Department of Nutrition and Food Science
Massachusetts Institute of Technology
Cambridge, Massachusetts 02139
Although the essential amino acids were recognized as a class early in the present century, it was not until the 1930's, when Rose assembled diets containing only purified amino acids in place of dietary protein, that progress could be made in identifying all the amino acids needed by man and in quantitating their requirements. For adult man, the eight essential amino acids are isoleucine, leucine, lysine, methionine, phenylalanine, threonine, tryptophan, and valine. Our ideas on the needs of man for each of these are largely based on N balance studies made by Rose and other investigators, most of which were already available to the authors of the 1965 Report on Protein Requirements. Irwin and Hegsted (1971 b) have recently summarized these studies in a detailed review and have reached the depressing conclusion that the estimated need of man for each of the essential amino acids falls far short of the quantities of these essential amino acids when consumed in the form of high quality protein in amounts sufficient to replace endogenous N output. Specifically, they find that the sum of essential amino acids required by young women is approximately 4 gm. daily, which would be provided by 8 to 10 gm of high quality dietary protein, whereas protein requirements measured with whole protein in the diet or endogenous N losses demand an intake of at least 20 gm of high quality protein daily.
In view of this conclusion, the published data on human amino acid requirements will be reviewed from three points of view. First, what conditions of the experimental design may have contributed to inaccurate estimates of essential amino acid needs ? Second, can we arrive at any reliable conclusions regarding the absolute or relative needs for essential amino acids on the basis of N balance data ? Third, what additional information is available to supplement estimates of requirements based on N balance studies ?
The limitations of N balance techniques have been frequently described. The most notable defect is that N intake tends to be overestimated and N output underestimated, so that there is a bias towards an erroneously positive balance. In addition, most or all N balance studies of amino acid requirements have considered only the urinary and fecal channels of N loss, and thus must appreciably underestimate true needs. To make comparisons of essential amino acid requirements with total protein requirements, one should use the mean urinary and fecal N outputs of subjects on a protein-free diet (Table I).
This would indicate a requirement of fully utilizable protein of about 20 gm/day for a 70 kg man.
A second problem is the criterion used to determine when amino acid intake has met such needs. Rose et al. (1954) accept as their end-point the lowest amino acid intake compatible with a "positive" N balance, whereas Leverton et al. (1956) designate an "equilibrium zone" within 5 % of balance. Clearly, these arbitrary rules leave room for some imprecision, as well as setting end-points at different levels of intake. In reading the final requirement, Rose chooses the highest need of any subject in a series, and doubles this value to provide a safe allowance.
Third, we have to consider the other components of the diet used in these balance experiments. Rose et al. (1954) used diets containing an amino acid mixture patterned after the requirements of the rat, whereas others (e.g. Leverton) have used the amino acid pattern of whole egg protein. The rest of the diet is usually made up of carbohydrate and fat sources low in protein, which have generally been regarded as introducing a negligible error into the estimate of amino acid needs. A recent study shows, on the contrary, that this may be an illusion. In 1966, Baker et al. performed a series of studies on adult swine, using a diet containing an amino acid mixture, with cornstarch as the main carbohydrate source. Under these conditions, they found that the swine could maintain positive N balances on a diet containing no leucine. In a subsequent study (Baker and Allee, 1970), they analysed their "leucine-free" diet and found considerable amounts of that amino acid to be present, mostly in the corn starch, which accounted for the apparent lack of leucine requirement by swine. This is obviously a hazard to be excluded in human studies; for example, Rose used corn starch in the diets employed for his studies of amino acid needs. Before accepting the reported amino acid requirements of man as accurate, it will be necessary to have further experimental data in which the amino acid content of the entire diet has been carefully checked. It may be remarked that most experimental work in this field was performed before the general availability of column chromatography of amino acids.
Finally, it is important to recognize that the other nutrients in the diet will influence amino acid utilization. In particular, N balance is affected by caloric intake and indeed Clark et al. (1960a) have demonstrated the considerable effect of caloric intake on the estimated lysine needs of their subjects. To arrive at accurate needs for essential amino acids, it would be necessary to have some guarantee that the caloric requirements of each subject were neither underestimated nor overestimated during the period of the study.
