SACCHARIN, CALCIUM, POTASSIUM AND SODIUM SALTS*
Saccharin was evaluated by the Joint Expert Committee on Food
Additives in 1967, 1974, 1977, 1980 and 1982 (Annex I, Refs. 14, 15,
35, 44, 53, 59, and 60). In 1977, the Committee changed the ADI from
5 mg/kg to a temporary ADI of 2.5 mg/kg and withdrew the conditional
ADI of 15 mg/kg for dietetic purposes only. The decision to reduce the
ADI and to restrict the use of saccharin was based primarily on the
results of animals studies which indicated that excessive and
long-term ingestion of saccharin was potentially a carcinogenic hazard
for humans. At the 1980 and 1982 meetings, the temporary ADI of
2.5 mg/kg was extended pending the completion of current
investigations, including a long-term feeding study in rats and
epidemiological studies. Since the previous evaluation, additional
data has become available and is summarized and discussed in the
following monograph addendum.
Previous studies have shown that the absorption of ingested
saccharin in animals and man is rapid and this is confirmed by
observations that the peak plasma concentration occurs soon after oral
administration to rats (Matthews et al., 1973; Sweatman & Renwick,
1980) and to man (Colburn et al., 1981; Sweatman et al., 1981;
Pantarotto et al., 1981a,b). The presence of food in the gut was
associated with a reduced initial peak plasma concentration in animals
(Matthews et al., 1973; Sweatman & Renwick 1980) and in man (Sweatman
et al., 1981).
Recent studies on the distribution of saccharin have given
attention to the nature and amounts of radioactivity in the bladder
tissue after administration of radiolabelled saccharin and on
concentrations in this tissue during chronic intake.
* Monograph addendum
In a recent two-generation study (Sweatman & Renwick, 1982) using
3H-saccharin, it was shown that there was a slower decrease in the
saccharin content of fetal tissues than of maternal tissue, and in
particular, the concentration of saccharin in the fetal bladder wall
decreased relatively slowly during a 48h period following a single
oral dose to the dam. Despite this, the steady state concentration of
saccharin in the liver and kidneys of fetuses from mothers fed a 5%
saccharin diet were lower than the maternal values while the
concentrations in the fetal bladder were similar or slightly higher.
It was concluded that there was no evidence of excessive accumulation
in the bladder wall or other tissues of male rats during in utero
exposure or during lactation which could explain sex and generation
specificity of the tumorigenic response.
The saturation of renal tubular secretion of saccharin in rats
fed high dietary levels was previously demonstrated by comparison of
plasma concentrations following intravenous infusion and chronic
dietary intake (Sweatman & Renwick, 1980) and, more recently, Sims and
Renwick (1983) found a marked decrease in renal clearance rates in
rats with high plasma concentrations (200-300 µg/ml) of saccharin.
Effects on enzymes and other biochemical parameters
Sodium saccharin, at concentrations similar to those in urine of
rats fed 1-5% sodium saccharin in their diet, markedly inhibited
urease and the proteases pepsin, thermolysin and papain (Lok et al.,
1982) and trypsin (Sims & Renwick, 1983a). Inhibition of proteolysis
in vivo was the probable cause of the high levels of protein and
tryptophan in the caeca of rats fed saccharin-containing diets (Sims &
Renwick, 1983). In this latter study, the metabolism of tryptophan by
the caecal bacteria was altered with increase degradation to indole
and indolelactic acid.
Increased metabolism of protein to tryptophan and indole in the
caecum occurred throughout a two-generation rat feeding study and the
lactating dams showed increased excretion of indican (the main urinary
metabolite of indole) via the milk; the pups also showed caecal
enlargements, increased protein and tryptophan in the caecum and an
increased excretion of indican immediately on weaning on to a
saccharing-containing diet. These changes persisted throughout life
as seen by subsequent analysis of urine from rats in the IRDC
carcinogenicity study (see special studies on carcinogenicity) at 13,
18, 24, 28 m (Renwick, 1983).
In a study of 15 human volunteers (Renwick, 1983) administration
of saccharin (1 g/d for 1 month) did not significantly increase the
daily excretion of indican in urine compared with the pre- and
post-administration control periods.
Special studies on renal function
Renal slices from rats fed diets containing 5% or 7.5% saccharin
showed a reduced accumulation of para-animohippurate (PAH) in vitro
but feeding these diets did not result in a reduced renal clearance of
PAH in vivo (Berndt et al., 1981).
The renal clearance of endogenous indican in saccharin treated
rats showed a highly significant inverse relationship to the plasma
concentration of saccharin (Sims & Renwick, 1983).
Special studies on urine volume and composition
Rats fed high dietary levels of saccharin showed an increase in
fluid intake and in urine volume which was accompanied by a decrease
in osmolality (Anderson, 1979; West & Jackson, 1981; Demers et al.,
1981; Berndt et al., 1981). Dose related decreased osmolality and
increased urine volume showed a strong correlation to the occurrence
of bladder tumours in the IRDC carcinogenicity study (see special
studies on carcinogenicity). These changes were primarily observed at
dietary concentrations of 3% saccharin and above. The increased daily
urinary volume was accompanied by both an increased volume per
micturition and an increased frequency of micturition. Saccharin-fed
rats showed a greater maximal distension of the urinary bladder
(Renwick & Sims, 1983) and these authors concluded that the increased
bladder distension would increase the possibility of interaction
between the epithelium and endogenous urinary metabolites, especially
during hours of daylight.
Administration of saccharin at a dose of 1 g/d to human
volunteers for 1 m did not affect urine volume when compared to pre-
and post-treatment control periods (Roberts & Renwick, unpublished
The effects of sodium saccharin on mineral and water balance and
a number of related parameters were studied over a 10-day period in
seven month old rats (Schoeing & Anderson, 1983). The study included
eight groups, each consisting of 10 males and 10 females. Rats in four
of the groups were from the second generation, the parental generation
having been exposed to dietary concentrations of 1.0, 3.0, 5.0, or
7.5% saccharin prior to and during gestation and lactation; the second
generation weanlings received the corresponding diets. The treatment
in two other groups was modified so that rats in one group were
exposed only in utero (via dams fed diets containing 5% sodium
saccharin) while exposure of the second group was started at birth
(via lactation dams fed similar diets) and continued at a dietary
saccharin concentration of 5%. The purpose of these modifications was
to evaluate the role of in utero exposure on the study parameters. A
group of second generation rats fed diets containing 5% soldium
hippurate was included to evaluate the specificity of sodium saccharin
and/or the effect of sodium ion on the study parameter. A group of
untreated animals served as controls. At dietary sodium saccharin
concentrations >1%, increases in water consumption and urine volume
were noted. At dietary concentrations >3.0% decreased urine
osmolality, changes in water and mineral balance, increased mass of
the caecum and bladder, and increases in bladder tissue mineral
concentrations were observed; the latter effect was noted only in male
rats. The evaluation of these parameters in rats with and without
in utero exposure indicated that in utero exposure played little
or no role in the occurrence or severity of these changes.
