ASPARTAME*
Explanation
Aspartame was first evaluated by JECFA in 1975 (see Annex,
Ref. 37). At that time a special problem was posed by the presence
of the conversion product, 5-benzyl-3, 6-dioxo-2-piperazine
(diketopiperazine, DKP) and no ADI for man was allocated. It was again
considered by JECFA in 1976 and its consideration was deferred in view
of the incompleteness of the information available (see Annex,
Ref. 40).
In 1977 JECFA had evidence that the problem with diketopiperazine
was of no significance and concluded that the safety of aspartame had
been adequately demonstrated; the Committee was prepared to establish
an ADI for man, but because of the assertion that the data base from
which the conclusions were drawn required validation the Committee
deferred its decision pending an assurance that the toxicological data
were valid (see Annex, Ref. 43).
In 1979 JECFA was presented with evidence of validation of the
toxicological data and accepted the validation; however, the Committee
did not have sufficient time to reassess the data on aspartame which
were evaluated by the previous meeting (see Annex, Ref. 51). In 1980
JECFA evaluated additional toxicity animal studies and several human
studies and an ADI of 0-40 mg/kg bw for aspertame and an ADI of
0-7.5 mg/kg bw for its breakdown, diketopiperazine, were established
(see Annex, Refs. 54 and 56).
Since that evaluation, additional studies have become available
and are summarized and discussed in the following monograph addendum.
BIOLOGICAL DATA
TOXICOLOGICAL STUDIES
Long-term studies
Rat
In a study designed to evaluate and characterize the effects
of long-term administration of aspartame or aspartame and
diketopiperazine in Wistar rats, groups of 86 male and 86 female rats
were fed a powdered basal diet containing 0, 1, 2, 4 g/kg/day of
aspartame or 4 g/kg/day of aspartame and diketopiperazine (3:1). Each
group was divided into a main and a satellite group for interim
* Monograph addendum to the monograph appearing in Ref. 56 (Annex).
clinical and post mortem examination. In satellite groups, 10 males
and 10 females were examined after 26 weeks and 16 males and 16
females of each group after 52 weeks. The remaining survivors were
killed at 104 weeks. No spontaneous deaths were observed at 26 and 52
weeks. Mortality rate of the various test groups was comparable at 104
weeks. A significant increase in urinary specific gravity and a
decrease in urinary pH were noted in the 4 g/kg aspartame and 4 g/kg
aspartame plus diketopiperazine groups. Urinary calcium excretion was
increased in both male and female at 2 and 4 g/kg aspartame and 4 g/kg
aspartame plus diketopiperazine throughout the study. Relative kidney
weights were increased at the higher dose levels in both sexes at 26
and 52 weeks. Histopathology of the kidneys revealed a high incidence,
over 95%, of chronic nephropathy in all groups including control. The
incidence of nephrocalcinosis including pelvic and medullary and
metastic mineralization appeared to be increased mainly in the females
of the aspartame treated groups when compared to the control group.
The spontaneous incidence of nephrocalcinosis in the controls was
relatively high, particularly in females (Ishii et al., 1981). The
incidence of brain tumours in this study was reported separately. No
brain tumours were detected at 26 or 52 weeks. The incidence of brain
tumours in rats exposed to the test material for more than one year
was as follows:
Control - 1/119 (0.8%) - 1 female astrocytoma
atypical at 99th week
1 g/kg - 1/119 (0.8%) - 1 male oligodendroglioma
at 75th week
2 g/kg - 2/120 (1.7%) - 1 female astrocytoma and
1 female ependymoma at
terminal sacrifice
4 g/kg - 1/120 (0.8%) - 1 male astrocytoma at
93rd week
4 g/kg - 1/120 (0.8%) - 1 female oligodendroglioma
(APM plus DKP) at 51st week
(Ishii, 1981)
OBSERVATIONS IN MAN
Four normal subjects and four normal obligate PKU heterozygotes
received 34 mg/kg aspartame in 8 oz of orange juice in a fasting
state. The normal controls were healthy young women varying in age
from 20 to 28 years. The PKU heterozygote mothers were selected to
match the age of the controls. Blood samples were obtained by
venipuncture at 0, 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4 and 8 hours after
ingestion and analysed for amino acid content of plasma and
erythrocytes. The data on normals revealed that plasma and
erythrocyte phenylalanine levels remained within the normal range
(6.06-18.18 µm/dl). The values for PKU heterozygotes were higher than
for the controls, but were generally within the range considered
normal. Peak values occurred between 0.5 and 2 hours and returned to
near pretest values during the period of testing. This was generally
the case for all the amino acids analysed (Koch & Blaskovics, 1978).
Six normal healthy adults (three male and three female) and five
female subjects heterozygous for phenylketonuria were administered
aspartame at 100 mg/kg bw dissolved in 500 ml of orange juice.
Aspartame was administered to subjects in a fasting state. Plasma and
erythrocyte amino acid levels were measured at 0, 0.25, 0.5, 0.75, 1,
1.5, 2, 3, 4, 5, 6, 7 and 8 hours after test load. Plasma levels of
aspartate were not significantly affected in either group. Similarly,
levels of glutamate, asparagine and glutamine, which are readily
derived from aspartate, were essentially unchanged. Plasma
phenylalanine levels were significantly increased after aspartame load
in both groups. Mean maximum phenylalanine levels observed in normal
subjects were approximately 20 µmol/dl at 30-90 minutes after loading
while those noted for heterozygous PKU subjects were twice as large,
ranging from 36.5 µmol/dl at 30 minutes to 41.7 µmol/dl at 90 minutes.
