Proc. Nadl. Acad. Sci. USA Vol. 88, pp. 8091-8095, September 1991 Biochemistry Uniformly 13C-labeled algal used to determine essentiality in vivo (Spirua paensis/chicken/labeled egg /stable isotopes/I[U-_3C] labeling) HEINER K. BERTHOLD*t, DAVID L. HACHEYt, PETER J. REEDSt, OWEN P. THOMASt, SCOT HOEKSEMA§, AND PETER D. KLEINt tUnited States Department of Agriculture/Agricultural Research Service, Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, Houston, TX 77030; tDepartment of Poultry Science, University of Maryland, College Park, MD 20742; and §Martek Corporation, Columbia, MD 21045 Communicated by Vernon R. Young, June 6, 1991

ABSTRACT The edible alga Spirulina platensis was uni- Apart from the extremely high sensitivity with which formly labeled with 13C by growth in an atmosphere of pure uniformly labeled substrates can be detected, they can be 3C02. The labeled biomass was then incorporated into the diet used to great benefit in studies of intermediary metabolism. of a laying hen for 27 days. The isotopic enrichment of For example, several recent reports describing hepatic glu- individual amino acids in egg white and yolk proteins, as well cose metabolism in vivo have depended on the use of as in various tissues of the hen at the end of the feeding period, [U-_3Clglucose (9-11) and have included a detailed analysis was analyzed by negative chemical ionization gas chromatog- of the mass isotopomer labeling patterns in the circulating raphy/mass spectrometry. The amino acids of successive eggs glucose and its metabolites. This technique could be equally showed one of two exclusive enrichment patterns: complete useful in studies of amino acid metabolism and synthesis. preservation of the intact carbon skeleton or extensive degra- Although many aspects of amino acid metabolism are well dation and resynthesis. The same observation was made in understood, there are still important qualitative and quanti- tissue proteins. These patterns were cleanly divided according tative issues to be studied. Such issues include the scale and to known nutritional amino acid essentiality/nonessentiality regulation of the synthesis of an amino acid subgroup (ex- but revealed differences in labeling among the nonessential emplified by , , , and ) that has amino acids: most notable was that proline accretion was been categorized as conditionally essential (12). The avail- derived entirely from the diet. Feeding uniformly 13C-labeled ability of uniformly 13C-labeled amino acid precursors would algal protein and recovering and analyzing de novo-synthesized enable these questions to be studied experimentally. protein provides a useful method to examine amino acid Over the last 10 yr, our group has used bacteria (13), higher metabolism and determine conditional amino acid essentially in plants (14), and lactating women (15) to produce dietary vivo. carbohydrates and proteins labeled with stable isotopes. The carbohydrates and proteins have then been used to study The use carbohydrate digestion and fermentation (16-18) and to of stable isotopically labeled amino acids has been probe the details of the absorption of intact milk proteins by the key to many developments in our understanding of the premature infants (19). protein metabolism of humans (1-4). Stable isotopically The present paper reports on the practicality of producing labeled amino acids have also been used to examine some highly enriched, uniformly labeled proteins by growing alga important practical questions concerned with human nutri- in closed hydroponic culture under conditions in which the tion. In a recent series ofreports, Young and coworkers (5-7) sole source of carbon was 13C02. The method enabled the have reported the use of amino acids labeled with 13C on the isolation of algal biomass that was efficiently enriched and carboxyl group to examine the require- uniformly labeled. In >97% of the amino acid molecules of ments of adult men; the studies have provided invaluable the alga, all C atoms of the carbon skeleton were uniformly information about the obligatory metabolic "needs" forthese 13C-labeled. By incorporating the labeled alga into the diet of amino acids (3, 8). This approach, however, does not directly a laying hen, we were able to isolate highly enriched