CALCIUM CYCLAMATE, SODIUM CYCLAMATE AND CYCLOHEXYLAMINE
Explanation
Calcium cyclamate, sodium cyclamate and cyclohexylamine (CHA)
were previously evaluated at the Joint Meeting of the FAO/WHO Expert
Committee on Food Additives in 1967, 1970, 1976, 1977 and 1980 (see
Annex I, Refs. 14, 22, 40, 43 and 54) and the temporary ADI for
cyclamate of 4 mg/kg bw was extended pending completion of a
reproduction study and a conversion study. Since the previous
evaluation, additional data have become available and are summarized
in this report. Cyclohexylamine (CHA) is the principal metabolite of
cyclamate.
CALCIUM CYCLAMATE AND SODIUM CYCLAMATE
BIOLOGICAL DATA
TOXICOLOGICAL STUDIES
Special studies on mutagenicity
Groups of 2 men and 2 women consisting of subjects receiving no
cyclamate, subjects who did not convert cyclamate to cyclohexylamine,
and those who did convert cyclamate to cyclohexylamine were studied.
In the groups receiving cyclamate, the subjects were given sodium
cyclamate 3 times a day for 4 days at total daily doses of 5 g for men
and 4 g for women. Blood was collected on the first and fifth days of
the study for chromosome analysis. All chromosome values were within
normal limits prior to and after treatment (Dick et al., 1974).
A rat study was conducted. The group distribution, dosages and
administration routes are presented in the table that follows.
The cyclohexylamine treated animals were dosed daily for 5 days
and positive control dosed daily for 2 days. On the day following the
last dose, animals were killed, femurs removed, and bone marrow cells
collected for chromosome analysis. There was no apparent increase in
chromosomal abnormalities in treated, when compared to controls (Dick
et al., 1974).
Four groups of mice each containing 150-160 animals were used.
One group was untreated while the other 3 received by gavage 300 mg/kg
of methylurea and 15 mg/kg of sodium nitrite. One of the test groups
received 1.9 g/kg cyclamate in their diet. Treatment continued for
1 week and then males placed on a normal basal diet for 3 weeks. Each
male was mated for 4 days with 3 females 1 and 3 weeks after
treatment. About 13 days after mating, the females were killed and
scored for dead and living implants. The males of the test groups
received 0.18% or 3.6% methylurea in their diet for 3 months after
mating. One of the test groups also received 1.4% sodium cyclamate in
the diet. At termination, animals were killed and the bladders,
livers, and lungs were removed for histopathological examination.
A significant increase of dominant lethals was observed in groups
treated with sodium nitrite and methylurea but were only slightly
increased with sodium cyclamate. No histopathological changes in the
bladder, liver or lungs relating to sodium cyclamate were observed.
Observed lesions were related to the intake of sodium nitrite and
methylurea (Aeschbacher et al., 1979).
Cyclamate and cyclohexylamine were assayed for their mutagenic
activity using the Ames test with and without microsomal activation.
Salmonella tester strains TA 1535, TA 100, TA 1537, and TA 98 were
used. Neither chemical was mutagenic under the conditions of this
assay (Herbold & Lorke, 1980).
Special studies on combined reproduction and carcinogenicity
Groups of 5-7 females were treated orally by gavage with
cyclamate at doses of 0, 0.2, 1 or 5 g/kg bw on days 14, 17 and 20 of
pregnancy and observed for life. Complete necropsies were performed.
All urinary bladders and organs with macroscopic abnormalities were
examined histologically. There were no treatment-related adverse
effects. Tumour incidence in treated animals was comparable with
controls. In particular, no carcinoma of the urinary bladder was
observed (Schmähl & Habs, 1980).
