ABSTRACT
Background: Branched-chain amino acids have been reported to improve fetal brain development in a rat model in which maternal phenylketonuria (PKU) is induced by the inclusion of an inhibitor of phenylalanine hydroxylase, DL-p-chloropheny-lalanine, and L-phenylalanine in the diet.
Objective: We studied whether a dietary mixture of several large neutral amino acids (LNAAs) would improve fetal brain growth and normalize the fetal brain amino acid profile in a rat model of maternal PKU induced by DL-a-methylphenylalanine (AMPhe). Design: Long-Evans rats were fed a basal diet or a similar diet containing 0.5% AMPhe + 3.0% L-phenylalanine (AMPhe + Phe diet) from day 11 until day 20 of gestation in experiments to test various mixtures of LNAAs. Maternal weight gains and food intakes to day 20, fetal body and brain weights at day 20, and fetal brain and fetal and maternal plasma amino acid concentra-tions at day 20 were measured.
Results: Concentrations of phenylalanine and tyrosine in fetal brain and in maternal and fetal plasma were higher and fetal brain weights were lower in rats fed the AMPhe + Phe diet than in rats fed the basal diet. However, fetal brain growth was higher and concentrations of phenylalanine and tyrosine in fetal brain and in maternal and fetal plasma were lower in rats fed the AMPhe + Phe diet plus LNAAs than in rats fed the diet contain-ing AMPhe + Phe alone.
Conclusion: LNAA supplementation of the diet improved fetal amino acid profiles and alleviated most, but not all, of the depression in fetal brain growth observed in this model of mater-nal PKU. Am J Clin Nutr 1999;69:687–96.
KEY WORDS Phenylketonuria, phenylalanine, brain, plasma, amino acids, gestation, fetus, rats, DL-a-methylphenylalanine, large neutral amino acids
INTRODUCTION
Children of phenylketonuric mothers may suffer physical anomalies and permanent effects on mental function that stem from abnormal development in a high-phenylalanine milieu dur-ing pregnancy (1–3). This problem can be minimized if phenylke-tonuric women adhere to a low-phenylalanine diet from the time of conception (2, 4–7). Such diets are not highly palatable, how-ever, and strict adherence to the diet is often not achieved (8, 9).
An alternative to a low-phenylalanine diet would be a more nor-mal diet that is altered so that the phenylalanine content is less problematic; hence, there is an interest in the possible use of large neutral amino acids (LNAAs), which compete with phenylalanine for membrane transport sites in the brain, to alleviate the prob-lems associated with hyperphenylalaninemia (10–12).
Maternal hyperphenylalaninemia can be induced in rats by inclusion in the diet of an inhibitor of phenylalanine hydroxylase, such as DL-p-chlorophenylalanine (PCPA) or DL- a-methylpheny-lalanine (AMPhe), in combination with a large supplement of L-phenylalanine. In one model of maternal phenylketonuria (PKU), supplements of 0.12% PCPA and 3.0% L-phenylalanine are included in the diet of rats from day 10 to day 20 of gestation (13, 14). Fetuses of phenylketonuric dams have lower body weights, lower brain weight relative to body weight, and lower concentrations of certain branched-chain amino acids in the brain than do fetuses of control dams (13). Supplementation of the diet with branched-chain amino acids increases fetal growth and brain weight (13, 14) and improves the performance of prog-eny in maze tests (15, 16).
PCPA, which appears to be more toxic than AMPhe (17–19), depresses fetal body and brain weights by nearly 20% (14). As an inhibitor of tryptophan hydroxylase it may alter serotonin synthesis (20). In an attempt to avoid these problems, Brass et al (21) developed a model of maternal PKU that is similar to the PCPA model except that the diet is supplemented with 0.5% AMPhe instead of 0.12% PCPA. With use of this model, body and brain weights of pups were reduced by <7% and 8%, respectively, in phenylketonuric fetuses (21). Progeny from phenylketonuric dams were also reported to have behavioral deficits (22) and impaired learning ability (23, 24).
The addition of isoleucine or a mixture of leucine, isoleucine, and valine to the diet of gestating rats in the PCPA model results
Am J Clin Nutr 1999;69:687–96. Printed in USA. © 1999 American Society for Clinical Nutrition
Effects of dietary mixtures of amino acids on fetal growth and
maternal and fetal amino acid pools in experimental maternal
phenylketonuria
1–3
Richard E Austic, Chun-Li Su, Barbara J Strupp, and David A Levitsky
687 1From the Department of Animal Science and the Division of Nutritional Sciences, Cornell University, Ithaca, NY.
2Supported in part by USPHS grant HD222558 from the NICHD. 3Address reprint requests to RE Austic, Department of Animal Science, 248 Morrison Hall, Cornell University, Ithaca, NY 14853. E-mail: bjs12@cornell.edu.
Received September 15, 1998.
in improved fetal brain growth (13, 14), but it is not clear whether these amino acids completely prevent the adverse effects of PCPA and phenylalanine on brain growth. The objec-tives of this investigation were to determine the amino acid pro-files in plasma and brain of fetuses from gestating rats fed a diet containing 0.5% AMPhe and 3.0% L-phenylalanine and to deter-mine whether the effects of AMPhe and phenylalanine on fetal brain size and amino acid profile could be prevented by supple-menting the diet with a mixture of LNAAs.
MATERIALS AND METHODS Animals
Long-Evans rats at 4–8 d of gestation were obtained from Blue Spruce Farms (Altamont, NY). They were housed individ-ually in stainless steel cages with raised wire floors in animal facilities that provided an ambient temperature of 228C, 57% rel-ative humidity, and a daily light period from 0800 to 2000. Food having the same composition as the basal diet in all experiments was provided ad libitum in glass jars that had lids and followers to minimize spillage. Water was provided ad libitum.
Food intakes and weight gains were monitored until day 11 of gestation. Only rats that consumed the diet well and gained weight at the expected rate were used in experiments. On the first day of the experiment, rats were ranked by body weight and randomly assigned to treatments within narrow weight group-ings (blocks). Daily food intake (corrected for daily food spillage) and cumulative weight gains were measured during the experimental period from day 11 to day 20 of gestation. Animals were maintained under conditions approved by the Cornell Uni-versity Institutional Animal Care and Use Committee; the exper-imental protocols were also approved by this committee.