Qualitatively similar, but quantitatively less severe changes were
observed in rats fed sodium hippurate.
Special studies on caecal enlargement and stool hydration
In an attempt to determine why dietary sodium saccharin causes
caecal enlargement and increased stool hydration, Anderson (1983)
analyzed stools from rats fed diets containing 1, 3, 5, or 7.5%
saccharin. Saccharin ingestion resulted in a small increase in stool
ash but no change in lipid or non-saccharin nitrogen concentrations
(mg/g dry stool). Saccharin treatment also resulted in a dose-
dependent increase in the stool content of a carbohydrate soluble in
1M-NaOH. The author suggested that the source of the stool
polysaccharide was either undigested dietary polysaccharide or a
product of intestinal microbial synthesis and that the hygroscopic
carbohydrate together with the high stool saccharin content caused
caecal enlargement and increased stool hydration. In this context,
Shibata et al., (1983) showed that a strain of Streptococcus
obtained from the rat caecum produced an extra-cellular hygroscopic
glucan when grown in the presence of sucrose.
Special studies on effects on the bladder epithelium
Sodium saccharin was fed to male F344 rats at dietary levels of
0, 0.1, 0.5, 1.0, 2.5 or 5% for 10 weeks. Food consumption and body
weight gain were similar in all groups and no gross signs of toxicity
were observed. Sodium saccharin induced a dose-dependent proliferation
of the urinary bladder mucosa as assessed by autoradiography and
scanning electron microscopy. All rats in all groups had cells with
ropy microridges and uniform microvilli; sodium saccharin at dietary
levels >1% increased the number and size of these foci. In addition,
pleomorphic microvilli were observed at the two highest dose levels
but not in controls (Murasaki & Cohen, 1981).
Strain and species differences in the response of the urinary
bladder to sodium saccharin were observed by Fukushima et al. (1983a).
Male ACI, Wistar, F344 and Sprague-Dawley rats were given a diet
containing 5% sodium saccharin for 52 weeks. In ACI rats, sodium
saccharin induced not only preneoplastic lesions but also bladder
tumours; in other strains it did not. The urinary bladder of ACI rats
had the most marked lesions under scanning electron microscopy, with
less marked changes in Wistar and F344 rats; Sprague-Dawley rats were
resistant to these changes.
Male F344 rats, B6C3F1 mice, Syrian hamsters and Hartley guinea
pigs were given 5% sodium saccharin in the diet for 20 weeks. Animals
from each group and respective controls were sacrificed at 0, 4, 8,
12, 6 and 20 weeks after commencement of feeding. The rats developed
urinary bladder lesions as detected by light and electron microscopy
and increased DNA synthesis of the urinary bladder epithelium was
detected by autoradiography. Mice, hamsters and guinea pigs were
resistant to sodium saccharin.
Male and female Sprague-Dawley rats eight weeks of age were
given saccharin according to the standard IRDC protocol and the
effects on endogenous mitotic activity in the bladder assessed by
autoradiographic measurement of the thymidine labelling index (Tsing,
1983). In contrast to earlier reports (Fukushima & Cohen, 1980;
Murasaki & Cohen, 1981) no treatment related effects on the thymidine
labelling index were observed; the index was higher in male than
female rats. The differences between the results of these different
studies may have been due to genetic differences between the strains
of rat used. Reitz et al. (1983) also reported no significant
differences in the thymidine labelling index between controls and
treated Fisher 344 rat pups exposed to 7.5% saccharin in the diet
in utero and subsequently up to 35 days. The thymidine labelling
index was determined 8 days and 35 days post partum.
The changes in membrane potential of the epithelium of the F344
rat bladder have been measured following treatment of the animals with
BBN or saccharin (Iamida et al., 1983). Dietary concentrations of O,
0.04, 0.2, 1 or 5% sodium saccharin were administered but only the
highest dose level caused a significantly higher membrane potential
than the control group.
Recently, El Gerzawi et al. (1982) obtained normal human bladder
tissue and studies the effects of N-methyl-N-nitroso-urea (MNU) and
saccharin on the histology of the epithelium in long-term explant
cultures. In MNU-treated cultures, a dose-response was observed.
Single doses of 1-100 µg MNU/ml induced a typical hyperplasia, however
the changes reverted to a single or double cell layer as seen in
controls. In contrast, after multiple doses of MNU the hyperplastic
changes persisted. The doses of MNU in the presence of saccharin
gave cellular changes similar to those seen with multiple doses of
MNU alone although the nuclei appeared more pleomorphic and
hyperchromatic. Continuous exposure of the explants to saccharin alone
did not result in any changes from the controls.
Special studies on food consumption patterns
Utilizing a protocol recommended for two-generation bioassays,
Reitz et al (1983) reported that when rats were given a diet
containing 7.5% sodium saccharin through gestation and lactation, the
very young animals received considerably more sodium saccharin on a
mg/kg body weight per day basis than did adults consuming the same
diet; the young animals suffered weight depression and early mortality
when exposed to a dietary concentration which was well tolerated by
adults. These authors concluded that failure to maintain the dose of
the MTDS of 5000-6000 mg/kg body weight per day throughout a
two-generation study would compromise its usefulness in the
formulation of human risk estimates.
Three groups of 5 Rhesus monkeys were used in a study designed to
determine the maximum amount of sodium saccharin which monkeys would
voluntarily consume and/or tolerate (Westland & Helton, 1983); sodium
saccharin was incorporated into the dry diet of one group and the
drinking water of a second group with the third group serving as
controls. An increasing dose regimen in which the concentration of
sodium saccharin in the diet or drinking water was doubled every three
days was employed. Concentrations between 0.125 and 8.0% were
evaluated in the diet, and between 0.015 and 0.48% in the drinking
water. Very little rejection of either diet or drinking water was
observed but severe diarrhoea precluded further treatment when the
concentration of sodium saccharin reached 8.0% (approximately
1600 mg/kg per day) in the diet and 0.48% (between 900 and 2400 mg/kg
per day) in the drinking water. An increase in fluid intake occurred
in monkeys in both treatment groups. In addition, an increase in urine
volume and a decrease in urine osmolality were seen in monkeys given
sodium saccharin in drinking water. No effects on body weight, food
consumption or urine pH were observed in either group. All animals
recovered rapidly after being returned to untreated diet or water.
