Contents - Previous - Next

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

Recent 24-hour tracer studies of amino acid requirements

Our earlier C-tracer studies, which provided the basis for the new requirement values for adults given in table 2, involved relatively short-term tracer infusion protocols lasting from three to eight hours. However, because of the existence of major rhythms in amino acid metabolism throughout the day, the value of short-term C-tracer studies for predicting the daily amino acid balance can be questioned [25, 26]. Other criticisms of the design of our tracer studies, including concerns about dietary amino acid imbalances, the level of intake of so-called non-specific or non-essential nitrogen, and the criterion of adequacy (balance calculations), have been addressed elsewhere [2, 4, 8, 27]. The various criticisms of our work do not appear to be of sufficient consequence or to help explain the major differences between our tentative new requirement figures (table 2) and the much lower values that were judged by FAO/WHO/UNU [9] to be adequate for maintenance of long-term nutritional health. Similar conclusions were drawn by Waterlow [28] and by Fuller and Garlick [29], largely on the basis of their reviews of our studies. However, because we believe that rhythms in amino acid metabolism are of great importance, our more recent studies have included the use of a continuous 24-hour C-amino acid tracer protocol. This modification in our experimental design has been of major value in improving the strength of estimates of human amino acid requirements.

TABLE 2. Revised estimates of amino acid requirements in adults and a proposed new amino acid scoring pattern for this age group

Amino acid 1985 FAO/WHO/UNU [9] Revised estimates [4, 6]
Requirement (mg/kg/day) Pattern (mg/g protein) Requirement (mg/kg/day) Pattern (mg/g protein)
Isoleucine 10 13 23 38
Leucine 14 I 9 39 65
Lysine 12 16 30 50
Methionine and cystinea 13 17 15 25
Phenylalanine and tyrosine 14 19 39 65
Threonine 7 9 15 25
Tryptophan 3.5 5 6 10
Valine 10 13 20 35
Total 83.5 111 187 313

a. Sulphur amino acids.

These 24-hour studies have shown that the isotope tracer balance concept is valid [30-32] and have fully supported the conclusions of our earlier, short-term studies that produced far higher estimates of the leucine requirements in adults than those proposed by FAO/WHO/UNU [9]. A summary of our findings for leucine is given in table 3. From these results it is entirely reasonable to conclude that an intake level of 39 mg leucine/kg/day approximates the minimum physiological requirement for this amino acid. Hence, this new intake level, rather than the FAO/WHO/UNU [9] figure of 14 mg/kg/day, should be adopted for use in any further quantitative consideration of adult human protein nutrition, including issues of dietary protein quality. Waterlow [33] has reached a similar conclusion with respect to our reassessment of the adult requirement for leucine.

TABLE 3. Leucine balance based on measured 24-hour leucine oxidation in adult subjects with different intakes of leucinea

Leucine intake (mg/kg/day) Leucine balance
Daily (mg/kg/day) Percent of intake
14 (FAO)b - 13 5 - 96
39 (MIT)c - 6 4 - 16
89.4 0.76 3 1

a. Summarized from ref. 31.
b. Requirements estimated by FAO/WHO/UNU [9]
c. Requirements estimated by Young et al [4].


We have also recently completed three major studies related to C-phenylalanine kinetics over the 24-hour day and estimates of the phenylalanine requirement (in the absence of dietary tyrosine) [34, 35]. Using the estimated rate of conversion of C-phenylalanine to its end product CO2 in expired air as an estimate of the irreversible oxidation rate of phenylalanine, we find (table 4) that healthy adult subjects are in distinctly negative phenylalanine balance when they are given an intake of phenylalanine (without tyrosine) that is equivalent to about 150% of the FAO/WHO/UNU [9] upper requirement level for aromatic amino acids. However, as would be anticipated, the mean phenylalanine balance was found to be at equilibrium when the phenylalanine intake was either generous (100 mg/kg/day) or at the newly proposed (MIT) aromatic amino acid requirement level of 39 mg/kg/day (see table 2). Again, these new studies are consistent with our previous tentative new requirement estimates and they offer further evidence to indicate that the FAO/WHO/UNU [9] values are inappropriate and should now be discarded.

TABLE 4. Phenylalanine balance based on 24-hour [1-13C]phenylalanine tracer studies in healthy adult men

Phenylalanine intake (mg/kg/day) Phenylalanine oxidation, mean SD (mg/kg/day) Phenylalanine balance (mg/kg/day)
100 106 11 0.2 19
39 (MIT)a 37 7 + 1.9 6.5
23 (~FAO)b 47 13 - 23 14c

a. Requirements estimated by Young et al [4].
b. Close to requirements estimated by FAO/WHO/UNU |9].
c Significantly different (p < .05) from zero balance.