proteins address questions related to the efficiency with which differ- that are common components in human diets. Moreover, ent dietary proteins support metabolic requirements. Such analysis of the isotopomer pattern of the hen tissue and eggs information is necessary if biological needs are to be con- gave insights into avian amino acid metabolism under average verted into dietary recommendations. Questions about the feeding conditions. These results will enable us to gain extent to which dietary proteins support metabolic require- additional information about human amino acid metabolism. ments can best be answered by comparing the use of dietary protein with free amino acids administered i.v. This method would enable an estimate of the significance of protein MATERIALS AND METHODS digestibility and absorptive processes for overall require- Methods. Spirulina platensis is a ubiquitous microorgan- ments of dietary amino acids. ism, which, when grown in the laboratory, contains :60- In general, 13C-, 2H-, and '5N-isotopes are inserted into 65% protein in its dry matter (20, 21). Although the content single amino acids via specific chemical reactions. This oflysine, cysteine, and especially is lower than in process limits the number of positions within a given mole- "good quality" proteins (refs. 21, 22; Table 1), there are cule that can be labeled, which, in turn, limits the metabolic individuals (e.g., dwellers around Lake Chad) who habitually information that can be obtained from studies with commer- consume S. platensis as a major protein source. In the cially available amino acids. present study, we produced 500 g ofS. platensis biomass 97% uniformly labeled with 13C by growing it in closed hydroponic The publication costs of this article were defrayed in part by page charge culture and gassing it with 99% 13C02 (23). payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact. *To whom reprint requests should be addressed. 8091 8092 Biochemistry: Berthold et a!. Proc. Natl. Acad. Sci. USA 88 (1991) Table 1. Amino acid composition (% of total amino acids) of S. 5988A GC/MS system using negative chemical ionization platensis, and the S. platensis/feed mixture with methane as the reagent gas. Selected ion monitoring was S. platensis, S. platensisi done on the [M-HF] fragment from all ions [M] to [M + x + Amino acid % feed mixture 2], where x is the number of C atoms in the respective amino acid. (The mathematical procedures needed to deconvolute Glycine 9.83 4.72 the overlaying mass spectra of individual isotopomers can be 11.66 7.65 obtained upon request from the authors.) Thus, in an amino ND 5.75 acid of six carbons, the uniformly labeled molecule is de- 7.66 11.40 tected as a molecule with a mass number 6 more than the 4.92 5.09 [12C6]amino acid. In this example, the labeled amino acids in Proline 4.74 6.55 the diet were highly uniformly labeled with 13C; thus, mass 2.92 4.9 increments of [M + 1] . . . [M + 5] represent amino acids ND 4.25 derived by synthesis de novo within the hen. The ratio of ND 1.74 these lower mass numbers to the unlabeled [Ml and uniformly Methionine 0.56 2.19 labeled [M + 6] is a the 5.82 4.30 measure of proportion that has been 5.72 3.67 synthesized de novo. Cysteine ND 1.30 Aspartate/ 9.61 7.60 RESULTS Glutamate/ 14.55 15.53 The provision of '3CO2 to the photosynthetic organism S. 4.76 4.23 platensis during photosynthesis resulted in virtually com- Arginine 6.59 7.52 plete and uniform labeling of the algal amino acids. The 1.55 2.80 labeling of two representative amino acids, phenylalanine Composition of the diet mixture was calculated on the basis of and , in the alga is shown in Fig. 1. The separate analyses ofS. platensis, corn, and the amount oftryptophan dominating labeled species are the [M + 9] (phenylalanine) added to the diet. ND, not determined. and the [M + 5] (glutamic acid), with small contributions from species with mass numbers lower by 1 or 2. Fig. 2 shows the Feeding Trial. The uniformly labeled S. platensis was time course during which the two amino acids were incor- combined witkcorn and other dietary components to produce chicken feed.The composition of the feed was as follows (% by weight): corn 77.26, salt 0.12, limestone 4.016, dical 2.72, Phe'Ilic vitamin premix 