Effect of Transferrin Glycation on the Use of Carbohydrate-Deficient Transferrin as an Alcohol Biomarker

• Published in: Alcohol and alcoholism (Oxford) Vol. 48; no. 4; pp. 478 - 482
• Main Authors: HELANDER, Anders, MODEN, Naama Kenan
• Format: Journal Article
• Language: English
• Published: Oxford Oxford University Press 01.07.2013
• Subjects: Alcohol consumption; Alcohol Drinking - blood; Alcoholism and acute alcohol poisoning; Biological and medical sciences; Biological markers; Biomarkers - blood; Diabetes Mellitus - blood; Diabetics; Endokrinologi och diabetes; Farmakologi och toxikologi; Glucose; Glycerophospholipids - blood; Glycosylation; Klinisk medicin; Medical sciences; Medicin och hälsovetenskap; Medicinska och farmaceutiska grundvetenskaper; Quantification; Serum; Sialoglycoproteins - blood; Toxicology; Transferrin - analogs & derivatives; Transferrin - metabolism; Consumption; Biological marker; Alcoholic beverage; Ethanol; Glycation; Carbohydrate deficient transferrin
• ISSN: 0735-0414; 1464-3502; 1464-3502
• DOI: 10.1093/alcalc/agt045

Measurement of an alcohol-induced shift in the serum transferrin glycosylation pattern, termed carbohydrate-deficient transferrin (CDT), is used as a biomarker for sustained high alcohol consumption. The present work examined whether non-enzymatic reaction of transferrin with glucose (glycation) might interfere with the use of CDT as an alcohol biomarker. The blood specimens were leftover volumes from the routine sample pool. Plasma and serum were collected among samples submitted for hemoglobin A1c (HbA1c) and CDT testing. Quantification of individual transferrin glycoforms in percentage of total transferrin was performed by an HPLC candidate CDT reference method. Incubating serum spiked with 20 or 200 mmol/l glucose caused time- and dose-dependent changes in the chromatographic profile of transferrin glycoforms, resulting in gradually wider peaks and reduced relative amounts of disialo- and trisialotransferrin. No similar chromatographic effects were seen in samples collected from diabetic patients with elevated HbA1c (>68 mmol/mol) values. These samples instead showed slightly higher mean %disialotransferrin levels (1.21%) compared with low HbA1c (<44 mmol/mol) samples (mean 1.06%; P = 0.023), pointing at a higher alcohol consumption level in the former group. Altogether ∼5% of the CDT values exceeded the cutoff. There was no significant difference in phosphatidylethanol (PEth) levels between the high and low HbA1c samples, but several (∼14%) showed elevated PEth concentrations. Glycation of serum transferrin in vivo was indicated to differ from that in vitro, and suggested not to interfere with %CDT testing by the HPLC method. The results indicated that CDT and PEth are useful as objective, complementary alcohol biomarkers to identify risky drinking also in diabetic subjects.