Special studies on combined reproduction, embryotoxicity,
teratogenicity and carcinogenicity
A six-generation experiment with SPF Swiss mice was carried out
to examine the short-term (reproductive, perinatal, teratogenic) and
long-term (carcinogenic and toxicological) effects of sodium
cyclamate, saccharin (as free acid) and cyclohexylamine. The following
dietary treatments were used:
(1) Control
(2) 2% sodium cyclamate
(3) 5% sodium cyclamate
(4) 0.2% saccharin
(5) 0.5% saccharin
INCIDENCE OF BONE MARROW CHROMOSOME ABNORMALITIES IN RATS
Average %
Expt. Group Number Compound Dose Administration Number of Gaps and Reunion Total
of rats Route metaphases breaks figures and
examined fragmented
metaphases
I. 1 12 CHA.HCl 50 mg/kg i.p. 1 200 2.1 2.1
2 12 CHA.HCl 50 mg/kg p.o. 1 200 2.4 2.4
3 12 Control i.p. 1 200 3.2 3.2
(water)
II. 1 17 CHA.HCl 68.1 mg/kga i.p. 850 1.8 1.8
2 14 CHA base 50 mg/kg i.p. 700 1.3 1.3
3 12 Control i.p. 600 2.8 2.8
(water)
4 10 TEMb 0.5 mg/kg i.p. 470 3.0 8.9 11.9
5 8 METEPAb 20 mg/kg i.p. 400 12.0 22.0 34.0
a CHA.HCl was given at 68.1 mg/kg to be equimolar with 50 mg/kg CHA base.
b Positive controls were dosed for 2 days; all other groups were dosed for 5 days.
CHA: cyclohexylamine; CHA.HCl: cyclohexylamine hydrochloride; TEM: triethylenemelamine;
METEPA: tris-(2-methyl-1-aziridinyl)phosphine oxide
(6) 2% sodium cyclamate + 0.2% saccharin
(7) 5% sodium cyclamate + 0.5% saccharin
(8) 0.5% cyclohexylamine
The P generation consisted of 50 males and 50 females per group.
Five weeks after the start of the study, 20 females and 10 males per
group were selected to produce the F1 litter. The litter size was
inadequate and the remaining 30 females were mated with 15 males to
produce the F1a' generation. Subsequently, 20 females and 10 males
were chosen to produce litters for each succeeding generation. The P,
F3b and F6a generation consisted of 50 males and 50 females and were
used to study carcinogenicity. The remaining generations consisted of
10 males and 20 females. Reproduction was studied in generations F1a,
F2b, F3a, F4a, F5a. Perinatal studies were conducted with the F1a',
F2b, F4a, and F5b generations. A teratogenic study was undertaken
with the F6b generation.
In 5 short-term experiments of the subsequent generations, sodium
cyclamate and saccharin, or its combinations, did not have any toxic,
carcinogenic, embryotoxic or teratogenic effect in the dosages used.
Cyclohexylamine administration led to growth retardation and embryonal
death, when given in 0.5% concentration in the diet.
In 3 long-term studies performed with the P generation, the F3b
and F6a generations only 7 bladder tumours were seen in a total
number of 2400 animals. The tumours were divided among the different
groups, including the control groups (Kroes et al., 1977).
Short-term studies
Monkey
Out of 12 monkeys fed 100 mg/kg and 11 monkeys fed 500 mg/kg
cyclamate 5 days a week for an average of 94 and 95 months, 10 and 9
respectively survived and did not exhibit any evidence of toxicity.
Two monkeys at each dose level died. The 2 monkeys at 100 mg/kg died
within 4 months and necropsy findings showed fatty degeneration of the
kidney. One of the deaths at 500 mg/kg was accidental, the other died
after 84 months with necropsy showing vacuolar degeneration of the
kidney. During the 16-year existence of this colony only 4 monkeys
have developed spontaneous neoplasms out of a total of 211 control
monkeys. Three tumours were lymphomas, the fourth a carcinoma of the
gall bladder (Sieber & Adamson, 1978).
Long-term studies
Mouse
Groups of 30 males and 30 females were fed diets containing 0,
0,7, 1.75, 3.5 or 7.0% sodium cyclamate for 80 weeks. A slight
reduction in body weight gain was observed in female rats at the 1.75
and 3.5% cyclamate levels in the diet during the last 6 months of the
study. Haemoglobin concentrations of both sexes fed 7% cyclamate were
significantly lower than in the controls at week 80. No apparent
differences between controls and treated mice with respect to
mortality, organ weights, incidence of histological changes or tumours
were noted. The no-effect level in the diet of mice under the
conditions of this study was 3.5% (approximately 5 g/kg bw per day)
(Brantom & Gaunt, 1973).
