Caballero B. (ed.) Encyclopaedia of Food Science, Food Technology and Nutrition. Ten-Volume Set

Подождите немного. Документ загружается.

result in 2-aminoacrylic acid (dehydroalanine) or

2-aminocrotonic acid (dehydroaminobutyric acid).

In the case of cystine, the eliminated thiolcysteine

can form a second dehydroalanine residue. Alterna-

tively, cleavage of the cystine disulfide bond can

occur by nucleophilic attack on sulfur, yielding dehy-

droalanine residues through thiol and sulfinate inter-

mediates. Intra- and interchain cross-linking of

proteins can occur through addition of amino and

thiol groups containing amino acid side-chains to

dehydroalanine residues. Ammonia may also react

via an addition reaction. Acid hydrolysis of such a

cross-linked protein yields the unusual amino acids

listed in Table 5. Ornithine is formed by cleavage of

arginine.

0024 Reaction of the imidazole ring of protein-bound

histidine with dehydroalanine residues results in

two histidinoalanine isomers, depending on the nitro-

gen that reacts. The two isomers, 3-N

- and 3-N

histidinoalanine, are formed in a constant ratio

of about 8:1. The histidinoalanine content of

commercial foods containing milk protein

(50–1800 mg kg

1

protein) is comparable with their

lysinoalanine content.

0025Histidinomethylalanine and lysinomethylalanine

do not contribute significantly to protein cross-

linking. Their demonstration failed in numerous

acid hydrolysates of various milk products. Model

studies showed that b-elimination of threonine takes

place much more slowly than that of serine and that

the reactivity of the resulting dehydroaminobutyric

acid towards nucleophiles is also much lower.

0026The formation of d-amino acids occurs through

abstraction and subsequent back addition of a proton

via the C

carbanion. For example, the extent of d-

amino acid formation is 30.1, 24.9, 18.8, 15.8, 8.0,

6.6, and 5.7% for aspartic acid, phenylalanine, glu-

tamic acid, alanine, leucine, valine, and proline, re-

spectively, when soya protein is exposed to 0.1 N

sodium hydroxide (pH * 12.5) at 65



C for 3 h. The

reaction with l-isoleucine is particularly interesting.

l-Isoleucine is isomerized to d-allo-isoleucine, which,

unlike other d-amino acids, is a diastereomer and

thus has a retention time different from l-isoleucine,

tbl0006 Table 6 Lysinoalanine content of various foods

Food Origin/treatment Lysinoalanine (mg kg

1

protein)

FrankfurterCP

,aspurchased0

Boiled 50

Oven-baked 170

Chicken thigh CP, raw 0

Oven-baked 110

Cooked in microwave oven 200

Egg white Fresh 0

Boiled (3 min) 140

(10 min) 270

(30 min) 370

Pan-fried (10 min at 150



C) 350

(30 min at 150



C) 1100

DriedeggwhiteCP160–1820

Condensed milk CP 360–540

Milk, evaporated CP 590–860

Skim milk, evaporated CP 520

Milk product for infants CP 150–640

Infant food CP <55–150

Simulated cheese CP 1070

Corn chips, pretzels, tortillas, etc. CP 170–500

Cocoa powder CP 130–190

Whipping agent CP 860–50 000

Soya protein isolate CP 0–370

Hydrolyzed vegetable protein CP 40–500

Sodium caseinate CP 45–6 900

Calcium caseinate CP 250–4 320

Yeast extract CP 120

Commercial product.

Variation range for different brand name products.

Data from Sternberg M and Kim CY (1977) Lysinoalanine formation in protein food ingredients. Advances in Experimental Medicine and Biology 86B: 73–84

and Haagsma N and Slump P (1978) Evaluation of lysinoalanine determinations in food proteins. Zeitschrift fu

r Lebensmittel-Untersuchung und -Forschung

167: 238–240.

PROTEIN/Interactions and Reactions Involved in Food Processing 4845

making its direct determination by amino acid

analysis possible.

0027 Heating proteins in a dry state at neutral pH results

in the formation of isopeptide bonds between the

e-amino groups of lysine residues and the b-org-

carboxamide groups of asparagine or glutamine resi-

dues. These isopeptide bonds are cleaved during acid

hydrolysis of protein and, therefore, do not contrib-

ute to the occurrence of unusual amino acids. But

their formation reduces protein digestibility. A more

drastic heat treatment of proteins in the presence of

water leads to more pronounced degradation. Alka-

line treatment of a protein isolate of sunflower seeds

shows that serine, threonine, arginine, and isoleucine

concentrations are markedly decreased with increas-

ing concentrations of sodium hydroxide. New amino

acids (ornithine and allo-isoleucine) are formed. Ini-

tially, the lysine concentration decreases but increases

at higher concentrations of alkali. Lysinoalanine

behaves in the opposite manner. The formation of

lysinoalanine is influenced not only by pH but also

by the protein source. A pronounced reaction occurs

in casein, even at pH 5.0, due to the presence of

phosphorylated serine residues, whereas noticeable

reactions occur in gluten from wheat or in zein from

corn only in the pH range of 8–11. Table 6 lists the

content of lysinoalanine in food products processed

industrially or prepared under ‘usual household con-

ditions.‘ The lysinoalanine content is obviously

affected by the food type and by the processing

conditions.