Experimental diets
The basal diet was AIN-76A (25, 26), which contained L-methionine. L-Methionine was substituted for DL-methionine to eliminate the possibility that the utilization of D-methionine would be impaired by AMPhe (27). The experimental diets were mixed by adding AMPhe, L-phenylalanine, and LNAAs to the AIN-76A diet at the expense of cornstarch on a weight-for-weight basis. The ingredients included casein from National Casein (Riverton, NJ); cornstarch, cellulose, and corn oil (Mazola) from Corn Products (Englewood Cliffs, NJ); sucrose from Domino Sugar Corporation (New York); mineral mix, vitamin mix, and choline bitartrate from Dyets, Inc (Bethlehem,
PA); L-methionine, L-phenylalanine, L-leucine, and L-tryptophan from United States Biochemical Corporation (Cleveland); L-thre-onine, L-valine, L-isoleucine, and L-tyrosine from Kyowa Hakko USA, Inc (New York); and AMPhe from Sigma Chemical Com-pany (St Louis).
Experimental design
Four experiments were conducted to determine whether dietary supplements of LNAAs would reduce the severity of PKU in fetuses from dams fed the basal diet supplemented with 0.5% AMPhe + 3.0% L-phenylalanine (AMPhe + Phe diet). Four LNAA supplements were used that differed in the profile of amino acids included or in the amount of the mixture added to the diet (Table 1). A summary of the experimental designs is pre-sented in Table 2. Experiment 1 was carried out to test the effects of dietary supplements of AMPhe + Phe and LNAA1, using a factorial arrangement of treatments, on rats and their fetuses. Modifications of the LNAA mixture were compared in experiments 2 and 3. Experiment 4 was conducted to confirm the effects of the LNAA4 supplement on body and brain growth and to determine effects of LNAA4 on brain protein synthesis of fetuses from phenylketonuric mothers. Basal control and LNAA-supplemented groups were pair fed to the rats fed the AMPhe + Phe diet in experiments 1 and 3. Food intake was not restricted in experiment 2. In experiment 4, the basal control and AMPhe + Phe groups were pair fed to the AMPhe + Phe + LNAA4 group. Pair-fed animals received the mean weight of food consumed on the previous day by the rats of the experimental group to which they were pair fed. There were 12, 8, 9, and 12 rats per treatment in experiments 1–4, respectively. The actual number of animals from which data were obtained, however, was sometimes less as a result of failure of pregnancy or unusually small litter size.
Biochemical analyses
Rats were subdued with carbon dioxide gas and killed by guil-lotine on day 20 of gestation. Trunk blood was collected in heparin-containing tubes for analysis of maternal blood amino acids. Fetuses were removed quickly and weighed. One-half of the fetuses (<4–5 from each uterine horn) were used for collec-tion of brain tissue for amino acid analysis and the remaining half were used for collection of trunk blood after decapitation and collection of brain tissue for measurement of protein syn-thesis. Fetal blood was collected by use of heparin-containing capillary tubes and was pooled by dam.
Blood held on ice after sampling was centrifuged at 1500 3 g for 10 min (maternal) or at 1400 3 g for 4 min (fetal) at 48C, and 0.5 mL plasma was combined with 0.25 mL 8% sulfosalicylic acid containing 750 mmol norleucine/L and mixed. Brain tissue pooled from one-half of the fetuses for each dam was frozen under dry ice immediately. While still frozen, tissue was homog-enized in a tissue homogenizer in a 4%-sulfosalicylic acid solu-tion containing norleucine. After several hours of storage in a refrigerator to allow for precipitation of protein, the samples were centrifuged at 15 000 3 g for 10 min at 48C in a microcen-trifuge and the supernate was stored frozen (2208C) before analysis of free amino acids.
Amino acids were measured in experiments 1 and 3 by ion-exchange chromatography with ninhydrin detection by using an HPLC system (Hitachi Instruments, Inc, Danbury, CT) and the Pickering system for physiologic amino acid analysis (Pickering Laboratories, Inc, Mountain View, CA). A peak integration sys-TABLE 1
Composition of the large neutral amino acid (LNAA) supplements
Amino acids LNAA1 LNAA2 LNAA3 LNAA4
g/kg diet L-Thr 5.8 0.0 0.0 8.7 L-Val 5.7 5.7 8.5 8.5 L-Ile 6.3 6.3 9.5 9.5 L-Leu 6.3 6.3 9.5 9.5 L-Met 7.2 7.2 10.9 10.9 L-Trp 9.9 9.9 14.9 14.9 L-Tyr 8.8 0.0 0.0 0.0 Total 50.0 35.4 53.3 62.0
tem (Spectra-Physics, San Jose, CA) was used to quantitate the amino acids. A Beckman Instruments (Palo Alto, CA) ion-exchange HPLC system with postcolumn o-phthalaldehyde derivatization, fluorescence detection, and a similar peak inte-gration system was used in the fourth experiment.
Fetal brain tissue was held on ice during collection in exper-iment 4. The procedures of Lerner and Johnson (28) were used to prepare microsomes and a pH 5 enzyme fraction. Briefly, the assay procedure involved incubation of microsomes and the pH 5 enzyme fraction prepared from each litter at 378C in tris-HEPES buffer in the presence of amino acids including [3H]phenylalanine and polyuridylic acid, GTP, ATP, phospho-enolpyruvate, Mg2+, K+, mercaptoethanol, and pyruvate kinase. After 30 min, incubations were terminated with 10% trichloroacetic acid and the mixture was centrifuged at 900 ×g for 10 min at room temperature to separate the peptide fraction containing [3H]phenylalanine. The pellet was washed 3 times with trichloroacetic acid to remove residual [3H]phenylalanine sub-strate, solubilized in a solution of 0.2 mol NaOH/L in 0.15 mol NaCl/L and transferred to vials for scintillation counting.
Samples of the amino acid mixture containing tritiated pheny-lalanine were subjected to scintillation counting. In addition, the pH 5 enzyme fraction from fetal brain tissue in experiment 4 was analyzed for free phenylalanine by using procedures similar to those used in the analyses of plasma and brain amino acids. These analyses were used to calculate the total phenylalanine content of the incubation mixture and, subsequently, the specific activity of phenylalanine in the mixture. Tritiated toluene was used as an internal standard in each experiment in the determi-nation of the efficiency of tritium counting. Protein was meas-ured in brain homogenates by the method of Lowry et al (29).
Statistical analyses
Data were subjected to analysis of covariance. Treatment means were compared by single degree of freedom contrasts
controlling for block (initial weight groups) and litter size as the continuous variable by using the general linear models proce-dure of SAS (30). Differences between means were considered significant at P < 0.05; individual means in experiment 1 were compared by using contrasts as described above when the P value for the interaction was < 0.08.
