Determination of the in vivo and in vitro protein quality of pulse protein concentrates and isolates.

Currently, the use of a rodent bioassay is the officially recognized method for generating data in support of protein content claims for food ingredients. Although there are in vitro methods for calculating protein digestibility, to date no method has been accepted by North American regulatory agencies for use in establishing protein content claims, likely due to a lack of evidence in support of their validity. To this end, we compared the in vivo (%True Fecal Protein Digestibility; %TFPD) and the in vitro protein digestibility (%IVPD) of faba bean, lentil and pea protein concentrates and isolates. Sprague‐Dawley rats (n=70, ~70 g) were randomized to one of seven diets (10% crude protein) based on: casein (reference control), the concentrates of faba bean (FB60), lentil (L55) or pea (P55) or the protein isolates of the same pulses (FB85, L85 and P85 respectively). The %TFPD and protein digestibility corrected amino acid score (PDCAAS) were calculated. The %IVPD was determined via enzymatic digestion with subsequent pH drop, and used to calculate the in vitro PDCAAS (IVPDCAAS). Except for L55, in which the sulfur amino acids were limiting, all experimental diets were limited by tryptophan content. FB60 had a higher AAS than FB85 (0.46 vs 0.58) whereas the AAS of the other diets had the opposite trend (L55‐0.75 vs L85‐ 0.52, P55 – 0.58 vs P85‐ 0.54). Digestibility was greater in the protein isolates than the concentrates, and the in vivo rat bioassay reported higher digestibility than the in vitro method; casein (95 vs 84), FB60 (94 vs 74), FB85 (97 vs 75), L55 (91 vs 75), L85 (96 vs 77), P55 (93 vs 78), P85 (97 vs 78). There was no correlation found between %TPD and %IVPD (R 2 = 0.027). Interestingly, a strong correlation was found between PDCAAS and IVPDCAAS (R 2 =0.99, p<0.0001), however the values from the rodent model were greater than the in vitro casein (89 vs 79), FB60 (43 vs 33), FB85 (56 vs 44), L55 (68 vs 56), L85 (50 vs 40), P55 (54 vs 46), P85 (53 vs 43). Excepting faba bean, all pulse concentrates had higher AAS, lower digestibility and greater PDCAAS values than their isolate counterparts. As such, processes used in the isolation of pulse protein sources increased digestibility, but may have led to shifts in protein composition, leading to a lower PDCAAS value. Although the strong correlation found between PDCAAS and IVPDCAAS suggests that in vitro analysis could reduce the requirement of animal experimentation for protein claims, a larger dataset combined with investigation into other potential in vitro methods is required. Support or Funding Information Pulse Canada