Sulfur amino acid metabolism in infants on parenteral nutrition

Clinical Nutrition (1995) 14:381-387

© Pearson Professional Ltd 1995

Sulfur amino acid metabolism in infants on parenteral nutrition R. A. HELMS,* R. W. CHESNEY* and M. C. STORM t From the Center for Pediatric Pharmacokinetics and Therapeutics and the Pediatric Pharmacology Research Unit at The University of Tennessee, The Departments of Clinical Pharmacy and Pediatrics at The University of Tennessee, Memphis, LeBonheur Children's Medical Center, Memphis, TNt, and McGaw, Inc., Irvine, CA, ft. (Correspondence to: RAH, Pharm.D. Professor and Vice Chairman Department of Clinical Pharmacy and Pediatrics 26 South Dunlap, Suite 506 Memphis, TN 38163) ABSTRACT--In the infant on parenteral nutrition, cysteine supplementation has been suggested

due to low levels of hepatic cystathionase activity limiting synthesis from methionine. We have examined the plasma concentrations of sulfur amino acids in four groups of postsurgical infants requiring parenteral nutrition receiving (A) a low methionine + cysteine + taurine formula, (B) a high methionine formula (non-steady state), (C) a high methionine formula (steady state), and (D) a high methionine + cysteine formula. Plasma methionine concentrations were above the normal reference range (2.2-4.9 micromole/dL) of normal breast-fed infants in Groups B (15.9 _+ 10.7 micromole/dL) and D (5.7 + 1.9 micromole/dL) and at the upper limit for Group C (4.9 + 1.7 micromole/dL). Total cysteine/cystine concentrations (normal reference range, 10.2-20.4 micromole/dL) were highest in Groups A (18.9 + 3.5 micromole/dL) and D (16.8 + 5.3 micromole/dL) that received cysteine HCI supplementation, and lowest in Group B (8.6 + 3.7 micromole/dL) that received no cysteine in non-steady state. All plasma free cystine concentrations were below the normal reference range (3.6-6.8 micromole/dL). Plasma taurine concentrations were not significantly different among the four groups and all were within the normal reference range (0.6-16.2 micromole/dL). The strikingly elevated methionine and low total cysteine/cystine values in Group B suggested the existence of a feedback loop of methionine conversion below the level of homocysteine. Equilibrium of methionine and cysteine/cystine plasma concentrations did occur, in time. Parenteral cysteine administration resulted in a greater proportion of plasma free cysteine concentration, but not cystine. The proportion of free to bound cysteine/cystine, as well as the proportion of free cystine to cysteine, was not normal during parenteral nutrition with or without cysteine HCI supplementation. Little benefit in plasma concentrations was derived from cysteine HCI supplementation to a high methionine formulation.

tional benefit of cysteine supplementation in a small group of newborns on TPN (5). These investigators showed that nitrogen retention, weight gain, and growth in length and head circumference were not influenced by cysteine supplementation at 77 mg/kg/d (0.64 mmole/kg/d). They documented an increase in plasma cystine concentration but noted that neither plasma methionine nor taurine concentrations were affected. Similar results have been reported by Malloy et al suggesting that cysteine supplementation to TPN may not be necessm-y (6). However, these earlier studies were completed with a parenteral amino acid preparation which was not designed for pediatric use and was possibly deficient in tyrosine compared to

Introduction

Sturman, Gaull and Raiha have documented a deficiency of cystathionase enzyme levels in the premature infant liver (1, 2). This deficiency renders cysteine and taurine as perhaps 'conditionally essential' in this population. Cysteine is often supplemented in the total parenteral nutrition (TPN) solution of this patient population as a means to enhance calcium and phosphate solubility and hence to increase the intake of these minerals (3). A beneficial effect on nutritional status with cysteine supplementation has been demonstrated in orally fed infants (4), but not in infants on TPN. Zlotkin et al were unable to document a nutri381

