Contents - Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
R. Bressani, D.A. Navarrete, and 0. Gutierrez
Division of Food and Agricultural Chemistry, Institute of Nutrition of Central America and Panama (INCAP) Guatemala City, Guatemala
It is a common practice to evaluate the protein quality of diets, ingredients, protein supplements. and new protein sources by the use of animal assay. A test with human subjects is of interest to reinforce animal data and to demonstrate their applicability to protein quality for human consumption. Protein quality assays in humans are. however, costly and of relatively long duration. Therefore, it would be desirable to have an assay of shorter duration, practical and valid, which would be capable of giving results comparable to those of other assays. This report presents additional information on a shortterm assay for protein quality that may be useful for routine use.
1. To compare the relationship of nitrogen intake to nitrogen retention using ascending and descending sequences of protein feeding.
2. To study the effect of variations in the descending sequence on the relationship between Nl and NB.
Study 1: Standard Short-Term Ascending-Descending Assay
Ten young men from 19 to 30 years of age, in good physical health with an average weight of 55 kg, were the subjects of study 1. The physical characteristics of these subjects are described in table 1.
Study 2: Modified Descending Assay
Thirteen healthy young men from 23 to 30 years old, with an average weight of 55.1 kg, were the subjects of study 2, whose physical characteristics are described in table 2. In both studies the subjects were selected on the basis of their previous protein and energy intakes, as well as for their physical and general health conditions.
TABLE 1. Physical Characteristics of Experimental Subjects: Standard Short-Term AscendingDescending Assay
|Subject||Age (years)||Weight (kg)|
|± S. error||24.8 ± 1.0||55.6 ± 1.9||54.3 ± 2.0|
TABLE 2. Physical Characteristics of Experimental Subjects: Modified Short-Term Descending Assay
|Subject||Age (years)||Weight (kg)|
|± S. error||25.4 ± 0.6||55.1 ± 1.1||55.4 ± 1.0|
TABLE 3. Composition of the Protein-Free Basal Diet: First Assay
|Artificial drink fruit flavour (glasses)||2|
|Mineral and vitamin supplement (tablets/day)a||1|
|Calorie Sources to Meet Requirements||Units|
|Carbonated drink (Pepsi-Cola)||1|
Study Environment: The subjects lived in Guatemala City and worked at INCAP. All of their meals were eaten in the Metabolic Unit of the Division of Food and Agricultural Chemistry. The daily ambient temperature ranged from 21 to 25°C, and relative humidity ranged from 72 to 85 per cent. Guatemala City is 1,510 m above sea-level.
Physical Activity: All men performed their usual chores at INCAP.
The diet described in table 3. Plantain flour made from half-ripe, drum-dried plantains was used to provide 25 per cent of the total energy intake equivalent to 45 kcal/kg/day. The protein source in all studies was provided by a pre-cooked bean flour containing 22.8 per cent protein. The protein intake per day was provided in three equal amounts at 7.30 a.m., 12 noon, and 5.30 p.m.
Standard Ascending Assay
This is described in table 4. It includes an adjustment period with animal protein included in the basal diet for six days, with faecal and urine collections for the last three days for nitrogen balance calculation. The low N basal diet follows for three days with faecal and urine collections during the last two days. This is followed by three levels of protein feeding of 0.2, 0.4, 0.6 kg/day each for two days with no adaptation in between. Faeces and urine are collected daily and pooled for two-day nitrogen balance calculation.
