Dietary fish oil enriched in very-long-chain polyunsaturated fatty acid reduces cardiometabolic risk factors and improves retinal function

Abstract

Very-long-chain polyunsaturated fatty acids (VLCPUFAs; C24-38) constitute a unique class of PUFA that have important biological roles, but the lack of a suitable dietary source has limited research in this field. We produced an n-3 C24-28-rich VLCPUFA-oil concentrated from fish oil to study its bioavailability and physiological functions in C57BL/6J mice. The serum and retinal C24:5 levels increased significantly compared to control after a single-dose gavage, and VLCPUFAs were incorporated into the liver, brain, and eyes after 8-week supplementation. Dietary VLCPUFAs resulted in favorable cardiometabolic changes, and improved electroretinography responses and visual performance. VLCPUFA supplementation changed the expression of genes involved in PPAR signaling pathways. Further in vitro studies demonstrated that the VLCPUFA-oil and chemically synthesized C24:5 are potent agonists for PPARs. The multiple potential beneficial effects of fish oil-derived VLCPUFAs on cardiometabolic risk and eye health in mice support future efforts to develop VLCPUFA-oil into a supplemental therapy.

Keywords: Cell biology; Dietary supplement; Physiology.

Graphical abstract

Figure 1

Dietary VLCPUFAs are bioavailable and incorporated into target tissues (A) Serum and retinal levels of 24:5 n-3, 24:6 n-3, 26:6 n-3, and 28:8 n-3 at 0, 2, 4, 8, and 24 h in C57BL/6J mice after a single-dose oral gavage of VLCPUFA-oil (250 mg/kg body weight). Data are represented as mean ± SEM (n = 5 per time point), ∗p < 0.05 vs. 0 h. (B) Changes in major n-3 PUFA/VLCPUFAs in plasma, liver, brain and eyes in C57BL/6J mice on chow diet supplemented with 1% or 5% (w/w) VLCPUFA-oil, or none (control) for 8 weeks. Decreased (p < 0.05), increased (p < 0.05), and unchanged fatty acids in 1% and 5% VLCPUFAs groups compared with control were colored by blue, red, and gray, respectively (n = 8). (C) Total VLCPUFAs composition (%) in PL, SL, GL, ST, and FFA in the plasma, liver, brain and eyes in C57BL/6J mice on chow diet supplemented with 1% or 5% (w/w) VLCPUFA-oil, or none (control) for 8 weeks. Data are represented as mean ± SEM (n = 8). ^a-cMeans within a column without a common superscript letter differ (p < 0.05). (D) Relative abundance (%) of PL, SL, GL, ST, and FFA in plasma, liver, and eyes in C57BL/6J mice on chow diet supplemented with 1% or 5% (w/w) VLCPUFA-oil, or none (control) for 8 weeks. Data are represented as mean ± SEM (n = 8). ^a-cMeans within a column without a common superscript letter differ (p < 0.05). Con: control; V1: 1% VLCPUFA-oil; V5: 5% VLCPUFA-oil; PL: phospholipids; SL: sphingolipids; GL: glycerolipids; ST: sterol lipids; FFA: free fatty acids. See also Figures S12, Table S3 and S4.

Figure 2

Dietary VLCPUFAs improve visual function (A) Amplitudes of photopic ERG a- and b-waves, and scotopic ERG a- and b-waves plotted versus flash intensity. Data are represented as mean ± SEM (n = 8). ∗p < 0.05, ∗∗p < 0.01 vs. control. (B) Photopic and scotopic visual acuities. Data are represented as mean ± SEM (n = 8). ∗∗p < 0.01 vs. control. (C) Photopic and scotopic contrast sensitivity. Averaged data were from both eyes of each animal. Data are represented as mean ± SEM (n = 8). ∗∗p < 0.01 vs. control. Mice received a daily oral gavage of VLCPUFA-oil (250 mg/kg body weight) or vehicle (control) for 15 consecutive days. See also Figure S13.

Figure 3

Dietary VLCPUFAs cause cardiometabolic changes in plasma and liver (A) Plasma glucose metabolism biomarkers: glucose, insulin, and HOMA-IR levels. Data are represented as mean ± SEM (n = 8). ^a-cMeans within a column without a common superscript letter differ (p < 0.05). (B) Representative plasma phospholipid and cholesterol FPLC profile from plasma samples pooled from 8 mice per group. (C) Plasma lipid metabolism biomarkers: triglycerides, phospholipid, and total cholesterol levels. Data are represented as mean ± SEM (n = 8). ^a-cMeans within a column without a common superscript letter differ (p < 0.05). (D) Representative images of Oil Red O staining of liver sections and enzymatic measurement of hepatic triglycerides, phospholipid, and cholesterol contents. Data are represented as mean ± SEM (n = 8). ^a-cMeans within a column without a common superscript letter differ (p < 0.05). Scale bar represents 100 μm. Nine-month-old C57BL/6J mice were fed a chow diet supplemented with 1% or 5% (w/w) VLCPUFA-oil, or none (control) for 8 weeks. Con: control; V1: 1% VLCPUFA-oil; V5: 5% VLCPUFA-oil; HOMA-IR: homeostatic model assessment for insulin resistance; FPLC: fast protein liquid chromatography. See also Figures S3, S4, S14, and Table S2.

Figure 4

Dietary VLCPUFAs alter hepatic transcriptional profile and activates PPARs (A) 3D-PCA plot showing samples clustering by control (white dots), 1% VLCPUFAs (gray dots), and 5% VLCPUFAs diet (black dots) (n = 4). (B) Differential gene expression summarized in a volcano plot with gene names highlighted for top genes by significance or magnitude. (C) Heatmap of expression trends of significant differentially expressed genes. (D) Expression changes among genes involved in glucose and lipid metabolism. ∗p < 0.05 between the control and VLCPUFA groups. (E) Pathway enrichment analysis of VLCPUFA-related differential expression. Pathway impact represents percentage of KEGG pathway overlapping over- or under-expressing genes colored by red and blue bubbles, respectively. Nine-month-old C57BL/6J mice were fed a chow diet supplemented with 1% or 5% (w/w) VLCPUFA-oil, or none (control) for 8 weeks. (F) Fold activation of PPARα and PPARγ by different dosages of VLCPUFA-oil in CHO cells that have a luciferase reporter gene functionally linked to a PPARα or PPARγ-response element. Data are represented as mean ± SEM (n = 3). ∗∗p < 0.01, ∗∗∗p < 0.001 vs. vehicle control ethanol (set to 1.0). (G) Structure of 24:5 n-3. (H) Fold activation of PPARα and PPARγ by chemically synthesized 24:5 n-3 compared with EPA and DHA in CHO cells. Data are represented as mean ± SEM (n = 3). ^a-cMeans within a column without a common superscript letter differ (p < 0.05). PCA: principal component analysis; Con: control; V1: 1% VLCPUFA-oil; V5: 5% VLCPUFA-oil; PPAR: Peroxisome proliferator-activated receptor. See also Figures S2 and S15.