Increased serum levels of methylglyoxal-derived hydroimidazolone-AGE are associated with increased cardiovascular disease mortality in nondiabetic women

Abstract

Objective

To investigate the association of the levels of methylglyoxal-derived hydroimidazolone AGE modified proteins (MG-H1-AGE) with cardiovascular disease (CVD) mortality in an 18-year follow-up study in Finnish nondiabetic and diabetic subjects.

Methods

The study design was a nested case-control study. Serum MG-H1-AGE levels in samples drawn at baseline were measured with a DELFIA type immunoassay in 220 diabetic subjects and 61 nondiabetic subjects who died from CVD during the follow-up, and age- and gender-matched 157 diabetic subjects and 159 nondiabetic subjects who did not die from CVD.

Results

In type 2 diabetic subjects serum MG-H1-AGE levels were similar in subjects who died from CVD and in subjects who did not, 32.6 (24.6–42.1) (median (interquartile range)) vs. 31.3 (22.5–40.7) U/mL (p = 0.281). In nondiabetic subjects serum MG-H1 levels were significantly higher in subjects who died from CVD than in subjects who were alive, 35.4 (28.1–44.7) vs. 31.3 (24.2–38.6) U/mL (p = 0.025). Corresponding MG-H1 levels were 41.2 (35.6–58.7) vs. 31.1 (26.7–35.7) U/mL, p = 0.003, in women, and 34.4 (26.3–41.2) vs. 32.0 (22.8–40.3) U/mL, p = 0.270, in men. Multivariate logistic regression analysis showed a significant association of serum levels of MG-H1-AGE with CVD mortality in nondiabetic women (adjusted p = 0.021), but not in nondiabetic men.

Conclusions

Our 18-year follow-up study shows that high baseline serum levels of MG-H1 type of AGE modified proteins were associated with CVD mortality in nondiabetic women, but not in nondiabetic men or in diabetic subjects.

Introduction

Advanced glycation end products (AGEs) formation has been linked to atherosclerosis in patients with diabetes [1] as well as in nondiabetic subjects [2]. AGEs are short- and long-term modification products of glycation or glycoxidation of proteins and lipids [1]. These end products are a heterogeneous group of compounds with different biological effects, some of which are mediated by interacting with receptors, including receptor for AGE (RAGE) on endothelial cells, smooth muscle cells and macrophages [3]. Several AGE compounds have also been localized in atherosclerotic plaques [4]. AGEs may contribute to atherosclerosis by activating the transcription factor NF-κB through the binding to RAGE, thus initiating induction of cellular adhesion molecule expression and cytokine activation [3], and through glycoxidation of lipoproteins and increased foam cell formation [5].

AGEs can be formed extracellularly as well as intracellularly. In endothelial cells biologically reactive dicarbonyl methylglyoxal (MG) has been identified as the major intracellular precursor in the formation of AGEs [6]. Methylglyoxal is formed by non-enzymatic elimination of phosphate from triosephosphates as well as enzymatically from dihydroxyacetone during glycolysis, and MG is detoxified to d-lactate by the glyoxalase system. Methylglyoxal reacts reversibly with arginine, lysine and cysteine residues in proteins [7] to form among others, the non-fluorescent arginine-modification Nδ-(5-hydro-5-methyl-4-imidazolon-2-yl)-ornithine (MG-H1 or methylglyoxal-derived hydroimidazolone). Methylglyoxal-derived hydroimidazolone residues represent the major glycation type protein-modification in diabetes, and occur both in extra- and intra-cellular proteins. Plasma levels of MG-H1 modified proteins have been shown to be increased in patients with type 1 diabetes compared to nondiabetic controls [8].

Limited information is available on the effects of increased serum levels of methylglyoxal-derived hydroimidazolone proteins on the development of atherosclerosis in humans, although in vitro evidence suggests possible links [9]. We have previously demonstrated that high levels of AGEs measured with a polyclonal anti-AGE antibody predicted increased coronary heart disease mortality in nondiabetic [10] and type 2 diabetic women [11]. In the present nested case-control study including diabetic and nondiabetic subjects, we investigated whether baseline levels of the specific AGE compound MG-H1 were associated with CVD mortality in an 18-year follow-up study.