Reactions Under Oxidative Conditions

0028 Oxidative changes in proteins primarily involve me-

thionine, which forms methionine sulfoxide relatively

readily. After in vivo reduction to methionine,

protein-bound methionine sulfoxide is apparently

biologically available.

See also: Amines; Amino Acids: Properties and

Occurrence; Determination; Metabolism; Browning:

Nonenzymatic; Toxicology of Nonenzymatic Browning;

Carbohydrates: Classification and Properties;

Interactions with Other Food Components; Fatty Acids:

Properties; Flavor (Flavour) Compounds: Structures

and Characteristics; Mutagens; Oxidation of Food

Components; Peptides

Further Reading

Chen C, Pearson AM and Gray JI (1990) Meat mutagens.

Advances in Food and Nutrition Research 34: 387–449.

Eriksson C (ed.) (1981) Maillard Reactions in Food. Chem-

ical, Physiological and Technological Aspects. Oxford:

Pergamon Press.

Friedman M (ed.) (1991) Nutritional and Toxicological

Consequences of Food Processing. Advances in Experi-

mental Medicine and Biology 289. New York: Plenum

Press.

Fujimaki M, Namiki M and Kato H (eds) (1986) Amino-

carbonyl Reactions in Food and Biological Systems.

Amsterdam: Elsevier.

Gardner HW (1979) Lipid hydroperoxide reactivity with

proteins and amino acids: a review. Journal of Agricul-

tural and Food Chemistry 27: 220–229.

Haagsma N and Slump P (1978) Evaluation of lysino-

alanine determinations in food proteins. Zeitschrift fu

Lebensmittel-Untersuchung und -Forschung 167: 238–

240.

Henle T, Walter AW and Klostermeyer H (1993) Detection

and identification of the cross-linking amino acids N

and N

-(2

-amino-2

-carboxy-ethyl)-l-histidine (‘histidi-

noalanine,‘ HAL) in heated milk products. Zeitschrift

r Lebensmittel-Untersuchung und -Forschung 197:

114–117.

Henle T, Schwarzenbolz U and Klostermeyer H (1997)

Detection and quantification of pentosidine in foods.

Zeitschrift fu

r Lebensmittel-Untersuchung und -For-

schung A 204: 95–98.

Hofmann T (1998) Studies on melanoidin-type colorants

generated from the Maillard reaction of protein-bound

lysine and furan-2-carboxaldehyde – chemical character-

isation of a red coloured domaine. Zeitschrift fu

Lebensmittel-Untersuchung und -Forschung A 206:

251–258.

Hofmann T, Bors W and Stettmaier K (1999) Radical-

assisted melanoidin formation during thermal process-

ing of foods as well as under physiological conditions.

Journal of Agricultural and Food Chemistry 47: 391–

396.

Labuza TP, Reineccius GA, Monnier VM, O’Brien J and

Baynes JW (eds) (1994) Maillard Reactions in Chemis-

try, Food, and Health. Cambridge: Royal Society of

Chemistry.

Ledl F (1990) Chemical pathways of the Maillard reaction.

In: Finot PA, Aeschbacher HU, Hurrell RF and Liardon

R (eds) The Maillard Reaction in Food Processing,

Human Nutrition and Physiology, pp. 19–47. Basel:

Birkha

user.

O’Brien J, Nursten HE, Crabbe MJC and Ames JM (eds)

(1998) The Maillard Reaction in Foods and Medicine.

Cambridge: Royal Society of Chemistry.

Phillips RD and Finley JW (eds) (1989) Protein Quality and

the Effects of Processing. New York: Marcel Dekker.

Sternberg M and Kim CY (1977) Lysinoalanine formation

in protein food ingredients. Advances in Experimental

Medicine and Biology 86B: 73–84.

Waller GR and Feather MS (eds) (1983) The Maillard

Reaction in Foods and Nutrition. ACS Symposium

Series 215. Washington, DC: American Chemical

Society.

4846 PROTEIN/Interactions and Reactions Involved in Food Processing

Walter AW, Henle T, Haeßner R and Klostermeyer H (1994)

Studies on the formation of lysinomethylalanine and

histidinomethylalanine in milk products. Zeitschrift fu

Lebensmittel-Unterschung und -Forschung 199: 243–

247.

