Glimepiride reduces mononuclear activation of the redox-sensitive transcription factor nuclear factor-kappa B

Aims: Glimepiride has the lowest ratio of insulin release to glucose decrease compared with other sulphonylureas. This prompted us to study in vitro and in vivo in a placebo-controlled study the effect of glimepiride on the redox-sensitive transcription factor nuclear factor-kappa B (NF-κB). Methods: Fifteen patients with type 2 diabetes on glibenclamide with a stable HbA1c over the last 6 months were included. After sampling for determination of baseline values, 10 patients were changed to an equivalent dose of glimepiride, while the placebo group was maintained at glibenclamide plus placebo. The glimepiride dose in these patients was adjusted so that no change in glucose control occurred, allowing for direct comparison. The others were kept on glibenclamide and received additional placebo. After 4 weeks of glimepiride or glibenclamide plus placebo, a second blood sample was taken. Mononuclear cells were isolated and assayed in a tissue-culture-independent electrophoretic mobility shift assay (EMSA)-based detection system for NF-κB binding activity, and by Western Blot for nuclear localization of NF-κB-p65, the cytoplasmic content of IκBα and the NF-κB-controlled haemoxygenase-1. Glimepiride dose-dependent inhibition of carboxymethyllysin (CML) albumin or tumour necrosis factor α (TNFα)- and H₂O₂-induced activation of NF-κB binding were determined, using isolated peripheral blood mononuclear cells from healthy volunteers, and transcriptional activity of bovine aortic endothelial cells either left untreated or induced with CML albumin incubated with or without glimepiride. Furthermore, in-vitro studies were implemented to demonstrate radical quenching properties of glimepiride in the cell-free 2,2′-azo-bis(2-aminopropane -dihydrochloride system. Results: Baseline glucose and HbA1c remained stable in the patients switched from glibenclamide to a corresponding dose of glimepiride or kept on glibenclamide plus placebo. While in the group of patients only taking glibenclamide plus placebo the NF-κB binding activity did not change significantly (p = 0.58), the NF-κB binding activity in the group of patients taking glimepiride was reduced from 19.3 relative NF-κB-p65-equivalents to 15.5 relative NF-κB-p65-equivalents (p = 0.04). The nuclear translocation of NF-κB-p65 was reduced from 100% at baseline to 58% after 4 weeks (p = 0.04); the cytoplasmic localization of NF-κB-p65 increased from 100% to 129% (p = 0.03) and the cytoplasmic content of IκBα increased from 100% to 109% (p = 0.06). The redox-sensitive haemoxygenase-1 antigen was reduced from 100% to 82% (p = 0.04). To prove directly that glimepiride reduces NF-κB activation, we isolated peripheral blood mononuclear cells (PBMC) from healthy volunteers. In vitro, glimepiride reduced TNFα-(1 nmol/l) and CML albumin (800 nmol/l)-induced NF-κB activation dose dependently, being half maximal at 120 μmol/l. H₂O₂-mediated NF-κB activation was only partially reduced. In addition, glimepiride reduced NF-κB-dependent gene expression using a NF-κB-driven luciferase reporter system. Finally, a cell-free detection system showed that glimepiride has radical quenching properties. Conclusion: Glimepiride can affect the activation of the redox-sensitive transcription factor NF-κB in vitro and in vivo.