Human Nutrition and Metabolism

Transcription

1 Human Nutrition and Metabolism Regulation of Sulfur Amino Acid Metabolism in Men in Response to Changes in Sulfur Amino Acid Intakes 1,2 Marco Di Buono,* Linda J. Wykes,** David E.C. Cole, Ronald O. Ball* and Paul B. Pencharz* 3 *Department of Nutritional Sciences and Paediatrics, University of Toronto, Toronto, Canada M5S 3E2; The Research Institute, The Hospital for Sick Children, Toronto, Canada M5G 1X8; **School of Dietetics and Human Nutrition, McGill University, Montreal, Canada H9X 3V9; Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Canada M5G 1L5; and Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, Canada T6G 2P5 ABSTRACT We showed previously that 64% of the total dietary sulfur amino acid (SAA) requirement could be supported by dietary cysteine (Cys). However, the observation of such a sparing effect may be affected by the dietary intakes of SAA provided. The aim of this study was to compare methionine (Met) metabolism and transsulfuration (TS) in five healthy men fed three different diets (in random order) for 3 d each, with varying combinations of Met and Cys: 24 mg Met/(kg d) and no Cys (diet A); 13 mg Met/(kg d) and 11 mg Cys/(kg d) (diet B); and 5 mg Met/(kg d) and 19 mg Cys/(kg d) (diet C). On d 3, Met kinetics and TS were assessed using orally administered L-[1-13 C, methyl- 2 H 3 ]methionine. Met demethylation (transmethylation, TM) significantly decreased as the dietary Met to Cys ratio decreased. Met TS was significantly lower during diets B [ mol/(kg h)] and C [ mol/(kg h)] than during diet A [ mol/(kg h)] (P 0.05). The results of the present study indicate that when the ratio of Met to Cys fed is typical of that found in major food proteins and total SAA are sufficient to meet requirements, TS is significantly reduced compared with the case in which SAA needs are supplied by Met alone. We conclude that Cys sparing occurs through an increase in the fraction of the homocysteine pool destined for RM relative to TS (RM:TS). J. Nutr. 133: , KEY WORDS: methionine cysteine sparing homocysteine stable isotope men The ability of cysteine (Cys) to quantitatively reduce the requirement for methionine (Met) is well established in animals (1 5). Early studies in rats showed that Cys feeding resulted in a 20 70% reduction in dietary Met requirements (4,5). Early studies in humans using growth and nitrogen balance as indicators identified a Cys-sparing effect ranging from 50 to 90% (6 8). However, a recent series of reports that evaluated the responses of Met kinetics and transsulfuration (TS) 4 to varying intakes of Met and Cys suggested that there 1 Presented in part at Experimental Biology 2002, April 2002, New Orleans, LA [Di Buono, M., Wykes, L. J., Ball, R. O. & Pencharz, P. B. (2002) Replacement of dietary methionine by cysteine suppresses homocysteine transsulfuration in humans. FASEB J 16:A 258 (abs.)]. 2 Mead Johnson Nutritionals (Canada) provided the protein-free powder for the experimental diets. Whitehall Robins (Canada) provided the multivitamin supplements. The study was funded by the Canadian Institutes of Health Research, grant MT 10321, and the Heart & Stroke Foundation of Ontario (DECC), grant T4340. M.DeB. was supported by a Fonds pour la formation de Chercheurs et l Aide à la Recherche (FCAR) doctoral scholarship. 3 To whom correspondence should be addressed. paul.pencharz@sickkids.ca. 4 Abbreviations used: APE, atom % excess; CBS, cystathionine -synthase; GC/MS, gas chromatography/mass spectrometry; GNMT, glycine N-methyltransferase; IAAO, indicator amino acid oxidation; MPE, mole % excess; MTHFR, methylene tetrahydrofolate reductase; RDA, recommended dietary allowance; RM, remethylation; SAA, sulfur amino acids; SAM, S-adenosylmethionine; TCEP, tris(2-carboxyethyl) phosphine; thcy, total homocysteine; TTR, tracer-to-tracee ratio; TM, transmethylation, TS, transsulfuration. was no Cys sparing occurring in humans (9 14). These studies typically provided sulfur amino acids (SAA) at a level consistent with the FAO/WHO/UNU population safe level of intake for total SAA (15), which we have recently identified as being too low (16,17). The current population recommended dietary allowance (RDA) for total SAA in healthy adults, as proposed by the FAO/WHO/UNU (15), is 13 mg/(kg d). We carried out a stable isotope tracer study, using indicator amino acid oxidation (IAAO), to define the total SAA requirements as Met alone in healthy adult men (16). We found a mean requirement of 12.6 mg/(kg d) and derived a population RDA, meant to represent the 95% confidence interval of the mean, of 21 mg/(kg d) (16). Subsequently we determined that the mean requirement for Met was 5 mg/(kg d) in the presence of excess dietary Cys, which is only 64% of the mean requirement for Met in the absence of dietary Cys (17). When comparing population RDA, we found dietary Cys could furnish 52% of dietary SAA requirement (17). In an editorial written in response to the findings of Di Buono et al. (16,17), Young (18) stated that there are several limitations to the interpretation of the data, making it difficult to conclude definitively that the Met requirement can be reduced by Cys to as great a degree as that proposed. One of the major limitations cited by Young (18) is that the Cys intake studied relative to that of Met at the breakpoint is not characteristic of typical /03 $ American Society for Nutritional Sciences. Manuscript received 7 October Initial review completed 1 November Revision accepted 4 December

