FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS	ESN:FAO/WHO/UNU/ EPR/81/23 October 1981
	WORLD HEALTH ORGANIZATION
	THE UNITED NATIONS UNIVERSITY

Item 3.1.1 of the Provisional Agenda

Joint FAO/WHO/UNU Expert Consultation on Energy and Protein Requirements

Rome, 5 to 17 October 1981

ADAPTATION TO DIFFERENT INTAKES AND ENVIRONMENTS

by

J.C. Waterlow

London School of Hygiene and Tropical Medicine
University of London
London, U.K.

Definition and general characteristics of adaptation*

There may be differences of opinion about what is meant by adaptation. My definition is as follows: ‘adaptation is a process by which a new or different steady state is reached in response to a change or difference in environment’. The word ‘new’ covers the case of an individual who is responding to a change in environment, as when I go from low to high altitude. The word ‘different’ relates to individuals who are habitually exposed to different environments. In both cases there is a comparison between two states, A and B. Adaptation is therefore relative; A considers himself normal and B adapted, while B takes the opposite view⁽¹⁾. I shall return to this point later.

The concept of a steady state is also relative. The word ‘steady’ seems to me to include two ideas: first, a stable state, which is not easily disturbed; secondly, one which is maintained and does not alter with time. Neither stability nor permanence are absolute. A person with a somewhat low haemoglobin level may be in a steady state, with no incapacity under usual conditions, but he will be less stable than an individual with a ‘normal’ haemoglobin concentration because less able to withstand a loss of blood. Constancy over time is also relative: body weight may be stable, within the limits of measurement, over a period of 6 months but not from day to day⁽²⁾ nor over 10 years.

*This paper represents a personal point of view. It is impossible in one working paper to cover the enormous literature.

The word ‘response’ covers many different kinds of changes, biological and social, short-term and long-term. The concept of adaptation should be wide enough to include the effects of biological selection over long periods of time. It may therefore be useful to distinguish between physiological, biological and social or behavioural adaptations.

Any consideration of different states of adaptation inevitably involves value judgments. Each state has advantages and disadvantages, which must be related to a particular environment. This is obvious enough, but it creates great difficulties. In the first place, we only have very crude methods of assessing success or failure of an adaptation in terms of the maintenance of health, working capacity, etc. Secondly, it becomes very difficult, in determining nutrient requirements, to make any general statements at all.

Body size

Body size is of crucial importance in any consideration of requirements. For example, the average young adult man in West Java is 158 cm tall and weighs 51 kg⁽³⁾, while his counterpart in the UK is 176.5 cm tall and weighs 70 kg. The Englishman is probably a little fatter (wt/ht² (body mass index) 22.5 compared with 20.4 for the Javanese).

If, as previous committees have always recommended, requirements for food are related to body size, on the basis of either weight or the cube of height the Englishman will need some 40% more food. In view of the secular changes in height which have been well documented in many countries, I myself believe that these differences in stature between groups are determined by a combination of two factors: the immediate effects of the environment and biological selection, it being an advantage to be small if food is scarce. Both within and between communities low height is associated with various kinds of social deprivation. However, I know of no evidence that per se it carries any handicap except for a very limited range of physical activities. On this view, the difference in stature would represent a successful biological adaptation.

Coming to weight, I think it is fair to say that for adults we do not know the lower limit of body mass index (BMI) which is compatible with optimum health. Results of prospective studies quoted by Keys⁽⁴⁾ suggest that in middle-aged men the risk of death begins to rise when the BMI falls below 20–22. It is doubtful, of course, whether one can extrapolate from the USA and Europe to countries where people have been for all their lives exposed to relatively low food intakes and may have adapted to them. Nevertheless, the range of BMI in healthy people in different countries and ethnic groups seems to be quite small.

From the data listed by Eveleth & Tanner⁽³⁾, the lowest BMI in adult males is found in a national sample from India, in which the value is approximately 18.3 (calculated as mean weight/(mean height)²). In a study of body composition of male adults in Colombia⁽⁵⁾, there was a group described as mildly malnourished, with normal albumin and haemoglobin concentrations, but body weight 90% of the US standard. In this group the mean BMI was 21.4. Their body fat was on average 18% of body weight. If one supposes that it would be compatible with health for five-sixths of this fat to be lost, provided there was no loss of lean tissue, the BMI at this lower limit would be 18.2 - that is almost identical with that of the Indian group. Perhaps we could regard a value of 18 for the BMI as the lower limit of normal: below this adults would be at risk of malnutrition.

As conditions improve, average stature, and hence weight, will probably increase, with or without an increase in BMI, so that the requirement per head will rise. Over a generation the effect may be quite substantial, and should be taken into account by planners.