382 SULFURAMINOACID METABOLISMIN INFANTSON PARENTERAL NUTRITION

infant requirements. Since optimal protein synthesis requires adequate availability of all amino acids, pediatric nutritional studies using an amino acid mixture deficient in another essential amino acid are not likely to demonstrate a nutritional benefit of cysteine supplementation (7). Clinical investigation, completed in preterm neonates on parenteral nutrition, has suggested that ideal sulfur amino acid nutrition is best addressed with the addition of L-cysteine HC1 to a pediatric amino acid formula (8). It is assumed, in preterm neonates, that little transsulfuration activity is ongoing. When enzyme activity increases to the extent that cysteine supplementation may no longer be beneficial has not been clearly defined. We wished to compare the plasma sulfur amino acid responses in infants receiving a low methionine pediatric amino acid formula containing tanrine and admixed cysteine to a high methionine standard amino acid formula with and without cysteine supplementation. In addition, we observed sulfur amino acid levels in infants shortly after being switched from a low methionine pediatric formula to the high methiohine formula (non-steady state). These data are interpreted in terms of the known metabolic pathway for the conversion of methionine to cysteine and taurine. Additional inhibitory feedback loop(s) in this metabolic pathway are proposed as a possible explanation for these observations.

Methods

Forty-eight plasma amino acid evaluations were completed in 35 surgical infants, weight 2.56 + 1.00 kg and 67 + 57 postnatal days, receiving as part of parenteral nutrition, TrophAmine (TA) plus cysteine HC1 [Group A, n = 10 (20 steady state observations)], FreAmine III (FA-III) 6-8 hours after TA plus cysteine HC1 [Group B; n =10 (10 non-steady state observations)], FA-III [Group C, n = 9 (10 steady state observations)] and FA-III plus cysteine HC1 [Group D, n = 6 (8 steady state observations)]. The amino acid composition of TrophAmine and FreAmine III can be found in Table 1. Diagnoses included necrotizing enterocolitis (n = 28), omphalocele! gastroschisis (n -- 4), and gut atresia (n = 3). No differences were found in the proportion of infants with a given diagnosis among groups. All infants had parenteral intakes of > 60 kcal/kg/day and > 2 g/kg/ day protein at the time of amino acid analysis, and all were > 14 days post-surgery. No infant had fever or other constitutional signs of infection at the time of the study, and none had major organ dysfunction. Because these subjects were studied at nearly two

Table 1 III

Amino acid composition of TrophAmine and FreAmine

Essential Amino Acids

rag/100 ml TrophAmine 6%*

FreAmine III 8.5%

Isoleucine Leucine Lysine Methionine Phenylalanine Threonine Tryptophan Valine Cysteine HC1.H20 Histidine Tyrosine N-Acetyl-L-Tyrosine

490 840 490 200 290 250 120 470 < 20 290 44 120

590 770 620 450 480 340 130 560 < 20 240 0 0

Nonessential Amino Acids Alanine Arginine Proline Serine Glycine Aspartic Acid Glutamic Acid Taurine

320 730 410 230 220 190 300 15

600 810 950 500 1190 0 0 0

* Marketed in Europe as ThomaeAminPAD.

months of age on average, some mild elevations (> 1 < 2.5 mg/dL) in direct bilirubin were found in seven infants. Two infants were studied as part of more than one group (one in groups A, C, D, and one in groups A, C). Subjects were enrolled in IRB reviewed (The University of Tennessee, Memphis) and approved protocols evaluating other aspects of nutrition therapy requiring amino acid monitoring. Informed consent was obtained from all study subjects' parents or legal guardians. Samples for amino acid analysis were obtained at approximately the same time each morning (08:00). Steady-state samples were obtained after 48 h of continuous infusion with no dosage change. Non-steady state was defined as observations made 6-8 h after a formulation change. Whole plasma was used for the determination of total cysteine/cystine. Deproteinized plasma was prepared immediately for amino acid and free cysteine/cystine analyses by the addition of 5'sulfosalicylic acid (40 mg/mL), vortex mixing, and refrigerated centrifugation. Samples were stored at 70°C until they were analyzed. Plasma methionine, cystine, cystathionine, and taurine as well as other amino acids were determined on a Beckman 6300 automated amino acid analyzer using a lithium citrate buffer system and post-column derivitization with ninhydrin. Free cysteine-cystine was determined on deproteinated plasma while total cysteine-cystine was