2 734 DI BUONO ET AL. FIGURE 1 Kinetic model of sulfur amino acid metabolism. I is the exogenous intake of methionine through diet or tracer administration; TM is the transmethylation rate, or rate of conversion of methionine to homocysteine; TS is the transsulfuration rate, or the rate at which methionine is oxidized; RM is the rate of methionine appearance from remethylation of homocysteine; S is the rate at which methionine is incorporated into newly synthesized proteins; B is the rate at which methionine appears from endogenous protein breakdown. Points of disappearance of labeled elements 13 C and 2 H 3 are indicated. The upper panel indicates the kinetics and isotopic determination of methyl (Q m ) and carboxyl (Q c ) fluxes; see also the methods section under experimental model. The lower panel shows how RM, TM and TS are determined. Met, methionine; SAM, S-adenosylmethionine; SAH, S-adenosylhomocysteine; HCY, homocysteine; Cys, cysteine. diets in which the Met to Cys ratio ranges from 1:1 to 2:1. The present experiment was designed to take this point into consideration when choosing the dietary levels of Met and Cys fed. In the first step of SAA metabolism (Fig. 1), Met and ATP condense to form S-adenosyl-methionine (SAM), the primary biological methyl donor in vivo. SAM is demethylated to S-adenosyl-L-homocysteine and subsequently converted to homocysteine, which is either remethylated by one of two enzymes to form Met or condensed with serine to form cystathionine. The synthesis of cystathionine by cystathionine -synthase (CBS) is the first reaction in the irreversible pathway for the catabolism of homocysteine, resulting in the formation of Cys, and then sulfate. This TS pathway has a limited distribution and is found primarily in the liver, kidney, small intestine and pancreas (19). Finkelstein and Mudd (20) and Finkelstein et al. (5) showed that Cys feeding leads to a decrease in tissue CBS content, thus providing a biochemical basis for Cys sparing of the Met requirement. More recent evidence from cell culture experiments of human CBS supports the notion of transcriptional regulation of CBS (21), and the short half-life of the major CBS isoforms (22,23) suggests that dietary Cys could possibly lead to changes in CBS mrna transcription, and therefore CBS activity, in the time frames of the experiments conducted by Finkelstein and Mudd (20) and Finkelstein et al. (5).

3 REGULATION OF SULFUR AMINO ACID METABOLISM IN HUMANS 735 Our recent data suggest that dietary Cys is capable of reducing the quantitative requirement for Met in healthy adult men. We hypothesized that providing Met and Cys in a ratio characteristic of a typical diet would result in adaptive changes in SAA metabolism, as shown by Met kinetics and TS rates, when total SAA intake was slightly more than adequate. SUBJECTS AND METHODS Subjects. Five healthy adult male volunteers participated in the study on an outpatient basis in the Clinical Investigation Unit at the Hospital for Sick Children (HSC), Toronto, Canada. Subject characteristics are described in Table 1. None of the subjects had a history of recent weight loss, unusual dietary practices or endocrine disorders, and none was using medication at the time of entry into the study. The design and aims of the study, as well as potential risks involved, were fully explained to each subject and informed written consent was obtained. All procedures were approved by the Ethics Review Board of HSC. Subjects received financial compensation for their participation. Experimental protocol. The study design was based on the Met kinetics model of Storch et al. (24). In a randomized crossover design, each subject received each of three dietary combinations of Met and Cys with a fixed and adequate total sulfur amino acid intake of 24 mg/(kg d). We chose to balance SAA intake at 24 mg/(kg d) instead of 21 mg/(kg d), which is the RDA for total SAA found in our previous study (16) because we wanted to test the effects on SAA metabolism of 13 mg/(kg d) Met and 11 mg/(kg d) Cys (Diet B). These intakes of Met and Cys represent a weight:weight ratio of 1.2 to 1 and were deemed representative of major dietary proteins. This particular combination of Met and Cys is also based on our previous findings (16,17). The dietary levels of SAA fed included: Diet A: 24 mg methionine/(kg d) and no Cys, representing the same total SAA intake as Diets B and C; Diet B: 13 mg methionine/(kg d), an intake representing both the mean requirement for Met in the absence of exogenous Cys determined using IAAO in young men (16) and also the FAO/WHO/UNU (15) population RDA and 11 mg Cys/(kg d), an intake representing the population RDA for Cys determined using IAAO in young men (17); Diet C: 5 mg methionine/(kg d), an intake that approximates that mean requirement for Met in the presence of excess exogenous Cys determined using IAAO in young men (17) and 19 mg Cys/(kg d), representing an excess of exogenous Cys. Each study consisted of 56 h of adaptation to one of the three diets described above and a protein intake of 1 g/(kg d) fed