Rat
Groups of 50 female rats whose bladders had been instilled with a
freshly prepared solution containing 2 mg of N-methyl-N-nitrosourea
(MNU) were fed 0 or 2% sodium cyclamate which was increased to 4%
after 10 weeks for a period of 21 years. A normal basal diet was given
to 100 untreated controls. There was no significant effect on body
weight gain, food or water consumption between groups. Animals that
survived more than 10 weeks after the administration of MNU exhibited
a wide range of pathological changes in the urinary tract such as
necroses, haemorrhages and calcium deposits in the collecting ducts of
the renal papillae, dysplasia, and focal hyperplasia of the papillary
and pelvic urothelium. Neoplasms of the urinary tract were observed
with equal frequency between the MNU treated control and cyclamate
treated group. No difference in the latency periods of the tumours was
found between the MNU treated groups (Green et al., 1980).
Groups of rats untreated and treated with N-methyl-N-nitrosourea
(a single intravesicular dose of 2 mg) were fed diets containing 0 or
2 g/kg bw per day of sodium cyclamate for a lifetime or until killed
because of a palpable tumour. The incidence of bladder tumours was as
follows:
Treatment No. of No. of animals %
animals with tumours Incidence
Nil 98 0 0
MNU 124 0 0
Cyclamate 228 3 1.3
Cyclamate & MNU 54 31 57.5
Under the conditions of this study, sodium cyclamate was found to
promote MNU-induced bladder tumours at the applied high dose level.
OBSERVATIONS IN MAN
Four groups of 8 men each were administered sodium cyclamate in
hard gelatin capsules at 0 (sucrose placebo), 5, 10 or 16 g/day for
213 days. Capsules were taken with regular meals. Blood, urine and
semen were collected twice prior to commencing study. Samples of blood
and urine were then collected on days 2 and 8 and weekly thereafter
while semen was obtained every 2 weeks. Blood was examined for
glucose, BUN, cholesterol, Na+, K+, protein bound iodine (PBI),
SGPT, haemoglobin, haematocrit and complete blood count. The
concentration and motility of sperm in the semen were estimated. Stool
was collected on days 2, 4, 9 and 11 and on 2 non-consecutive days and
graded as to consistency. Body weight and blood pressure were
monitored weekly.
The only untoward effect observed was softening of the stool
which occurred in 7/8 men receiving 16 g/day and 2/8 receiving
10 g/day. All subjects receiving cyclamate excreted cyclohexylamine in
varying amounts on one or more occasions. The percentage of
cyclohexylamine that appeared in the urine as free cyclohexylamine
varied between 0.25 and 75.4% with a mean value of 17.2%. A
significant increase in the concentration of PBI in the serum was also
observed. This was shown to be an artifact arising from iodine in the
colour used in colouring the gelatin capsules (Wills et al., 1981).
CYCLOHEXYLAMINE
BIOLOGICAL DATA
TOXICOLOGICAL STUDIES
Special studies on mutagenicity
Two groups of 20 male mice (30-35 g) received either
150 mg/kg/day of CHS orally as a 0.6% solution in demineralized water
or 25 ml water/kg/day on 5 successive days. Males were then mated with
3 untreated females. Each week 3 other untreated females were placed
for mating with each male for a total of 8 mating periods. On the
14th day of gestation the uterus was examined for pro- and post-
implantation losses. No apparent effect on the mating capacity or
fertility of the treated males was observed. Pre- and post-
implantation losses of the treated group were comparable to the
control. Thus under the conditions of this study there would appear to
be no dominant lethal effects (Lorke & Machemer, 1974).
A group of Chinese hamsters, 13 males and 7 females, weighing
between 30-40 g were given a dose of 200 mg/kg/day of CHA by gavage
for 3 successive days. Blood samples for lymphocyte culture were taken
from the orbital plexus prior to and 4 days after treatment. A
significant increase in structural aberrations of the chromosomes
(ring chromosome, exchange figures, fragments and breaks) were
observed. There was no significant increase in aneuploid or polyploid
cells (Van Went-de Vries et al., 1975).
