Saccharin

• Chem. Abstr. Serv. Reg. No.: 81-07-2
• Deleted CAS Reg. Nos: 474-91-9; 61255-27-4; 126987-83-5
• Chem. Abstr. Name: 1,2-Benzisothiazol-3(2H)-one, 1,1-dioxide
• IUPAC Systematic Name: 1,2-Benzisothiazolin-3-one, 1,1-dioxide
• Synonyms: Acid saccharin; anhydro-ortho-sulfaminebenzoic acid; 3-benzisothiazolinone 1,1-dioxide; benzoic sulfimide; benzoic sulphinide; benzosulfimide; benzosulfinide; ortho-benzoic acid sulfimide; ortho-benzoic sulfimide; ortho-benzosulfimide; ortho-benzoyl sulfimide; 1,2-dihydro-2-ketobenzisosulfonazole; 2,3-dihydro-3-oxobenzisosulfonazole; 1,1-dioxo-1,2-benzisothiazol-3(2H)-one; 3-hydroxybenzisothiazole-S,S-dioxide; saccharimide; saccharin acid; saccharin insoluble; saccharine; saccharinol; saccharinose; saccharol; ortho-sulfobenzimide; ortho-sulfobenzoic acid imide

Sodium saccharin

• Chem. Abstr. Serv. Reg. No.: 128-44-9
• Deleted CAS Reg. No.: 38279-26-4
• Chem. Abstr. Name: 1,2-Benzisothiazol-3(2H)-one, 1,1-dioxide, sodium salt
• IUPAC Systematic Name: 1,2-Benzisothiazolin-3-one, 1,1-dioxide, sodium salt
• Synonyms: ortho-Benzoylsulfimide sodium salt; saccharin sodium; saccharin sodium salt; saccharin soluble; sodium ortho-benzosulfimide; sodium saccharide; sodium saccharinate; sodium saccharine; soluble saccharin

Calcium saccharin

• Chem. Abstr. Serv. Reg. No.: 6485-34-3
• Deleted CAS Reg. No.: 17105-05-4
• Chem. Abstr. Name: 1,2-Benzisothiazol-3(2H)-one, 1,1-dioxide, calcium salt
• IUPAC Systematic Name: 1,2-Benzisothiazolin-3-one, 1,1-dioxide, calcium salt
• Synonyms: Calcium ortho-benzosulfimide; calcium saccharinate

3.1.1. Single-generation exposure

Mouse: Groups of 50 female Swiss mice, 9–14 weeks of age, were given 0 or 5% saccharin made by the Remsen-Fahlberg method in the diet for 18 months, at which time the survivors were killed. The average survival rates were not affected, and the tumour incidences were similar in tested and control animals. No pathological alterations were observed macroscopically in the urinary bladder (Roe et al., 1970). [The Working Group noted that the urinary bladders were not examined histologically.]

As part of a multigeneration study, two groups of 50 male and 50 female Swiss SPF mice [age unspecified] were fed diets containing 0.5 or 0.2% saccharin (free acid) made by the Remsen-Fahlberg method (containing 0.5% ortho-toluenesulfonamide) for up to 21 months. A concurrent control group of 50 males and 50 females received a standard diet. At 18 months, 62, 64 and 66 animals were still alive in the groups receiving 0.5 and 0.2% saccharin and in the control group, respectively. One control female developed an anaplastic carcinoma of the bladder, and one male fed 0.2% saccharin had a noninvasive transitional-cell carcinoma of the bladder (Kroes et al., 1977) (see also section 3.1.2).

Groups of 25 male and 25 female Charles River CD mice, eight weeks of age, received diets containing sodium saccharin (containing 345 mg/kg (ppm) ortho-toluenesulfonamide) at concentrations of 0, 1 or 5% for up to two years. Animals that died before six months were not examined, and the survival times were not reported. Animals were killed when obvious tumours were seen or when they were moribund; all survivors were killed at two years. All animals that survived six months or longer were examined grossly, and any tissues with abnormal changes were examined histologically; in addition, all vital organs from at least 12 animals in each group were examined histologically. In high-dose males, two cases of papillary hyperplasia of the bladder and two small papillomas were found. One transitional-cell carcinoma of the bladder associated with a stone was found in male controls. Vascular tumours were seen at increased frequency in male mice at the high dose, while lung tumours, hepatomas and lymphomas occurred with apparently equal incidence in control and treated groups. Any differences in the incidence of tumours were considered not to be significant, and none were found in a duplicate experiment for which no data were given (Homburger, 1978). [The Working Group noted the inadequate reporting of the experiment.]

Fifty male B6C3F₁ mice, six weeks of age, were fed sodium saccharin (purity, 99.5%; with 7 ppm ortho-toluenesulfonamide) at a dose of 5% in Oriental M diet for 52 weeks with interim kills of five mice at 0, 4, 8, 16 and 20 weeks after the beginning of the experiment; 20 mice were still alive at the end of the experiment. A control group of 35 mice was fed Oriental MF diet only. There was no effect on growth or survival, and no bladder lesions were detected by autoradiography, histology or scanning electron microscopy (Fukushima et al., 1983a). [The Working Group noted the small number of animals and the short duration of the experiment.]

