Contents - Previous - Next

2. The effects of different levels of energy intake on protein metabolis

2.1. Adults
2.2. Children

2.1. Adults

2.1.1. Studies relating protein and energy intakes to nitrogen balance

Changes in food energy intake, below or above energy needs, affect nitrogen balance so that any change in food energy intake results in a corresponding alteration in nitrogen balance. The improvement in nitrogen balance caused by an increase in energy intake, however, can be attenuated if the intake of protein is grossly inadequate; conversely, the beneficial effects of an increase in protein intake can be inhibited by an inadequate energy intake. Thus, nitrogen balance must be seen as a result of levels of both energy and protein intakes and both must be defined before a precise interpretation of N-balance results can be made with respect to establishing physiologic protein requirements. In this section of the review, a number of studies relating protein and energy intakes to nitrogen balance will be examined. These mainly involve adult humans, although investigations in other species will also be considered.

A summary of a series of studies in adult human subjects is shown in Table 1 with the relationships between N balance and energy intake summarized in Table 2. The data presented here include the range of nitrogen and food energy intakes, the number of subjects involved and linear regression equations (both linear and multiple) relating N balance to both nitrogen and energy intakes. We have expressed the results obtained in various experiments in the same units, (nitrogen intake (ma N/kg), food energy intake (kcal/kg) and nitrogen balance (mg N/kg)), irrespective of how these data were reported in the original experiments. This has permitted, as will be described later, an integration of the results into a single data set, allowing a more extensive examination of the statistical associations between nitrogen and energy intakes and nitrogen balance. First, however, we will describe briefly the various studies used in our analysis, before we discuss a coalescence of these data in the following section.

The relationship between energy supply and N balance was examined and quantified by CALLOWAY and SPECTOR (1954) and was more recently again reviewed by CALLOWAY (1981). They concluded that a change of 1 kcal in food energy intake produced a change of about 1.5 mg in the amount of nitrogen retained. It should be noted that the CALLOWAY and SPECTOR (1954) review is the only study included here in our revaluation where body weight data for the subjects are lacking. Hence an average body weight of 71.2 kg (mean body weight from all US young men in the other cited experiments) was assumed to convert to kcal, N intakes and N balance per unit body weight for the 18 composite data sets used by these investigators.

CALLOWAY and SPECTOR (1954) selected studies from the literature, dating back to at least 1907, which demonstrated the effects of food-energy intakes on protein metabolism as reflected by nitrogen balance in normal active young men. The research was sponsored by the Department of Defense because of its relevance to military feeding in warfare situations. These authors were able to gather many data which were not widely available to others engaged in nutrition research. Some 460 observations were cited for subjects receiving varied intakes of both protein and energy, together with 186 observations for those receiving non-protein diets at differing energy intakes. Major conclusions were drawn from 'representative data' where 9 values were cited for non-protein data and 18 values for those receiving both protein and food energy.

Table 1. Correlations between N balance, N intake and energy intake in selected studies

^a 18 composite periods derived from 460 observations.

Table 2. Effect of energy intake on nitrogen balance in selected studies

^a 18 composite periods derived from 460 observations.
^b Includes additional studies of Inoue et al. 1981 and Yanez et al. 1984.

In these experimental situations (Figure 1), daily intakes ranged from 400 to 3342 kcal (6 to 47 kcal/kg) and from 1 to 15.4 g N (14 to 216 mg N/kg). These produced nitrogen balances from -6.5 g per day to + 1.1 g (-91 to + 15 mg N/kg) per day. For the non-protein diets, the N balances were as low as -12 g per day (-168 mg N/kg) when associated with very low food energy intakes, and they rose to -7 g per day (-98 mg N/kg) at higher levels of energy intake. Among the author's conclusions were:

"... to the general principles that on a fixed adequate protein intake, energy level is the deciding factor in nitrogen balance and that with a fixed adequate caloric intake protein level is the determinant may be added a corollary. That is, at each fixed inadequate protein intake there is an individual limiting energy level beyond which increasing calories without protein or protein without calories is without benefit".

Source: Calloway and Spector (1954)

In more specific terms Calloway and Spector asked why 660-800 kcal/d should maximally reduce the destruction of body protein but not permit the utilization of dietary protein, or why 3 g of dietary N should be most efficiently used at it 1000 kcal or 6 g N at 1600 kcal. They further concluded that there were no ready answers. These conclusions are broadly congruent with those of MUNRO (1951) and supported much later in a review by WATERLOW (1986), who also concluded that an increase in the energy supply will not promote N retention unless the amino acid supply is adequate, and conversely an increased amino acid supply will be useless if energy is limiting. In this paper, Waterlow emphasized that, in adults who are not growing, these relationships can only be investigated in the region near or below nitrogen equilibrium, but for children growth extends the range over which these interactions may be observed.