Special studies on mutagenicity of impurities in saccharin
The known impurities in sodium saccharin produced by the Maumee
(Sherwin-Williams) and the Remsen-Fahlberg processes have been
tabulated, and saccharin produced by the former process was analysed
to determine the concentration and identities of impurities (Riggin
& Kinzer, 1983). Most of the contaminants were found to be derived
from the polythene bags in which the saccharin was packed.
N-methylsaccharin (0.15 ppm) and methyl anthranilate (0.05 ppm) were
the predominant impurities. The major polar impurities in the Maumee
product were identified as 5-, 6-, and 7-aminosaccharin which were
present at a combined level of approximately 150 ppm (Radford et al.,
1983); these polar metabolites occurred to only a minor extent in
Remsen-Fahlberg saccharin (Wolf & Voigt, 1983).
Recently, an evaluation of the mutagenic activity of a mixture
of the major polar impurities of Maumee saccharin and of 5- and
6-aminosaccharins was carried out using the Ames Salmonella assay;
these compounds were found to be non-mutagenic with or without
metabolic activation by S9-mix (Litton Bionetics, Inc., 1983,a,b,c).
Similarly, Riggin et al. (1983) examined the mutagenic activity of
solvent extracts of specific manufacturing lots of saccharin produced
by the Sherwin-Williams process. All the individual components
identified were found to be non-mutagenic in the Ames assay. A weak
mutagenic activity was associated with chloroform extracts from one
lot of saccharin representing less than 1.5 ppm of the sample but was
not attributable to a single component. The possibility of artefact
formation from solvent interaction with impurities could not be
ruled out. The authors concluded that these impurities are of no
toxicological significance in animal feeding studies.
Special studies mutagenicity of saccharin
Saccharin was classified as highly mutagenic in an in vivo
mammalian spot test for detection of genetic alterations in mouse
embryo pigment cells (Mahon & Dawson, 1982). Offspring heterozygous
for several coat-colour genes were exposed in utero by
administration of saccharin to the dams at intra-gastric doses of
0.075, 0.75, 1.5, 3.0, 5.0, or 7.5 g/kg body weight on days 8, 9, or
10 of pregnancy. The presence of colour spots on the cost of the
offspring was taken as indicative of expression of a recessive
phenotype resulting from alteration of loss of a wild-type allele from
a prospective pigment cell. The frequency of saccharin-treated mice
(all dose levels) with colour spots was 3.6% compared to a control
frequency of 0.9% (P = 1 × 10-6) but there was no significant
variation in frequency due to dose over the wide dose range used. The
lack of a proportional dose response was acknowledged to be unusual in
In contrast, Fahrig (1982) categorized saccharin (Remsen-
Fahlberg, containing 27 ppm OTS) as non-mutagenic in the mammalian
spot test. In this study, saccharin (1 g/kg body weight) was
administered by i.p. injection on day 10 of pregnancy. Only one spot
of genetic relevance was found among 701 saccharin treated offspring
and this did not differ from the spontaneous frequency. The effect of
1 g/OTS kg body weight given orally was also evaluated in three
replicate tests, only one of which gave statistically significant
positive results. A clear classification of OTS as mutagenic or
non-mutagenic in this test was not possible.
The genotoxic potential of sodium saccharin and of 1-naphthalene
sulphonic acid was evaluated in the rat hepatocyte unscheduled DNA
synthesis assay at concentrations ranging from 1 × 10-4 to 1 × 10-1
M. Both materials were toxic to hepatocyte cultures at 3.16 × 103 to
1 xz 10-1 M. Decreasing concentrations of sodium saccharin or
1-naphthalene sulphonic acid resulted in decreased toxicity and the
culture exposed to 1 × 10-3 to 1 × 10-4 M concentrations resembled
negative controls. Neither compound elicited significant unscheduled
DNA synthesis at any of the concentrations tested (Reitz & Medrala,
Special studies on carcinogenicity-promoting or co-carcinogenic effects
The promoting effects of sodium saccharin and of phenobarbital
(PB) on all organs of rats were studied after initiation with
N-mitrosomethylurea (NMU) (Tsuda et al., 1983). Three groups of 25
male F344 rats were given 20 mg NMU/kg body weight i.p. twice a week
for 4 weeks, then given a diet containing 0.05% PB or 5% sodium
saccharin for the next 32 weeks. The animals were then killed, a
complete necropsy performed and sections of all tissues were stained
(H&E) and examined microscopically. The results indicated that PB
promoted the induction of neoplastic or preneoplastic changes in the
thyroid and liver but that sodium saccharin acted exclusively as a
promoter in the urinary bladder. There were significant increases in
the incidence and number of PN-hyperplastic changes (P <0.01) in the
bladders of animals given NMU + sodium saccharin. No papillomas were
observed in the bladders of any rats in this study.
The possibility of sodium saccharin acting as a co-carcinogen was
studies by co-administration of sodium saccharin and N-(4-(5-nitro-2-
furyl)-2-thiazolyl)-formamide (FANFT) at dietary levels of 5% and
0.005% respectively to male Fisher rats for 2 years. The effects of
L-tryptophan (2% of the diet) were also studied alone or
co-administered with sodium saccharin. Five of the sixteen rats given
sodium saccharin plus FANFT developed bladder tumours whereas none of
the animals given FANFT, L-tryptophan or sodium saccharin alone,
sodium saccharin plus tryptophan or control diets, developed bladder
tumours. Two of the rats receiving only FANFT developed papillary and
nodular hyperplasia. The results were stated to indicated that sodium
saccharin had no-carcinogenic activity when given simultaneously with
FANFT (Murasaki & Cohen, 1983a).
Rats were fed diets containing 0 or 5% sodium saccharin
immediately, 2 weeks or 8 weeks after freeze ulceration of the urinary
bladder and the effects studied by light and scanning electron
microscopy and by measurement of the thymidine labelling index. Sodium
saccharin prolonged the regenerative hyperplastic changes following
ulceration and maintained an increased proliferative rate in the
epithelium. Delaying saccharin administration for 8 weeks after
ulceration still resulted in nodular and papillary lesions, surface
abnormalities detected by scanning electron microscopy and an
increased labelling index. These changes were thought to contribute to
the eventual induction of bladder neoplasma in rats fed sodium
saccharin following ulceration (Murasaki & Cohen, 1983b).