RESULTS
The weight gains of the rats and the body weights of their fetuses in experiment 1 were lower when the diets contained AMPhe + Phe than when the diets did not contain these amino acids (Table 3). There was a possible interaction of AMPhe + Phe and LNAA1 on fetal brain weight. Fetal brain weights were lower in rats fed the AMPhe + Phe diet than in rats fed the basal diet. Fetal brain weights of rats fed the diet containing LNAA1 alone were not significantly different from those of rats fed the basal diet. Rats fed the AMPhe + Phe diet supplemented with LNAA1, however, had higher fetal brain weights than rats fed the diet containing AMPhe + Phe alone. There also was an inter-action of LNAA1 and AMPhe + Phe on food intake. Intakes of rats fed diets containing AMPhe + Phe or LNAA1 alone did not differ significantly from the intake of rats fed the basal diet, but the intake of rats receiving the combined supplement was lower than the intakes of rats in all other treatments.
Concentrations of valine and methionine were lower in fetal brain in rats fed diets containing AMPhe + Phe than in rats fed diets that did not contain the phenylalanine mixture; addition-ally, the concentration of methionine was higher when the diet contained LNAA1 than when it did not (Table 4). There were numerous interactions of AMPhe + Phe and LNAA1 on amino acid concentrations in fetal brain. Concentrations of serine, glycine, tyrosine, phenylalanine, and lysine were higher and the concentration of isoleucine was lower in rats fed the AMPhe + Phe diet than in rats fed the basal diet. The concentrations of TABLE 2
Experimental designs Experiment and
treatment group Diet composition1 Food allocation Measured variables
Experiment 1 (n = 12)2
Group 1 Basal Pair fed to group 2 Food intake, initial body weight, final body weight, litter size,
Group 2 AMPhe + Phe Ad libitum fetal body weight, fetal brain weight, and free amino acids in
Group 3 LNAA1 Pair fed to group 2 fetal brain, fetal plasma, and maternal plasma
Group 4 AMPhe + Phe + LNAA1 Pair fed to group 2
Experiment 2 (n = 8)
Group 1 AMPhe + Phe + LNAA1 Ad libitum Food intake, initial body weight, final body weight, litter
Group 2 AMPhe + Phe + LNAA2 Ad libitum size, fetal body weight, and fetal brain weight
Group 3 AMPhe + Phe + LNAA3 Ad libitum
Experiment 3 (n = 9)
Group 1 Basal Pair fed to group 2 Food intake, initial body weight, final body weight, litter
Group 2 AMPhe + Phe Ad libitum size, fetal body weight, fetal brain weight, fetal brain protein
Group 3 AMPhe + Phe + LNAA3 Pair fed to group 2 concentration, and free amino acids in fetal brain
Group 4 AMPhe + Phe + LNAA4 Pair fed to group 2
Experiment 4 (n = 12)
Group 1 Basal Pair fed to group 3 Food intake, initial body weight, final body weight, litter
Group 2 AMPhe + Phe Pair fed to group 3 size, fetal body weight, fetal brain weight, fetal brain protein
Group 3 AMPhe + Phe + LNAA4 Ad libitum concentration, and fetal brain protein synthesis
1Basal diet, AIN76-A with
L-methionine (25, 26); AMPhe + Phe, basal diet supplemented with 0.5% DL-a-methylphenylalanine and 3.0% L -phenylala-nine; LNAA, large neutral amino acid. See Table 1 for composition of LNAAs.
these amino acids did not differ significantly between rats fed the basal diet and those fed the basal diet plus LNAA1. Rats fed the diet containing AMPhe + Phe plus LNAA1, however, had significantly lower concentrations of serine, glycine, phenylala-nine, and lysine and higher concentrations of isoleucine in fetal brain than rats fed the diet containing AMPhe + Phe alone. The concentration of threonine appeared to be higher in fetal brains of rats fed AMPhe + Phe or LNAA1, but not the combined sup-plement, than in rats fed the basal diet. Leucine concentrations in fetal brain did not differ significantly between treatments.
Concentrations of threonine, valine, and methionine were lower and the concentration of glycine was higher in fetal plasma of rats fed diets containing AMPhe + Phe than in rats fed diets that did not contain the phenylalanine mixture (Table 5). There were several interactions of AMPhe + Phe and LNAA1. Concen-trations of phenylalanine and tyrosine in fetal plasma were higher and the concentration of isoleucine was lower in rats fed the AMPhe + Phe diet than in rats fed the basal diet. The concentra-tions of these 3 amino acids did not differ significantly between rats fed the basal diet and those fed the basal diet plus LNAA1.
However, concentrations of phenylalanine were lower and those of isoleucine were higher in fetal plasma when LNAA1 was added to the diet containing AMPhe + Phe than when the diet contained AMPhe + Phe alone. Tyrosine concentrations were higher in fetal plasma of rats fed the AMPhe + Phe diet than in plasma of rats fed the basal diet, but were not significantly dif-ferent between rats fed the AMPhe + Phe plus LNAA1 diet and rats fed the basal diet supplemented with LNAA1. Serine con-centrations were lower in rats fed the combined supplement than in rats fed AMPhe + Phe alone but did not differ significantly between rats fed the basal diet or the diets containing AMPhe + Phe or LNAA1 alone. Leucine and lysine concentrations in fetal plasma did not differ significantly between treatments.
Concentrations of valine, isoleucine, and leucine were higher in plasma of dams fed diets containing AMPhe + Phe than in dams fed diets that did not contain the phenylalanine mixture (Table 6). Rats fed diets containing LNAA1 had higher plasma threonine concen-trations and lower plasma lysine concenconcen-trations than did rats fed diets that did not contain LNAA1. There were significant interac-tions of AMPhe + Phe and LNAA1 on phenylalanine and tyrosine TABLE 3
Influence of large neutral amino acids (LNAAs) on selected fetal and maternal measures in rats subjected to experimental phenylketonuria (experiment 1)1
Treatment2
1: 2: 3: 4: Main effects and interactions: P
Basal AMPhe + Phe LNAA1 AMPhe + Phe + LNAA1 Pooled AMPhe + AMPhe + Phe
(n = 11) (n = 10) (n = 12) (n = 11) SEM Phe LNAA1 3 LNAA1
Fetal
Brain weight (g) 0.158a,3 0.136c 0.159a 0.146b 0.00245 < 0.001 < 0.05 < 0.08
Body weight (g) 3.40 3.17 3.37 3.17 0.0723 < 0.01 NS NS
Maternal
Food intake (g/d) 13.5a 13.8a 13.4a 12.7b 0.191 NS < 0.01 < 0.05
Initial body weight (g) 307 306 310 307 2.54 NS NS NS
Body weight gain (g) 29.7 25.5 27.0 21.9 2.17 < 0.05 NS NS
1AMPhe + Phe, basal diet supplemented with 0.5%
DL-a-methylphenylalanine and 3.0% L-phenylalanine. Means within a row with different superscript letters are significantly different, P < 0.05.