as a crystalline amino acid mixture, followed by a 5-h measurement of Met kinetics using orally administered L-[1-13 C, methyl 2 H 3 ]methionine. On the morning of each study day, an indwelling catheter was inserted in a retrograde direction into a forearm vein of the subject s nondominant arm for blood sampling. Dietary study periods were separated by at least 1 wk, with all subjects completing all three studies within 1 mo. Diet and energy intakes. Dietary intakes during the 2-d adaptation period were provided in the form of an experimental formula TABLE 1 Subject characteristics and energy intakes Subject Age Weight BMI Energy intake1 y kg kg m 2 MJ d Mean SD Energy intake was calculated from the resting metabolic rate, as determined by indirect calorimetry, multiplied by an activity factor of 1.7. developed for amino acid kinetic studies (25). Briefly, a flavored liquid formula (protein-free powder, Product 80056, Mead Johnson, Evansville, IN; Tang, Don Mills, Canada; Koolaid, Don Mills, Canada) and protein-free cookies (25) supplied the main source of energy in the diet. A crystalline amino acid mixture, based on the amino acid composition of egg protein, was fed at 1.0 g/(kg d) and provided the only source of amino nitrogen in the diet. Energy intakes were based on each subject s resting metabolic rate as determined by indirect calorimetry (2900 Energy Measurement System-Paramagnetic, Sensormedics, Yorba Linda, CA), multiplied by an activity factor of 1.7 (Table 1). The macronutrient composition of the experimental diet, expressed as a percentage of dietary energy, was 55% carbohydrate, 35% fat and 10% protein fed as a crystalline amino acid mixture. The diet contained 77 mg choline/100 g, contributing mg/d choline. The diets were prepared and weighed and were portioned into isoenergetic, isonitrogenous meals. The diet was consumed as 12 hourly meals, with each meal representing one twelfth of the subjects total daily protein and energy requirement. Subjects also consumed a daily multivitamin supplement (Centrum, Whitehall- Robins, Mississauga, Canada) containing 0.4 mg folic acid, 3 mg vitamin B-6 and 9 g vitamin B-12 for the entire duration of the three studies. Subjects had free access to water during the adaptation period. No other food or beverages, including diet products with artificial sweeteners, were consumed during the adaptation period. On each of the three study days, the diet was provided as an experimental formula developed for amino acid kinetic studies (25) as described above. Subjects had free access to water throughout the study day. Dietary Met intake was reduced by the amount of L-[1-13 C, methyl 2 H 3 ]Met infused. Tracers. Tracers used in this study were NaH 13 CO 3 (99 atom %) and L-[1-13 C, methyl 2 H 3 ]Met (99 atom %, Mass Trace, Woburn, MA). Isotope solutions were prepared in deionized water and stored at 20 C. Oral priming doses of NaH 13 CO 3 (0.176 mg/kg) and L-[1-13 C, methyl 2 H 3 ]Met (0.306 mg/kg) were given with the third hourly meal. An hourly oral dosing protocol of L-[1-13 C, methyl 2 H 3 ]Met [0.612 mg/(kg h)] was commenced simultaneously and continued throughout the remaining 5hofthestudy. Analytical measurements. Baseline samples of breath CO 2 and blood were collected at 30, 15 and 5 min before initiation of the isotope protocol. Both samples were also collected at 30-min intervals during isotopic steady state, during the period min after initiation of the isotope protocol, and stored at 20 C. Breath samples were collected in disposable Haldane-Priestley tubes (Exetainers, PDZ Europa, Cheshire, UK) using a collection mechanism that permits the removal of dead-space air (26). Samples were stored at room temperature pending analysis. Carbon dioxide production (VCO 2 ) was measured during each study day for 30 min with a variable flow indirect calorimeter (2900 Energy Measurement System-Paramagnetic, Sensormedics). Blood samples taken from the indwelling catheter were collected in heparinized tubes and were centrifuged immediately at 1500 g for 10 min at 4 C; plasma was stored at 20 C until analyzed. The enrichment of 13 C in breath CO 2 was measured on a continuous flow isotope ratio mass spectrometer (PDZ Europa). Breath 13 CO 2 enrichments were expressed as atom % over a reference standard of compressed CO 2 gas. Amino acids were separated by cation exchange (Dowex 50W-X8, mesh H form, Bio Rad, Richmond, CA). Plasma Met was derivatized to its isopropyl, N- heptofluorobutyramide derivatives (27). Isotopic enrichment was measured by negative chemical ionization gas chromatography (GC)/ mass spectrometry (MS) (Hewlett-Packard 5890 series; GC; Hewlett- Packard 5988A MS system, Mississauga, Canada). Selected-ion chromatograms were obtained by monitoring [m HF ] ions at m/z 367, 368 and 371 for methionine, [1-13 C]Met and [1-13 C, methyl- 2 H 3 ]methionine, respectively. The areas under the peaks were integrated using Hewlett-Packard G1034C MS-CHEM software. Tracerto-tracee ratios (TTR, mol/100 mol) were determined by inserting raw ion abundances into a matrix containing data for the relative abundances of the three mass istopomers using the method of Brauman (28). Net TTR of a sample was calculated by subtracting the background TTR obtained from the sample collected before isotope administration.