Recent studies by Nakanishi et al. (1982) were conducted to
determine if a 4 week pre-treatment with N-butyl-N-(4-hydroxybutyl)
nitrosamine (BBN) or N-2fluorenylacetamide (2-FAA) followed by 5%
saccharin or 0.05% phenobarbital (PB) in the diet for 32 weeks induced
or enhanced liver and/or bladder tumours. Male F344 rats were
pretreated with 0.02% 2-FAA or 0.01% BBN in the drinking water. The
results indicated that, while 2-FAA and BBN have tumour-initiating
effects in both the liver and urinary bladder, the promoting effects
of saccharin and PB are organ-specific. Similar results were reported
by Tsuda et al. (1983).
The effect of partial cystectomy on the occurrence of
pre-neoplastic lesions, papillary or nodular hyperplasia
(PN hyperpl.asia) of the bladder in male F344 rats was studied in an
experiment in which bladder carcinogens and promoters were given to
the rats after initiation with BBN. The bladder carcinogens tested
were N-ethyl-N-(4-hydroxybutyl) nitrosamine (EHBN) at a level of 0.01%
in drinking water or FANFT at a dietary level of 0.2%. The promoters
used were sodium saccharin (5%), sodium cyclamate (2.5%), or
DL-tryptophan (2%) in the diet. Partial cystectomy significantly
decreased the occurrence of PN hyperplasia in rats treated with EHBN
and tended to inhibit that in rats given saccharin or tryptophan i.e.,
partial cystectomy inhibited rather than enhanced the induction of PN
hyperplasia (Fukushima et al., 1982).
Fukushima et al. (1983b) studied the promoting effects of various
chemicals on bladder carcinogenesis in 22 groups of 30 males F344 rats
after initiation by BBN. The rats were initially given 0.01% BBN rats
in the drinking water for 4 weeks and then the test compounds in the
diet for 34 weeks. The test compounds used were: sodium saccharin (0.5
and 5%) sodium ascorbate (5%), calcium carbonate (5%), DL-tryptophan
(5%) allopurinol (0.02%), acetazolamide (0.35%), quercetin (5%),
sodium hippurate (5%) and vitamin D (0.002%). Each of these compounds
was given to two groups of 30 rats. Effects were judged by measuring
the formation of preneoplastic lesions - papillary or nodular
hyperplasia (PN hyperplasia) of the urinary bladder. Administration of
5% but not 0.5% sodium saccharin in the diet significantly increased
the incidence and extent of PN hyperplasia; sodium ascorbate,
DL-tryptophan and allopurinol also increased the extent of PN
hyperplasia but the other test compounds did not at the dietary
concentration used. The results with sodium saccharin and
DL-tryptophan were consistent with earlier findings and the results
with sodium ascorbate and allopurinol suggest that these compounds
have promoting activities in urinary bladder carcinogenesis in the
rat. No correlation was found between the extent of crystalluria and
promotion of pre-neoplastic lesions.
In a two-stage bladder carcinogenesis study, Ito et al. (1983a)
evaluated the promoting effects of 16 test chemicalsby their ability
to induce PN hyperplasia in F344 rats. Male rats were given 0.01% BBN
in drinking water for four weeks followed by one of the test compounds
for 32-34 weeks. The dose response of saccharin was also studied in
rats of both sexes at dietary concentrations of 0, 0.04, 0.2, 1.0 and
5% for 32 weeks after BBN treatment. Dose-response curves showed
enhanced hyperplasic responses in both sexes given 0.2 to 5%
saccharin. Ito et al. (1983a) also studied the organ specificities of
phenobarbital or saccharin after initiation with BBN or 2AAF; the
promoting effects were found to be organ specific. Similar findings
were reported in other studies (Nakanishi et al., 1982; Tsuda et al.,
A number of tumour promoters in the two-stage mouse-skin
carcinogenesis system are known to be reversible inhibitors of nerve
growth factor-induced neurite out-growth while their non-promoting
structural congeners are not. A 50 mM concentration of Maumee sodium
saccharin inhibited neurite out-growth; the inhibition was completely
and rapidly abolished by washing out the saccharin. Saccharin also
inhibited binding of 125I-nerve growth factor in embryonic chick
sensory ganglia cells in a concentration dependent manner (Ishii,
1982). It was postulated that alteration of cellular differentiation
by tumour promoters may result from interactions with receptor systems
that regulate cellular function.
Special studies on carcinogenicity
A two-generation carcinogenicity study has been performed with
the primary objective of investigating the dose-response curvre for
urinary bladder tumours in male Charles River CD rats. The study was
also designed to evaluate the role of in utero exposure, the
specificity of sodium saccharin and the role of excess sodium in the
occurrence of urinary bladder tumours.
First generation (Fo) parental animals were given diets
containing 0, 1.0, 3.0, 4.0, 5.0, 6.25 and 7.5% sodium saccharin from
approximately six weeks of age for approximately 4 1/2 months. During
this time, the animals were mated (one male to two females) commencing
with 110-114 days old and the females were allowed the nurse the
offspring for 21 days. When the second generation (F1) offspring were
between 28 and 38 days old, second generation male rats were randomly
selected from each treatment group for the chronic phase of the
bioassay. At this point, all Fo male and female rats, all F1 female
rats and F1 males not selected for the carcinogenicity study were
removed from the study. F1 males selected for the chronic phase were
maintained on the same diet which their parents received for a period
of up to 30 months.
Two further treatment groups were also included to study the
possible role in in utero exposure in the production of urinary
bladder tumours. The Fo animals in one group were fed sodium
saccharin at a dietary concentration of 5% only during mating and
gestation. After parturition, the dams were fed control diet and
selected F1 males were continued on the control diet for a period of
30 months. This group was designated "5% saccharin through gestation".
In the second group, Fo animals were maintained on control diet until
parturition after which the dams were placed on a sodium saccharin
test diet beginning at 1% and increasing to 5% during lactation.
Selected F1 males from these parents were fed sodium saccharin at a
dietary concentration of 5% for 30 months. This group was designated
"5% saccharin following gestation".
A third treatment group was included to study the effect of
excess sodium and to determine the specificity of high doses of
saccharin to the occurrence of bladder tumours. Animals in this group
were fed a diet containing 5% (reducing to 3%) sodium hippurate
through two generations. This compound was selected because of its
similarity to saccharin in physical, chemical and pharmacokinetic
properties, e.g. both are sodium salts of organic acids of almost
identical molecular weight and are filtered and actively secreted into
urine by the kidney.