Whitaker JR and Fujimaki M (eds) (1980) Chemical Deteri-

oration of Proteins. ACS Symposium Series 123. Wash-

ington, DC: American Chemical Society.

Quality

G S Gilani and N Lee, Bureau of Nutritional Sciences,

Health Canada, Ottawa, Ontario, Canada

Background

0001 The quality of a dietary protein is determined by the

pattern and concentration of indispensable or es-

sential amino acids, the digestibility of the protein,

and the bioavailability of its amino acids. To satisfy

the protein requirements of the body, the diet must

supply both enough of the indispensable amino acids

(IAA) in required proportions and enough of the

total amino nitrogen for synthesis of dispensable or

nonessential amino acids. These are needed to sup-

port the synthesis of body protein and the production

of other nitrogen-containing compounds such as hor-

mones and neurotransmitters involved in a range of

physiological functions.

0002 An IAA is defined as an amino acid that cannot

be synthesized in the body, or at least not in adequ-

ate amounts. Nine amino acids – histidine, isoleucine,

leucine, lysine, methionine, phenylalanine, threonine,

tryptophan, and valine – are not synthesized in ad-

equate amounts by mammals and are, therefore, IAA

for humans. Dispensable amino acids (alanine, argin-

ine, aspartic acid, cyst(e)ine, glutamic acid, glycine,

hydroxyproline, proline, serine, and tyrosine) are not

essential in the diet, as they can be synthesized from

IAA and amino nitrogen. Since cystine can replace

part of the requirement for methionine, and tyrosine

a part of the requirement for phenylalanine, cystine

and tyrosine are also included when considering IAA

contents of diets, and these pairs are then expressed

as total sulfur amino acids (methionine þ cystine)

and total aromatic amino acids (phenyla-

lanine þtyrosine). Most dietary proteins contain a

mixture of all 20 common amino acids in varying

proportions.

0003The reported quality of a protein may vary

depending on the function measured, for instance

growth versus maintenance, because these differ in

the amount and pattern of IAA they require. In all

cases, however, the quality of a protein source is

limited by the IAA available in least amounts (the

first limiting amino acid) relative to requirements.

Occasionally, a protein source may be colimiting in

more than one IAA. There are several different ways

that have been developed to assess how well a protein

could be expected to meet human protein and amino

acid requirements.

Methods of Assessing Protein Quality of

Foods

0004The only true measurement of protein quality for

human use is an evaluation, in humans, of growth,

nitrogen balance, or metabolic balance or some other

appropriate test, preferably performed with suitable

subjects from the target population of interest. Such

evaluations directly reflect how well the food or food

product meets human needs and this is a function of

the proportion and content of IAA, the digestibility

of the protein, and the bioavailability of amino acids

in the food or food product.

0005Conventional or ‘long-term‘ nitrogen-balance ex-

periments which measure dietary nitrogen retained in

the body (nitrogen intake  fecal nitrogen output 

urinary nitrogen output) in humans are often con-

sidered to be the standard or reference method for

assessing protein quality of foods. Nitrogen balance is

used as a surrogate for protein balance and is assumed

to reflect how well protein intake is maintaining body

protein homeostasis and organ function. Nitrogen

balance can be expressed in two ways: one relates

nitrogen intake to nitrogen retained, as measured by

the net protein utilization (NPU) method, and the

second relates nitrogen absorbed to nitrogen retained,

as measured by the biological value (BV) (see Table 1).

0006Nitrogen-balance studies are used to determine the

minimum intake of a test or food protein that will

maintain nitrogen balance. The amount of a protein

required to maintain nitrogen balance is a measure of

its quality. Therefore, the accepted nitrogen balance

technique to measure protein quality in humans is to

feed graded levels of a test protein for 10–21 days.

Shorter-term human nitrogen balance assays or those

based on single test levels can provide useful infor-

mation, but at this time, it is not clear without further

validation and critical evaluation whether these

methods have the capacity to predict, in quantitative

terms, the nutritional value of a source of protein.

0007More recently, some research has focused on pro-

tein and nitrogen metabolism during different periods

PROTEIN/Quality 4847

tbl0001 Table 1 Comparison of features of rat assays for protein quality evaluation

Credits protein used in both growth

andmaintenance

Description/equation Comments

Growth methods

PER (standardized) (protein

efficiency ratio)

No, growth only; no protein-free

diet group included

Weight gain of rats on test or casein control diets

per gram of protein consumed, standardized to

casein PER of 2.5

Simple but relatively costly and lacking other attributes

including proportionality to protein quality, accuracy,

precision, and reproducibility

NPR (net protein ratio) Yes, protein-free diet group

included

Weight gain on test or control diet plus weight loss

on nonprotein diet per gram of protein consumed

Overcomes the major weakness of PER, but the assumption is

that the quality of protein required to prevent weight loss is

equivalent to that needed for maintenance; usually

conducted in 2 weeks, half the time required for

standardized PER; comparable to NPU (below) except

estimated from body weight change, not nitrogen balance

RNPR (relative net protein

ratio)