The question of size is of more immediate importance in children because it is more sensitive to environmental and nutritional conditions. In some communities as many as 50% of children are classified as ‘stunted’⁽⁶⁾, defined as less than 90% of the Harvard median, or less than 2SD below the NCHS median. Without going into too much detail, the following points are perhaps relevant to the present discussion.

1. Stunting is associated with a generally deprived environment, and can perhaps be regarded as a ‘proxy indicator’ of social handicap.
2. Retarded growth in height is often associated with deficits in mental and behavioural development. The view is gaining ground that this relationship depends to a large extent on social as well as nutritional factors⁽⁷⁾.
3. Stunting can be ‘cured’ when young children are moved from a deprived to a good environment⁽⁸⁾.
4. Stunting can be prevented by supplementation with food, without other intervention⁽⁹⁾.
5. There is some evidence that the rate of linear growth in children may be affected by their intake of milk^(10,11).
6. There is evidence that stunted children do not show a defect in immune response⁽¹²⁾, nor increased susceptibility to common infections⁽¹³⁾. In one study it has been found that severe stunting (less than 85% of standard height for age) is associated with increased risk of death, but moderate stunting is not⁽¹⁴⁾.

I myself believe that stunting is the result of limitation in nutrient supply, but whether the nutrient is protein, calcium, zinc, etc., we do not know. I incline to the view that stunting should be regarded as an adaptation, within the definition proposed above. Although it may be a response to a limited intake of nutrients, it does not necessarily follow that it is a deficiency state, or that it should be classified as a form of malnutrition.

Adaptation to Low Energy Intakes

Variations in intake in individuals and groups

As is very well known, in normal healthy adults in the steady state and with similar levels of physical activity, there is a very wide range of energy intakes, so wide that the largest eaters may be consuming almost twice as much as the smallest. Even infants, who could perhaps be considered a more homogeneous group than adults, show a similar range of intakes apparently compatible with good health^(15,16).

Measurements of energy expenditure of people in various occupations ⁽¹⁷⁾ in general show good agreement with average intakes, although, as Edholm et al. found⁽¹⁸⁾, even in individuals under strict control exact balance may not be achieved in periods as long as 3 weeks.

Because of this general agreement, I do not think we can criticize the approach of previous FAO/WHO committees, who built up their estimates of energy requirements from physiological measurements of energy expenditure. However, when the estimates so obtained are applied to people in developing countries, some remarkable inconsistencies are apparent, particularly in the case of pregnant and lactating women. Such women in New Guinea⁽¹⁹⁾ or the Gambia⁽²⁰⁾ may have an intake of 1500 kcal (6.3 MJ) compared with 2900 kcal (12.1 MJ) for lactating women in the UK⁽²¹⁾. Allowance must be made for differences in body weight but, from the papers quoted, it seems that the woman in New Guinea may be eating only about 30 kcal per kg per day, compared with 40 consumed by her UK counterpart. I have not been able to make a survey of the literature, but I think it would be accepted that many of the intakes recorded in developing countries are very low by comparison with the usual standards, and this cannot be attributed to errors of measurement. Weight may be lost in the hungry season, but it is regained again. If a steady state is maintained over a period of years, there must be some kind of adaptation.

It does not seem possible to explain these low intakes on the hypothesis put forward earlier⁽⁶⁾, that these individuals with low intakes simply represent the lower end of the normal range of energy requirements. If this were so, one would expect a smaller range of variation of the energy intake in such a group, but in fact this is not found. Both in New Guinea and in the Gambia the inter-individual variation in energy intake appears to be about ±20% - much the same as in developed countries.

The question which has to be answered is whether the mechanisms which underlie, or explain, the variations between individuals are the same as those underlying the differences between groups in developed and developing countries. The question is of importance, because if the answer is yes, it means that all the work going on in industrialized countries on regulation of energy expenditure in obesity is highly relevant to the problem of economy of energy utilization in people in the third world.

Possible mechanisms of adaptation to low energy intakes

1. The effect of small body size in reducing energy requirements has already been mentioned, and quantitatively is of great importance. However, this is clearly not a physiological mechanism which can be used to respond to short-term changes, since, once stature has been determined, the limits within which body weight can be reduced, consistent with normal function, are probably quite narrow.

2. A second obvious adaptation, if one can call it that, is a reduction in physical activity. This is difficult to quantify, but from the pattern of life and work in developing countries, it would be impossible to maintain that there is a generally low level of physical activity. In fact, we probably have the paradox that people work more and eat less. Thus Ferro-Luzzi showed that rural Italian children, in spite of lower energy intakes, performed better in a step-test than urban children⁽²²⁾. If they were fitter, they presumably must have had a higher habitual level of physical activity. However, in young children a marginally inadequate level of energy intake has been shown to limit physical activity before it limits growth⁽²³⁾. This is a very undesirable adaptation from the point of view of social and mental development.

