James D. House1,2,3,4 | Matthew Nosworthy1
Consumers continue to demand products that feature protein claims on front of the package, with protein featured as one of the top 10 key trends in food, nutrition and health for 2015.1 Factors driving increased interest in protein include perceptions by consumers of the linkage between protein consumption and increased satiety (feeling of fullness), sustained energy, and maintenance of a healthy immune system. During food product formulation, the nature of any protein is a critical decision point, particularly when balancing issues such as allergens (i.e. milk, soy, wheat) and perceptions regarding healthfulness and sustainability of production practices. Plant-based protein sources can allow the food industry to capture the protein trend, however foods need to meet the regulatory requirements necessary to qualify for a protein content claim. In Canada, the Protein Rating2 system is used which is based on a rodent growth bioassay to determine the Protein Efficiency Ratio (PER) of a given protein source. Protein ratings between 20 and 40 qualify foods for a “Source of Protein” claim, while ratings above 40 qualify for “Excellent Source of Protein” claims. This method, while simple, has severe limitations including the fact that the results derived from individual protein sources cannot be used to calculate a protein rating for blends of different proteins, in any meaningful way. In the United States, source claims are based on the Protein Digestibility-Corrected Amino Acid Score method, or PDCAAS3, a method that does permit calculation of values for protein blends. This method yields a value derived as the product of: a) an estimate of the digestibility of the protein, determined using a rodent bioassay; and b) the amino acid score, a value calculated from the amino acid composition of the protein source when compared against human amino acid requirement estimates.
For both the Protein Rating and the PDCAAS methods, the use of a rodent bioassay is required. A new method, called the Digestible Indispensable Amino Acid Score or DIAAS, has been positioned as an alternative assessment method for establishing protein quality estimates4. However, the DIAAS method also relies on the use of an animal bioassay, in this case a surgically-modified swine model, complete with a cannula inserted through the abdominal wall of the animal into the terminal end of the small intestine. While theoretically a refinement of the PDCAAS method, to date, no convincing evidence exists to substantiate that protein nutrition of the human population would be substantially enhanced by moving to the DIAAS method. It is important to note that, from a regulatory perspective, protein is the only nutrient that requires the use of an animal model for the establishment of nutrient content claims. Perhaps the time has come to focus on alternative methods to measure protein quality in support of protein content claims. This is particularly important for those companies within the food sector that have policies in place to prohibit the use of bioassays. Any positioned alternative should be: a) sufficiently robust to ensure that any risks to human health are minimized; b) flexible to permit the determination of the quality of novel protein blends; c) widely available and accessible to permit innovation via the food industry; and d) validated against an existing, approved methodology. Alternatives exist, but unlike current international efforts to develop consensus around a new in vivo approach4, the same cannot be said for in vitro approaches.
IN VITRO MEASUREMENTS OF PROTEIN QUALITY
The simplest approach to expressing the ability of a food product to contribute to human protein needs is the use of crude protein content. Effectively, this approach is used in those jurisdictions that don’t mandate the use of either PER or PDCAAS to reflect the protein quality of foods.5 Given that current national databases examining nutrient intake in the human population focus primarily on crude protein intake, without adjusting for protein quality, perhaps this approach is sufficient. However, simply maintaining a focus on crude protein (i.e. nitrogen x 6.25) does not necessarily protect consumers from exposure to ingredients that cannot contribute to human protein needs, including severely damaged proteins, ingredients that contain anti-nutritive factors or ingredients that are high in non-protein nitrogen. Shifting the focus to the amino acid composition of the ingredient is a significant advancement and reflects the primary approach used in both the PDCAAS and DIAAS methods. Existing nutrient databases6 contain the amino acid composition of thousands of food items, and amino acid analysis of novel foods or food ingredients is readily available via commercial laboratories. The amino acid composition data can then be positioned against human amino acid requirement reference patterns to establish the amino acid score. A score of 1.0 or greater would indicate that the food or ingredient is not deficient in any amino acid with respect to its ability to support human amino acid needs. A score less than 1.0 is indicative of a deficiency in one or more of the essential amino acids. The amino acid score (also called chemical score) for defining protein quality could be used instead of the PDCAAS, essentially following the same procedures for correcting the protein content for quality estimates. The obvious limitation in using just the score is the fact that the value does not reflect any factors that may limit the digestibility or availability of the amino acids contained in the protein source. In order to offer a more robust alternative to PDCAAS, the continued use of the amino acid scoring concept coupled with an in vitro assessment of protein/amino acid digestibility may fit the bill.
