Proc. Natl. Acad. Sci. USAVol. 92, pp. 300-304, January 1995Applied Biological Sciences

Can amino acid requirements for nutritional maintenance in adulthumans be approximated from the amino acid composition ofbody mixed proteins?

(obligatory amino acid losses/amino acid patterns)

VERNON R. YOUNG AND ANroiNE E. EL-KHOURYLaboratory of Human Nutrition, School of Science and Clinical Research Center, Massachusetts Institute of Technology, Cambridge, MA 02139

Contributed by Vernon R. Young, September 15, 1994

ABSTRACT The quantitative needs for the dietary indis-pensable amino acids in adult human protein nutrition arestill poorly established. Tracer studies with 13C-labeled aminoacids have been undertaken previously in our laboratories toreevaluate and further determine the minimum physiologicalneeds for selected indispensable amino acids in healthy adultvolunteers. For those amino acids that have not yet beenstudied by this approach we have proposed a tentative set ofrequirement figures based on considerations ofthe amino acidcomposition of body mixed proteins and the rate of obligatoryamino acid losses (i.e., losses when the diet contains noproteins or amino acids). Here we provide an argument for,and a justification of, this approach as an interim measureuntil more comprehensive data become available on the quan-titative aspects of amino acid metabolism in healthy humans.

We have carried out a series of stable nuclide-enriched, aminoacid tracer studies to reevaluate the minimum physiologicaldietary requirement for the indispensable (essential) aminoacids (IAA) in healthy adult humans (1-3). Further, we haveconcluded that the current international minimum amino acidrequirement values are far too low for this population agegroup. For those amino acids that we have not yet examinedin tracer studies we estimated the requirements for them basedon an estimation of the obligatory oxidative losses (OOL;oxidative losses when the diet contains no proteins or aminoacids) and minimum dietary intakes required to balance these(2). The pattern of total daily obligatory amino acid loss(OAAL) reflects that of the relative concentrations of aminoacids in whole-body mixed proteins and so we used this patternto approximate the required dietary intakes of some of thespecific IAA (2, 3). However, our assumption that the aminoacid requirement pattern for maintenance of protein nutri-tional status in adults is similar to that for mixed proteins in thewhole body has been criticized by a number of investigators(4-7). Furthermore, Millward et al (8) state that we (9) do notjustify, or explain, our use of the OOL as the basis for ourpattern, although we have attempted to do this, to a limitedextent, in a number of our more recent papers (3, 10). Thesecriticisms must be taken seriously since our tentative aminoacid requirement pattern (11) has potentially important im-plications for the design of sound food and nutrition policiesand programs, including agricultural research directed towardhuman nutrition. In this paper, we will further explain, andjustify, the basis for this amino acid requirement pattern. Indoing so, we will evaluate the basis for, and strength of, thearguments that have been used to disagree with our proposedpattern (4, 5, 7).

The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked "advertisement" inaccordance with 18 U.S.C. §1734 solely to indicate this fact.

Initial Proposal

In adults, when intakes of energy and other nutrients areadequate but the diet is essentially protein-free, the rate ofbody nitrogen (N) loss, pritcipally via urine and feces, reachesa new, relatively steady-state level within about 1 week (12, 13).This new level is called the obligatory N loss (ONL) (14) andamounts to 54 mg ofN kg-lday-1 in healthy adults (15). It canbe assumed reasonably that the amounts of the different IAAcontributing to these N losses occur in proportion to theirconcentrations in body mixed proteins, providing that thoseproteins contributing to the major proportion of the total Nloss do not have amino acid patterns (concentrations) thatdiffer markedly from the average of the body mixed proteins.