The available estimates obtained by nitrogen balance measurements are shown in Table II In view of the comments made above, it is not surprising to find lack of unanimity between investigators, not to mention a wide range of estimated needs even within a single study. In order to extract useful figures from the discrepant literature, it is suggested that the mid-range values obtained by Rose and his colleagues should be tentatively accepted; these have the merit that all the essential amino acids were studied by a single investigator employing constant criteria. Table III shows these values. When added together, they amount to 5.2 gm of essential amino acids. This can be compared with an average daily requirement of 19 gm of high-quality protein by an adult male (based only on obligatory urinary and fecal N outputs shown in Table I). In other words, essential amino acids account for only 27 % of the total protein needs of the adult. On the other hand, the eight essential amino acids with cystine and tyrosine constitute almost 50 % of whole egg protein, a source of high biological value. There is thus a two-fold discrepancy between the concentration of essential amino acids in the minimal amount of whole egg needed for N equilibrium, and the sum of the requirements for each amino acid measured separately.
Only two alternatives seem possible: either the estimated essential amino acid needs of man are only about half the true values, or else whole egg provides twice the minimum levels of essential amino acids needed, so that 73% of protein requirement is for nonessential nitrogen.
Table III also displays these estimated requirements for essential amino acids relative to tryptophan, and compares them with those of the young and adult rat, and with the ratios of essential amino acids in whole egg protein. The most striking feature of these data is the absence of agreement on any single point except that the requirements for tryptophan is less than that for any other essential amino acid. Even the two estimates shown for adult rate reveal considerable discrepancies. This is partly a reflection of the inaccuracy of balance determinations, but it may also be contributed in part by small amounts of plant proteins in the dietary carbohydrate sources, which will not affect the estimates of requirements uniformly. Thus corn protein is high in leucine but low in lysine and tryptophan, so that, as a consequence of contamination of corn starch with protein, leucine needs will appear to be much reduced whereas the apparent requirements for lysine and tryptophan will be less seriously affected.
It now seems established for some amino acids in the case of the rat and of the chick that, as the dietary level of the essential amino acid is raised, the plasma concentration of that essential amino acid remains low until the point of requirement is reached. Thereafter, the plasma concentration rises rapidly. The point of inflection seems often to be quite sharp, and thus marks the requirement of the animal. This principle has been used in man by Young et al. (1971) in their studies on human tryptophan requirements. With increasing increments in tryptophan intake, the plasma level of tryptophan remained steady until an intake of 3 mg/kg weight was reached, when there was a sharp rise in blood tryptophan level. Balance studies made by measuring urinary and fecal N output showed that equilibrium was achieved at 2.6 mg tryptophan/kg body weight. However, if cutaneous N losses are also taken into account, this would approximate a requirement of 3 mg/kg for N equilibrium. It is also probable that oxidation to CO2 and catabolic enzyme activity will show a similar sharply defined increase at the point of requirements, and will provide further evidence, at least in animal studies.
Extension of such procedures to all essential amino acids may help to resolve the problem of identifying the amino acid requirements of man more accurately than the rather arbitrary end-point in N balance studies. Unfortunately, they do not absolve the investigator from the tedious work of removing small amounts of essential amino acid contamination from the non-amino acid components of the test diets, and the problem of adjusting caloric intakes accurately for each subject.
The dielmma presented by published data on human amino acid requirements has already been stated, namely that only 27 % of the total protein needs can be accounted for as essential amino acid requirements, whereas our best protein standard (whole egg) contains 50 % essential amino acids. This discrepancy may occur because our estimates of individual essential amino acid needs are too low, or because whole egg has twice the essential amino acid content needed by the body, that is, the E/T ratio is grossly excessive. In this connection, it is interesting to examine results obtained on adult rats. Hartsook and Mitchell (1956) found that 3.2 % of whole egg protein was needed in the diet to achieve N equilibrium. By the same criterion, Benditt et al. (1950) and Smith and Johnson (1967) have measured the requirement of adult rats for nine essential amino acids, with cystine and tyrosine. Calculations based on their data show that the combined requirements for these amino acids amount to 1.2-1.3 % of the diet and of the caloric intake.