Important experiments on energy and protein relationships were reported in 1958 by ALLISON. Depleted dogs were studied and the rate of repletion was examined when fed different proteins at different levels of both protein and food energy. Additionally, lysine-supplemented wheat gluten was also fed under the same conditions. Again, it was found that both protein and energy were separately effective in promoting the rate of repletion, but that the type of repletion differed; for the most effective repletion, high levels of both protein and food energy were found to be necessary. It was also shown that casein gave more rapid repletion than did wheat gluten, although the latter, when supplemented with lysine, behaved in an essentially similar manner to casein.

MILLER (1973) gave particular consideration to protein-energy relationships in the determination of protein quality and remarked:

"... it is of little practical value to study protein in isolation from the rest of the diet, especially in isolation from the energy component. The efficiency of utilization of dietary protein is, in practice, less influenced by its amino acid composition than by the energy value of the diet. Protein can act both as an energy source and as a nutrient... the proportion burnt to meet energy needs is increased when protein is in surfeit or when food is limited."

It had been long known that it was possible to spare protein, i.e., reduce N losses, by the addition of various energy sources to the diet, and Miller noted that the net result of these additions was not only to increase the energy intake, but also to lower the protein/energy ratio. Both factors influenced N balance, but they were not wholly independent. Miller gave examples from various experimental designs where, respectively, the protein percentage, the energy intake and the protein intake were held constant. As a result of the experimental changes made, variations were observed in energy and nitrogen intakes and in dietary protein concentration. In other words, in all experimental conditions there was always more than a single change of either protein or energy intake, thus complicating possible interpretations of the data. As an example, in the subsequent and official recommendations (e.g., AOAC, 1975) for the PER procedure, which had been originated by OSBORNE, MENDEL and FERRY in 1919, it became accepted that a constant level of protein (9-10%) should be fed ad libitum. This, of course, would lead to variable protein and energy intakes depending on the food intake behaviour of the individual rat.

For experiments where the protein intake was held constant and energy levels were varied (e.g., MUNRO, 1951; CALLOWAY and SPECTOR, 1954), inevitably protein concentration (Pcal%) would also change.

Human N-balance data were reported by INOUE et al. (1973) where the relationships of excess energy to nitrogen balance were examined by feeding egg and rice diets to Japanese young men. Seventeen men were fed the egg diets and 24 the rice diets for 10-day balance periods. Energy levels were 45 and 57 kcal/kg, respectively, during the maintenance and excess energy periods. During the study, protein levels varied from 45 to 120 mg/N kg in different subjects and diets. The results obtained for the egg diet are shown in Figure 2. Again, their experiments showed that excess energy spared protein but also demonstrated that the degree of sparing differed for rice and for egg. Thus, protein quality was shown to be an additional factor in the relationships between energy intake and nitrogen balance.

Further human studies were reported by NAGESWARA RAO et al. (1975) using five subjects whose weights ranged from 47.4 to 64.4 kg and who were given diets containing either 40 or 60 g protein/d. Each N-balance period was of 11 days duration. Although there were five subjects, dietary groups never exceeded four, and there were only two subjects for the high and low food energy levels on the 40 g protein diet. Energy levels varied from 1800 to 3000 kcal/d with the 40 g protein diet and from 2100 to 2700 kcal/d on the 60 g protein diet. From regression equations, it was calculated that the energy intake for N equilibrium would be 2066 kcal at the 60 g/d protein level and 2249 kcal when 40 g/d protein was fed. At 60 g/d protein intake, balance was improved by 1.60 mg/kcal against 0.87 mg/kcal on the 40 g/d intake. Despite wide differences in the body weights of the experimental subjects, it appears that they had the same protein and energy intakes. Thus, protein intakes ranged from 0.93 to 1.22 g/kg for the 60 g protein diet and from 0.62 to 0.84 g/kg for the 40 g protein diet. For food energy, intakes apparently ranged widely from 28 to 57 kcal/kg. This reduces the precision of the estimates and limits the value of the study.