Initially, we assumed that the OOL were due entirely tooxidative catabolism and so we referred to these as obligatoryoxidative losses of amino acids. Although the major route ofloss of these amino acids is, indeed, via oxidative catabolism,the amino acids are also lost in urine in small quantities or viathe intestine (particularly for threonine) in the ileal digesta(16). However, this is not a crucial issue for our proposalwhether all of the amino acids are terminally oxidized orwhether some are lost intact; body amino acid losses have beenpredicted from total ONL. Nevertheless, it would have beenmore appropriate to have referred to these as OAAL ratherthan to call them OOL; we will use the term OAAL in ourfurther considerations below.Having estimated the OAAL, we then assumed, from

whole-body N balance studies in humans (14, 15), that at abouta requirement level of intake, individual IAA would beretained with an efficiency of about 70%. Further research isneeded, however, to determine whether it is valid to use asingle figure for all of the amino acids for this purpose and sowe need to know the actual value(s) with which the variousabsorbed IAA coming from the diet are retained (and beforethey are lost from the body). For present reference, however,our proposed amino acid requirement pattern is given in Table1 and it is compared here with the adult amino acid require-ment pattern prepared by the 1981 Food and AgricultureOrganization/World Health Organization/United NationsUniversity (FAO/WHO/UNU) Expert Consultation on En-ergy and Protein (15). It can be seen that the internationalrecommendation is about one-half to one-third of the level thatwe have proposed as a necessary minimum intake to maintainprotein nutritional status in healthy adult humans.

Pattern of Amino Acid Requirements

Amino Acid Requirements in Preschool Children. To assessthe validity of the approach described above it is instructive to

Abbreviations: OOL, obligatory oxidative loss(es); IAA, indispensableamino acids; OAAL, obligatory amino acid loss(es); ONL, obligatoryN loss(es); SAA, sulfur amino acids; FAO/WHO/UNU, Food andAgriculture Organization/World Health Organization/United Na-tions University.

300

Dow

nloa

ded

by g

uest

on

May

7, 2

020

Applied Biological Sciences: Young and El-Khoury

Table 1. Newly revised estimates of the amino acid requirements in adults and a proposed aminoacid scoring pattern for this age group compared with the FAO/WHO/UNU requirements

Revised estimates' 1985 FAO/WHO/UNURequirement, Scoring pattern, requirement,

Amino acid mg-kg-l.day-1 mg/g of proteint mg-kg-l.day-lIsoleucine 23 38 10Leucine 39 65 14Lysine 30 50 12Methionine and cystine 15 25 13Phenylalanine and tyrosine 39 65 14Threonine 15 25 7Tryptophan 6 10 3.5Valine 20 35 10

*From ref. 11.tHere and in Table 2 we use the term protein according to the conventional definition ofN x 6.25 (crudeprotein).tSee ref. 15.

apply it to the preschool age group, for which amino acidrequirements have been established using multiple criteria,including N balance and plasma amino acid levels, by workersat the Institute for Nutrition for Central America and Panamain Guatemala (17). These data were used by FAO/WHO/UNU (15) to establish the recommended amino acid require-ment for preschool and young school-age children. Hence, wefelt that it would be of interest to learn whether it is possibleto predict, with reasonable accuracy, these requirements fromOAAL, also taking into account the amounts of amino acidsthat are deposited as a net body protein gain. We have madethese calculations (10) and, as summarized in Table 2, it isevident that our predicted requirement values are close tothose proposed by the international group (Table 2, columns1 and 2). It appears, therefore, that the requirements forpreschool children can be approximated, initially, fromOAAL. Further, the experimentally determined pattern ofamino acid requirements is similar to that of body mixedproteins (Table 2, columns 6 and 7). Although there is a dailynet deposition of body protein in this age group, it should beappreciated that "maintenance" accounts for about 90% of thetotal daily protein (nitrogen or IAA) requirement in the5-year-old child (15). On this basis, the amino acid requirementpattern (essentially for maintenance) in the young child ap-pears to be similar to that for mixed proteins in the whole body.

Perhaps it might be pointed out that, in spite of the similarityin the requirement pattern as discussed above, the absoluteconcentration of each of the IAA per unit of dietary protein (orN) required is lower than for the concentration of that amino

acid in mixed proteins of the body (Table 2, compare columnsC and D). This does not weaken our argument, however,because the efficient utilization of IAA requires an adequatelevel of dietary, nonspecific nitrogen, supplied in the form ofthe nutritionally dispensable amino acids or other, nontoxic,utilizable nitrogen sources, including urea (20, 21). Our (22)recent experiments reveal that the quantitative relationshipsbetween the indispensable and dispensable amino acid nitro-gen components of the total protein requirement deservefurther investigation in human subjects.Amino Acid Metabolism and Requirements in the Pig: A

