Determination of the protein quality of cooked Canadian pulses

Abstract A study to determine the protein digestibility‐corrected amino acid score and protein efficiency ratio of nine different cooked Canadian pulse classes was conducted in support of the establishment of protein quality claims in Canada and the United States. Split green and yellow pea, whole green lentil, split red lentil, Kabuli chickpea, navy bean, pinto bean, light red kidney bean, and black bean were investigated. Protein digestibility‐corrected amino acid score (PDCAAS) and the protein efficiency ratio (PER) were determined using the appropriate rodent models. All pulses had high digestibility values, >70%, with PDCAAS values greater than 0.5, thereby qualifying as a quality protein in the United States, but only navy beans qualified as a good source of protein. All pulses except whole green lentils, split red lentils, and split green peas would qualify as sources of protein with protein ratings between 20 and 30.4 in Canada. These findings support the use of pulses as protein sources in the regulatory context of both the United States and Canada.

the lowest amount relative to a reference requirement pattern reflects the value of the amino acid score. Values less than 1.0 reflect the fact that at least one amino acid is limiting relative to the requirement pattern. If a food protein is completely digested and absorbed, then the amino acid score reflects the inherent limitations of a given protein for productive purposes. Most food proteins, particularly plant-based food proteins, are not completely digestible. The PDCAAS reflects an attempt to measure the overall quality of a protein as the product of the digestibility of the protein and its amino acid score. Recently, a new method has been proposed for determining protein quality called the Digestible Indispensable Amino Acid Score (DIAAS) (FAO/ WHO, 2013). This approach uses ileal digestibility rather than fecal, and considers the digestibility of individual amino acids instead of a single value; however, this method has yet to be officially adopted by any jurisdiction. Interestingly, the European Union does not use PER, PDCAAS, or DIAAS for its regulation of protein content claims, rather if a food product contributes 12% of its total energy via protein it is considered a 'source of protein'; if the contribution of protein to total energy is 20% that product has 'high protein' (European Commission, 2017).

| Sourcing of pulse samples
For the current study, composite pulse samples were prepared from individual lots collected from several processors across Canada from the 2010 cropping year. Composite samples were prepared by the Canadian International Grains Institute by blending an equal amount of each processors sample for 5 min using a Hobart mixer (Model HL 200). A description of the pulse samples is presented in Table 1.

| Cooking of pulse samples
All beans and chickpeas were soaked for 16 hr prior to cooking. The lentils and split peas were cooked without soaking. Cooking times for the pulses are listed in Table 2. For chickpea, navy bean, black bean, kidney bean, and pinto bean cooking times were determined via an automated Mattson cooker (Wang & Daun, 2005). In brief, plunger weight was adjusted according to market class and each tip placed onto one seed of the appropriate market class. The loaded apparatus was then incubated in boiling water until all plungers had dropped.
Cooking time was defined as the time required for 80% of the

| Analytical procedures
Prior to analysis, all samples were ground in a hand-held electric mill.

| Protein digestibility-corrected amino acid score determination
A rat bioassay, as described previously, was used to determine the PDCAAS of the pulse samples (AOAC International, 1995 Feed was restricted to a maximum of 15 g/day over a 4-day acclimation period followed by a 5-day balance period, during which daily feed intake was calculated. Total fecal output was collected during the balance period, air-dried, and analyzed for its dry matter and nitrogen content. True protein digestibility (TPD%) was calculated as follows where nitrogen intake and fecal nitrogen loss represent the product of food intake or fecal weights and their respective nitrogen values: The value for metabolic nitrogen loss was determined as the amount of fecal nitrogen produced by rats consuming a protein-free diet. The PDCAAS was calculated as the product of the amino acid score and TPD%. For the purposes of establishing a protein content claim in the United States, the products of the protein content of a 90 g representative serving size and the PDCAAS of each pulse were compared to the daily reference value (DRV) of 50 g of protein.
Values between 10 and 19.9% of the DRV constitute "Good Sources" of protein.

| Protein efficiency ratio
Protein Efficiency Ratio (PER) values were determined over a 28-day growth period for rats consuming feed ad libitum (Health Canada, 1981). For the current study, the 28-day growth period included the 9-day protein digestibility study period. Rat weights were recorded throughout the acclimation and balance periods, and feed intake was recorded throughout the study. contained in the reference serving (g). Protein ratings between 20 and 39.9 qualify for a "Source of Protein" claim, while values of 40 or greater qualify for an "Excellent Source of Protein Claim" (Health Canada, 1981).

