Reprinted from THE JOURNAL OF PEDIATRICS, SI. LouisVol. 109. No.4, PP. 668-671, October, 1986 (Printed in the U. S. A.)(Copyright@ 1986 by The C. V.MosbyCompany)

Plasma amino acid levels after single-doseaspartame consumption in phenylketonuria, mildhyperphenylalaninemia, and heterozygous statefor phenylketonuria

Benjamin Caballero. M.D,.Barbara E. Mahon. A.B..Frances J. Rohr. M.S.. R.D..Harvey L. Levy. M.D..and Richard J. Wurtman. M.D.

From the Laboratory of Neuroendocrine Regulation. Department of Applied BiologicalSciences, Massachusetts Institute of Technology, and the Department of Neurology. HarvardMedical School. Cambridge. Massachusetts; and the IEM-PKUProgram of the DevelopmentalEvaluation Clinic. The Children's Hospital. Boston

Aspartame (N-aspartyl-phenylalanine methyl ester), awidely used artificial sweetener, is hydrolyzed in theintestinal lumen to methanol and to its constituent amino

.acids phenylalanine and aspartic acid. Ingestion of thissweetener causes a sharp increase in plasma phenylalanineI vels in normal persons. and in patients with phenylketon-u ia.2The label on aspartame-containing products includesa warning to patients with PKU, but no information on thea tual sweetener content, and there is concern that aspar-t me ingestion by those who do not have PKU couldp oduce phenylalanine levels sufficiently high to harm theb ain, especially among the estimated 2% of the general

pulation who carry the gene for PKU and thus have a

r~uced capacity for phenylalanine metabolism. Studies inrAtsshow that elevations of plasma phenylalanine levels, as

Supported in part by Grants NS 21231 and NS 05096 from theational Institute of Neurological, Communicative Diseases, and

Stroke, the Center for Brain Sciences and Metabolism Charitableust, and by Grant MCJ-250501 from the Division of Maternal

a

f

d Child Health, Department of Health and Human Services.

S bmitted for publication Jan. 20, 1986; accepted May 13,I 86.

~eprint requests: B. Caballero, M.D., Rm. EJ7-445, MIT, Cam-bfidge, MA 02139.

I

occur in humans after' aspartame ingestion, produce anincrease in brain phenylalanine concentration~3.4On theother hand, the peak plasma phenylalanine levels afterloading doses of aspartame in normal persons and in PKUheterozygotes are still considerably lower than those inmentally retarded patients with PKU.s

Most of the studies in humans on the effects of aspar-tame on plasma amino acids have used high doses of the

PhePKULNAA

Phenylalanine

PhenylketonuriaLarge neutral amino acids

sweetener, I.S and have reported increases in plasmaphenylalanine levels up to fourfold in normal and tenfold inPKU heterozygotes. Moreover, no studies in humans haveaddressed the effects of aspartame on the plasma ratio ofphenylalanine to the other large neutral amino acids(valine, leucine, isoleucine, tyrosine, and tryptophan),which is the determinant of phenylalanine availability tothe brain.6

We examined the effect of an estimated average singledose of aspartame on plasma phenylalanine levels and theplasma phenylalaninejLNAA ratios in people with PKU,

Volume 109Number 4

Table I. Basal plasma amino acid concentrations

Clinical and laboratory observations 669

PhenylalanineValineIsoleucineLeucine

TyrosineTryptophanPhe/LNAA

Normalen = 10)

44.5 :t 12.9'

182.0 :t 48.7'b49.3 :t 18.9'b

103.8 :t 32.2'

57.6 :t 23.0'57.6 :t 20.4

0.100:t 0.011'

PKU

en = 15)

1369.1 :t 240.1'

151.1 :t 33.3b40.0:t 12.lb77.2 :t 20.6b

32.3 :t 6.7b42.4 :t 9.6'

4.212 :t 1.302'

Hyper-phenylalanine

en = 10)

411.7 :t 207.9b

201.6 :t 28.6'

56.5 :t 8.2'

112.7 :t 17.8'56.9 :t 11.9'63.0 :t 12.9'

0.849 :t 0.4 J4b

PKUcarrieren = 14)

68.8 :t 13.8'

168.8 :t 43.2'b

44.1 :t 9.4b93.7 :t 22.1'b

47.2 :t 9.7'56.2 :t 10.8'

0.170 :t 0.027'

Values represent mean:!: SD "mol/L.Group means with different letters are significantly different by Duncan multiple range test (P <0.05).Phe/LNAA, Phenylalanine/large neutral amino acids ratio.

Table II. Plasma amino acid concentration 1 hour after aspartame (10 mgjkg) ingestion

tho$e with non-PKU mild hyperphenylalaninemia, thosecarrying the gene for PKU, and in normal individuals.

METHODS

The subjects studied were from the population of par-ents and children observed in the IEM-PKU Program ofthe Children's Hospital, Boston, and all were ingesting anormal diet. The group included patients with classicPKU. with atypical hyperphenylalaninemia, and PKUcarriers (heterozygotes). Ten normal young adults werestudied as controls.