Model for the Human? Millward et aL (7) have based theircriticisms of our approach taken to arrive at a tentative set ofrequirement figures for adults largely on the basis of estimatesof amino acid requirements for maintenance and growth in thegrowing 40- to 45-kg (body weight) pig [as reported by Fulleret aL (23)]. Hence, Millward et aL (7) have said that the patternof the OOL is more like the dietary pattern of amino acidsrequired for growth than for the amino acid pattern that theyconsider to be required for maintenance in this animal model.From such a comparison these investigators (7) have acceptedthat "the pattern of the OOL must (present authors' emphasis)be quite different from the minimum requirement pattern."However, we consider that this conclusion is based on a flawed,quantitative extrapolation from observations made with a pigmodel to the human nutritional context. Our reasoning is asfollows. First, the proportionate contributions made to thetotal amino acid (and nitrogen) requirement by that formaintenance or for growth are profoundly different for the

Table 2. Comparison of amino acid requirements in 2-year-old children: Estimations from OAALs with those of FAO/WHO/UNU/1985Requirement basis

Predictions 1985 FAO/WHO/UNUfrom OAALs, Body protein,mg-kg-l-day- * mg-kg-l-day-1 g/100 g of proteint Ratio g/16 g of Nt C/D§ C/Dt

Amino acid (A) (B) (C) A/B (D) (% of Lys) (% of SAA)

Leucine 53 73 6.6 0.7 6.9 130 (108)11 130 (108)11Isoleucine 31 31 2.8 1.0 4.1 92 94Valine 38 38 3.5 1.0 5.0 95 95Lysine 60 64 5.8 0.94 7.8 (100) 100Threonine 32 37 3.4 0.86 4.2 109 110Aromatic (Phe + Tyr) 58 69 6.3 0.84 7.6 112 113Sulfur (Met + Cys) 26 27 2.5 0.96 3.4 99 (100)

See ref. 15 for FAO/WHO/UNU data. SAA, sulfur amino acids.*Details for predictions given in ref. 10.tValues are g/100 g of protein required, based on the FAO/WHO/UNU (15) estimate of 0.99 g of protein per kg of body weight per day.tFrom Reeds (18).§Figures adjusted with the ratio for lysine taken as 100.1Figures adjusted with the ratio for SAA (Met + Cys) taken as 100.'lValue in parentheses for an assumed body mixed protein leucine concentration of 8.2% (19).

Proc- NatL Acad ScL USA 92 (1995) 301

Dow

nloa

ded

by g

uest

on

May

7, 2

020

302 Applied Biological Sciences: Young and El-Khoury

growing pig as compared to both the young child and humanadult. From the data of Carr et at (24), we can calculate thatthe growth component accounts for about 90% of the totalrequirement in the 40-kg pig and this component continues toremain high as the animal progresses toward a mature weight(>140 kg) (10). In the growing rat the maintenance need mightbe as low as about 5% of the total requirement (25). In thehuman subject, in contrast, the growth component of the totalprotein requirement is considerably lower than that for pigsand rats, being about 50% at 6 months of age and thendeclining to about 20% in the 2-year-old child and 15% or lessin the 4-5 year old (15). Thus, on this basis alone, thequantitative importance of the metabolic processes responsi-ble for the amino acid requirement in the growing pig (and rat)differs profoundly from that for the preschool child. In addi-tion, the efficiencies of dietary nitrogen (protein) utilization,and presumably that of dietary amino acid retention, atmaintenance intakes in the 40-kg pig (i.e., Millward's model forman) and the 2- to 5-year-old child are substantially different.According to Fuller et at (23), the young pig uses dietarynitrogen with an approximate 100% efficiency for mainte-nance. Here we define efficiency as the dietary intake of Nrequired to balance the losses of N that would occur with a

N-free diet. In the young child, on the other hand, it is substan-tially less than the estimate for the pig, being about 70%according to FAO/WHO/UNU (15). Furthermore, this valuealso applies to that for healthy adults (15). Therefore, the dataof Fuller et at (23) cannot be used to resolve problemsconcerning quantitative aspects of human amino acid require-ments.