Yes, protein-free diet group

included

NPR for test protein divided by NPR for reference

protein expressed as a fraction of 100

Values are proportional with respect to protein quality within

reasonable limits; more accurate and reproducible and

cheaper than PER

NU (nitrogen utilization) Yes, no protein-free diet group

included; factor of maintenance

calculated

Body weight gain þ0.1 (initial þfinal body weight)

divided by nitrogen consumed

Factor of maintenance is calculated instead of measured

RNU (relative nitrogen

utilization)

Yes, no protein-free diet group

included; factor of maintenance

calculated

100 NU (test protein)/NU(reference protein) RNU gives almost the same results as RNPR

Nitrogen balance methods

NPU (net protein utilization) Yes, protein-free diet group

included

The proportion of ingested protein (N6.25) that is

retained for maintenance and/or growth

Relates N intake to N retained; this index includes the factor of

digestibility; nitrogen retained can be measured either by

the balance method or by carcass analysis; nitrogen

balance is used as a surrogate for protein balance and is

assumed to reflect how well protein intake is maintaining

body protein homeostasis and organ function; NPU is the

product of BV and TPD

Equation: 100 (N retained on protein diet  N

retained on protein-free diet/g N ingested on

protein diet)

Note: N retained ¼N intake  fecal N output 

urinary N output

BV (biological value) Yes, protein-free diet group

included

The proportion of nitrogen absorbed that is

retained for maintenance and/or growth

purposes

Relates N absorbed to N retained; thus, as an index, it

excludes the factor of digestibility

Equation: 100 (N retained on protein diet  N

retained on protein-free diet)/N ingested  (fecal

N on protein diet  fecal N on protein-free diet)

TPD (true protein

digestibility coefficient)

Yes, protein-free diet group

included

The proportion of ingested protein (N6.25) that is

absorbed

This is not a measure of protein quality, only digestibility; it is

needed for calculating protein digestibility-corrected amino

acid score; this test requires an adaptation period of 4 days

and a fecal collection period of 5 days

Equation: 100 (N ingested on protein diet  (fecal

N on protein diet  fecal N on nitrogen-free

diet) )/N ingested on protein diet

of the day in relation to eating and fasting. The meta-

bolic fate and retention of dietary nitrogen during

the postprandial phase may be of particular import-

ance in determining the quality of a given protein. An

experimental approach, termed ‘net postprandial

protein utilization,‘ measures the efficiency of post-

prandial nitrogen retention in humans after the inges-

tion of different protein sources and appears to be

able to differentiate between proteins of similar high

quality. Postprandial nitrogen retention may offer a

sensitive in-vivo method to discriminate between

protein sources of differing quality. It is too early,

however, to say whether this will become a routine

method of protein-quality evaluation.

0008 Although they are the gold standard, human stud-

ies cannot be carried out on a routine basis for

reasons of cost and ethics. Therefore, in-vitro and

animal assay techniques have been developed that

correlate well with data from human experimentation

for each food product.

Routine Protein-quality Evaluation

Methods

0009 Criteria for a valid and useful assay for routine

protein-quality evaluation include accuracy, preci-

sion, reproducibility, proportionality to protein qual-

ity, and low cost. The ability of assays to accurately

measure protein quality depends on a number of

factors including whether they account for protein

requirements for growth alone or for growth plus

maintenance.

Chemical and Microbiological Assays

0010 The chemical assays are the amino acid scoring

methods which involve chemical analysis of the

amino acid composition of the protein of interest

and comparison of its IAA profile to that of a refer-

ence profile. As a routine assay, this method has been

standardized as the protein digestibility-corrected

amino acid score (PDCAAS), which will be discussed

later. Microbiological, or in-vitro, assays use micro-

organisms that require the same amino acids as

humans. The most successful of these uses the proto-

zoans, Tetrahymena pyriformis or T. thermophila,

which require the same 10 IAA as the growing rat

and human. The use of this method is limited mainly

by the fact that the growth of T. pyriformis can be

inhibited by factors in the food unrelated to protein

quality, including several food additives and spices.

Rat Assays

0011 The rat is the most common species used in a protein-

quality bioassay. Rat assays are based on the

measurement of either growth or nitrogen balance

in young animals in response to feeding defined

amounts of one or more test proteins or a control

protein (see also Table 1).

