The source of circulating selenoprotein S and its association with type 2 diabetes mellitus and atherosclerosis: a preliminary study.

BACKGROUND Selenoprotein S (SelS) is a transmembrane protein that is expressed in the liver, skeletal muscle, adipose tissue, pancreatic islets, kidney, and blood vessels. In addition to its transmembrane localization, SelS is also secreted from hepatoma HepG2 cells (but not L6 skeletal muscle cells, 3T3-L1 adipocytes, Min6 pancreatic β cells and human embryonic kidney 293 cells) and has been detected in the serum of some human subjects, with a detection rate of 31.1 %. These findings prove that serum SelS is secreted by hepatocytes. However, whether vascularly expressed SelS can be secreted has not been reported. Transmembrane SelS has been suggested to play different roles in the pathogenesis and progression of diabetes mellitus (DM) and atherosclerosis (AS), but the association of secreted SelS with DM and macroangiopathy remains unclear. RESEARCH DESIGN AND METHODS Supernatants were collected from human umbilical vein endothelial cells (HUVECs), human aortic vascular smooth muscle cells (HA/VSMCs) and human hepatoma HepG2 cells that were untransfected or transfected with the indicated plasmid and concentrated for western blotting. Serum samples were collected from 158 human subjects with or without type 2 DM (T2DM) and/or AS. Serum SelS levels were measured using an enzyme-linked immunosorbent assay. RESULTS Secreted SelS was only detected in the supernatants of hepatoma HepG2 cells. The SelS detection rate among the 158 human serum samples was 100 %, and the average SelS level was 64.81 ng/dl. The serum SelS level in the isolated DM subjects was lower than the level in the healthy control subjects (52.66 ± 20.53 vs 70.40 ± 21.38 ng/dl). The serum SelS levels in the DM complicated with SAS subjects (67.73 ± 21.41 ng/dl) and AS subjects (71.69 ± 27.00 ng/dl) were significantly increased compared with the serum SelS level in the isolated DM subjects. There was a positive interaction effect between T2DM and AS on the serum SelS level (P = 0.002). Spearman correlation analysis showed that the serum SelS level was negatively correlated with fasting plasma glucose. CONCLUSIONS Vascular endothelial and vascular smooth muscle cells could not secrete SelS. Serum SelS was primarily secreted by hepatocytes. SelS was universally detected in human serum samples, and the serum SelS level was associated with T2DM and its macrovascular complications. Thus, regulating liver and serum SelS levels might become a new strategy for the prevention and treatment of DM and its macrovascular complications.