Assessment of Protein Quality

Metabolic Studies

The most accurate assessment of protein quality of foods for humans is through clinical or metabolic studies that measure nitrogen balance. A fixed amount of protein is fed to a group of individuals until a steady state is reached. At that point, excreta are collected and analyzed for their nitrogen content, and integumental nitrogen losses are generally estimated at approximately 5mgNkg—1 to calculate NB as follows: NB = IN — UN — FN — IntegN (See abbreviations in Table 1). Measurements are repeated with different amounts of food protein and the relationship between nitrogen intake and nitrogen balance is evaluated (Figure 1). The slope of the line before NB reaches a plateau and the amount of dietary protein needed to attain ''0'' nitrogen balance are indicators of protein quality: The steeper the slope and the lower the amount of dietary protein to achieve balance, the higher the quality of the protein being tested.

33 "to



0.60 0.80 1.00 Protein intake, g/kg/d




0.60 0.80 1.00 Protein intake, g/kg/d

Figure 1 Plot of a nitrogen balance study. Nitrogen retention increases with protein intake until a plateau is reached.

Influence of Energy Intake on Nitrogen Balance

When food energy intake is insufficient to satisfy energy needs, amino acid oxidation increases in an effort by the human body to satisfy energy requirements. This raises urinary nitrogen excretion and reduces nitrogen balance. On the other hand, increased energy intake may reduce amino acid oxidation and urinary nitrogen excretion, thereby improving N balance until it reaches a plateau. This response, known as the protein-sparing effect of dietary energy, can be attenuated if the quantity or quality of food protein intake is inadequate. It has been postulated that the protein-sparing effect of dietary carbohydrates is mediated by increased insulin secretion, which inhibits proteolysis, hepatic gluconeogenesis, and renal ammoniagenesis. The protein-sparing effect of dietary fat may be due to a reduction of amino acid oxidation through an effect of free-fatty acid oxidation in the liver, whereby the increase in NADH/NAD inhibits branched-chained keto-acid dehydrogenase. For these reasons nitrogen balance must not be used to estimate protein quality when the amount of dietary energy is such that it produces weight loss or gain in an otherwise well-nourished individual.

Because of their high cost and experimental complexity, metabolic studies are done mainly to evaluate new, nonconventional protein sources and novel food processes that may affect protein quality. Other methods that can predict protein quality for humans rapidly and at low cost are used to evaluate diets and conventional foods routinely.

Assays in Laboratory Animals

Biological assays in laboratory animals have been used to assess food protein quality, based either on a protein's ability to support growth in young rats (protein efficiency ratio) or on nitrogen retention (net protein utilization). However, these assays underestimate the quality of some vegetable and animal proteins for humans. For example, the proteins of pulses and milk casein have a lower quality for rats than for humans because rats have a higher requirement of sulfur-containing amino acids. Thus, application of rat assay results to human nutrition can result in important quantitative errors. The discrepancy usually has economic rather than public health implications because rat assays generally err by underestimating protein quality for humans, but the value of certain animal proteins can be overestimated because of higher efficiency of utilization by the rat.

Amino Acid Score Adjusted for Digestibility

The concept of assessing protein quality on the basis of a protein's constituent amino acids was introduced in the late 1940s. It was later suggested that the calculations be corrected by the protein's digestibility. The validity of this approach and its correlation with results of metabolic and clinical studies were initially limited by lack of accurate procedures to measure tryptophan and sulfur amino acids, insufficient information on digestibility of proteins from various sources, and uncertainty about human amino acid requirements to prepare an adequate scoring pattern. Significant scientific and technological advancements now allow the use of an amino acid scoring procedure adjusted for digestibility as a good and practical predictor of protein quality for humans.

This method is recommended by expert committees of the World Health Organization (WHO), United Nations Food and Agriculture Organization (FAO), United Nations University (UNU), and the Codex Committee on Vegetable Proteins (CCVP), as well as by regulatory agencies of several countries, for routine evaluation of protein quality for humans. The elements required for its application are knowledge about the amino acid composition and digestibility of the food protein(s) under evaluation and a scoring pattern based on human amino acid requirements.