Special studies on reproduction
Two separate studies were conducted. In the first study rats were
administered orally cyclohexylamine sulfate (CHS) as follows: groups
of 5 males and 15 females received 22.26 and 44.52 mg/kg/day and 8
males and 25 females received 89.04 mg/kg/day. A control group of 7
males and 17 females received distilled water. Survival of progeny at
birth and 24 hours post-partum was noted. Viable members of each
litter at 2 days of age were weighed and killed for skeletal visceral
and chromosomal studies. During the second cycle females were allowed
to deliver normally and at 2 days post-partum each litter was randomly
reduced to 8 pups and their body weight gains were measured until
weaning (21 days). Pups and mothers were then sacrificed. In the
second study 2 groups of 10 males and 10 females each were
administered 44.52 or 89.04 mg/kg/day of CHS. A group of 12 males and
12 females received 178.08 mg/kg/day and a group of 13 males and 13
females were used as controls. Females showing pregnancy were
sacrificed after 11-15 days of separation from males. The number of
viable embryos and resorption sites in the uteri were counted. The
rats which were discontinued from the breeding studies were bled from
the abdominal aorta 24, 57, 77, 101 and 119 days after initiation of
study and within 6 hours of last administered dose. There was no
detectable effect on body weight gain. Blood analysis for CHS gave
evidence of its absorption. Male fertility appeared to be impaired.
However, there was no evidence of adverse effects on female fertility.
No deleterious effects were noted on embryo viability, litter size,
litter weight, postnatal viability, weight gain of pups or somatic
cell chromosomes of pups and dams (Khera et al., 1971).
Short-term studies
Rat
Groups of 100 male weanling Sprague-Dawley rats were randomly
assigned to 9 treatment groups. Animals were observed for a 1-week
baseline feeding period followed by a 15-day diet acclimation period
from which time the animals were fed 0, 50, 100, 200 and 300 mg/kg CHA
base admixed to basal diet. Four control groups were assigned for
pair-feeding with each test group. The animals in these control groups
received an amount of basal diet equal to that consumed by its pair-
fed partner in the test group during the previous week. The animals
were fed their respective diets for 90 days. Body weight gain and food
consumption were recorded weekly. On termination of the study necropsy
was performed on each animal. Testicular weights were recorded and
each testis was subjected to histological examination. The body
weights of the treated groups were found to be significantly less than
the body weights of the untreated controls. When compared to their
corresponding pair-fed control the body weights of the treated animals
were still significantly less except for those receiving 50 mg/kg bw
of CHA. The mean percentage decrease, however, was only significant
for the 200 and 300 mg/kg bw groups.
In treated rats both right and left testicular weights were found
to be significantly less than in the untreated controls. Analysis
comparing treated to corresponding pair-fed group showed a significant
difference only at the highest dose group (300 mg/kg bw). The only
compound-related changes noted at autopsy were in relation to the
testes of the highest dose group (atrophy, shrinking, soft
consistency, livid, glassy).
At termination of the study, 3 cross-sections were prepared from
both left and right testicle of each rat. Each of these 6 sections
were evaluated histopathologically. The extent of tubular alteration
was scored as follows:
0 - tubular alterations not discernible;
1+ - up to 5% of tubules within section affected;
2+ - 6-20% of tubules within section affected;
3+ - 21-60% of tubules within section affected;
4+ - 61% or more of tubules within section affected.
For each rat, the 3 left scores were averaged and the 3 right scores
averaged to reduce each rat's testicular data to a pair of
observations. The left and right average scores were analysed
separately. The average left and right testicular scores for the 2
highest test groups were significantly higher than the untreated
control group. When each of the treated groups was compared to its
pair-fed control, once again, the only treatment groups showing a
significantly different average testicular score for both left and
right testis were the 200 and 300 mg/kg groups (Brune & Mohr, 1978).
Two groups of 15 male rats, 35 days of age, were administered
CHA as a suspension in corn oil by gavage at dose levels of 0 or
200 mg/kg/day. Five rats from each group were killed after weeks 4 and
9 of treatment, the remaining 5 rats/group were maintained untreated
for 13 weeks. Body weight gain, food and water intake were recorded
weekly. Serum follicle stimulating hormone (FSH), luteinizing hormone
(LH) and testosterone levels were determined after 1 day, 4 and 9
weeks of treatment and after the 13 week withdrawal period. Terminal
examination included autopsy. Pituitary glands, testes (including
epididymides), prostate and seminal vesicles were weighed and
preserved for histological examination. A quantitative assessment of
spermatogenesis was performed.
Body weight and food consumption were reduced. FSH levels
increased and testosterone levels decreased in the treated rats. One
rat examined after 13 weeks withdrawal of CHA showed bilateral
testicular atrophy. There were no statistically significant effects on
the weights of the pituitaries, testes or secondary sex organs of the
treated animals. Quantitative assessment of testicular spermatogenesis
showed a significant reduction in the number of late spermatids in the
treated group both during and after withdrawal periods. After 13 weeks
withdrawal of CHA, rats were found to have a reduced number of
spermatocytes and early spermatids (James et al., 1981).