3. The extent to which a reduction in basal or resting metabolic rate may contribute to a low requirement is uncertain. The early Expert Committees on energy requirements allowed for a lower BMR in tropical countries, but this recommendation was abandoned by the 1973 Committee. The evidence on BMR will be reviewed in other working papers. If there is a climatic effect, it is probably not very large. More important may be the lack of need in tropical countries of a thermogenic response to cold⁽²⁴⁾.

4. The evidence for a greater efficiency of physical work, in the sense that the same task is performed with a lower expenditure of energy, seems to be contradictory. Norgan et al.⁽¹⁹⁾ in their studies in New Guinea found no difference in the energy cost of various tasks from the values obtained in the UK. On the other hand, Ashworth⁽²⁵⁾ in Jamaica found that subjects on a low energy intake performed a standard step-test with a lower energy expenditure than controls. This does not, of course, rule out a training effect, which may be very important - that is, a greater economy of physical effort for habitual tasks. In no other way can one explain, for example, the ability of the Nepali porter to carry twice his body weight (100 kg) up several thousand feet in a day.

5. An interesting possibility arises from the fact that the body is never exactly in energy balance. Edholm's classical study on cadets⁽¹⁸⁾, already referred to, showed that there was no significant relationship between food intake and energy expenditure on the same day. We have evidence that within the 24 hours there is a cycle, with positive energy and protein balance during the day (feeding) and negative balance during the night (fasting)⁽²⁶⁾. The process of storing amino acids as protein, and then breaking down the protein and oxidizing it, is energetically wasteful, since the synthesis of protein requires energy (see working paper on the energy cost of growth). Payne & Dugdale⁽²⁷⁾ have suggested that the tendency to store protein rather than fat distinguishes people who maintain their body weight constant from those with a tendency to become obese, since calorie for calorie the storage of fat uses much less energy than the storage of protein. Therefore it is possible that the propensity to store fat rather than protein, and so to conserve energy, could be an adaptive mechanism.

6. Related to this is the possibility that the protein turnover rate may be low in subjects on low energy intakes. According to our measurements⁽²⁸⁾, there is a fairly wide range of variation between individuals in their rates of protein turnover, of which we do not know the significance. On the basis of the ATP and GTP requirement for peptide bond formation, the energy needed for protein synthesis is about 1 kcal (4 kJ) per g.^(28,29). As a result, the energy cost of protein turnover could account for about 15% of the BMR. Thus an individual whose rate of protein turnover was two-thirds of the average might save 5% of his BMR - not a very large amount, but still not unimportant.

7. Resting energy expenditure is mainly for chemical work, including the maintenance of ion gradients. The final common pathway is the generation of high energy phosphate bonds. Krebs⁽³⁰⁾ has pointed out that the efficiency of this process (heat produced per mole ATP formed) is greater when the substrate oxidized is carbohydrate than when it is amino acids or fat. Since people in third world countries tend to have diets high in carbohydrate and low in fat, this could be one route by which energy is economized.

8. There are a number of possible biochemical mechanisms which allow for variations in the amount of heat produced in the course of intermediary metabolism without any net change in chemical or physical work. The common feature of such mechanisms is that a substrate is shuttled round a circular path (substrate cycles). It has been suggested that what used to be called the specific dynamic action of food, now usually referred to as the thermic effect of food, can be explained by an increase in the rate of cycling, stimulated by the rise in substrate concentrations that follows a meal⁽³¹⁾. Such a mechanism would allow a great deal of flexibility in the amount of free energy converted to heat. How far it actually operates as a mechanism of regulation is not known.

9. Recent work has shown that the thermic response to a meal is less in obese than in non-obese subjects⁽³²⁾. To this extent obese people utilize energy more economically. It has been proposed^(33,34) that the site of thermogenesis is brown fat, a tissue rich in mitochondria which are active in the oxidation of fatty acids. Whether differences between individuals result from differences in the amount of brown fat or in its responsiveness is not yet clear.

10. Since heat production in response to a meal remains low in obese people who have returned to normal weight by dieting⁽³²⁾, it seems possible that this capacity for economy may be an individual characteristic which is either inborn or determined and fixed at an early stage of life, thus distinguishing the big from the small eaters. This idea is supported by a study which showed that children 3–4 years old with an obese parent had lower intakes of energy and lower expenditure, both at rest and throughout 24 hours, than children of non-obese parents⁽³⁵⁾. There were no differences in body weight between the two groups of children, but the former would have a tendency to become obese.

To summarize:

1. It is clear that there are differences between individuals in their efficiency of energy utilization, and we are beginning to learn something about the mechanisms of this, although much remains speculative. At the biochemical level there are a number of possible ways of increasing or decreasing the amount of energy used for the same end result.

2. It seems reasonable to suppose that any explanation found for differences between individuals may apply also to differences between groups.

3. I think it is possible that very substantial economies in overall energy expenditure could be achieved by summation of relatively small changes in a number of different systems.