CLINICALNUTRITION 383 determined on whole plasma. Both were assayed spectrophotometrically after reduction with dithiothreitol by a modification (9) of the method of Gaitonde (10). Previous work has demonstrated that cysteine, its dimer, cystine, and its adduct, D-glucoseL-cysteine, are measured by this assay methodology (8). No distinction can be made between the adduct and cysteine/cystine. The normal plasma amino acid reference range (95% confidence limits) was established in term infants, breast fed, at 1 month of age (11). These infants were rapidly growing and healthy. Plasma amino acid concentrations were completed 2 h postprandial. The human milk fed, term infant amino acid pattern has been validated by Pohlberger and co-workers (12) as an appropriate plasma concentration range to use as a goal to maximize growth in preterm infants. Statistical analysis of plasma amino acids and demographic parameters was completed by analysis of variance with a Duncan Multiple Range Test. Homogeneity of variances was evaluated by a Bartlett's Test. Chi Square analysis was utilized for assessing differences in sex distribution. Significance was assumed at p < 0.05. All values are reported as mean + S.D.

Results

The sex distribution, weight and postnatal age upon study entrance, and mean amino acid, cysteine, and non-protein energy intakes for the study subjects are summarized in Table 2. All subjects received nutrient infusions at similar daily volume per kg body weight. All subjects received similar mineral, vitamin, and

trace element intakes. Enteral intake, as Pedialyte or Pregestimil (Mead Johnson, Evansville, IN) occurred in seven of 35 infants, but represented small quantities of protein (on average, 0.2 g/kg/day) and energy (7 kcal/kg/day). No differences among groups in enteral intake, or proportion of infants receiving enteral intake were found. The plasma concentrations for the sulfur containing amino acids are summarized in Table 3. Plasma methionine was at the upper limit of normal or above tile normal reference range (11) in all groups receiving FA-III. Group B had the highest plasma methionine concentration (15.9 + 10.7 micromole/100 mL) of any of the groups. This non-steady state level was significantly elevated when compared to the steady state levels of the other two groups receiving FA-III (Group C; FA-III only, 4.9 + 1.7 and GrouP D; FA-III with cysteine HC1, 5.7 + 1.9 micromole/100 mL). Total cysteine/cystine was highest in the two groups (A and D) that received cysteine HC1 supplementation. However, all plasma free cystine concentrations were below the normal reference range. Total cysteine/cystine was significantly lower and below the normal reference range in Group B when compared to other treatment groups. Plasma taurine levels were not significantly different among the four groups and all were within the normal reference range. Percent total cysteine/cystine as free (unbound) was marginally above the normal value of ~50% in all treatment groups (see Table 3). However, the percent free cysteine/cystine as cystine was distinctly different from normal values (23.4 to 39.6% in the treatment groups versus ~70-90% in normals), and resulted in free cystine:cysteine ratios that were widely dispro-

Table 2 Patient characteristics Group A

Sex (M/F) Entrance Weight (kg) Entrance Postnatal Age (d) Mean Amino Acid** Intake (g/kg/d) Mean Methionine Intake (mg/kg/d) Mean CysteineHC1 Intake (mg/kg/d) Mean Taurine Intake (mg/kg/d) Mean Non-ProteinCalories (kcal/kg/d) %Non-Protein Calories as Lipid

(TA + Cys) 4/6 2.48 ± 0.85 56 ± 13 2.20 ± 0.62 70 ± 20

Group B*

Group C

Group D

(FA-IIINSS) 3/7 2.32 ± 1.23 75 ± 76 2.44 ± 0.10

(FA-III) 5/4 3.02 -+ 1.04 89 ± 96 2.43 ---0.18

(FA-III+ Cys) 313 2.42 + 0.81 40 + 42 2.31 + 0.44

NS NS NS NS

129 ---9.5

122 + 14

B, C, D > A

99+34

A,D,>B,C

129 ± 5.3

85 ± 18

0

0

5.5 ± 1.2

0

0

0

p

A>B,C,D,

97 ± 29

107 ± 16

92 ± 34

78 ± 22

NS

15 ± 6

17 ± 6

18 ± 8

21 ± 9

NS

* Non-Steady State (NSS). ** Mean amino acid intake = amino acids plus L-cysteineHC1.

