Abstract
Fructose-lysine linkage is formed by the non-enzymatic binding of glucose to the s-amino group of lysine in plasma protein. Furosine derived from fructose-lysine of plasma protein by acid hydrolysis was identified by HPLC for the estimation of glycosylated plasma protein (GPP). An equimolar amount of glucose was added to a lysine solution. Then the mixture was heated at 100°C for 1 hour and was hydrolyzed in 6 mol/l HCL at 95°C for 30 hours. Hydrolysates of plasma proteins from nondiabetics and diabetics were determined by HPLC using a μ Bondapak C18 column. The area under the furosine peak which appeared at a retention time of 4.1 minutes under the following chromatograhic conditions was measured: flow rate, 1 ml/min; UV detector wavelength, 280 nm and 254 nm. The furosine appeared at a peak height A 280 nm/A 254 nm ratio of 3.9. This peak could not be found by reduction with NaBH4 prior to hydrolysis. Therefore, this peak was considered a peak of furosine. The tyrosine peak was used as an internal standard. The levels of GPP were expressed as the ratio of furosine peak area to tyrosine peak area, i.e., furosine/tyrosine × 100 (%). The intra-assay and inter-assay coefficients of variation of furosine values were I. 4% and 3.6%, respectively. GPP values paralleled plasma protein concentrations. GPP values reached a plateau at a content of 30 mg plasma protein and after 24 hours' hydrolysis. The furosine value of GPP was 4.6 ±1.5%(mean ± S.D.) in diabetics, significantly higher than that in normal subjects, i.e., 2.6 ± 0.5%. A significant positive correlation was found between GPP and fasting blood glucose (r= 0.66, p<0.001) or HbA1 (r=0.62, p<0.001) The results indicate that furosine derived from fructose-lysine in glycosylated plasma protein may become an indicator of long-term blood glucose control in diabetic patients.