Amino Acid Analysis of Food Proteins

Modern methods that involve acid or alkaline hydrolysis of the protein followed by separation and quantification of the released amino acids by ion exchange, gas-liquid or high-performance liquid chromatography, and other chemical and microbiological methods for specific amino acids, such as lysine, methionine, cysteine, and tryptophan, provide data with a repeatability within laboratory of approximately 5% and a reproducibility between laboratories of approximately 10%. Although several national and international food composition tables include amino acid contents of foods, it is preferable to use analytical results from a reliable laboratory owing to technical shortcomings in the preparation of some tables and to the considerable variability between the reported values, especially for tryptophan, cysteine, and methionine.

Amino acid data are usually calculated as milligrams amino acid per gram of protein. If they are reported as milligrams amino acid per gram of nitrogen, they can be converted to the protein equivalents multiplying by specific protein factors that range from 5.7 (17.5% nitrogen) to 6.4 (15.6% nitrogen) for the major protein sources in the diet. The factor used for a mixture of protein sources is 6.25, corresponding to a nitrogen content of 16%.

Table 2 Calculating the lysine (lys) content of a rice, lentil, and chicken mixture

1. Protein sources in 100g of the cooked mixture:

10 g dry polished rice 10 g dry lentil

20 g raw white chicken meat

2. Chemical composition:

10 g dry polished rice 10 g dry lentil

20 g raw white chicken meat

2. Chemical composition:




Protein (g per 100 g food)




Lysine (mg per 100 g food)




3. Lysine content of the mixture (mg per g protein):

Food mg lys per g protein g component per

100 g mixture

3. Lysine content of the mixture (mg per g protein):

Food mg lys per g protein g component per

100 g mixture

Chicken (l590/19.2) x 20

Weighted mean = (364 + 734 +1656)/(10 +10 + 20) = 60 mg lys per g protein

To calculate the amino acid content of a combination of food proteins, as in a processed food based on several protein sources or in a mixed diet, a weighted mean of the published or analytical results of each component should be used, as illustrated in Table 2.

Amino Acid Scoring Pattern

For infants younger than 1 year, the scoring pattern should be based on the amino acid composition of breast milk, even if some EAAs in human milk exceed minimum requirements for infants of this age. For example, infants consuming cow's milk proteins, which have less sulfur-containing amino acids than human milk, show adequate growth and nitrogen balance. Thus, although the use of a scoring pattern based on human milk composition may somewhat underestimate the protein quality of some foods for infants, there is consensus to accept errors on the side of safety for this highly vulnerable age group.

International expert committees (WHO, FAO, UNU, and CCVP) have agreed that the scoring pattern proposed in the 1980s for preschool children—based on studies of amino acid requirements at the Institute of Nutrition of Central America and Panama and on recommendations of protein intake by FAO, WHO, and UNU—is robust and represents the best available estimate of EAA requirements for this age group. Published nitrogen balance studies on older children and adults have experimental flaws, and only a limited

Table 3 Amino acid scoring patterns for infants under 1 year and for older children and adults (mg amino acid per g protein)3

Amino acid Infant Older children Egg, cow's milk,

Table 3 Amino acid scoring patterns for infants under 1 year and for older children and adults (mg amino acid per g protein)3

Amino acid Infant Older children Egg, cow's milk,

<1 year

and adults

and beef protein

















Methionine +





Phenylalanine +

















^Composition of animal proteins shown for comparison. ^Essentiality of histidine not clearly determined after 1 year of age. cCow's milk proteins have less sulfur-containing amino acids and tryptophan than human milk.

^Composition of animal proteins shown for comparison. ^Essentiality of histidine not clearly determined after 1 year of age. cCow's milk proteins have less sulfur-containing amino acids and tryptophan than human milk.

number of EAAs have been studied with amino acid oxidation techniques in adults. Since proteins with amounts of EAAs that satisfy the needs of young children will probably be adequate for older children and adults, the scoring pattern for preschool children is currently used for all after 1 year of age.

