Modulation of the endogenous omega-3 fatty acid and oxylipin profile in vivo-A comparison of the fat-1 transgenic mouse with C57BL/6 wildtype mice on an omega-3 fatty acid enriched diet

Abstract

Dietary intervention and genetic fat-1 mice are two models for the investigation of effects associated with omega-3 polyunsaturated fatty acids (n3-PUFA). In order to assess their power to modulate the fatty acid and oxylipin pattern, we thoroughly compared fat-1 and wild-type C57BL/6 mice on a sunflower oil diet with wild-type mice on the same diet enriched with 1% EPA and 1% DHA for 0, 7, 14, 30 and 45 days. Feeding led after 14-30 days to a high steady state of n3-PUFA in all tissues at the expense of n6-PUFAs. Levels of n3-PUFA achieved by feeding were higher compared to fat-1 mice, particularly for EPA (max. 1.7% in whole blood of fat-1 vs. 7.8% following feeding). Changes in PUFAs were reflected in most oxylipins in plasma, brain and colon: Compared to wild-type mice on a standard diet, arachidonic acid metabolites were overall decreased while EPA and DHA oxylipins increased with feeding more than in fat-1 mice. In plasma of n3-PUFA fed animals, EPA and DHA metabolites from the lipoxygenase and cytochrome P450 pathways dominated over ARA derived counterparts.Fat-1 mice show n3-PUFA level which can be reached by dietary interventions, supporting the applicability of this model in n3-PUFA research. However, for specific questions, e.g. the role of EPA derived mediators or concentration dependent effects of (individual) PUFA, feeding studies are necessary.

Fig 1. Fatty acid profile in blood.

Shown are relative amounts of EPA, DHA and ARA as well as the relative distribution of n3- and n6-PUFA, MUFA and SFA in transgenic fat-1 mice and wild type animals (WT-STD) on a sunflower oil based diet, as well as in wild type mice on the same diet enriched with EPA and DHA (WT-STD+n3) during the course of the feeding period (45 days) in (A) whole blood, (B) plasma and (C) blood cells. Analytes that were below the limit of quantification are marked with a white filling in the diagram. Results of the statistical analyses for selected fatty acids during the course of the feeding with the n3-PUFA enriched diet as well as for WT and fat-1 mice after 30 days on the experimental diets are shown in S6+S7 Tables.

Fig 2. n6/n3 ratio in blood as well as tissues and %EPA+DHA in blood cells.

Shown is the n6/n3 ratio in blood (A) and in tissues (B), as well as %EPA+DHA in blood cells (C) in transgenic fat-1 mice and wild type animals (WT-STD) on a standard sunflower oil based diet, as well as in wild type mice on the same diet enriched with EPA and DHA (WT-STD+n3) after 30 days of feeding. The n6/n3 ratio was calculated as Ʃ%(C18:2 n6, C18:3 n6, C20:3 n6, C20:4 n6, C22:4n6)/ Ʃ%(C18:3 n3, C20:5 n3, C22:5 n3, C22:6 n3). Statistical differences were determined using one-way ANOVA followed by Tukey’s post test (*** p<0.001, **** p<0.0001).

Fig 3. Concentrations of oxylipins.

Presented are concentrations of selected prostanoids, 5-LOX, 12-LOX, 15-LOX, CYP4 and CYP2 products of ARA, EPA and DHA as well as 18-HEPE in (A) plasma, (B) brain and (C) colon in transgenic fat-1 mice and wild type animals (WT-STD) on a sunflower oil based diet, as well as in wild type mice on the same diet enriched with EPA and DHA (WT-STD+n3) after 30 days of feeding. The lower limit of quantification (LLOQ) for the analyte is indicated in case it was not exceeded in >50% of the samples per group. Results of the statistical analyses for the comparison of oxylipins between the feeding groups after 30 days on the experimental diets are shown in S8 Table.

Fig 4. Correlation between plasma n3-PUFA/n6-PUFA oxylipins and the ratio of their precursor fatty acids.

The ratio of CYP metabolites (sum of epoxy-FA and dihydroxy-FA) from EPA (A) and DHA (B) to the respective ARA metabolites are plotted against the ratio of their precursor PUFA. In panel C-D the same correlation is shown for selected 12-LOX metabolites (12-HETE and 14-HDHA). The slope of the linear regression and the correlation coefficient were calculated based on all feeding groups of the experiment.