4. It also seems possible that the degree of economy which can be achieved by an individual may be an inborn characteristic. If so, there may have been, in the course of history, selection of those best able to adapt to a low intake.

Adaptation to low protein intakes

It is undeniable that people can adapt from a high protein intake to a lower level. The question is: what is the minimum, and how far does it depend not only on individual variation but also on the environment and life history? Nicol & Phillips⁽³⁶⁾ said: ‘The protein requirement of all apparently healthy men can only be established in the context of their ecological, socio-economic and nutritional backgrounds’.

The protein requirement depends in principle on two factors: the magnitude of the obligatory N loss, and the efficiency with which dietary protein is utilized to make good that loss and to provide for growth, lactation, etc. Our ability to measure these factors is limited by variations within the individual, which may themselves be determined by adaptive changes⁽³⁷⁾, and by variations between individuals⁽³⁸⁾.

Mechanisms of adaptation

The body has powerful mechanisms for conserving nitrogen, and a good deal is known about them. On a low protein diet there is a decrease in the amount and activity of the urea cycle enzymes, both in animals^(39,40) and in man⁽⁴¹⁾; a decrease in the activity of the amino acid dehydrogenases⁽⁴²⁾; and an increased activity of the enzymes which catalyze the first stage of amino acid uptake into protein^(43,44). The reduced oxidation of branched chain amino acids on low protein diets ^(45,46) is of particular interest because the consistently low levels of these amino acids in plasma and tissues in human kwashiorkor⁽⁴⁷⁾ and experimental protein deficiency⁽⁴⁸⁾ suggest that in practice they might be the limiting amino acids.

It may be noted in passing that the concept of chemical score, depending on the concentration of a limiting amino acid, implies that essential amino acids are needed more or less in the proportions in which they occur in tissues. It is assumed that all essential amino acids are economized, and spared from oxidation, with equal efficiency. There is no reason why this should be so. As Hegsted⁽⁴⁹⁾ observed, ‘… animals have the ability to conserve essential amino acids to varying degrees when they are in short supply .. ’. The fact that it is so difficult to determine the maintenance requirement for lysine may be because lysine is very efficiently conserved. The minimum protein requirement must in the end be determined by the amino acid whose conservation is least efficient. In general, essential amino acids must be conserved more efficiently than non-essentials, otherwise it would not be possible to maintain growth and N balance after ‘dilution’ of food protein with non-specific nitrogen^(50,51).

The first stage in adaptation is probably through changes in substrate concentration. As Krebs pointed out, the concentrations of many amino acids in plasma are close to the kMs of the degradative enzymes⁽⁵²⁾. Changes in substrate concentration may also operate indirectly. An example is the effect of arginine concentration on the level of acetylglutamate, which is an activator for carbamyl phosphate synthetase⁽⁵³⁾.

The second stage in adaptation is through changes in the amount of enzymes. It is unknown how this is regulated. Some of these enzyme changes occur very rapidly, in a matter of hours in the case of urea cycle enzymes in the rat⁽⁴⁰⁾. The activity of some liver enzymes is closely related to the intake of food⁽⁵⁴⁾. In man there is an immediate fall in amino acid oxidation during the night, when subjects are receiving no food⁽²⁶⁾, and similar responses have been observed in rats⁽⁵⁵⁾. From the present point of view, long term enzyme adaptations would be of greater importance than these short-term changes, but I know of little information about them.

It is clear that the capacity for adaptation exists. The next question is the extent to which this capacity is used in vivo and the degree to which N losses can be reduced.

Obligatory losses

Results summarized in FAO/WHO (1979)⁽⁵⁶⁾ show, for adults, very uniform findings obtained by different workers in different places. In round figures the losses amount to about 60 mg N per kg day (urine 35–40; faeces 10–20; skin 5). Recent unpublished values in infants are very similar⁽⁵⁷⁾. It may be contended that all these results were for ethical reasons obtained after relatively short periods (6–10 days) on a protein-free diet, which do not allow for complete adaptation. Nevertheless, I do not see much scope for further economy in obligatory loss.

Faecal loss is perhaps the most variable. It is higher in subjects in developing countries, and seems to depend largely on the fibre content of the diet⁽⁵⁸⁾. This is not something which can readily be manipulated.

Loss of N from skin is in desquamated cells and in sweat, mainly as urea. It is true that the N concentration in sweat falls in people acclimatized to heat⁽⁵⁹⁾, but this adaptation is of no relevance, since it represents a shift in the pathways of urea excretion, rather than a reduction in the amount produced.