Table 3 shows the internationally accepted patterns for amino acid scoring applicable to infants and to persons after lyear of age; the composition of high-quality animal foods is shown for comparison. The content of each EAA in a food protein is evaluated relative to the age-specific scoring pattern, to determine the protein's amino acid score and to identify the limiting amino acids as shown in Table 1. All EAAs present in proportions that exceed requirements are assigned a fractional score of 1.00 (or a percentage score of 100%), even if mathematical calculation gives a higher value. The EAA with the lowest value (i.e., the most limiting amino acid) determines the protein's amino acid score.

The only EAAs that are likely to limit the protein quality of mixed diets for humans are lysine, the sulfur-containing amino acids (methionine and cysteine), threonine, and tryptophan. Consequently, when information on all EAAs is not available, protein quality can be estimated on the basis of its score for these four amino acids.

Correction for Protein Digestibility

A protein may have a good amino acid composition relative to the scoring pattern, but if it is not fully digested and its constituent amino acids are not absorbed, its capacity to provide nitrogen and EAAs for human function will diminish. Not all food proteins are digested, absorbed, and utilized to the same extent because of inherent differences in their source (e.g., inside vegetable cells with indigestible membranes), their physicochem-ical nature (e.g., protein configuration and amino acid binding), the presence of food constituents that modify digestion (e.g., dietary fiber, tannins, and other polyphenols), the presence of anti-physiological factors that interfere with protein breakdown (e.g., trypsin inhibitors and lectins), and processing conditions that alter the nature or release of amino acids (e.g., Maillard reaction and formation of polyamino acids and methylmercap-tan). Consequently, amino acid scores as predictors of protein quality must be adjusted for protein digestibility and amino acid availability.

The standard for obtaining digestibility data is through metabolic studies in humans, in which the nitrogen excreted in the feces is subtracted from the amount ingested with the diet and expressed as a percentage of intake. This apparent digestibility value must be corrected for the amount of fecal nitrogen excreted when a person is consuming a protein-free diet to calculate "true" digestibility (Table 1). Ethical constraints and practical complexities do not permit the determination of obligatory fecal nitrogen losses on a protein-free diet in all age and physiological groups. It is recommended that existing published values for daily obligatory fecal losses in preschool children (approximately 20 mg Nkg-1) and adults (approximately 14mgNkg_1) be used to correct apparent protein digestibility values.

Protein digestibility values of specific foods and well-defined diets may be taken from reliable published data. Table 4 shows some examples. When such data are not available for a mixed diet, a weighted average can be calculated from the true digestibilities of its constituent protein sources, as illustrated in Table 5. For new or novel products or processes, digestibility must be determined, preferably in humans. When cost and practicality do not permit performing metabolic studies in humans, standardized fecal balance methods in rats have been used. These methods have given true protein digestibility values of 93-100% for animal foods or food products (casein, beef salami, skim milk, tuna, and chicken sausage) and soya protein isolate; 86-92% for beef stew, chick peas, rolled oats, and whole-wheat cereal; and 70-85% for lentils and different types of beans. These value ranges are similar to those from human studies. Nevertheless, rat data must be used with caution for foods and diets that are known or suspected of being handled differently by the human and rat intestines. In vitro procedures have also been developed using combinations of trypsin, chymotrypsin, peptidase, and bacterial protease. Further research is needed to

Table 4 True protein digestibility of selected foods and diets

True protein digestibility (%)

Table 4 True protein digestibility of selected foods and diets

True protein digestibility (%)

Egg white


Whole egg, milk, beef, poultry, fish


Wheat, refined flour


Soya protein isolate


Polished rice


Soya flour


Wheat, whole


Maize products


Rice, whole




Mixed diets





Colombia, high-income


Phillippines, urban


Chile, middle class


Mexico, rural


Guatemala, rural


Brazil, rural


India, vegetarian


Table 5 Calculation of true digestibility of a mixed diet of rice,

beans, wheat, and egg

Diet True protein

Proportion of total protein


(g per 100 g protein in


whole diet)

Polished rice



Black beans



Whole wheat



Whole egg




(0.88 x

40) + (0.69 x 35) + (0.86 x 15) +

digestibility of


x 10) = 82%

whole diet

validate their use as predictors of protein digestibility in humans.

