Cover Image
close this bookEnergy and Protein Requirements, Proceedings of an IDECG workshop, November 1994, London, UK, Supplement of the European Journal of Clinical Nutrition (International Dietary Energy Consultative Group - IDECG, 1994, 198 pages)
close this folderThe requirements of adult man for indispensable amino acids
View the document(introductory text...)
View the document1. Introduction
View the document2. The problem in defining requirements
View the document3. Protein quality
View the document4. The maintenance requirement (MR)
View the document5.Diurnal cycling: the Millward-Rivers model
View the document6. Theoretical basis of the MIT tracer balance studies
View the document7. Technical problems of tracer balance studies
View the document8. Results of the MIT tracer balance studies
View the document9. Relation between leucine oxidation and nitrogen excretion
View the document10. Factors relating to the design of tracer balance experiments
View the document11. Breakpoint analysis
View the document12. Effect of protein/amino acid intake on protein synthesis and breakdown
View the document13. The colon: losses or gains?
View the document14. Conclusion
View the documentReferences
View the documentDiscussion
View the documentReferences

(introductory text...)

JC Waterlow

Centre for Human Nutrition, London School of Hygiene and Tropical Medicine, 2 Taviton Street, London, WC1H OBT, UK

Descriptors: indispensable amino acids, requirements, carbon balance, leucine

1. Introduction

Until less than a decade ago all quantitative estimates of human requirements for protein and indispensable amino acids (IAAs) were derived from measurements of nitrogen balance. The FAO/WHO/UNU Expert Consultation, which met in Rome in 1981 (henceforth referred to as the Rome Committee) and which reported in 1985, reviewed the extensive short-term and long term balance studies on healthy young men and women that had been carried out by Young and Scrimshaw at MIT, by Calloway at Berkeley and many others. The results of selected studies, summarized in the Rome report, were fairly consistent in establishing the average total protein requirement, in terms of egg or beef protein, as 0.625 g protein/kg/d with a mean coefficient of variation (CV) of 12.5%. Studies with typical mixed diets from a variety of countries gave a mean requirement of 0.745 g/kg/d, with a mean CV within studies of 16.3%. Long-term studies, carried on for 2-3 months, showed that balances were marginal or negative in most individuals on intakes slightly below 0.6 g/kg/d. The matter of the total protein requirement thus seemed to have reached a reasonable solution, but a question mark still hung over the IAA requirements and the assessment of protein quality.

For the IAAs the Rome Committee based their recommendations on the balance studies of Rose (1957). These estimates (Table 1A) were so low that in the view of the Rome Committee no adult diet anywhere in the world would be limited by its IAA content. In other words, protein quality was not a problem for adults. Leverton's work on young women led to similar values (Leverton, 1959).

Young et al (1989) have stressed the drawbacks of the nitrogen balance method on which these results were based (Table 2). To those listed in the Table should be added the unrealistically high positive balances almost universally obtained with high intakes of nitrogen. Young and coworkers therefore developed a radically new approach, based on the availability of amino acids labelled with stable isotopes. They argued that whereas the classical balance experiment involves determining the amount of amino acid needed to replace losses as measured by N excretion, it would be equally valid to determine the losses by measuring the oxidation of carbon from a labelled amino acid. The general procedure was to infuse the test amino acid labelled with 13C in the 1-carbon position and determine the amount of amino acid oxidized from the output of 13CO2 divided by the isotope abundance in plasma. The details of the method, which has been called the 'tracer balance', and its implications have been discussed by Young (1987), Young et al (1989), Young and Marchini (1990) and Young (1991). The approach was initially applied to determine the requirements for leucine (Meguid et al, 1986a); Lysine (Meredith et al, 1986); valine (Meguid et al, 1986b); and threonine (Zhao et al, 1986). The requirement of adults for these amino acids determined by tracer balance turned out to be two to three times greater than the estimates in the Rome report (Table 1). Young has argued in support of these new figures that their pattern (ma amino acid per g protein) is very similar to that of tissue protein and also to the requirement pattern of the pre-school child as proposed by the Rome Committee (Table 1B). These points are considered in more detail in sections 3 and 4.

Correspondence: JC Waterlow, 15 Hillgate Street, London W8 7SP, UK.

In 1989 an Expert Consultation of FAO/WHO on the evaluation of protein quality, impressed by the evidence against the low estimates of adult IAA requirements proposed by the Rome Committee, recommended as a provisional measure that the pattern of IAA requirements of the pre-school child, as set out by FAO/WHO/UNU (1985), should be adopted for all ages other than infants (FAO/WHO, 1990). This pattern is similar to that suggested by the MIT workers.*

* In the text and tables that follow, all references to the FAO pattern or to FAO levels of IAA requirements relate to the pattern of requirement levels proposed by FAO/WHO/UNU (1985).

Millward has pointed out an inconsistency in the Rome report, in that much larger absolute amounts of IAAs were recommended for pre-school children than for adults, even though by 2 years of age the growth component, which requires net protein deposition, is already small compared with the maintenance component (e.g. Millward & Rivers, 1988). The new proposals would meet this point by increasing the adult IAA requirements rather than by reducing those of children.

An analysis of diets in developing countries led to the conclusion that with the new estimates Lysine was likely to be limiting in cereal-based diets, and that for an individual to have an intake that met the requirement it would be necessary for about 40% of his protein intake to be derived from animal sources or legumes (Young & Pellett, 1990). Since in many developing countries cereals are the main source of protein, it was concluded that large numbers of people in these countries would be at risk of having deficient intakes of Lysine. Young & Pellett admitted that '... individuals consuming diets that we have characterized as being of poor quality have survived under these conditions. We do not know, however, whether this is because our predictions are invalid, whether a relatively benign physiological adaptation has occurred, or whether there has been a functionally costly accommodation to lower than apparently desirable amino acid (or Lysine) intake levels'. It is worth noting, however, that the calculations were made for populations of reference body weight, with no allowance for the much lower body weights prevailing in developing countries.

Table 1A Indispensable amino acid requirements: per unit body weight (mg/kg/d)


FAO/WHO/UNU adultsa

MIT new estimatesb

Predicted from obligatory lossesb

FAO/WHO/UNU pre-school chida

Leucine

14

40

39

73

Lysine

12

30

42

64

Threonine

7

15

21

37

Valine

10

20

24

38

Methionine


13

16


Methionine + cystine

13



27

Safe level of protein (g/kg/d)

0.75



1.1

a From FAO/WHO/UNU (1985).
b Young et al (1989). Average requirements.

Table 1B Indispensable amino acid requirements: scoring pattern (mg/g protein)


FAO/WHO/UNU adultsa

MIT new estimatesa

FAO/WHO/UNU pre-school childb

Leucine

19

53

66

Lysine

16

40

58

Threonine

9

20

34

Valine

13

26

35

Methionine + cystine

17

17

25

Assuming safe level of total protein:
a 0.75 g/kg/d.
b 1.1 g/kg/d.

These new recommendations of the MIT group have been strongly criticized by Millward and coworkers (Millward & Rivers, 1988; Millward et al, 1989; Millward, 1992, 1994), but mainly on conceptual and technical grounds, rather than on the basis of different quantitative data. In a series of studies published since 1986, Young et al (1989) have attempted to counter most of these criticisms. It is clear that the new recommendations, if accepted, have important implications for food and agricultural policy. I have therefore tried in this report to examine in detail the experimental observations of the MIT group as well as the ideas of Millward. It has also seemed necessary to consider more general questions, such as what is implied by the term 'requirement', the problem of adaptation, and the still mysterious mechanism by which intake and output are brought into balance.

Table 2 Sources of error in nitrogen balance studies used to estimate adults' amino acid requirements

Issue

Comment

Nitrogen balance

Can be obtained at various levels of intake and does not necessarily indicate adequacy of intake

Technical errors

Tendency to overestimate N balance and therefore underestimate requirements

Criterion of balance

Need to allow for miscellaneous unmeasured losses

High energy intakes

Excess energy will improve N balance and thus underestimate requirements at maintenance energy intakes

Sensitivity and precision

Changes in body N balance may not be detected promptly or with precision

Validation

No satisfactory validation available

After Young et al (1989).

2. The problem in defining requirements

At the meeting of the FAO/WHO Expert Committee on Protein Requirements in 1963 (FAO/WHO, 1965) I maintained that there was only one level of requirement worth talking about - the minimal requirement. I am now persuaded that that view is too restricted because it was confined to results obtained by nitrogen balance. Further, it was suggested that the requirements of all men are more or less equal, to which a committee member from former Czechoslovakia replied 'I am not so sure'. Nicol & Phillips wrote in 1976: 'The protein requirements of all apparently healthy men can only be established in the context of their ecological, socioeconomic and nutritional backgrounds'. Thus as long as 30 years ago doubts were being expressed about the way in which protein requirements should be formulated.

Millward said 'The established perception of the nature of protein requirements is inadequate' (Millward et al, 1990). For him there are three levels of requirement: the optimal, the operational and the minimal. The optimal requirement would be determined by functional criteria such as good health, growth, resistance to disease. These criteria are hard to define, although studies of immune status could be used at the population level. Chittenden lived a healthy and active life for many years on a protein intake of about 30 g per day. He maintained that the high protein intakes recommended by Voit (about 120 g per day) constituted 'individual and racial suicide'. This is the only example I know of in which health has been the criterion for recommending a specific level of protein intake.* It is an important task for the future to search for correlations between protein intake and functional criteria that can be stated in quantitative terms. For example, there is increasing evidence that linear growth in children may be influenced by diets that provide good quality protein (Allen, 1994; Golden, 1994), although we do not know whether the effect is due to vitamins, minerals or amino acids. Golden (1994) has put forward a convincing hypothesis for the role of sulphur amino acids.

* Chittenden observed that in soldiers and athletes who had been living on a generally high protein diet, change to a mainly vegetarian diet providing 0.75 g protein/kg/d for 5 months led to an increase of 38% in strength and work performance by 15 tests (Millward, 1994).

The operational requirement, a term introduced by Millward, although it has overtones of the NPUop of Miller & Payne (1961), will be discussed below in relation to the Millward-Rivers model. It takes account of the fact that nitrogen balance can be achieved over a wide range of protein intakes. This has long been recognized, at least since the time of Folin (1905), but I think it is fair to say that we still do not know how this balance is achieved (Waterlow, 1994).

The minimal requirement, which has been the object of innumerable measurements, is, as its name implies, the lowest level of protein or amino acid intake at which N balance can be achieved and maintained. The work of Sukhatme & Margen (1978), which at one time had a good deal of influence, seemed to imply that this minimal level could be variable in an individual. Millward et al (1989) have said that their work implies 'a regulatory mechanism which adjusts daily N balance over a period of several days ... adaptive mechanisms exist which adjust output to balance intake and limit the extent of any loss or gain of body protein. This is an alternative model defining the requirement as a range of intakes over which equilibrium can occur. In contrast, the conventional model is based on an intrinsic requirement which is a fixed function of body weight'.

It is necessary to comment on this statement. The range of intakes over which balance can be achieved is well recognized and the description of an 'alternative model' is unjustified. Sukhatme & Margen's theory of regulation is based on the finding of auto-correlation in urinary N output. This means that if the output on day 2 is lower than on day 1, it will be lower on day 3 than on day 2, and so on. This process would, if continued, lead to zero output (negative correlation) or infinite output (positive correlation). Obviously this does not Occur; after a few days the cycle is reversed. This reversal appears to be caused by random variation (Sukhatme, personal communication). Healy (1989) has criticized the concept of autocorrelation on theoretical grounds; Rand et al (1979) looked for it in a large series of long-term balance studies and found evidence of it in only a small minority. In any case, autocorrelation, if it relies on random variation to maintain a long-term steady state, would seem to be the reverse of a regulatory mechanism. A regulatory mechanism is one which, like a thermostat, manages to maintain a steady state by opposing or reducing the effect of random variations or imposed fluctuations (Waterlow, 1985).

In the statement that 'the conventional model is based on an intrinsic requirement which is a fixed function of body weight', the key words are 'intrinsic' and 'fixed'. All balance studies are conducted on a particular individual at a particular point in time with a particular body weight and total body nitrogen. It seems reasonable to suppose on physiological grounds that the nitrogen losses, which have to be balanced by the requirement, should depend on the body weight, or better, the lean body mass or total body N. However, I know of no studies that have attempted to establish how strong the correlation is, in the way that we have studies relating the BMR to body weight or lean body mass. I find it difficult to believe that such a correlation does not exist, but that in no way rules out the influence of other factors, such as body composition, age, sex and possibly height in relation to weight. For example, Egun & Atinmo (1993) showed that on a Nigerian diet women had a lower protein requirement per kg than men, but it was the same when related to lean body mass. If the measurement of nitrogen balance was as easy as that of BMR, we would be far further.

We have no information about whether the minimum requirement per unit body nitrogen is in fact fixed. Studies in third world countries, where people might be supposed to be existing on low protein intakes, have so far shown no significant differences in obligatory N losses from those found in industrialized countries (Torun et al, 1981; FAO/WHO/UNU, 1985). However, even if there is a strict physiological relationship between the daily obligatory losses and the amount of body N. there is still a possibility for adaptation in the efficiency with which amino acids are used (Nicol & Phillips, 1976). Millward (1992) contends that in the adult there is a set-point for the upper limit of body protein, which is determined by height and frame size. This idea of a set-point seems very reasonable. For example, in the normal adult neither plasma albumin nor haemoglobin concentration can be raised above a certain level by an increase in dietary protein. In experiments with rats, Henry et al (1953) showed that with increasing protein intake total liver protein rose towards an asymptote, with ever diminishing returns on the increased intake.

It is also well recognized that on moving from a higher to a lower protein intake there is a small loss of body protein, about 1% in the human adult (Young et al, 1968). This small loss can apparently be tolerated without harmful effects (Waterlow, 1985). It has been regarded as drawing on 'labile protein stores', but the concept of a store is inappropriate. It is probably better to regard it as a kinetic adjustment that allows constancy of body protein to be maintained at a new setting, the processes of protein synthesis and breakdown needing a little time to adjust to the new level of intake (Waterlow et al, 1978).

There is a further stage of adaptation. If the intake is too low there will be an exponential loss of body protein until balance is achieved at a new level of body weight (Waterlow, 1985). For example, if the requirement for maintenance is taken as 0.1 g N per kg per day, and a 70 kg man is on a diet that provides 5 g N, or 0.07 g per kg, other things being equal he will lose body N until his weight has fallen to 50 kg, when he will again be in balance. Of course, other things may not be equal; nitrogen may be used more efficiently, as suggested by Allison's work on dogs (Allison, 1951). One may ask, what degree of loss of body N is acceptable? If the subject initially had a height of 1.75 m and at 70 kg a body mass index (BMI) of 22.8, at the end of this second stage the BMI would be 16.3, which, according to current thinking, would be inacceptable (James et al, 1988). Moreover, it appears that such a loss would not be uniform, but would involve a disproportionate amount of muscle mass, visceral mass being well maintained (Soares et al, 1991). This is a further reason why the nitrogen balance at a given point in time cannot be regarded as giving a complete answer to the question of the protein requirement. If we regard the requirement per unit body weight as fixed, what is the ideal body weight at which it should be fixed, or is anything short of Millward's setpoint suboptimal?

3. Protein quality

In absolute terms the requirement for protein in a non growing adult is the minimum amount needed to meet that body's metabolic demands for nitrogen, i.e. to secure N balance, in a particular situation. One aspect of the particular situation is the energy intake. It has long been recognized that, within limits, increasing or decreasing energy intake alters N balance by 1.5-2 mg/ kg. It was a criticism of Rose's original estimates of IAA requirements that the energy intake was unrealistically high. Therefore all comparative studies of protein quality in recent years have been made with subjects as far as possible in energy balance.** Obviously also protein quality can only be measured at low intakes. As Jackson (1995) has pointed out, a distinction has to be made between nitrogen requirements and protein requirements. A striking example of the need for this distinction is the breast-fed infant, whose food contains 30% of its nitrogen as non-protein N (see report on infant protein requirements). The NPN is usually included with protein in estimates of 'protein' requirements, which should more accurately be called nitrogen requirements, as they are measured by nitrogen balance.

** At MIT, for example, the nutritionist estimated the habitual energy intake from a dietary history and added 10%. Subjects were required to maintain their usual level of physical activity throughout the study and to monitor it by an activity diary. Weight changes were monitored and, if any trend was apparent, the energy intake was adjusted.

Jackson's analysis of the older literature shows that the requirement for IAAs is influenced by both the amount and the nature of the non-essential (NEN) component, effectiveness decreasing progressively as the NEN is provided by non-essential amino acids, glycine + glutamic acid, ammonium salts and urea. He has proposed that this gradation reflects the relative capacity of these components to provide substrates for the synthesis of IAAs by gut bacteria (see section 13). Whether or not this is the explanation remains to be seen, but the fact is clear, as shown, for example by the experiments of Kies & Fox (1978) among others. In conventional estimates of protein quality, NEN is provided by the NEAAs, which are readily exchangeable by transamination and to a lesser extent by deamination (glycine, serine). Thus in practice it is assumed that protein quality depends only on the amount and pattern of the IAAs.

Estimates of protein quality by N balance are in fact measures of the efficiency of utilization. If a protein could produce balance at an intake exactly equal to the obligatory loss, the efficiency of utilization would be 100%. In fact the efficiency, corrected for digestibility, is never better than about 80%, even with proteins such as those of egg or beef, which have an IAA pattern very close to that of tissue proteins. The Rome Report used a value of 70%. The reasons for this 'inefficiency' are not clear, and this is an important gap in our knowledge.

Millward et al (1989) have summarized the results of eight studies designed to measure the quality of different proteins by balance measurements at a series of different intakes (Figure 1). It happens that the first two in time of these studies, by Young et al (1973) on egg and by Inoue et al (1974) on wheat gluten, show fairly clear differences in biological value at low protein intakes. These early experiences encouraged Young and Scrimshaw to record the view that 'regardless of the method of measurement, our findings indicate that differences in the quality of dietary protein are important in the protein nutrition of adult man' (Young et al, 1975). The less clear-cut results of the later studies shown in Figure 1 do not mean that there are no differences in quality between different proteins; rather that they may be unimportant in practice with diets containing a mixture of proteins, such as those listed in Table 39 of the Rome Report. On the other hand, if the difference between cereals and animal protein is of practical importance, as claimed by Young & Pellett (1990) because of the difference in lysine content, why did it not show up in the balance studies? Rand et al (1981) calculated that to demonstrate a significant difference in biological value in adult humans would require a totally unrealistic number of subjects. Therefore, if Young's claim is sustained that the tracer balance is more sensitive than the nitrogen balance, it will be an important advance.