384 SULFURAMINOACIDMETABOLISM1NINFANTSON PARENTERALNUTRITION Table 3 Plasma sulfur amino acid concentrations,percentages and ratios Normal (11)

Mean +_SD (micromoles/100mL) Group A Group B Group C (TA + Cys) (FA-III N S S ) (FA-III) n=20 n= 10 n= 10

Group D (F-III + Cys) n=8

Methionine Cystathionine Total cysteine/ cystine

3.58 + 0.67 N.A. 15.34 + 2.55

4.25 + 1.81 0.28 + 0.21 18.9 + 3.50

15.9 (J') + 10.7 0.50 + 0.26 8.63 61,)+ 3.73

4.95 +_1.68 0.56 + 0.51 13.5 + 6.32

5.66 ($) + 1.93 0.84 + 1.1 16.8 + 5.32

Free cysteine/ cystine Free cystine Tanrine & Total cysteine/ cystine as free % Free cysteine/ cystine as free cystine Free cystine: cysteine Free cysteine/ cystine to bound cysteine/cystine

N.A. B
13.1 + 2.48 5.90 + 1.96 0.01 3.07 ($) 1.16 2.20($) + 0.61 4.69 + 2.92 3.66 + 1.52 69.3 68.4

10.2 + 5.58

8.65 + 1.78

3.04 ($) + 1.18 4.16 + 3.26 75.6

3.43 (,1,)-+ 1.02 4.98 +-3.51 51.5

~70-90 (9)

23.4

37.3

29.8

39.6

2:1 (15)

0.3:1

0.6:1

0.4:1

0.7:1

1:1 (15)

2.3:1

2.1:1

3.1:1

1.1:1

B>A,C,D D>A B
0.001 0.05 0.001 0.05 0.05 0.001

B
0.05

(1")or (,l,)= mean plasma concentration above or below the normal range (95% confidence limits)u, N.A. = not available. portionate (0.3-0.7:1 in treatment groups versus 2:1 in normal infants). Free cysteine/cystine to bound cysteine/cystine were also higher in all treatment groups. The sulfiar amino acid metabolic pathway is depicted in the Figure. The major enzymatic conversion steps are shown as solid arrows. Only the conversion of methionine to homocysteine is thought to be readily reversible via a separate enzyme, homocystine methyltransferase. The proposed product feedback loops, which possibly moderate enzymatic activities of the pathway, are supported by the plasma sulfur amino acid concentrations and are shown as dotted lines.

Discussion Our results demonstrate the effects of cysteine HC1 supplementation on plasma sulfur amino acid concentrations in parenterally-nourished, ~2 month old, postsurgical infants receiving high and low intakes of methionine with or without the addition of tanrine. Providing cysteine HC1 to infants receiving high methionine intake increases plasma methionine above the normal reference range, but appears to have little effect on plasma total cysteine/cystine and taurine concentrations when compared to infants receiving high methionine intake but without cysteine HC1 under steady state conditions. Normal methionine (albeit at the upper limits of normal), total cysteine!

cystine and taurine plasma concentrations in Group C suggest activity through the transsulfuration pathway in study subjects. This is in contrast to developmentally less mature neonates where little conversion of methionine to cysteine to taurine is assumed (8). Although sulfur amino acid nutriture m a y be met with the high methionine containing formula, FA-III, the absence of other conditionally essential amino acids and the resultant abnormal plasma amino acid patterns have been associated with significantly lower nitrogen balance and retention, and less weight gain when compared to the pediatric amino acid formula, T A (13). The low concentration of plasma cystathionine found in subjects in this study (range 0.18-3.55 m i c r o m o l e / 1 0 0 m L ) are comparable to those of Valman, et al (14). In a study of oral feeding regimens they found a similar range (and similar variability) of plasma cystathione levels with the highest concentrations being observed in infants receiving the higher protein intake. Cystathionine is not a 'normal' constituent of plasma and these levels most likely reflect a basal level of hepatic cell leakage or turnover. Since the observed cystathionine concentrations are close to the limit of detection for this amino acid, it is not clear whether the observed variability is real or an artifact of the analytical technique. However, it is notable that the infants receiving T A had lower cystathionine values when compared to F A III-nourished infants (p < 0.05 for Group A vs Group D). It is possible