Calculations and Examples

The EAA composition and protein digestibility of the food or mixed diet being tested are determined. Then the percentage or fractional value of the most limiting EAA (noncorrected amino acid score) is multiplied by the percentage or fractional value of 'true' protein digestibility to obtain the corrected score, which is equivalent to protein quality. This value can be used as such or it can be expressed in relation to the corrected amino acid score of a reference protein or food, usually casein or an animal food (milk, egg, or beef).

Proteins that have no limiting amino acids are assigned an amino acid score of 100% (or 1.00) that must be only corrected for digestibility.

Table 6 Calculation of amino acid scores of single protein sources corrected for digestibility and in relation to the protein quality of cow's milk

Food Most limiting amino acid Noncorrected amino True protein Corrected amino Protein quality relative to acid score digestibility acid score milk

Polished rice Lysine 36 mg per g protein (36/58) x 100 = 62% x88% = 55% (55/95) x 100 = 58%

Egg white None >100! 100% x97% = 97% (97/95) x 100 = 102%

Similarly, if the clinical or experimental assessment of 'true' protein digestibility gives a value greater that i00% (generally due to experimental variability), a digestibility correction factor of i00% (or 1.00) is applied to the amino acid score. Table 6 shows examples of calculations for a single food as protein source. The same procedure can be used for food mixtures using a weighted average procedure based on the protein content, amino acid composition, and digestibility of the individual components. Table 7 shows an example of those calculations. For simplicity, the example uses only the four EAAs that are most often limiting.

Protein Concentration

Protein concentration or density (i.e., the amount of protein per unit of food) is another factor of a food's protein quality. Protein-dense foods are especially important for young infants, whose small gastric capacity limits the amount they can eat, and for elderly people with poor appetite. Evaluation of a food's protein concentration must be done for ready-to-eat preparations because food processing and cooking can result in significant changes relative to raw foods. Meats, poultry, and fish usually have a higher concentration of protein after cooking or frying, whereas vegetable food preparations contain more water and less protein than the raw products (Table 8).

Protein/Energy Ratio

The percentage of protein energy in the diet (P/E ratio) has been used to describe whether a diet provides adequate amounts of protein. The reasoning is that energy requirements are the main driving force for food intake. Therefore, a diet is adequate if it satisfies the requirements for all nutrients when it is eaten in amounts that will satisfy energy needs.

P/E ratio is calculated by dividing the amount of metabolizable energy derived from dietary protein (grams of protein x 16.7kJ or 4kcal) by the total amount of metabolizable energy in the diet, multiplied by 100 to avoid using fractional values. However, the use of P/E ratio as an index of food's protein adequacy may be misleading because it only gives information about protein concentration and does not indicate the biological value or quality of the proteins. Its usefulness improves when amino acid score is taken into account to calculate what can be defined as a desirable P/E ratio, as in the examples discussed later.

The P/E ratio indicates the amount of protein that the diet provides relative to energy and does not imply a constant relationship between protein and energy requirements. For example, the lower limit of the desirable P/E ratio of a diet with an amino acid score of 85% is 6.2 for a young child whose daily requirements are 16 g protein and 5.1 MJ energy ((16g x 16.7kJ/0.85)/ 5100 kJ x 100). For an adult male with daily requirements of 55 g protein and 12.8 MJ, the desirable P/E ratio is 8.4 ((55 x 16.7/0.85)/ 12,800 x 100).

Diets, especially those eaten by adults, often provide protein in amounts that surpass requirements, which elevates the P/E ratio. For example, almost all adult populations eat diets with P/E ratios between 10 and 15%. This is related to culture and food availability and does not reflect a biologically optimal ratio. Consistent with the calculations in the preceding paragraph, P/E ratios of 10 and 15 are adequate and it cannot be argued that one is nutritionally better than the other.

Keep Your Weight In Check During The Holidays

Keep Your Weight In Check During The Holidays

A time for giving and receiving, getting closer with the ones we love and marking the end of another year and all the eating also. We eat because the food is yummy and plentiful but we don't usually count calories at this time of year. This book will help you do just this.

Get My Free Ebook

Post a comment