0012The protein efficiency ratio (PER) was the first

method adopted for routine assessment of protein

quality of foods. The PER is a standardized method

in which a test diet and a casein control diet, both

containing 10% protein (N 6.25) are fed to wean-

ling rats for a period of 4 weeks. The PER values are

calculated by dividing the weight gain of the rats by

the amount of protein consumed over this period. To

compare PER values of diets determined by different

laboratories, the PER values are corrected to an as-

sumed value of 2.5 for casein control. It has been

recognized that the PER method, although simple,

does not meet any other criteria of a valid bioassay,

as noted above. The failure of the PER assay to

properly credit protein used for maintenance

purposes is the chief criticism of the method. The

poorer the quality of dietary protein, the larger

is the error that is introduced because of this

failure to make allowance for maintenance. For this

reason, PER values are not proportional to protein

quality, i.e., a PER of 2.0 is not twice as good as a

PER of 1.0.

0013The net protein ratio (NPR) method overcomes the

major weakness in the PER assay by adding the

weight loss of a group of rats fed a zero-protein diet

to the weight gain of the rats fed the test protein.

However, it has to be assumed that the protein re-

quired to prevent weight loss of rats fed the nonpro-

tein diet is equivalent to the protein needed for

maintenance. As normally carried out, NPR values

are uncorrected to any standard. However, relative

NPR (RNPR), where the NPR of a test protein is

expressed relative to a value of 100 for the NPR of

a reference protein, is expressed on a scale of 1–100.

NPU and BV, two nitrogen-balance methods men-

tioned earlier, are also expressed on a scale of 1–100.

0014The relative net protein ratio (RNPR) method at

2 weeks is shorter than the standard PER method

(4 weeks) and therefore is less expensive. Also,

RNPR values (unlike PER values) are proportional

to each other in protein quality within reasonable

limits. RNPR is more accurate and reproducible

than PER.

0015The nitrogen utilization (NU) method is a modifi-

cation of NPR, the factor of maintenance being cal-

culated instead of measured:

NU ¼body weight gain of test group

þ0:1 ðinitial þ final body weightÞ

=nitrogen consumed by test group:

PROTEIN/Quality 4849

The factor, 0.1, derives from experimental observa-

tions of the growth of rats on protein-free diets. The

relative NU (NU of test protein/NU of reference

protein 100) gives essentially the same values as

RNPR. For example: lactalbumin, RNPR of 87 vs.

RNU of 87; for egg white, RNPR of 97 vs. RNU of 97;

for wheat gluten, RNPR of 32 vs. RNU of 31; for

wheat gluten þsoy protein, RNPR of 64 vs. RNU of

64. The RNU method eliminates the feeding of a

‘zero‘ protein diet to a group of rats; thus, less labor

is required for the RNU procedure than for the RNPR

procedure.

Limitations of Rat Assays

0016 It is well known that the requirements of rats for

sulfur amino acids are much larger than those of

humans. Thus, a limitation of any rat growth assay

is that it will underestimate the protein quality of any

protein limiting in sulfur amino acids such as soybean

protein, peanuts, and other legumes or pulses. (See

Protein: Sources of Food-grade Protein.)

Protein Digestibility-corrected Amino

Acid Score (PDCAAS) Method

0017 A method based on a comparison of the amino acid

content of food with human requirements (amino

acid scoring system) is considered internationally to

be the most suitable approach for routine assessment

of the protein quality of foods. The amino acid

score should be corrected for incomplete digestibility

of protein, and for the unavailability of individual

amino acids, especially those that are susceptible to

damage by processing treatments.

0018 The FAO/WHO has recommended the PDCAAS as

the preferred method for routine prediction of protein

quality of properly processed and highly digestible

food products for human nutrition. In other words,

the foods should contain minimal amounts of residual

antinutritional factors, and the digestibility of the

protein should be a good approximation of the bio-

availability of individual amino acids.

Analyses Required for the Determination of

PDCAAS

0019 Proximate composition Levels of moisture, total

nitrogen, fat, and crude protein are determined

according to standardized AOAC methods. Total

protein is then calculated by using a nitrogen-

to-protein conversion factor of 6.25. When the true

nitrogen-to-protein conversion factor is known for a

given protein, this should be used.

0020 Amino acid profile Amino acids in foods have com-

monly been analyzed by ion-exchange chromatog-

raphy of protein hydrolysates. Hydrolysis with 6 M

HCl at 110



C for 24 h is routinely used for all amino

acids except tryptophan and sulfur amino acids.

Since sulfur amino acids and tryptophan are des-

troyed to varying degrees by 6 M HCl hydrolysis, a

performic acid followed by the HCl hydrolysis and

an alkaline (4.2 M NaOH) hydrolysis are widely used

for the accurate determination of sulfur amino acids

(methionine as methionine sulfone and cysteine/cyst-

ine as cysteic acid) and tryptophan, respectively.

Amino acids may also be determined accurately and

more rapidly by high-performance liquid

chromatography methods using precolumn phenyli-

sothiocyanate-derivatized amino acids. Amino acids

should be expressed on the basis of mg per gram of

protein.