CLINICAL NUTRITION 385

~Protein

~Methionine~

methionine adenosyltransferase Betaine ~Choline S-AdenosyI-Methionine Carnitine homocysteine 1 methyltransferase I SAM methyltransferase I S-AdenosyI-Homocysteine hydrolase Ho mocystei ne ,I

-~Ho mocysti ne

m

u

~~] cystathionine 13-synthase1 Cystathione ~,-cystathionase -Cysteine ~ ~Cystine~ .--Protein l -".Glutathione -~ cysteine dioxygenase I

t

Cysteinesulfinate

-'Pyruvate

I cysteinesulfinate decarboxylase I Hypotaurine I hypotaurineoxidase .Taurine Fig. Depicted is the transsulfuration pathway for methionine metabolism. Solid line ( line (- - -) indicates proposed inhibitory feedback loop.

) indicated known enzymatic pathway. Dashed

386

SULFUR AMINO ACID METABOLISM IN INFANTS ON PARENTERAL NUTRITION

that less transsulfuration pathway activity is ongoing with the use of the pediatric amino acid formulation resulting in less cystathionine generation and leakage. A consistent observation during these studies is that while cysteine supplementation increases plasma total cysteine/cystine, little change is noted in plasma free cystine. As proposed by Heird, parenteral cysteine supplementation appears to result in a greater free cysteine concentration (but not cystine) or a greater bound cysteine/cystine component (15, 16). Therefore, the normal proportion of free cystine to cysteine, approximately 2:1, or the normal proportion of free cysteine/cystine to bound cysteine/cystine, approximately 1:1, is distorted. As shown here, neither of these proportions is consistently normal during parenteral nutrition in neonates, with or without cysteine HC1 supplementation. The significance of these distortions is not known; however, some of these distortions may be influenced by plasma albumin (or other protein) concentrations since total cysteine! cystine does have a protein bound component. An important analytical concern relates to the ability to distinguish cysteine/cystine from its adduct, Dglucose-L-cysteine (DGC). DGC forms when cysteine is formulated in parenteral solutions with high concentrations of dextrose (17). In physiologic samples we were unable to distinguish cysteine/cystine from DGC. This would make interpretation of our results difficult if DGC was not usable as a cysteine source. However, several studies now suggest that DGC is bioavailable making intretation of our data possible (18, 19). Additional inhibitory feedback loops may exist between methionine and cysteine, and cysteine and taurine (Figure) in these older, more mature infants. Only in the presence of transsulfuration enzyme activity can feedback loops be proposed. Inhibition appears to be controlled by the concentration of product or intermediate product. The proposed feedback loops modulate steady state plasma amino acid concentrations to normal or near normal values. Only when intake suddenly changes, do plasma values fall outside the expected concentrations. One prominent comparison in the present study is between Groups A and B; TA versus FA-III in nonsteady state. The elevated concentrations of methionine observed in Group B (15.9 + 10.7 micromole/dL) versus that observed in Group A (4.3 _+ 1.8 micromole/ dL) can be rationalized in the following way. During delivery of a nutrient mixture containing balanced levels of methionine, cysteine/cystine, and tanrine, the feedback loops proposed in the Figure may reduce or eliminate the necessity for the enzymatic conversions of the transsulfuration pathway. Thus, more normal levels of taurine (documented in plasma and pre-