Estimates of the lysine requirement in adults

Lysine is the first limiting amino acid in most cereal-based diets characteristic of populations in certain areas of the developing world [6], and it is likely that the protein nutritional quality of these diets could be improved by increasing the supply and availability of lysine Therefore, it is highly important to resolve the uncertainties concerning the minimum physiological requirement for this indispensable amino acid. Twenty-four-hour C-labelled amino acid tracer studies comparable to those described above for leucine and phenylalanine, using both C-lysine and C-leucine tracers and estimations of their body amino acid balances as criteria for evaluating the lysine requirements of healthy adults, are now under way in our laboratories. However, they have not yet reached the point at which we are able to confirm our proposed tentative requirement estimate of 30 mg/kg/day (table 2) a value significantly higher than the FAO/WHO/UNU [9] figure of 12 mg/kg/day. Nevertheless, it is worthwhile overviewing the status of our current knowledge about the minimum requirements for lysine in healthy adults, particularly because it is consistent with the tentative requirement value that we propose.

Various criteria and methods have been used to estimate the lysine requirement in healthy adults. The nitrogen balance studies carried out by Rose et al. [36] and those by later investigators in men and women [10, 19], including our own nitrogen balance experiments [38, 39], are in general similar (table 5). These give estimates of the mean minimal lysine requirement on the order of 9 to 12 mg/kg/day. It might be noted that the FAO/WHO/UNU [9] value of 12 mg/kg/day was taken from the 1973 FAO/WHO report [11], and was derived from the upper range of the individual requirements found in men and women. Our own data on nitrogen balance are not inconsistent with these values, except that we estimated a nitrogen balance derived requirement figure to be somewhat higher, namely 17 mg/kg/day, when the nitrogen balance data included an estimate of miscellaneous and unmeasured nitrogen losses [37, 38]. These similar findings are not unexpected, because it might be anticipated that nitrogen balance studies carried out in different laboratories should give comparable estimates of the minimum needs for dietary lysine However, the apparent consistency in the data in no way validates the technique or the derived estimates of the requirements for specific amino acids. Indeed, we have illustrated previously why we believe some of the classical balance experiments, and those by Rose et al. [36] in particular, would lead to erroneous requirement estimates [3].

TABLE 5: Estimates of the lysine requirement in healthy adults

Authority Value (mg/kg/day) Criterion/method
Rose et al. [36] 9
N balance (mean; n = 5)
safe intake (double minimum)
FAO/WHO/UNU [9] 12 N balance; upper level for group
MIT studies to date
[37, 38. and this article]
~ 12
~ 30
~ 30
<= 26
> 15 < 80
N balance (17 with miscellaneous)
13C-lysine; short-term fed
factorial prediction
plasma amino acid responses
diurnal retention (after Millward et al. [49])
current 24-h 13C kinetics (incomplete)
Zello et al. [42] 37 indicator amino acid oxidation
Duncan et al. [43] 40 indicator amino acid oxidation


On the basis of earlier short-term C-lysine tracer infusion studies, we estimated the mean lysine requirement as approximately 30 mg/kg/day [38]; our factorial prediction gave a somewhat higher value of 42 mg/kg/day [4]. Furthermore, preliminary results from our recent 24-hour C-lysine kinetic studies indicate that the minimal requirement for lysine is likely to be much greater than the FAD/WHO/UNU value [9]. However, we have not completed a sufficient number of these C-lysine studies to establish what minimal level of lysine intake might be sufficient to maintain body lysine balance or protein nutritional status over the long term.

A group in Toronto has used the so-called indicator amino acid oxidation technique for determining amino acid requirements [39, 40]. This approach is based on the concept that the incorporation of an indispensable amino acid into proteins or its oxidation is dependent on the level of intake of the most limiting amino acid (see FIG. 1. Schematic presentation of the indicator amino acid oxidation approach for estimating the requirements for specific indispensable amino acids. Here the indicator used is C-phenylalanine [41]). Thus, when one amino acid is limiting for protein synthesis, the other indispensable amino acids whose intakes are in relative excess will be oxidized. With graded increases in the intake of the limiting amino acid, these other indispensable amino acids will be used to an increasingly greater extent for protein synthesis, and so their rates of oxidation will be gradually reduced. When the requirement for the initially limiting amino acid is reached, the oxidation of the other indispensable amino acids will have been reduced to the point at which no additional change in their oxidation rate occurs with further increases in the dietary level of the limiting amino acid. This point is taken to be the requirement level.

Using the oxidation of L-[13C]phenylalanine [42] as the indicator amino acid, Zello et al. [41] concluded that "the lysine requirement of adult males is three times greater than the World Health Organization recommendation of 12 mg/kg/day." Recent studies by this group [43] have confirmed their previous estimate of 40 mg/kg/day, which is consistent with, although somewhat higher than, our new tentative estimate of 30 mg/kg/day.