If we compare the amino acid requirement pattern for adultpigs (boars) (26) with that for body proteins (18, 27) we alsofind that, when related to the lysine requirement, the patternof dietary requirement is not too dissimilar from that for bodymixed proteins. However, we have been informed by D. H.Baker of the University of Illinois (personal communication)that the National Research Council requirement pattern (26)is based on very weak data and so we do not emphasize undulythese numbers in our account of the issue at hand. Neverthe-less, when we also compare the profile of amino acids requiredfor near maximum growth in the rat (28) with the amino acidcomposition of the body of the rat (27) the relative values agreeno better with each other than those based on our comparisonsabove for the mature pig. The point we are making here is thatthe pattern of amino acids required for rat growth does notnecessarily resemble any more precisely the amino acid com-position of mixed body proteins (or pattern of OAAL) ascompared with that for maintenance in the preschool child(Table 2) (and presumably the adult). Again, this is in contrastwith the opinion expressed by others (7).

Finally, the question can be raised as to the physiological, aswell as nutritional, significance of maintenance amino acidneeds for the growing organism; at these intake levels theorganism would effectively be receiving a deficient intake inrelation to its full metabolic potential. It will be of interest tolearn whether the maintenance amino acid requirement pat-tern proposed by Fuller et at (23), based on studies in thegrowing pig, is capable of efficiently maintaining protein ho-meostasis in the mature, nongrowing pig that is given an adequatebut not excessive intake of total nitrogen. We raise this issue alsobecause direct estimates of the maintenance requirements forthreonine, SAA, isoleucine, and lysine in 145-kg gilts (29, 30) donot agree with those proposed by Fuller et at (23). Indeed, thedifferences between these various published estimates may be as

much as 2-fold, further weakening the arguments and conclusionsdrawn by Millward et at (7). Also, while the University of Illinoispattern (UIP) of amino acids required by adult swine (29, 30)does not resemble that for mixed proteins in the body, the UIPis based on limited data, and the estimates of requirements showlarge differences depending upon whether a nitrogen balance

equilibrium or a relatively small positive body nitrogen balance of1 g is used as the criterion for evaluation. Finally, the rate ofwhole-body protein synthesis (WBPS) in growing pigs approxi-mates 8-15 g kg- 'day-' (31-33), in comparison with that in adulthumans of about 4 g kg- 'day-'; when protein and energy intakesare at about maintenance levels in young pigs the rates of WBPSare substantially below those achieved when the ration suppliestwice and three times the maintenance requirement (34) oradequate levels of protein (35). In contrast, according to Pacy etat (36), in adult human subjects there is relatively little effect onwhole-body protein turnover with changes in protein intake,during 2-week periods, over a wide range covering submainte-nance and supramaintenance levels. Hence, again differences inthe quantitative aspects of protein metabolism in the growing pigand adult human call for considerable caution when assessingamino acid metabolism and dietary requirements in adult humansubjects.

Other Considerations and Counterarguments

The So-Called "Anabolic Drive." Millward et at (8) haveproposed a descriptive model of whole-body protein andamino acid metabolism where a minimally adequate intake ofprotein (and IAA) is one that is sufficient to exert an acceptableanabolic drive, or the regulatory influence of amino acids inprotein metabolism. In this context, they further propose thatfor a single, rate-limiting IAA the magnitude of the obligatoryrequirement may be similar to its OOL. We accept that thiscould be so, providing that there is a highly effective mecha-nism for the conservation and retention of the amino acidwhen ingested at a physiologically limiting intake but not whenintakes of this or other IAA are just adequate. These inves-tigators (8) then use lysine, as an example, and compare thecurrent international human adult requirement value of lysineof 12 mg kg-1 day-1 with an estimated OAAL of 30mg kg-1'day-l (2), which they say indicates that there is a highcontent of lysine in body proteins but a relatively low metabolicneed. The effective conservation of lysine at limiting intakes isnow well recognized (37, 38) but this does not mean thatprotein nutriture can be adequately sustained at extremely lowamino acid intakes; indeed, improved conservation of lysine ingrowing animals is associated with diminished growth perfor-mance (37-39) and reduced protein turnover. However, wemight further consider the lysine requirement value of 12mg-kg-lday-1, which is accepted by Millward et at (8), inrelation to the necessary repletion of tissue protein that shouldoccur during the fed period of the day so that a balance isachieved following the loss during the fasting period of thetotal 24-hr period. We can do this in relation to a recent studyfrom Millward's laboratories (40) in which it was found that theloss of body N during the 12-hr fasting phase of the 24-hr daywas about 60 mg of N-kg-1.12 hr-1 for subjects receiving anadequate, but not excessive, 0.8 g of protein kg-l day-1 andwho were in an overall daily, negative body N balance of about-19 ± 4.5 mg of N-kg-1day-1. Taking, then, the value of 12mg of lysine-kg-1lday-1 above and also assuming a quantitativenet retention of this amino acid (i.e., no oxidative loss, whichwould be unlikely in reality) during the fed 12 hr, this repletionwould be equivalent of about 24 mg of protein N (N x6.25).kg-1 during this anabolic phase. Clearly, this would beinsufficient to balance the fasting loss of 60 mg of N kg-1 andso the subjects would remain presumably in a marked dailynegative N balance. This further assumes that the N loss isequivalent essentially to the rate of net urea production (orurea production from nonrecycled N), which we believe to bethe case (41), but if the latter rate is higher, as some havesuggested (42), then the subjects would be in even greaternegative N balance. Thus, from these N excretion data, it canbe concluded that a minimum intake of about 29 mg oflysine kg-1 during the day would be required, assuming the