TABLE 4. Experimental Design of the Ascending-Descending Protocols
|Protein Level (mg/kg/day)||Days||Marker|
a. Animal protein-based diet
TABLE 5. Experimental Design for the Modifications on the Descending Protocol
|Protein Level (mg/kg/day)||Days||Marker|
a. Animal protein-based diet
TABLE 6. Nitrogen Retention at Different Levels of Protein Intake
|Design||Subjects||Average Intake (mg/kg/day)||N Retention|
|Ascending standard||10||24.7||-66.3 ± 14.1|
|52.8||-43.6 ± 8.4|
|85.9||-9.5 ± 6.9|
|117.4||-2.2 ± 7.0|
|Descending standard||10||119.5||-11.4 ± 4.7|
|85.9||-43.9 ± 5.4|
|60.4||-42.4 ± 4.0|
|25.4||-54.7 ± 3.7|
|Descending modification 1||7||119.7||-25.3 ± 7.8|
|86.2||-43.1 ± 5.4|
|56.4||-33.1 ± 2.7|
|26.2||-62.4 ± 4.6|
|Descending modification 2||6||120.5||-29.8 ± 8.1|
|103.2||-33.0 ± 8.9|
|86.8||-33.5 ± 6.2|
|73.7||-36.1 ± 12.7|
|51.9||-51.9 ± 9.9|
|43.5||-45.3 ± 9.5|
TABLE 7. Regression Equations between Nitrogen Intake and Nitrogen Retention
NI = a + b (NR)
r (P < 0.01)
NI for NR = 0 (mg/kg/day)
- 81.68 + 0.73
- 68.98 + 0.42
|Descending mod. 1||
- 72.46 + 0.49
|Descending mod. 2||
- 66.62 + 0.39
TABLE 8. Comparison of Nitrogen Intake Values for Nitrogen Equilibrium of Various Proteins Using a Very Short and Short-Term Assays
|Egg||Young et al.||88||-|
|Inove et al.|
|INCAPa||83 (16)||70 (4)|
|INCAP||88 (7)||74 (4)|
|Supro 620||Inoue||130 (14)||-|
|Supro 710||Wayler||114 (36)||-|
|Supro 220||INCAP||-||90 (20)|
|Glandless||Thomas et al.||106||-|
|Glanded||Alford et al.||106||-|
a. INCAP: Very short term assays Bressani and Navarrete
Standard Descending Assay
This is described in table 5. It follows the ascending assay with a three-day rest period in between in which the basal diet with animal protein if fed at 0.6 9 P/kg/day. Protein intake is then decreased from at 0.6 g/kg/day, to 0.4 0.2, and the low N-basal diet with the test protein for two days each.
Modified Descending Assay
Two modifications were introduced into the standard descending design. One consisted of extending to four days the protein levels of 0.4 and 0.2. The second modification consisted of keeping the number of days constant, at two days per protein level, with protein levels decreasing from 0.6 to 0.5, 0.4, 0.3, 0.2, and 0.1 g/kg/day. These are shown in table 5.
Table 6 summarizes the average nitrogen balance at the various levels of protein intake in the four designs studied, and table 7 summarizes the regressions of nitrogen intake to nitrogen retention. The greatest linearity, with the highest coefficient of regression, was observed with the ascending design. Among descending designs, which had similar coefficients, the standard and modification 1 showed the best linearity. Modification 2 of the descending designs had the lowest coefficient of regression and the lowest correlation. The table also shows the nitrogen intake needed for nitrogen equilibrium. These were 112, 162, 147, and 171 mg N/kg/day for the ascending and descending standard assays and for the descending modification 1 and 2, respectively. The average value of the three designs showing the greatest linearity between intake and retention was 140.
Conclusions and Comments
The ascending short assay has a tendency to overestimate the quality of the protein as well as the amount of nitrogen needed for nitrogen equilibrium. On the other hand, the descending design tends to underestimate the same values in its present sequence. Therefore, it seems justified to assay proteins by both protocols and calculate average values. It is of interest to point out that the ascending standard ranks proteins for their quality rather well.
This is shown in the values of table 8, which compares the nitrogen intake for nitrogen equilibrium of animal and vegetable proteins assayed by the very-short-term method and by other longer-term assays. Two conditions should be tried for the descending protocol; one is to increase the adjustment period and the other to extend the length of intermediate protein feeding periods, as in modification one of this report.
Division of Nutrition, National Institute of Nutrition, Tlalpan, Mexico D.F, Mexico
Vernon R. Young, and Nevin S. Scrimshaw
Department of Nutrition and Food Science. Massachusetts Institute of Technology, Cambridge, Mass., USA
To estimate the magnitude of the nitrogen losses through sweat induced by physical exercise and to test the effect of nitrogen intake and previous training on such losses.
Seventeen healthy young male university students participated. On average, they were 20 years old and had a height of 178.5 cm, a 74.7 kg body weight, a 1.91 m? of body surface, and a ponderal index of 42.5 ± 1.1. Ten of them were trained and seven were untrained.
The subjects lived and exercised in the air-conditioned Metabolic Unit of the MIT Clinical Research Center (22 to 24.5°C).
Exercise and Training
The exercise consisted of treadmill walking at a speed of 4 mph (6.44 km/h) at a 10 per cent grade, performed in the post-absorptive state. This represents a vertical climb equivalent to 644 m and requires the expenditure of 8 kcal per minute. The trained subjects exercised two separate hours per day for 24 days, after an eight-day period of step-wise adaptation, while the untrained group walked only once and for only 30 minutes.