Two additional lines of evidence can be used in support of our view that the lysine requirement value for adults is considerably higher than that derived from nitrogen balance studies and, in consequence, the recommended values proposed earlier by national [10] and international [9, 11] expert groups. The first is in reference to the response of plasma free lysine concentrations to changes in the level of lysine intake. The reason for exploring this aspect of lysine metabolism is that plasma amino acid concentrations are known to be influenced by the level and adequacy of the amino acid intake, with particularly marked changes occurring for the limiting dietary amino acid [44, 45]. Indeed, plasma amino acid changes have been used as a basis for identifying the limiting amino acids in food proteins and for quantifying protein nutritional quality [45-47]. In our previous studies, we found that amino acid-containing meals produce an increase in the plasma concentration of the non-limiting indispensable amino acids in the diet, as compared with the concentrations found during the fasting (post-absorptive) state. In the case of the "limiting" amino acid, meal ingestion was associated with an increased concentration of the amino acid in plasma only when the meal provided an adequate supply of this amino acid. Below a given dietary level, meal ingestion reduced the concentration of the limiting amino acid below that of the fasting blood values. Hence, the intake level at which a changeover in the prandial/post-absorptive plasma amino acid response occurs may well correspond closely to the minimum physiological requirement. Thus, with lysine as the "limiting," or test, amino acid, the observed plasma lysine response was as shown in figure 2 (see FIG. 2. Relationship between plasma lysine concentrations in fasted and fed states and lysine intake in young adults. Drawn from data of Meredith [37] and Meredith et al. [48]). These data suggest, again, that the lysine requirement approximates 30 mg/kg/day (table 5).

The second additional line of evidence is based on considerations of the diurnal cycle of feeding and fasting, resulting in gains and losses of body proteins throughout the day in subjects who are in overall nitrogen balance. This area of protein and amino acid metabolism has been explored [49, 50], with the suggestion that diurnal cycling has important implications for understanding the metabolic basis of the protein requirement [26]. A recent study examined experimental work on nitrogen and amino acid homeostasis in relation to the determination and evaluation of post-prandial protein utilization (PPU) in the human adult [50]. This investigation represents a conceptual development that is quite valuable, in our opinion, and may also be used to further assess and develop an approximation of the minimum lysine requirement in adult subjects.

PPU is defined as protein deposition plus post-absorptive losses/intake [50]. For milk as the test dietary protein source, a value of 0.86 was obtained from metabolic studies [51]. Further, Millward and Pacy [50] have estimated from the results of our studies with amino acid-based diets [22] that when subjects receive the FAO/WHO/UNU [9] amino acid requirement pattern, the PPU is 0.53. This compares with higher values of 0.74 and 0.70 for amino acid-based diets supplying the MIT and egg patterns, respectively. Again, these calculations imply that the FAO/WHO/UNU [9] amino acid requirement values are inferior to those represented by the new, tentative MIT amino acid requirement pattern for adults.

We can apply the concepts that Millward and his colleagues have developed, concerning the magnitude of the post-prandial retention of protein necessary to balance subsequent post-absorptive losses, to estimate a minimum requirement level of lysine. This estimate is based on the following line of reasoning. First, it is accepted that, for the purposes of this argument, the minimum needs for total protein in adults can be estimated, as illustrated in figure 3 (see FIG. 3. Relationship between nitrogen balance, the intake of good-quality protein, and the requirement for protein), from the response of body nitrogen balances measured over relatively short experimental diet periods (e.g., 10 14 days) [9, 52]. Second, it follows that at the requirement intake level for total protein, the post-prandial retention of protein would be just sufficient to balance the subsequent loss of protein during the post-absorptive period in order to achieve a state of nitrogen equilibrium over the 24-hour period. Third, the magnitude of the post-prandial retention of nitrogen (protein= N x 6.25) for subjects consuming 0.77 g/kg/day for 10 days or more was reported to be equivalent to 250 mg protein/kg/12 h (21 mg protein/kg/h) [51]. These subjects, however, were in a daily negative balance of -9 mg N/kg, and so it is uncertain whether a post-prandial retention of the magnitude observed is actually sufficient to maintain protein homeostasis. Nevertheless, from earlier 13C-leucine tracer studies [53] we found that there was a net whole-body protein synthesis (protein synthesis minus protein breakdown) of about 23 mg protein/kg/h during the post-prandial period in healthy adults receiving an intake of 0.8 g egg protein/kg/day (see FIG. 4. Net protein synthesis during the post-prandial period in young adults receiving different intakes of egg protein. Based on the data of Motil et al. [53]). This tracer-derived estimate is remarkably close to the nitrogen-balance data of Price et al. [51], and so a prediction can be made of the amount of dietary lysine minimally required to achieve this level of post-prandial protein retention or of net protein synthesis. Hence, assuming that the lysine content of the protein retained is 78 mg per gram crude protein [54], and that at a requirement level of lysine intake the efficiency of lysine retention is 80%, then the minimum lysine intake to support this gain and subsequent loss of body protein would have to be 27 mg/kg/day. This value is also comparable to most of the lysine requirement estimates discussed above, except, of course, for the far lower estimates based on nitrogen balance and the FAO/WHO/UNU recommendations [9].

There is a cohesive though limited body of data indicating that the mean requirement value for lysine in adults is probably in the region of 30 mg/kg/day (table 5). When expressed per unit of the mean protein requirement, the lysine requirement pattern would be 50 mg per gram of protein. Again, this far exceeds the FAO/WHO/UNU [9] value of 16 mg lysine per gram of protein. The question now arises, therefore, as to whether there are additional experimental data to support this new tentative requirement value of 50 mg lysine per gram of protein.

Contents - Previous - Next