Characterization of Metformin-Associated Metabolic Gene Expression Alterations in Type 2 Diabetes: Evidence from Patient Blood Samples and HepG2 Cells


Sari-Ak D., Celik A. B., Nur A. M., Helvaci-Kurt N., Kural A., Senoymak M. C., ...Daha Fazla

BRATISLAVA MEDICAL JOURNAL, 2026 (SCI-Expanded, Scopus) identifier identifier

Özet

BackgroundMetformin is the first-line treatment for type 2 diabetes mellitus (T2DM), although its complete biochemical mechanisms of action remain incompletely understood.AimThis study aimed to characterize metformin-associated metabolic gene expression alterations in patients with T2DM and in a human hepatocellular carcinoma cell line (HepG2) using RT-qPCR-based profiling and bioinformatic analyses.MethodsBlood samples from four groups (n = 20 per group)-healthy controls, prediabetic individuals, severe untreated T2DM patients, and metformin-treated T2DM patients-were analyzed. In addition, metformin-treated HepG2 cells were evaluated. A total of 46 metabolism-related genes were assessed by RT-qPCR. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway and gene-metabolite network analyses were performed.ResultsMetformin treatment was associated with altered expression of genes involved in glucose uptake (SLC2A2, HK2), lipid metabolism (ACLY, SCD), and energy balance. Pathway analysis indicated effects on glycolysis, the tricarboxylic acid (TCA) cycle, fatty acid biosynthesis, and redox processes. Gene expression changes were linked to predicted metabolite nodes, including NADP, palmitic acid, and ATP. Metformin-treated patients exhibited gene expression patterns associated with relatively improved metabolic profiles compared to untreated patients.ConclusionMetformin treatment is associated with coordinated changes in metabolic gene expression in T2DM. This integrative approach combining patient-derived data, an in vitro HepG2 model, and gene-metabolite networks provides a systems-level perspective on metformin-associated metabolic alterations. Findings should be interpreted considering model limitations and predicted metabolite associations.