2Dams in treatment groups 1, 3, and 4 were pair fed to treatment group 2. 3Least-squares mean.
TABLE 4
Amino acid concentrations in fetal brain (experiment 1)1
Treatment2
1: 2: 3: 4: Main effects and interactions: P
Basal AMPhe + Phe LNAA1 AMPhe + Phe+ LNAA1 Pooled AMPhe + AMPhe + Phe
(n = 10) (n = 10) (n = 12) (n = 11) SEM Phe LNAA1 ×LNAA1
nmol/g tissue nmol/g tissue
Thr 869b,3 1190a 1154a 1058a,b 80.4 NS NS < 0.05 Ser 666b 1041a 614b 690b 38.0 < 0.001 < 0.001 < 0.001 Gly 1117c 2856a 1109c 1866b 171.2 < 0.001 < 0.01 < 0.05 Val 239 147 240 180 16.1 < 0.001 NS NS Met 110 50 141 99 13.9 < 0.01 < 0.05 NS Ile 107a 61b 103a 91a 7.8 < 0.01 NS < 0.05 Leu 196 168 206 200 13.5 NS NS NS Tyr 168c 338a 217b,c 272a,b 23.9 < 0.001 NS < 0.05 Phe 151c 2251a 224c 1046b 215.1 < 0.001 < 0.05 < 0.01 Lys 812b,c 1340a 792c 926b 42.4 < 0.001 < 0.001 < 0.001
1AMPhe + Phe, basal diet supplemented with 0.5%
DL-a-methylphenylalanine and 3.0% L-phenylalanine; LNAA, large neutral amino acid. Means within a row without a superscript letter in common are significantly different, P < 0.05.
2Dams in treatment groups 1, 3, and 4 were pair fed to treatment group 2. 3Least-squares mean.
concentrations in maternal plasma. Rats fed the AMPhe + Phe diet had higher concentrations of phenylalanine and tyrosine in plasma than rats fed the basal diet. Concentrations of phenylalanine and tyrosine in maternal plasma did not differ significantly between rats fed the basal diet supplemented with LNAA1 and rats fed the basal diet, but were significantly lower in rats fed the AMPhe + Phe diet supplemented with LNAA1 than in rats fed the diet containing AMPhe + Phe alone.
The composition of LNAA2 (Table 1) was similar to the com-position of LNAA1 except that it did not contain threonine and tyrosine, 2 of the amino acids in experiment 1 that were higher in concentration in fetal brain when rats were fed diets containing AMPhe + Phe than when fed the basal diet. LNAA3 contained the same amino acids as LNAA2, but the amount of inclusion in the diet was 50% higher. When responses to all 3 supplements were compared in a single experiment in which all diets also contained AMPhe + Phe, maternal weight gains and fetal body weights were
not significantly different between the LNAA1, LNAA2, and LNAA3 treatment groups (Table 7). Fetal brain weights were also not significantly different between the LNAA1 and LNAA2 treat-ment groups, but were significantly higher in the LNAA3 group than in the other 2 groups. Additionally, food intake was signifi-cantly lower in the LNAA3 group than in the other 2 groups.
In experiment 3, the LNAA3 mixture was modified by includ-ing threonine and responses to LNAA3 and LNAA4 were com-pared. Fetal brain weight, fetal body weight, initial maternal body weight, and maternal weight gain were not significantly affected by treatment (Table 8). The mean (±SEM) fetal brain protein concentrations were 90.5± 3.4, 82.3±3.1, 84.7 ±3.7, and 82.5±3.8 mg/g tissue, respectively, for the basal, AMPhe + Phe, AMPhe + Phe plus LNAA3, and AMPhe + Phe plus LNAA4 treatment groups; there were no significant differences between treatments. Food intake was significantly lower in rats fed the diet supplemented with LNAA4 than in rats fed the basal diet or TABLE 5
Amino acid concentrations in fetal plasma (experiment 1)1
Treatment2 Main effects and interactions: P
1: 2: 3: 4: Pooled AMPhe + AMPhe + Phe
Basal AMPhe + Phe LNAA1 AMPhe + Phe+ LNAA1 SEM Phe LNAA1 3 LNAA1
mmol/L plasma mmol/L plasma
Thr 538 [11]3 450 [9] 710 [12] 487 [10] 66.8 < 0.05 NS NS
Ser 257a,b[6] 284a[6] 258a,b[6] 234b[5] 6.74 NS < 0.05 < 0.05
Gly 320 [11] 424 [9] 326 [12] 394 [10] 20.8 < 0.001 NS NS Val 407 [11] 315 [9] 417 [12] 356 [10] 22.0 < 0.01 NS NS Met 151 [11] 81 [9] 171 [12] 113 [10] 14.6 < 0.001 NS NS Ile 173a[11] 126b[10] 176a[12] 170a[11] 8.97 < 0.01 < 0.05 < 0.05 Leu 280 [11] 253 [10] 290 [12] 288 [11] 15.6 NS NS NS Tyr 222c[11] 390a[10] 279b,c[12] 331a,b[11] 22.6 < 0.001 NS < 0.05 Phe 212c[11] 2336a[10] 278c[12] 1319b[11] 277.6 < 0.001 NS < 0.08 Lys 1656 [11] 1804 [10] 1765 [12] 1779 [11] 163.1 NS NS NS
1AMPhe + Phe, basal diet supplemented with 0.5% DL-a-methylphenylalanine and 3.0% L-phenylalanine; LNAA, large neutral amino acid. Means
within a row without a superscript letter in common are significantly different, P < 0.05.
2Dams in treatment groups 1, 3, and 4 were pair fed to those in group 2. 3Least-squares mean; n in brackets.
TABLE 6
Amino acid concentrations in maternal plasma (experiment 1)1
Treatment2
1: 2: 3: 4: Main effects and interactions: P
Basal AMPhe + Phe LNAA1 AMPhe + Phe + LNAA1 Pooled AMPhe + AMPhe + Phe
(n = 10) (n = 10) (n = 12) (n = 11) SEM Phe LNAA1 3 LNAA1
mmol/L plasma mmol/L plasma
Thr 13343 1377 1712 1811 177.6 NS < 0.05 NS Ser 783 689 700 712 72.7 NS NS NS Gly 426 404 441 502 37.2 NS NS NS Val 440 544 471 580 27.0 < 0.001 NS NS Met 121 147 123 134 10.5 NS NS NS Ile 245 293 260 321 13.8 < 0.001 NS NS Leu 340 411 364 441 20.4 < 0.01 NS NS Tyr 160c 511a 166c 322b 35.1 < 0.001 < 0.05 < 0.05 Phe 262b 3674a 113b 1107b 406.8 < 0.001 < 0.01 < 0.01 Lys 3604 3709 2934 2953 234.8 NS < 0.01 NS
1AMPhe + Phe, basal diet supplemented with 0.5% DL-a-methylphenylalanine and 3.0% L-phenylalanine; LNAA, large neutral amino acid. Means
within a row with different superscript letters are significantly different, P < 0.05.