4 736 DI BUONO ET AL. Total plasma homocysteine (thcy) and Met were measured using an HPLC method as previously described (29,30). In brief, thcy was obtained by converting all of the homocysteine to its free thiol form by reduction with tris(2-carboxyethyl) phosphine (TCEP). TCEPreduced serum samples were assayed using the Dionex DX-500 Ion Chromatograph outfitted with two pumps in parallel, valves and 2 columns (OmniPac PCX-500, 4 50 mm precolumn and PCX-500, mm analytical column) plumbed in serial to permit heartcut trapping of thcy and methionine. Detection was achieved using the ED40 electrochemical detector and pulsed integrated amperometry set to detect reduced sulfur. Detector output was analyzed with Peaknet (Dionex) software generating chromatograms, with homocysteine exiting the column and peaking at 8.1 min, and Met at 11.2 min. In addition, on the basis of recent findings by MacCoss et al. (31) concerning the measurement of intracellular sulfur amino acid enrichment, plasma homocysteine and cystathionine enrichments were analyzed post-hoc but none was detected. For plasma homocysteine and cystathionine enrichment, samples were treated with dithiothreitol and 4-vinylpyridine under vacuum according to MacCoss et al. (31) to reduce and alkylate homocysteine. After isolation by cation exchange (DOWEX 50W-X8, mesh H, BioRad), homocysteine and cystathionine were derivatized to their isopropyl, N- heptafluorobutyramide derivatives (27). Isotopic enrichment was measured by negative chemical ionization GC/MS (Hewlett-Packard 5890 series; GC). After split injection on a DB5 capillary column (J&W Scientific, Folsom, CA), selected-ion chromatograms were obtained by monitoring [m HF ] ions at m/z 469 and 470 for homocysteine and [1-13 C]homocysteine. For cystathionine analysis, m/z 678 and 679 were monitored for cystathionine and [1-13 C]cystathionine, respectively, after splitless injection. Experimental model. Steady-state isotopic plateaus for plasma [1-13 C, methyl 2 H 3 ] methionine, plasma [1-13 C]Met and breath 13 CO 2 were achieved 2.5 h after the start of isotope administration. Met kinetics (Fig. 1) were calculated according to the model of Storch et al. (24). Flux (Q) rates of Met carboxyl (Q c ) measured using the [ 13 C]-labeled carboxyl carbon of Met and methyl (Q m ) measured with the [ 2 H 3 ]-labeled methyl carbon of Met were calculated as described by Storch et al. (24). Briefly, Q m and Q c were calculated as follows: Q m I tr [(E tr /E 4 ) 1] (1) Q c I tr {[E tr /(E 1 E 4 )] 1} (2) where I tr and E tr are the infusion rate and the enrichment of the tracer [1-13 C, methyl 2 H 3 ]methionine, respectively, and E 1 and E 4 are the plasma enrichments of [1-13 C]Met and [1-13 C, methyl 2 H 3 ]methionine, respectively, from plasma samples obtained at isotopic steady state during the last 180 min of each study day. Q m represents Met methyl group flux via transmethylation, whereas Q c represents Met flux through the entire Met cycle. The equations below relate Q m and Q c rates to their individual components. The calculations and assumptions have been discussed previously (9,24): Q m I B RM S TM (3) Q c I B S TS (4) where I is dietary intake, B is plasma Met appearance via tissue protein breakdown, RM is Met appearance from remethylation of homocysteine, S is Met plasma disappearance via protein synthesis, TM is transmethylation (rate of demethylation of Met to homocysteine) and TS is transsulfuration, which is assumed to be equivalent to Met oxidation. Equations 3 and 4 can be used to derive the following two equations: RM Q m Q c (5) TM RM TS (6) Only Met sulfur appears in Cys after TS, whereas the remaining structure forms -ketobutyrate, which is ultimately oxidized to CO 2. As such, TS can be calculated from the oxidation of -ketobutyrate to CO 2. The TS rate was thus calculated as follows: TS F 13 CO 2 [(1/[ 13 C]Met pool enrichment)] [(1/[ 13 C]Met tracer enrichment)] (7) where F 13 CO 2 is the rate of 13 C output in expired air and is calculated as: F 13 CO 2 (VCO 2 APE )/(weight ) (8) The constants are 0.82 to account for the 13 CO 2 retained in the body, in the fed state, due to bicarbonate fixation (32), 44.6 to convert gas volume to moles, 60 to show time per hour and 100 to convert atom % excess (APE) to a fraction. Statistical analysis. Two-way repeated-measures ANOVA was performed on primary and derived variables to assess the effects of Met and Cys intake (SAS version 6.6, SAS Institute, Cary, NC), considering order in which the test intakes were administered, subject and interactions. Subject was significant but order and interactions were not. The final model contained only subject and intake. The significance of differences between intakes was examined using Tukey s post-hoc test. Results are expressed as least squares means SD. In all cases, differences were considered to be statistically significant at P RESULTS There were no significant changes in plasma homocysteine with the different diets (Table 2). Plasma Met was higher after consumption of Diets A and B than after Diet C (P 0.05), because relative to the minimum Met requirement of 5 mg/ (kg d), when subjects consumed Diets A and B, they were ingesting an excess of Met, whereas when they consumed Diet C, they were not. Plasma enrichment of [1-13 C, methyl- 2 H 3 ]Met and [1-13 C]Met increased significantly with a reduction in Met intake and an increase in Cys intake (Table 2). Breath 13 CO 2 enrichment decreased significantly when subjects consumed Diet C, which had the lowest Met to Cys ratio. Plasma Q m and Q c (Table 2) decreased significantly with decreasing dietary Met and increasing dietary Cys. The effects of reducing the Met content while simultaneously increasing the Cys content of the diet on Met oxidation (TS) are also summarized in Table 2. Met oxidation decreased significantly when diets containing Cys were given in comparison with the diet devoid of Cys with a constant SAA intake of 24 mg/(kg d). The carboxyl flux data are also compelling in showing a biochemical basis for the reduction in Met catabolism (oxidation) caused by Cys ingestion. The Met TM and homocysteine RM rates were calculated from the relative level of plateau enrichments of the Met tracer [1-13 C, methyl- 2 H 3 ]Met and the [1-13 C]Met metabolites in the plasma. Although the TM rates decreased significantly with a decrease in the ratio of dietary Met to Cys (Table 2), the homocysteine RM rates were significantly lower when Diet C rather than Diet A was consumed. No difference was observed for RM between Diets A and B, or Diets B and C. The relationships among Met flux rates, RM, TM and TS rates are summarized in Table 3. TS:TM, which is an index of the rate at which newly formed homocysteine molecules are diverted toward Cys biosynthesis, decreased significantly (P 0.05) with a decreasing ratio of Met to Cys. Conversely, RM:TM, which is an index of the rate at which newly formed homocysteine is remethylated to methionine, increased significantly (P 0.05) when the dietary Met:Cys ratio was lowest. As a result of Cys sparing at the highest dietary Cys intake, the fraction of homocysteine destined for RM relative to TS is