0.033, trace mineral premix 0.033, choline chloride 0.087, soybean oil 1.87, S. platensis (lyophilized alga) 13.83, and tryptophan 0.025. The protein content ofthe diet was -'16% by weight. The amino acid composition ofthe A ''I- diet (Table 1) was calculated on the basis ofseparate analyses of the corn and S. platensis and the extra tryptophan. - Although all amino acids were appropriate for current rec- ommendations, we calculated that tryptophan was probably first limiting, followed by cysteine, lysine, and methionine. The feed was offered to a laying hen (DeKalb White Leghorn) for 27 days. The average feed intake was 100 to 120 g/day. ..i!; The hen laid 23 eggs over this period. At the end ofthe feeding .. period, the hen was killed, and its internal organs and carcass were individually deep frozen at -70°C. Each egg was opened under sterile conditions and separated into white and yolk fractions. The fractions were lyophilized and stored at El -600C. Protein Isolation and Gas Chromatography/Mass Spec- trometry (GC/MS). The lipids were extracted from the *1 1 I-.1b1 1 \ V Iyophilized yolks [3 mg of lyophilized yolk was extracted twice in a mixture of 2 ml of water and 3 ml of hexane/1- FIG. 1. Examples of the 13C-labeling pattern in essential (phen- propanol, 3:1 (vol/vol) solution], and the dry proteins were ylalanine) and nonessential (glutamate) amino acids of S. platensis hydrolyzed either with 6 M hydrochloric acid (114°C for 24 grown in hydroponic culture with 13CO2 (hatched). The predominant hr) or with 4 M methanesulfonic acid to recover tryptophan isotopic species in all amino acids analyzed was [M + x] (where x was the respective number of carbon atoms). The predominant species [114°C for 24 hr (24)]. The amino acids were isolated by contributed between 56 and 83% of the enrichment and confirmed cation-exchange chromatography, and their functional uniformity oflabeling. The mean percent abundance of [M + x] in S. groups were derivatized with hydrochloride isopropanol and platensis amino acids was significantly different between essential heptafluorobutyric anhydride for GC/MS analysis (25, 26). (64.5 + 4.68) and nonessential amino acids (77.9 + 3.78) (mean + SD; Tissue samples (Ql mg) were taken from both the internal P < 0.001). Formulation of the pure alga into poultry feed resulted organs and the carcass, homogenized in water, and after acid in a "dilution" of uniformly labeled (UL) amino acids to between 20 hydrolysis, the amino acids were derivatized by using the and 60%6 abundance (white). Black bars show the natural abundance same procedures described above. for comparison. These values were calculated on the basis of the The isotopic enrichment of glycine, alanine, valine, leu- natural isotopic abundances of all atoms forming the molecule cine, isoleucine, proline, tryp- analyzed by GC/MS-i.e., the atoms forming the amino acid plus phenylalanine, tyrosine, the atoms added to derivatize the functional groups ofthe amino acid. tophan, methionine, serine, threonine, asparagine and aspar- Note that an [M + 10] or higher isotopomer in phenylalanine can be tate, glutamine and glutamate, lysine, arginine, and histamine seen due to the statistical probability of having 13C atoms in the in the S. platensis/feed mixture, the egg yolks and whites, derivatizing groups and/or having heavy isotopes of hydrogen, and the tissues was determined by GC/MS. We analyzed the nitrogen, or oxygen in addition to the 13C atoms of the carbon amino acid derivatives in triplicate on a Hewlett-Packard skeleton. Biochemistry: Berthold et al. Proc. Natl. Acad. Sci. USA 88 (1991) 8093 Virtually all of the amino acids examined were either conserved without alteration (Fig. 2 Phe) or underwent extensive dismutation and reassembly (Table 2). These pat- terns followed the conventional categorization ofamino acids into nutritionally indispensable (label pattern conserved) and dispensable (label pattern randomized). There were, however, three exceptions. (i) On the basis of the isotopomer distribution, proline behaved as if it were an essential amino acid for egg and tissue protein synthesis and reached a final enrichment of the [M + 5] species equivalent to that in the diet. (ii) The second exception was cysteine; most (76%) ofit was incorporated