Plasma tyrosine levels increased in both groups after loading, with
higher levels noted in normal subjects. This is to be expected since
heterozygous PKU subjects have a decreased ability to convert
phenylalanine to tyrosine. Erythrocyte levels of amino acids followed
the same pattern as those reported for plasma (Stegink et al., 1978).
A total of 12 infants, aged eight to 12 months, were administered
aspartame dissolved in Kool-Aid at 34 and 50 mg aspartame/kg bw (six
at each dose level). Blood samples were obtained by heel stick at 0,
30, 45, 60, 90, 120 and 150 minutes. A total of four samples was
obtained from each infant (a fasting sample and three subsequent
samples).
Each blood sample was analysed for plasma and erythrocyte free
amino acid levels and blood methanol concentration. Plasma aspartate
levels were higher in the infants than previously observed in adults.
There was, however, no increase in plasma aspartate after loading with
aspartame. No significant changes were noted in erythrocyte aspartate
levels. Plasma phenylalanine levels increased slightly when 34 mg
aspartame/kg bw was administered, rising from a mean of 6.3 µmol/dl at
zero time to 9.7 µmol/dl at 30 minutes. Erythrocyte phenylalanine
levels showed a similar, but smaller response. When a 50 mg/kg bw dose
was administered plasma phenylalanine rose from 5.7 at zero time to
11.6 µmol/dl at 60 minutes. Erythrocyte phenylalanine values increased
but response as lower. Blood methanol levels increased at both loading
doses from 0.07 to 0.19 and 0.3 mg/dl respectively, 45 to 90 minutes
after loading, followed by a decrease to baseline values. These data
show that a one-year-old infant handles aspartame as well as the
normal adult at these dosage levels. The failure to increase plasma
aspartate, phenylalanine and methanol above post prandial levels would
indicate little hazard to the infant from aspartame at the dosage
levels studied. Since the infant metabolized aspartame as well as an
adult and in previous studies no adverse effects were seen in the
adult when given 100, 150 and 200 mg/kg bw of aspartame, this study
was extended to giving a loading dose of 100 mg/kg bw of aspartame in
Kool-Aid to a total of eight infants, eight to 12 months of age. A
fasting and three subsequent blood samples were obtained from each
infant. Plasma and erythrocyte aspartate levels were unchanged after
aspartame loading. Plasma phenylalanine levels increased from 4.8 to
21.4 µmol/dl at 45 minutes. Erythrocyte phenylalanine showed a similar
but somewhat lower response. Blood methanol levels increased from
0.11 mg/dl to 1.02 mg/dl at 90 minutes (Stegink et al., 1977).
Comments
The Committee evaluated an additional long-term study in rats of
aspartame and diketopiperazine impurity and further biochemical
studies of aspartame in man. It appears that the increased urinary
excretion of calcium as well as the nephrocalcinosis are due probably
to the consequence of a protein overload induced by the high intake of
aspartame (2-4% in the diet). In accordance with other reported
studies, the rat, especially the female, was more prone to the
development of both functional and anatomical renal changes
attributable to slight imbalance in calcium metabolism. Since neither
hypercalciuria nor nephrocalcinosis was observed in mice or dogs with
chronic administration of aspartame, the effect in the rat would
appear to be species and sex specific. The probability of human renal
changes due to aspartame consumption within the limits of the proposed
ADI (40 mg/kg) would appear to be remote since this amount would not
significantly increase the daily amino acid or protein intake. The
incidence of brain tumours between the control and treated groups
was comparable. It was concluded that neither aspartame or
diketopiperazine caused brain tumours in rats in this study.
The evidence available to the Committee when it established an
ADI for aspartame at its twenty-fourth meeting was substantial. The
additional data summarized in this working paper serves to confirm the
previously established ADI.
EVALUATION
Estimated level causing no toxicological effect in the rat
Aspartame: 4 g/kg bw
Diketo piperazine: 750 mg/kg bw
Estimate of acceptable daily intake for man
Aspartame: 40 mg/kg bw
Diketo piperazine: 7.5 mg/kg bw
REFERENCES
Ishii, H. (1981) Incidence of brain tumours in rats fed aspartame,
Toxicology Letters (In press)
Ishii, H. et al. (1981) Toxicity of aspartame and its
diketo-piperazine for Wistar rats in dietary administration for
104 weeks, Toxicology (In press)
Koch, R. & Blaskovics, M. (1978) Effect of aspartame on plasma and red
cell amino acids of apparently healthy female adults and on
presume phenylketonuric heterozygotes. Unpublished report from
the Departments of Pediatrics at the Children's Hospital of Los
Angeles, and the University of Southern California School of
Medicine, Los Angeles, submitted to WHO by G. D. Searle & Co.
Stegink, L. D. et al. (1977) Effect of aspartame loading upon plasma
and erythrocyte free amino acid levels and blood methanol levels
in normal one-year-old children. Unpublished report from the
Departments of Pediatrics and Biochemistry, University of Iowa
College of Medicine, Iowa City, Submitted to WHO by G. D. Searle
& Co.
Stegink, L. D. et al. (1978) Effect of aspartame loading at
100 mg/kg body weight upon plasma and erythrocyte levels of free
amino acids in normal subject and subjects presumed to be
heterozygous for phenylketonuria. Unpublished report from the
University of Iowa College of Medicine, Iowa City, submitted to
WHO by G. D. Searle & Co.