Proc. Natl Acad ScL USA 92 (1995)

Dow

nloa

ded

by g

uest

on

May

7, 2

020

Applied Biological Sciences: Young and El-Khoury

lysine content of retained protein to be 7.8% and that nooxidation of lysine occurred during the 12-hr fed phase, whichwould again be unlikely at this lysine intake level (43). Fur-thermore, it may also mean that an anabolic drive would notoccur at 12 mg-kg-' level intake because plasma lysine levels,and presumably also tissue free amino acid pool concentra-tions (44, 45), would decline (43, 44) with feeding at thisrelatively low intake level. In sum, the various arguments byMillward et at (8) against the amino acid requirement patternbeing similar to that of body proteins do not stand up to closescrutiny.

Lysine may not be the rate-limiting amino acid, or the onethat "drives" the ONL. Hence, there is some evidence that thisdriver may be methionine and/or possibly threonine (46) in thegrowing rat. The currently stated requirements for the SAA(methionine plus cystine) and threonine in healthy adulthumans are 13 and 7 mg kg-l day-1, respectively (17) (Table1). In comparison, our proposed mean requirement values arealso 13 mg for SAA but they are higher, or 15 mg-kg- 'day-1,for threonine (2, 11). Using the same line of reasoning that weuse in the foregoing paragraph concerning lysine retention andrequirements, we predict that the 13-mg value for the SAArequirement (methionine plus cystine) would permit anachievement of body N equilibrium; on the other hand, with a7-mg threonine intake, the N balance would be distinctlynegative. However, a 15-mg threonine-kg-1 day-1 level wouldpermit a condition of close to body N equilibrium. Again,however, it seems unlikely that the oxidation rate of these twoamino acids would decline to essentially a zero value during theentire feeding period, although these rates may fall below thosefor the fasting state when the intakes are limiting (47).Actually, our OAAL procedure predicts a somewhat higherthreonine requirement level of about 21 mg kg-1 day-1. Thisfigure may somewhat exceed the minimum requirement ifthreonine oxidation adaptatively, and without untoward func-tional consequences, declines to a relatively low level during abrief period of the absorptive/postprandial phase of aminoacid metabolism.Amino Acid Use in Non-Protein Metabolic Pathways.

Amino acids are used to meet a plurality of metabolic func-tions, in addition to their well-established role as substrates forprotein synthesis (48). By essentially ignoring these additionalfunctions and metabolic pathways in our calculations andpredictions, this raises a potential criticism concerning thepresent approximation of adult human amino acid require-ments that we have based quite simply on the composition ofbody mixed proteins. Further, these functions probably bearlittle relationship to the pattern of amino acids involved inpolyribosome-directed protein synthesis and, thus, of the aminoacid composition and turnover of tissue proteins. Althoughthere is limited information on the quantities of amino acidsinvolved in these other metabolic pathways, we make thetentative conclusion that they do not have any major impact onthe quantitative values for irreversible amino acid losses anddaily amino acid requirements, at least when these are assessedfrom body amino acid balance criteria. To give just twoexamples: (i) although the approximate turnover of glutathi-one in adults is about 24 ,umol kg-l hr-1 (49, 50), there ispresumably an effective reutilization of the cysteine that isliberated, via the membrane-bound enzyme, 'y-glutamyl-transpeptidase, either into the plasma or following its intra-cellular release, with resynthesis of glutathione; (ii) themethionine used in creatine synthesis can be regeneratedeffectively via methylation of homocysteine. From a physio-logic standpoint these other metabolic pathways and/or ad-ditional functions are of vital importance but from the dataavailable they would not appear to represent quantitativelyimportant irreversible routes of loss of the IAA. Therefore,considerations of protein synthesis, balance and body proteinamino acid pattern would seem to be of paramount importance