into the avian proteins with a conserved carbon skeleton, but a small proportion entered Table 2. Percentage of nonessential and "semiessential" amino acids originating from the diet [M], [M + 1], [M + 2], [M + 3], [M + 4], [M + 5], Amino acid % % % % % % Egg white Glycine 71 6 23 Serine 69 9 6 16 Alanine 84 3 1 12 Cysteine 42 4 2 51 Aspartate/ asparagine 65 9 5 2 20 Glutamate/ glutamine 68 12 9 4 1 7 Methionine 43 2 1 2 15 377 Egg yolk Glycine 81 7 11 Serine 75 9 4 12 Alanine 82 5 1 11 FIG. 2. Incorporation oflabeled isotopomers in egg white protein 51 5 3 41 phenylalanine and glutamate over 27 days of feeding uniformly Cysteine labeled algal biomass to a laying hen. Note that because the ovula- Aspartate/ tion-oviposition cycle time in the chicken is >24 hr (27), an egg was asparagine 72 10 3 1 15 not laid on every day of the feeding period; to improve legibility of Glutamate/ the graphs, the gaps of missing data on days 9, 18, 21, and 26 have glutamine 76 12 7 2 0 4 been filled in by connecting the adjoining data points. Methionine 43 2 1 2 15 37 Liver porated into the egg white proteins. The egg protein isotopic Glycine 83 8 10 enrichments reached the asymptotic portion of the labeling Serine 80 12 5 3 time curve after -18 days. Although the amino acids in the Alanine 83 6 2 10 diet were labeled in virtually identical patterns, they were Cysteine 27 8 4 61 incorporated into egg proteins (data not shown) in two Aspartate/ discrete enrichment patterns. The two enrichment patterns asparagine 74 11 3 1 11 became apparent after correction of the distribution of iso- Glutamate/ tope in the two amino acids. (Mathematical procedures can glutamine 74 13 8 2 0 3 be obtained upon request from the authors.) Methionine 51 3 1 3 16 26 In the first isotopic enrichment pattern, the unlabeled Muscle phenylalanine in egg protein (Fig. 2 Phe) was replaced by a Glycine 90 3 8 single the [M + labeled form. Serine 87 6 3 4 isotopic species, 9] uniformly Alanine 95 2 0 3 The half-time ofthe incorporation process was 6.2 days in egg Cysteine 64 5 2 28 whites and 9.4 days in egg yolks. No other isotopic species Aspartate/ with less than nine '3C atoms could be detected during the asparagine 91 3 1 0 5 27-day period. The result, therefore, demonstrated in vivo the Glutamate/ essentiality of phenylalanine. glutamine 91 5 2 1 0 3 The second enrichment pattern ofnonessential amino acids Methionine 80 1 0 1 5 13 (exemplified by glutamate; Fig. 2 Glx) contrasted with that of Gut phenylalanine. Despite the fact that 35% of the dietary gluta- Glycine 86 4 10 mate was composed of a single isotopomer, ['3C5]Glx, the Serine 78 10 6 6 enrichment of this isotopic species in the egg white never Alanine 88 3 1 exceeded 6%, so that 94% of egg glutamate had been derived Cysteine 29 5 3 64 from de novo synthesis. Moreover, intermediate mass species Aspartate/ were clearly labeled, with the [M + 1], [M + 2], [M + 3], and asparagine 76 7 3 1 12 [M + 4] species accounting for '12%, 9%o, 4%, and <1% of Glutamate/ the total carbon, a further illustration of the extensive meta- glutamine 78 10 7 2 0 4 bolic degradation and de novo synthesis of glutamic acid Methionine 60 1 0 1 9 29 before its incorporation into protein. The uniformly labeled species is underlined. 8094 Biochemistry: Berthold et al. Proc. Natl. Acad Sci. USA 88 (1991) a pathway that led to enrichment in the [M + 1] and [M + 2] essential amino acids. Detailed data on lipid labeling will be isotopomers. (Mi) The third exception was methionine. In the reported elsewhere. egg protein, the enrichment of the [M + 4] species ap- proached 40% of the [M + 5] uniformly labeled methionine, DISCUSSION and there were, moreover, low levels of enrichment in the [M + 1] and [M + 2] species. The proportion of methionine The initial objective of our study was to examine the utility that had undergone replacement of the methyl group was ofusing algal biomass highly enriched with uniformly labeled determined by the ratio of [M + 4]/[M + 5]. The ratios in organic species to label proteins that are normally consumed liver, muscle, and gut proteins were 0.615, 0.385, and 0.31, by humans. Although the expense needed to obtain the stable respectively. isotopically labeled algal biomass limited the present exper- The