2Dams in treatment groups 1, 3, and 4 were pair fed to those in group 2. For serine, n = 6, 9, 7, and 4 in groups 1–4, respectively. 3Least-squares mean.
the AMPhe + Phe diet. Food intakes of rats fed diets supple-mented with LNAA3 and LNAA4 were not significantly different. Fetuses of rats fed the diet containing AMPhe + Phe alone had higher concentrations of glycine, tyrosine, phenylalanine, and lysine and lower concentrations of valine, isoleucine, and leucine in brain than fetuses of rats fed the basal diet (Table 9). Arginine, which was not measured in experiment 1, also was higher in the AMPhe + Phe group. Threonine and serine, which were measured in experiment 1, did not resolve sufficiently dur-ing chromatography to enable accurate measurement. Fetuses of rats fed the diet containing AMPhe + Phe plus LNAA3 had higher brain concentrations of valine, methionine, and isoleucine and lower concentrations of glycine, phenylalanine, lysine, and arginine than fetuses of rats fed the diet containing AMPhe + Phe alone. Fetuses of rats fed the diet containing AMPhe + Phe plus LNAA4 had higher concentrations of methionine and lower con-centrations of glycine, phenylalanine, and lysine in brain than did fetuses of rats fed the diet containing AMPhe + Phe alone. Glycine concentrations were higher in fetal brains of rats fed LNAA4 than in those fed LNAA3. In general, concentrations of LNAAs in fetal brain, other than threonine, which was not meas-ured, were not significantly different between the LNAA3 and LNAA4 treatments. Concentrations of valine and isoleucine in the fetal brains of rats fed AMPhe + Phe plus LNAA3 and con-centrations of valine, isoleucine, and leucine in the fetal brains
of rats fed AMPhe + Phe plus LNAA4 were lower than the con-centrations of these LNAAs in fetal brains of rats fed the basal diet. The concentration of phenylalanine was higher in fetal brains of rats fed the diets containing AMPhe + Phe plus LNAA3 or LNAA4 than in rats fed the basal diet.
The addition of LNAAs to diets containing AMPhe + Phe tended to result in lower food intake (Tables 3, 7, and 8). Because reduced food intake might limit fetal growth or improvements in amino acid profiles, a final experiment was conducted to test the effects of LNAA4 supplementation of the diet under conditions in which rats fed the basal and the AMPhe + Phe diets were pair fed to rats fed the AMPhe + Phe plus LNAA4 diet. Food intakes, weight gains of dams, and fetal body and brain weights were lower in rats fed the AMPhe + Phe diet than in rats fed the basal diet (Table 10). Maternal body weight gains were higher in rats fed the AMPhe + Phe plus LNAA4 diet than in rats fed the basal diet; however, fetal body and brain weights were not significantly different between rats fed the AMPhe + Phe plus LNAA4 diet and those fed the basal diet.
Protein concentrations in fetal brain did not differ signifi-cantly between rats fed the basal diet and those fed the AMPhe + Phe diet (Table 11). Protein concentrations were lower, how-ever, in fetal brains of rats fed the AMPhe + Phe plus LNAA4 diet than in those fed the basal diet. There were no significant differences between treatments in protein synthesis as measured TABLE 7
Effect of different mixtures of large neutral amino acids (LNAAs) on selected fetal and maternal measures in rats receiving 0.5% DL-a-methylphenylalanine (AMPhe) and 3% L-phenylalanine (Phe) in their diets (experiment 2)1
Treatment2
1: 2: 3:
AMPhe + Phe + LNAA1 AMPhe + Phe + LNAA2 AMPhe + Phe + LNAA3 Pooled
(n = 6) (n = 7) (n = 7) SEM Fetal Brain weight (g) 0.136b,3 0.138b 0.150a 0.00316 Body weight (g) 3.07a 3.07a 3.11a 0.0682 Maternal Food intake (g/d) 13.1a 13.0a 11.3b 0.405
Initial body weight (g) 314a 318a 308a 4.94
Body weight gain (g) 33.6a 39.1a 33.5a 3.89
1Means within a row with different superscript letters are significantly different, P < 0.05. 2Dams were fed ad libitum.
3Least-squares mean.
TABLE 8
Comparison of the effectiveness of large neutral amino acid supplement 3 (LNAA3) and LNAA4 in alleviating the adverse effects of 0.5% DL-a-methylphenylalanine (AMPhe) and 3% L-phenylalanine (Phe) on selected fetal and maternal measures (experiment 3)1
Treatment2
1: 2: 3: 4:
Basal AMPhe + Phe AMPhe + Phe + LNAA3 AMPhe +Phe + LNAA4 Pooled
(n = 9) (n = 9) (n = 8) (n = 8) SEM
Fetal
Brain weight (g) 0.153a,3 0.135a 0.146a 0.142a 0.00601
Body weight (g) 3.06a 2.98a 3.05a 3.06a 0.0748
Maternal
Food intake (g/d) 13.5a,b 14.1a 12.1b,c 11.7c 0.484
Initial body weight (g) 309a 309a 309a 307a 1.43
Body weight gain (g) 36.9a 36.2a 33.4a 27.2a 3.87
1Means within a row without a superscript letter in common are significantly different, P < 0.05. 2Dams in treatment groups 1, 3, and 4 were pair fed to those in treatment group 2.
in vitro by using a synthetic messenger RNA template and microsomal and pH 5 enzyme fractions from fetal brain.
DISCUSSION
The inclusion of 0.5% AMPhe and 3.0% L-phenylalanine in the diet from day 11 to day 20 of gestation resulted in lower fetal brain weights and, in 2 of 3 experiments, lower fetal body weights than were observed in rats fed the basal diet. This is con-sistent with the findings of Brass et al (21), who developed this model of maternal PKU, and others (14) who used PCPA to induce maternal PKU in rats.
In both gestational models and in postnatal models of PKU in rats and mice, the administration of phenylalanine with AMPhe or PCPA has been reported to increase the concentrations of phenylalanine and tyrosine and to decrease the concentrations of certain LNAAs in fetal brain (10–14, 21, 31). In agreement with these reports, AMPhe + Phe consistently resulted in lower fetal brain concentrations of isoleucine and valine and in 1 of 2 exper-iments resulted in significantly lower concentrations of leucine
than did the basal diet. The high glycine concentration in rats fed AMPhe + Phe agrees with the reports of Brass et al (21) and Su et al (31) and with the reports of Dienel (32) and Huether et al (33) involving postnatal rat models. The changes in leucine, isoleucine, valine, phenylalanine, tyrosine, threonine, glycine, serine, lysine, and arginine concentrations in fetal brain in this study are consistent with those observed previously in this labo-ratory (31).