Selected F1 males from each of the groups were allocated to the
second generation long-term study using the following unbalanced
F1 treatment group Number of F1 males
untreated control 350
1.0% sodium saccharin 700
3.0% sodium saccharin 500
4.0% sodium saccharin 200
5.0% sodium saccharin 125
6.25% sodium saccharin 125
7.5% sodium saccharin 125
5.0% sodium saccharin through gestation 125
5.0% sodium saccharin following gestation 125
5.0/3.0% sodium hippurate 125
The F1 animals were observed twice daily for signs of toxicity,
behavioural changes and survival. Individual body weights and food
consumption were measured weekly during the first 13 weeks and once
every two weeks thereafter. Urinanalyses were conducted on 30
rats/group on days 6, 30, 64, 92 and months 6, 13, 18, 24 and 29. The
urinanalysis included urinary pH, microscopic examination, bilirubin,
protein, glucose, ketones, urobilinogen, nitrite and occult blood.
Osmolality was also measured on fresh urine samples on day 6, 30 and
64, and on 24 h urine samples at each examination except day 6.
Individual 48 h water consumption was measured on the same 30 animals
per group for the first 13 weeks and approximately every two weeks
During the F1 phase, complete post-mortem examinations were
performed on all animals which died during the course of the study or
were sacrificed in extremis and on all terminally sacrificed animals.
A complete range of tissues was fixed for possible subsequent
examination and the urinary bladder, kidney, urethra and ureters of
all animals were examined histologically. Gross lesions and masses,
from all tissues, observed at autopsy were also examined
microscopically. Examinations of the fixed urinary bladder and
subsequent microscopic examinations of tissues were conducted in a
Results of the Fo generation and litters
The feeding of sodium saccharin at dietary concentrations up to
7.5% to male and female Fo rats from the post-weanling stage through
a single reproductive cycle had no effects on behaviour or survival.
However, at dietary concentrations <3.0% statistically significant
(P>0.05) body weight depressions were noted. The difference from the
untreated control group was as high as 11% in the 7.5% sodium
saccharin treatment group. The depressions in body weight were not due
to a decrease in food consumption or nutrient intake since the treated
rats compensated for the non-nutritive ingredient added to their diet
by consuming more total diet (g/rat/d) than untreated controls. At
dietary concentrations of sodium saccharin <3% there were significant
(P>0.05) reductions in the mean number of pups per litter. There was
a statistically significant increase in water intake and urine volumes
at dietary levels <1%; a decrease in urinary pH and visible increase
in the moisture content of faeces were observed at dietary
No effects on survival or behaviour were observed through the
first 28 to 38 days of life in the offspring receiving dietary
concentrations of sodium saccharin up to 7.5% and the mean body
weights at birth were also comparable to control animals. However,
statistically significant (P>0.05) body weight depressions were noted
in all sodium saccharin groups later during the lactation period and
in males at 28 days of age. The difference was noted only at days 21
and 28 in the 1.0% sodium saccharin group and was small (approximately
2%); by comparison, the average body weight depressions in the 3.0%
through 7.5% sodium saccharin groups were between 8% and 24% at day
21, and between 9% and 31% at day 28. The weanling rats (28-38 day
old) from the 5% and 7.5% sodium saccharin groups were found to be
The feeding of 5.0% sodium saccharin only during mating and
gestation caused statistically significant depressions (P<0.05) in
both food consumption and body weight of the Fo animals during the
treatment period. A statistically significant (P<0.05) decrease in
the mean number of pups born per litter was also observed which was
comparable to the decrease noted in the dose-response group in which
parental animals were fed 5.0% sodium saccharin continuously from the
weaning stage. No changes were observed in the offspring during the
first 28 to 38 days of life.
The feeding of sodium saccharin to lactating female rats on an
increasing dosage schedule of 1%, 3% and 5% during weeks 1, 2 and 3
post-parturition respectively had no observable effect until the
dietary concentration of sodium saccharin was increased to 3%. After
this time, the offspring showed statistically significant (P<0.05)
body weight depressions on days 14 and 21; the males also had lower
body weights than controls at day 28. The observed body weight
depressions were slightly less than in the dose-response group from
parents fed 5.0% sodium saccharin continuously but by day 28 the body
weights of the male rats in these two groups were similar.
The feeding of 5.0% sodium hippurate to Fo animals from weaning
through a single reproductive cycle caused statistically significant
(P<0.05) body weight depressions averaging up to 10% and 14% in male
and female rats, respectively. In females, the average depression was
only 4% until the beginning of gestation. Unlike the sodium saccharin
treated animals, Fo rats fed sodium hippurate consumed less diet than
controls. The feeding of sodium hippurate caused a statistically
significant (P<0.05) decrease in the mean number of pups per litter
at birth and aggressive behaviour and mortality in the lactating dams.
Teratogenic effects (microphthalmia, domed heads and hydrocephaly) and
statistically significant (P<0.05) depressions in body weight were
noted in the offspring. The body weight depressions were as high as
38% in 28 day-old males. Increased water intake and urine volume were
observed in the Fo rats but the changes were less than in the
corresponding 5% sodium saccharin group. Little or no visible changes
in the faecal moisture content were observed.
Results in the F1 generation
There were a statistically significant trend for increased
survival in the sodium saccharin treatment groups, being most evident
at the 5.0% and 7.5% treatment levels. The survival rates after 123
weeks of treatment were as shown:
Treatment group Survival rate (%)
1.0% sodium saccharin 24.5
3.0% sodium saccharin 23
4.0% sodium saccharin 19
5.0% sodium saccharin 37
6.25% sodium saccharin 26
7.5% sodium saccharin 34
5.0% sodium saccharin through gestation 20
5.0% sodium saccharin following gestation 36
3.0% sodium hippurate 30
No changes in behaviour or appearance were observed in the study.
A clear dose-response was observed for physiological effects at
treatment levels of 3.0% sodium saccharin or above. Changes such as
relative depressions in body weight, food consumption and water
consumption were seen but there was no direct statisticaly correlation
with the occurrence of bladder tumours. The 1.0% dietary
concentration was considered a no-effect level for these changes.
