Cereals & Grains Association
Log In

  How comprehensive does protein quality data need to be to predict gluten strength in a population of bread wheats?
H. SAPIRSTEIN (1), C. Isaak (1), R. Graf (2) (1) University of Manitoba, Winnipeg, MB, Canada; (2) Agriculture & Agri-Food Canada, Lethbridge, AB, Canada.

Gluten protein composition and small-scale dough mixing results were compiled for a set of 52 HRW wheats grown in one site and year. Flour protein content (FPC) ranged from ~11-14%. A mixograph was used w/o and with salt at constant absorption. Gluten strength was assessed by work input to peak development (WIP) and other parameters, and varied considerably across genotypes. FP was fractionated using a basic differential solubility method in 50% 1-propanol w/o and with reducing agent yielding propanol soluble protein (SP), insoluble glutenin (IG) and residue protein. SP and IG were quantified by UV absorbance. Gluten proteins were also fractionated using a more comprehensive procedure to distinguish gliadins, LMW soluble glutenin, HMW insoluble glutenin, with subunits quantified by RP-HPLC. Prediction of WIP ranged from R2 = 0.23 (for FPC alone) to > 0.80 depending on the type and number of protein factors determined by stepwise regression and whether salt was used in dough mixing. Salt had a non-linear and genotype-specific effect that improved discrimination of samples and prediction of WIP; e.g. the most influential protein quality factor = FPC*IG/SP had R2 w/o and with salt = 0.59 and 0.69, respectively.  Prediction of WIP could be improved to R2 ~ 0.76 and ~0.79 in 2- and 3-variable regression models by adding any number of protein quality parameters derived from detailed protein fractionation and HPLC results. Including individual subunits of reduced HMW or LMW glutenin did not enhance prediction of WIP.  Results indicated that many of the most effective protein quality parameters for predicting gluten strength were confounded by FPC. It was concluded that a relatively simple gluten protein fractionation and quantification procedure could be successfully used to explain a high proportion of the variation of gluten strength in this population of genotypes.