5 REGULATION OF SULFUR AMINO ACID METABOLISM IN HUMANS 737 TABLE 2 Estimated means of various components of the methionine (MET) cycle in young men who consumed diets differing in methionine and cysteine concentrations during the fed state based on infusion of L-[1-13C, methyl 2H 3 ]methionine1 Diet A (24 mg MET) Diet B (13 mg MET, 11 mg CYS) Diet C (5 mg MET, 19 mg CYS) Plasma MET, mol/l a,b Plasma HCY, mol/l Plasma MET enrichment, % [1-13C-methyl-2H 3 ]methionine a a,b [1-13C]methionine a a Breath 13CO 2 enrichment (APE) a,b MET4 flux Q m, mol kg 1 h a a,b Q c, mol kg 1 h a a,b Transmethylation, mol kg 1 h a a,b Transsulfuration, mol kg 1 h a a Remethylation, mol kg 1 h a Values are mean SD, n 5. a Different from Diet A, P b Different from Diet B, P MET, methionine; HCY, homocysteine; Q m, methyl flux; Q c, carboxyl flux. significantly increased (P 0.05) compared with the diet devoid of Cys, as can be seen by RM:TS. DISCUSSION The normal metabolic fates of Met include its incorporation into newly synthesized proteins and the biosynthesis of Cys by TS (Fig. 1) (32,33). These two fates also represent the major outlets in the Met cycle; polyamine flux represents only 1 2% of methyl flux per day (5). In animals, transamination has been identified as an initial step in Met catabolism and an alternate pathway for Met oxidation that does not involve its activation to S-adenosyl-Met (34 36). However, evidence suggests that the transaminative pathway is minor in humans (37 40). In contrast to methionine, Cys is synthesized by mammals and is therefore considered to be fully dispensable (37). Cys sulfur is derived from Met, whereas the carbon and nitrogen are derived from serine. This dependence on Met sulfur implies that Cys is a dietary dispensable amino acid only if Met intake is sufficient to meet the total SAA requirement (37). Conversely, the dietary requirement for Met can be reduced, but not eliminated, when Cys intake is adequate (1,6 8,37). However, the ability of Cys to reduce, or spare the Met requirement in adult humans has recently come into question (9 14). We estimated the RDA for SAA intake in adult men as 21 TABLE 3 Relations among methionine metabolic indexes in young men given diets with different methionine content1 Ratio Diet A (24 mg MET) Diet B (13 mg MET, 11 mg CYS) Diet C (5 mg MET, 19 mg CYS) TS:TM a RM:TM a RM:TS a 1 Values are mean SD, n 5. a Different from Diet A, P TM, rate of conversion of methionine to homocysteine; RM, methionine appearance from remethylation of homocysteine; TS, transsulfuration rate. mg/(kg d) (16), and subsequently identified a sparing effect of Cys on the Met requirement (17). This report regarding the sparing capacity of Cys toward the Met requirement estimated a Cys sparing effect of 64% toward the mean Met requirement (17). In the present study, the replacement of 11 mg/(kg d) of dietary Met with Cys (Diet B), an amount previously identified as being equivalent to a population RDA for Cys (17), to a diet containing only Met (Diet A), resulted in a 64% reduction in TS. Compared to the Cys-devoid diet (Diet A), TS was reduced by 81% upon further replacement of dietary Met by Cys (Diet C). This latter reduction in TS is commensurate with early observations, based on nitrogen balance, of a Cys-sparing effect of 89% in adult humans (6). It is important to note that the ratio of dietary Met to Cys (by weight) at which our observed sparing effect occurs is 1.2:1, which is consistent with the ratio of these SAA in most of the major food proteins. Our results support and amplify those published in recent kinetic studies using doubly labeled Met (9 14). For example, in the study by Storch et al. (9), a dietary intake of 2.4% methionine, with no Cys present in the diet resulted in a TS of 7.6%, which compares with our TS of 7.8% when Met was fed at 2%. Recent tracer studies (10 14) showed that when Met intake is reduced, without a change in Cys intake, TS decreases by 30 40%. A difference between these studies and this one is that the total SAA intake was below the RDA defined in our recent IAAO studies (16,17). In the present study, we used total SAA intakes approximating this newly defined RDA. Due to the design differences between previous studies (9 14) and this study, it is difficult to separate the effects of replacing Met with Cys; however, on the basis of animal work (5,20,41) and our earlier IAAO study (17), we tentatively conclude that the reduction in TS reported here is due to a combination of reduced Met intake and increased Cys intake. In their in vitro study of TS in rats, Finkelstein et al. (41) concluded that the dietary content of Met was an important determinant of the effect of supplemental cystine on hepatic Met metabolism. Furthermore, Finkelstein et al. (5) showed that changes in the tissue content of CBS explained less than half of the change in flow through the TS pathway consequent to their dietary alterations, with changes in the concentrations of substrates and effectors being of equal or greater importance (5). Our whole-body in vivo findings of the