absolute amount of uniformly labeled amino acid iment to one chicken, we were nevertheless able to obtain incorporated into tissue protein (Fig. 3) was largely a function gram quantities of highly enriched proteins. These proteins of the rate of protein turnover in each tissue. The highest can now be used in a variety ofways to examine questions of amounts occurred in liver, kidney, and spleen. The mass specific protein digestibility and the general use of dietary distribution of labeled species in both essential and nones- protein-bound amino acids. Uniformly labeled metabolic sential amino acids closely reflected the pattern in the eggs. substrates can also be used to study in vivo intermediary One possible exception was methionine in which the [M + 4] metabolism, and the present investigation provided some than in indications of possibilities that could be exploited. Through isotopomer was somewhat more enriched in hepatic detailed analysis of the mass isotopomer patterns of the egg proteins. protein-bound amino acids, we were able to examine some We also examined the labeling of selected lipid metabolites aspects of avian metabolic processes and to do so under in the egg. Although the S. platensis contained no enriched natural conditions. Although we have concentrated our ef- cholesterol, there was substantial enrichment of egg choles- forts on questions related to amino acid metabolism, the terol with surprisingly significant isotopic enrichment of examination offatty acid and cholesterol labeling shows that species as high as [M + 12]. Furthermore, examination of lipid metabolism could also be studied by this method. labeled unsaturated fatty acids in the egg triacylglycerol Over the last 3 yr, the estimation of amino acid needs, showed that the essential fatty acids linoleic and linolenic especially for adult humans, has been debated actively. were incorporated in the uniformly labeled form, as were the Previous approaches have been criticized on the basis that conditions under which measurements ofamino acid require- ments were made inevitably altered the nutritional and met- abolic status of the subjects (3, 8, 28, 29). This criticism implies that information so generated might have limited applicability to normal free-living humans consuming com- mon diets. The results of the present study offer an alterna- tive means by which to approach questions of essentiality of organic nutrients and, in particular, to address the specific problem of the so-called conditionally essential amino acids, such as cysteine and proline. In general, the labeling patterns in the egg and body proteins confirmed the well-established nature of the indispensable and completely dispensable amino acids for the laying hen. At the end ofthe 27-day study, the incorporation ofdietary essential amino acids into egg proteins was generally similar; the value suggested that approximately two-thirds had orig- inated from the diet. In contrast, the degree to which the 'IC label became randomized in the nonessential amino acids was remarkably high and showed the facility with which carbon is exchanged between these amino acids and the precursors and intermediates of the Krebs cycle. Even within a back- ground ofextensive metabolic interconversion, however, the degree to which the carbon skeleton ofdifferent nonessential amino acids was conserved varied not only among the amino acids but also among the various protein end-products. In the egg white proteins, for example, 23% of the glycine and 20% of the aspartate had been incorporated without metabolic transformation; in the same proteins, only 11% ofthe alanine and 7% of the glutamate had been transferred intact. In --Z.