Dicarbonyl Protein Adduction: Plasminogen as a Target and Metformin as a Scavenging Therapeutic in Type 2 Diabetes

Abstract

Formation of advanced glycation endproducts (AGEs) on proteins has been linked to the pathogenesis of diabetic complications. Importantly, elevated levels of methylglyoxal (MG) occur in type 2 diabetes mellitus (T2DM), and the resulting site-specific formation of stable adducts on arginine residues can cause protein damage. Using MG, site-specific modifications on the plasma protein plasminogen (Pg) were determined following protein digestion into peptides and liquid chromatography tandem mass spectrometry (LC-MS/MS) analysis, and 30 arginine sites were identified on the protein. Investigation into three of the most highly modified sites, R504, R530, and R561, using molecular modeling, identified likely functional changes to the Pg cleavage and the lysine binding pocket as a result of adduct formation at these sites. Overall functional changes to Pg were examined using silver staining and kinetic assays to examine normal protein cleavage by activator enzymes streptokinase (STK), tissue plasminogen activator (tPA) and urokinase plasminogen activator (uPA). MG-modified Pg exhibited decreased activation into plasmin (Pn), which is the active enzyme that forms via normal Pg cleavage, by all three activator enzymes. Activation into Pn by STK was significantly delayed by MG modification on plasminogen. Similar effects were observed with tPA and uPA. Efforts to identify the primary sites of MG adduction on Pg by two dimensional gel electrophoresis (2DGE) identified six sites, including R504 and R530, as the earliest modified sites. In order to probe MG site specific modification effects on lysine binding, MG-modified protein was run through a lysine-sepharose binding column and fractions were collected. The results indicated that MG-modified Pg bound more weakly to the column and eluted easier than unmodified Pg and LC-MS/MS using a LTQ Orbitrap Velos of the fraction indicated that R504 and R530 were the primary sites of MG adduction within the eluate. To assess MG-modification of Pg in humans, 12 plasma samples were immunodepleted of the top 14 abundant proteins and samples were analyzed by LC-MS/MS using a LTQ Orbitrap Velos. Nine of the 12 patient samples indicated the presence of MG-modified peptides. The antihyperglycemic drug metformin, a drug that scavenges MG and lowers formation of AGEs, was studied in order to better elucidate this scavenging mechanism. A novel reaction imidazolinone product, IMZ, was determined to be the primary product formed in the reaction between metformin and MG, confirmed unequivocally through x-ray diffraction analysis. In order to determine levels of IMZ in human patients on metformin therapy, multiple reaction monitoring (MRM) was employed to quantify the compound. Human urine samples from 92 patients on metformin treatment were analyzed. 66 of the 68 patients to exhibit high concentrations of metformin also indicated the presence of IMZ in their urine. The remaining samples either exhibited no metformin, or levels of metformin too low to detect the presence of IMZ. Importantly, IMZ was never identified in patients without a metformin signal, indicating the validity and quality of the assay. This dissertation builds upon the current knowledge of site-specific MG modifications, both in vitro, identifying for the first time Pg as a sensitive site-specific target of glycation, with functional effects, and importantly in humans, as this is the first identification of MG-modified Pg in vivo. The functional effects associated with this modification may provide a link between elevated MG in T2DM, and resulting cardiovascular complications. Additionally, the identification of the novel reaction product IMZ is important, as it helps to fully elucidate the role metformin plays in treating T2DM patients. The detection of IMZ in the urine of human patients on metformin therapy indicates that metformin plays a role in the reducing MG levels through scavenging in vivo, and the developed MRM method allows for future rapid, sensitive study of cohorts to better understand this mechanism and the role it plays in reducing AGEs and diabetic complications.