sumed for the intracellular levels) might decrease the irreversible conversion of cysteine to taurine. Similarly, normal concentrations of cysteine might be expected to moderate the conversion of methionine to cysteine via homocysteine and cystathionine. Hence, during steady state therapy with normal plasma concentrations, as in Group A, enzymatic activity is down regulated. In the short-term, non-steady state observations in Group B, a sharp increase in plasma methionine was observed with a coincident fall in plasma cysteine/cystine. These changes appear to mirror the composition of amino acid solution which contains increased methionine (resulting in an 84% increase in intake), but little or no cysteine, and no taurine. Thus, in the short-term, the down regulated enzymes are unable to adequately metabolize the available methionine, leading to the observed, elevated plasma level. This suggestion is supported by the work of Storch, et al (20). They examined dilabeled methionine kinetics in healthy, normal-weight adult males. Their data suggest two important points in methionine conservation in vivo. First is the distribution of methionine between protein anabolism and transmethylation. Second is the distribution of homocysteine between remethylation back to methionine and continuation down the transsulfuration pathway. These two loci are likely influencing our findings. Enzyme down regulation, as we have suggested, would prevent homocysteine from irreversible catabolism via transsulfuration, while enhancement of protein synthesis would divert methionine from catabolism. Finkelstein and coworkers (21) demonstrated in mammals reduced cystathionine synthesis when cystine was substituted for methionine in the diet of rats. Decreases in cystathionine synthase activity explained, in part, the finding and seem consistent with these reported findings in infants. It is unlikely that the acute drop in total cysteine/ cystine concentration in Group B is the result of enhanced cysteine dioxygenase activity in response to a 6-8 h increase in methionine intake (Figure). Bagley and Stipanuk demonstrated increased cysteine dioxygenase activity and increased taurine production in rat hepatocytes when fed excess L-methionine or Lcystine over time (22). Taurine concentrations in the Group B are not significantly increased over Group A, nor are the taurine concentrations significantly increased in any group receiving high methionine or high methionine and cysteine intake making enhanced cysteine dioxygenase activity improbable. In steady state (Group C), under the same conditions of hourly intake of nutrients, the feedback loops have been removed. Enzymatic activities are increased and a drop in plasma methionine is observed to coincide with increases in plasma cysteine/cystine. Although

CLINICAL NUTRITION 387

plasma methionine was at the upper limit of the normal reference range, it was significantly lower than Group B, while total cysteine/cystine have moved towards more normal concentrations. In Group D, the effect of adding cysteine to a high methionine intake (but otherwise equivalent to Group C) can be observed. Plasma total and free cysteine! cystine concentrations are not significantly increased; however, consistent with the proposed feedback loops, mean plasma methionine concentrations are increased above the normal reference range. In summary, little benefit is derived from supplementation of cysteine HC1 to a standard amino acid formulation in infants and may result in elevations in plasma methionine concentrations. However, supplementation of cysteine to a pediatric, low methionine, taurine-containing formulation allows for a more normal sulfur amino acid profile, but with distinct distortions of the normal proportions of bound and free cysteine/cystine. Normal sulfur amino acid concentrations in infants receiving a standard amino acid formulation suggests transsulfuration activity in these older infants. The data support the existence of enzyme regulatory forces that control the availability of methionine for catabolism via the transsulfuration pathway.

Acknowledgement The authors would like to thank Ms Kristi Boehm and Dr Elizabeth Mauer for their technical assistance, and McGaw, Inc. for analytical support. Supported in part by a Center of Excellence grant from the State of Tennessee, and the ASHP Research and Education Foundation.

References 1. Sturman J A, Gaull G E, Raiha N C R. Absence of cystathionase in human liver: is cystine essential? Science 1970; 169:74-76 2. Gaull G E, Sturman J A, Raiha N C R. Development of mammalian sulfur metabolism. Absence of cystathionase in human fetal tissues. Pediatr Res 1972; 6:538-47 3. Eggert L D, Rusho W J, McKay M W e t al. Calcium and phosphorus compatibility in parenteral nutition solutions for neonates. Am J Hosp Pharm 1982; 39:49-53 4. Snyderman S E 1971 The protein and amino acid requirements of the premature infant. In: Jonxis J H P, Visser H K A, Troelstra J A (eds), Metabolic Processes in the Foetus and Newborn Infant. Leiden: Stenfert Kroese, pp. 128-43 5. Zlotkin S H, Bryan M H, Anderson G H. Cysteine supplementation to cysteine-free intravenous feeding regimens in newborn infants. Am J Clin Nut 1981; 34:914-23