in predicting approximate dietary requirements for IAA inadult humans. Nevertheless, we accept that an analysis of theimpact of altered levels of amino acid intake on the quanti-tative activity of these other metabolic pathways, and theirfunctional significance, should help to improve upon currentapproaches taken to estimate the minimum dietary intakes ofspecific amino acids necessary for maintenance of physiologicfunction and performance.

Relationships Between Amino Acid Oxidation of the Vari-ous Amino Acids. In his evaluation of our previous 13C tracerstudies (2, 3, 10) on the amino acid requirements of adulthuman subjects, Waterlow (5) suggested that the 13C tracerdata implied that when the intakes of all the IAA, other thanthe limiting one under test, are high then the activity of variousamino acid catabolizing enzymes is maintained at a high leveland the oxidation of the test amino acid would be incompletelysuppressed. His view suggests the requirements for the indi-vidual amino acids that have been determined in preschoolchildren (15, 17) and in adults (15) by altering the level of asingle, test amino acid while maintaining intakes of other IAAat adequate or generous levels, potentially would have led tooverestimates of actual needs. Further, Waterlow's (5) pointalso implies that the pattern of amino acid requirements isdetermined by the pattern of amino acids in an experimentaldiet, which, in turn, could be very different from the aminoacid composition of body mixed proteins. However, the sug-gestion made by Waterlow (5) is not supported by extensiveunpublished data (V.R.Y.) in which the relationship betweenthe rate of leucine oxidation in the fed state and the amino acidcomposition of various diets was examined. It can be con-cluded that the dietary level of leucine is the most important,quantitative determinant of its whole-body oxidation rate, withthe amino acid composition (pattern) and level of totalnitrogen in the diet having relatively little or no effect. Thispoint is supported by findings, even under conditions ofmarked dietary disproportions of the branched-chain aminoacids in human subjects (51) and of amino acid imbalances inrats (52), that the oxidation of the limiting amino acid is notnecessarily increased and may be reduced to quite low levels(52, 53).

Conclusions

From the foregoing we conclude that (i) the amino acidrequirements predicted from OAAL agree well with thosedetermined from metabolic studies in young, preschool-agechildren; (ii) the pattern of the amino acid requirement asderived from metabolic studies in this young age group issimilar to the composition of the mixed proteins in the body,despite the large maintenance and small growth componentsto the overall requirement for total nitrogen and amino acids;(iii) the arguments made by others against use of OAAL for,at least, a first approximation of the adult human amino acidrequirements are not supported by the available data; (iv) theprediction of adult amino acid requirements from OAAL hasmerit as a tentative basis for establishing the adult humanrequirements for those IAA that have not yet been reevaluatedand studied directly. Although the pattern of amino acidsrequired by the adult subject that is ultimately established, andlater agreed upon by authoritative national and internationalgroups, may differ from that of whole-body mixed proteins itappears that the proposed approach we have taken to arriveat potentially more realistic amino acid requirement figurescan be justified, in the interim. Thus, the answer to thequestion posed in the title to this paper is a qualified "yes;"indeed, there seems to be little alternative, at present.

We appreciate greatly the advice and direction given to us by Profs.D. H. Baker (University of Illinois, Urbana) and N. J. Benevenga

Proc Natt Acad ScL USA 92 (1995) 303

Dow

nloa

ded

by g

uest

on

May

7, 2

020

304 Applied Biological Sciences: Young and El-Khoury

(University of Wisconsin, Madison). This study was supported byNational Institutes of Health Grants DK 15856 and DK 42101.