- addition, the degree of conservation was generally higher in '-". _-z egg protein compared with tissue protein, and the nonessen- tial amino acids in the egg white proteins had a generally higher uniformly labeled 13C enrichment than the yolk. In this FIG. 3. Percent incorporation from the diet ofthe essential amino context, however, it is important to remember that the acid phenylalanine and the nonessential amino acid glutamate into measurements were obtained after a prolonged labeling pe- tissue proteins. Enrichment of [13C9]Phe in muscle was -36% that in riod. The results suggest a tendency, even among nonessen- liver and egg white. The enrichment pattern for muscle glutamate was similar to that for egg proteins-i.e., the presence of a high tial amino acids, for the dietary amino acids to be channeled proportion of unlabeled material and all other isotopic species. The toward the pathway of net protein production-i.e., the egg. relatively high enrichment of [13C,]Glx in liver (13%) demonstrated The results also suggest that dietary amino acids escaped that this organ is particularly active in the synthesis of nonessential complete first-pass catabolism in the gastrointestinal tract amino acids. and liver, despite the adequate protein intake ofthe chicken. Biochemistry: Berthold et al. Proc. Natl. Acad. Sci. USA 88 (1991) 8095 Our study produced three particularly interesting results. States Department of Agriculture, Agricultural Research Service The first concerned proline, an amino acid for which growing under Cooperative Agreement 58-6250-1-003. birds have a distinct dietary requirement (30) and which is believed to be conditionally essential, at least for growing 1. Schoenheimer, R., Ratner, S. & Rittenberg, D. (1939) J. Biol. birds (31, 32). The feed used in our study was relatively rich Chem. 130, 709-714. in proline. In our study, however, proline behaved isotopi- 2. Waterlow, J. C. (1984) Q. J. Exp. Physiol. 69, 409-438. 3. Young, V. R. (1987) Am. J. Clin. Nutr. 46, 709-725. cally as an essential amino acid for both hen tissue and egg 4. Irving, C. S., Thomas, M. R., Malphus, E. W., Marks, L., protein synthesis. The results showed that no protein-bound Wong, W. W., Boutton, T. W. & Klein, P. D. (1986) J. Clin. proline had passed through a pool that allowed label equi- Invest. 77, 1321-1331. librium with glutamate (the precursor for proline synthesis). 5. Meguid, M. M., Matthews, D. N., Bier, D. M., Meredith, Early work suggested that egg production could be main- C. N. & Young, V. R. (1986) Am. J. Clin. Nutr. 43, 781-786. tained on a proline-free diet (33, 34); either no proline was 6. Meredith, C. N., Wen, Z.-M., Bier, D. M., Matthews, D. M. & synthesized in the well-nourished bird in the present exper- Young, V. R. (1986) Am. J. Clin. Nutr. 43, 787-794. iment or proline synthesized de novo was used in pathways 7. Zhao, X.-H., Wen, Z.-M., Meredith, C. N., Matthews, D. E., other than those for protein synthesis. Although similar Bier, D. M. & Young, V. R. (1986) Am. J. Clin. Nutr. 43, information could perhaps have been obtained by conven- 795-803. tional balance experiments involving graded deletions of 8. Reeds, P. J. (1990) Proc. Nutr. Soc. 49, 489-497. dietary proline, the experiments would have involved sub- 9. Katz, J., Lee, W.-N. P., Wals, P. A. & Bergner, E. A. (1989) have J. Biol. Chem. 264, 12994-13004. stantial alterations to the diet and presumably would 10. Kalderon, B., Gopher, A. & Lapidot, A. (1986) FEBS Lett. 204, induced metabolic adaptations that could have obscured the 29-32. general significance of the results. 11. Des Rosiers, C., Landau, B. R. & Brunengraber, H. (1990) Am. The second result that we found interesting was that J. Physiol. 259, E757-E762. cysteine behaved isotopically in a manner consistent with 12. Laidlaw, S. A. & Kopple, J. D. (1987) Am. J. Clin. Nutr. 46, that of a conditionally essential amino acid, despite the fact 593-605. that the cysteine/methionine ratio was adequate. Thus, a 13. Irving, C. S., Cooney, C. L., Brown, L. T., Gold, D., Gordon, high proportion of egg cysteine had been incorporated with a J. & Klein, P. D. (1983) Anal. Biochem. 131, 93-98. conserved carbon skeleton, but this proportion was lower 14. Boutton, T. W., Bollich, C. N., Webb, B. D., Sekely, S. L. & than that of the essential amino acids. Moreover, there was Klein, P. D. (1987) Am. J. Clin. Med. 45, 844 (abstr). labeling of the [M + 1] and [M + 2] isotopomers in the egg 15. Irving, C. S., Malphus, E. W., Thomas, M. R., Marks, L. & and tissue proteins, even though the diet contained a more Klein, P. D. (1988) Am. J. Clin. Nutr. 47, 49-52. than adequate cysteine/methionine ratio. In view of the high 16. Shulman, R. J., Boutton, T. W. & Klein, P. D. (1991) J. Pediatr. 