Administrations of individual LNAAs or mixtures of LNAAs have been reported to increase brain protein synthesis in postnatal murine models of PKU (12, 34), and to improve fetal brain growth in a PCPA model of maternal PKU in rats (13, 14). This led various investigators to suggest that competition between phenylalanine and other LNAAs for uptake into the brain limits the availability of one or more LNAAs for protein synthesis and, consequently, for brain growth and development (10–15).
The results of the present investigation indicate that supple-mentation of the diet of gestating dams with a mixture of LNAAs ameliorates, but does not fully prevent, the adverse effects of AMPhe + Phe on fetal brain growth and tends to normalize the TABLE 9
Amino acid concentrations in fetal brain (experiment 3)1
Treatment2
1: 2: 3: 4:
Basal AMPhe + Phe AMPhe + Phe + LNAA3 AMPhe + Phe + LNAA4 Pooled
(n = 9) (n = 8) (n = 7) (n = 8) SEM mmol/L plasma Gly 1268c,3 3218a 1463c 1843b 101.0 Val 308a 170c 227b 199b,c 13.1 Met 97a,b 53b 135a 129a 19.6 Ile 147a 74c 106b 89b,c 8.71 Leu 256a 185b 226a,b 211b 14.3 Tyr 182b 280a 247a,b 258a 23.0 Phe 226c 3069a 1204b 1499b 224.6 Lys 936b 1422a 796b 979b 75.3 Arg 78c 132a 92b,c 108a,b 9.30
1AMPhe + Phe, basal diet supplemented with 0.5%
DL-a-methylphenylalanine and 3.0% L-phenylalanine; LNAA, large neutral amino acid. Means within a row without a superscript letter in common are significantly different, P < 0.05.
2Dams in treatment groups 1, 3, and 4 were pair fed to those in group 2. For arginine, n = 8, 8, 7, 8 in groups 1–4, respectively. 3Least-squares mean.
TABLE 10
Influence of large neutral amino acid supplement 4 (LNAA4) on selected fetal and maternal measures in rats subjected to experimental phenylketonuria (experiment 4)1
Treatment2
1: 2: 3:
Basal AMPhe + Phe AMPhe + Phe +LNAA4 Pooled
(n = 9) (n = 10) (n = 12) SEM
Fetal
Brain weight (g) 0.149a,3 0.126b 0.149a 0.00216
Body weight (g) 2.98a 2.68b 3.15a 0.0736
Maternal
Food intake (g/d) 10.0a 9.4b 10.0a 1.57
Initial body weight (g) 285a 283a 287a 1.19
Body weight gain (g) 24.3b 17.8c 30.3a 1.60
1AMPhe + Phe, basal diet supplemented with 0.5%
DL-a-methylphenylalanine and 3.0% L-phenylalanine. Means within a row with different superscript letters are significantly different, P < 0.05.
2Dams in treatment groups 1 and 2 were pair fed to those in treatment group 3. 3Least-squares mean.
fetal brain amino acid profile. This response cannot be attributed solely to transport interactions at the blood-brain interface because there was a tendency for the concentrations of isoleucine, valine, and methionine to be lower in fetal plasma as well as fetal brain with AMPhe + Phe treatment. Furthermore, the addition of LNAAs to the diet containing AMPhe + Phe resulted in markedly lower phenylalanine concentrations not only in fetal brain but also in fetal blood and maternal blood.
Concentration ratios of amino acids in fetal brain relative to fetal plasma and in fetal plasma relative to maternal plasma, cal-culated from treatment means in Tables 4, 5, and 6, are shown in Table 12. The inclusion of AMPhe + Phe to the basal diet tended to result in lower fetal brain–to–fetal plasma and fetal plasma–to–maternal plasma concentration ratios of isoleucine, valine, leucine, and methionine than observed with the basal diet alone. The ratios averaged 16% lower for fetal brain to fetal plasma and 41% lower for fetal plasma to maternal plasma (Table 13). In a previous study (31), the concentration ratios of isoleucine, valine, leucine, and methionine in response to AMPhe + Phe averaged 15% lower for fetal brain to fetal plasma and 46% lower for fetal plasma to maternal plasma than with a basal diet. The addition of LNAAs to the diet containing AMPhe + Phe tended to prevent changes in the fetal brain–to–fetal plasma and fetal plasma–to–maternal plasma concentration ratios of these amino acids (Table 12 and Table 13), although only leucine and methionine fetal brain–to–fetal plasma concen-tration ratios appeared to be restored fully to the ratios extant in
animals fed the basal diet. It is tempting to speculate that the pri-mary effects of the mixture of LNAAs are twofold: 1) to reduce the phenylalanine concentration in maternal blood, thereby lim-iting the interference of phenylalanine with the transport of other amino acids to the fetus, and 2) to reduce, secondarily, the con-centration of phenylalanine in fetal blood, thereby limiting the interference of phenylalanine with the transport of other amino acids into fetal brain.
The LNAA1 mixture was based on the profile of amino acids that Binek-Singer and Johnson (12) administered by injection into mice in a preweaning model of PKU. The amount of the mixture included in the diet in the present study was an educated guess based on an investigator’s experience with amino acid mixtures in diets nutritionally adequate in total protein but marginally ade-quate or excessive in a single amino acid (35). It was obvious in experiment 1 that LNAA1 improved, but did not restore, fetal brain amino acid profiles compared with those of rats consuming the basal diet. Tyrosine clearly was not needed in the mixture because the addition of AMPhe + Phe to the basal diet resulted in higher tyrosine concentrations in fetal brain and maternal and fetal plasma. Brain threonine concentrations also were higher. Because threonine is a precursor of glycine (36, 37), the concentration of which was markedly higher in brains of fetuses from rats fed the AMPhe + Phe diet than in those fed the basal diet, it seemed desir-able to omit threonine from the mixture of LNAAs. The mixture lacking threonine and tyrosine (ie, LNAA2) did not improve fetal brain growth in rats receiving AMPhe + Phe (Table 7), but
increas-TABLE 12
Concentration ratios of selected large neutral amino acids (LNAAs) in fetal brain (FB) relative to fetal plasma (FP) and in FP relative to maternal plasma (MP) (experiment 1)1
Treatment
1: 2: 3: 4:
Basal AMPhe + Phe LNAA1 AMPhe + Phe + LNAA1
FB:FP FP:MP FB:FP FP:MP FB:FP FP:MP FB:FP FP:MP Thr 1.61 0.40 2.64 0.34 1.62 0.42 2.17 0.27 Ser 2.59 0.33 3.66 0.41 2.38 0.37 2.94 0.33 Gly 3.49 0.75 6.74 1.05 3.40 0.74 4.73 0.78 Val 0.59 0.91 0.47 0.57 0.58 0.88 0.51 0.63 Met 0.75 1.26 0.64 0.56 0.82 1.39 0.75 0.88 Ile 0.62 0.70 0.46 0.42 0.58 0.68 0.54 0.54 Leu 0.70 0.81 0.65 0.61 0.71 0.80 0.70 0.65 Tyr 0.79 1.36 0.90 0.73 0.78 1.68 0.80 1.06 Phe 0.71 0.81 0.96 0.64 0.80 2.46 0.79 1.19 Lys 0.49 0.46 0.74 0.88 0.49 0.60 0.52 1.30
1AMPhe + Phe, basal diet supplemented with 0.5% DL-a-methylphenylalanine and 3% L-phenylalanine. All values are derived from the means
in Tables 4, 5, and 6.