In urine it has been known since the time of Folin that the amounts of N metabolites other than urea are reduced very little on low protein intakes. Even the amount of ammonia-N is relatively constant⁽⁶⁰⁾. It seems that it is not possible to reduce urea excretion to zero. On a protein-free diet urea still accounts for about 50% of urinary N. Some of the urea is derived from the metabolism of nucleotides⁽⁶¹⁾; however, even if we assume that all of it is the end-product of amino acid oxidation, the amount of amino-N oxidized is only about 5% of the amino acid flux⁽²⁸⁾. This represents a remarkable degree of economy, if 95% of amino acids derived from protein breakdown are being reutilized for synthesis. Whether further economy is possible one cannot tell, but there is clearly not much room for manoeuvre.

There is good evidence that urea-N recycled through the gut can be taken up into protein⁽⁶²⁾. However, this mechanism cannot produce any net economy of N unless, in parallel with the conservation of urea-N there is reduced oxidation of essential amino acids or their carbon skeletons. It is the rate of irreversible degradation of these amino acids which determines the minimum rate of N loss. The same principle applies to the concept that N might be made available through fixation of atmospheric N by bacteria in the gut⁽⁶³⁾.

Capacity for adaptation in vivo

If the argument is accepted, that there is little scope for reduction in obligatory losses, then the only remaining way of economizing is by increasing the efficiency with which N is used for covering these losses. In the balance studies summarized by FAO/WHO (1973)⁽⁶⁴⁾, the ‘inefficiency factor’ (requirement for balance/obligatory loss) was 1.3–1.4. Is it possible to bring this down closer to the theoretical limit of 1?

The evidence that I have been able to find is scanty and contradictory. The question is in any case difficult to answer because of the inherent limitations of balance studies⁽⁶⁵⁾. A small deviation from linearity at the point of zero balance can make a great difference to the numbers⁽³⁸⁾. In the classical experiments of Allison⁽⁶⁶⁾ depleted dogs not only excreted less N on a protein-free diet than control dogs; there was also, in other experiments, an improvement in the efficiency of utilization of various sources of protein⁽⁶⁷⁾. For example, the N balance index of wheat gluten was 0.44 in normal dogs and 0.70 in depleted ones. These dogs are described as depleted rather than adapted. For the moment I make no distinction (see below). Such studies at least show what can be achieved by the biochemical mechanisms discussed above. On the other hand, the experiments of Dreyer⁽⁶⁸⁾ on rats lead to an opposite conclusion. The weight gain which young rats were able to achieve on a 5% casein diet was less when they had previously been exposed to a low protein intake than when they had been exposed to a high one, suggesting the reverse of adaptation.

In man I can find only one experiment specifically designed to test the capacity for adaptation, i.e. to find out whether a preliminary period on a low intake improves the capacity for economizing N in a subsequent period of partial deprivation. That is the experiment of Fisher et al.⁽⁶⁹⁾. These authors were particularly interested in adaptation to shortage of one amino acid - tryptophan - in the presence of adequate amounts of total N. They concluded that ‘There was an inverse relationship between prior protein nutrition and the subsequent utilization of the high-N low-tryptophan diet’. This finding seems inherently reasonable in view of the known adaptive capacity of the enzyme tryptophan oxygenase. However, because of the design of the experiment, in which periods on each diet lasted only 7 days, I doubt if firm conclusions can really be drawn.

Comparisons of protein utilization in the same individual under different conditions require a purposely designed long-term experiment, which is not easy to organize in human subjects. Useful information can also be obtained by comparisons of different people in particular dieatry situations, some of which may produce either adaptation or depletion. In our own studies on children recovering or recovered from malnutrition, nitrogen from milk was retained with an efficiency of almost 100%⁽⁷⁰⁾. Nicol & Phillips⁽³⁶⁾ showed that Nigerian farmers accustomed to eating diets containing only moderate amounts of protein, mainly from vegetable sources, were able to maintain balance on an intake of egg protein which provided only 3.9 g N per day, equivalent to 0.45 g protein per kg per day. On this intake the NPU of the protein was 0.90, compared with 0.75⁽⁷¹⁾ and 0.64⁽⁷²⁾ in similar studies in subjects in the USA. Moreover, the Nigerians were in balance on intakes of egg protein roughly equal to the average requirement, as estimated by FAO/WHO (1973), and well below the safe level (average × 1.3). By contrast, American subjects were not in balance at the safe level⁽⁷³⁾. The obvious conclusion is that the Nigerians were adapted. I myself tend to believe that this conclusion is justified. Nevertheless, there are many difficulties about such comparisons: there may be differences in energy intake in relation to activity; skin losses have not been measured; balance periods differ; results at one level of intake are not strictly comparable with those obtained at several different levels⁽³⁸⁾. Therefore at the time of writing I regard this question as still open. Further light will hopefully be thrown on it by recent studies organized by UNU.

In man what we are really interested in is the capacity for long-term adaptation to low protein intakes. Garza et al.⁽⁷³⁾ found a continuing negative balance in American subjects who were kept for 80 days or more on the safe level of egg protein; they observed that there was no evidence of a ‘drift’ towards equilibrium.On the other hand, in the studies of Sukhatme & Margen⁽³⁷⁾ there is some indication of a drift from negative balance towards equilibrium in subjects maintained for 30 days on an extremely low protein intake (0.64 g N per day). Garza et al.⁽⁷³⁾ maintain that no evidence of a long-term drift has ever been obtained. Perhaps it has not been looked for over long enough periods.

Of more practical relevance is the extent to which people can apparently exist and work on protein intakes which are totally inadequate by usual standards. The FAO/WHO moderately active reference woman has a body weight of 55 kg, an energy expenditure of 2200 kcal per day and a safe level of protein intake (NPU 80) of 36 g, or 0.65 g per kg per day⁽⁶⁴⁾. We might contrast with this, as an archetype of the adapted woman, the non-pregnant Kaul woman aged between 18 and 29 described by Norgan et al.⁽¹⁹⁾. This woman weighs 49 kg and consumes on average 1400 kcal and 24 g protein per day, of which only 3.5 g is animal protein, so that the NPU must be about 70 or even less. Presumably over the long term she is in N balance, otherwise she could not survive. Her N output must then average 80 mg per kg per day. This gives an NPU of 75, similar to that found in western subjects on a maintenance intake of good quality protein⁽⁶⁴⁾. On the New Guinea diet, with 15% animal protein, one would expect the NPU to be lower. In this example, therefore, if the figures are correct, there seems to have been a successful adaptation to a low intake of poor quality protein. But what is the cost?

The Cost of Adaptation: how is it distinguished from depletion?

Metabolic changes in the adapted organism

The enzyme changes discussed above are a means to an end. Here we are concerned with the effects rather than the mechanism. Twenty to thirty years ago there was a great deal of discussion about labile protein and protein stores⁽⁷⁴⁾. The transition from a high to a low protein intake involves a net loss of body protein which is proportionately greater in rat than in man. In the rat the total loss amounts to 3–5% of total body N, initially from liver and viscera, later mainly from muscle and skin⁽⁷⁵⁾.

In man the loss that occurs before a new steady state is reached appears to be less than 1% of total body N^(75,76,77). We do not know from what tissues this protein comes, but it seems likely that in any case the N loss per se is too small to be of significance, provided that the subject does come into a steady state. However, there are metabolic changes which may have an influence on function. The best known is the reduction in the rate of plasma albumin synthesis⁽⁷⁸⁾, which in the steady state is balanced by a reduced rate of breakdown⁽⁷⁹⁾. The rate of muscle protein synthesis is depressed much more than that of liver protein⁽²⁸⁾. Even so, if the diet is high in carbohydrate, as it is likely to be, stimulation of insulin secretion will tend to divert amino acids to muscle at the expense of liver⁽⁸⁰⁾. These are indications that subjects on a low intake, although in balance, are not metabolically the same as those on a high intake.

Implications for function

The concept of protein stores led to a number of experiments to test whether animals with ‘depleted stores’ were more vulnerable to stresses of various kinds^{(81, 82)}. The results were essentially negative. I know of no more recent studies of this kind.

In man it is, of course, well established that malnourished children are more vulnerable to infection. However, malnutrition in the sense of wasting can in no way be regarded as an adapted state. By contrast, it has been argued above that stunting may represent a successful adaptation. As already mentioned, from the scanty evidence available it seems that stunted children are not more susceptible to infection⁽¹³⁾, and moderate stunting does not carry an increased risk of death, although severe stunting does⁽¹⁴⁾.

Morbidity and mortality are crude criteria. What we really need to know is how far the kind of adaptation represented by what I have called the ‘archetype of an adapted woman’ carries with it functional handicaps which can be attributed to the low protein intake, and not to deficiency of energy and other nutrients, or the non-nutritional stresses of the environment. Chittenden⁽⁸³⁾ was enthusiastic about the better sense of wellbeing on his low protein diet (about 0.6 g per kg per day) but his previous diet provided excessive amounts of energy, and the wellbeing may perhaps be accounted for by loss of excess body fat. This illustrates the obvious difficulty of making comparisons which meet the criterion that other things should be equal. However, one possibility suggests itself: a detailed comparison of the life style and performance of the woman in New Guinea with that of her counterpart in the Gambia, whose energy intake is similar but whose protein intake is about twice as high⁽²⁰⁾.

Conclusions

1. At the biochemical level the body possesses powerful mechanisms for economizing nitrogen. Although I think it unlikely that obligatory losses can be significantly reduced, I know of no theoretical reason why food protein should not be utilized with 100% efficiency to cover those losses, provided that enough essential amino acids are supplied to replace the amounts oxidized; the intake needed to maintain balance would then approach the obligatory loss. There are some examples of real-life situations in which this is nearly achieved.

2. Adaptation to low protein intakes carries with it metabolic changes. We do not know what, if any, are their functional implications.