0021Amino acid score Amino acid ratios (milligrams of

an IAA in 1.0 g of test protein per milligram of the

same IAA in 1.0 g of reference protein 100) for nine

IAA are calculated by using the FAO/WHO’s sug-

gested pattern of amino acid requirements as the

reference protein for all ages except for infants. This

pattern contains, in milligrams per gram of protein:

histidine, 19; isoleucine, 28; leucine, 66; lysine, 58;

methionine þcystine, 25; phenylalanine þtyrosine,

63; threonine, 34; tryptophan, 11; and valine, 35.

For infants under 1 year of age, the amino acid com-

position of human milk is used as the basis for

scoring: (milligrams per gram of protein): histidine,

26; isoleucine, 46; leucine, 93; lysine, 66; methioni-

ne þcystine, 42; phenylalanine þtyrosine, 72; threo-

nine, 43; tryptophan, 17; and valine, 55. The lowest

amino acid ratio is termed the amino acid score.

0022Protein digestibility The classic method for deter-

mining the digestibility of proteins and amino acids

has been the balance method, in which the protein or

amino acid excreted in the feces is subtracted from the

amount ingested and the value expressed as a percent-

age of protein intake. This calculation gives an appar-

ent digestibility value. To determine true digestibility,

it is necessary to correct for the amount of fecal

protein or amino acid excreted when the subject or

animal is consuming either a protein-free diet or a diet

with just enough of a highly digestible protein to

prevent excessive loss of body protein. Therefore,

the true digestibility of protein and amino acids can

be calculated by the following equations:

True protein digestibility

¼½PI ðFP  MFPÞ=PI100,

where PI is the protein intake, FP is fecal protein, and

MFP is metabolic fecal protein.

4850 PROTEIN/Quality

True amino acid digestibility

¼½AAI ðFAA  MFAAÞ=AAI100,

where AAI is the amino acid intake, FAA is fecal

amino acid, and MFAA is metabolic fecal amino acid.

0023 The amounts of protein and amino acids in the

feces of rats fed the protein-free diet are used as the

estimates for MFP and MFAA, respectively.

0024 Since true digestibility measurements take into ac-

count the fecal protein or amino acids which are not

of dietary origin, the true digestibility of a protein or

amino is always higher than the apparent digestibility.

Unlike apparent digestibility, the true protein digest-

ibility of a food is not affected by the dietary condi-

tions under which the food is fed to the animal such as

the protein content of the test diet. It is a more accur-

ate measure of amino acids that are absorbed from

the gut and therefore gives a better representation of

protein quality than apparent digestibility.

0025 Protein/amino acid digestibility is most frequently

determined using rats. The rat balance method is well

established, and the procedure for the determination

of true protein digestibility has been standardized. It

has been recommended by the FAO/WHO that when

human balance studies cannot be used, the standard-

ized rat fecal-balance method should be used. The

true digestibility of protein is considered to be a rea-

sonable approximation of the true digestibility of

most amino acids (as determined by the rat balance

method) in mixed diets. Therefore, the FAO/WHO

has recommended that amino acid scores be corrected

only for true digestibility of protein.

Protein Digestibility-corrected Amino Acid Score

(PDCAAS)

0026 The PDCAAS of a test food is calculated as follows:

True protein digestibility coefficient

amino acid score:

PDCAAS above 100 are considered as 100.

Protein Digestibility of Some Diets and Common

Foods

0027 Values for the true digestibility of protein in diets

from India (54–75%), Guatemala (77%), and Brazil

(78%) are markedly lower than the values in North

American diets, including vegetarian diets, 88–94%,

suggesting that protein digestibility is of greater con-

cern in diets of some developing countries. The poor

digestibility of protein in the diets of some developing

countries is due to the use of less-refined cereals and

pulses (such as beans and lentils). Low true digestibil-

ity values (63–65%) have also been reported in stud-

ies with children fed millet and ragi-based diets in

India.

0028A comparison of the true digestibility of protein in

some common foods for human adults has indicated

that animal protein sources (meat, fish, poultry, eggs,

properly processed milk protein products) and flours

and breads made from low-fiber wheats, wheat

gluten, farina, peanuts, and soy-protein isolates have

high true protein digestibilities of 94–99%. Whole

corn (except high amylose containing opaque-2) and

flour or bread of high-fiber wheat, polished rice, oat-

meal, triticale, cottonseed, soy flour, and sunflower

have intermediate true protein digestibilities of over

85%. Ready-to-eat cereals (corn, wheat, rice, or oat)

have lower true protein digestibilities of 70–77%,

probably due to the high heat involved in their

processing. Various types of dry beans (Phaseolus

vulgaris) and millet also have low true protein digest-

ibilities of 75–79%.