Urinanalysis revealed a dose-dependent increase in the mean 24 h
urine volume and decrease in osmolality. These changes were primarily
observed at dietary concentrations of sodium saccharin of >3.0% and
showed a strong statistical correlation to the occurrence of bladder
tumours. During the first 92 days, the urinary pH was significantly
depressed for all groups in the dose-response portion of the bioassay
in which sodium saccharin was fed at dietary concentrations >4%;
significantly lower urinary pH values were also recorded in the 3%
sodium saccharin group at the 64 and 92 day time intervals. No
dose-related lowering of pH was evident later in the study (6-29 m)
except for the 4% and 7.5% sodium saccharin groups at the 13 m time
interval. At no stage in the study was the pH of the 1% sodium
saccharin group significantly lower than control values. The urine pH
values in the 5.0% sodium saccharin through gestation group were
comparable to control values but significant depressions were noted in
the 5% sodium saccharin following gestation group at the 6-, 30-, 64-
and 92-day time intervals, and a statistically significant depression
was also observed at 24 m for the 3% sodium hippurate group. No
treatment-related changes were seen in the analyses for urinary
urobilinogen, protein, glucose, ketones, bilirubin, occult blood and
nitrite. Analysis of the urine for calcium ion was performed at the
24 m interval and no treatment-related effects on calcium ion
concentration were observed; however, due to the increased urine
volume, there was a treatment-related increase in total calcium ion
excreted (mg/24 h) in the groups receiving >3.0% sodium saccharin.
Examination of the urine for microcrystals showed that at 13, 18, and
24 m, the control group had significantly higher crystal scores than
the 7.5% sodium saccharin group, and crystalluria did not appear to be
involved in bladder tumour formation.
At autopsy, there was a significant increase in absolute and
relative bladder weights at sodium saccharin dietary concentrations of
3% or above. No urinary bladder weight changes were noted in animals
exposed to sodium saccharin only in utero nor in animals fed sodium
Histopathological examination revealed a treatment- and dose-
related mineralization of the renal pelvis but no treatment-related of
the ureter or urethra were observed.
A clear dose response for bladder tumours was observed. The slope
of the dose-response curve was steep indicating that the incidence of
bladder tumours declined rapidly with decrease in dose. The incidence
of bladder tumours observed in the various treatment groups was as
Treatment group Incidence of primary bladder tumours %
Benign Malignant Total
Control 0.0 0.0 0.0*
1.0% sodium saccharin 0.6 0.2 0.8
3.0% sodium saccharin 0.8 0.8 1.6
4.0% sodium saccharin 2.1 4.2 6.3
5.0% sodium saccharin 3.3 9.2 12.5
6.25% sodium saccharin 10.0 6.7 16.7
7.5% sodium saccharin 15.3 16.1 31.4
* Tumour incidence in 863 control male rats of the same used in
this study from 10 recent in utero lifetime studies conducted
at IRDC ranged between 0.0-2.5% for papillomas and 0.0-0.8% for
carcinomas. The mean incidence for total urinary bladder tumours
In this study, the lowest dosage level of 1.0% was considered a
no-effect level for bladder tumours based upon pairwise statistical
analyses with the concurrent untreated control group and a comparison
with background bladder tumour incidence for this strain of rat at the
IRDC laboratory utilizing an in utero lifetime design. At the 3.0%
sodium saccharin treatment level, the incidence of benign bladder
tumours alone or of malignant bladder tumours alone was not
significantly increased, but the combined incidence was significantly
higher than in concurrent controls. The incidence of benign and/or
malignant bladder tumours was significantly increased at dosage levels
of 4% or greater. No increase in the incidence of hyperplasia or other
treatment-related effects were observed in the ureter, urethra or
The animals exposed to saccharin only in utero were comparable
to controls but the animals whose exposures began at birth (5% sodium
saccharin following gestation) had an incidence of urinary bladder
tumours similar to that of animals fed diets containing 5% sodium
saccharin whose exposure included the in utero period. Therefore it
appeared that in utero exposure was not essential to the development
of urinary bladder tumours in sodium saccharin treated rats. No
bladder tumours were seen in the group fed sodium hippurate although
the incidence of kidney mineralization was similar to that in the 3.0%
sodium saccharin group.
No other treatment-related toxic or neoplastic lesions were
observed in either the genito-urinary tissues or in the various
tissues examined only in the event of a macroscopic lesion.
A detailed statistical analysis of the data from the IRDC
dose-response carcinogenicity study was carried out by Carlborg
(1983). The data were examined using four types of mathematical model
for carcinogenic dose/response viz:
the threshold level (no-effect-level) model,
the one-bit model,
the Weibull model, and
the polynomial (multi-stage) model.
Three versions of the polynomial model were considered; the first
with a cubed power of the dose as the lowest term, the second with a
squared power of the dose as the lowest term and the third with a
linear power of the dose as the lowest term.
A dose of 0.01% sodium saccharin in the diet was chosen as the
level at which low-risk assessments were made.
The data overwhelmingly rejected the one-hit model; all the other
models fitted the data in the statistical sense. The threshold level
method yielded 1.0% as a lower bound on the threshold Á. The
linearized polynomial model yielded a best estimate of 5.9 × 10-5 for
the excess risk at a saccharin dose of 0.01% of the diet with an upper
confidence limit of 9.1 × 10-5. The Weibull model yielded a best
estimate of 2.5 × 10-10 for the excess risk at a saccharin dose of
0.0l% with an upper confidence limit of 1.2 × 10-8.
The author concluded that, even under the conservative assumption
of low-dose linearity, the results from the IRDC study have reduced
the estimated risk by roughly one order of magnitude relative to the
estimates based on previous experiments; a risk assessment based on
the observable dose-response pattern showed that saccharin is
virtually safe at an exposure of societal concern. Some of the
measurable characteristics of the urine of the rat taken very early in
life appeared to be predictive of tumourigenicity.
In addition to the long-term rat study described above, Adamson &
Sieber (1983) administered saccharin (25 mg/kg bw) orally to two
groups of 10 monkeys each on 5 days/week. One group received saccharin
for an average of 122 months and the second group for 36 months. Since
the inception of the study, none of the animals have died and there is
no evidence of toxicity or tumours in any of the animals.
OBSERVATIONS IN MAN
Morgan (1983) has reviewed the epidemiological studies carried
out in relation to ingestion of saccharin by man, including new
studies and re-analyses not previously available to JECFA (Walker et
al., 1982; Hoover & Hartge, 1982; Jensen & Kamby, 1982; Morrison et
al., 1982; Najem et al., 1982). In an attempt to summarize the studies
done to date, a statistical power analysis of previous case-control
studies was carried out. Based on this analysis, it was calculated
that, if the true relative risk of bladder cancer as a result of using
artificial sweeteners were 1.13 or larger, there was a 95% probability
that the studies reviewed, in toto would have detected such a risk
as statistically significant.
The results reviewed demonstrated that saccharin is not a strong
or even a moderate carcinogen for man and the author concluded that
the remarkable approximation to unity of the summary relative risk
from all studies was impressive.