6 738 DI BUONO ET AL. effects of adding Cys to the diet are consistent with earlier animal studies of Finkelstein et al. (5,41). In addition to the findings that demonstrate a sparing effect of Cys on the Met requirement, our results also address the inefficiency of whole-body SAA metabolism in conserving methionine. TS was reduced by 81% when subjects consumed Diet C (compared with Diet A) (Table 2), suggesting that excess dietary Cys is capable of reducing but not arresting TS when dietary Met intake is set a level of 5 mg/(kg d) and may be limiting whole-body protein synthesis in some subjects (17). Another indication of this inefficiency is that TM rates, although reduced, were not arrested when dietary Met intakes might have been limiting for protein synthesis, as in Diet C. Overall, these observations demonstrate that the use of Met in methyl donor reactions via SAM may take priority over the use of Met for protein synthesis. The emerging links between SAA and diseases of relevance to public health indicate that knowledge of the regulation of SAA metabolism is of growing importance. In the present study, Met was replaced by two levels of Cys, both of which reduced flux via the TS pathway (Table 2). Unless compensatory mechanisms are active, this could result in an expansion of the homocysteine pool, which it did not (Table 2). When Diet B (a moderate replacement of Met by Cys) was consumed, TM of Met to form homocysteine was reduced (Table 2) and fractional RM of homocysteine to Met was maintained and even mildly increased (Table 3). When Diet C (in which most of the Met was replaced by Cys) was consumed, TS (Table 2) and the fractional conversion of homocysteine to Cys (TS:TM) (Table 3) were maximally inhibited, whereas the ratio of RM:TS was maximized (Table 3). Although SAM increases CBS activity and decreases methylene tetrahydrofolate reductase (MTHFR) activity, and increased Met intake will increase the concentration of SAM in vivo, the results presented here cannot be explained by the effects of elevated Met intake resulting in elevated SAM alone. The increased methionine intakes when Diets A and B were consumed did not change homocysteine concentrations; hence, although SAM concentrations may have been increasing as a result of higher methionine intakes, SAM did not exert its full potential on either CBS or MTHFR. This can be explained by the documented existence, in humans, of the glycine-sarcosine overflow pathway, which represents a means of catabolizing excess methionine (and SAM) via the enzyme glycine N- methyltransferase (GNMT) (42). GNMT is inhibited by 5-methyltetrahydrofolate pentaglutamate (43). Thus, when SAM is elevated and MTHFR activity is reduced, 5,10 methylene tetrahydrofolate is not converted to 5-methyltetrahydrofolate and this latter compound will no longer inhibit GNMT (43). By this mechanism, excess SAM can be metabolized via the GNMT pathway, thereby removing the effects of SAM on the CBS and MTHFR pathways. The tissue distribution of GNMT parallels that of the TS enzymes (43). Although plasma homocysteine concentrations did not change significantly with a reduction in the methionine:cys ratio of the diet, TM accounted for significantly less of the Met methyl flux (TM:Q m ) as the dietary methionine:cys ratio decreased. This can be explained by the changes observed in plasma Met concentrations, which were significantly reduced in those consuming the diet containing the lowest methionine:cys ratio; therefore, the observed decrease in TM may be explained by the level of Met in Diet C. The Met content of Diet C was similar to our estimate of the minimum mean Met requirement (17) and thus may have been low enough to affect conservation of Met for protein synthesis in some subjects for whom this intake of Met was insufficient to meet their dietary requirement. In fact, Storch et al. (9) showed that when dietary Met intake was inadequate, conservation of endogenous Met was achieved primarily by a shift of Met metabolism away from the transmethylation pathway and toward that of protein synthesis. The study reported in this paper was designed using the model of Storch et al. (24). In light of the recent report by MacCoss et al. (31), we also carried out measurements of isotope enrichment in plasma homocysteine and cystathionine and were unable to detect enrichment, having taken into account all the precautions outlined by the authors. Nonetheless, we were able to achieve isotopic steady state in breath 13 CO 2 enrichment, which reflects oxidation of Met to -ketobutyrate and CO 2 via the TS pathway. It is therefore important to highlight the differences between the two studies. First, our studies were conducted in the fed state and used oral isotope administration, whereas the studies by MacCoss et al. (31) were conducted in the fasted state and used intravenous isotope administration. Next, our study period was 5 h whereas that of MacCoss et al. (31) was up to 8 h, and the plasma enrichments of [1-13 C]homocysteine and [ 13 C]cystathionine rose exponentially to a plateau by 6 h (31). A final point concerns the plateau value of plasma L-[1-13 C]Met, which was mol % excess (MPE) (31), whereas the comparable value in our subjects (Table 2) ranged from a mean of 1.6 when Diet A was consumed to 3.4 for Diet C. Our conclusion is that under our experimental conditions, plasma enrichments of [1-13 C]homocysteine and [ 13 C]cystathionine did not reach a level of enrichment, detectable above a background by GC/ MS, of MPE. Although the corrections proposed by MacCoss et al. (31) would increase TM, TS and RM by 40%, they should not alter the ratios among metabolic indexes at a given intake of Met and Cys. However, it has yet to be determined whether the relationship between plasma Met and homocysteine/cystathionine enrichments remains unchanged at different intakes of Met and Cys. Hence, we cannot say with complete certainty that the corrections proposed by MacCoss et al. (31) would not change our conclusions. We conclude that the ratio of dietary Cys to Met regulates whole-body SAA metabolism in adult humans when total SAA intake is adequate and held constant at 24 mg/(kg d). At high Met intakes, the Met pool is regulated by high rates of TS. Replacement of dietary Met with Cys results in increased RM at the expense of TS. Because it is not possible from this study alone to assess the specific effect of Cys on TS, the results presented do not offer definitive evidence of a Cyssparing effect. However, the data are consistent with previous findings (16,17) and indicative of a Cys-sparing effect in humans. Therefore, we postulate that the mechanism by which dietary Cys spares the requirement for Met is through an increase in the fraction of homocysteine destined for RM relative to TS. Thus, the regulation of SAA metabolism at the point of homocysteine can be studied with a Met tracer alone and is relevant for chronic diseases that have been linked to SAA metabolism, such as cancer, cardiovascular disease, dementia and renal disease. ACKNOWLEDGMENTS We acknowledge the technical expertise of Deborah Martin. We thank the subjects who participated in the study and Linda Chow in the Department of Nutrition and Food Services (HSC) for preparing the protein-free cookies.