3. An adaptation which perhaps may be regarded as successful, without significant functional loss, is a decrease in the growth rate of children (stunting), with a decrease in final adult size. This would have an important effect in reducing requirements, but it is not, of course, an effect which can be produced by design.

4. The attainable limit of adaptation to low protein intakes is probably of little practical importance. It is likely that in most cases where this limit is approached, there is also a low energy intake, which is much more serious from a functional point of view. Correction of this deficit with natural diets will automatically improve the protein intake.

REFERENCES

1. 1. Waterlow J C (1971)
   In: Metabolic Adaptation and Nutrition. Pan American Health Organization, Scientific Publication no. 222, Washington DC p76

2. 2. Edholm O G, Adam J M & Best T W (1974) Ann Hum Biol 1, 3

3. 3.Eveleth P B & Tanner J M (1976) Worldwide Variation in Human Growth. Cambridge University Press, Cambridge

4. 4. Keys A (1980) Nutr Rev 38, 297

5. 5. Barac-Nieto M, Spurr G B, Lotero H & Maksud M G (1978)
   Am J Clin Nutr 31, 23

6. 6. Waterlow J C (1981) Proc Nutr Soc 40, 195

7. 7. Cravioto J & Delicardie E R (1976) Fd Nutr 2, 2

8. 8. Graham G G & Adrianzen T B (1972) Johns Hopkins Med J 131, 204

9. 9. Mora Sellers Suescum & Herrera (1981) Nutr Res (in press)

10. 10. Fomon S F, Filer L J, Ziegler E E, Bergmann K E & Bergmann R L (1977)
    Acta Paediat Scand 66, 17

11. 11. Graham G G, Creed H M, MacLean W C, Kallman C H, Rabold J & Mellits E D (1981) Am J Clin Nutr 34, 539

12. 12. Reddy V (1979)
    Ind J Nutr Dietet 16, 165

13. 13. Graitcer P L, Duckett E M, Nichaman M Z & Lane J M (1981)
    Am J Clin Nutr (in press)

14. 14. Chen L C, Chowdhury A K M & Huffman S L (1980) Am J Clin Nutr 33, 1836

15. 15. Wallgren A (1944–45) Acta Paediat 32, 778

16. 16. Fomon S J, Thomas L N, Filer L J, Ziegler E E & Leonard M T (1971)
    Acta Paediatr Scand Suppl 223

17. 17. Passmore R & Durnin J V G A (1955) Physiol Rev 35, 801

18. 18. Edholm O G, Adam J M, Healy M J R, Wolff H S, Goldsmith R & Best T W (1970) Br J Nutr 24, 1091

19. 19. Norgan N G, Ferro-Luzzi A & Durnin J V G A (1974)
    Phil Trans R Soc Lond Ser B 268, 309

20. 20. Paul A A, Muller E M & Whitehead R G (1979)
    Trans R Soc Trop Med Hyg 73, 686

21. 21. Thomson A M, Hytten F E & Billewicz W Z (1970) Br J Nutr 24, 565

22. 22. Ferro-Luzzi A, d'Amicis A, Ferrini A M & Maiale G (1979)
    Biblthca Nutr Dieta 27, 85

23. 23. Torun B (1980) Unpublished observations in: Report on the informal gathering of investigators to review the collaborative research programme on protein requirements and energy intake. ESN/MISC/80/3 FAO, Rome

24. 24. Dauncey M J (1981) Br J Nutr 45, 257

25. 25. Ashworth A (1968) Br J Nutr 22, 341

26. 26. Garlick P J, Clugston G A, Swick R W & Waterlow J C (1980)
    Am J Clin Nutr 33, 1983

27. 27.Payne P R & Dugdale A E (1977)
    Ann Hum Biol 4, 525

28. 28. Waterlow J C, Garlick P J & Millward D J (1978)
    Protein Turnover in Mammalian Tissues and in the Whole Body Elsevier/North Holland, Amsterdam

29. 29. Millward D J, Garlick P J & Reeds P J (1976) Proc Nutr Soc 35, 339

30. 30. Krebs H A (1964)
    In: Mammalian Protein Metabolism. Eds H N Munro & J B Allison Academic Press, New York & London. ch.3, p.125

31. 31. Newsholme E A & Crabtree B (1976) Biochem Soc Symp 41, 61

32. 32. Shetty P S, Jung R T, James W P T, Barrand M A & Callingham B A (1981) Clin Sci 60, 519

33. 33. Jung R J, Shetty P S, James W P T, Barrand M & Callingham B A (1979) Nature (Lond.) 279, 322