Figure 1
Measurements of biological value by the nitrogen balance method at different levels of protein intake.

The FAO/WHO reports of 1963 and 1973 gave much attention to the development of an IAA scoring pattern, with its linked concept of a limiting amino acid. The method can be applied to the protein of any diet whose amino acid composition is known, and the measurement is far simpler than the nitrogen balance. It is probably not of great importance whether the scoring pattern is taken to be that of milk, egg, beef, etc. whose relative contents of IAAs are similar to that of human tissue protein. However, the scoring pattern does not tell us anything about the absolute requirement for IAAs as a proportion of the total protein or N requirement. It is agreed that this proportion changes with age, being greater in the growing infant than in older children and adults. There is disagreement about whether the pattern changes with age, and is different for growth and maintenance (section 4). If, as Young maintains, it is not different, and is close to that of tissue protein, then to know the requirement for all IAAs it will be sufficient to know the requirement for one of them.

A third method of examining protein quality has recently been proposed by Millward et al (1991), as an outcome of his work on diurnal cycling: the slope of protein deposition vs intake (see section 5).

4. The maintenance requirement (MR)

This is a subject on which there is still a difference of opinion. Clearly when growth is rapid the IAA requirement must to a large extent be determined by the pattern of IAAs in the tissue that is being laid down. In the non-growing adult the MR derives from the need to replace IAAs that are consumed in a variety of irreversible pathways (Table 3). There is no a priori reason why this consumption pattern should bear any relation to that of deposition; a view that has been widely held by nutritionists in the past, e.g. Osborne & Mendel (1916, quoted by Millward, 1992) and Said & Hegsted (1970), and which is accepted in principle by Young (personal communication). Thus Fuller et al (1989), in experiments on young pigs, have assessed separately the pattern of requirements for growth and the pattern when there is no growth (maintenance), and found the two patterns to be quite different.

Table 3 Some non-protein pathways of amino acid utilization

Amino acid

Pathway

Methionine

Methylation reactions


Creatine


Choline

Cysteine

Glutathione


Taurine

Tyrosine

Neurotransmitters

Glutamate

Neurotransmitters

Lysine

Carnitine

Glycine

Nucleic acid bases


Haem


Creatine

From Reeds (1990).

Nevertheless, Young and El Khoury (1995) maintain that in practice in man the maintenance IAA pattern resembles that of tissue protein, on the grounds that there is a close correspondence between the tissue pattern and that of the IAA requirements of pre-school children, as observed by the workers at INCAP (Pineda et al, 1981) (Table 1B). Since in the pre-school child growth represents only some 10% of the requirement, it is reasonable to extrapolate these results to adults.

Is it possible to reconcile these two opposing points of view? On a protein-free diet or fasting the obligatory N loss (ONL) must reflect the pattern of tissue proteins, since the only source of nitrogen is from protein breakdown. In this situation one may suppose that the IAA with the largest consumption pathway 'drives' the obligatory loss. If protein breakdown provides more of a particular IAA than is needed to make good its loss through irreversible pathways, this extra amount must nevertheless be oxidized. When protein is fed, the pattern changes and the IAAs are only needed in a pattern that balances their losses in the consumption pathways. Thus if fasting and feeding occur in a 12 h cycle, 50% of the 24 h requirement will reflect the composition of body protein and 50% that of the consumption pathways.

The picture, however, is complicated by the diurnal cycling of deposition and loss of body protein (see below). Fed state deposition can be regarded as a form of temporary growth, and therefore requires that amino acids should be provided in the concentrations in which they occur in body protein, although Fuller (personal communication) has suggested that temporary protein storage could have a composition quite different from that of body protein as a whole. To a large extent these amino acids must be derived from protein breakdown; if the food intake is, say, one fifth of the flux, four fifths of the amino acids deposited will be derived from breakdown. At maintenance levels of intake, deposition is relatively small, but at higher intakes, when deposition is increased, the rate of protein breakdown is greatly reduced (section 11), so that intake from the diet becomes more important. Millward et al (1991) have discussed whether amino acids liberated from body protein during the fasting period could be held over, as it were, and be available for meeting needs in the fed state. However, the free amino acid pools, particularly those of the branched chain amino acids (BCAs), are too small for this to be likely. Therefore, in the fed state, particularly with generous intakes, the intake from food plays an essential role in topping up the amino acid supply for protein deposition, and to the extent that it is used in this way, this intake probably must have the pattern of tissue protein.

It follows that conceptually the IAA requirement pattern will be some kind of halfway house between the pattern of irreversible losses and the pattern of body protein, and the relative proportions of the two patterns will be influenced by the level of protein intake. It seems to me that the next step must be to measure the individual IAA losses through irreversible pathways. In the meantime, because the two processes of fasting loss and diurnal cycling tend to shift the maintenance pattern towards that of body protein, it may be considered that Young's proposal represents a reasonable working compromise until more direct observations become available. It should be noted that on Young's hypothesis, if the requirement of one IAA is established, those of all the others follow, whereas if an amino acid, e.g. Lysine, is differentially conserved, the requirement of each IAA must be determined separately.

5.Diurnal cycling: the Millward-Rivers model

As mentioned in section 2, Millward starts from the position that there are three levels of requirement: optimal, operational and minimal. No attempt is made to put numbers to the optimal requirement, and little more can be said about it, except to emphasize that the minimal requirement, which has engaged so much attention, is not the be-all and end-all.

The operational requirement is the amount needed to secure balance at a range of intakes above the minimal. That such a range exists has long been recognized. The contribution of the new model (Millward & Rivers, 1988; Millward et al, 1989) is to suggest that the oxidative losses of amino acids can be divided into two parts, obligatory losses, LO and regulatory losses, LR. The important point is that the regulatory losses are no longer regarded simply as wasted amino acids; rather they have a beneficial and indeed necessary role. This role is the 'anabolic drive', which includes all the effects of amino acids in stimulating the production of hormones such as insulin and in promoting the deposition of protein after a meal (Millward, 1989, 1990). The fact that the replacement of the obligatory losses (50-60 mg N/kg) requires an intake that is almost twice as great (100 mg N/kg) has previously been taken as an effect of inefficiency, the nature and causes of which have never been defined. For Millward, LR makes a necessary contribution to the control of protein metabolism although he now (personal communication) considers it not very important in adults. Nevertheless a state in which oxidative losses were reduced to a level that exactly compensated for the obligatory loss would not be a desirable one. Thus 'a scoring pattern based on minimal obligatory needs is of little practical use' (Millward, 1992).

The consequence of different patterns for growth and maintenance (section 4) is that in the fed state, even at minimal levels of protein intake, some IAAs may be provided in excess of needs. Millward et al (1990, 1991) have pointed out that the branched chain and aromatic IAAs are potentially toxic; their concentrations cannot be allowed to rise and any excess has to be disposed of by oxidation, through high capacity highly regulated oxidative pathways. One might suppose that although the oxidative enzymes are readily inducible, teleologically it would be inadvisable for the expression of them ever to be reduced to zero. This could be regarded as a further reason for the so-called inefficiency of amino acid utilization.

A second contribution of the Millward-Rivers model is that it emphasizes the importance of the diurnal cycle of fasting and feeding. Clugston & Garlick (1982) originally showed that deposition of protein during feeding is balanced by loss in the fasted state. Millward predicted from his model that the amplitude of the fasted-fed swings would increase with increasing protein intake, and his recent studies have shown that this is indeed the case, as shown in Figure 2 (Pacy et al, 1994). As Millward (1992) has said 'The increasing fasting loss with increasing protein intake generates an increasing demand for fed-state protein deposition to balance these losses'. Thus it is impossible to define any particular figure for the operational requirement. However, his dictum 'the more you eat the more you need' seems to imply that the intake can never catch up with the need, which is the logic of Achilles and the tortoise. This must be regarded as artistic license - the intake certainly can overtake the need. Millward's own 24 h balances are positive at high intakes; the same problem that has always plagued conventional nitrogen balance studies (Hegsted, 1976).

Although the mechanism is not understood by which the balance of body protein is maintained over the cycle of feeding and fasting, a good deal is known about the kinetic changes that characterize the cycle. In the original studies of Clugston & Garlick (1982) feeding appeared to cause an increase in whole body synthesis, but it was later suggested that this was an artefact resulting from recycling of amino acids during a prolonged 24 h infusion. The subject has been reviewed by Pacy et al (1994) and by McNurlan & Garlick (1989), and is discussed further in section 12. Most workers (e.g. Nissen & Haymond, 1986) now believe that the main effect of food is to decrease the rate of protein degradation through the combined influences of amino acids, insulin and perhaps other hormones (Millward et al, 1990).

The other component of the diurnal cycle is amino acid oxidation. The fasting oxidation rate is influenced by the preceding diet, as shown in Figure 3, from Price et al (1994). It increased almost 2-fold when the intake rose from 0.36 to 2.3 g/kg/d. There has been some disagreement over fasting oxidation. The MIT group, in their early studies, took the view that it remained constant when the habitual amino acid intake varied. It is apparent from Figure 3 that the change in fasting oxidation rate is only about 7.5 mmol/kg/h when the change in protein intake is 1 g/kg/d. The MIT group were working over a narrower range of leucine intakes than Price et al (section 8) so that, given the variability, the relation to habitual intake was not apparent. The difference between fed and fasted oxidation rates in Figure 3 represents the effect of food. As Millward predicted from his model, the difference increases with increasing intake.


Figure 2
Diurnal changes in nitrogen balance in subjects habituated to four levels of protein intake.


Figure 3
Leucine oxidation in the fed and fasted states in subjects habituated to four levels of protein intake. Data from Table 8.

This work on the diurnal cycle is important and relevant It firmly implicates the processes of protein synthesis and breakdown in the regulation of protein balance. Previously it would have been possible to regard the twin cycles (synthesis-breakdown and input output) as operating to a large extent independently (Waterlow, 1994), but not any more.

6. Theoretical basis of the MIT tracer balance studies

So far we have been concerned more with ideas than with numerical estimates of requirements. We come now to the tracer balance studies of Young and coworkers (Young et al, 1989). The new MIT pattern was supported by two general propositions. The first is that the amount of the IAA requirement is determined by the obligatory N loss (ONL) and that the pattern reflects the composition of body protein. 'The oxidation rates of individual amino acids occur in proportion to the pattern or concentration of amino acids in mixed body protein' (Young et al, 1989). Thus, if the ONL is 54 mg/kg/d (FAO/WHO/UNU, 1985), equivalent to 0.34 g protein, and if the leucine content of body protein is 80 mg/g, then the loss of leucine will be 27 mg/kg/d. Of course, a factor also has to be applied for the efficiency with which dietary amino acids are able to replace this loss. I have already argued (section 4) that this reasoning is doubtful and that the adult requirement pattern of the IAAs cannot be determined simply by the composition of body protein.

The second general proposition is that the ONL represents that proportion of total protein turnover that is not recycled back into protein synthesis. We know from tracer studies that, at intakes in the region of maintenance, recycling is about 90% (Waterlow, 1968). If the ONL of 54 mg/kg/d represents the 10% of protein turnover that is not recycled, then total protein turnover must be 540 mg N or about 3.4 g protein/kg/d. This estimate fits well with actual values obtained by various methods (Waterlow, 1984). However, the argument tends to be circular, because although it is consistent with what we know about protein turnover, it is still based, as before, on the ONL and the composition of body protein. It is therefore not surprising that the two approaches give identical results, give or take a milligram or two for rounding off, as shown in Tables 7 and 8 of Young et al (1989).

These criticisms, like the arguments themselves, are really irrelevant. The MIT pattern stands or falls by the extensive body of well planned and well executed studies that have been carried out by Young and his colleagues over more than a decade. The original tracer balance studies of 1986 were followed by others designed to eliminate possible sources of error and to extend the data base.

7. Technical problems of tracer balance studies

The general procedure in these studies has been for subjects to be placed on a diet in which nitrogen is provided as crystalline amino acids to provide about 0.8 g amino acids per kg, with the level of the test amino acid only being varied. The subjects are kept on each test level for 6-7 days. In the most widely used protocol, after an overnight fast, they are given a primed constant infusion for 8 h of amino acid labelled with 13C in the carboxyl carbon. For the first 3 h the subjects remain fasting, and in the third hour, when isotopic equilibrium has been reached, measurements are made of isotopic enrichment in plasma and expired CO2. For the remaining 5 h the subjects are fed at frequent intervals and the measurements repeated in the final hour. Total CO2 output is determined by standard indirect calorimetry to give the amount of amino acid oxidized.

As mentioned above, early studies were made with Lysine, valine and threonine, as well as with leucine. These results are difficult to interpret, since no fasting measurements were made and there was no correction for precursor activity (see below). Moreover with Lysine and threonine the first step in oxidation does not involve the carboxyl carbon. However, the important point at this stage is to be satisfied about the validity of the method, and for this one amino acid is enough. The following remarks are therefore mainly confined to leucine.

7.1. Mass of tracer

Since stable isotopes are not mass-less, it is necessary in balance measurements to correct for the input from the infusion, particularly in the fasting state and at low intakes. It is unclear whether this was always done, or done correctly, in the early studies.

7.2. Retention of CO2

A proportion of the CO2 derived from oxidation of the labelled amino acid is not excreted in the breath but retained in the body in a form that is not known. The recovery varies a little with feeding state, duration of infusion (Table 4) and exercise (Wolfe et al, 1982; Barstow et al, 1990). The recovery factor is determined beforehand by a separate primed infusion of NaH13 CO3. Use of the same factor from one experiment to another could involve a small error.

Table 4 Percentage recovery of labelled CO2 in constant infusions

Author

Fasting

Feeding


Clugston & Garlick (1983)

0.94

0.88

12 h fasted, 12 h fed

Elia et al (1992)

0.86

3-6 h of infusion

Wenham et al (1991)

0.76

0.90

fasted 0 5 h, fed 4-8 h

Hoerr et al (1989)



fasted 0-4 h, fed 4-8 h


0.70

0.82

intravenous infusiona


0.74

0.79

intragastric infusion

Melville et al (1989)


0.56-0.76

during final 4h of feeding

El Khoury et al (1984a)

0.77

0.85

12 h fasted, 12 h fed

a The difference between i.v. and i.g. infusions is not significant.

It has also been suggested that the recovery from infused bicarbonate may not be a valid measure of the recovery of carbon generated by intramitochondrial enzyme activity. The evidence on this point is conflicting (Marsolais, 1987; Hamel et al, 1993), but the effect, if it exists, is not large.

7.3. Precursor enrichment

It has long been recognized that the labelling of the intra-cellular amino acid, which is presumably the precursor of both protein synthesis and oxidation, is lower than that of the amino acid in plasma, because of dilution with unlabelled amino acids derived from protein breakdown (Aub & Waterlow, 1970). If the intracellular labelled amino acid produces a metabolite which appears in the plasma, the labelling of that metabolite can be taken as that of the amino acid from which it was derived. Thus Fern & Garlick (1974) used the labelling of plasma serine as a measure of the labelling of intracellular glycine which was its precursor. This has come to be known as the 'reciprocal pool' method. For leucine the keto acid produced by transamination, i.e. 1-oxoisocaproic acid (KIC), is the precursor of oxidation, and the labelling of KIC in plasma is regarded as giving a better estimate of the precursor activity than that of the amino acid itself (Matthews et al, 1980, 1982; Schwenk et al, 1985). The ratio APE* of KIC/APE of leucine is usually in the region of 0.7-0.8 (Table 5), but is somewhat variable, with no clear relation to protein intake or fasted/fed state. The keto acid of valine, 1-oxoisovaleric acid (KIV), can be used in the same way, but suitable precursors for the oxidation of Lysine and threonine have not yet been used in tracer balance studies, although Arents & Bier (1991) have suggested that a-amino-adipic acid in urine might be a suitable reciprocal for Lysine. For phenylalanine the oxidation rate can be obtained from the rate of hydroxylation to tyrosine.

* Atom percent excess.

With these and other amino acids, where in contrast to the BCAs the liver is the main site of oxidation, we know from animal experiments (Simon et al, 1978; Lobley et al, 1980) that the isotope abundance of amino acids in the liver may be as little as half that in plasma, but liver biopsies are unlikely to be possible in man. A solution to the problem now seems to be available, based on the finding that apolipoprotein B-100 is synthesized so rapidly in the liver that it reaches isotopic equilibrium with its precursor amino acids in a matter of hours (Reeds et al, 1992). Some of the results obtained by this method are shown in Table 6. Thus, on the assumption that the products of the same precursor have the same enrichment (Waterlow et al, 1978), apo-B can now be used to give a. measure of the precursor activity in the liver of a range of amino acids.

Table 5 Ratio of enrichment of KIC to enrichment of leucine: isotope administered intravenously


Fasted

Fed

Not stated

Pelletier (1991)

0.79

0.69


Hoerr (1991)

0.72

0.79


Reeds (1992)

0.91

0.92


Thompson (1989)



0.77

Bido (1992)



0.81

Price (1994)



0.69

Cortiella (1988)




20-40 mg leucine



0.71

10 mg leucine



0.63

Hoerr (1993)




High protein



0.82/0.77a

Low protein



0.88/0.67a

a Infusions with 2H3-leucine. All others with 13C-leucine.

However, this does not solve the precursor problem. If the enrichment of the tissue amino acid is lower in liver, for example, than in muscle (as many workers, though not all, have found in animal experiments), then no single value for the enrichment of a metabolite such as KIC in plasma can accurately reflect the enrichment of the precursor in both liver and muscle. It is difficult to predict the extent of the error that may arise from this source. One might hazard a guess that, with KIC, the enrichment in plasma is likely to be lower than that of the precursor in muscle and higher than that in liver. The consequence would be that, by using the plasma value across the board, oxidation in muscle would be over-estimated and in liver it would be under-estimated, but these two errors would tend to cancel out. The balance will also vary according to the tracer that is used and its main site of oxidation. For a more extended discussion of the precursor problems see Waterlow (1995).

7.4. Extrapolation from the period of measurement to 24 hours

In the standard protocol of infusion for 8h, 3 h fasted and 5 h fed, measurements made in the last hour of each period are extrapolated to a whole day. Marchini et al (1993) proposed two methods for doing this, which differ according to the weight given to the oxidation of tracer in the fasted state. In my opinion, which agrees with that of Fuller & Garlick (1994), the correct method is Marchini's method 1. This involves calculating the fasting balance as: (infusion - observed oxidation) × 12, and the fed balance as (infusion + dietary intake -observed oxidation) × 12. The 24 h balance is the sum of these two. Marchini's method 2, with a tracer infusion rate of 4 mmol/kg/h, gives a more positive balance of 64 mmol, or about 8 mg/kg/d, which, at low intakes is far from negligible.

Table 6 The ratio of isotopic enrichment of amino acids in apolipoprotein B100 to that of free amino acids in plasma


Fasted

Fed

Leucine

0.90/0.94

0.69/0.72

Lysine

0.73/0.74

0.64/0.64

Phenylalanine

1.07/1.03

0.79/0.86

Data of Reeds et al (1992). The pairs of numbers represent results from two experiments.

Table 7 Comparison of measured and predicted leucine oxidation rates. Data of El Khoury et al (1994a,b)


Leucine oxidation


Leucine intake (mg/kg/d)

Measured (mmol/kg)

Predicted (mmol/kg)

Difference predicted-measured (%)

89 Fast 12 h

34.7

32.5a


Fed 12 h

54.8

58.7b


Sum 24 h

89.5

91.2

+ 1.9

38 Fast 12 h

20.9

25.5a


Fed 12 h

24.3

25.0b


Sum 24 h

45.2

50.5

+ 11.7

14 Fast 12 h

15.0

18.7a


Fed 12 h

12.8

14.2b


Sum 24 h

27.8

31.2

+12.2

a From last hour of fasting × 12.
b From fifth hour of feeding × 12.

In a praiseworthy attempt to solve this problem, El Khoury et al (1994a,b) did two heroic experiments in which leucine was infused for 24 h in order to compare the measured 24 h oxidation with that predicted from the shorter protocol. On the whole, agreement between measured and predicted rates was satisfactory (Table 7), with the prediction tending to overestimate the oxidation rate. Figure 4 shows that there is a temporal pattern of 13CO2 output in both fed and fasted periods, particularly with low intakes, so that the predicted oxidation is critically dependent on which hour is chosen to be representative of the whole period. It would seem that the choice, imposed by the protocol, of the fifth hour of feeding, was rather fortunate, since at this time the 13CO2 output more or less represents the mean, but it is the mean of quite wide variations, e.g. for the 38 mg intake, from about 12 down to about 5 mmol/kg/30 min.

Figure 4 Pattern of 24-h leucine oxidation rate for subjects receiving 38.3 mg (a) or 14 mg (b) leucine/kg/d. Mean ± s.d. Minutes (0-1440) refer to experimental time; time of day is indicated by hours. Reproduced from El Khoury et al (1994) by courtesy of the authors and the publishers of the American Journal of Clinical Nutrition.


Pattern of 24-h leucine oxidation rate - a


Pattern of 24-h leucine oxidation rate - b

8. Results of the MIT tracer balance studies

8.1. Initial studies (1986)

The initial tracer balance studies were concerned with leucine (Meguid et al, 1986a), Lysine (Meredith et al, 1986), valine (Meguid et al, 1986b), and threonine (Zhao et al, 1986). The test amino acids were fed at six to nine different levels. The estimates of IAA requirements derived from these experiments were given in Table 1A. The problem with these studies is that no correction was made for precursor enrichment; it is not clear whether in calculating the balance the dose of tracer was included on the input side; and oxidation in the fasted state was not measured directly but estimated from oxidation at the lowest level of amino acid intake. Even a small error in the estimate of fasting oxidation can have quite an important effect on the calculated balance. This can be corrected retrospectively from later MIT studies in which fasting oxidation was measured directly. Those results are shown in Figure 3 and Table 8. In Table 9 the left column gives the balances recorded in the original papers; in the right-hand column I have attempted to make corrections for these various sources of error. For valine and threonine the recalculation makes very little difference to the estimated intake for zero balance. For leucine it increases the estimate; for Lysine the balances become more negative.

I do not claim that these recalculated estimates are any more valid than the original ones. I think that the lesson from this exercise is that quite small variations in fasted oxidation may have quite a profound effect on the tracer balance.

Table 8 Fed and fasted oxidation rate: data from the MIT studies


Oxidation rates


Leucine intake (mg/kg/d)

Fasted (mmol/kg/h)

Fed (mmol/kg/h)

Difference fed-fasted

Reference

120

22a

57a

+35

Motil (1981a)

80

14a

34a

+20

Young (1987)

80

18

32

+ 14

Marchini (1993)

48

28a

27a

+ 1

Motil (1981a)

40

17.5

15

+2.5

Hiramatsu (1994a)

40

19

25

+6

Pelletier (1991b)

40

14

18

+4

Marchini (1993)

30

15a

26a

+11

Young (1987)

14

14

9

- 5

Hiramatsu (1994a)

14

15

16.5

+ 1.5

Pelletier (1991)

14

13

9

- 4

Marchini (1993)

14

7.5a

11a

+3.5

Young (1987)

8

16a

15a

- 1

Motil (1981a)

7

12a

15a

+ 3

Young (1987)

In this table, where several studies have been done at the same intake, the results have been averaged.
a Oxidation rates corrected on the assumption that APE KIC/APE leucine = 0.8, since APE KIC was not measured in the original studies.

Table 9 Original and recalculated balances from MIT studies (1986)



Balance (mg/kd/d)


Intake (mg/kg/d)

Original

Recalculated

Leucine

60

+ 18.0

+ 11.8


40

+2.6

-2.2


20

-4.6

+0.5


16

-8.6

-5.6


12

-9.6

-6.1

Lysine

58

+ 16.0

- 1.4


35

+ 12.6

0


30

+3.9

-9.8


20

-4.4

- 12.4


12

-4.8

- 15.7

Valine

40

+ 13.0

+7.2


20

+5.4

+7.2


16a*

+4.1

+6.6


16b*

+ 1.3

+ 3.5


12

-2.6

-0.6


8

-3.9

-0.4

Threonine

40

+ 25

+ 11.4


30

+ 10.5

+ 3.3


20

+8.2

+5.9


10

-0.5

- 1.7


3

-7.1

-8.9

Recalculated on the assumptions:
1. Fed-state oxidation rates divided by 0.8 as a correction for precursor activity.
2. Fasted-state oxidation rates calculated from values shown in Table 8; taken as 16 mmol/kg/h for intakes above 20, and as 13 mmol/kg/h for intakes of 20 or below. Infused tracer deducted from these estimates of oxidation in calculating balances.
The values used in the original paper were (mmol/kg/h); leucine 7, Lysine 5, valine 4, threonine 5 for net oxidation after deducting input of tracer.
* a and b are replicate studies.

8.2. Further MIT studies with leucine

A number of other studies have been undertaken at MIT to validate the original results and to meet certain criticisms. All these were done with leucine and the calculations based on the enrichment in plasma of a-ketoisocaproic acid (KIC), the transamination product of leucine, as a more valid measure of precursor activity. In most cases the protocol was extended to an 8 h infusion, with 3h in the fasted and 5h in the fed state. In the results quoted here, extrapolation to 24 h has been done by the preferred (Marchini et al, 1993) method 1, so there may be some differences from the figures given by the authors.

(i) The experiments of Cortiella et al (1988) were designed primarily to compare results obtained with intravenous and intragastric tracers. This aspect is considered in section 10. Measurements were made at four levels of leucine intake, but fasting oxidation was determined only at the lowest intake (10 mg/kg/d). This study differed from previous ones in that it was the oxidation of the intragastric tracer that was measured. The results, as far as concerns the tracer balances, are shown in Table 10.

(ii) Young et al (1987) measured leucine balance after 1 week on either the FAO or the MIT level for leucine (see Table 1). The special point about this experiment is that the measurements were repeated after 3 weeks on the diets, and again during recovery. This part of the study will be discussed later under the heading 'adaptation'. Here I report only the results after the usual period of 1 week on the diets (Table 11).

(iii) The criticism had been made that in the amino acid diet modelled on egg the relatively high amounts of valine and isoleucine would stimulate the branched chain dehydrogenase complex and artificially elevate the rate of leucine oxidation. Pelletier et al (1991a,b) made measurements in both fed and fasted states at the FAO and MIT levels of leucine requirements (14 and 38 mg/kg/d). Varying the concentrations of valine and isoleucine in the diet had no effect on leucine oxidation at either level of intake. The dehydrogenase complex appears to be more sensitive to leucine or KIC than to the other branched chain amino acids. In a separate experiment oxidation of valine was measured (from the APE of the precursor KIV) and found to be unaffected by leucine intakes between the minimum physiological requirement and what would be supplied by a good quality diet.

Table 10 Results of Cortiella et al (1988) with graded levels of leucine intake given intragastrically

Leucine intake (mg/kg/d)

Balance (mg/kg/d)

40

+0.2

30

-8

20

-15

10

-24

13C-leucine was given intragastrically and results calculated from the APE of 13C-KIC in plasma.
Fasted-state oxidation was measured only at the 10 mg intake and amounted to 15 mmol/kg/h. This value, less the tracer, has been used for calculating the balances at the other levels.

Table 11 Leucine balances at FAO (14 mg/kg/d) and MIT (38 mg/kg/ d) levels of intake
Group 1: studies in which the total amino acid intake was kept at a level corresponding to about 0.8 g egg protein/kg/d, and only the level of the test amino acid was altered.
Group 11: studies in which all the IAAs were fed in amounts equivalent to the FAO or MIT pattern.


Leucine balance (mg/kg/d)


FAO

MIT

Group I

Young (1987)a

- 9

- 32

Pelletier (1991a,b)



low valine

- 13

- 13

high valine

- 13

- 19

low isoleucine

- 12.5

- 9.5

high isoleucine

- 20

- 11

El Khoury (1994)b

- 13

- 6

Mean (unweighted)

- 13.4

- 15.1

Group II

Hiramatsu (1994a)



low NENc

- 8

-

high NEN

- 9

+ 5

low glutamine

-

- 0.5

high glutamine

-

- 5

Marchini (1993)

- 5

- 0.5

Mean (unweighted)

- 7

- 2.5

a Corrected assuming precursor factor of 0.8.
b 24 h infusions.
c non-essential nitrogen.
All studies were made after 6 days on the diet. Except for that of El Khoury et al, the protocol involved an 8h infusion: 3h fasted followed by 5 h fed with measurements of oxidation during the last hour of each period.

(iv) The experiments of Hiramatsu et al (1994a) were designed to test the effect on the oxidation of the IAAs of altering the dietary intake of non-essential nitrogen. Leucine, phenylalanine and tyrosine kinetics were studied, but only the results with leucine will be reported in this section.

The first phase of the study involved four diets; in the first two all the IAA were fed at the FAO level and the nonessentials were provided in amounts that would bring the total N to the equivalent of either 0.65 or 1 g protein/kg/d. With the next two diets the same procedure was followed but all the IAAs were fed at the MIT level. In a second study the IAAs were given at the MlT level and the total N was brought up to the equivalent of 1 g protein/kg/d with varying amounts of glutamine, ranging from 0 to 100% of the non-essential nitrogen. The glutamine content of the diet made no difference to the leucine balance. The results are shown in Table 11

(v) Marchini et al (1993) adopted the same procedure as Hiramatsu et al (1994a) in which all the IAAs were fed at either the FAO or the MIT level. A control group received the egg pattern of amino acids. The total N intake was made up to the equivalent of 0.9 g protein/ kg/d. The ratio of IAAs to total AAs was 0.11 for the FAO diet, 0.21 for MIT diet and 0.53 for the egg diet. As with the study of Young et al (1987), the diets were fed for 3 weeks and measurements of leucine balance made at 1 week, 3 weeks and after recovery; 80 infusions in all, a phenomenal experiment. Only the findings at 1 week are reported here (Table 11). Those at 3 weeks are discussed under the heading of 'adaptation'.

(vi) Finally, we have two papers by El Khoury et al (1994a,b) in which the infusion of leucine was given for 24 h 12 h fasting and 10h fed, followed by a further 2h in which no food was given. In the first study leucine was provided at a single generous level (80 mg/kg/d). The object of the experiment was to check a number of technical points: the baseline 13CO2 enrichment was measured for 24 h to check the isotope abundance of carbon in the food, and a correction applied for this baseline; new correction factors were calculated for 13CO2 recovery during fasting and feeding (see Table 8); the 24 h infusion allowed a check on the accuracy of extrapolation over 24 h from the shorter infusion protocol of the previous studies. The observed oxidation of leucine over 24 h showed satisfactory agreement with that predicted from the third hour of fasting and the fifth hour of feeding. Finally, the opportunity was taken to measure urea production with 15N15N urea and to compare the oxidation of leucine with that of total N (see section 9).

The second paper by El Khoury et al explored further the differences between the FAO and MIT leucine intakes when the infusions were continued for 24 h For both levels of intake the 24 h oxidation predicted from the third hour of fasting and the fifth hour of feeding was 12% greater than the measured oxidation over 24 h. Somewhat better agreement was obtained when the fasting oxidation was taken as that observed 12 h after the last meal (Table 7).

These studies show very beautifully the temporal changes with fasting and feeding in the enrichment of plasma KIC and of CO2 in breath (see section 7, Figure 4). The magnitude of deviations from baseline was greater, the higher the leucine intake, in broad agreement with the predictions of Millward (1992) and the findings of Price et al (1994).

8.3. Tracer balances with other amino acids

Lysine. In the experiments of Meredith et al (1986) the requirement for zero balance was between 20 and 39 mg/kg/d. Halliday & McKeran (1975) found that with unprimed infusions it took some 12 h for Lysine labelling to reach a plateau, and this is the order of time that one would expect from the large size of the free Lysine pool (Waterlow & Fern, 1981). It seems odd that in Meredith's experiment, with a prime equivalent to 2.5h of infusion, plateau should have been reached in 2h. Perhaps there is a large pool of free Lysine which mixes only slowly with the pool into which the infusion is given, producing a situation like that found with glutamine (Daumann et al, 1986). However, the flux and synthesis rates calculated from Meredith's data are entirely reasonable.

The only other isotopic study of Lysine requirement in man* is that of Zello et al (1993), which was based on what may be called a 'reverse breakpoint' analysis (see section 11). In this system phenylalanine was used as an indicator amino acid and its oxidation measured at different levels of Lysine intake. The idea is that at inadequate Lysine intakes the rate of uptake of phenylalanine into protein will be reduced and that of oxidation will be increased; oxidation will fall to a minimal level when the Lysine intakes become adequate. Analysis of the data fitted by least squares produced two straight lines which met at a Lysine intake of 37 mg/kg/d.

* Studies of Lysine turnover in relation to protein intake have been made by Conway et al (1980) and Motil et al (1981a,b)

A criticism of this approach might be that the test diets were only fed for the day on which measurements were made, since Zello maintains that a period of adaptation is unnecessary. However, in confirmation of his result, calculation of protein synthesis rates from the data of Meredith et al (1986) show an abrupt fall at Lysine intakes below 35 mg/kg/d (section 12; Figure 9).

Threonine. Table 9 shows very low rates of oxidation of threonine, so that the requirement for zero balance is only of the order of 10 mg. My calculation assumes a precursor correction of 0.8. Ballevre et al (1990), working on young pigs, found a value of 0.76 for the ratio of specific radioactivity of threonine in liver to that in plasma, so that calculation should not involve a serious error. However, threonine is catabolized by two routes: (1) through ketobutyric acid to propionic acid + CO2 and (2) to glycine and acetyl CoA. In the pigs of Ballevre et al (1990) on high threonine intakes of about 150 mg/kg/d, 80% of threonine oxidation was through the glycine pathway, and an increase in oxidation with increasing intake occurred almost exclusively through this pathway. If the same conditions hold in man, CO2 production will seriously under-estimate threonine oxidation, so that even the requirement of 15 mg/kg/d suggested by Young et al (1989) may be an underestimate. It is noteworthy also that intestinal secretions are rather rich in threonine. Measurements on ileostomy fluid showed a daily loss by this route of 300 mg, or 5 mg/kg/ d (Fuller et al, 1994).

Table 12 Valine balance at different levels of valine and leucine intake. Data of Pelletier et al (1991a)


Valine intake (mg/kg/d)


10

20

Leucine intake (mg/kg/d)

Valine balance (mg/kg/d)

40

+ 0.2

+ 1.7

80

4.8

+ 1.9

Valine. The study of Meguid et al (1986b) showed a valine requirement of about 16 mg/kg/d (Table 9). Pelletier et al (1991a) measured valine balance at two levels of valine intake, 10 and 20 mg/kg/d, and two levels of leucine intake, 40 and 80 mg/kg/d. For precursor activity they used the labelling of 1-keto-isovaleric acid in plasma, so there can be no criticism of the results on that score. The results tend to confirm those of Meguid et al. At the lower intake of valine the higher level of leucine significantly depresses the valine balance (t = 3.07).

Methionine. Two recent papers provide new information on the requirement for methionine. In both of them the precursor factor was taken as 0.8. In the first study (Young et al, 1991) methionine was fed at the single level of 13 mg/kg/d, and balance was just achieved. In the second experiment (Hiramatsu et al, 1994b) methionine was fed at two levels, 13 and 6.5 mg/kg/d, and at the lower level cystine intake was varied from 0 to 21 mg/kg/d. Table 13 shows the balances extrapolated by the authors to 24 h by a method that I consider incorrect, compared with those recalculated correctly. This is a good illustration of the uncertainty introduced by the method of extrapolation.

Phenylalanine. The first step in the catabolism of phenylalanine is its hydroxylation to tyrosine. Clarke & Bier (1982) established the method for measuring phenylalanine hydroxylation, and Thompson et al (1989) suggested that it would be a useful method for studying whole body protein turnover in situations where the collection of expired CO2 was not possible. An alternative approach is to measure the production of 13CO2 from phenylalanine labelled in the 1-carbon position. It seems that this oxidation only occurs after the conversion of phenylalanine to tyrosine. However, if say 20% of the phenylalanine flux is oxidized and 20% of the tyrosine flux is oxidized, the overall oxidation would be only 4%, and 13CO2 production would grossly underestimate phenylalanine oxidation. Zello et al (1990) therefore proposed that the tyrosine derived from phenylalanine was preferentially oxidized, without mixing with the free pool of tyrosine. This introduces an additional complication, particularly since the results of Cortiella et al (1992) showed that the oxidation rate estimated from 13CO2 was 10-15% lower than the rate of hydroxylation.

Table 13 Methionine balance. Data of Young et al (1991) and Hiramatsu et al (1994b)



Balance (mg/kg/d)


Intake (mg/kg/d)

Original calculation

Recalculated

Study 1

13

- 0.7

+ 6.6

Study 2

13

+0.7

+3.5


6.5

-1.3

+2.1

Recalculated by correct method of extrapolation (see section 7.5)

In the opinion of Millward (personal communication) Zello's data are uninterpretable.

There are three studies which allow calculation of the tracer balance from the hydroxylation rate: those of Cortiella et al (1992); of Hiramatsu et al (1994a); and of Price et al (1994). The results that we can glean from them are shown in Table 14.

In none of the balances listed in this table has a correction for the precursor been applied. Price et al (1994) calculated that the hydroxylation rates should be divided by 0.69 in the fasted and 0.56 in the fed state. Application of these factors, however, leads to impossibly negative balances. Table 14 suggests that the requirement for phenylalanine for balance is in the region of 20 mg/kg/d, when tyrosine is provided in the same proportion as it occurs in protein.

In addition, Zello et al (1990) have carried out a study based on the principle of breakpoint, which does not require an estimate of precursor activity (section 11). Phenylalanine flux and oxidation rates were measured at seven levels of phenylalanine intakes, from 5 to 60 mg/kg/d, with generous amounts of tyrosine. The subjects were on the test intake for only the day of the measurement; no period of adaptation was considered necessary. When oxidation was plotted against intake, there was a breakpoint at 9 mg/kg/d. Plasma phenylalanine concentrations showed a breakpoint at a similar level. The 1985 FAO/WHO/UNU report estimated that 1/3 of the requirement for the aromatic amino acids must be provided by phenylalanine. If the same ratio is applied to the estimate of Zello, the aromatic amino acid requirement would come out at about 30 mg/kg/d.

8.4. Summary of the results

When one compares the FAO and MIT levels (Table 11), the studies fall into two groups: in those of Cortiella (1988) (Table 10), and of Young (1987), Pelletier (1991a,b) and El Khoury (1994b) (Table 11), only the leucine intake, or that of the branched-chain amino acids was altered. By contrast, in the experiments of Hiramatsu (1994a), and Marchini et al (1993) the whole IAA pattern was fed at two different levels. The balances in the first group are consistently more negative than in the second group. Moreover, in group I there is no indication of any difference in balance between the FAO and MIT patterns, whereas in group II the MIT pattern does have an advantage, although I have not tested it statistically.

Table 14 Phenylalanine balance, based on measurements of hydroxylation. Data from Price et al (1994), Cortiella et al (1992) and Hiramatsu et al (1994a)


Phenylalanine




Intake (mg/kg/d)

Balance (mg/kg/d)

Nitrogen intake (mg/kg/d)


Price

80

+ 32

254

Balanced protein

Cortiella

54

+ 18

160

Balanced protein

Price

37

+ 11

123

Balanced protein

Hiramatsu

26

+ 1.6

107

MIT pattern


26

+6.6

160

MIT pattern + NENa

Cortiella

22

- 1.4

160

Low Phe/Tyr


22

+ 9.2

64

Low Phe/Tyr, low total N

Price

16

- 11

57

Balanced protein

Hiramatsu

8

- 8.2

107

FAO pattern


8

- 12.2

160

FAO pattern + NENa

In none of these studies has a correction been made for the precursor. All calculations by method 1 (section 7.5).
a Non-essential nitrogen.

The difference between the results with the two experimental designs suggests that when the intakes of the IAAs, other than the one under test, are high, the activity of the various oxidative enzymes is maintained at a high level across the board, so that oxidation of the test amino acid is increased. The experiments of Cortiella et al (1992) also showed some indication of sparing of oxidation when a diet low in total N and all IAAs was compared with one in which only phenylalanine and tyrosine were low. On the other hand, the experiments of Pelletier and of Hiramatsu show that the high rates of leucine oxidation cannot be explained either by excessive amounts of BCAs or by the level of NEN.

Young (personal communication) has commented: 'An analysis across our multiple experiments of leucine oxidation during the fed state reveals a linear relationship between intake of leucine and its oxidation, with no significant effect of level of N intake, source of nonspecific N and the amino acid patterns involved in these experiments. Hence I do not think that the available data support the implication proposed here'. I do not accept this conclusion, because, as I have pointed out above, fasted state oxidation, not included in this analysis, has an important effect on the balances. Moreover, as Table 11 shows, if Young's argument is correct it weakens his advocacy of the MIT pattern.

Another problem in interpreting the results is the between-subject variation within each experimental group. For example, in the experiment of Marchini et al (1993), with n = 6 or 7, for the FAO diet after l week the s.d. of leucine balance was 6.5 mg/kg/d, and for the MIT diet 13 mg/kg/d (Table 7 in Marchini et al, 1993). On the MIT diet this amounts to 34% of the intake, which means that it will be difficult to get a safe level (mean requirement + 2 s.d.), according to the principles applied to total protein by FAO/WHO/UNU (1985). This kind of variability seems to be almost unavoidable in human studies.

9. Relation between leucine oxidation and nitrogen excretion

It would help to validate estimates of leucine oxidation if parallel changes were found in N excretion as a measure of total amino acid oxidation. Prediction of the oxidation of total N from that of leucine has traditionally relied on the assumption that leucine represents 8% of body protein. As far as I can make out, this value was introduced into the isotope literature by Golden & Waterlow (1977), based on the old analyses of Block & Weiss (1956). It corresponds also to the leucine content of beef muscle (FAO, 1970) and that of the human fetus (Widdowson et al, 1979). Since the body contains about 20% of structural protein that includes very little leucine, the value of 6.5% proposed by Reeds & Harris (1981), based on carcass analysis, is probably not appropriate, although Young has used it in a recent paper (1991)

The data of Reeds & Harris (1981) in pigs show excellent correspondence between measured balances and those predicted from leucine. However, in man Clugston & Garlick (1982) were able to account for only 68% of total N excretion from their measurements of the oxidation of leucine. Young et al (1987) measured N balances in the third week on diets providing, 7, 14 and 30 mg leucine/kg/d and again after repletion. There was qualitative agreement with the leucine balances, but no attempt was made at a quantitative comparison. In the experiments of Marchini et al (1993) N balance as well as leucine balance was measured on the FAO, MIT and egg diets. No significant differences were found between 1 and 3 weeks on the diets (cf. Young et al, 1987), so the results have been averaged in Table 15. The leucine balances put the results in the expected order, which the N balances hardly do, so this may justify the authors' claim for the 'poor reliability' of N balances. However, the s.ds of both sets of balances are so large that the difference between the FAO and MIT diets is significant in only one comparison. Furthermore, there are disturbing quantitative discrepancies between predicted and observed balances.

Price et al (1994) investigated this point in some detail, using the figure of 8% for the leucine content of body protein. Their results are shown in Figure 5. The ratio of predicted to measured N loss was the same over the whole range of protein intakes, with a mean of 0.79. The measured N loss included estimated faecal and miscellaneous losses. It could be argued that in this context it would have been more appropriate to include only the amounts of urea and ammonia N excreted. This would bring the ratio closer to 1. The predicted loss was calculated from leucine oxidation after correcting for any difference between the leucine content of body protein and of food. The important point is that correspondence, even if not exact, is the same over a wide range of intakes. These data, therefore, provide valuable confirmation of the validity of the leucine balances.

El Khoury et al (1994a), during their 24 h infusions, measured urea production with 15N15N urea, as well as urea excretion. The results in Table 16 show a remarkable correspondence between N intake, observed N output and predicted N oxidation, calculated on the basis that leucine is 8% of body protein. One might have expected a better correspondence between predicted oxidation and urea production (which does not include urea recycled from the gut) rather than urea excretion, since urea production is presumably a measure of the total amino-N oxidized. However this is a controversial point on which Young and Millward are in agreement and I differ (see section 13). As with Price's data, if the faecal and non-urea N are excluded from the calculation, leucine oxidation comes into good agreement with urea production.

Table 15 Average of leucine and nitrogen balances after 1 and 3 weeks on egg, MIT or FAO diets. Data of Marchini et al, (1993)

Diet

Leucine balance (mg/kg/d)

Measured N balance (mg/kg/d)

Predicted N balance (mg/kg/d)

egg

+164

+5

+33

MIT

+2.5

+7

+5

FAO

-10.9

- 0.5

- 22

Predicted N balance = leucine balance × 100/8 × 1/6.25.


Figure 5
Predicted/measured nitrogen excretion.

10. Factors relating to the design of tracer balance experiments

10.1. The model

The most important factor affecting the calculation of amino acid oxidation is the enrichment of the precursor, since the measurement of labelled CO2 output is straightforward. Is there any possibility that the values that have been used for precursor activity are too low? One suggestion of how this could happen is that there is a pool of protein turning over with lifetime kinetics (Waterlow et al, 1978) with an average lifetime of 6-9 h in effect a delay pool (Slevin et al, 1991). This idea arose from the finding of a 'step' in the enrichment curve of urinary ammonia during hourly dosage of 15N-glycine without a prime. The first step or plateau occurred at 6-9 h, the second from 21 h onwards. This kind of behaviour could not be explained by any model in which exchanges of tracer occur by first order kinetics, nor could it be explained by recycling. Later results have confirmed our preliminary observations (Jackson, unpublished). The APE at the first plateau was 30% less than at the second plateau, and would lead to higher estimates of the flux. We have suggested that this pool, which is turning over rapidly by lifetime kinetics, may be associated with the gastro-intestinal tract, but at present these observations are preliminary and speculative.

Table 16 Leucine oxidation and nitrogen excretion. Calculated from data of El Khoury et al (1994a)


N (mg/kg/d

Nitrogen intake

161

Urea N excretion


+ non-urea N


+ faecal N

153

Urea N excretion alone

124

Urea N production

155

Predicted N excretion from leucine balance

160

We have to consider whether an effect of this kind could be occurring in the tracer balance studies. In these studies it is not possible to identify a step in the enrichment curve of the tracer, because all the infusions were primed and the fluxes calculated from the plateau at 2-3 h. The reason it seems unlikely that there is any serious defect in the model is that the fluxes and protein turnover rates calculated from this early plateau are in good agreement with those in the literature, including results obtained with 15N (Waterlow, 1995). However, it would surely be a mistake to suppose that the model currently used is the last word and that no more research is needed to establish the validity of the fluxes derived from it (Bier, 1989).

10.2. Route of administration of tracer; the 'first pass' effect

It has been suggested that it is artificial to give the tracer by vein when what is being measured is oxidation of amino acid in the food. Several investigators have given one tracer by vein and another by mouth (Cortiella et al, 1988; Beaufrere et al, 1989; Hoerr et al, 1991, 1993; Biolo et al, 1992; Matthews et al, 1993). These studies are in remarkable agreement in showing that with leucine as tracer 70-80% of the tracer given by mouth enters the plasma and 20-30% is 'sequestered' in the tissues of the splanchnic bed. Of the latter, about 60% is used for protein synthesis and 40% emerges as KIC (Matthews et al, 1993).

According to the data of Yu et al (1990), in the dog in the fed state 14% of the leucine taken up in the splanchnic bed was oxidized. With phenylalanine the proportions taken up and oxidized in the splanchnic bed were much greater (Biolo et al, 1992; Sanchez et al, 1995).

The significance of what has been called the 'first pass effect' for estimates of flux and oxidation is at present far from clear. According to my calculations (which may not be correct), the route of administration of tracer will have no effect on the estimate of flux from measurements on plasma, provided that the enrichment of the precursor is the same in splanchnic tissues, plasma and peripheral tissues. As discussed in section 7.3, one would expect on the basis of most animal experiments that with intravenous infusion the precursor activity would be lower in liver than in plasma, but we do not know how it will behave with an intragastric infusion. If the enrichment is lower with i.g. than with i.v. infusion, then i.v. administration of tracer will lead to an underestimate of flux. This, in fact, has been found, as Table 17A shows. Unfortunately, the corollary, that precursor activity should be lower with i.g. infusion of tracer, is not so far supported by the data. If with leucine infusions KIC is taken as precursor, Hoerr et al (1991) found virtually identical enrichments of plasma KIC by both routes in both fed and fasted states (Table 17B). There is therefore a puzzle here which has not yet been resolved, but which emphasizes the uncertainty still existing about the appropriate precursor for measurements of whole body turnover. One may also recall the old studies of Golden & Waterlow (1977) in which 14C leucine was infused for 24 h the route of administration being switched from i.v. to i.g., or vice versa, half-way through the infusion, with no effect on the plateau labelling of leucine in plasma. From the practical point of view of the present enquiry, the effect of underestimating the flux in i.v. infusions would be to underestimate oxidation and hence the requirement.

Table 17(A) Rates of flux with either intravenous (i.v.) or intragastric (i.g.) infusions of 1-13C or 3H-leucine. All measurements were in the fed state and based on enrichment of plasma KIC


Flux (mmol/kg/h)



i.v.

i.g.

i.g./i.v.

Melville et al 1989

97.5

135

1.38

Cortiella et al 1988

113

131

1.16

Hoerr et al 1991

137

184

1.34

Hoerr et al 1993




3H-leucine

179

210

1.17

13C-leucine

196

209

1.07

Table 17(B) Enrichments in plasma of leucine and KIC with intragastric or intravenous administration of tracer. Data of Hoerr et al (1991)

Route of administration

E leucine

E KIC

Ratio E

KIC/E leucine

intragastric





fasted

3.34

3.27


0.98

fed

2.15

2.02


0.94

intravenous





fasted

4.18

3.02


0.72

fed

2.84

2.24


0.79

E = enrichment.

10.3. Adaptation

In the classical studies of Nicol & Phillips (1976) in Nigeria, village men were in negative N balance on an essentially protein-free diet (14 mg/kg/d), but came into positive balance and gained weight when the diet was supplemented with egg to provide an intake of 61 mg N/ kg/d, which is little more than the obligatory loss. The Nigerians utilized the egg protein more efficiently (NPU 0.90) than subjects in California (NPU 0.75). A similar high efficiency of utilization of milk protein was observed in children recovering from malnutrition (Chan & Waterlow, 1966).

Two points arise from these studies. The first is whether the state of 'adaptation' described in the Nigerians is acceptable. My view, discussed earlier (section 2), is that this kind of long-term adaptation, with a low body weight and probably a reduced muscle mass, is not acceptable. The fact that the Nigerian men gained weight when given a marginal protein supplement suggests that they were well below Millward's set-point for body protein. Others might contend differently; it is very interesting that, in the Minnesota experiment, when the body weight of the subjects had come down to a body mass index of 16.7, they developed severe psychological and physical symptoms, whereas Indians and Nigerians, who have lived all their lives at this BMI, cope quite well. But the problem remains, that in evaluating states of adaptation it is almost impossible to avoid subjective judgements (Waterlow, 1990).

On the other hand, adjustment of protein metabolism to variations in intake is a part of normal life, and so the second question is how long does this physiological adaptation take? Rand et al (1976) and Bodwell et al (1979) have produced curves showing the change in urinary N output on moving from a normal to a zero protein intake. Rand showed that urinary N decreased according to a single exponential, in agreement with the older findings of Martin & Robison (1922). He suggested that the output might be regarded as 'stabilized' when it was within 1 s.d. of the theoretical output at infinite time. The mean time to reach stability was 4.5 d (range 3-10). The reduction in N output is accompanied, at least in the rat, by coordinated decreases in the activity of the urea cycle enzymes, glutamate dehydrogenase and alanine and as part ate aminotransferases , which presumably play a role in feeding N into the urea cycle (Schimke, 1964; Stephen & Waterlow, 1968; Das & Waterlow, 1974). This adaptation, which we may regard as physiological, takes 30 h in the rat and, from the data of Rand and Bodwell, 6-7 d in man. Presumably the effect of the reductions in enzyme activity is not only to reduce the amount of N excreted but also to improve the efficiency with which N is utilized. In the balance studies of Atinmo et al (1988), in which diets of different protein content were fed for 10-day periods, the NPU improved from 52 at an intake of 0.75 g/kg/d to 91 at 0.3 g over a period of 3-4 weeks.

Further evidence of the need for a period of adaptation in tracer balance studies is the sensitivity of fasting oxidation rates to preceding intake (Figure 3). Therefore it would be unwise to accept the contention of Zello et al (1990) that a period of adaptation is not necessary. Incidentally, Zello recalls that in Rose's experiments, exclusion of a single amino acid from the diet led to anorexia, irritability and fatigue. No such symptoms have been reported in the MIT studies.

To test the adequacy of the 6 d period of adaptation, Young et al (1987) explored the effects of keeping the subjects on the test diets (7, 14 and 30 mg/kg/d) for a further 2 weeks, after which they received leucine at a generous level. The results of this study deserve to be given in some detail. Table 18 shows that leucine balance improved after 3 weeks on the diet compared with 1 week as a result of a fall in fed-state oxidation. However, in the two low intake groups this was achieved at a cost of a sharp decline in the rate of protein synthesis. Young, following Beaton (1985), calls this an 'accommodation', i.e. making the best of a bad job after the limits of physiological adaptation have been passed. A normal rate was restored after 3 days on the generous intake. The groups on the lower intakes went into positive N balance on refeeding, which indicates a degree of depletion or deviation from their setpoint (see also Fisher et al, 1965).

There are some further points of interest about this study. On the two lower intakes plasma leucine concentrations fell, as did the rates of appearance (Ra) of leucine in plasma as a result of protein breakdown (not shown in Table). One might then postulate the following regulatory mechanism: Persistently low leucine intake ® reduced Ra ® fall in plasma leucine ® reduced oxidation. It is noteworthy also from the figure in Young's paper that there is no evidence of longer exposure to the low diet altering the sensitivity of the relation between oxidation rate and plasma leucine level.

Table 18 Leucine kinetics after 1 and 3 weeks on three levels of leucine intake. After Young et al (1987)

Intake (mg/kg/d)

weeks on diet

Oxidation (mmol/kg/h)

Fed state synthesis (mmol/kg/h)

Breakdown (mmol/kg/h)

Balance (mg/kg/d)

Plasma leucine (mM)

7

1

15

79

84

- 16

54


3

4

56

50

- 11

30

14

1

11

74

70

+4

69


3

7

65

65

+ 11

33

30

1

26

90

91

- 16

98


3

14

96

85

+ 6

83

Rates of oxidation synthesis (non-oxidative disposal and breakdown (appearance) have been corrected with a precursor factor of 0.8.

Marchini et al (1993) carried out what was essentially a repetition of Young et al's study, except that the intakes of all the IAAs were modified together, according to the FAO or MIT patterns. In this experiment there were no significant differences between 1 week and 3 weeks on the diet at either level, although balances on the FAO diet were lower than on the MIT diet. This again suggests that an unbalanced diet in which the level of only one amino acid is low is less well tolerated than one in which the intakes of all the amino acids are reduced (see section 8).

On the basis of these results it looks as if, on marginal or inadequate intakes, a longer period of adaptation brings the subjects more nearly into balance by reducing oxidative losses. Nevertheless, this balance may not be an adequate criterion. The reduction in the rate of protein synthesis suggests that the physiological is merging into the pathological. Even the 'undernourished' labourers of Soares et al (1991), who presumably have had a marginal intake all their lives, show no reduction in protein synthesis per unit body weight or lean body mass.

From the practical point of view it seems that one week on the diet is probably about right to allow for physiological adaptation. The experiments of Marchini et al (1993) support this view.

10.4. The nature of the meal

It is possible that a diet in which the protein component consists of crystalline amino acids that are very rapidly absorbed might have a different metabolic effect than one in which the proteins are being continuously hydrolysed as they pass down the digestive tract, and much of the nitrogen absorbed as di- and tri-peptides. The only direct comparison that I know of is the study of Bailey & Clark (1976). They obtained identical small positive balances on a rice-wheat diet and an isonitrogenous diet of the same amino acid composition, made up of crystalline amino acids. Millward (personal communication) reports recent experiments by French workers showing that the efficiency of post-prandial protein deposition was some 30% greater with casein than with a hydrolysate containing oligopeptides.

Another problem is whether small hourly meals have a different effect from food eaten in a normal meal pattern. This is a subject that needs further research.

10.5. Immobilization

The standard protocol of the tracer balance studies requires immobilization of the subject in bed for at least 8h, and it is possible that this may have a negative effect on the balance. In people with fractures or in children with muscular dystrophy, muscles that are immobilized have a reduced rate of protein synthesis with atrophy of type I (glycolytic) muscle fibres (Gibson et al, 1987, 1988; Tucker et al, 1981). Not very much research seems to have been done on the effects of immobilization on whole body metabolism. In a study on mobile and immobile elderly people (Lehmann & James, 1985) the immobile had higher rates of protein turnover and lower rates of oxidation than the mobile. This result, which is the opposite of what one might expect, may be because the immobile are more likely to be sick. Deitrick et al (1948), in the aftermath of World War II, carried out a remarkable experiment in which four subjects were immobilized in bed for 6-7 weeks with a pelvic girdle and their legs encased in plaster casts. The urinary N rose to a maximum on the 5th to the 8th day, with an average increase of 2.5 g N/d by the 5th day. There was also an increased output of calcium, phosphorus and potassium and a small increase in the urinary ratio of nitrogen to sulphur. Schonheyder et al (1954) did a similar experiment in which three subjects were put to bed with both their legs in plaster. They showed an even larger increase in N loss after several days, averaging 9 g/d, but inspection of the charts shows no effect at all on the first day.

In another study with a single dose of 15N-glycine, bed rest and ambulatory conditions were compared (Ang et al, 1992). Synthesis and breakdown both fell during bed rest, but there were no differences in nitrogen excretion.

I conclude that partial immobilization for less than a day is unlikely to have had a significant effect on the tracer balances.

11. Breakpoint analysis

For the sake of completeness we must look briefly at a completely different method of assessing amino acid requirements. There were attempts in the 1970s to use plasma amino acid concentrations as an indicator, the idea being that if concentration is plotted against intake, there will be a sharp inflexion at an intake corresponding to the requirement. Although these expectations were not fulfilled, it seems worthwhile to look at the results of the early tracer balance experiments, since they involved measurements over a wide range of intakes. The results (Figure 6) show a more or less continuous fall of amino acid concentration with decreasing intakes, but nothing that could be interpreted as a breakpoint.


Figure 6
Fed state plasma amino acid concentrations in relation to amino acid intake. Redrawn from data of Meguid et al (1986a,b): Meredith et al (1986) and Zhao et al (1986). For Lysine concentration ÷ 2; for threonine ÷ 3; for valine ÷ 2.

Brooks et al (1972) extended the method to oxidation. They plotted Lysine oxidation against intake in the rat and found a clear-cut breakpoint at an intake of about 100 mg/d (although it is not clear how they calculated the Lysine oxidized from measurement of the output of labelled CO2, since there was no analysis of precursor activity). They considered that the method was more sensitive and specific than similar breakpoint measurements on plasma.

The fed-state oxidation rates plotted in Figure 7 show so much variation that it is impossible to predict the requirement from such data. Bayley and coworkers extended the idea. Instead of measuring the oxidation of the test amino acid at different levels of intake, they used oxidation of a labelled indicator amino acid at different levels of the test amino acid. This method gives curves which are the inverse of the original method; at the point where the requirement is fulfilled, oxidation of the indicator amino acid is at its lowest. Some of the breakpoints obtained on pigs are very clear; for example with 14C-phenylalanine as indicator, there were sharp breakpoints for tryptophan and histidine, but not for Lysine (Kim et al, 1983; Ball & Bailey, 1984, 1986).

This approach has been used by Zello et al (1993) (section 8) to measure the Lysine requirement with phenylalanine as indicator amino acid, but I know of no other studies in man.


Figure 7
Leucine oxidation rate in the fed state in relation to leucine intake.

12. Effect of protein/amino acid intake on protein synthesis and breakdown

This subject is relevant to IAA requirements because maintenance of a 'satisfactory' level of protein synthesis, however one defines 'satisfactory', might be a criterion that requirements are being met. This has been suggested by both Young (1987) and by Millward (1994).

As McNurlan & Garlick (1989) have pointed out, there are two questions to be examined. The first is the immediate effect of food, which is given by the difference between fasted and fed rates of turnover. The second is the effect on protein synthesis and breakdown of the prevailing protein or amino acid intake, taken to be the intake over the previous 6-10 days and called the 'habitual' intake.

The first point has been examined in detail by McNurlan & Garlick (1989). The prevailing opinion is that deposition of protein in the fed state results mainly from a decrease in degradation, with only a small increase in synthesis (Nissen & Haymond, 1986). The recent findings of Pacy et al (1994) (Figure 8) lead to a similar conclusion. The original studies of Clugston & Garlick (1982) had indicated an increase in synthesis as a result of feeding, but it now turns out that this was probably the result of recycling during an infusion which lasted 24 h, 12 h fasted and 12 h fed.

For examining the effect of 'habitual' intake on synthesis and degradation it seems best to look at results in the fed state. In the original study of Motil et al (1981b) fed state synthesis rates derived from incorporation of leucine were similar (113 and 102 mmole/kg/h) at protein intakes of 1.5 and 0.6 g/kg/d, but when the intake was reduced to 0.1 g/kg/d, synthesis fell sharply to 64 mmole/ kg/in. This suggests a breakpoint at or near an intake of 0.6 g protein/kg/d. With Lysine as the tracer the flux fell from 107 to 75 mmole/kg/h as the intake was reduced from 1.5 to 0.4 g protein/kg/d (Conway et al, 1980).


Figure 8
Response to feeding of protein synthesis, breakdown and amino acid oxidation measured with 1-13C leucine.

Figure 9 summarizes the findings of the 1986 studies. With Lysine there was a dramatic fall in fed-state synthesis at intakes below 35 mg/kg/d. With the other amino acids there was a continuous fall, and it would take the eye of faith to discern a breakpoint. The results of the other leucine balance studies that have been reviewed here are shown in Table 19 and Figure 10. At all levels of intake the variability is large; the mean synthesis rate at the leucine intake of 30-40 mg/kg/d is 85.5 mmol/kg/ h, and at 14-15 mg it is 75.5, a difference that is clearly not significant. It might be worthwhile to go back to the original data for further analysis. It makes no difference whether only the level of leucine in the diet is altered, or that of all the IAAs. The main difference is that the intake of 40 mg/kg/d allows for modest protein deposition in the fed state, whereas this hardly occurs at lower intakes.

At this intake the fasted synthesis rate, like that in the fed state, averages 85 mmol/kg/h (data not shown). The two rates taken together result in a synthesis rate over 24 h of 3.4 g protein/kg. This is towards the lower end of the range of values usually found in normal men. It is noteworthy, although it is a pure coincidence, that this is exactly the value Young et al (1989) obtained from their calculation that the obligatory loss represents 10% of the protein turnover rate.

Figure 9 Rates of amino acid uptake into protein at different levels of amino acid intake. Data from 1986 MIT studies; (a) Lysine; (b) leucine; (c) threonine and valine.


Lysine


Leucine


Threonine and valine

Figure 11 shows the rates of synthesis and breakdown in the fed and fasted states at the three levels of intake: egg, MIT and FAO pattern. These results are very similar to those of Pacy et al (1994) (Figure 12). His experiments covered a wider range of intakes than those of the MIT group, and show that increasing the protein intake produced virtually no change in the fasting synthesis rate and only a small increase in the rate during feeding (Figure 12). In his experiments the upper two intake levels, 1.6 and 2.1 g protein/kg/d, are higher than any of those in the MIT series. The sharp fall in the degradation rate at these high intakes does not come through in the MIT studies because their highest intake, 80 mg leucine/kg/d, equivalent to 1 g protein, is represented by only two studies, which show very divergent rates of degradation.

Table 19 Fed state rates of protein synthesis and breakdown after 1 week at different levels of leucine intake

Author

Leucine intake (mg/kg/d)

Synthesis (S) (mmol/kg/h)

Breakdown (B) (mmol/kg/h)

S - B

Young (1987)a

80

119

79

40

Marchini (1.993)

83

83

57

26

Mean





Pelletier (1991b)

40

86

81

5

Pelletier (1991b)

40

76

71

5

Marchini (1993)

40

75

63

12

Hiramatsu

40

101

85

16

Mean




9.5

Young (1987)a

30

90

91

- 1

Young (1987)a

14

74

70

4

Pelletier (1991b)

15

76

75

1

Pelletier (1991b)

15

70

73

- 3

Marchini (1993)

14

66

62

4

Hiramatsu

14

92

89

3

Mean




1.3

Young (1987)a

7

79

84

- 5

a Calculated assuming correction for precursor = 0.8.
In the studies of Young and Pelletier only the leucine content of the diet was altered
In those of Marchini and Hiramatus all the IAAs were altered according to the FAO or MIT pattern.

The conclusion from the 13C leucine studies is that although the synthesis rate does show some variation with level of intake, it is not a sensitive indicator of protein or amino acid status.

This, perhaps, is not surprising, since the flux from which the amino acids for protein synthesis are derived is several times greater than the intake from the food. The synthesis rate is maintained even though there may be a substantial fall in plasma amino acid concentration during feeding, as happens particularly with leucine, presumably because of its small pool size. One may speculate that the synthesis rate is preserved because the amino-acyl-t-RNA synthetases, which catalyse the first step in protein synthesis, have such low KM values that they are saturated at all times.


Figure 10
Rate of leucine uptake into protein in the fed state in relation to leucine intake (later MIT studies: with 1- 13C leucine, comparing FAO, MIT and egg patterns).

Finally, it is interesting to consider the results obtained by Pacy et al (1994) by the end-product method with 15Nglycine (Table 20). We have suggested (Fern et al, 1984, 1985a,b) that the estimate of flux obtained by the labelling of ammonia is biased by the metabolism of the peripheral tissues, specially muscle; and the estimate obtained from urea is biassed by the metabolism of the viscera, particularly the liver. It should be made clear that these are independent estimates of whole body flux and are not additive. We consider that the best estimate is given by the average of the two.

The results with 15N glycine differ strikingly from those with 13C leucine; the synthesis rate increases with increasing protein intake, and the degradation rate shows very little change. The reason for this difference clearly needs to be investigated. On the whole, most workers do not have a very high opinion of the endpoint method and relatively little work has been done with it.

Be that as it may, it is claimed that the method can give information about metabolic activity in two different parts of the body and that it is an advantage of the method that the two estimates of flux do not agree. The ratio of the two estimates, QA/QU, is of some interest (Q being the notation generally used for the flux). From a recent analysis (to be published) of the values obtained under various conditions, it looks as if QA/QU is in the region of 0.7-0.8 in normal subjects, whereas in undernourished adults or children it tends to be lower. Thus in the study of Soares et al (1991), referred to at the beginning of this report, undernourished Indian labourers had lost muscle mass but not visceral mass. They maintained a rate of whole body protein synthesis per kg equal to or greater than that of controls, but QA/QU was 0.58, compared with 0.77 in controls. Table 20 suggests that in Pacy's subjects the lowest intake, equivalent to 27 mg leucine/kg/d, fails to support a normal rate of protein turnover, and the reduction is at the expense of turnover in muscle.


Figure 11
Rates of protein synthesis (uptake) and breakdown (release) in the fed and fasted states in subjects fed three levels of leucine intake: FAO,MIT and egg patterns. Total Amino acid intake equivalent to 0.9 g protein/kg/d. Drawn from data of Marchini et al (1993)

There is no independent evidence I know of that it is in some way better to have a high rather than a low rate of protein synthesis, although both Millward and Young favour the idea. All one can say is that the data in Table 19 show that the FAO pattern of IAAs, or the leucine intake equivalent to it, seems not quite able to support a normal or usual rate of protein turnover. I do not think that this can be ignored. From this point of view the MIT pattern is probably close to the lower limit of adequacy.

13. The colon: losses or gains?

Nitrogen is, of course, lost in the faeces, the obligatory loss being 10 - 20 mg/kg/d. Reeds & Harris (1981) and Fuller & Garlick (1994) have published data on the IAA content of ileostomy fluid, showing that small but not negligible amounts are delivered to the colon. It is immaterial, from the point of view of balance, whether these amino acids are endogenous or derived from the food, but if these amounts pass into the faeces, then they should be added to the losses given by the tracer balances. The fact is, we do not know the fate of these IAAs.


Figure 12
Influence of protein intake on protein synthesis and breakdown, measured with 1- 13C leucine in the fed and fasted states. Data of Pacy et al (1994), reproduced by courtesy of the authors and the publishers of Clinical Science.

The possibility has been raised that amino acids synthesized by the microbial flora in the colon may be available to the body. One of the sources from which this synthesis could occur is urea. It is well recognized since the pioneer work of Walser & Bodenlos (1959) that in the normal adult about 20% of the urea produced passes into the colon and is hydrolysed to ammonia by bacterial urease. Jackson and coworkers have shown that the proportion of urea production that is 'salvaged' in this way (T/Pu) tends to increase in conditions where N balance is under stress, as in subjects on low protein diets or children recovering from malnutrition (Jackson et al, 1990; Langran et al, 1992). If all the ammonia produced by hydrolysis of urea was delivered to the liver, our studies (unpublished) with 15N ammonium chloride suggest that about 70% of it would be immediately converted to urea. In fact, the fraction of urea transferred to the colon that is recycled to urea is only about 25% or less (Jackson et al, 1990; Langran et al, 1992). It has usually been assumed that the ammonia which has not been recycled is taken up into the amino-N pool by transamination. However, there is also the possibility that the ammonia is utilized by bacteria for the synthesis of IAA as well as non-IAA.

Table 20 Fed-state protein turnover rates with 15N-glycine. Data of Pacy et al (1994)

Protein intake (g/kg/12h)

0.34

0.72

1.50

mg N/kg 12 h




N loss

49

74

151

Qu

426

494

651

Qa

165

328

453

Qa/Qu

0.39

0.66

0.70

g protein/kg/12 h




Qav

1.85

2.57

3.45

Sav

1.54

2.11

2.51

Bav

1.51

1.85

1.95

Qu = flux calculated from enrichment of urea.
Qa = flux calculated from enrichment of ammonia.
Qav = arithmetic mean of Qu and Qa
Sav = synthesis rate calculated as Qav -N loss.
Bav = breakdown rate calcuated as Qav-N intake.

This possibility must be seen against the background of a very large nitrogen flux through the colon. Jackson's calculations (Jackson et al, 1984) estimate this flux at about 15 g N/d in a normal adult, or about 1/3 of the whole body flux (this, incidentally, is why we speculate that the substantial pool turning over by lifetime kinetics (section 12) might be related to the gastrointestinal tract). This evidence for a large flux through the colon is supported by an experiment of Wrong et al (1985), which showed that the enrichment of faecal ammonia was many-fold less than that of the plasma urea from which the ammonia must have been derived. Thus we have the picture of a large intra-colonic flux, of which only about 10%/d is disposed of into the faeces, the remainder being recycled back into the body.

The validity of this picture depends on the possibility of reabsorption of amino acids from the colon. Moran & Jackson (1990a,b) found that in adults 60-80% of 15N-urea instilled into the colon was retained in the body. Heine et al (1987) instilled 15N-yeast protein into the colon of infants with Hirschsprung's disease and found that about 90% of the 15N was retained, with significant labelling of plasma proteins. This could have been ammonia that had been fixed by transamination; even so, on this preliminary evidence it looks as if amino acids as well as urea can pass out of the colon. Read et al (1974) suggested this possibility as early as 1974 but could not pursue it because of the civil war in Lebanon. On the basis of an experiment with 15N yeast in infants they wrote: 'Although the colon is not capable of digesting the intestinal flora it can absorb amino acids and there is a high probability that a percentage of the amino acids synthesized by the bacteria will be released into the surrounding medium'.

The final stage in the hypothesis that the body can utilize IAAs synthesized by colonic bacteria must be the direct demonstration of their appearance in body protein. So far the evidence is preliminary. Japanese workers in 1980 claimed to have isolated 15N-labelled Lysine from plasma proteins in subjects in Papua New Guinea to whom they had given 15N-urea (Tanaka et al, 1980), but this claim was greeted by most workers with some scepticism. In a recent study (Torrallardona et al, 1993) a pig was given 15N-ammonium chloride and 14C polyglucose for 10 d. Significant labelling of both N and C was found in the carcass protein. It was estimated that synthesis by the bacterial flora provided Lysine to the amount of 43 mg/kg/d if calculated from 15N and 33 mg/kg/d if calculated from 14C (see also Fuller & Garlick, 1994).

Young has commented that if bacterial synthesis of IAAs occurred to any significant degree, El Khoury et al (1994a) would not have found an almost exact correspondence between leucine intake from the food and leucine oxidized (Table 16). It is difficult to suppose any mechanism by which the colonic bacteria could 'recognize' that their host's dietary intake of an amino acid was high, and therefore reduce their own production. If, then, we visualize a more or less constant and significant addition of IAA from the colon, in striking a balance this addition will be taken into account on the output side (oxidation) but not on the input side (food). The result will be that when oxidation is substracted from the amino acid intake in the food, there will be a negative tracer balance, and this balance will always be negative, however high the intake. In fact this does not happen: the work of both Young and Millward shows that at higher intakes, e.g. 80 mg/kg/d and above, tracer balances become progressively more positive - a well known fact that has never been explained.