7 REGULATION OF SULFUR AMINO ACID METABOLISM IN HUMANS 739 LITERATURE CITED 1. Womack, M. & Rose, W. C. (1942) The partial replacement of dietary methionine by cystine for purposes of growth. J. Biol. Chem. 141: Graber, G. & Baker, G. H. (1971) Sulfur amino acid nutrition of the growing chick: quantitative aspects concerning the efficacy of dietary methionine, cysteine and cystine. J. Anim. Sci. 33: Baker, D. H., Clausing, W. W., Harman, B. G., Jensen, A. H. & Becker, D. E. (1969) Replacement value of cysteine for methionine for young pigs. J. Anim. Sci. 29: Aguilar, T. S. (1982) Studies in vivo on the methionine-sparing effect of Cys in rats. Arch. Latinoam. Nutr. 32: Finkelstein, J. D., Martin, J. J. & Harris, B. J. (1988) Methionine metabolism in mammals. The methionine-sparing effect of cystine. J. Biol. Chem. 263: Rose, W. C. & Wixom, R. L. (1955) The amino acid requirements of man. XII. The sparing effect of cystine on the methionine requirement. J. Biol. Chem. 215: Reynolds, M. S., Steel, D. L., Jones, E. M. & Baumann, C. A. (1958) Nitrogen balance of women maintained on various levels of methionine and cysteine. J. Nutr. 64: Clark, H. E., Howe, J. M., Shannon, B. M., Carlson, K. & Kolski, S. M. (1970) Requirements of adult human subjects for methionine and cystine. Am. J. Clin. Nutr. 23: Storch, K. J., Wagner, D. A., Burke, J. F. & Young, V. R. (1990) [1-13 C; methyl- 2 H 3 ]methionine kinetics in humans: Methionine conservation and cystine sparing. Am. J. Physiol. 258: E790 E Hiramatsu, T., Fukagawa, N. K., Marchini, J. S., Cortiella, J., Yu, Y. M., Chapman, T. E. & Young, V. R. (1994) Methionine and cysteine kinetics at different intakes of cystine in healthy adult men. Am. J. Clin. Nutr. 60: Raguso, C. A., Ajami, A. M., Gleason, R., & Young, V. R. (1997) Effect of cysteine intake on methionine kinetics and oxidation determined with oral tracers of methionine and cysteine in healthy adults. Am. J. Clin. Nutr. 66: Fukagawa, N. K., Yu, Y. M. & Young, V. R. (1998) Methionine and cysteine kinetics at different intakes of methionine and cysteine in elderly men and women. Am. J. Clin. Nutr. 68: Raguso, C. A., Pereira, P. & Young, V. R. (1999) A tracer investigation of obligatory oxidative amino acid losses in healthy, young adults. Am. J. Clin. Nutr. 70: Raguso, C. A., Regan, M. M. & Young, V. R. (2000) Cysteine kinetics and oxidation at different intakes of methionine and cystine in young adults. Am. J. Clin. Nutr. 71: ,. 15. FAO/WHO/UNU (1985) Energy and Protein Requirements. World Health Organization, Geneva, Switzerland. 16. Di Buono, M., Wykes, L. J., Ball, R. O. & Pencharz, P. B. (2001) Total sulfur amino acid (SAA) requirement in young men by indicator amino acid oxidation (IAAO). Am. J. Clin. Nutr. 74: Di Buono, M., Wykes, L. J., Ball, R. O. & Pencharz, P. B. (2001) Dietary cysteine (CYS) reduces the methionine (MET) requirement in adult males. Am. J. Clin. Nutr. 74: Young, V. R. (2001) Got some amino acids to spare? Am. J. Clin. Nutr. 74: Finkelstein, J. D. (2000) Pathways and regulation of homocysteine metabolism in mammals. Semin. Thromb. Hemost. 26: Finkelstein, J. D. & Mudd, S. H. (1967) Trans-sulfuration in mammals. The methionine-sparing effect of cystine. J. Biol. Chem. 242: Ge, Y., Konrad, M. A., Matherly, L. H. & Taub, J. W. (2001) Transcriptional regulation of the human cystathionine beta-synthase-1b basal promoter: synergistic transactivation by transcription factors NF-Y and Sp1/Sp3. Biochem. J. 357: Kraus, J. P., Le, K., Swaroop, M., Ohura, T., Tahara, T., Rosenberg, L. E., Roper, M. D. &. Kozich, V. (1993) Human cystathionine beta-synthase cdna: sequence, alternative splicing and expression in cultured cells. Hum. Mol. Genet. 