In 2-week balance periods both additional N and additional energy improved N balance. The effect of energy was generally more significant. (From INOUE et al., 1973)

In a subsequent study by CALLOWAY (1975), designed to evaluate the relative importance of energy and protein on N equilibrium in the near-adequate range of intakes, healthy men were given two levels of protein with energy constant and three levels of energy with protein constant. In the first two 12-day experimental periods, diets provided 5 and 7% of energy from egg white protein with enough energy (ca. 40 kcal/kg) to maintain body weight almost constant. N-balance data obtained in feeding these diets were used to select an individual protein intake level nearest to need (5, 6 or 7%), and that level was fed for the next three periods with the same energy intake as before (100%) or at 85 or 115% of it. Predicted minimum N need to maintain crude N balance (average = + 0.12 g) at 100% energy level was 89 mg/kg body weight or 3.76 mg/basal kcal. N balance fell to 0.61 g/d at the 85% level and increased to + 0.59 g/d at the 115 % level. N balance changed by 1.74 mg/kcal between 85 and 100% energy levels and 1.12 mg/kcal between 100 and 115% energy levels. It was thus concluded that energy intake had a much greater effect on N balance than did protein in the marginally adequate range of intakes. CALLOWAY (1975) emphasized that in many studies of minimum N need, energy intakes have been relatively high and participants have either maintained or gained body weight.

GARZA, SCRIMSHAW and YOUNG (1976) also noted that protein requirement studies in man had generally avoided deficient dietary energy intakes, because they decreased the efficiency of nitrogen utilization, but that the opposite effect of excess dietary energy had often been overlooked. In their study, four young men were fed the 1973 FAO/WHO safe level of egg protein (0.57 g/kg; 91 mg N/kg) at several levels of dietary energy. The experiment was divided into two or three dietary periods of 21 to 32 days during which the different energy intakes ranging between 43 and 53 kcal/kg were provided. Increases in energy intake generally improved nitrogen balances which, however, were not closely correlated with body weight changes. The four subjects needed 47 to 50 kcal/kg to achieve nitrogen balance at the test protein intake level. Since this level (0.57 g/kg) was the prevailing recommended protein requirement level, the authors concluded that excess energy intakes would generally be required for the population at large in order to maintain nitrogen balance at this level of dietary protein.

In a subsequent experiment (GARZA, SCRIMSHAW and YOUNG, 1978), subjects were given the same level of protein but with additional non-essential amino acids equivalent to 0.23 g/kg (total intake = 128 mg N/kg) at varying energy levels from 39 to 56 kcal/kg. This group required significantly less energy to maintain nitrogen balance than those fed 0.57 g/kg in the earlier experiment.

Important studies, which confirm and extend these various observations made in humans, were reported by BLACK and GRIFFITHS (1975). These involved milk-fed lambs, with essentially no rumen function, which were fed from a teat when less than 15 kg and by abdominal infusion for those of greater weight. The results of 298 N-balance studies, using male cross-bred lambs weighing 3-38 kg which had been either fasted or fed entirely on liquid diets of various protein content at various energy intakes up to an ad libitum level were used to describe the effects of the amount and quality of absorbed protein, energy intake and live weight on N balance and the total N requirement of lambs.

As is illustrated in Figure 3, it was demonstrated that, when N intake was below the requirement level, N retention was linearly related to N intake and was unaffected by energy intake, i.e., growth was in a protein-dependent phase. When, however, N intake was in excess of requirement, an energy-dependent phase came into play and N retention was influenced by energy intake and unaffected by N intake. For lambs receiving excess dietary N and a constant metabolizable energy intake, N balance decreased as the body weight increased, probably because less of the energy available for growth was directed towards protein synthesis and more towards lipogenesis as the animals became heavier.

It was additionally noted by BLACK and GRIFFITHS (1975) that protein accretion was directly related to the digestibility and BV of the protein consumed. Similar phenomena have also been reported for pigs (CAMPBELL, 1988; HARRIS et al., 1989) and for broiler chickens (CAMPBELL, 1988). The relationships between protein and energy balance delineated in these farm livestock species are essentially identical to those described for humans by MUNRO (1951) and by CALLOWAY and SPECTOR (1954).

Another investigation on the utilization of egg protein by Japanese young men with marginal energy intakes was reported by KISHI et al. (1978). In their first experiment, involving 31 subjects, egg protein at 32, 64, and 80 mg N/kg was given together with excess energy (48 kcal/kg). In the second experiment, involving 15 subjects, submaintenance energy (40 kcal/kg) was provided at these same egg protein levels. Once again, it was concluded that nitrogen balance was markedly affected by energy intake and that balance was poorer at lower energy intake levels. Numerical relationships between N balance and nitrogen and energy intakes at similar levels of energy expenditure, for this and other experiments, are shown in Tables 1 and 2.

