Bioinformatic Determination of Potential Protein Targets of Metformin Activity in Hyperlipidemia

Abstract ID 128649 Poster Board 204 Type 2 diabetes mellitus affects hundreds of millions globally, primarily characterized by hyperglycemia. Dyslipidemia, prevalent in 50%-80% of diabetic patients, presents as elevated triglycerides and reduced high-density lipoprotein (HDL) cholesterol, exacerbating cardiovascular risks. Metformin, a first-line treatment for type 2 diabetes, effectively lowers glucose levels without hypoglycemia or weight gain, possibly by inhibiting gluconeogenesis and activating adenosine monophosphate kinase (AMPK). Beyond its glucose-lowering effects, metformin exhibits pleiotropic benefits, including potential anti-cancer and geroprotective properties, suggesting broader regulatory roles beyond its primary targets. Metformin's metabolic advantages extend to lipid profiles, with evidence indicating reductions in triglycerides and low-density lipoprotein (LDL) cholesterol, and increases in HDL cholesterol, contributing to its efficacy in managing type 2 diabetes. However, the underlying pharmacological targets of metformin in lipid-lowering are poorly understood. We used a network pharmacology approach to elucidate the potential protein targets of metformin that underlie its lipid-lowering activity. Target genes of metformin (208 genes) were compiled using SuperPred, SwissTargetPrediction, TargetNet, and DrugBank databases. The genes linked with hyperlipidemia (282 genes) were retrieved from the DisGeNET and GeneCards databases. The putative targets of metformin activity in hyperlipidemia were revealed by the examination of the overlap between genes identified as metformin targets and genes known to have a role in hyperlipidemia. Twenty-five such putative target genes were identified. Enrichment analysis performed using the REACTOME pathway database revealed that putative targets were enriched in pathways for nuclear receptor transcription, extra-nuclear estrogen signaling, and metabolism of lipids, among others. Protein-protein interaction network for the putative target genes were visualized using STRING (interaction score of > 0.4). Network topology and hub genes in the interaction network were determined using the cytoHubba plugin in the Cytoscape software. The top ten hub target genes based on maximal clique centrality (MCC) for metformin activity in hyperlipidemia were: HIF1A, SIRT1, EGFR, AR, PIK3CA, NFE2L2, ESR2, NOS3, PTGS1, and PLG. Follow-up studies are essential to validate the role of these proteins in the lipid-lowering activity of metformin. Overall, these findings provide insights into the potential mechanisms of metformin's beneficial effects on plasma lipids in patients with type 2 diabetes.