New information presented to the Committee included biochemical,
pharmacokinetic, mutagenicity, and epidemiological data; the results
of special studies on urine volume and composition and the effect of
saccharin on the bladder epithelium; the results of studies on
saccharin as a promoter or co-carcinogen; and the results of a
carcinogenicity study in rats designed to investigate the dose-
response relationship in the development of bladder tumours and the
outcome of in utero exposure.
In the Committee's opinion the available evidence indicated that
saccharin is not mutagenic. An in utero phase of exposure is not
essential for a carcinogenic response to saccharin in the bladder of
the male rat. There was a definite carcinogenic effect at levels of
dietary inclusion of 3% and above in the long-term study with in
utero exposure. There was also a carcinogenic effect at a level of
5%, the only level tested, in the 1-generation study with exposure
from birth, which included pups suckled by dames receiving saccharin
in their diets. The Committee considered that a 1% dietary inclusion
level could be taken as a no-effect level. Further data on the bladder
histopathology in the carcinogenicity study mentioned above were
received too late to be reviewed by the Committee. Within the
statistical limitations of the studies, the epidemiological data do
not show any evidence that saccharin is a bladder carcinogen.
Level causing no toxicological effect
Rat: 1% (10.000 ppm) in the diet, equivalent to 500 mg/kg bw.
Estimate of temporary acceptable daily intake for man
0-2.5 mg/kg bw.
FURTHER WORK OR INFORMATION
(Information to be submitted when it becomes available.)
1. Data on the bladder histopathology.
2. Information to elucidate the mechanism by which the compounds
produce bladder tumours, including the possible significance of
exposure through lactation, the influence of gastrointestinal
tract microorganisms, the effect of osmolar changes in the urine,
and species specificity in the development of urothelial changes.
ADAMSON, R.H. & SIEBER, S.M. (1983) Chemical carcinogenesis studies in
non-human primates. In: Organ and Species Specificity in Chemical
Carcinogenesis. R. Langenbach et al., eds. p. 129, New York: Plenum.
ANDERSON, R.L. (1979) Response of male rats to sodium saccharin
ingestion: urine composition and mineral balance. Fd. Cosmet.
Toxicol., 17; 195.
ANDERSON, R.L. (1983) Effect of saccharin ingestion on stool
composition in relation to caecal enlargement and increased stool
hydration. Fd. Chem. Toxicol., 21: 255.
BERNDT, W.O., REDDY, R.V., & HAYES, A.W. (1981) Evaluation of renal
function in saccharin treated rats. Toxicology, 21: 305.
CARLBORG, F.W. (1983) Statistical considerations in the design of the
IRDC experiment with saccharin. Unpublished report presented at the
Scientific Review Conference, Duke University Medical Center, 4th-6th
COLBURN, W.A., BEKERSKY, I., & BLUMENTHAL, H.P. (1981) A preliminary
report on the pharmacokinetical of saccharin in man: single oral
administration. J. Clin. Pharmacol. 21: 147.
DEMERS, D.M., FUKUSHIMA, S., & COHEN, S.M. (1981) Effect of sodium
saccharin and L-tryptophan on rat urine during bladder carcinogenesis.
Cancer Res., 41: 108.
EL-GERZAWI, S., HEATFIELD, B.M., & TRUMP, B.F. (1982) N-methyl-
N-nitrosourea and saccharin: effects on epithelium of normal human
urinary bladder in vitro. J. Natl. Cancer Inst., 69: 577.
FAHRIG, R. (1982) Effects in the mammalian spot test: cyclamate versus
saccharin. Mutation Res., 103: 43.
FUKUSHIMA, S. & COHEN, S.M. (1980) Saccharin-induced hyperplasia of
the rat urinary bladder. Cancer Res., 40: 734.
FUKUSHIMA, S., HIROSE, M., OKUDA, M., NAKANOWATARI, J., HATANO, A.,
& ITO, N. (1982) Effect of partial cystectomy on the induction of
pre-neoplastic lesions in rat bladder initiated with N-butyl-N-
(4-hydroxybutyl) nitrosamine followed by bladder carcinogens and
promoters. Utol. Res., 10: 115.
FUKUSHIMA, S., ARAI, M., NAKANOWATARI, J., HIBINO, T., OKUDA, M., &
ITO, N. (1983a) Differences in susceptibility to sodium saccharin
among various strains of rats and other animal species. Gann,
FUKUSHIMA, S., HAGIWARA, A., OGISO, T., SHIBATA, M., & ITO, N. (1983b)
Promoting effects of various chemicals in rat urinary bladder
carcinogenesis initiated by N-nitroso-n-butyl-(4-hydroxybutil)
amine. Fd. Chem. Toxicol., 21: 59.
HOOVER, R.1N. & HARTGE, P. (1982) Non-nutritive sweeteners and bladder
cancer. Am. J. Publ. Hlth., 72; 382.
IMAIDA, K., OSHIMA, M., FUKUSHIMA, S., ITO, N., & HOTTA, K. (1983)
Membrane potentials of urinary bladder epithelium in F344 rats treated
with N-butyl-n-(4-hydroxybutil) nitrosamine or sodium saccharin.
Carcinogenesis, 4: 659.
IRDC (1983) Evaluation of the dose response and in utero exposure of
saccharin in the rat. Unpublished report of the International Research
& Development Corporation, Mattawan, MI.
ISHII, D.N. (1982) Inhibition of iodinated nerve growth factor binding
by the suspected tumour promoters, saccharin and cyclamate. J. Natl.
Cancer Inst., 68: 299.
ITO, N., FUKUSHIMA, S., SHIRAI, T., & NAKANISHI, K. (1983a) Effects
of promoters on N-butyl-N-(4-hydroxybutyl) nitrosamine-induced urinary
bladder carcinogenesis in the rat. Environ. Hlth. Pers., 50: 61.
JENSEN, O.M. (1983) Artificial sweeteners and bladder cancer:
epidemiological evidence. Unpublished paper presented at the Third
European Toxicology Forum, Geneva, Switzerland. October 18-22, 1983.
LITTON BIONETICS INC. (1983a) Mutagenicity evaluation of FDC 12 in the
Ames Salmonella/Microsome plate test. Unpublished Report submitted to
FDC Consultants Inc.
LITTON BIONETICS INC. (1983b) Mutagenicity evaluation of FDC 14 in the
Ames Salmonella/Microsome plate test. Unpublished Report submitted to
FDC Consultants Inc.
LITTON BIONETICS INC. (1983c) Mutagenicity evaluation of Lot No.2517
in the Ames Salmonella/Microsome plate test. Unpublished Report
submitted to FDC Consultants Inc.
