Optimisation of a solvent for the complete extraction of prolamins from heated foods

Wheat bread was lyophilised, ground, extracted and centrifuged. The supernatants were analysed for gluten content by RP-HPLC and a commercial sandwich ELISA. Prolamin extraction solvents contained tris(2-carboxyethyl)phosphine (TCEP; 1, 2, 5, 10, 20, 50 mmol/L), guanidine hydrochloride (GUA; 0 or 2 mol/L) and a buffered salt solution. A commercial cocktail solution (250 mmol/L mercaptoethanol (ME), 2 mol/L GUA, buffered salt solution) as well as 60% (v/v) ethanol were used as control solvents. Wheat flour was the control for the extractability of the native prolamins. 60% ethanol only extracted 37% of the prolamins from wheat bread (cocktail = 100%). When ME was replaced by TCEP the protein yield increased from 35% at the lowest TCEP-level to 95% when 20-50 mmol/L TCEP were used. The use of GUA was essential to extract prolamins quantitatively. Comparative protein analysis using RP-HPLC and ELISA showed that both methods provided comparable prolamin (gliadin) concentrations of the wheat flour (40.3-45.7 mg/g), when 60% ethanol was used as extraction solvent. The extraction yields from bread were considerably lower (16.7-24.7 mg/g). Cocktail and TCEP extracted almost the same amount of protein from flour and bread with TCEP showing slightly lower yields. Total extractable protein (gliadin glutenin) as determined by RP-HPLC was 70.5-75.3 mg/g, and total gliadin as determined by ELISA was 42.7-44.2 mg/g. Thus, the study has shown that TCEP in combination with GUA extracts proteins from heated, gluten-containing foods as effective as the commercial cocktail solution. (C) 2010 Elsevier Ltd. All rights reserved.