2: Roper, M. D. & Krasu, J. P. (1992) Rat cystathionine beta-synthase: expression of four alternatively spliced isoforms in transfected cultured cells. Arch. Biochem. Biophys. 298: Storch, K. J., Wagner, D. A., Burke, J. F. & Young, V. R. (1988) Quantitative study in vivo of methionine cycle in humans using [methyl- 2 H 3 ]- and [1-13 C]methionine. Am. J. Physiol. 255: E322 E Zello, G. A., Pencharz, P. B. & Ball, R. O. (1990) The design and validation of a diet for studies of amino acid metabolism in adult humans. Nutr. Res. 10: el-khoury, A. E., Fukagawa, N. K., Sanchez, M., Tsay, R. H., Gleason, R. E., Chapman, T. E. & Young, V. R. (1994) Validation of the tracer-balance concept with reference to leucine: 24-h intravenous tracer studies with L-[1-13 C]leucine and [ 15 N- 15 N]urea. Am. J. Clin. Nutr. 59: Patterson, B. W., Hatchey, D. L., Cook, G. L., Amann, J. M. & Klein, P. D. (1991) Incorporation of a stable isotopically labeled amino acid into multiple human apolipoproteins. J. Lipid. Res. 32: Brauman, J. I. (1966) Least squares analysis and simplification of multi-isotope mass spectra. Anal. Chem. 38: Evrovski, J., Callaghan, M. & Cole, D. E. (1995) Determination of homocysteine by HPLC with pulsed integrated amperometry. Clin. Chem. 41: Cole, D. E., Lehotay, D. C. & Evrovski, J. (1998) Simplified simultaneous assay of total plasma homocysteine and methionine by HPLC and pulsed integrated amperometry. Clin. Chem. 44: MacCoss, M. J., Fukagawa, N. K. & Matthews, D. E. (2001) Measurement of intracellular sulfur amino acid metabolism in humans. Am. J. Physiol. 280: E947 E Hoerr, R. A., Yu, Y-M., Wagner, D, Burke, J. F. & Young, V. R. (1989) Recovery of 13 C in breath from NaH 13 CO 3 infused by gut and vein: effect of feeding. Am. J. Physiol. 257: E426 E Finkelstein, J. D. (1974) Methionine metabolism in mammals: the biochemical basis for homocystinuria. Metabolism 23: Mudd, S. H., Ebert, M. H. & Scriver, C. R. (1980) Labile methyl group balances in the human: the role of sarcosine. Metabolism 29: Blom, H. J., Boers, G. H. J., van den Elzen, J. P. A. M., Gahl, W. A. & Tangerman, A. (1989) Transamination of methionine in humans. Clin. Sci. (Lond.) 76: Mitchell, A. D. & Benevenga, N. J. (1978) The role of transamination in methionine oxidation in the rat. J. Nutr. 108: Griffith, O. W. (1987) Mammalian sulfur amino acid metabolism: an overview. Methods Enzymol. 143: Gahl, W. A., Bernardini, I., Finkelstein, J. D., Tangerman, A., Martin, J. J., Blom, H. J., Mullen, K. D. & Mudd, S. H. (1988) Transsulfuration in an adult with hepatic methionine adenosyltransferase deficiency. J. Clin. Investig. 81: Blom, H. J., Boers, G. H., Trijbels, J. M., van Roessel, J. J. & Tangerman. A. (1989) Cystathionine-synthase-deficient patients do not use the transamination pathway of methionine to reduce hypermethioninemia and homocystinemia. Metabolism 38: Bugianesi, E., Tangerman, A., Ronchi, M., Bianchi, G. & Marchesini, G. (1996) Transamination of methionine after loading in patients with cirrhosis. J. Hepatol. 24: Finkelstein, J. D., Martin, J. J. & Harris, B. J. (1986) Effect of dietary cysteine on methionine metabolism in rat liver. J. Nutr. 116: House, J. D., Jacobs, R. L., Stead, L. M., Brosnan, M. E. & Brosnan, J. T. (1999) Regulation of homocysteine metabolism. Adv. Enzyme Regul. 39: Mudd, S. H., Cerone, R., Schiaffino, M. C., Fantasia, A. R., Minnitti, G., Caruso, U., Lorini, R., Watkins, D., Matiaszuk, N, Rosenblatt, D. S., Schwann, B., Rozen, R., LeGros, L., Kotb, M., Capdevila, A., Luka, Z., Finkelstein, J. D., Tangerman, A., Stabler, S. P., Allen, R. H. & Wagner, C. (2001) Glycine N-methyltransferase deficiency: a novel inborn error causing persistent isolated hypermethioninemia. J. Inherit. Metab. Dis. 24: