Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

2. Analytical methods for the determination of nitrogen and amino acids in foods

Protein and Other Nitrogen Components of Foodstuffs
Analyses of Individual Amino Acids in Foods
Estimation of "Available" Amino Acids in Foods 1

Protein and Other Nitrogen Components of Foodstuffs

Nitrogen in foods not only comes from amino acids in protein, but also exists in additional forms that may or may not be used as a part of the total nitrogen economy of humans and animals. The nitrogen content of proteins in foods can vary between 150 and 180 g/kg 115-18 per cent), depending on the amino acids they comprise. In addition, purines, pyrimidines, free amino acids, vitamins, creatine, creatinine, and amino sugars can all contribute to the total nitrogen present. In meat, a portion of the nitrogen occurs as free amino acids and peptides; fish may contain these and volatilebase nitrogen and methyl-amino compounds 11). Marine elasmobranchs may also contain urea. Half of the nitrogen of the potato may not be in the form of protein (2), and even in human milk as much as 50 per cent of the total nitrogen may be urea nitrogen (3). Because the nutritional significance of much of the non-amino acid and non-peptide nitrogen is unclear, nitrogen analysis of a food is usually much more precise than the nutritional significance that can be attached to it.

In practice, most biological methods for evaluating protein quality (chaps. 4 and 5) are, in fact, evaluating nitrogen but are expressed as crude protein (N x 6.25). Nitrogen data are also used for amino acid scores (chap. 3) when amino acids are expressed in terms of mg/g N. When results are to be expressed in terms of protein, as for example in protein efficiency ratio (PER) and when amino acid scores are to be expressed per 16 9 N, then the average conversion factor of 6.25, defined as crude protein, is again used (4). Since requirements are also expressed in terms of N x 6.25, conversion factors are not needed and no confusion should exist.

For other purposes, however, such as labelling regulations and food composition data, conversion from nitrogen to protein is wide)y used and a range of conversion factors exists.

Most food composition tables derive estimates of protein content by applying

TABLE 3. Factors Used in Converting Nitrogen to Protein

Foodstuffs Conversion factor for protein content as reported in food Correction factor for conversion of reported value to
composition tables "crude protein"  
Wheat (hard, medium, or soft)    
whole 5.83 1.07
flour (medium or low extraction) 5.70 1.10
macaroni, spaghetti, wheat pastes 5.70 1.10
bran 6.31 0.99
Rice (all varieties) 5.95 1.05
Rye, barley, and oats 5.83 1.07
Pulses, nuts, seeds    
Groundnuts 5.46 1.14
Soya 5.71 1.09
Tree nuts    
almonds 5.18 1.21
Brazil nuts 5.46 1.14
coconut, chestnuts 5.30 1.18
Seeds - sesame, safflower, sunflower 5.30 1.18
Milk (all species) and cheese 6.38 0.98
Other foods 6.25 1.00

From FAO Nutritional Studies No. 24, "Amino Acid Content of Foods and Biological Data on Proteins" (FAO, Rome, 1970). different factors to the nitrogen content of individual foods. Some of these factors are shown in table 3. These are mostly standard factors; however, some of the values for cereals, legume foods, and oilseed meals reported by Tkachuk (5) are based on direct calculation from amino acid analyses.

To compare the reported protein content of foods with protein requirements, a correction of the reported protein content must be made. The correction factors to convert reported protein to crude protein are also shown in table 3. Considerable care is needed when protein data from food composition tables are used in conjunction with determined protein quality and/or requirement values. Factors different from those in table 3 may have been used, and it is necessary to ascertain how the data were derived.

Some new products such as single-cell protein (SCP) contain high levels of purine nitrogen, some of which may be used, and cell-wall nitrogen, most of which is probably not utilized (6). Recommendations have been made by the Protein-Calorie Advisory Group (PAG) (7) for the calculation of protein nitrogen in SCP products where substantial parts of the total nitrogen may come from nucleic acid. The calculation of "crude protein" by multiplying total nitrogen by 6.25 can give a serious overestimation of protein content. Purine nitrogen should be determined separately and the nucleic acid content estimated by multiplication by a factor to make allowance for the pyrimidine content. As the ratio of the nitrogen in pyrimidine to that in purines approximates 0.40, and both are present in equimolecular amounts in most nucleic acids, the purine nitrogen should be multiplied by 1.4 to obtain nucleic acid nitrogen. Nucleic acid nitrogen multiplied by 9.0 can then be considered as nucleic acids. Consideration of purine nitrogen alone will give an underestimation for nucleic acid nitrogen.

An example of the calculation of protein nitrogen in an SCP product is given below:


Total N content of a yeast SCP product = 1,000 mg

Purine N =160 mg


Total nucleic acid N = 160 x 1.4 = 224 mg

(allowance made for pyrimidine N)

Corrected protein N = 1,000 - 224 =776 mg

Crude protein content = 1,000 x 6.25 = 6,250 mg

Corrected protein content = 776 x 6.25 = 4,850 mg

Total nucleic acids = 224 x 9.0 = 2,016 mg

For certain foodstuffs, total nitrogen values have sometimes been partitioned into "true protein" and "nonprotein nitrogen." This partitioning has been according to the quantities of nitrogen that were recovered in the precipitate and filtrate, respectively, after extraction of the food with various protein solvents followed by precipitation with trichloroacetic acid or some other protein precipitant. This approach is not recommended, because free amino acid nitrogen may be of the same nutritional value as the protein. Amino acid analyses are now usually feasible for the expression of total amino acids in food. Thus the conventional measure of "protein" or "crude protein" in foods is N x 6.25, and it is recommended that this one factor be used in nutritional studies in which whole diets contain more than one source of nitrogen.

The protein content of foodstuffs is conventionally estimated from the nitrogen content determined by the Kjeldahl technique. Numerous modifications of the original procedure have been proposed (8). A recommended method based on the procedure of the Association of Official Analytical Chemists (AOAC) (9) is fully described in chapter 8 (p. 86).

When the equipment is available, the determination of ammonia in digests may be carried out on an autoanalyser system using a colorimetric method based on reaction with an alkaline phenolate-hypochlorite reagent. This method has proved reliable and can save time. Earlier hopes that food samples could also be rapidly digested with complete recovery in an autoanalyser system have not been supported, and it is still necessary to digest samples prior to autoanalysis.

Other methods, such as those using Biuret and the Folin-Ciocalteu reagent, and fluorimetry have been reviewed by Cole (10). Where many samples of a single, unprocessed material are being screened for their protein content, a dye-binding procedure may be the most appropriate (11, 12).

Analyses of Individual Amino Acids in Foods

Hydrolysis Prior to Chemical Analysis

All methods to be discussed require preliminary treatment of the test material to hydrolyse the proteins to the free amino acids. A major problem of amino acid analysis in foodstuffs is the destruction of amino acids during acid hydrolysis. Unfortunately, this problem can be greatest with the essential amino acids likely to be limiting in practical diets: "methionine + cystine," Iysine, threonine, and tryptophan. Proteins and protein foodstuffs differ so widely in their composition that "ideal" hydrolytic procedures would need to be almost specific for each material. Thus, compromises between the ideal and practical are often necessary. Of the many procedures available in the literature, those shown in chapter 8, section B, have proved suitable for routine analysis of the amino acid composition of wide ranges of pure proteins and protein foodstuffs. Many reviews (13-20) exist in the )iterature where analytical problems are discussed.

Amino acids are released and destroyed at different rates that depend upon the amino acid composition and characteristics of the sample. Assessment of amino acid composition has been recommended (21) as being most accurate when derived from five separate hydrolyses - three acid hydrolyses of different time durations (usually 24, 48, and 72 hours) a special acid hydrolysis following performic acid oxidation for cysteic acid and methionine sulphone, and an a)kaline hydrolysis for tryptophan determination. The three hydrolysis times are designed to allow selection of specific times for certain amino acids as well as extrapolation to zero time for the most labile amino acids. Separate procedures for sulphur amino acids and for tryptophan are always essential; however, in most cases a single 24-hour acid hydrolysis can give adequate information for scoring purposes (chap. 3).

Detailed procedures have been described for cereals (21). An ideal analysis for cereal products would use values for threonine and serine from extrapolation to zero time, tyrosine after 24 hours, isoleucine and valine values after 72 hours, and the remaining amino acids as the mean of the three determinations. Samples other than cereals would not necessarily behave in a similar manner. Caution should be used in the interpretation of sulphur amino acid data following oxidation because certain food samples (22) may already contain unavailable oxidized sulphur amino acids as a consequence of procedures used for their processing. Availability, however, is dependent on the degree of oxidation, methionine sulphoxide being generally fully available whereas methionine sulphone is not. As tryptophan is destroyed during normal acid hydrolysis, a separate analytical procedure is needed. Alkaline hydrolysis (p. 92) followed by a special short-column procedure to separate tryptophan from Iysinoalanine and the dibasic amino acids (23) is routinely used in many laboratories. An alternative procedure is to use one of the variants of the Spies and Chambers colour reaction with dimethylaminobenzaldehyde (24-27). It is important with foodstuffs to check the recovery of pure added tryptophan.

Chromatographic Techniques

Most procedures for amino acid analysis depend on the use of chromatography. Early techniques using oneandtwo-dimensional paper chromatography, while at best only semi-quantitative, gave much information on amino acid composition and amino acid metabolism. These procedures have almost all been replaced by column techniques, although thin-layer chromatography has certain applications. Within the various techniques using columns for separation, a major subdivision can be made into preand post-column derivatization procedures. Ion-exchange chromatography uses post-column derivatization in that the amino acids are separated by means of ion exchange, and derivatives are formed after they have emerged from the column so that they may be quantified. The most common derivatization procedure is that using ninhydrin with subsequent determination of optical density. In contrast, pre-column derivatization, as is used for gas-liquid chromatography (GLC) and high-performance liquid chromatography (HPLC), uses columns to separate the amino acid derivatives. These derivatives, after emerging from the column, are then quantified by various detection devices. GLC and HPLC procedures are frequently more rapid than ion exchange procedures, but their major limitations often lie in the preparation of the derivatives rather than in the chromatography as such.

High-Performance Liquid Chromatography

As high pressures are no longer an integral part of HPLC, the general procedure has recently been renamed "high-performance" rather than "high-pressure" liquid chromatography. Pre-column derivatization is required for amino acid analysis. Dansyl chloride (5-dimethylamino-1-naphthalene sulphonyl chloride) is frequently used for this purpose, producing fluorescent dansyl derivatives that are separated by a reversed phase column chromatographic procedure. The column employs silica gel with attached non-polar hydrocarbon functional groups (e.g., octadecyl moieties) as the stationary phase (28) and uses a multi-step non-linear elusion procedure. Among other eluents, acetonitrile and water mixtures have been suggested for the separation of dansylated amino acids (29). These are then detected and measured by a fluorescence detector, which can provide detection limits in the picogram range. It has been demonstrated (28) that HPLC, utilizing various non-polar stationary phases, is superior to ion-exchange chromatography for separating peptides. However, because of limitations in the separation of the polar amino acids, it is at present stir) inferior to ion-exchange for protein compositional studies. Procedures can, however, be extremely rapid and sensitive. Separation of some 24 amino acids in physiological fluids in a 40minute period has been demonstrated (29). Thus, while the procedure currently suffers from some limitations, it seems highly likely that the method will develop into a rapid and sensitive routine procedure for amino acid analysis. One major advantage of HPLC over all other chemical procedures lies in its ability to distinguish D and L forms of amino acids. Therefore, for certain research purposes, its use is already essential.

Ion-Exchange Chromatography lon-exchange chromatography remains the most utilized method of amino acid analysis. Many of the commercial analysers currently available use the automated procedure introduced by Spackman, Stein, and Moore in 1958 (30, 31). Modifications include use of only one column and gradient elusion instead of stepwise elusion (32, 33). With recent advances, such as the use of lithium instead of sodium buffers, higher pressures, narrow columns with fine spherical resin particles, and fluorescamine reagents replacing ninhydrin, the newer commercial equipment will give good separation of the common amino acids in picogram quantities in two hours or less. General reviews of the use of these procedures are available (34, 35). It is essential to introduce an internal standard into each hydrolysate to check the potency of ninhydrin solutions. Norleucine is convenient for acid and neutral runs and for singlecolumn systems. Alpha-amino-beta-guanidino-n-propionic acid is recommended where a separate, short column is employed for basic amino acids. An alternative to the use of internal standards when automated peak area calculations are available is the running of amino acid standards as every third or fourth run.

Automation has progressed to include sequential sample loading and regeneration of the columns. Laborious calculation of peak areas can also be avoided by the integration of photocell signals by computer integration techniques. Methodology of analysis is provided by equipment manufacturers. Some brief notes for users of these techniques are provided in chapter 8 (p. 94).

Gas-Liquid Chromatography

Analysis of amino acids using gas-liquid chromatography requires the quantitative conversion of the amino acids to volatile derivatives (36, 37). The method has the promise of a speedier output and relatively low capital outlay for equipment. Hydroxy acid methyl esters (38), trimethylsilyl esters (39), and n-butyl-N-trifluoroacetyl esters (40) appear to offer some promise. Before their conversion to volatile esters, hydrolysates of foods require a preliminary separation of the amino acid fraction to remove interfering substances. Good agreement with the values obtained by other procedures has been reported with maize and soy bean mea) (41).

Thin-Layer Chromatography

Because of the high expense of modern ion-exchange chromatography equipment, amino acids are often separated by thin-layer chromatography (TLC). TLC techniques are inexpensive, and, although the processing of each plate may be lengthy, many plates can be treated at one time. Hence, output can be quite considerable. The early work (42, 43) has been improved to a level at which the technique may be routinely applied to protein hydrolysates (44, 45). The results agree well with values obtained by ion-exchange chromatography.

Microbiological Assays

Microbiological assays can be extremely useful when other equipment is not available, or when many assays of only one amino acid, for which there is not a convenient, specific colour reaction, are required. An example of the latter is in the preliminary selection of Phaseolus beans for methionine content from large numbers of lines (46).

Detailed procedures for microbiological assays have been published (47-49). Continuous vigilance is required over standardization of organisms, maintenance of cultures, composition of media, and general techniques, because all of these factors have been shown to influence results and to vary from one laboratory to another. High apparent Iysine values for animal products have been traced to a synergistic effect of hydroxylysine; this may be avoided by the inclusion of hydroxylysine in the basal medium 150, 51).

Workers have sometimes used only mild hydrolytic procedures for test materials in order to reduce the risk of losses of amino acids, and have relied on the ability of organisms to utilize soluble peptides. However, this can lead to stimulation of growth with final calculated values being too high, as judged by chemical procedures (52), so the practice needs careful scrutiny.

Contents - Previous - Next