MUNRO (1951; 1964; 1978) had observed that the degree of protein sparing was highly dependent on the type of energy source, especially in short-term balance studies, since there was a specific insulin-dependent effect of dietary carbohydrate on protein metabolism. This short-term specific effect on protein metabolism was not shared by fat, probably because, through the action of insulin released by dietary carbohydrate, plasma amino acids were diverted into muscle protein. The quantitative effect of isoenergetic exchange of fat for carbohydrate on dietary protein utilization in healthy young men was also investigated by RICHARDSON et al. (1979). Milk protein (0.57 g/kg body weight, the then safe level of protein intake (FAO/WHO, 1973)), was given for two 21-day experimental periods with two ratios of carbohydrate to fat calories. The first provided an equal proportion of energy from carbohydrate (Fcal% = 48%); considerably above the mean US value of 38%, while the 'high'-carbohydrate diet had a Fcal% = 33%, with about twice as much of the food energy coming from carbohydrate as from fat. Total food energy intake was constant (average = 45 kcal/kg) for each subject during the two experimental periods. Nitrogen balance and dietary protein utilization were significantly different between the two diets with the mean nitrogen balance increasing from -0.25 g/d on the high-fat diet to + 0.23 g N/d on the high-carbohydrate diet. The results are illustrated in Figure 4.

It was also of interest that, while weight gain and nitrogen balance were correlated in a statistically significant manner (R = 0.74; p<0.001; Nbal = 0.010 X Weight Change - 0.041), the individual subjects varied considerably with half of the subjects gaining and half losing weight on both diet regimes. It was additionally noted that the protein-sparing effect of carbohydrate, relative to fat, was more pronounced for those subjects whose energy intakes were lowest.

A study of a different type (UAUY et al., 1984) can also be used to examine the relationships of food energy to nitrogen balance, although it was not designed specifically for that purpose. In this study, concerned with protein requirements for young men in Chile, most of 53 subjects (army recruits) were given diets providing 3200 kcal per day, although a few received 3500 kcal. Protein intake was provided at a constant level of 183 mg N/kg derived from varied sources typical of a mixed Chilean diet. Since the body weights varied from 49 to 83 kg, food energy intakes, when expressed on a body weight basis, showed considerable variation and ranged from 45 to 68 kcal/kg. It was observed that the subjects with the largest positive N balances were those with the highest energy intakes.

A study from Thailand (TONTISIRIN et al., 1984a) which produced interesting data relating protein and energy intakes also originated as an investigation on the adequacy of the recommended protein intake for adults. In this study, 0.72 g/kg protein (115 mg N/kg), supplied from a rice-fish mixture, was fed to 12 healthy men at energy intake levels ranging from 36 to 55 kcal/kg. Diet periods were 10 days in duration, and N balance improved with increasing energy intakes. Nevertheless, all subjects lost weight (average 0.7 kg) even when N balance was positive at the highest energy intake levels.

Consequently, a second experiment was conducted where subjects were given similar protein levels as in the first, but with rice being allowed on an ad libitum basis. There were three experimental periods in this second phase. In the first period, the typical Thai rice-fish-vegetable diet, but with as much rice as was desired, supplied average intakes of 170 mg N/kg and 57 kcal/kg. For the second 21-day period, the protein level was reduced and intakes now averaged 164 mg N/kg and 50 kcal/kg, again with ad libitum rice. In the final period, a fixed energy intake at the average of the last 14 days of the first period (58 kcal/kg) also provided less protein at an intake of 131 mg N/kg.

Despite a considerable range of protein intakes, there were few differences between the periods, and 10 of the 11 subjects were in positive N balance throughout. Poor agreement was observed with simultaneous energy expenditure measurements that the authors believed to be relatively insensitive, nevertheless large energy effects on N balance were observed, and it was concluded that, provided energy needs were met with ad libitum rice, relatively low protein supplies from mixed sources could meet the needs of young Thai men. This study has important implications for assessing protein needs in developing countries.

The final study to be discussed, examined excess energy and nitrogen balance at protein intakes above the requirement level in young men (CHIANG and HUANG, 1988). The subjects were given test diets with a fixed protein level of 1.2 g/kg (192 mg N/kg) but with three successive energy (E) levels at usual, ad libitum activity (1.0 E), 15% above (1.15 E) and 30% above (1.30 E) both in ascending and in descending sequences. These averaged 46, 53 and 60 kcal/kg. The duration of each dietary period was ten days. Nitrogen balance increased from 7.2 to 23.8 to 33.3 mg N/kg in the ascending series and decreased from 27.8 to 17.6 to 4.8 mg N/kg in the descending series. The changes in N balance per 100 kcal change in energy intake range between 144 and 243 mg N with smaller changes at the higher energy levels.

All of the studies described above showed significant effects of energy intake on protein metabolism, as evidenced by changes in nitrogen balance, but the degree of the change (Tables 1 and 2) depended on the experimental design. However, we have organized the data using constant units of measurement and now attempt to integrate the data and to draw some general conclusions. This exercise is reported in the next section.

Continue

Contents - Previous - Next