Dog
Four sexually mature male Beagle dogs were administered CHA
suspended in corn oil by gavage. Initially the CHA suspension was
administered at 75 mg/kg on days 1 and 2, 75 mg/kg twice daily on days
3 and 4, 150 mg/kg on days 5 and 6 and then 150 mg/kg twice daily on
subsequent days. After day 9 the dose was reduced to 125 mg/kg twice
daily to avoid appetite suppression. Two dogs were killed after week
9 of treatment and the two remaining dogs maintained untreated for
13 weeks. Body weight gain was recorded weekly. Food and water
consumption were measured daily. Testicular measurements and semen
examinations were made prior to and after 2, 4 and 8 weeks of
treatment and again during withdrawal period. Serum luteinizing
hormone (LH) and testosterone concentrations were determined prior to
and after 1, 2, 4 and 8 weeks of treatment and during weeks 4, 8 and
12 of withdrawal period. At termination full macroscopic examinations
were performed. Pituitary glands, testes and prostate were weighed and
preserved for histologic examination. A quantitative assessment of
spermatogenesis was performed.
Body weight gain and food consumption were reduced during
treatment; however, both returned to normal during the withdrawal
period. Serum LH and testosterone concentrations were not affected.
Sperm count was decreased and a significantly increased percentage of
abnormal spermatozoa was observed, both parameters returning to normal
after a 12-week withdrawal period. Weights of pituitary glands, testes
and secondary sex organs of treated rats were not affected. CHA
administration caused a reduction in the counts of spermatocytes and
early and later spermatids after 9 weeks of treatment. These counts
exhibited remarkable improvement at the end of the recovery period
(James et al., 1981).
Long-term studies
Mouse
Groups of 48 male and 50 female weanling mice were fed diets
containing 0, 300, 1000 or 3000 ppm (0, 0.03, 0.1 or 0.3%)
cyclohexylamine hydrochloride for 80 weeks. There were no apparent
adverse effects on general health, behaviour, body weight gain, food
or water consumption, haematology or in the incidence of tumours.
However, there was a slight increase in some minor histological
changes of the liver at 3000 ppm (0.3%). The no-effect level of
CHA.HCl was found to be 1000 ppm (0.1%) (Hardy et al., 1976).
Rat
Groups of 48 males and 48 females were fed a diet containing 0,
600, 2000 or 6000 ppm (0, 0.06, 0.2 or 0.6%) cyclohexylamine
hydrochloride (CHA.HCl) for 104 weeks. A dose-related decrease in body
weight gain, food consumption, and water intake was observed.
Haemoglobin concentration of females given 6000 ppm (0.6%) CHA.HCl was
reduced during the first 52 weeks of the study. There were, however,
no apparent effects later in the study or in the males. Total
leucocyte count was reduced in males after 104 weeks. Blood urea (BUN)
concentrations were significantly reduced in males at all dose levels
while serum albumin concentrations were increased at the 2 higher dose
levels. At 6000 ppm (0.6%) CHA.HCl, rats produced a more dilute urine.
The relative organ weights of liver, spleen, and kidneys in males fed
the diet containing 6000 ppm (0.6%) CHA.HCl were lower than the
control values and the relative weight of the thyroids in females was
reduced at 2000 and 6000 ppm (0.2% and 0.6%) CHA.HCl. There was no
indication of a tumorigenic effect at any treatment level.
Testicular changes, such as atrophy of the tubules with few
spermatids, were noted in rats given 2000 or 6000 ppm (0.2 or 0.6%)
CHA.HCl. The changes, other than histological effects observed in the
testes, were considered explicable in terms of the changes resulting
from a lowered body weight gain and reduced food consumption. The
no-effect level was 600 ppm (0.06%) on the basis of effects on the
testes (Gaunt et al., 1976).
Comments
The evaluation of recent studies permitted the change in the
previously established ADI. The temporary status of the ADI maintained
at the twenty-fourth meeting of the Joint Committee in 1980 is removed
and the ADI increased.
EVALUATION
Level causing no toxicological effect
Rat: 100 mg/kg bw in the diet.