| Digestible indispensable amino acid score (DIAAS)
Digestible indispensable amino acid score (DIAAS) was calculated using the amino acid reference pattern for children aged 6 months to 3 years which was used in conjunction with the following equation

| Proximate analysis and amino acid composition of cooked Canadian pulses
The data for percentage crude protein (%CP) and crude fat (%Fat) content of the various pulse classes are provided in 72%-75%) as compared to the beans and chickpeas (approx. 65%).
Given the fact that serving sizes are expressed on a wet weight basis, the moisture content of the pulses can influence the grams of protein present in a given volume of a serving. Due to displacement, the moisture content will have a significant impact on the protein rating of a cooked pulse. By standardizing the cooking times through the use of automated Mattson cooker procedure, this study sought to achieve a level of "doneness" that reflected those typically encountered in consumer settings. As such, the density of the pulses, and thus the serving sizes, likely reflects those relevant for protein content claims at the consumer level. However, further opportunity exists to examine the impact of cooking methods on moisture content and, ultimately, the density calculations relative to standard serving sizes.

| Amino acid scores of cooked Canadian pulses
The amino acid scores were calculated according to the 1991 FAO reference pattern (Table 4). Using the 1991 Reference Pattern, on the basis of the lowest ratios observed, the sulfur amino acids, methionine and cysteine, were limiting for red kidney beans, whole green lentils, split red lentils, split green peas, and black beans. Alternatively, the amino acid tryptophan was limiting for navy beans, split yellow peas, chickpeas, and pinto beans. Other groups have found similar results when determining limiting amino acids in pulses (Jackson, 2009;Sarwar & Peace, 1986;Wu et al., 1996). The amino acid scores of the cooked pulses ranged from 0.59 (split red lentils and split green peas) to a high of 0.83 for navy beans.

| True protein digestibility & PDCAAS
True protein digestibility values are presented in Table 4 and reflect the amount of nitrogen digested by rats during a bioassay, with corrections for endogenous nitrogen losses. In general, the protein in cooked pulses is highly digestible, with values being 70% or greater.
True protein digestibility (TPD %) was lowest for black beans (70%), and highest for the split red lentils (90.6%). Previous investigation into the digestibility of autoclaved pulse flour found similar digestibilities of peas 88%, common beans 79% (average), lentils 85%, and black beans 72% (FAO/WHO, 1991) . The lower value observed in black beans may be related to the phenolic fractions present in these beans, or other factors not yet recognized. With respect to chickpea, raw seed has been found to have a digestibility of 72% which increases to 84% after cooking (Clemente, Shnchez-vioque, Vioque, Bautistab, & Millin, 1998), similar to the results of this study. Pulses contain a wide variety of antinutritional factors that reduce digestibility and nutrient availability including proteolytic inhibitors (Gupta, 1987;Oomah et al., 2011) and tannins (Hahn et al., 1984). Cooking has been shown to reduce trypsin inhibitor activity and tannin concentrations (Wang et al., 2008(Wang et al., , 2009(Wang et al., , 2010, which contribute to increased protein digestibility. The product of the true protein digestibility and the amino acid score provides the PDCAAS values, and these are presented in for noninfant foods, and greater than 0.4 (40%) to qualify as a quality protein for foods intended for infants (FAO/WHO, 2013). In order to make a protein content claim in the United States, foods must first meet the definition of a quality protein and all of the pulses tested met these criteria (Table 6)

| Protein efficiency ratios and protein ratings
Within Canada, the official method for determining the protein quality of foods is the Protein Efficiency Ratio (PER) (Health Canada, 1981).
The PER is a rat bioassay method in which rats consume a semipurified diet where the protein of interest represents the sole source of protein (present at 10% of the diet). In order to calculate PER both protein consumption and weight gain are monitored over a 28-day period with the PER being calculated as the total weight gain divided by the total protein intake. In order to provide consistency between studies, casein is used as a reference protein, and all PER values are adjusted to a common PER for casein of 2.5. The data for the adjusted PER values for all cooked pulses are presented in Table 6.
The adjusted PER values for the pulses ranged from a low of 0.86 for Split Green Peas to a high of 2.32 for Chickpeas. In order to successfully obtain a Canadian protein content claim, a serving of a given food must have a Protein Rating of 20.0-39.9 to qualify as a "Source of Protein". Protein Ratings above 40.0 qualify for an "Excellent Source of Protein" claim. The Protein Ratings are calculated as the product of the Adjusted PER and the amount of crude protein (grams of N × 6.25) in a representative serving of the food. The data given in Table 6 have been derived using a serving size of 250 ml. contain protein that is highly digestible (typically greater than 80%).
Limiting amino acids within the proteins will depend on the choice of reference amino acid pattern, but typically the sulfur amino acids or tryptophan is the first limiting amino acid. Protein Rating values presented support the use of "Source of Protein" content claims for most pulses. The use of the PDCAAS will support the establishment of "Good Source of Protein" content claims in the United States for Navy Beans, Split Yellow Peas, Chickpeas, and Pinto Beans.

ACKNOWLEDGMENTS
Funding for this research was provided from a grant-in-aid from Pulse Canada (Winnipeg, MB, Canada).

CONFLICTS OF INTEREST
The authors declare that they have no conflicts of interest.

23.27
Bolded values qualify for "Good Source of Protein" claim in United States and Canada, respectively. The reference serving size used for pulses in the United States was 90 g and protein content is provided on an as is basis.