1. Young, V. R. (1987) Am. J. Clin. Nutr. 46, 709-725.2. Young, V. R., Bier, D. M. & Pellett, P. L. (1989)Am. J. Clin. Nutr.

50, 80-92.3. Young, V. R. & Marchini, J. S. (1990) Am. J. Clin. Nutr. 51,

270-289.4. Fuller, M. F. & Garlick, P. J. (1994)Annu. Rev. Nutr. 14,217-241.5. Waterlow, J. C. (1994) The Requirements ofAdult Man for Indis-

pensableAminoAcids (Int. Dietary Energy Consult. Grp., NestleFoundation, Lausanne, Switzerland), in press.

6. Carpenter, K. J. (1994) Protein and Energy:A Study of ChangingIdeas in Nutrition (Cambridge Univ. Press, New York).

7. Millward, D. J., Price, G. M., Pacy, P. J. H. & Halliday, D. (1990)Proc. Nutr. Soc. 49, 473-487.

8. Millward, D. J., Jackson, A. A., Price, G. & Rivers, J. P. W.(1989) Nutr. Res. Rev. 2, 109-132.

9. Young, V. R. & Pellett, P. L. (1989) in Milk Proteins in HumanNutrition, eds. Barth, C. A. & Schlimme, E. (Steinkopff, Darm-stadt, F.R.G.), pp. 7-36.

10. Young, V. R. (1991) Zeit. Ernaehrungswiss. 30, 239-267.11. Young, V. R. & Pellett, P. L. (1990) Food Nutr. Bull. 12, 289-300.12. Rand, W. M., Young, V. R. & Scrimshaw, N. S. (1976) Am. J.

Clin. Nutr. 10, 639-644.13. Rand, W. M., Scrimshaw, N. S. & Young, V. R. (1989) in Protein

Quality in Humans: Assessment and in Vitro Estimation, eds.Bodwell, C. E., Adkins, J. S. & Hopkins, D. T. (AVI, Westport,CT), pp. 59-97.

14. FAO/WHO (1973) Energy and Protein Requirements: Report ofaJoint FAO/WHO ad hoc Expert Committee (WHO, Geneva),Technical Report Series No. 522.

15. FAO/WHO/UNU (1985) Energy and Protein Requirements: Re-port of a Joint Expert Consultation (WHO, Geneva), TechnicalReport Series No. 724.

16. Fuller, M. F., Milne, A., Harris, I. A., Reid, T. M. S. & Keeran,R. (1994) Am. J. Clin. Nutr. 59, 70-73.

17. Pineda, O., Torun, B., Viteri, F. F. & Arroyave, G. (1981) inProtein Quality in Humans: Assessment and in Vitro Estimation,eds. Bodwell, C. E., Adkins, J. S. & Hopkins, D. T. (AVI, West-port, CT), pp. 29-42.

18. Reeds, P. J. (1990) Proc. Nutr. Soc. 49, 489-497.19. Widdowson, E. M., Southgate, D. A. T. & Hey, E. N. (1979) in

Nutrition and Metabolism of the Fetus and Infant, ed. Visser,H. K. A. (Nijhoff, The Hague, The Netherlands), pp. 169-177.

20. Kies, C. (1972) Fed. Proc. Fed. Am. Soc. Exp. Biol. 31,1172-1177.21. Snyderman, S. E., Holt, L. E., Jr., Dancis, J., Roitman, E., Boyer,

A. & Baylis, M. E. (1962) J. Nutr. 78, 57-72.22. Hiramatsu, T., Cortiella, J., Marchini, J. S., Chapman, T. E. &

Young, V. R. (1994) Am. J. Clin. Nutr. 59, 1347-1355.23. Fuller, M. F., McWilliam, R., Wang, T. C. & Giles, L. R. (1989)

Br. J. Nutr. 62, 255-267.24. Carr, J. R., Boorman, K. N. & Cole, D. J. A. (1977) Br. J. Nutr.

37, 143-155.25. Benevenga, N. J., Gahl, M. J., Crenshaw, T. D. & Finke, D. J.

(1994) J. Nutr. 124, 451-453.