Estimate of acceptable daily intake for man*
0-11 mg/kg bw as calcium and sodium salts, expressed as cyclamic acid.
* Calculation of ADI
(1) approximately 37% cyclamate absorbed. 63% available for
conversion to CHA by intestinal flora. Absorbed cyclamate is not
metabolized.
(2) human conversion rate of cyclamate to cyclohexylamine - 30%.
(3) mol wt. cyclamate = 2
mol wt. cyclohexylamine
(4) no-effect level for cyclohexylamine = 100 mg/kg bw
(5) no-effect level for cyclamate
x = 100 × 2 = 200 = 1058
0.63 × 0.3 0.189
(6) ADI for cyclamate = NOEL = 1058 = 10.6 mg/kg
Safety factor 100 bw per day
REFERENCES
Aeschbacher, H. U. et al. (1979) Effect of simultaneous administration
of saccharin or cyclamate and nitrosamide (MNU) on bladder
epithelium and the dominant lethal test, Toxicology Letters,
3, 273
Brantom, P. G. & Gaunt, I. F. (1973) Long-term toxicity of sodium
cyclamate in mice, Fd. Cosmet Toxicol., 11, 735-746
Brune, H. & Mohr, U. (1978) Establishment of the no-effect dosage of
cyclohexylamine hydrochloride in male Sprague-Dawley rats with
respect to growth and testicular atrophy. Unpublished report from
Abbott Laboratories, Ltd.
Dick, C. E. et al. (1974) Cyclamate and cyclohexylamine: Lack of
effect on the chromosomes of man and rats in vivo, Mutation
Res., 26, 199-203
Gaunt, I. F. et al. (1976) Long-term toxicity of cyclohexylamine
hydrochloride in the rat, Fd. Cosmet. Toxicol., 14, 255-267
Green, U. et al. (1980) Syncarcinogenic action of saccharin or sodium
cyclamate in the induction of bladder tumours in MNU-pretreated
rats, Fd. Cosmet. Toxicol., 18, 575-580
Hardy, J. et al. (1976) Long-term toxicity of cyclohexylamine
hydrochloride in mice, Fd. Cosmet. Toxicol., 14, 269-276
Herbold, B. A. & Lorke, D. (1980) On the mutagenicity of artificial
sweeteners and their main impurities examined in the Salmonella/
microsome test, Mutation Res., 74, 155-156
Hicks, R. M., Wakefield, J. St J. & Chowaniec, J. (1975) Evaluation of
a new model to detect bladder carcinogen or co-carcinogens,
results obtained with saccharin, cyclamate and cyclophosphamide,
Chem.-Biol. Interactions, 11, 225-233
James, R. W., Heywood, R. & Crook, D. (1981) Testicular responses of
rats and dogs to cyclohexylamine overdosage, Fd. Cosmet.
Toxicol., 19, 291-296
Khera, K. S. et al. (1971) Reproduction study in rats orally treated
with cyclohexylamine sulfate, Toxicol. Appl. Pharmacol., 18,
263-268
Kroes, R., Peters, P. W. J., Berkvens, Johanna M., Verschuuren, H. G.,
De Vries, Th. & Van Esch, G. J. (1977) Long-term toxicity and
reproduction study (including a teratogenicity study) with
cyclamate, saccharin and cyclohexamine, Toxicol., 8, 285-300
Lorke, D. & Machemer, L. (1974) Investigation of cyclohexylamine
sulfate for dominant lethal effects in the mouse, Toxicol.,
2, 231-237
Schmähl, D. & Habs, M. (1980) Absence of carcinogenic response to
cyclamate and saccharin in Sprague-Dawley rats, Drug Res.,
30, 1905-1906
Sieber, S. M. & Adamson, R. H. (1978) Long-term studies on the
potential carcinogenicity of artificial sweeteners in non-human
primates. Proc. ERGOB Conf. Health and sugar substitutes, Geneva,
266 pp.
Van Went-de Vries, G. F. et al. (1975) In vivo chromosome damaging
effect of cyclohexylamine in the Chinese hamster, Fd. Cosmet.
Toxicol., 13, 415-418
Wills, J. H., Serrone, D. M. & Coulston, F. (1981) A 7-month study of
ingestion of sodium cyclamate by human volunteers, Regulatory
Toxicol. Pharmacol., 1, 163-176