The Novel Lipid Emulsion Vegaven Is Well Tolerated and Elicits Distinct Biological Actions Compared With a Mixed-Oil Lipid Emulsion Containing Fish Oil: A Parenteral Nutrition Trial in Piglets

Abstract

Background: Vegaven is a novel lipid emulsion for parenteral nutrition (PN) based on 18-carbon n-3 (ω-3) fatty acids, which elicits liver protection via interleukin-10 (IL-10) in the murine model of PN.

Objectives: In a preclinical model of PN in neonatal piglets, Vegaven was tested for efficacy and safety and compared with a mixed-oil lipid emulsion containing fish oil (SMOFlipid).

Methods: Male piglets 4-5 d old were randomly allocated to isocaloric isonitrogenous PN for 14 d, which varied only by the type of lipid emulsion (Vegaven, n = 8; SMOFlipid, n = 8). Hepatic IL-10 tissue concentration served as primary outcome. Secondary outcomes were organ weights, bile flow, blood analyses, plasma insulin and glucagon concentrations, insulin signaling, proinflammatory cytokines, tissue lipopolysaccharide concentrations, and fatty acid composition of phospholipid fractions in plasma, liver, and brain.

Results: Total weight gain on trial, organ weights, and bile flow were similar between the Vegaven and the SMOFlipid group. Vegaven elicited higher hepatic IL10 (Δ = 148 pg/mg protein; P < 0.001) and insulin receptor substrate-2 amounts (Δ = 0.08 OD; P = 0.012). Plasma insulin concentrations (Δ = 1.46 mU/L; P = 0.003) and fructosamine (glycated albumin, Δ = 12.4 μmol/g protein; P = 0.003) were increased in SMOFlipid as compared with those of Vegaven group, indicating insulin resistance. Higher hepatic injury markers were observed more frequently in the SMOFlipid group than those in the Vegaven group. Lipopolysaccharide, tumor necrosis factor-α, and IL-6 concentrations increased in pancreatic and brain tissues of SMOFlipid-treated compared with those in the Vegaven-treated piglets. Insulin signaling reduced in the brains of SMOFlipid-treated piglets. Vegaven and SMOFlipid elicited distinct fatty acid profiles in the phospholipid fractions of the rapidly growing brains but showed similar accretion of docosahexaenoic acid and arachidonic acid after 2 wk of PN.

Conclusions: Vegaven is well tolerated in this piglet model of PN, demonstrating distinct biological actions compared with SMOFlipid, namely lower liver, pancreas, and brain inflammation, enhanced insulin signaling, and improved whole body glucose control.

Keywords: 18-carbon n–3 fatty acids; hyperinsulinemia; insulin resistance; interleukin-10; lipid mediators; liver; mixed-oil lipid emulsions; neonatal piglet; neuroinflammation; pancreas; parenteral nutrition; stearidonic acid; α-linolenic acid.

FIGURE 1

Study flow diagram. Secondary outcomes included growth, bile flow, metabolism and insulin signaling, fatty acid profiles, liver lipid mediator profile, and safety data. JVC, jugular venous catheter; SMOF, piglets treated with SMOFlipid-based PN for 14 d; VEGA, piglets treated with Vegaven-based PN for 14 d.

FIGURE 2

Cytokine profile in liver. (A) Interleukin-10 (IL10). (B) Interleukin-6 (IL6). (C) Interleukin-6 to interleukin-10 ratio. (D) Tumor necrosis factor-α (TNFα). (E) Interferon-γ (IFNγ).. ∗Significantly different. Bars represent means ± SDs. Dots indicate individual experiments. n = 8 per group. SMOF, piglets treated with SMOFlipid-based PN for 14 d; VEGA, piglets treated with Vegaven-based PN for 14 d.

FIGURE 3

Hepatic injury markers. (A) γ-glutamyl transferase (GGT). (B) Alkaline phosphatase (ALP). (C) Aspartate aminotransferase (AST). (D) Alanine transaminase (ALT). (E) Bilirubin (total). (F) Bile acids. Bars represent means ± SDs. Dots indicate individual experiments. n = 8 per group. There were no differences in the individual injury markers. However, when counting the 3 highest measurements (red dots) for each of the 6 liver parameters (“liver panel,” composite liver injury), 14 counts were observed in SMOFlipid-treated piglets but only 4 counts in Vegaven-treated piglets (SMOF 14/18 vs. VEGA 4/18, P = 0.002), suggesting a lower potential for liver injury in Vegaven-treated piglets. SMOF, piglets treated with SMOFlipid-based PN for 14 d; VEGA, piglets treated with Vegaven-based PN for 14 d.

FIGURE 4

Blood glucose control and liver glycogen. (A) Insulin plasma concentration. (B) Glucagon plasma concentration. (C) Fructosamine (glycated albumin) plasma load. (D) Liver glycogen content. ∗Significantly different. Bars represent means ± SDs. Dots indicate individual experiments. n = 7–8 per group. One SMOFlipid sample was excluded from glycogen analysis because of prolonged PN line occlusion overnight. SMOF, piglets treated with SMOFlipid-based PN for 14 d; VEGA, piglets treated with Vegaven-based PN for 14 d.

FIGURE 5

Hepatic insulin signaling. (A) Insulin receptor β-subunit abundance (IRβ). (B) Insulin receptor substrate (IRS) 2 abundance. (C) Tyrosine phosphorylation of insulin receptor substrate (pY-IRS) 2. (D) Ratio of tyrosine phosphorylated insulin receptor substrate-2 to total IRS-2 (pY-IRS2/IRS2). ∗Significantly different. Bars represent means ± SDs. Dots indicate individual experiments. n = 7–8 per group. One SMOFlipid sample was excluded from IRS2 and pY-IRS2 analyses because of prolonged PN line occlusion overnight. OD, optical density; SMOF, piglets treated with SMOFlipid-based PN for 14 d; VEGA, piglets treated with Vegaven-based PN for 14 d.

FIGURE 6

Pancreatic injury. (A) LPS (endotoxin) tissue accumulation. Note logarithmic scale of x-axis. (B) Interleukin-10 (IL10). (C) Interleukin-6 (IL6). (D) Interleukin-6 to interleukin-10 ratio. (E) Tumor necrosis factor-α (TNFα). (F) Interleukin-1β (IL1β). ∗Significantly different. Bars represent means ± SDs. Dots indicate individual experiments. n = 8 per group. SMOF, piglets treated with SMOFlipid-based PN for 14 d; VEGA, piglets treated with Vegaven-based PN for 14 d.

FIGURE 7

Brain LPS concentrations, neuroinflammation, and insulin signaling. (A) LPS (endotoxin) tissue concentrations (logarithmic scale). (B) TNF-α. (C) IL-6. (D) Insulin receptor β-subunit (IRβ). (E) Insulin receptor substrate (IRS)-1. (F) IRS-2. (G) Tyrosine phosphorylated insulin receptor substrate-2. (H) Tyrosine phosphorylated insulin receptor substrate-2 to total insulin receptor substrate-2 (pY-IRS2/IRS2).. Note that tyrosine phosphorylations of IRβ and IRS-1 were below detection limit. ∗Significantly different. Bars represent means ± SDs. Dots indicate individual experiments. n = 7–8 per group. One Vegaven sample was excluded from LPS analysis because it was below detection limit. SMOF, piglets treated with SMOFlipid-based PN for 14 d; VEGA, piglets treated with Vegaven-based PN for 14 d.