The measurement of in vitro protein digestibility can be accomplished through either static or dynamic measurement systems. Static systems reflect methods that seek to measure the release of amino acids from dietary proteins upon exposure to gastric and/or intestinal proteases, under discrete temperature and pH conditions. The measurement of amino acid release can be monitored by measuring individual amino acids, free amino groups, or by changes in pH7. These methods have advantages in terms of cost, throughput and ease of implementation, however the static nature of the systems distances them from normal physiological processes. Several methods have been proposed and, to date, consensus has not been reached on a single method. Dynamic systems for measuring protein digestibility involve the use of computer-controlled chambers designed to mimic the human gastrointestinal tract, where pH, temperature, enzyme addition, mixing and residence times are precisely controlled8. While the systems tend to model gastric physiology more accurately, sample throughput and cost per analysis make routine use challenging. However, the coupling of in vitro protein digestibility with the amino acid score, to yield an in vitro PDCAAS, creates the opportunity to move away from conducting animal-based studies for protein quality evaluation. What is needed to advance this concept are studies designed to directly compare in vivo and in vitro PDCAAS measurements and evidence to support that such a move would not compromise human protein nutrition. Approved or accepted in vitro techniques would offer the food industry opportunities to quickly assess new product formulations, while affording the confidence that the values could be used to support regulatory claims. This is particularly true for the plant protein sector, given the variety of new plant protein sources and the strong interest expressed towards the inclusion of plant proteins into food products.
CASE STUDY: PROTEIN QUALITY OF NOVEL PLANT PROTEIN BLENDS
As an example of how in vitro PDCAAS could be used to assist the food industry, a study was conducted to examine the in vivo and in vitro PDCAAS of two novel plant flours: a) extruded whole buckwheat flour; and b) extruded whole pinto bean flour. The amino acid score of both protein sources was determined (3), and corrected for either in vivo3 or in vitro9 protein digestibility. The data in Figure 1 shows how the PDCAAS values, regardless of how digestibility was determined, can be influenced by blending the two protein sources. The addition of extruded pinto bean flour complements the lower lysine level in the extruded buckwheat flour, while the sulphur amino acid content of buckwheat mitigates the limiting nature of the pinto beans. As depicted in Figure 1, despite a slightly lower overall PDCAAS for the in vitro method, the optimal blend estimates were the same (45:55 buckwheat: pinto bean).
A PATH FORWARD
The ability of the food industry to adopt new innovations in supplying plant-based protein sources to consumers is, to a certain extent, limited by current regulations mandating the use of animal data to support protein quality estimates. Positioning alternatives should become a priority for the food industry. However, any new approaches must provide regulatory agencies the confidence that risks to human health have been mitigated.
 Mellentin, J. (2014). 10 Key trends in food, nutrition and health. New Business Journal. December, 2014. (accessed: https://www.new-nutrition.com/report/showReport/1235)
 Health Canada, Health Protection Branch. Official Method F0-1: The determination of protein rating. (accessed 11/15: http://www.hc-sc.gc.ca/fn-an/alt_formats/hpfb-dgpsa/pdf/res-rech/fo-1-eng.pdf)
 Joint FAO/WHO (1990). In Joint FAO/WHO expert consultation on protein quality evaluation. Food and Agriculture Organization of the United Nations; Rome, Italy.
 FAO Expert Consultation (2013). Dietary protein quality evaluation in human nutrition. Food and Agriculture Organization of the United Nations. Food and Nutrition Paper 92. Rome, Italy.
 Lewis, J.L. (2012). Brit. J. Nutr. 108: S212.
 Canadian Nutrient File. Online database. (Accessed: http://webprod3.hc-sc.gc.ca/cnf-fce/index-eng.jsp)
 Hsu, H.W. et al. (1977). J. Food Sci. 42:1269.
 Minekus, M. et al. (1995). ATLA 23:197.
 Tinus, T. et al. (2012). J. Food Eng. 113:254.
1Department of Human Nutritional Sciences, University of Manitoba, Winnipeg, MB
2Department of Animal Science, University of Manitoba, Winnipeg, MB
3Richardson Centre for Functional Foods and Nutraceuticals, University of Manitoba, Winnipeg, MB
4Canadian Centre for Agri-Food Research in Health and Medicine, Winnipeg, MB
(*Corresponding author email: James.House@umanitoba.ca)