Multi-omics profiling and cellular validation reveal distinct lipid networks of krill oil and its predicted mechanisms against insulin resistance

Insulin resistance is a major contributor to metabolic disorders. We compared krill oil (KO), fish oil (FO), and soybean lecithin (SL) using untargeted metabolomics (UHPLC-HRMS), untargeted lipidomics (UPLC-QTOF-MS), network pharmacology, molecular docking, and a TNF-alpha-induced 3T3-L1 adipocyte model. Metabolomics detected 3,314 annotated features. In the KO versus FO comparison, 378 features were up-regulated and 593 were down-regulated in positive-ion mode, whereas 365 were up-regulated and 925 were down-regulated in negative-ion mode. Lipidomics detected 4,667 lipid-associated features and revealed distinct product-specific patterns of phospholipids, free fatty acids, glycerolipids, and sphingolipids. Network pharmacology identified 394 intersecting candidate targets, with AKT1, IL6, PPARG, and SRC prioritized as high-centrality nodes. Molecular docking was used to prioritize predicted component-target interactions. In mature 3T3-L1 adipocytes, TNF-alpha reduced insulin-stimulated 2-NBDG uptake. KO, FO, and SL increased glucose uptake relative to the TNF-alpha plus insulin group, with KO showing the largest response among the tested oils. KO also increased the p-AKT/AKT ratio and altered the p-IRS1/IRS1 ratio. These findings support compositional differences among the tested oils and cell-level modulation of insulin signaling by KO. They do not establish in vivo efficacy, direct target engagement, or an effect attributable solely to phospholipid form because EPA/DHA exposure was not dose-matched.