In relation to a total protein requirement of 3.2 % of the diet, this is once more much lower than the 50 % essential amino acid content of egg protein used as a reference standard. The question of whether whole egg contains a higher proportion of essential amino acids than is nutritionally necessary can best be determined by reviewing the literature on dilution of proteins with non-essential nitrogen. This topic is being covered in a separate report.
Factorial Calculation of Protein Needs of Adults
|1965 Report||Revised estimates|
|Obligatory N Losses:|
|mg N/kg b.w.|
|+ Stress factor (10 %)||95||-|
|g protein/kg b.w.|
|Obligatory N expressed as protein||0.59||0.34|
|+ 20 %||0.71||
|+ 30 %||
|Replacement by dietary protein (NPU 70)||1.01||0.64|
|Total protein intake (NPU 70) for 65 kg subject||66||42|
Estimates of the Amino Acid Requirements of Young Adults
|Essential Amino Acid||Sex||Estimated Requirement mg/day||Author|
|Isoleucine||4 M||650-700||Rose et al. (1955a)|
|7 F||250-450||Swendseid and Dunn (1956)|
|11 M+F||> 422||Linkswiller et al. (1960)|
|Leucine||5 M||500-1100||Rose et al. (1955a)|
|13 F||170-710||Leverton et al. (1956a)|
|Lysine||6 M||400-800||Rose et al. (1955b)|
|14F||400-500||Jones et al. (1956)|
|10 M+F||500-900||Clark et al. (1957)|
|5 M||400-1200||Clark et al. (1960b)|
|5 F||300-700||Clark et al.(1960b)|
|5 F||50||Fisher et al. (1969)|
|Methionine||6 M||800-1100*||Rose et al. (1955c)|
|8 F||150-350**||Swendseid et al. (1956)|
|20 F||300-550***||Reynolds et al. (1958)|
|Phenylalamine||6 M||800-1100||Rose et al. (1955d)|
|(no tyrosine)||6 F||834-1184||Tolbert and Watts (1963)|
|13 F||600-700||Burril and Schuek (1964)|
|9 M||900-1000||Burril and Schuek (1964)|
|Threonine||3 M||300-500||Rose et al. (1955c)|
|15 F||103-305||Leverton et al. (1956d)|
|Tryptophan||7 M||240||Denko and Grundy (1949)|
|3 M||150-250||Rose et al. (1954)|
|5 M||225||Baldwin and Berg (1949)|
|8 F||82-157||Leverton et al. (1956c)|
|5 F||50||Fisher et al. (1969)|
|2 M||6-9 mg/kg||Holt et al. (1944)|
|5 M||2-2.6 mg/kg||Young et al. (1971)|
|Valine||5 M||400-800||Rose et al. (1955e)|
|7 F||465-650||Leverton et al. (1956c)|
|7 F||230-480||Linkswiler et al. (1958)|
* No cystine
** 200 mg. cystine
*** Total S-amino acids
PATTERNS OF AMINO ACID REQUIREMENTS
|Essential Amino Acid||Adult Man||Adult Rat Ratio(b)||Adult Rat Ratio(c)||Young Rat Ratio(d)||Egg Protein Ratio|
|Methionine & Cystine||950||4.2||-||7.3||3.3||3.2|
|Phenylalanine & Tyrosine||950||4.2||-||3.2||4.8||6.1|
(a) Middle of range for studies for Rose et al. (see Table II).
(b) Benditt et al. (1950).
(c) Smith and Johnson (1967).
(d) Rama Rao et al. (1959).
Andik, I., Donhoffer, S., Farkas, M. and Schmidt, P. (1963). Brit. J. Nutr., 17, 257.
Ashworth, A. and Harrower, A.D.B. (1967). Brit. J. Nutr., 21, 833.
Baker, D.H., and Allee, G.L. (1970). J. Nutr., 100, 277.
Baker, D.H., Becker, D.E., Norton, H.W., Jensen, A.H. and Harmon, B.G. (1966). J. Nutr., 88, 382 and 391.
Baldwin, H.R., and C.P. Berg (1949). J. Nutr. 39: 203.
Benditt, E.P., Woolridge, R.L., Steffee, C.H. and Frazier, L.E. (1950). J. Nutr., 40, 335.
Bourges, H. (1968). Ph. D. Thesis, M.I.T.
Bricker, M.L., Shively, R.F., Smith, J.M., Mitchell, H.H. and Hamilton, T.S. (1949). J. Nutr., 37, 1963.