TABLE 11
Effects of large neutral amino acids (LNAAs) on brain protein content and protein synthesis of fetal brain from rats subjected to experimental phenylketonuria (experiment 4)1
Treatment
1: 2: 3: Pooled
Basal AMPhe + Phe AMPhe + Phe + LNAA4 SEM
Brain protein (mg/g tissue) 74.7a[9]2 72.6a,b[10] 69.4b[12] 1.13
Phenylalanine incorporated (pmol/incubation)3 892a[7] 1080a[9] 1202a[12] 106.0
1AMPhe + Phe, basal diet supplemented with 0.5%
DL-a-methylphenylalanine and 3% L-phenylalanine. Means within a row without a superscript let-ter in common are significantly different, P < 0.05.
2Least-squares mean; n in brackets.
ing the dietary supplementation by 50% (ie, LNAA3) resulted in significantly higher brain weights. This higher level of LNAA sup-plementation, however, resulted in lower food intake.
Diets for rats and other simple-stomached animals are easily imbalanced with regard to threonine when other amino acids are provided in excess (38–40). Because of the possibility that the supplemental amino acids (LNAA3) increased the need for thre-onine, this amino acid was reintroduced into the mixture (LNAA4). In an experiment to compare LNAA3 and LNAA4 (Table 8), the negative control (AMPhe + Phe treatment) did not result in low fetal brain weights (P > 0.05) and, therefore, the intended comparison was not possible. The tendency of LNAA4 to depress food intake may have partly offset its otherwise posi-tive effects. Therefore, a final experiment was carried out to test the effect of LNAA4 under conditions in which all treatments were pair fed to the AMPhe + Phe plus LNAA4 group rather than to the AMPhe + Phe group as in the first 3 experiments. With use of this pair-feeding regimen, the adverse effect of AMPhe + Phe on fetal brain weight was prevented by supplementation of the diet with LNAA4; however, protein concentrations of fetal brain were <7% lower than those of the untreated controls.
Weight gains and the efficiencies of food utilization for weight gain of dams were higher in rats fed the AMPhe + Phe plus LNAA4 diet than in those fed the basal diet. The reason for this is not clear. All rats had limited food intake as a result of pair-feeding. Perhaps the intake of one or more amino acids was particularly limiting for the growth of dams fed the basal diet and this limitation was overcome by the amino acids contained in LNAA4. Alternatively, or in addition, the increase in weight gain may have reflected a change in body composition.
Brain protein synthesis as assessed in a cell-free system with a synthetic template did not differ significantly between treat-ments. This suggests that neither ribosomal activity nor the activity of the pH 5 enzyme fraction was reduced by AMPhe + Phe treatment or increased by LNAA. This contrasts with the results of Binek-Singer and Johnson (12) in a preweaning mouse model in which they observed alterations in the polyribosomal profile indicative of reduced initiation of protein synthesis and a decreased rate of peptide elongation in brain tissue of mice injected with AMPhe and phenylalanine—effects that were pre-vented by the injection of a mixture of LNAAs.
The present studies indicate that the adverse effect of a dietary supplement of 0.5% AMPhe plus 3.0% L-phenylalanine from day 11 to day 20 of gestation on fetal growth and fetal brain weight
in a rat model of maternal PKU was ameliorated but not com-pletely prevented by supplementation of the diet of the dam with a mixture of several LNAAs. Significant alterations in the con-centrations of branched-chain amino acids (leucine, isoleucine, and valine) and other amino acids persisted in fetal brain when the mixture of LNAAs was included in the diet of phenylke-tonuric dams.
REFERENCES
1. Mabry CC, Denniston JC, Nelson TL, Son CD. Maternal phenylke-tonuria. A cause of mental retardation in children without the meta-bolic defect. N Engl J Med 1963;269:1404–8.
2. Lenke RR, Levy HL. Maternal phenylketonuria and hyperphenylala-ninemia. An international survey of the outcome of untreated and treated pregnancies. N Engl J Med 1980;303:1202–8.
3. Rouse B, Lockhart L, Matalon R, et al. Maternal phenylketonuria pregnancy outcome: a preliminary report of facial dysmorphology and major malformations. J Inherit Metab Dis 1990;13:289–91. 4. Lenke RR, Levy HL. Maternal phenylketonuria—results of dietary
therapy. Am J Obstet Gynecol 1982;142:548–53.
5. Koch R, Hanley W, Levy H, et al. A preliminary report of the col-laborative study of maternal phenylketonuria in the United States and Canada. J Inherit Metab Dis 1990;13:641–50.
6. Smith I, Glossop J, Beasley M. Fetal damage due to maternal phenylketonuria: effects of dietary treatment and maternal pheny-lalanine concentrations around the time of conception. J Inherit Metab Dis 1990;13:651–7.
7. Drogari E, Beasley M, Smith I, Lloyd JK. Timing of strict diet in relation to fetal damage in maternal phenylketonuria. Lancet 1987;2:927–30.
8. Pueschel SM, Hum C, Andrews M. Nutritional management of the female with phenylketonuria during pregnancy. Am J Clin Nutr 1977;30:1153–61.
9. Kecskemethy HH, Lobbregt D, Levy HL. The use of gelatin capsules for ingestion of formula in dietary treatment of maternal phenylke-tonuria. J Inherit Metab Dis 1993;16:111–8.
10. Andersen AE, Avins L. Lowering brain phenylalanine levels by giv-ing other large neutral amino acids. Arch Neurol 1976;33:684–6. 11. Berry HK, Bofinger MK, Hunt MM, Phillips PJ, Guilfoile MB.