Two diets were used: a nitrogen-free formula diet and a diet based on whole fresh eggs as the only nitrogen source, fed at different levels (either 0.3, 0.4, 1.7 or 2.0 9 of protein/kg BWt).
In both cases the diet was divided into four equal meals. The energy intake was based on each subject's dietary history plus the appropriate allowance for exercise; dextromaltose and corn oil provided twothirds and one-third of the non-protein calories, respectively. Sodium, potassium, iron, and vitamins were supplemented daily in excess of the requirements.
Sweat Collections and Determinations
The subjects thoroughly washed both arms. Preweighed plastic bags were placed around the arms, their upper ends fixed 10 cm below the acromion using surgical tape. The sweat collected was measured and frozen for analysis of total nitrogen (microKjeldahl), urea nitrogen, (Berthelot's reaction), and creatinine (Jaffe's reaction).
Body sweat rates were estimated by the difference between the pre- and post-exercise body weights. A correction (-5 per cent) was made to allow for respiratory water losses.
1. Sweat rates were estimated in every exercise session, i.e., 50 sessions for each of the 10 trained subjects (n = 500) and 2 to 3 sessions for the seven untrained subjects (n = 19).
2. Collection and analysis of sweat were done in only four of the trained subjects and in all seven untrained subjects as follows:
The four trained subjects were fed a diet providing 0.4 9 egg protein (64 mg N) per kg bwt for 44 days. The last 24 days they exercised for two hours a day and, on days 5 and 11, sweat was collected from both arms. On the second day after switching to a diet providing 2 9 egg protein (320 mg N) per kg bwt, sweat was again collected.
The seven untrained subjects were fed a nitrogen-free diet for 8 days. On the last day they exercised for 30 minutes and sweat was collected. Five of these subjects switched to a diet providing 0.3 9 egg protein (48 mg N) per kg bwt for seven days; they exercised for 30 minutes on days 1 and 7 and sweat was collected. The other two subjects switched to a diet containing 1.7 9 egg protein (270 mg N) per kg bwt; they exercised for 30 minutes on day 2 and sweat was collected.
TABLE 1. Sweat Rates in Trained and Untrained Subjects (mean ± S.D.)
|(g/hour)||(CV %)||(g/100 g bwt)||(CV %)|
|Trained||2 h/day x 25 days||611 ± 107||(17)||0.83 ± 0.10||(12)|
|n = 10||n = 500|
|Untrained||30 minutes only||739 ± 150||(20)||1.00 ± 0.10||(10)|
|n = 7||n = 19|
Results on trained and untrained subjects are shown in table 1. For the untrained subjects, one-hour rates were extrapolated from 30-minute rates.
Sweat rates were considerable with this kind of exercise. The rates of the untrained subjects, extrapolated from 30 minutes to one hour, were 20 per cent higher. Inter-and intra-individual variation was high (pooled CV = 17-20 per cent); expressing the results per kg bwt reduced the pooled CV to 10-12 per cent. Intra-individual variation in the trained group was less, averaging 7.8 per cent (range from 2 to 17 per cent: median 6.1 per cent). Some subjects were "profuse" sweaters and some were not throughout the study. A general tendency to lower sweat rates by the end of the 24-day period was apparent.
Arm-Sweat Volume and Composition
The sweat collected from the two arms represented 10.8 + 2.8 per cent of the total body-sweat volume. Sweat volumes and composition from the right and the left arms were almost identical.
The results of the analysis of arm sweat are presented in table 2. In all cases creatinine was present in only trace amounts.
Sweat nitrogen content ranged from 0.24 to 2.07 mg/ml, depending on protein intake and training, but most values were within the range reported in the literature (0.5 to 1.25 mg/ml). Inter-individual variability was considerable both in total nitrogen and in urea nitrogen content.
A higher protein intake resulted in higher nitrogen concentration in sweat in the trained subjects (925 mg/hour when on 2 9 protein intake/kg bwt, as compared with 310 mg/hour when on 0.4 9 of protein intake/kg bwt, but this effect was not observed in the untrained group.
TABLE 2 Arm Sweat Composition and Total Sweat Nitrogen Loss in One Hour in Trained and Untrained Subjects Fed Different Levels of Egg Protein Intake (Mean S.D.)