LOK, E., IVERSON, F., & CLAYSON, D.B. (1982) The inhibition of urease
and proteases by sodium saccharin. Cancer Letts., 16: 163.
MAHON, G.A.T., & DAWSON, G.W.P. (1982) Saccharin and the induction of
presumed somatic mutations in the mouse. Mutation Res., 103: 49.
MATTHEWS, H.B., FIELDS, M., & FISHBEIN, L. (1973) Saccharin:
distribution and excretion of a limited dose in the rat. J. Agr. Fd.
Chem., 21: 916.
MORGAN, R.W. (1983) A review of epidemiologic studies of artificial
sweeteners and bladder cancer. Unpublished Report presented at the
Scientific Review Conference, Duke University Medical Center, 4th-6th
MORRISON, A.S., VERHOEK, W.G., LECK, I., AOKI, K., OHNO, Y., & OBATA,
K. (1982) Artificial sweeteners and bladder cancer in Manchester,
U.K., and Nagoya, Japan. Br. J. Cancer, 45: 332.
MURASAKI, G. & COHEN, S.M. (1981) Effect of dose of sodium saccharin
on the induction of rat urinary bladder proliferation. Cancer Res.,
MURASAKI, G. & COHEN, S.M. (1983a) Co-carcinogenicity of sodium
saccharin and N-(4-(5-nitro-2-firyl)-2-thiazolyl) formamide for the
urinary bladder. Carcinogenesis, 4: 97.
MURASAKI, G. & COHEN, S.M. (1983b) Effect of sodium saccharin on
urinary bladder epithelial regenerative hyperplasia following freeze
ulceration. Cancer Res., 43: 182.
NAJEM, G.R., LOURIA, D.B., SEEBODE, J.J., THIND, I.S., PRUSAKOWSKI,
J.M., AMBROSE, R.B., & FERNICOLA, A.R. (1982) Life-time occupation,
smoking, caffeine, saccharine, hair dyes and bladder carcinogenesis.
Int. J. Epid., 22: 212.
NAKANISHI, K., FUKUSHIMA, S., HAGIWARA, A., TAMANO, S., & ITO, N.
(1982) Organ specific promoting effects of phenobarbital sodium and
sodium saccharin in the induction of liver and urinary bladder tumors
in male F344 rats. J. Natl. Cancer Inst., 68: 497.
PANTAROTTO, C., SALMONA, M., & GARATTINI, S. (1981a) Plasma kinetics
and urinary elimination of saccharin in man. Toxicol. Lett.,
PANTAROTTO, C., SALMONA, M., FANELLI, R., BIACHI, M., & SZCZAWINSKA,
K. (1981b) ' GLC-mass fragmentographic determination of saccharin in
biological fluids. J. Pharmaceut. Sci., 70: 871.
RADFORD, T., COOK, J.M., DALSIS, D.E., WOLF, E., & VOIGT, M. (1983)
Identification and quantitation of aminosaccharins in sodium saccharin
produced by the Maumee process. Fd. Chem. Toxicol., in press.
REITZ, R.H., FOX, T.R., QUAST, J.F., EMERSON, J., LEWKOWSKI, J., &
STANLEY, J. (1983) Food consumption patterns and effects of saccharin
upon urinary bladder tissue in very young rats. Unpublished report,
Dow Chemical, U.S.A.
REITZ, R.H. & MENDRALA, A.L. (1983) Evaluation of soldium saccharin
and 1-naphthalene sulfonic acid in the rat hepatocyte unscheduled DNA
synthesis assay. Unpublished report, Dow Chemical, U.S.A.
RENWICK, A.G. & SIMS, J. (1983) Distension of the urinary bladder in
rats fed saccharin containing diet. Cancer Letts., 18: 63.
RIGGIN, R.M. & KINZER, G.W. (1983) Characterization of impurities in
commercial lots of sodium saccharin produced by the Sherwin-Williams
process 1. Chemistry. Fd. Chem. Toxicol., 21: 1.
SCHOENIG, G.P. & ANDERSON, R.L. (1983) The effect of high dietary
levels of sodium saccharin on selected physiological parameters in
rats. Unpublished report.
SHIBATA, S., GOLDSTEIN, I.J., & KIRKLAND, J.J. (1983) Structure of a
water-insoluble D-glucan isolated from a Streptococcal organism.
Carbohydrate res., 120: 77.
SIMS, J. & RENWICK, A.G. (1983a) Unpublished observations cited by
Renwick, A.G. (1983) Saccharin and tryptophan metabolism. Paper
presented at the Expert Review Panel Meeting, Duke University, 4th-6th
SWEATMAN, T.W. & RENWICK, A.G. (1980) The tissue distribution and
pharmaeokinetics of saccharin in the rat. Toxicol. appl. Pharmacol.,
SWEATMAN, T.W., RENWICK; A.G., & BURGERS, C.D. (1981) The
pharmacokinetics of saccharin in man. Xenobiotica, 11: 531.
SWEATMAN, T.W. & RENWICK, A.G. (1982) Tissue levels of saccharin in
the rat during two-generation feeding studies. Toxicol. appl.
Pharmacol., 62: 465.
TSING, M.T. (1983) Final report on the preliminary autoradiographic
evaluation of the urinary bladder from saccharin treated rats.
TSUDA, H., FUKUSHIMA, S., IMAIDA, K., KURATA, Y., & ITO, N. (1983)
Organ-specific promoting effect of Phenobarbital and Saccharin in
induction of thyroid, liver and urinary bladder tumours in rats after
initiation with N-nitrosomethylurea. Cancer Res., 43: 3292.
WEST, R.W. & JACKSON, D.C. (1981) Saccharin effects on the urinary
physiology and urothelium of the rat when administered in diet or
drinking water. Toxicol. Lett., 7: 409.
WALKER, A.M., DREYER, N.A., FRIEDLANDER, E., Loughlin, J., ROTHMAN,
K.J., & KOHN, H.I. (1982) An independent analysis of the National
Cancer Institute Study on non-nutritive sweeteners and bladder cancer.
Am. J. Pub. Hlth., 72: 376.
WESTLAND, J.A. & HELTON, E.D. (1983) Evaluation of sodium saccharin
administered in dry and liquid diets in Rhesus monkeys. Unpublished
report of the Primate Research Institute, New Mexico State University.
WOLF, E. & VOIGT, M. (1983) Determination of polar impurities in
saccharin and saccharin sodium by HPLC. Fresenius Z. Anal. Chem.,