34. 34. Rothwell N J & Stock M J (1979) Nature (Lond.) 281, 31

35. 35. Griffiths M & Payne P R (1976) Nature (Lond.) 260, 698

36. 36. Nicol B M & Phillips P G (1976) Br J Nutr 35, 181

37. 37. Sukhatme P V & Margen S (1978) Am J Clin Nutr 31, 1237

38. 38. Rand W M, Scrimshaw N S, & Young V R (1977) Am J Clin Nutr 30, 1129

39. 39. Schimke R T (1962) J Biol Chem 237, 459

40. 40. Das T K & Waterlow J C (1974) Br J Nutr 32, 353

41. 41. Stephen J M L & Waterlow J C (1968) Lancet i, 118

42. 42. Harper A E (1965) Canad J Biochem 43, 1589

43. 43. Gaetani S, Paolucci A M, Spadoni M A & Tomassi G (1964)
    J Nutr 84, 173

44. 44. Stephen J M L (1968) Br J Nutr 22, 153

45. 45. Sketcher R D, Fern E B & James W P T (1974) Br J Nutr 31, 333

46. 46. Reeds P J (1974) Br J Nutr 31, 259

47. 47. Holt L E, Snyderman S E, Norton P M, Roitman E & Finch J (1963)
    Lancet ii, 1343

48. 48. Lunn P G, Whitehead R G & Baker B A (1976) Br J Nutr 36, 219

49. 49. Hegsted D M (1973) In: Proteins in Human Nutrition. Eds J W G Porter & B A Rolls. Academic Press, London & New York. p.282

50. 50. Snyderman S E, Holt L E, Dancis J, Roitman E, Boyer A & Balis M E (1962) J Nutr 78, 57

51. 51. Scrimshaw N S, Young V R, Schwartz R, Piche M L & Das J R (1966)
    J Nutr 89, 9

52. 52. Krebs H A (1972) In: Advances in Enzyme Regulation. Ed G Weber. vol.10. Pergamon Press, New York. p.397

53. 53.Tatibana M & Shigesada K (1976) In: The Urea Cycle. Eds S Grisolia, R Baguena, F Mayor. John Wiley & Sons, New York. p.301

54. 54. Potter V R (1968) Fed Proc 27, 1238

55. 55. Wannemacher R W & Dinterman R E (1980) Biochem J 190, 663

56. 56. FAO/WHO (1979) Bull Wld Hlth Org 57, 65

57. 57. Torun B & Viteri F E (1980) Obligatory nitrogen losses and factorial calculations of protein requirements of pre-school children.Paper presented to UNU/IUNS Workshop, MIT, May.

58. 58. Calloway D H & Kretsch M J (1978) Am J Clin Nutr 31, 1118

59. 59. Ashworth A & Harrower A D B (1967) Br J Nutr 21, 833

60. 60. Waterlow J C (1963) Am J Clin Nutr 12, 235

61. 61. Golden M H N, Waterlow J C & Picou D I M (1981)
    In: Nitrogen Metabolism in Man. Eds J C Waterlow & J M L Stephen. Applied Science Publishers, London. (in press)

62. 62. Picou D & Phillips M (1972) Am J Clin Nutr 25, 1261

63. 63. Oomen H A P C (1970) Proc Nutr Soc 29, 197

64. 64. FAO/WHO (1973) FAO Nutrition Meetings Report Series no. 52 FAO, Rome

65. 65. Hegsted D M (1976) J Nutr 106, 307

66. 66. Allison J B (1951) Fed Proc 10, 676

67. 67. Allison J B, Anderson J A & White J I (1949)
    Trans Am Assoc Cereal Chem 7, 24

68. 68. Dreyer J J (1970) S Afr Med J 6, 335

69. 69. Fisher H, Brush M K, Griminger P & Sostman E R (1965)
    J Nutr 87, 306

70. 70. Chan H C & Waterlow J C (1966) Br J Nutr 20, 775

71. 71. Calloway D H & Margen S (1971) J Nutr 101, 205

72. 72. Young V R, Taylor Y S M, Rand W M & Scrimshaw N S (1973)
    J Nutr 103, 1164

73. 73. Garza C, Scrimshaw N S & Young V R (1977) J Nutr 107, 335

74. 74. Munro H N (1964) In: Mammalian Protein Metabolism. Eds H N Munro & J B Allison.Vol. I. Academic Press, London & New York.ch. 10, p.381

75. 75. Chan H C (1968) Br J Nutr 22, 315

76. 76. Gopalan C & Narasinga Rao B S (1966) J Nutr 90, 213

77. 77. Young V R, Hussein M A & Scrimshaw N S (1968) Nature 218, 568

78. 78. James W P T & Hay A M (1968) J Clin Invest 47, 1958

79. 79. Hoffenberg R, Black E & Brock J F (1966) J Clin Invest 45, 143

80. 80. Munro H N (1951) Physiol Rev 31, 449

81. 81. Vaughan O W, Filer L J & Churella H (1962) Pediatrics 29, 90

82. 82. Halac E (1962) Am J Clin Nutr 11, 574

83. 83. Chittenden R H (1907) The Nutrition of Man.Heinemann, London