After an overnight fast, a baseline blood sample wasobtained, and each subject ingested, over 10 to 15 minutes,an amount of cherry-flavored Sugar-Free Kool-Aid (Gen-eral Foods, White Plains, N.Y.) that provided 10 mgaspartame/kg body weight. A second blood sample wasobtained 1 hour later. Plasma amino acids were analyzedin duplicate by high-performance liquid chromatography

. withfluorometricdetection,and resultswerecomparedbyt test or by one-way analysis of variance.

RESUL TS

As shown in Table I, pretreatment plasma phenylala-nine levels and Phe/LNAA ratios were markedly elevated

in subjects with PKU or non-PKU hyperphenylalaninemiabut were not significantly elevated in the carrier group.Basal plasma levels of the other LNAAs examined tendedto be depressed in patients with PKU; these decreasesattained statistical significance for tyrosine, tryptophan,and leucine.

Aspartame administration significantly increased plas-ma phenylalanine levels and Phe/LNAA ratios in thecontrol and carrier groups but not in non-PKU hyperphe-nylalaninemic or patients with PKU (Table II). Theabsolute increases in plasma phenylalanine levels weresimilar in the control and carrier groups (13.5 j.tM),producing concentrations that were still within 1 and 2 SD,respectively, of the mean for a general adult population.7The percent increases were smaller in the carrier groupbecause basal levels tended to be higher. The artificialsweetener failed to affect the plasma levels of otherLNAAs. However, its administration did cause the differ-ence in plasma tyrosine betwecn normal and carriersubjects to become significant (Table II), because absoluteplasma tyrosine levels were increased in normal subjectsbut not in the carrier group.

No serious adverse effects were noted in any of thesubjects during the study period. One subject in the control

% % Hyper- % %Normal Change PKU Change phenylalanine Change PKUcarrier Change

Phenylalanine 58.0 :t 9.5'. +30 1323.3 :t 209.7' -3 412.8 :t 184.lb 0 82.3 :t 17.3'. +20Valine 175.5 :t 27.2'b -4 154.5 :t 26.7' +2 198.6 :t 39.lb -I 159.8 :t 41.2' -5Isoleucine 44.0 :t 10.2' -11 40.5 :t 9.2' +1 54.7 :t 12.3b -3 41.5 :t 9.2' -6Leucine 92.9 :t 20.8' -10 78.3 :t 13.4' +2 109.4 :t 17.4b -3 89.3 :t 22.5' -5

Tyrosine 60.7 :t 15.0' +5 34.0 :t 7.5' +5 54.6 :t 11.3'b -4 46.0 :t 9.6b -2

Tryptophan 54.5 :t 12.5' -6 43.5 :t 8.3b +3 58.7 :t 12.4' -7 52.0:t 14.I'b -7

Phe/LNAA 0.137 :t 0.017'. +37 3.847 :t 0.824' -9 0.892 :t 0.430b +5 0.214 :t 0.028'. +26

Valuesrepresentmean:!:SD "mol/L. Group meanswithdifferentlettersare significantlydifferentby Duncanmultiplerange test (P <0.05).'Significant increaseoverbaselinevalues,Student t test for pairedsamples(P <0.05).

670 Clinical and laboratory observations

group complained of severe headache after ingestion of thetest dose of aspartame.

DISCUSSION

These observations show that an estimated averl1gedoseof aspartame does not accentuate hyperphenylalaninemiain patients with untreated PKU or non-PKU hyperphenyl-alaninemia, but does produce statistically significant eleva-tions in the plasma phenylalanine level :n normal subjectsand in PKU carriers. The dose of aspartame used, 10mg/kg, is one fifth the present AD! (acceptable dailyintake) level set by the Food and Drug Administration; itwas selected to approximate the expected maximal amountof aspartame that an adult would consume in 10 to 15

minutes. For a 60 kg adult, 10 mg/kg aspartame is presentin a little more than one quart of Kool-Aid (34.5 oz) or inthree cans of an aspartame-sweetened soft drink.

The plasma Phe/LNAA ratios paralleled the phenylala-nine levels, because the other LNAAs were not affected byaspartame. Thus these ratios were also increased byaspartame in both normal and PKU carrier groups butwere unchanged in those with PKU and non-PKU hyper-phenylalaninemia. AmongPKU carriers this ratio wasincreased by aspartame to values above the normal rangereported in persons consuming meals containing varyingamounts of protein and carbohydrate.8 This effect reflectsthe decreased ability of this group to hydroxylate phenyl-alanine to tyrosine.