6. Malloy M H, Rassin D K, Richardson C J. Total parenteral nutrition in sick preterm infants: effects of cysteine supplementation with nitrogen intakes of 240 and 400 mg/kg/ day. J Pediatr Gastroenterol Nutr 1984; 3:239-244 7. Heird W C 1989 Essentiality of cyst(e)ine for neonates: clinical and biochemical effects of parenteral cysteine supplementation. In: Kinney J M, Borum P R (eds): Perspectives in Clinical Nutrition. Urban & Schwarzenberg: Baltimore 8. Helms R A, Christensen M L, Storm M C, Chesney R W. Adequacy of sulfur amino acid intake in infants receiving parenteral nutrition. J Nutr Biochem 1995, (September, in press) 9. Malloy M H, Rassin D K, Gaull G E. A method for measurement of free and bound plasma cyst(e)ine. Anal Biochem 1981; 113:407-15 10. Gaitonde M K. A spectrophotometric method for the direct determination of cysteine in the presence of other naturally occurring amino acids. Biochem J 1967; 104:627-33 11. Wu P Y K, Edwards N, Storm M C. Plasma amino acid pattern in normal term breast-fed infants. J Pediatr 1986; 109:347-9 12. Polberger S K T, Axelsson I E, Raiha N C R. Amino acid concentrations in plasma and urine in very low birth weight infants fed protein-uneuriched or human milk protein-enriched human milk. Pediatrics 1990; 86:909-915 13. Helms R A, Christensen M L, Mauer E C, Storm M C. Comparison of a pediatric versus standard amino acid formulation in pre-term neonates requiring parenteral nutrition. J Pediatr 1987; 110:466-70 14. Valman H B, Brown R J K, Palmer T, Oberholzer V G, Levin B. Protein intake and plasma amino acids of infants of low birth weight. Brit Med J 1971; 4:789-91 15. Heird W C, Dell R B, Helms R A e t al. Amino acid mixture designed to maintain normal plasma amino acid patterns in infants and children requiring parenteral nutrition. Pediatrics 1987; 80:401-8 16. Heird W C 1989 Essentiality of cyst(e)ine for neonates. Clinical and biochemical effects of parenteral cysteine supplementation. In: Kinney J M, Borum P R (eds), Perspectives in Clinical Nutrition. Urban & Schwarzenberg: Baltimore, pp 275-82 17. Gomez M R, Rogers L K, Smith C V, Heird W C. The fate of parenteral D-glucose-L-cysteine (DGC) in infants. Pediatr Res 1993; 32: 303A 18. Kashyap S, Abildskov K, Heird W C. Cysteine (Cys) supplementation of very low birth weight (VLBW) infants receiving parenteral nutrition (TPN). Pediatr Res 1992; 31: 290A 19. Gomez M R, Benzick A E, Rogers L K, Heird W C, Smith C V. Attenuation of acetaminophen hepatotoxicity in mice as evidence for the bioavailability of the cysteine in D-glucoseL-cysteine in vivo. Toxicol Lett 1994; 70:101-8 20. Storch K J, Wagner D A, Burke J F, Young V R: [1J3C; methyl-2H3] methionine kinetics in humans: methionine conservation and cystine sparing. Am Physiol Soc 1990; 258:E790-E798 21. Finkelstein J D, Martin J J, Harris B J. Methionine metabolism in mammals. J Biol Chem 1988; 263:11750-11754 22. Bagley P J, Stipanuk M H. Supplementation of low protein diet with sulfur amino acids results in increased cysteine dioxygenase activity and increased taurine production in rat hepatocytes. J Nutr, in press

Submission date: 20 December 1994; Accepted after revision: 14 August 1995