26. National Research Council (1988) Nutrient Requirements ofSwine(Natl. Acad. Press, Washington, DC).

27. Davis, T. A., Fiorotto, M. L. & Reeds, P. J. (1993) J. Nutr. 123,947-956.

28. Gahl, M. J., Finke, M. D., Crenshaw, T. D. & Benevenga, N. J.(1991) J. Nutr. 121, 1720-1729.

29. Baker, D. H., Becker, D. E., Norton, H. W., Jensen, A. H. &Harman, B. G. (1966) J. Nutr. 88, 391-396.

30. Baker, D. H., Becker, D. E., Norton, H. W., Jensen, A. H. &Harman, B. G. (1966) J. Nutr. 89, 441-447.

31. Simon, O., Munchmeyer, R., Bergner, H., Zebrowska, T. &Buraczewska, L. (1978) Br. J. Nutr. 40, 243-251.

32. Salter, D. N., Montgomery, A. J., Hudson, A., Quelch, D. B. &Elliott, R. J. (1990) Br. J. Nutr. 63, 503-513.

33. Reeds, P. J., Fuller, M. F., Cadenhead, A., Lobley, G. E. &McDonald, J. D. (1981) Br. J. Nutr. 45, 539-546.

34. Reeds, P. J., Cadenhead, A., Fuller, M. F., Lobley, G. E. &McDonald, J. D. (1980) Br. J. Nutr. 43, 445-455.

35. Fuller, M. F., Reeds, P. J., Cadenhead, A., Sere, B. & Preston, T.(1987) Br. J. Nutr. 57, 287-301.

36. Pacy, P. J., Price, G. M., Halliday, D., Quevedo, M. R. & Mill-ward, D. J. (1994) Clin. Sci. 86, 103-116.

37. Brookes, I. M., Owens, F. N. & Garrigus, U. S. (1972) J. Nutr.102, 27-36.

38. Yamashita, K & Ashida, K. (1969) J. Nutr. 99, 267-273.39. Stockland, W. L., Mead, R. J. & Melliere, A. L. (1970) J. Nutr.

100, 925-934.40. Price, G. M., Halliday, D., Pacy, P. J., Quevedo, M. R. & Mill-

ward, D. J. (1994) Clin. Sci. 86, 91-102.41. El-Khoury, A. E., Fukagawa, N. K., Sanchez, M., Tsay, R. H.,

Gleason, R. E., Chapman, T. E. & Young, V. R. (1994) Am. J.Clin. Nutr. 59, 1000-1011.

42. Jackson, A. A. (1993) Proc. Nutr. Soc. 52, 1-10.43. Meredith, C. N., Wen, Z.-M., Bier, D. M., Matthews, D. E. &

Young, V. R. (1986) Am. J. Clin. Nutr. 43, 787-794.44. Munro, H. N. (1970) in Mammalian Protein Metabolism, ed.

Munro, H. N. (Academic, New York), Vol. 4, pp. 299-386.45. Bergstrom, J., Furst, P. & Vinnars, E. (1990) Clin. Sci. 79,

331-337.46. Yoshida, A. & Moritoki, K. (1974) Nutr. Rep. Int. 9, 159-168.47. El-Khoury, A. E., Fukagawa, N. K., Sanchez, M., Tsay, R. H.,

Gleason, R. E., Chapman, T. E. & Young, V. R. (1994) Am. J.Clin. Nutr. 59, 1012-1020.

48. Young, V. R. (1992) Scand. J. Nutr. 36, 47-56.49. Aebi, S., Assereto, R. & Lauterberg, B. H. (1991) Eur. J. Clin.

Invest. 21, 103-110.50. Bergunder, J.-M. & Lauterburg, B. H. (1981) Eur. J. Clin. Invest.

17, 408-414.51. Pelletier, V., Marks, L., Wagner, D. A., Hoerr, R. A. & Young,

V. R. (1991) Am. J. Clin. Nutr. 54, 395-401.52. Harper, A. E., Benevenga, N. J. & Wohlheuter, R. M. (1970)

Physiol. Rev. 50, 428-558.53. Yoshida, A., Leung, P. M. B., Rogers, Q. R. & Harper, A. E.

(1966) J. Nutr. 89, 80-90.

Proc- NatL Acad Sci USA 92 (1995)

Dow

nloa

ded

by g

uest

on

May

7, 2

020