PDCAAS for Some Common Foods or Food

Products

0029The availability of amino acid composition and pro-

tein digestibility data for a number of foods has made

it possible to calculate their PDCAAS. Animal protein

products such as egg white powder, casein, ground

beef, beef salami, skim milk powder, tuna and

chicken frankfurters have high PDCAAS of about

100 (97–100). Soybean protein products also have

high PDCAAS (89–99) and are marginally deficient

in sulfur amino acids in some cases. The scores for

various types of beans, lentils, and peas range from 47

to 71, and these products are first limiting in sulfur

amino acids for human nutrition. Wheat gluten and

sunflower-protein isolate are severely limiting in

lysine and have PDCAAS of 25 and 37, respectively.

Effects of Supplementation and

Complementation on Protein Quality

0030The low protein quality of a vegetable protein source

can be improved by the addition of supplementary

protein or a limiting amino acid, or by protein

complementation. Protein supplementation means

to increase the protein quality of a food having a

low-quality protein by the addition of a moderate

amount of another food having a high content of the

indispensable amino acid which is limiting in the low-

quality protein. For example, egg, milk, meat, and

fish proteins are rich in lysine and sulfur amino

acids, and so they would be supplementary for lysine

deficient proteins such as cereals, and for proteins

deficient in sulfur amino acids such as beans, peas,

and lentils. These animal proteins can also fully sat-

isfy all indispensable amino acid requirements for

adults when ingested as the sole protein source at a

level of 0.7 g per kilogram of body weight per day.

PROTEIN/Quality 4851

0031 Protein complementation means that the quality of

two proteins is higher than either of the two compon-

ents. Proteins which complement each other have

different limiting amino acids. When blended in the

right proportions, cereals and legumes (beans and

peas) complement each other because cereals are

limiting in lysine, and legumes are deficient in sulfur

amino acids. As an example, the BV of pea flour is

43% and is limited by methionine, with a surplus of

lysine. Maize meal protein with a BV of 35% is

limited by lysine but with a surplus of sulfur amino

acids. The average of the BV of these two protein

sources is 39%. In practice, however, the BV of an

equal mixture of these two proteins sources is found

to be 70% due to the complementary effects. In the

same way, traditional combinations of vegetable pro-

teins consumed in some countries, such as rice–chick

peas in Asia, wheat–bean in the Near East, and

maize–bean in the Americas, have a good protein

quality.

Applications of Protein-quality

Measurement

0032 There are two distinct purposes for measuring protein

quality: to determine whether a food or diet will meet

human amino acid needs, and for regulatory purposes

to ensure that foods are of the appropriate protein

quality for the intended use and that protein claims

are valid.

Protein-quality Measurement of Mixed Diets

0033 It is recognized that the consumption of mixed diets

in the developed world poses no problem of inad-

equacy in protein content and quality. However, the

food supply is undergoing many changes, which may

significantly alter the types and quality of protein

available. Many new foods are being produced for

various purposes through genetic modification. In-

creasing numbers of foods consumed in the West are

processed, and new methods of processing are con-

stantly being developed. Products of the plant and

animal world that have not been previously used as

human food are being developed for that purpose.

These situations point to a continuing need for sur-

veillance of potential effects on protein quality of

foods in the diet, if not the diet as a whole.

0034 Some subgroups in developed countries consume

protein from a single source, rely on a small variety of

foods, or eat foods with proteins of inferior quality.

To the extent that these people do not have control

over the source of the proteins in their foods because

they may be relying on commercially formulated

products, they must be considered at risk, and the

quality of their dietary proteins must be known.

Examples of such groups are infants and young

children relying on commercial formulas; people

using formulated products to lower their energy/

food intake, generally for weight reduction; people

changing their eating patterns to become vegetarian,

for example, and possibly relying on prepackaged

substitute foods of lower protein quantity and qual-

ity; and individuals with various illnesses requiring

therapeutic formulated diets. There are also subpo-

pulations who, for economic or other sociocultural

reasons, might become at risk because of their de-

creased ability to procure high-quality proteins or

because of making uninformed dietary choices.

These might include the poor, the elderly, teenagers

who are undergoing rapid growth and often eating

unbalanced diets, and pregnant and lactating women,

especially the growing number of teenage mothers.

Depending on the issue of concern, either protein-

quality assessment or protein-status assessment will

have an application.

