05 as compared to the control group. (B) Body weight gain (g) of mice fed a standard diet or an EPA-rich diet. Mice were killed on days 0 and 4. Results are means 6 sem (n = 8). *P,0.05 as compared to the control group. (C) Effects of eicosapentaenoic acid on adiponectin secretion by 3T3-L1 adipocytes. Cells were incubated for 2 h (gray) or 4 h (black)EPA-Derived Prostaglandin and AdiponectinFigure 3. HPLC profile of metabolites formed from PGD2 (A) or PGD3 (B). 1 mM of PGD2 or PGD3 was incubated in phosphate-buffered saline at 37uC for 24 h. PGD2/PGD3 and their metabolites were chromatographed on a Waters Xbridge C18 column (4.66250 mm, 3.5 mm) at a flow rate of 1 ml/min starting at 100 solvent A (acetonitrile/water acidified to pH 3, 2/8 v/v) to 100 solvent B (acetonitrile) from 1 to 30 min. The elution profiles were monitored by UV absorbance at 195 nm. doi:10.1371/journal.pone.0063997.gExposure of 3T3-L1 Adipocytes to PGD3 and 15d-PGJ3 Leads to Increased Adiponectin LevelsWe then examined whether PGD3 and 15d-PGJ3 could increase adiponectin secretion by 3T3-L1 adipocytes. Exposure of cells to 1 mM PGD3 and 100 nM 15d-PGJ3 resulted in increased adiponectin levels in 3T3-L1 adipocytes medium by 55 and 28 , respectively compared to control cells (Fig. 5). These concentrations were chosen since the increase in adiponectin secretion was observed from 1 mM EPA. We also observed an increased adiponectin secretion (+25 ) after incubation of cells for 24 h with 100 nM 15d-PGJ3 compared to control cells (not shown).Formation of 15d-PGJ3 in Cell Medium after Incubation of Cells with EPAWe then sought to determine whether 15d-PGJ3 could be detected in the culture medium of cells incubated with EPA. 3T3L1 were incubated with 10 mM EPA. Products were separated by HPLC; the HPLC fraction supposed to contain 15d-PGJ3 was collected and 15d-PGJ3 was analyzed by GC-MS and GC-MS/ MS. As shown in Fig. 6A, a significant amount of 15d-PGJ3 was detected in the culture medium (nanomolar concentrati.