Burrill, L.M., and C. Schuck (1964). J.Nutr. 83: 202.
Calloway, D.H. and Margen, S. (1971). J. Nutr. 101, 205.
Calloway, D.H., Colasito, J. and Mathews, R.D. (1966). Nature, 212: 1238.
Clark, H.E., E.T. Mertz, E.H. Kwong, J.M. Howe and D.C. DeLong (1957). J. Nutr. 62: 71.
Clark, H.E., S.P. Yang, W. Walton and E.T. Mertz (1960a). J. Nutr. 71: 229.
Clark, H.E., S.P. Yang, L.L. Reitz and E.T. Mertz (1960b). J. Nutr. 72: 87.
Consolazio, C.F., et al. (1963). J. Nutr., 79: 399.
Costa, G. (1960). Nature, 188, 549.
Costa, G., Ullrich, L., Kantor, F. and Holland, J.F. (1968) Nature, 218: 546.
Cuthbertson, D.P. and Munro, H.N. (1937). Biochem. J., 31: 694.
Cuthbertson, D.P., McGuir, J.L. and Munro, H.N. (1937). Biochem. J., 31: 2293.
Darke, S.J. (1960). Brit. J. Nutr. 14, 115.
Denko, C.W., and W.E. Grundy (1949) J. Lab. Clin. Med. 34: 839.
Deuel, H.J., Jr., Sandiford, I., Sandiford, K. and Boothby, W.M. (1928). J. Biol. Chem. 76, 391.
Fisher, H., M.K. Brush and P. Griminger (1969) Amer. J. Clin. Nutr. 22: 1190.
Forbes, G.B. and Reina, J.C. (1970). Metabolism, Clin. and Exptl., 19, 653.
Gillett, L.H., Wheeler, L. and Yates, A.B. (1918). Am. J. Physiol., 47, 25.
Gopalan, C. and Narasinga Rao, B.S. (1966). J. Nutr., 90, 213.
Graham, N.M., Wainman, F.W., Blaxter, K.L. and Armstrong, D.G. (1959). J. Agr. Sci., 52, 13.
Hardy, J.D. (1961). Physiol. Rev., 41: 521.
Hartsook, E.W. and Mitchell, H.H. (1956). J. Nutr., 60, 173.
Hegsted, D.M., Tsongas, A.G., Abbott, D.B. and Stare, F.J. (1946). J. Lab. Clin. Med. 31: 261.
Hoffmann, L., and Schiemann, R. (1964). Arch. Tierernaehrung., 14: 23.
Holt, L.E., Jr., A.A. Albanese, J.E. Frankston and V. Irbay (1944). Bull. Johns Hopkins Hosp. 75: 353
Ingle, D.J., Meeks, R.C., and Humphrey, L.M. (1953). Am. J. Physiol., 173, 387.
Inoue, G., Fujita, Y. and Niiyama, Y. (1971). Unpublished results.
Irwin, M.I. and Hegsted, D.M. (1971a). J. Nutri., 101, 286.
Irwin, M.I. and Hegsted, D.M. (1971b). J. Nutri., 101, 360.
Issekutz, B., Rodahl, K. and Birkhead, N.C. (1962). J. Nutr. 78: 189.
Jones, E.M., C.A. Baumann and M.S. Reynolds (1956). J. Nutr. 60: 549.
Keys, A., et. A. (1950). The Biology of Human Starvation, Vol. I. Univ. Minn. Press.
Kraut, H. and Muller-Wecker, H. (1960). Hoppe-Seyler's Z. 320: 241.
Lathe, G.H. and Peters, R.A. (1949). Quart. J. Expl. Physiol. 35: 55.
Leverton, R.M. J. Ellison, N. Johnson, J. Pazur, F. Schmidt and D. Geschwender (1956a) J. Nutr. 58: 355.
Leverton, R.M., N. Johnson, J. Pazur and J. Ellison (1956b). J. Nutr. 58: 219.
Leverton, R.M., M.R. Gram, E. Brodovsky, M. Chaloupka, A. Mitchell and N. Johnson (1956c). J. Nutr. 58: 83.
Leverton, R.M., M.R. Gram, M. Chaloupka, E. Brodovsky and A. Mitchell (1956d). J. Nutr. 56: 59.