I conclude that, in people living on Western diets, colonic synthesis and release of IAAs is unlikely to be significant. However, the microbial flora of the large gut is variable, depending, among other things, on the nature of the carbohydrate and the fibre content of the diet. An effect which does not occur in habitually well nourished subjects might well become significant in those whose diets are less privileged. Research on this subject is now active and the results will be awaited with interest.

14. Conclusion

This report has been stimulated by Young's production of the MIT pattern of IAA requirements and the very large difference between it and the FAO pattern. There is said to be a conflict between his views and those of Millward. My reading of their publications suggests that this is not so. Both are concerned, to an extent that has not been discussed before, with the meaning of 'requirement'. Both seem to feel that the traditional minimum requirement, as measured by balances, is not an adequate measure of human needs in a real environment. Millward with his model has tried to translate this idea into metabolic terms, with his experiments on diurnal cycling and his concept of the anabolic drive: that an intake above the minimum is necessary to keep the machinery of protein turnover operating in a flexible way and with some reserve capacity; but he has not produced alternative figures. His criticisms of Young's experiments are at the technical level and have for the most part been met in the more recent studies of the MIT group.

Young's application of tracer methods to amino acid kinetics for the determination of individual amino acid balances is a new departure and should in principle provide a more precise scoring pattern for the assessment of protein quality. However, it involves more assumptions and more technical difficulties than the traditional N balance.

The point at which the results are most vulnerable is the extrapolation to 24 h from shorter periods of observation, for two reasons. Figure 4 shows that neither in the fed nor in the fasted state is the output of labelled CO2 constant. The pattern may vary under different conditions; thus the times chosen from which to extrapolate, determined by the experimental protocol, may not always be appropriate, although they happen to be in the present case (El Khoury et al, 1994a,b). For example, the pattern of 13CO2 production may be different when food is given in a normal meal rather than in small amounts every hour.

The second problem is how, in the extrapolation, to allow for the periods when the tracer is not being infused. The choice of method can make a significant difference. In my calculations I have taken the view that of the two methods proposed by Marchini et al (1993), the first is preferable (see section 7.5).

As far as I am aware there has been no previous detailed comparison of the various studies that have been done at MIT. The results are consistent in supporting an increased requirement for leucine, according to the MIT pattern. There is much less information about the other IAAs, but to the extent that the pattern of their requirements resembles that of body protein (section 4), if the estimate for leucine has to be increased, so must that for all the others.

The main technical problem with the tracer balance studies is that at the level of the whole body, which is a mixture of tissues, to identify a single value for enrichment of the precursor at all the sites of oxidation is impossible. In my opinion the use of KIC is a reasonable compromise, and it is difficult to imagine that any error introduced by it could be large enough to account for the difference between the new and the old estimates of requirements.

The results of all the studies show a rather large amount of variation between individuals, with coefficients of variation for the rate of oxidation, which is the ultimate measurement, that are often of the order of 25% (e.g. Marchini et al, 1993). This is somewhat greater than the variability of N excretion in traditional balances. It would be interesting to know how far the results in an individual are reproducible, and if they are, whether any characteristic can be identified that is accompanied by high or low rates of flux and protein synthesis. This might help us in the debate about the validity of using the rate of protein turnover as a criterion of adequacy.

The criticism that the meals and their pattern are artificial could in principle be investigated by the single dose method in a subject taking a normal meal. Mathematically this is equivalent to a constant infusion if the analysis is based on the areas under the curves for precursor and product (Waterlow et al, 1978), but a larger number of blood and breath samples would have to be collected. Although an oral dose would avoid the invasiveness of an infusion, the problem of extending these studies to developing countries remains very difficult.

Since I have failed to find any source of error large enough to account for the 2-3-fold difference from Rose's estimates, it is logical to look for sources of error in the old rather than the new figures. The shortcomings of nitrogen balances are well recognized (Table 2). The three main points made by Young in his McCollum Award Lecture (1987) are as follows:

(i) 'Body N equilibrium does not necessarily reflect an adequate state of organ protein metabolism or of nutritional status'. Agreed: but cannot exactly the same thing be said about amino acid equilibrium, as measured in the tracer balance studies? Admittedly those studies give us, as a bonus, information on protein turnover and deposition, but, as discussed in section 12, we cannot with any confidence translate these findings into quantitative estimates.

(ii) The second problem, for which no explanation has been found, is that, at intakes above maintenance, N balances tend to be unrealistically positive. Exactly the same applies to the tracer balance studies; for example, with intakes of leucine of 80 mg/kg/d, positive leucine balances were found ranging from 13 to 25 mg/kg/d (Marchini et al, 1993). Presumably it can at least be concluded that these positive balances are not due to collection errors, because what is collected and how it is collected are completely different in the two systems.

(iii) 'The high energy intakes would be expected to lead to an underestimation of the actual requirement where energy intakes were just sufficient to maintain energy balance'. This is probably the most important point. The energy intakes of Rose's subjects were about 55 kcal/kg/d, compared with about 45 kcal/kg in the MIT subjects. On the basis of the classical figure of 2 mg N retained per extra kcal (Calloway & Spector, 1954), this extra energy should spare 20 mg N, and it is by this substantial amount that Rose's balances could have been underestimates. If we accept for the moment Young's thesis that the IAAs contribute to the obligatory loss according to their pattern in body protein, these 10 kcal would correspond to an underestimate of leucine requirement of 10 mg/kg/d (20 × 6.25 × 8/100). This is a maximum figure, since the contribution of leucine to obligatory loss may be less than its contribution to body protein (section 4), but it brings us closer to the mark. If in addition an allowance is made for inefficiency of utilization at 60%, the extra requirement for leucine comes to some 17 mg over and above that proposed by Rose. It is not clear, however, that the same explanation would apply to Leverton's findings in young women (1959), which led to much the same estimates of requirements as those of Rose, because she had lower energy intakes.

Finally, if the IAAs are lost intact, i.e. without first being metabolized, in the faeces and through the skin or in the urine, the tracer balances will be underestimating the total requirement.

I have not attempted to draw any conclusions about the practical action that should result from this body of work, although Young & Pellett (1990) have done so in no uncertain terms. It is an occupational hazard of scientists to say that more research is needed, but I think that it is so in this case. Anyone who has worked in this field for some time will remember how in the 1970s the emphasis on high protein feeds led to mistaken policy decisions and exaggerated reactions. I suggest that there is a need for more balance studies, such as those of Atinmo et al (1988) in Nigeria, where habitual diets in which Lysine is likely to be inadequate are fed at the usual level, linked to such measures of functional performance as we are able to develop. Many interesting examples in relation to physical activity in different African populations are given in the book 'Capacity for work in the tropics' (Collins & Roberts, 1988), but these are not related to intakes of protein or IAAs. Children are outside the scope of this report, but I believe that the importance of protein quality for their growth and development still needs further investigation.

It is interesting that whereas up to about 1980 protein requirements dominated our thinking, ever since the Rome meeting more and more attention has been given to energy metabolism, and our colleagues in the energy field are making good progress in defining chronic energy deficiency and in research on its functional effects. It seems strange that in reports, for example about seasonal energy deficiency, a great deal of information is given about changes in energy intake but little or nothing about intake or quality of dietary protein. There to be an opportunity here for promising collaboration.

Meanwhile practical decisions have to be taken. Even to do nothing is a policy decision. Young & Pellet (1990) have given their views about the implications of the MIT results. They suggest that diets based on cereals are :likely to be deficient in Lysine and that there is a need for more animal protein and legumes. The inexorable logic of the increase in the world's population means that there is no future in diverting cultivable land and resources, already stretched nearly to their limit (Blaxter, 1987), to the inefficient production of animal protein, and the stocks in the sea are already dwindling. The development of legumes with a higher yield and a higher biological value is another matter altogether, and it is to be hoped that it will be stimulated by the MIT studies. These questions, however, are outside my brief; they will no doubt be taken up in due course by the international agencies and other bodies.

References

Allen LH (1994): Nutritional influences on linear growth: a general review. In Causes and mechanisms of linear growth retardation, eds JC Waterlow & B Schürch Eur. J. Clin. Nutr. 48, Suppl 1, S75-S89.

Allison JB (1951): Interpretation of nitrogen balance data. Fed. Proc. 10, 676-683.

Ang BCN, Halliday D, Georgiannos S & Powell-Tuck J (1992): Bed rest decreases whole body turnover in the post-absorptive state. Proc. Nutr. Soc. 51, 118A.

Arents J & Bier DM (1991): Labeled amino acid infusion studies of in vivo protein synthesis with stable isotope tracers and gas chromatography mass spectrometry. Analytica Chimica Acta 247, 255-263.

Atinmo T. Mbofung CMF, Egun G & Osotomehin BO (1988): Nitrogen balance study in young Nigerian male adults using 4 levels of protein intake. Br. J. Nutr. 60, 451-458.

Aub M & Waterlow JC (1970): Analysis of a five-compartment system with continuous infusion and its application to the study of amino acid turnover. J. Theoret. Biol. 76, 243-250.

Bailey LB & Clark HE (1976): Plasma amino acids and nitrogen retention of human subjects who consumed isonitrogenous diets containing rice and wheat or their constituent amino acids with and without additional Lysine. Am. J. Clin. Nutr. 29, 1353-1358.

Ball RO & Bailey HS (1984): Tryptophan requirement of the 2.5 kg piglet determined by oxidation of an indicator amino acid. J. Nutr. 114, 1741-1746.

Ball RO & Bailey HS (1986): Influence of protein concentration on the oxidation of phenylalanine by the young pig. Br. J. Nutr. 55, 651-658.

Ballevre O. Cadenhead A Calder AG, Rees WD, Lobley GE, Fuller MF & Garlick PJ (1990): Quantitative partition of threonine oxidation in pigs: effect of dietary threonine. Am. J. Physiol. 259, E483-E491.

Barstow TJ, Cooper DM, Sobel EM, Landau EM & Epstein S (1990): Influence of increased metabolic rate on 13C bicarbonate washout kinetics. Am. J. Physiol. 259, R163-R171.

Beaton GH (1985): The significance of adaptation in the definition of nutrient requirements and for nutrition policy. In Nutritional adaptation in man, eds KL Blaxter & JC Waterlow, pp 219-232. London: John Libbey.

Beaufrere B. Horber FF, Schwenk FW, Marsh HM, Matthews DE, Gerich JE & Haymond MH (1989): Glucocorticoids increase leucine oxidation and impair leucine balance in humans. Am. J. Physiol. 257, 712-721.

Bier DM (1989): Intrinsically difficult problems: the kinetics of body protein and amino acids in man. Diabetes Metab. Rev. 5, 111-132.

Biolo G. Tessari P. Inchiostro S. Brutomesso D, Fongher C, Sabadin L, Fralton MG, Valerio A & Tiengo A (1992): Leucine and phenylalanine kinetics during mixed meal ingestion: a multiple tracer approach. Am. J. Physiol. 262, E455-E463.

Blaxter KL (1987): Future hunger. Lancet i, 309-313.

Block RJ & Weiss KW (1956): Amino acid handbook: methods and results of protein anlaysis. Springfield, IL: Thomas.

Bodwell CE, Schuster EM, Kyle E, Brooks B. Womack H. Steele P & Ahrens R (1979): Obligatory urinary and faecal nitrogen losses in young women, older men and young men and the factorial estimation of adult human protein requirements Am. J. Clin. Nutr. 32, 2450-2459.

Brooks IM, Owens FN & Garrigus US (1972): Influence of amino acid level in the diet upon amino acid oxidation by the rat. J. Nutr. 102, 27-36.

Calloway DH & Spector H (1954): Nitrogen balance as related to caloric and protein intake in active young men. Am. J. Clin. Nutr. 2, 405-412.

Chan H & Waterlow JC (1966): The protein requirement of infants at the age of about 1 year. Br. J. Nutr. 20, 775-782.

Clarke JTR & Bier DM (1982): The conversion of phenylalanine to tyrosine in man. Direct measurement by continuous intravenous tracer infusions of L-ring 2H5 phenylalanine and L-14I-13C tyrosine in the post-absorptive state. Metabolism 31, 999-1005.

Clugston GA & Garlick PJ (1982): The response of protein and energy metabolism to food intake in lean and obese men. Hum. Nutr. Clin. Nutr. 36C, 57-70.

Clugston GA & Garlick PJ (1983): Recovery of infused 14C bicarbonate as respiratory CO2 in man. Clin. Sci. 64, 231-233.

Collins KJ & Roberts DF (eds) (1988): Capacity for work in the tropics. Cambridge: Cambridge University Press.

Conway JM, Bier DM, Motil KJ, Burke JF & Young VR (1980): Whole body Lysine flux in young adult men: effects of reduced total protein intake and of Lysine intake. Am. J. Physiol. 239, E192-E200.

Cortiella J. Marchini JS, Branch S. Chapman TE & Young VR (1992): Phenylalanine and tyrosine kinetics in relation to altered protein and phenylalanine and tyrosine intakes in healthy young men. Am. J. Clin. Nutr. 56, 517-525.

Cortiella J. Matthews DE, Hoerr RA, Bier DM & Young VR (1988): Leucine kinetics at graded intakes in young men: quantitative fate of dietary leucine. Am. J. Clin. Nutr. 48, 998-1009.

Das TK & Waterlow JC (1974): The rate of adaptation of urea cycle enzymes, aminotransferases and glutamate dehydrogenase to changes in dietary protein intake. Br. J. Nutr. 32, 353-373.

Daumann D, Matthews DE & Bier DM (1986): Glutamine and glutamate kinetics in humans. Am. J. Physiol. 251, E117-E126.

Deitrick JE, Wheddon DG & Shorr E (1948): Effects of immobilization upon various metabolic and physiologic functions of normal men. Am. J. Med. 4, 336.

Egun GN & Atinmo T (1993): Protein requirement of young adult Nigerian females on habitual Nigerian diet at the usual level of energy intake. Br. J. Nutr. 70, 439-448.

Elia M, Fuller NJ & Murgatroyd PR (1992): Measurement of bicarbonate turnover in humans: applicability to estimations of energy expenditure. Am. J. Physiol. 263, E676-E687.

El Khoury AK, Fukagawa NK, Sanchez M, Tsay RM, Gleason RE, Chapman TE & Young VR (1994a): Validation of the tracer balance concept with reference to leucine: 24 h intravenous tracer studies with L-1-13C leucine and 15N-15N urea. Am. J. Clin. Nutr. 59, 1000-1011.

El Khoury AK, Fukagawa NK, Sanchez M, Tsay RM, Gleason RE, Chapman TE & Young VR (1994b): The 24-h pattern and rate of leucine oxidation, with particular reference to tracer estimates of leucine requirements in healthy adults. Am. J. Clin. Nutr. 59, 10121020.

FAO (1970): Amino acid content of foods and biological data on proteins. FAO Nutrition Studies No 24. Rome: FAO.

FAO/WHO (1965): Protein requirements: Report of a joint FAO/WHO Expert Group. WHO Technical Report No 301. Geneva: WHO.

FAO/WHO (1990): Protein quality evaluation. Report of a joint FAO/WHO Expert Consultation. Rome: FAO.

FAO/WHO/UNU (1985): Energy and protein requirements: Report of a joint FAO/WHO/UNU Expert Consultation. WHO Technical Report No 724. Geneva: WHO.

Fern EB & Garlick PJ (1974): The specific radioactivity of the tissue free amino acid pool as a basis for measuring the rate of protein synthesis in the rat in vivo. Biochem. J. 142, 413-419.

Fern EB, Garlick PJ, Sheppard HG & Fern M (1984): The precision of measuring the rate of whole body nitrogen flux and protein synthesis in man with a single dose of 15N-glycine. Hum. Nutr. Clin. Nutr. 38C, 63-73.

Fern EB, Garlick PJ & Waterlow JC (1985a): The concept of the single body pool of metabolic nitrogen in determining the rate of whole body protein turnover. Hum. Nutr. Clin. Nutr. 39C, 85-99.

Fern EB, Garlick PJ & Waterlow JC (1985b): Apparent compartmentation of body nitrogen in one human subject: its consequences in measuring the rate of whole body synthesis with 15N. Clin. Sci. 68, 271-282.

Fisher H. Bresh MK, Griminger P & Sostman ER (1965): Amino acid balance and nitrogen retention in man as related to prior protein nutrition. J. Nutr. 87, 306-316.

Folin O (1905): A theory of protein metabolism. Am. J. Physiol. 13,117-135.

Fuller MF & Garlick PJ (1994): Human amino acid requirements: can the controversy be resolved? Ann. Rev. Nutr. 14, 217-241.

Fuller MF, McWilliam R. Wang TC & Giles LR (1989): The optimum dietary amino acid pattern for growing pigs 2: requirements for maintenance and for tissue protein accretion. Br. J. Nutr. 62, 255-267.

Fuller MF, Milne A, Harris CI, Reid TMS & Keenan R (1994): Amino acid losses in ileostomy fluid on a protein-free diet. Am. J. Clin. Nutr. 59, 7073.

Gibson JNA, Halliday D, Morrison WL, Stoward PJ, Hornsby GA, Watt PW, Murdoch G & Rennie MJ (1987): Decrease in human quadriceps muscle protein turnover consequent upon leg immobilization. Clin. Sci. 72, 503-509.

Gibson JNA, McMaster MJ, Scrimgeour CM, Stoward PJ & Rennie MJ (1988): Rates of muscle protein synthesis in paraspinal muscles: lateral disparity in children with idiopathic scoliosis. Clin. Sci. 75, 79-83.

Golden MHN (1994): Is complete catch-up possible for stunted malnourished children ? In Causes and mechanisms of linear growth retardation, eds JC Waterlow & B Schürch. Eur. J. Clin. Nutr. 48, Suppl 1, S58-S71.

Golden MHN & Waterlow JC (1977): Total protein synthesis in elderly people: a comparison of results with 15N-glycine and 14C -leucine. Clin. Sci. Mol. Med. 53, 277-288.

Halliday D & McKeran RO (1975): Measurement of muscle protein synthesis rate from serial muscle biopsies and total protein turnover in man by continuous infusion of L- a 15N Lysine. Clin. Sci. Mol. Med. 49, 581-590.