118, 39-43. cysteine concentration in gut secretions, it was interesting to 17. Lifschitz, C. H., Wolin, M. J. & Reeds, P. J. (1990) Pediatr. find that uniformly labeled cysteine was greatly conserved in Res. 27, 165-169. the gut proteins. 18. Lifschitz, C. H., Torun, B., Chew, F., Boutton, T. W. & Klein, The third finding was a significant enrichment of the P. D. (1990) Pediatr. Res. 27, 539-544. [M + 4] isotopomer of methionine. The enrichment probably 19. Hutchens, T. W., Henry, J. F., Yip, T. T., Hachey, D. L., represents substitution of 12C for '3C in the methyl group as Schanler, R. J., Motil, K. J. & Garza, C. (1991) Pediatr. Res. a result of at least one cycle of de- and remethylation before 29, 243-250. incorporation into protein. There was also evidence that a 20. Chung, P., Pond, W. G., Kingsbury, J. M., Walker, E. F. J. & small proportion of the methionine was enriched in the Krook, L. (1978) J. Anim. Sci. 47, 319-330. + 1] and + 2] isotopomers, indicating the possibility 21. Clement, G., Giddey, C. & Menzi, R. (1967) J. Sci. FoodAgric. [M [M 18, 497-501. that a minor metabolic pathway, possibly related to polyam- 22. Aaronson, S. & Dubinsky, Z. (1982) Experientia 38, 36-40. ine metabolism (35), can lead to methionine synthesis in vivo. 23. Berthold, H. K., Hachey, D. L., Reeds, P. J. & Klein, P. D. Our investigation has shown that the use of ensembles of (1990) FASEB J. 4, A806 (abstr). nutrients uniformly labeled with 13C, coupled with the ability 24. Simpson, R. J., Neuberger, M. R. & Liu, T.-Y. (1976) J. Biol. to perform selected ion monitoring over the entire labeled Chem. 251, 1936-1940. mass range, enabled an examination of various aspects of 25. Liechtenstein, A. H., Cohn, J. S., Hachey, D. L., Millar, nutrient metabolism and interconversion. In effect, the pro- J. S., Ordovas, J. M. & Schaefer, E. J. (1990) J. Lipid Res. 31, cedure enabled the quantification of endogenous synthesis 1693-1701. under defined physiological conditions. Thus the conditional 26. Hachey, D. L., Patterson, B. W., Ghiselli, G., Marks, L., Cook, G., Brown-Booth, L., Gotto, A. M. & Klein, P. D. essentiality of some specific organic nutrients can be deter- (1987) in Proceedings ofthe 3Sth Annual Conference on Mass mined in real-life conditions without depriving the organism Spectrometry andAllied Topics, ed. McEwen, C. N. (Am. Soc. ofthe respective metabolites. In this way, adaptive metabolic Mass Spectrom., East Lansing, MI), pp. 269-270. responses that may alter the pathways under study can be 27. Johnson, A.-L. (1986) in Avian Physiology, ed. Sturkie, P. D. avoided. Short-term feeding of proteins uniformly labeled (Springer, New York), pp. 403-431. with '3C followed by the isotopic analysis of rapidly turning 28. Young, V. R. & Pellett, P. L. (1988) in Milk Proteins, eds. over plasma proteins could be used, for example, as a probe Barth, C. A. & Schlimme, E. (Springer, New York), pp. 7-36. of conditional essentiality for individual amino acids in hu- 29. Millward, D. J. & Rivers, J. P. W. (1988) Eur. J. Clin. Nutr. 42, mans. Such findings would be of particular use in studies of 367-393. 30. Graber, G. & Baker, D. H. (1973) Poult. Sci. 52, 892-899. infants in whom some important questions regarding amino 31. Rannels, D. E., Wartell, S. A. & Watkins, C. A. (1982) Life acid metabolism remain unanswered (36). Finally, the avail- Sci. 30, 1679-1685. ability of highly labeled dietary protein sources offers addi- 32. Griminger, P. & Scanes, C. G. (1986) in Avian Physiology, ed. tional opportunities for the study of protein digestibility. Sturkie, P. D. (Springer, New York), pp. 326-344. 33. Johnson, D. & Fisher, H. (1958) Br. J. Nutr. 12, 276-285. The authors thank E. Roseland Klein for editorial supervision of 34. Johnson, D. & Fisher, H. (1956) J. Nutr. 60, 275-282. this manuscript. This work is a publication of the United States 35. Edwards, C. H., Wade, W. D., Freeburne, M. M., Jones, Department of Agriculture/Agricultural Research Service, Chil- E. G., Stacey, R. E., Sherman, L., Seo, C.-W. & Edwards, dren's Nutrition Research Center, Department of Pediatrics, Baylor G. A. (1977) J. Nutr. 107, 1927-1936. College of Medicine and Texas Children's Hospital, Houston. This 36. Jackson, A. A., Shaw, J. C. L., Barber, A. & Golden, project has been funded, in part, with federal funds from the United M. H. N. (1981) Pediatr. Res. 15, 1454-1461.