Regulatory Use of Protein-quality Assessment of

Foods for Human Nutrition

0035Methods currently used by regulatory bodies Since

the adoption of the PDCAAS method by the FAO/

WHO, this method has been widely accepted. For

example, in the USA, PDCAAS has replaced PER as

the official method of protein-quality assessment, in-

cluding for protein declaration in nutritional labeling

of all foods except infant formulas. However, PER

remains the official regulatory biological method for

protein-quality assessment of infant formulas sold in

USA. PER also remains the official method for evalu-

ating protein claims for foods sold in Canada. The

chemical index (amino acid score using human

breast milk as reference protein) is used for assessing

the protein quality of infant formulas in the UK. The

Codex Alimentarius Commission requires that the

quality of protein in an infant formula not be less

than 85% of that of casein as determined by the

PER method. The EC requires that, for an equal

energy value, an infant formula must contain an

available quantity of each essential and semiessential

amino acid at least equal to that contained in the

reference protein (breast milk), but for calculation

purposes, the concentrations of methionine and cyst-

ine may be added together. For infant formulas based

on partial protein hydrolysates, the EC has an add-

itional requirement, specifying that the PER and the

NPU must be at least equal to those of casein.

0036Limitations of the PDCAAS As adopted by the

FAO/WHO, the PDCAAS method is truncated in

that it requires the adjustment of all values that

4852 PROTEIN/Quality

exceed 100 down to 100. This means that, unlike

biological assays such as PER, the PDCAAS does

not recognize the additional value of high-quality

animal proteins (see Protein: Sources of Food-grade

Protein) and their potential for acting as supplements

to low-quality proteins.

0037 Since the PDCAAS method was adopted, it has

been found that there are differences in the metabolic

fate and retention of dietary nitrogen from different

protein sources when measured acutely during the

postprandial period in humans. One method, called

‘net postprandial protein utilization,‘ is designed to

measure the efficiency of postprandial nitrogen reten-

tion following consumption of various dietary pro-

teins. Results obtained so far have placed several food

protein sources in the order: milk > soy > pea > wheat,

with values of 120 (sulfur amino acids), 99 (sulfur

amino acids), 73 (sulfur amino acids), and 36 (lysine),

respectively. This order is the same as that of the

nontruncated PDCAAS values for these sources, sug-

gesting that there may be valid reasons to use the

nontruncated PDCAAS values. The official PDCAAS

method puts both milk and soy at 100.

0038 PDCAAS includes a rat-feeding study, but this

only measures protein digestibility. Thus, it may not

adequately reflect the adverse effect on the bioavail-

ability and utilization of amino acids caused by nat-

urally occurring growth-depressing factors such as

glucosinolates and isothiocyanate in mustard flour,

trypsin inhibitors, and hemagglutinins in soy protein

and beans (Phaseolus spp.), or antinutritional factors

formed during processing such as lysinoalanine in

alkaline-/heat-processed lactalbumin and soy protein.

As a result, the PDCAAS method would overestimate

the protein quality of such sources. In the absence of

upper safe limits for these antinutritional factors, the

use of the PDCAAS method may be inappropriate for

predicting the protein quality of such protein prod-

ucts and any foods where these products may be the

sole source of protein, such as infant formulas and

enteral nutritionals.

0039 The PDCAAS method assumes complete biological

efficiency of supplemental amino acids in improving

protein quality which may not be true, especially

in the case of low-quality proteins. Markedly lower

relative protein efficiency ratio or RNPR values

have been seen with amino acid-supplemented zein

(3–44%) compared to the PDCAAS value (71%),

which suggests incomplete biological efficiency of

the supplemental amino acids and overestimation of

protein quality by the PDCAAS method. The crystal-

line amino acid added to a low-quality protein that

also has poor digestibility may be absorbed more

quickly than the amino acids from the protein, so

protein synthesis does not take place.

0040The PDCAAS method does not accurately predict

the quality of proteins colimiting in more than one

IAA such as soy when used in infant formulas. The

amino acid requirements for infants are not the same

as the FAO/WHO’s suggested pattern of amino acid

requirements used in PDCAAS. Since the identifica-

tion of the true first limiting amino acid would be

difficult in cases where the protein is colimiting in

more than one IAA, an improvement in protein qual-

ity with supplementary amino acid(s) should not be

assumed without biological testing.

0041Recommended method for regulatory purposes The

PDCAAS is the most appropriate routine method for

properly processed food where protein digestibility is

a good approximation of bioavailability of the amino

acids and where there is known to be no or only low

levels of antinutrient factors. Where these conditions

do not apply, a biological method should be used,

whether this is PER, RNPR, or relative nitrogen util-

ization (RNU). Probably the least expensive and yet

still valid and reliable method, and one that gives

results that are proportional across a range of protein

qualities, is the RNU method. The RNU method gives

essentially the same values as the better known RNPR

rat assay for routine protein-quality assessment of

foods except that, as described in the section Rat

Assays, it replaces measurement of protein needed

for maintenance with a factor for maintenance. The

elimination of the need for feeding a ‘zero‘ protein

diet to a group of rats makes the RNU method less

costly than the RNPR method.

See also: Protein: Sources of Food-grade Protein