Nitrogen Intake (mg N/kg/day)
Day of Collection
Number of Measurements
Arm Sweat Composition
Total Sweat Nitrogen Loss (mg/hour) (CV %)
Total Nitrogen mg/ml sweat (CV %)
Urea Nitrogen (% of total N)
|Trained (4 subjects)||64||5||4||0.64 ± 0.37 (57)||49 ± 23.5||376 ± 235 (62)|
|11||4||0.38 ± 0.12 (31)||43 ± 5.2||232 ± 74 (32)|
|5 & 11||8||0.50 ± 0.29 (58)||52 ± 20.4||310 ± 149 (48)|
|320||2||4||1.25 ± 0.67 (50)||64 ± 15.0||925 ± 463 (50)|
|7||0||7||7||0.90 ± 0.54 (62)||26 ± 10.6||633 ± 410 (65)|
|5||48||1 & 7||10||0.70 ± 0.22 (31)||35 ± 10.9||518 ± 200 (39)|
|2||270||1||2||0.80 ± 0.56 (70)||49 ± 34.3||586 ± 317 (54)|
Expressed as mg per kg bwt per hour, sweat nitrogen was 4.8 ± 2.2 in the trained subjects fed a low protein diet (0.4 g/kg bwt) and 11.3 ± 5.9 when these subjects were fed 2g of protein/kg bwt. In the untrained group it was 7.7 ± 4.1.
In the trained subjects the nitrogen concentration was lower in the sweat obtained on day 11 compared to sweat obtained on day 5, suggesting a possible adaptation to conserve nitrogen as training progressed.
Nitrogen losses through sweat in one hour varied, reflecting the variability in sweat nitrogen concentration. These losses were only 232 mg/hour in the trained subjects on the 11th day of exercise, but reached 925 mg/hour in the same subjects fed a high-protein diet; the untrained group showed higher mean losses.
Heavy physical exercise of the type used in these experiments may be difficult to sustain for more than three or four hours, and the nitrogen losses observed in two hours do not necessarily hold for longer periods. If the data for two hours of exercise are extrapolated to four hours, assuming similar sweat volumes and concentration of nitrogen, 1.2 to 4 9 nitrogen - equivalent to 8 and 26 9 of protein - would be lost, making sweat a major route for nitrogen excretion.
Training and acclimatization have been reported to lower sweat nitrogen losses, and urinary nitrogen excretion is known to decrease when sweat losses are high. Until these adapting mechanisms are better tested under different conditions, however, sweat nitrogen losses deserve careful consideration when estimations are made of dietary protein allowances of manual workers in tropical environments favouring profuse sweating.
Although urea, which is a metabolic end-product, represents a large proportion (50-60 per cent) of the nitrogen found in sweat, creatinine and uric acid are present in negligible amounts, and ammonia may represent no more than 10 per cent, leaving amino acids as the most likely source of the remaining 30 40 per cent of the nitrogen; in the subjects studied, amino nitrogen would then represent about 200 mg/hour, equivalent to 1.25 9 of amino acids, or 5 grams in 4 hours.
1. Heavy dynamic exercise induced sweat losses in the order of 600 700 ml/hour. The values tended to vary little within subjects, but showed a higher inter-individual variation. Training caused some reduction in sweat rates.
2. The nitrogen content of sweat varied from 0.24 to 2.07 mgiml and urea accounted for about 50 per cent of it. A high protein intake resulted in higher nitrogen losses in sweat in the trained group, but it had no effect in the untrained group. Training caused some reduction in sweat nitrogen losses.
3. Under conditions favouring abundant sweat production, it became a major route of nitrogen excretion, much above the usual allowance factors for tegumental nitrogen used in balance experiments, and this needs special consideration in calculations of protein requirements.
Imran Ozalp, Meral Ozgüç, Suzan Tokol, Nevin Tasci, Ayse Baysal, and Turgay Coskin
Nutritional Metabolic Unit, Institute of Child Health Hacettepe Medical School and Department of Nutrition and Dietetics, Hacettepe University, Ankara, Turkey
In the short-term studies of nitrogen balance carried out previously, the duration of the adaptation period was only five days for the estimation of mean protein requirement. This study was carried out to investigate whether a longer time is necessary for adaptation to a given level of protein. Therefore, the nitrogen balances of 15 young adults on a safe level of protein intake, as estimated from a previous study, were investigated for ten days following a five-day adaptation period.
Subjects. Fifteen 19- to 28-year-old university students were studied. They were healthy, based on medical history, physical examination, and a routine complete blood count. Their characteristics are described in table 1.
Physical Activity. The subjects were allowed to continue their usual activities, except for participating in competitive, heavy sports.
Environment. The study was carried out in Ankara during the last two weeks of September. Average temperature was 15°C.