Reduction of cerebrospinal fluid phenylalanine after oral administra-tion of valine, isoleucine, and leucine. Pediatr Res 1982;16:751–5. 12. Binek-Singer P, Johnson TC. The effects of chronic
hyperpheny-lalaninaemia on mouse brain protein synthesis can be prevented by other amino acids. Biochem J 1982;206:407–14.
13. Berry HK, Butcher RE, Brunner RL, Bray NW, Hunt MM, Wharton CH. New approaches to treatment of phenylketonuria. In: Mittler P, ed. Research to practice in mental retardation. Biomedical aspects.
TABLE 13
Percentage decreases in the distribution ratios of selected large neutral amino acids (LNAAs) in fetal brain (FB) relative to fetal plasma (FP) and in FP relative to maternal plasma (MP) (experiment 1)1
Percentage decrease due to Percentage decrease due to
AMPhe + Phe AMPhe + Phe + LNAA1
FB:FP FP:MP FB:FP FP:MP % % Val 20 37 14 31 Met 15 56 0 30 Ile 26 40 13 23 Leu 7 25 0 20
Val + Met + Ile + Leu 16 41 6 27
Val + Ile + Leu 17 51 8 25
Val + Ile 23 38 13 27
1AMPhe + Phe, basal diet supplemented with 0.5%
Vol 3. Baltimore: University Park Press, 1977:229–39.
14. Brunner RL, Vorhees CV, McLean MS, Butcher RE, Berry HK. Ben-eficial effect of isoleucine on fetal brain development in induced phenylketonuria. Brain Res 1978;154:191–5.
15. McSwigan JD, Vorhees CV, Brunner RL, Butcher RE, Berry HK. Amelioration of maze deficits from induced hyperphenylalaninemia in adult rats using valine, isoleucine, and leucine. Behav Neural Biol 1981;33:378–84.
16. Vorhees CV, Acuff-Smith KD, Weisenburger WP, Minck DR, Berry HK. Branched chain amino acids improve radial-arm maze acquisi-tion and water maze forced-choice learning in rat offspring exposed in utero to hyperphenylalaninemia. Neurotoxicol Teratol 1992;14:35–41.
17. Kelly CJ, Johnson TC. Effects of p-chlorophenylalanine and a-methylphenylalanine on amino acid uptake and protein synthesis in mouse neuroblastoma cells. Biochem J 1978;174:931–8.
18. Delvalle JA, Dienel G, Greengard O. Comparison of a-methylpheny-lalanine and p-chlorophenya-methylpheny-lalanine as inducers of chronic hyper-phenylalaninaemia in developing rats. Biochem J 1978;170:449–59. 19. Lane JD, Schöne B, Langenbeck U, Neuhoff V. Characterization of
experimental phenylketonuria. Augmentation of hyperphenylalanine-mia with a-methylphenylalanine and p-chlorophenylalanine. Biochim Biophys Acta 1980;627:144–56.
20. Koe BK, Weissman A. p-Chlorophenylalanine: a specific depletor of brain serotonin. J Pharmacol Exp Ther 1966;154:499–516. 21. Brass CA, Isaacs CE, McChesney R, Greengard O. The effects of
hyperphenylalaninemia on fetal development: a new animal model of maternal phenylketonuria. Pediatr Res 1982;16:388–94.
22. Hanulak AT, Hull EM. Behavioral deficits in a rat model of maternal PKU. Psychobiology 1987;15:75–8.
23. Sadava D, Sutcliffe D. The effects of maternal hyperphenylalanine-mia on learning in mature rats. Life Sci 1988;43:1119–23.
24. Strupp BJ, Bunsey M, Levitsky DA, Hamberger K. Deficient cumu-lative learning: an animal model of retarded cognitive development. Neurotoxicol Teratol 1994;16:71–9.
25. Ad Hoc Committee on Standards for Nutritional Studies. Report of the American Institute of Nutrition Ad Hoc Committee on Standards for Nutritional Studies. J Nutr 1977;107:1340–8.
26. Ad Hoc Committee on Standards for Nutritional Studies. Second report of the American Institute of Nutrition Ad Hoc Committee on Standards for Nutritional Studies. J Nutr 1980;110:1726.
27. Umbreit WW. Reactions of a-methyl amino acids. In: McElroy WD, Glass HB, eds. A symposium on amino acid metabolism. Baltimore: The Johns Hopkins Press, 1955:48–54.
28. Lerner MP, Johnson TC. Regulation of protein synthesis in developing mouse brain tissue. Alteration in ribosomal activity. J Biol Chem 1970;245:1388–93.
29. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measure-ment with the folin phenol reagent. J Biol Chem 1951;193:265–75. 30. SAS Institute, Inc. SAS user’s guide: statistics, version 5 ed. Cary, NC:
SAS Institute, Inc, 1985.
31. Su C-L, Austic RE, Strupp BJ, Levitsky DA. Improved model of maternal phenylketonuria in rats by use of lower dietary concentra-tions of a-methylphenylalanine and L-phenylalanine in a semipurified diet. Nutr Res 1998;18:989–1002.
32. Dienel GA. Chronic hyperphenylalaninemia produces cerebral hyper-glycinemia in immature rats. J Neurochem 1981;36:34–43.
33. Huether G, Kaus R, Neuhoff V. Amino acid depletion in the blood and brain tissue of hyperphenylalaninemic rats is abolished by the admin-istration of additional lysine: a contribution to the understanding of the metabolic defects in phenylketonuria. Biochem Med 1985;33:334–41. 34. Hughes JV, Johnson TC. Experimentally-induced and natural recovery from the effects of phenylalanine on brain protein synthesis. Biochim Biophys Acta 1978;517:473–85.
35. Davis AJ, Austic RE. Dietary protein and amino acid levels alter thre-onine dehydrogenase activity in hepatic mitochondria of Gallus domesticus. J Nutr 1997;127:738–44.
36. Bird MI, Nunn PB, Lord LAJ. Formation of glycine and aminoacetone from L-threonine by rat liver mitochondria. Biochim Biophys Acta 1984;802:229–36.
37. Ballèvre O, Cadenhead A, Calder AG, et al. Quantitative partition of threonine oxidation in pigs: effect of dietary threonine. Am J Physiol 1990;259:E483–91.
38. Leung PM-B, Rogers QR, Harper AE. Effect of amino acid imbalance on plasma and tissue free amino acids in the rat. J Nutr 1968;96:303–18.
39. Tews JK, Kim Y-WL, Harper AE. Induction of threonine imbalance by dispensable amino acids: relation to competition for amino acid trans-port into brain. J Nutr 1979;109:304–15.
40. Davis AJ, Austic RE. Dietary threonine imbalance alters threonine dehydrogenase activity in isolated hepatic mitochondria of chicks and
