Non-transferrin-bound iron and protein glycation in type 2 diabetes

Abstract

Background and Methods The involvement of iron in the risk for, and complications of, type 2 diabetes has generated substantial interest over the past 15 years, initially sparked by an association with raised serum ferritin, and the observation that people with iron overload diseases frequently develop diabetes. Considerable advances have since been made in understanding the effect glucose has on molecules, cells, and tissues; and the role that oxidative stress plays in the development of the pathologies of long-term diabetes. Poorly liganded iron is potentially both a contributor to, and consequence of, these complications. In vitro experiments with glucose-incubated transferrin by earlier workers have demonstrated loss of function with increasing glycation, leading to the suggestion that the failure of this key iron-binding protein may contribute to diabetic pathology, via the presence of redox active non-transferrin-bound iron (NTBI). In vitro glycated transferrin is examined here by ultrafiltration, to assess loss of function and possible oxidative fragmentation. Mass spectrometry is used to identify a range of amino acid glycation sites on in vitro glycated transferrin for the first time. Finally, several groups have previously measured NTBI in people with diabetes, finding little agreement in results. NTBI is measured here in a cohort of people with type 2 diabetes, using a new adaptation of earlier NTBI assays. NTBI is also assessed in pre-dialysis chronic kidney disease (CKD) stages I to III for the first time. Results and Conclusions Experiments with glycated transferrin in vitro demonstrate oxidative fragmentation, explaining the loss of function reported by earlier groups. In vitro glycated transferrin examined by mass spectrometry reveals a substantial number and range of amino acids subject to glycation. Comparison with in vivo glycated transferrin suggests that many of the in vitro glycation sites are not glycated in vivo, and that there are many oxidized methionine residues which are potential artefacts, or likely to be repaired by methionine sulphoxide reductases in vivo. A study of people with type 2 diabetes finds no direct association between NTBI and protein glycation. Unexpected correlations between NTBI and LDL, and LDL and haemoglobin with increasing protein glycation, are reported for the first time. NTBI is suggested to be iron sourced from haemoglobin or haem, from erythrocyte haemolysis prior to sample collection. In people with pre-dialysis CKD stages I to III no significant difference in NTBI level compared to controls is seen, or correlations with markers of renal function. No link between NTBI and kidney function at this stage of disease is indicated.