Diet and Experiments Design. Mixed habitual Turkish foods were given. As in the other balance studies, bread, preboiled wheat, and yogurt were the main foods consumed during the study. Table 2 shows the main composition of the diet. The subjects' customary dietary energy intake (three-day recall of dietary history! and protein/energy ratio, as suggested in the UNU standard protocol, were taken into account.
After one day of a protein-free diet, the subjects were fed a diet providing 0.8 g protein/kg/day. The first five days were for adaptation and this was followed by two five-day balance periods on the same intake. Four meals were provided under the close supervision of dietitians.
TABLE 1. Characteristics of the Young Adults Consuming Safe Level of N and Changes in their Body Weight on this Diet
Initial Body Weight
Final Body Weight
|Mean + SD||21.76 ± 2.52||171.86 ± 6.11||62.96 ± 5.79||62.68 ± 5.62||0.09 ± 0.56|
TABLE 2. Composition of the Diet (Mean Values for 15 Subjects)
|Kcal/kg/day||Percentage Caloric Fat||Percentage Intake CHO||Percentage Protein||Percentage Animal||Percentage Vegetable|
|49.5 ± 2.26||28.42 ± 5.33||65.07 ± 5.33||6.51 ± 0.29||15.92 ± 1.71||84.14 ± 1.58|
Measurements. After the adaptation period. 24-hour urine samples were collected for 10 days in HCl acid. An aliquot was analysed for total nitrogen, urea nitrogen, and creatinine. Faeces were collected daily and pooled during the last 10 days. Vegetable charcoal was used as a marker for pooled faeces collections.
At the beginning and end of the dietary periods, fasting blood samples were taken, and total protein, albumin, total lipid, cholesterol, SOOT, and SGPT determinations were performed.
TABLE 3 Nitrogen Balance and its Components
First Balance Period
Second Balance Period
N Intake (mg/kg)
N Urine (mg/kg)
N Faecal (mg/kg)
N Balance (mg/kg)
N Intake (mg/kg)
N Urine (mg/kg)
N Faecal (mg/kg)
N Balance (mg/kg)
Apt. Dig. (%)
N Balance (mg/kg) 1 st 2nd
|9||128.55||81.85||26.23||15.47||79.6||125.02||97.14||26.23||- 3.35||79.0||- 18.82|
|10||128.60||56.38||35.87||31.35||72.1||125.04||89.38||35.87||- 5.21||71.3||- 36.56|
|11||128.57||96.75||32.29||- 5.47||74.9||125.05||109.44||32.29||- 21.68||74.2||- 16.21|
|15||128.43||82.22||24.56||16.65||80.9||125.04||111.07||24.56||- 15.59||80 4||- 32.24|
TABLE 4. Heamoglobin and Some Serum Constituents of the Subjects (Mean Value of 15 Subjects) Consuming Safe Level of Nitrogen (128 mg N/kg/day)
|X ± SD||X ± SD|
|Hb (g/dl)||16.50 ± 1.61||15.19 ± 1.46|
|Total protein (g/dl)||7.81 ± 0.25||7.88 ± 0.19|
|Albumin (g/dl)||5.47 ± 0.47||5.51 ± 0.45|
|Total lipid (mg/dl)||647.33 ± 36.48||590.67 ± 123.03|
|Cholesterol (mg/dl)||102.60 ± 36.48||146.07 ± 42.61|
|SGOT (U)||9.67 ± 2.16||6.93 ± 2.60|
|SGPT (U)||11.73 ± 3.97||7.27 ± 3.77|
Nitrogen values of urine, stool, and liquefied diet samples were determined by the micro-Kjeldahl method, and nitrogen balance was calculated by subtracting the main daily urine and faecal nitrogen excretion from the daily nitrogen intake. Integumental and miscellaneous nitrogen losses were estimated at 5 mg N/kg/body weight/day.
Table 3 shows individual and mean values of nitrogen intake, faecal and urinary nitrogen output, and nitrogen balances of the subjects during the second and third five-day periods. Nitrogen excretion of the subjects did not change significantly throughout the last 10 days. There was no significant difference in the mean values of urinary excretion of 15 subjects between the second and third fiveday periods (78.96 ± 14.21 and 80.76 ± 23.42 mg/kg/day, respectively).
No significant changes were observed in body weights or in blood chemistries of the subjects during the study period (tables 1 and 4).
Since the N excretion of subjects did not change significantly in the second and third five-day periods, five days is long enough to adapt to a certain level of protein intake, and 0.8 9 protein/kg/day meets the daily nitrogen allowances of Turkish young adults consuming their habitual mixed diet.
Contents - Previous - Next