Lewis, M. and Evans, R.A. (1970). Proc. Nutr. Soc. (Eng. Scot), 30: 30 A.
Linkswiler, H., H.M. Fox, D. Geschwender and P.C. Fry (1958). J. Nutr. 65: 455.
Linkswiler, H., H.M. Fox and P.C. Fry (1960). J. Nutr. 72: 397.
Martin, C.J. and Robison, R. (1922). Biochem. J., 16: 407.
Mitchell, H.H. (1949). Arch. Biochem. 21, 335.
Mitchell, H.H. and Edman, M. (1962). Am. J. Clin. Nutr., 10: 163.
Mitchell, H.H. and Hamilton, T.S. (1949). J. Biol. Chem. 178: 345.
Munro, H.N. (1964). in "Mammalian Protein Metabolism", Vol. I, p. 381. (H.N. Munro, and J.B. Allison eds.). Academic Press, New York.
Munro, H.N. (1969) in "Mammalian Protein Metabolism," Vol. III, p. 133. (H.N. Munro, Ed.,) Academic Press, New York.
Murlin, J.R., Edwards, L.E., Hawley, E.E. and Clark, L.C. (1946). J. Nutr., 31, 533.
Playfair, L. (1865). Med. Times and Gazette, 1, 459.
Rama Rao, P.B., Metta, V.C. and Johnson, B.C. (1959). J. Nutr., 69: 387.
Reynolds, M.S., D.L. Steel, E.M. Jones and C.A. Baumann (1958). J. Nutr. 64: 99.
Robertson, F.W. (1962). Proc. Nutr. Soc. (Eng. Scot.) 21: 169.
Rose, W.C., C.F. Lamvert and M.J. Coon (1954). J. Biol. Chem. 211: 815.
Rose, W.C., C.H. Eades, Jr. and M.J. Coon (1955a). J. Biol. Chem. 216: 225.
Rose, W.C., A. Borman, M.J. Coon and G.F. Lambert (1955b). J. Biol. Chem. 214: 579.
Rose, W.C., M.J. Coon, H.B. Lockhart and G.F. Lambert (1955c). J. Biol. Chem. 215: 101.
Rose, W.C., B.E. Leach, M.J. Coon and G.F. Lambert (1855d). J. Biol. Chem. 213: 913.
Rose, W.C., R.L. Wixom, H.B. Lockhart and G.F. Lambert (1955e). J. Biol. Chem. 217: 987.
Rubner, M. (1902). "Die Gesetze des Energieverbrauchs", Deuticke, Leipzig.
Scherman, H.C. (1920). J. Biol. Chem., 41: 97.
Shock, N.W. (1970). J. Am. Diet. Assoc., 56: 491.
Sirbu, E.R., Margen, S., and Calloway, D.H. (1967). Am. J. Clin. Nutr., 20: 1158.
Smith, E.B. and Johnson, B.C. (1967). Brit. J. Nutr., 21: 17.
Smith, M. (1926). J. Biol. Chem., 68: 15.
Smuts, D.B. (1935). J. Nutr., 9: 403.
Steggerda, F.R. and Dimmick, J.F. (1966). Am. J. Clin. Nutr., 19: 120.
Swendseid, M.E. and M.S. Dunn (1956). J. Nutr. 58: 507.
Swendseid, M.E., I. Williams and M.S. Dunn (1956). J. Nutr. 58: 495.
Tolbert, B., and J.H. Watts (1963). J. Nutr. 80: 111.
Voit, C. (1878). Z. Biol. 14: 80.
Voit, E. (1930). Z. Biol., 90: 508.
Waterlow, J.C. (1969) in "Mammalian Protein Metabolism," Vol. III, p. 325. (H.N. Munro, Ed.) Academic Press, New York.
Yoshimura, H. (1955). J. Japan. Soc. Food and Nutr., 7: 155.
You, S.S., You, R.W. and Sellars, E.A. (1950). Endocrinology, 47: 156.
Young, V.R. and Scrimshaw, N.S. (1968). Brit. J. Nutr. 22: 9.
Young, V.R., Hussein, M.A., Murray, E. and Scrimshaw, N.S. (1971). J. of Nutr. 101: 45.