Hamel N. Divertie G. Silverberg J. Persson M & Miles J (1993): Tracer disequilibrium in CO2 compartments during NaH 14CO3 infusion. Metabolism 42, 993-997.

Healy MJR (1989): Nutritional adaptation and variability: comments on the paper of PV Sukathme. Eur. J. Clin. Nutr. 43, 209-210.

Hegsted DM (1976): Balance studies. J. Nutr. 106, 307-311.

Heine W. Wutzke KD, Richter I, Walther F & Plath C (1987): Evidence for colonic absorption of protein nitrogen in infants. Acta Paed. Scand. 76, 741-744.

Henry KM, Kosterlitz HW & Quenouille MH (1953): A method for determining the nutritive value of a protein by its effects on liver protein. Br. J. Nutr. 7, 51-67.

Hiramatsu T. Cortiella J. Marchini JS, Chapman TE & Young VR (1994a): Source and amount of dietary nonspecific nitrogen in relation to whole-body leucine, phenylalanine and tyrosine kinetics in young men. Am. J. Clin. Nutr. 59, 1347-1355.

Hiramatusu T. Fukagawa NK, Marchini JS, Cortiella J. Yu Y-M, Chapman TE & Young VR (1994b): Methionine and cysteine kinetics at different intakes of cystine in healthy adult men. Am. J. Clin. Nutr. 60, 525-533.

Hoerr RA, Matthews DE, Bier DM & Young VR (1991): Leucine kinetics from 2H3 and 13C leucine infused simultaneously by gut and vein. Am. J. Physiol. 260, E111-E117.

Hoerr RA, Matthews DE, Bier DM & Young VR (1993): Effects of protein restriction and acute refeeding on leucine and Lysine kinetics in young men. Am. J. Physiol. 264, E567-E575.

Hoerr RA, Yu Y-M, Wagner DA, Burke JF & Young VR (1989): Recovery of 13C in breath from NaH13CO3 infused by gut and vein: effect of feeding. Am. J. Physiol. 257, E426-E438.

Huang P-C & Lin CP (1982): Protein requirements of young Chinese male adults on ordinary Chinese mixed diet and egg diet at ordinary levels of an egg intake. J. Nutr. 112, 897-907.

Inoue G. Fujita Y. Kishi K, Yamamoto S & Niiyama Y (1974): Nutritive values of egg protein and wheet gluten in young men. Nutr. Rep. Int. 10, 201-207.

Jackson AA (1995): Salvage of urea nitrogen and protein requirements. Proc. Nutr. Soc. 54, 535-547.

Jackson AA, Doherty J. de Benoist M-H, Hibbert J & Persaud C (1990): The effect of the level of dietary protein, carbohydrate and fat on urea kinetics in young children during rapid catch-up weight gain. Br. J. Nutr. 64, 371-385.

Jackson AA, Picou D & Landman J (1984): The non-invasive measurement of urea kinetics in normal man by a constant infusion of 15N15N-urea. Hum. Nutr. Clin. Nutr. 38C, 339-354.

James WPT, Ferro-Luzzi A & Waterlow JC (1988): Definition of chronic energy deficiency in adults. Eur. J. Clin. Nutr. 42, 969-981.

Kies C & Fox HM (1978): Urea as a dietary supplement for humans. Adv. Exp. Med. Biol. 105,103-118.

Kim K, McMillan I & Bayley HS (1983): Determination of amino acid requirements of young pigs using an indicator amino acid. Br. J. Nutr. 50, 369-382.

Langran M, Moran BJ, Murphy JL. & Jackson AA (1992): Adaptation to a diet low in protein: effect of complex carbohydrate on urea kinetics in normal man. Clin. Sci. 82, 191-198.

Lehmann AB & James OFW (1985): Protein turnover in active and inactive elderly subjects using a double isotope infusion technique. Clin. Sci. 69,188.

Leverton RM (1959): Amino acid :requirements of young adults. In Protein and amino acid nutrition, ed AA Albanese, pp 477-506. New York: Academic Press.

Lobley GE, Milne V, Lovie JM, Reeds PJ & Pennie K (1980): Whole body and tissue protein synthesis -in cattle. Br. J. Nutr. 43, 491-502.

Marchini JS, Cortiella J, Hiramatsu T, Chapman TE & Young VR (1993): Requirements for indispensable amino acids in adult humans: longer term amino acid kinetic study with support for the adequacy of the Massachusetts Institute of Technology amino acid requirement pattern. Am. J. Clin. Nutr. 58, 670-683.

Marsolais C, Huot S, France D, Garneau M & Brunengraber M (1987): Compartmentation of 14CO2 in the perfused rat liver. J. Biol. Chem. 262, 2604-2607.

Martin CJ & Robison R (1922): The minimum nitrogen expenditure of man and the biological value of various protein for human nutrition. Biochem. J. 16, 407-442.

Matthews DE, Marano MA & Campbell RG (1993): Splanchnic bed utilization of leucine and phenylalanine in humans. Am. J. Physiol. 264, E109-E118.

Matthews DE, Schwarz HP, Yang RD, Motil KJ & Young VR (1982): Relationship of plasma leucine and a-ketoisocoproate during a L 1-13C leucine infusion in man: a method for measuring human intracellular tracer enrichment. Metabolism 31, 1105-1112.

McNurlan MA & Garlick PJ (1989): Influence of nutrient intake on protein turnover. Diabetes Metabol. Rev. 5,165-189.

Meguid MM, Matthews DE, Bier DM, Meredith CN, Soeldner JS & Young VR (1986a): Leucine kinetics at graded leucine intakes. Am. J. Clin. Nutr. 43, 770-780.

Meguid MM, Matthews DE, Bier DM, Meredith CN & Young VR (1986b): Valine kinetics at graded valine intakes in young men. Am. J. Clin. Nutr. 43, 781-786.

Melvine S, McNurlan MA, McHardy KC, Broom J, Milne E, Calder AG & Garlick PJ (1989): The role of degradation in the acute control of protein balance in adult man: failure of feeding to stimulate protein synthesis as assessed by L 1-13C leucine infusion. Metabolism 38, 248-255.

Meredith CN, Wen Z-M, Bier DM, Matthews DE & Young VR (1986): Lysine kinetics at graded intakes in young men. Am. J. Clin. Nutr. 43, 787794.

Miller DS & Payne PR (1961): Problems in the prediction of protein values of diets: the use of food composition tables. J. Nutr. 74, 413-419.

Millward DJ (1989): The endocrine response to dietary protein: the anabolic drive on growth. In Milk proteins, eds CA Barth & E Schlimme, pp 49-61. Darmstadt: Steinkopf.

Millward DJ (1990): The hormonal control of protein turnover. Clin. Nutr. 9, 115-126.

Millward DJ (1992): The metabolic basis of amino acid requirements. In Protein-energy interactions, eds NS Scrimshaw & B Schürch, pp 31-56. Lausanne: IDECG.

Millward DJ (1994): Can we define indispensable amino acid requirements and assess protein quality in adults? J. Nutr. 124, 1509S-1516S.

Millward DJ, Jackson AA, Price G & Rivers JWP (1989): Human amino acid and protein requirements: current dilemmas and uncertainties. Nutr. Res. Rev. 2,109-132.

Millward DJ, Price GM, Pacy PJM & Halliday D (1990): Maintenance protein requirements: the need for conceptual reevaluation. Proc. Nutr. Soc. 49, 473-487.

Millward DJ, Price DM, Pacy PJH, Queredo RM & Halliday D (1991): The nutritional sensitivity of the diurnal cycling of body protein deposition to be measured in subjects at nitrogen equilibrium. Clin. Nutr. 10, 239244.

Millward DJ & Rivers JPW (1988): The nutritional role of indispensable amino acids and the metabolic basis for their requirement. Eur. J. Clin Nutr. 42, 367-394.

Moran BJ & Jackson AA (1990a): 15N-urea metabolism in the functioning human colon: luminal hydrolysis and intestinal permeability. Gut 31, 454457.

Moran BJ & Jackson AA (1990b): Metabolism of 15N-labelled urea in the functioning and defunctional human colon. Clin. Sci. 79, 253258.

Motil KJ, Bier DM, Matthews DE, Burke JF & Young VR (1981a): Whole body leucine and Lysine metabolism studies with 1-13C leucine and a-15N Lysine: response in healthy young men given excess energy intake. Metabolism 30, 783-791.

Motil KJ, Matthews DE, Bier DM, Burke JF, Munro HN & Young VR (1981b): Whole-body leucine and Lysine metabolism: response to dietary protein intake in young men. Am. J. Physiol. 240, E712-E721.

Nicol BM & Phillips PG (1976): The utilization of dietary protein by Nigerian men. Br. J. Nutr. 36, 337-351.

Nissen S & Haymond MW (1986): Changes in leucine kinetics during meal absorption: effect of dietary leucine availability. Am. J. Physiol. 250, E695-E701.

Osborne TB & Mendel LB (1916): The amino acid minimum for maintenace and growth, as exemplified by further experiments with Lysine and tryptophan. J. Biol. Chem. 25, 1-8.

Pacy PJ, Price GM, Halliday D & Millward DJ (1994): Nitrogen homeostasis in man: the diurnal response to protein synthesis and degradation and amino acid oxidation for diets with increasing protein intakes. Clin. Sci. 86,103-118.

Pelletier V, Marks L, Wagner DA, Hoerr RA & Young VR (1991a): Branched-chain amino acid interactions with reference to amino acid requirements in adult men: valine metabolism at different leucine intakes. Am. J. Clin. Nutr. 54, 395-401.

Pelletier V, Marks L, Wagner DA, Hoerr RA & Young VR (1991b): Branched-chain amino acid interactions with reference to amino acid requirements in adult men: leucine metabolism at different valine and isoleucine intakes. Am. J. Clin. Nutr. 54, 402-407.

Pineda O, Torun B. Viteri FE & Arroyave G (1981): Protein quality in relation to estimates of essential amino acids requirements. In Protein quality in humans: assessment and in vitro estimation, eds CE Bodwell, JS Adkins & DT Hopkins, pp 29-42. Westport, CT: AVI.

Price GM, Halliday D, Pacy PJ, Queredo MR & Millward DJ (1994): Nitrogen homeostasis in man: influence of protein intake on the amplitude of diurnal cycling of body nitrogen. Clin. Sci. 86, 91-102.

Rand WM, Young VR & Scrimshaw NS (1976): Change of urinary nitrogen excretion in response to low-protein diets in adults. Am. J. Clin. Nutr. 29, 639-644.

Rand WM, Scrimshaw NS & Young VR (1979): Analysis of temporal pattern in urinary nitrogen excretion of young adults receiving constant diets at two nitrogen intakes for 8-11 weeks. Am. J. Clin. Nutr. 32, 14081414.

Rand WM, Scrimshaw NS & Young VR (1981): Conventional long term nitrogen balance studies for protein quality evaluation in adults: rationale and limitations. In Protein quality in humans: assessment and in vitro estimation, eds CE Bodwell, JS Adkins & DT Hopkins, pp 59-97. Westport, CT: AVI.

Read WWC, McLaren DS & Tchalian M (1974): 15N studies of endogenous faecal nitrogen in infants. Gut 15, 29-33.

Reeds PJ (1990): Amino acid needs and protein scoring patterns. Proc. Nutr. Soc. 49, 489-497.

Reeds PJ (1992): Isotopic estimation of protein synthesis and proteolysis in vivo In Modern methods in protein nutrition and metabolism, ed S Nissen, pp 249-273. New York: Academic Press.

Reeds PJ & Harris Cl (1981): Protein turnover in animals: man in his context. In Nitrogen metabolism in man, eds JC Waterlow & JML Stephen, pp 391-408. London: Applied Science Publishers.

Reeds PJ, Hachey DL, Patterson BW, Motil KJ & Klein PD (1992): LDL Apolipoprotein B-100, a potential indicator of the isotope labeling of the hepatic protein synthetic pool in humans: studies with multiple stable isotopically labeled amino acids. J. Nutr. 122, 457-466.

Rose WC (1957): The amino acid requirements of adult man. Nutr. Abstr. Rev. 27, 631-647.

Said AK & Hegsted DM (1970): Response of adult rats to low dietary levels of essential amino acids. J. Nutr. 100, 1363-1376.

Sanchez M, EI-Khoury AK, Castillo L, Chapman TE, Young VR (1995): Phenylalanine and tyrosine kinetics in young men throughout a continuous 24-h period, at a low phenylalanine intake. Am. J. Clin. Nutr. 61, 555-570.

Schimke RT (1964): The importance of both synthesis and degradation in the control of arginase levels in rat liver. J. Biol. Chem. 239, 3808-3817.

Schonheyder F, Heilskov NSC & Olesen K (1954): Isotopic studies on the metabolism of negative nitrogen balance produced by immobilization. Scand. J. Clin. Lab. Invest. 6,178-188.

Schwenk WF, Beaufrere B & Haymond MW (1985): Use of reciprocal pool specific activities to model leucine metabolism in humans. Am. J. Physiol. 249, E646-E650.

Scrimshaw NS, Wayler AH, Murray E, Steinke FH, Rand WM & Young VR (1983): Nitrogen balance response of young men given one of two isolated soy proteins or milk proteins. J. Nutr. 113, 2492-2497.

Simon O, Bergner H & Wolf E (1978): Studies on the distribution of radioactivity in the organism during the constant intravenous infusion of tracer amino acids and on the calculation of the rate of the protein synthesis. Arch. Tierernährung 28, 629-639.

Slevin K, Jackson AA & Waterlow JC (1991): A model for the measurement of whole-body protein turnover incorporating a protein pool with lifetime kinetics. Proc. R. Soc. London B 243, 87-92.

Soares MJ, Piers LS, Shetty PS, Robinson S, Jackson AA & Waterlow JC (1991): Basal metabolic rate, body composition and whole body protein turnover in Indian men with differing nutritional status. Clin. Sci. 81, 419425.

Stephen JML & Waterlow JC (1968): Effect of malnutrition on the activity of two enzymes concerned with amino acid metabolism in human liver. Lancet i, 118-119.

Sukhatme PV & Margen S (1978): Models for protein deficiency. Am. J. Clin. Nutr. 31, 1237-1256.

Tanaka N, Kubo K, Shiraka K, Koishi M & Yoshimura H (1980): A pilot study on protein metabolism in Papua New Guinea highlanders. J. Nutr. Sci. Vitaminol. 26, 247-259.

Thompson GN, Pacy PJ, Merritt H, Fond GC, Read MA, Cheng RN & Halliday D (1989): Rapid measurement of whole body and forearm protein turnover using a 2H5-phenylalanine model. Am. J. Physiol. 256, E631-E639.

Torrallardona D, Harris Cl, Milne E & Fuller MF (1993): Contribution of intestinal flora to Lysine requirements in non ruminants. Proc. Nutr. Soc. 52, 135A.

Torun B, Young VR & Rand WM (1981): Protein energy requirements of developing countries: evaluation of new data. Tokyo: UNU.

Tucker KR, Seiler MJ & Booth FW (1981): Protein synthesis rates in atrophied gastrocnemius muscles after limb immobilization. J. Appl. Physiol. 51, 73-77.

Walser M & Bodenlos L (1959): Urea metabolism in man. J. Clin. Invest. 38, 1617-1626.

Waterlow JC (1968): Observations on the mechanism of adaptation to low protein intakes. Lancet ii, 1091-1097.

Waterlow JC (1984): Protein turnover with special reference to man. Quart. J. Exp. Physiol. 69, 409-438.

Waterlow JC (1985): What do we mean by adaptation? In Nutritional adaptation in man, eds KL Blaxter & JC Waterlow, pp 1-12. London: John Libbey.

Waterlow JC (1990): Nutritional adaptation in man: general introduction and concepts. Am. J. Clin. Nutr. 51, 259-263.

Waterlow JC (1994) Emerging aspects of amino acid metabolism: where do we go from here? J. Nutr. 124, 1524S-1528S.

Waterlow JC (1995): Whole body protein turnover in humans - past, present and future. Annul Rev. Nutr. 15, 57-92.

Waterlow JC & Fern EB (1981): Free amino acid pools and their regulation In Nitrogen metabolism in man, eds JC Waterlow & JML Stephen, pp 116. London: Applied Science Publishers.

Waterlow JC, Garlick PJ & Millward DJ (1978): Protein turnover in mammalian tissues and in the whole body. Amsterdam: Elsevier.

Wenham D, Pacy P, Price G, Millward DJ & Halliday D (1991): Bicarbonate recovery during feeding and fasting. Proc. Nutr. Soc. 50, 47A.

Widdowson EM, Southgate DAT & Hey En (1979): Body composition of the fetus and infant. In Nutrition and metabolism of the fetus and infant, ed HKA Visser, pp 169-178. The Hague: Martinus Nijhoff.

Wolfe RR, Goodenough RD, Wolfe MH, Royle GT & Nadel ER (1982): Isotopic analysis of leucine and urea metabolism in exercising humans. J. Appl. Physiol. Respirat. Environ. Exercise Physiol. 52, 458-466.

Wrong OM, Vince AJ & Waterlow JC (1985): The contribution of endogenous urea to faecal ammonia in man, determined by 15N labelling of plasma urea. Clin. Sci. 68, 193-199.

Young VR (1987): 1987 McCollum Award Lecture: Kinetics of human amino acid metabolism: nutritional implications and some lessons. Am. J. Clin. Nutr. 46, 709-725.

Young VR (1991): Nutrient interactions with reference to amino acid and protein metabolism in non-ruminants; particular emphasis on protein-energy relations in man. Z. Ernährungswiss. 30, 239-267.

Young VR, Bier DM & Pellett PL (1989): A theoretical basis for increasing current estimates of the amino acid requirements in adult man with experimental support. Am. J. Clin. Nutr. 50, 80-92.

Young VR & El Khoury AE (1995): Can amino acid requirements for nutritional maintenance in adult humans be approximated from the amino acid composition of body mixed proteins? Proc. Natl Acad Sci. USA 92, 300-304.

Young VR, Fajardo L, Murray E. Rand WM & Scrimshaw NS (1975): Protein requirements of man: comparative nitrogen balance response within the submaintenance to maintenance range of intakes of wheat and beef proteins. J. Nutr. 105, 534-542.

Young VR, Gucalp C, Rand WM, Matthews DE & Bier DM (1987): Leucine kinetics during three weeks at submaintenance-maintenance intakes of leucine in men: adaptation and accommodation. Hum. Nutr. Clin Nutr. 41C, 1-18.