The addition of protein to a meal actually lowers theplasma Phe/LNAA ratio (and presumably, brain phenyl-alanine concentration in humans), because even though theprotein contributes phenylalanine to the bloodstream, itcontains and contributes far larger amounts of the otherLNAAs. Thus, if an increased Phe/LNAA ratio is unde-sirable, ingesting aspartame with protein might be prefer-able to ingesting it only with carbohydrate: the insulinsecretion induced by dietary carbohydrates causes majorreductions in plasma levels of leucine, isoleucine, andvaline, thereby elevating the plasma Phe/LNAA ratio andpotentiating the effects of aspartame on brain phenylala-nine levels.4

The precise levels of plasma phenylalanine required tocause damage to the central nervous system have not beenestablisl.1ed experimentally, although based on observa-tions of patients, a "safe threshold" of 600 J.tMhas beenproposed. 9,Similarly, data are lacking on the plasma orbrain phenylalanine levels needed to impair the synthesisof brain neurotransmitters in humans, a possible mecha-nism for the brain damage in PKU.IOThe plasma phenyl-alanine levels and Phe/LNAA ratios produced in normalpersons and in tpose who carry the gene for PKU by theamount of aspartame we administered were far below

The Journal oj PediatricsOctober 1986

those noted in the basal state among individuals who arebrain damaged from PKU, or even those who have a

seemingly benign form of non-PKU hyperphenylalaninem-ia. However, it is possible that an aspartame-induced risein the plasma Phe/LNAA ratio may have different effectson the brain of a person whose plasma Phe/LNAA ratioand brain phenylalanine levels are usually normal than insomeone with a lifetime elevation in plasma phenylalanine.In the latter, adaptive changes could occur at the transportsystem that carries phenylalanine across the blood-brainbarrier. The considerable variability in the extent offunctional brain damage observed among people withuntreated PKU suggests that additional factors can modu-

late the neurotoxic effects of high phenylalanine concen-tration.1I

Our data indicate that 10 mg/kg doses of aspartamemay safely be consumed by persons with hyperhenylalani-nemia or PKU who are not on a phenylalanine-restricteddiet. This study, however, did not address the question oflong-term effects of aspartame use nor the consequences ofconsuming higher doses of aspartame, alone or with otherfoods. Moreover, plasma phenylalanine concentrationsmay increase to unacceptable levels when patients withPKU on phenylalanine-restricted diets consume aspar-tame-containing soft drinksl2 or after loading doses of thesweetener.s The appearance of severe headaches afteraspartame intake has also been reported in a non-PKUpatient receiving neuroactive drugs.13

REFERENCES

I. Stegink LD, Filer LJ Jr, Baker GL. Effect of aspartame andaspartate loading upon plasma and erythrocyte free aminoacid levels in normal adult volunteers. J Nutr 1977;107:1837-1845.

2. Koch R, Schaeffier G, Shaw KNF. Results of loading doses ofaspartame by two phenylketonuric (PKU) children comparedwith two normal children. J Toxicol Environ Health 1976;2:459-469.

3. Gleaser BS, Maher TJ, Wurtman RJ. Changes in brain levelsof acidic, basic and neutral amino acids after consumption ofsingle meals containing various proportions of protein. JNeurochem 1983;41:1016-1021.

4. Yokogoshi H, Roberts C, Caballero B, Wurtman RJ. Effectsof aspartame and glucose on plasma and brain large neutralamino acids and brain hydroxyindoles. Am J Clin Nutr1984;40:1-7.

5. Stegink LD, Filer LJ, Baker GL, McConnell JE. Effect of an

abuse dose of aspartame upon pla~ma and erythrocyte levelsof amino acids in phenylketonuric, heterozygous and normaladults. J Nutr 1980;110:2216-2224.

6. Partridge WM. Regulation of amino acid availability to thebrain. In: Wurtman RJ, Wurtman JJ, eds. Nutrition and the

brain, vol 1. New York: Raven Press, 1977;141-204.7. Scriver CR, Gregory DM, Sovetts D, Tissenbaum G. Normal

plasma free amino acid values in adults: influence of some

common physiological variables. Metabolism 1985;34:868.

Volume 109Number 4

8. Maher TJ, Glaeser BS, Wurtman RJ. Diurnal variations in

plasma concentrations of basic and neutral amino acids in redcell concentrations of aspartate and glutamate: effects of

dietary protein intake. Am J Clin Nutr 1984;39:722-729.9. Levy HL, Shih YE, Karolkewicz Y, et al. Persistent mild

hyperphenylalanimeitia in the untreated state: a prospectivestudy. N Engl J Med 1984;285:424-429.

10. Kaufman S. Phenylketonuria: biochemical mechanisms. In:

Agranoff BW, Aprison MH, eds. Advances in neurochemis-

try, vol 2. New York: Plenum Press, 1977;1-132.

Clinical and laboratory observations 671

II. Waisbren SE, Schnell R, Levy HL. Intelligence and person-

ality characteristics in adults with untreated atypical phenyl-ketonuria and mild hyperphenylalaninemia. J PEDlATR 1984;105:955-958.

12. Guttier F, Lou H. Aspartame may imperil dietary control of

phenylketonuria. Lancet 1985; I :525-526.

13. Ferguson JM. Interaction of aspartame and carbohydrates inan eating-disordered patient. Am J Psychiatry 1985;142:271.