Young VR, Hussein MA & Scrimshaw NS (1968): Estimate of loss of labile body nitrogen during acute protein deprivation in the adult. Nature 218, 568.

Young VR & Marchini JS (1990): Mechanisms and nutritional significance of metabolic responses to altered intakes of protein and amino acids with reference to nutritional adaptation in humans. Am. J. Clin. Nutr. 51, 270289.

Young VR, Meredith C, Hoerr R. Bier DM & Matthews DE (1985): Amino acid kinetics in relation to protein and amino acid requirements. In Substrate and energy metabolism in man, eds JS Garrow & D Halliday, pp 119-133. London: John Libbey.

Young VR & Pellett PL (1990): Current concepts concerning indispensable amino acid needs in adults and their implications for international nutrition planning, Food Nutr. Bull. 12, 289-300.

Young VR, Puig M, Quieroz E. Scrimshaw NS & Rand WR (1984): Evaluation of the protein quality of an isolated soy protein in young men: relative nitrogen requirements and effect of methionine supplementation. Am. J. Clin. Nutr. 39, 16-24.

Young VR, Taylor YSM, Rand WR & Scrimshaw NS (1973): Protein requirements of man: efficiency of egg protein utilization at maintenance and sub-maintenance levels in young men. J. Nutr. 103 1164-1174.

Young VR, Wagner DA, Burini R & Storch KJ (1991): Methionine kinetics and balance at the 1985 FAO/WHO/UNU intake requirement in adult men studied with L 2H3methyl-1-13C methionine as a tracer. Am. J. Clin. Nutr. 54, 377-385.

Yu Y-M, Wagner DA, Tredget EE, Walaszewski JA, Burke JF & Young VR (1990): Quantitative role of splanchnic region in leucine metabolism: L-1-13C, 15N leucine and substrate balance studies. Am. J. Physiol. 259, E36-E51.

Zello GA, Pencharz PB & Ball RO (1990): Phenylalanine flux, oxidation and conversion to tyrosine in humans studied with L- 13C phenylalanine. Am. J. Physiol. 259, E835-E843.

Zello GA, Pencharz PB & Ball RO (1993): Dietary Lysine requirement of young adult males determined by oxidation of L- 1-13C phenylalanine. Am. J. Physiol. 264, E677-E685.

Zhao X-H, Wen Z-M, Meredith CM, Matthews DE, Bier DM & Young VR (1986): Threonine kinetics at graded threonine intakes in young men. Am. J. Clin. Nutr. 43, 795-802.

Discussion

Rose's estimates of IAA requirements

The problem which Waterlow tried to examine in his paper is why the MIT estimates are in most cases so much higher than those of Rose. It is logical therefore, to begin by looking at what Rose actually did; a point that was not dealt with in any detail in Waterlow's paper. The group considered that these studies were flawed in many ways. If the results were corrected for the high energy intakes (55 kcal/kg/d) and for skin and other losses, how much difference would it make? In Young's opinion, not much.* Rose's figures, as reproduced by FAO/WHO in their 1973 report, and quoted again by FAO/WHO/UNU in 1985, represented the highest estimate of individual requirement needed to achieve a positive N balance; this Rose called 'the tentative minimum requirement'. He then doubled these figures to give 'a definitely safe intake', which, for many IAAs would bring this safe level closer to the MIT estimates. However, the MIT figures represent average requirements, so that a comparison with Rose's 'highest individual requirement' is not an exact comparison of like with like.**

* On the question of the effect of the high energy intake in Rose's studies: if as evidence suggests, 1 kcal spares' about 1.5 mg N. a reduction of energy intake to 45 kcal/kg d, as in the MIT studies (Marchini et al, 1993) would increase the requirement by about 15 mg N/kg/d, or about 20% of the obligatory losses (54 mg/kg d) after correction for efficiency at 70% This does not go very far towards filling the gap A further problem is that Rose's N balances included faecal nitrogen, whereas the MIT figures take no account of possible faecal losses of I AA. This difference would further enlarge the gap.

** Table 17 of the 1973 report, dealing with the IAA requirements of adults, gives three columns, all in mg/d.

(a) Rose's figures for young men;
(b) a variety of data for women;
(c) data for women recalculated by regression analysis.

A fourth column gives 'combined adult values' as mg/kg/d. It is not clear, without going back to the original papers, what body weights were used to calculate mg kg and how the male and female data were combined. As a result there seem to be some discrepancies, but they are not very important. e.g.


Rose, man, assume 70 kg body wt mg/kg

Table 17,1993 mg/kg

Isoleucine

10

10

Leucine

15.7

14

Lysine

11.4

12

Sulphur amino acids

15.7

13

Aromatic amino acids

15.7

14

Threonine

7.1

7

Tryptophan

3.6

3.5

Valine

11.4

10

The next question was whether it makes any difference if the studies are done in ascending or descending order of amino acid intake. Millward referred to the work of Atinmo et al (1988), who used both ascending and descending designs and found quite large differences. These differences, however, are not consistent in showing more positive balances with either design. Scrimshaw and Torun quoted studies by UNU (Rand et al, 1984; Torun et al, 1981) and at INCAP (Pineda et al, 1981) done in randomized ascending and descending order; there was no difference. However, Scrimshaw considered that the ascending pattern was preferable for establishing a minimum requirement, since the subjects will be starting from a lower level and therefore will be better adapted. Young, in his carbon balance studies, tried to replicate the design of Rose's work, but could not understand how he did it. At the end of each period he determined the balance and then decided whether to go higher or lower; but how did he do it, since it takes time to analyse urine and faeces?*

*What Rose actually says is: 'to arrive at an amount representing the minimum requirement of the organism for a given amino acid, a negative N balance was induced at some stage of the test, and then the intake of the substance under investigation was raised until a slight but distinct positive balance, as measured by the average for a period of several days was achieved' (Rose's italics) (Rose, 1957).

Allen made the point that these designs, whether of carbon or nitrogen balances, are in any case artificial, since they take no account of the day-to-day and meal-to-meal fluctuations in intake that are found in real life. Her point fits in with Millward's insistence that if the question is whether a particular level is adequate for maintenance, then the measurements have to be made in subjects adapted to that level. However, for Young that is not the question: the whole aim of his work has been to establish a minimum requirement to balance the obligatory loss. In either case, whether the aim is to find a minimum requirement or a requirement at some other level, the question of adaptation and the time needed to achieve it is crucial, but was not discussed in any detail. Scrimshaw just pointed out that in the long term studies sponsored by the United Nations University (UNU) and lasting as much as 6 months, adaptation occurred in a few days.

Protein quality

The discussion began with a consideration of whether there are different patterns of amino acid requirements for growth and for maintenance, as proposed by Waterlow in his paper. There was a general consensus (Reeds, Young, Scrimshaw) that it is illogical to use biological assays in rapidly growing animals such as rats and pigs to make inferences about the amino acid needs of adult humans, and the continuing use of PER in many countries is to be deplored. This conclusion is in line with the recommendations of the 1990 FAO/WHO Expert Consultation.

How should protein quality be measured? The FAO/ WHO/UNO 1985 report concluded that this could appropriately be done by using a chemical score. For Reeds the chemical score, which is an intrinsic property of a protein depending on its amino acid composition, provides a measure that is accurate, precise and objective. Others considered that the chemical score is not appropriate. It is easy to show a difference in protein quality, for example between milk and Incaparina, if the proteins are fed at sub-maintenance level, but at levels above maintenance the difference disappears (Scrimshaw). Studies carried out by UNU (Rand et al 1984; Torun et al, 1984) showed that with diets actually consumed in developing countries, based on rice, wheat or maize, differences in the amount of protein needed resulted from differences in digestibility and not in protein score. The reason that these diets were apparently not limiting in any indispensable amino acid was probably complementation by other components of the diet. Thus the 1985 Report concluded that with real diets quality is not important for adults, once a correction has been made for digestibility. However, this might not apply in situations such as that of refugees, whose diet is limited in composition and may not fulfill their energy requirements.

In balance studies at MIT for the comparative evaluation of wheat protein vs milk protein, the predicted NPU of wheat, based on amino acid composition and the MIT pattern of requirements, was about 70, which agreed with the balance data (Young). This was contested by Millward, who claimed that the amount of wheat protein that achieved balance (0.74 g/kg/d) did not provide an amount of Lysine that matched the MlT estimate of Lysine requirement. Millward described a study in which subjects were maintained on their usual level of protein intake; there was no difference between wheat and milk, although the Lysine content of the wheat diet was below the level of adequacy according to the MIT pattern. This was not an experiment set up to disprove the MIT pattern: it showed that there is enough Lysine in the body pool in the early postprandial phase to supplement the wheat Lysine and enable it to be used with 90% efficiency.

Millward reiterated the importance of establishing a biological model of how amino acids are utilized to maintain balance. When agreement has been reached on whether amino acid needs are fixed or variable, then it will be possible to design experiments to evaluate the need. In response (Waterlow): is this duality between 'fixed' and 'variable' needs a real question? The amount of an amino acid needed to secure balance is evidently influenced by the preceding or habitual intake, and in this sense requirements are variable; however, most of the discussion, notably that of Young, had been focused on the minimum requirement, which seems to be a reasonably clear-cut concept, even though this minimum may be less than the amounts usually consumed in real life, and less than some so far undefined optimum.

Reeds raised the question of whether, with leucine for example, the requirement for leucine carbon was the same as the quantity of N that the body needs to come from leucine. (The point was not taken up, but the question seems to be answered, at least in part, by the experiments of Walser's group on the relative utilization of leucine and ketoisocaproic acid (Kang & Walser, 1985)).

The MIT pattern

Millward began by stressing the importance of differentiating between values obtained experimentally and those generated theoretically from the pattern of amino acids in body tissues. He himself had focused on Lysine because of Young and Pellett's (1990) conclusion that Lysine was likely to be limiting in dietaries based largely on cereals. There are no reliable isotopic data for Lysine; Zello et al's paper (1993) had been referred to, but the method is inappropriate; to use phenylalanine oxidation rates is very complicated and the results debatable. Only a few of the MIT values are experimentally derived. In reply Young said that there are stable isotope data for leucine, valine, threonine, Lysine and methionine, although the early studies can be criticized. The early studies might be criticized. The best experimental data-base is that for leucine, particularly the more recent work of Marchini et al (1993) and el Khoury et al (1994). The measurements of phenylalanine oxidation were entirely consistent with the data for leucine. Balances were consistently negative at the FAO level. The experimentally determined values were compared with those predicted from the composition of body protein (Young & Pellett, 1990). The requirements for those amino acids that had not been determined experimentally were also predicted in the same way. It turned out that these predicted values agreed reasonably well with the estimates for preschool children derived by Torun at INCAP (Pineda et al, 1981) and adopted in the 1985 report. The ratios of predicted requirement: preschool child requirement are: leucine 0.7; isoleucine 1.0; valine 1.0; Lysine 0.94; threonine 0.86; aromatic amino acids 0.8; sulphur amioacids 0.96. For this reason the FAO/WHO 1990 report recommended as an interim measure that the preschool pattern should be used for all age-groups.

A number of criticisms were raised. The FAO/WHO 1990 report was criticized because it was the outcome of a small, not widely representative meeting, and the discussion of the amino acid reference protein was relatively brief. However, Torun pointed out that it was an official FAO/WHO consultation. Even if the discussion and review were not satisfactory, it is now the report accepted by the Agencies, and represents the established pattern until there is another formal UN meeting. A second criticism was that the INCAP studies on preschool children have never been published in peer reviewed journals (Pineda et al, 1981). A question can be raised about those studies because there were substantial positive N balances, greater than would be expected in normally growing children of the same age; they were children who had recovered from malnutrition, but they may still have been depositing lean tissue at a greater than normal rate. Torun replied that the studies were done with diets containing a certain amount of milk protein and a mixture of amino acids in the pattern of milk protein, except for the amino acid being tested. They were done in both ascending and descending order. The criteria used were: maintenance of a positive plateau in N balance, a plateau in urinary urea and changes in free circulating amino acids. The diets were fed at a level of 1.2 g protein/kg/d, with glycine to make up the amount of non-essential N. Other studies were done with various sources of protein (milk, soy) and the results were in good agreement. Young had back-calculated the amounts of Lysine that would be needed to achieve a lower and more reasonable level of N balance: this amount fitted exactly with his prediction. The INCAP data are entirely consistent with classical fortification studies that have been done at MIT.

There followed a brief discussion of individual amino acids. Questioned about the confidence that he would put on his isotopic estimate of the Lysine requirement, Young conceded that his data were limited; however, new 24 h studies, not yet published, clearly show that the FAO level is far too low. The Lysine content of the MIT pattern is, in fact, less than that of body protein. Reeds expressed concern about the sulphur amino acids (SAA). The N balance literature is consistent in suggesting a high SAA requirement for maintenance. It was noted that the Rose value for SAA was quite close to that proposed by Young et al even though his values for the other amino acids were much lower.

Policy implications

Ferro-Luzzi was apprehensive about changing from the 1985 recommendations to higher values. The key question here is whether the conclusion of the 1985 report, that almost all real diets are adequate in IAAs, would still hold if the MIT pattern is adopted. Scrimshaw referred again to the UNU studies showing that natural diets, eaten in usual amounts, fulfilled IAA requirements, as judged by the criterion of N balance. However, the 'key' question is not answered unless the IAA content of these diets is compared with the MIT pattern. Ferro-Luzzi was concerned that the conclusions of these studies, based on a small number of subjects, may not hold at the national level, because of the poor quality of FAO data on national diets. Young pointed out that in some countries, particularly in Africa, where cereals provide more than two-thirds of the protein intake, if there is a risk of IAA deficiency, the MIT pattern suggests that it would be of Lysine. However, a small amount of milk powder or fish would make up for any deficiency. In the study of Nicol and Phillips in Nigeria (Nicol & Phillips, 1976), about 5 g of dried fish used as a condiment apparently made it possible to achieve N balance on protein intakes that otherwise seemed remarkably low. Ferro-Luzzi was still worried about the implications of adopting the MIT pattern, even though Young & Pellet (1990) refer to risk rather than actual deficiency of Lysine. In China she had seen people sprinkle Lysine on top of rice, like Parmesan cheese.

The final question was whether the MIT pattern should be endorsed, at least as an interim recommendation for practical purposes. Torun considered that if the recommendation, e.g. for Lysine, should prove to be too high, it will be erring on the side of safety. Thus, there would be no danger from a public health point of view. Millward dissented; he agreed that a scoring pattern is needed to supplement balance data that are inevitably limited. However, in his view the MIT pattern is not sufficiently secure, based as it is on predictions from body composition for which the justification is disputed, on balance data on pre-school children, which are open to some criticism, and on limited isotope data. In his view, if we are not able to come up with an agreed scoring pattern, we should say so. There is no intellectual justification for accepting the MIT pattern, just because it is the best we have. As for it being an interim recommendation, any decision made here will have important implications. Once even a grudging commitment is made it is very difficult to change it. In the present state of knowledge he was not in favour of adopting the recommendation.

Scrimshaw summed up the debate as follows: 'Those dealing with developing countries do feel the need for a scoring pattern. Not having one would cause a serious problem for the agencies and users. Secondly, every expert group has been faced with the problem of having to make the best recommendations that they can with inadequate data, and have always come up with a long list of research recommendations. We are no different in this respect. Thirdly, there is clear evidence that when new data become available, groups do change their recommendations and views. There will be another meeting in the next 3 to 4 years, and one of the issues that will be discussed is precisely this issue of amino acid requirements; hopefully in the meantime your criticisms of this work will have helped to generate new data that will put the new group into a better position to take a decision.'

References

Atinmo T. Mbofung CMF, Egun G & Osotumehin BO (1988): Nitrogen balance study in young Nigerian male adults using four levels of protein intake. Br. J. Nutr. 60, 451-458.

Calloway DH & Spector H (1954): Nitrogen balance as related to calorie and protein intake in active young men. Am. J. Clin. Nutr. 2, 405-412.

El Khoury AK, Fukagawa NK, Sanchez M, Tsay RM, Gleason RE, Chapman TE & Young VR (1994): The 24-h pattern and rate of leucine oxidation, with particular reference to tracer estimates of leucine requirements in healthy adults. Am. J. Clin. Nutr. 59, 1012-1020.

FAO/WHO (1973): Energy and protein requirements. FAO Nutrition Meetings Report Series no. 52. Rome: FAO.

FAO/WHO/UNU (1985): Energy and protein requirements. Technical Report Series no. 724. Geneva: WHO.

FAO/WHO (1990): Protein quality evaluation. Rome: FAO.

Kang CW & Walser M (1985): The nutritional efficiency of a-ketoisocaproate relative to leucine, assessed isotopically. Am. J. Physiol. 249, E355-E359.

Marchini JS, Cortiella J. Hiramatsu T. Chapman TE & Young VR (1993): Requirements for indispensable amino acids in adult humans: long-term amino acid kinetic study with support for the adequacy of the Massachusetts Institute of Technology requirements pattern. Am. J. Clin. Nutr. 58, 670-683.

Nicol BM & :Phillips PG (1976): The utilization of dietary protein by Nigerian men. Br. J. Nutr. 36, 337-351.

Pineda O. Torun B. Viteri FE & Arroyave G (1981): Protein quality in relation to estimates of essential amino acid requirements. In Protein quality in humans: assessment and in vitro estimation, eds CE Bodwell, JS Adkins & DT Hopkins, pp 29-42. Westport, CT: AVI.

Rand WM, Uauy R & Scrimshaw NS, eds (1984): Protein-energy requirement studies in developing countries: results of international research. Food Nutr. Bull. Suppl. 10. Toyko: UNU.

Rose WC (1957): The amino acid requirements of adult man. Nutr. Abstr. Rev. 27, 631-647.

Torun B. Young VR & Rand WM, eds (1981): Protein-energy requirements of developing countries: evaluation of new data. Food Nutr. Bull. Suppl. 5. Tokyo: UNU.

Young VR, Bier DM & Pellett PL (1989): A theoretical basis for increasing current estimates of the amino acid requirements in adult man, with experimental support. Am. J. Clin. Nutr. 50, 80-92.

Young VR & Pellett PL (1990): Current concepts concerning indispensable amino acid needs in adults and their implications for international nutrition planning. Food Nutr. Bull. 12, 289-300.

Zello GA, Pencharz PB, Ball RO (1993): Dietary Lysine requirement of young adult males determined by oxidation of L-1-13C-phenylalanine. Am. J. Physiol. 264,E677-E685.