Groups of 10 male and 10 female inbred ICR Swiss mice, six weeks of age, received 0, 0.5, 1 or 1.5 g/kg bw per day saccharin dissolved in 1 mL distilled water by oral gavage for one year, at which time all remaining mice were killed. No deaths occurred in any of the treated groups. The mice fed 1.5 g/kg bw per day showed slight weight loss when compared with controls (48 versus 55 g); both groups consumed 10 g of food per day per mouse. Five males and three females at the high dose had papillary adenocarcinomas of the thyroid. No thyroid tumours were reported in the other groups, and no tumours of other sites were reported (Prasad & Rai, 1986). [The Working Group noted the inadequate number of mice and the incomplete reporting of this experiment. It also noted that the finding of thyroid tumours has not been replicated in any other study in mice or in other species.]

As part of study of two-stage carcinogenesis, female BALB/c StCrlfC3H/Nctr mice were randomly divided into five groups of 192, 192, 192, 144 and 96 mice and were fed 0, 0.1, 0.5, 1, or 5 sodium saccharin, respectively, in Purina Lab Chow beginning at 19 weeks of age and continuing until 135 weeks of age. A slight but nonsignificant statistically increase in the length of survival was observed in treated mice. No bladder neoplasms were observed in any of the groups, and the incidence of bladder hyperplasia was similar: 8/164 (5%), 10/162 (6%), 9/161 (5%), 7/130 (5%) and 3/79 (4%), respectively. The incidences of Harderian gland neoplasms were 27/163 (17%), 32/172 (19%), 29/160 (18%), 22/132 (17%) and 22/84 (26%), respectively (p < 0.04, test for trend). A significant, dose-related reduction in the time to onset of lymphomas was observed, although the incidences were similar among groups (Frederick et al., 1989; see also section 3.4). [The Working Group noted that Harderian gland tumours are common age-related, spontaneous neoplasms, the variable percentage of Harderian glands examined, the variability of the incidence of these tumours in control animals, the unequal numbers of animals per group and the long survival time.]

Rat: Groups of 10 male and 10 female Osborne-Mendel rats, 21 days of age, received diets containing 0, 1 or 5% saccharin [source and purity unspecified] for up to two years. The mortality rate in pooled controls was 14% at one year and 68% at two years; at one year, seven males and nine females in the control group, 10 males and 10 females in the group receiving 1% and nine males and nine females at 5% were still alive; the two-year survival rates were not given. It was reported that 7/18 rats [sex unspecified] at 5% had abdominal lymphosarcomas, and that four of the seven also had thoracic lymphosarcomas. The urinary bladders were not examined (Fitzhugh et al., 1951). [The Working Group noted the multiple inadequacies of this study, including the small number of animals in each group.]

Groups of 20 male and 20 female Boots-Wistar rats [age unspecified] were fed 0, 0.005, 0.05 or 5% saccharin made by the Remsen-Fahlberg method [purity unspecified] for two years. At 18 months, 15 male and 14 female controls and 10 male and 10 female rats at the highest dose level were still alive. No statistically significant differences in tumour incidence were found between treated and control animals. Only five bladders, all from animals at the highest dose, were examined histologically. Urothelial hyperplasia was found in one male and one female, and a bladder papilloma was found in another female. Bladder parasites were not found. Bladder calculi were found in four male and one female rats fed 5% saccharin (Lessel, 1971).

Groups of 52 male and 52 female BD rats [age unspecified] were fed 0 (control), 0.2 or 0.5% sodium saccharin made by the Remsen-Fahlberg method [purity unspecified] for up to 30 months starting between 70 and 90 days of age, providing average total doses of 0, 83 and 210 g/kg bw. The survival rates at 18 months were 55/104 controls, 50/104 at the low dose and 41/104 at the high dose; at 24 months, the survival rates were 6/104, 3/104 and 5/104, respectively. Sixteen percent of all animals had parasites (Strongyloides capillaria) in the urinary tract. Benign and malignant mesenchymal tumours were found with similar frequency in all groups. No bladder tumours were observed (Schmähl, 1973).

In a study reported in an abstract, groups of Charles River CD male and female rats [number and age unspecified] received saccharin [source and purity unspecified] by an unspecified route (in the diet or by gastric intubation thrice weekly) for 18 months, followed by a six-month period of observation. A high incidence of benign tumours of the pituitary and mammary glands was found in surviving controls and experimental animals. The survival times, types of pathological examination, tumour types and other important experimental details were not reported (Ulland et al., 1973). [The Working Group noted the inadequacy of this experiment.]

In a study reported in an abstract, groups of 54–56 male Wistar rats [age unspecified] were fed 0 or 2.5 g/kg bw per day sodium saccharin [source and purity unspecified] for up to 28 months. Ten to 16 rats from each group were killed at 12 months, 11 from each group at 24 months and all survivors [number unspecified] at 28 months. No urinary bladder tumours were observed (Furuya et al., 1975). [The Working Group noted the incomplete reporting of this experiment.]

Groups of 60 male and 60 female Charles River CD rats [age unspecified] were fed diets containing sodium saccharin made by the Remsen-Fahlberg method (purity conformed to United States Pharmacoepia, British Pharmacoepia and Food Chemicals Codex specifications) for 26 months, to give daily intakes of 0, 0.09, 0.27, 0.81 or 2.4 g/kg bw. Saccharin treatment did not affect the survival of female rats: at 18 months, approximately 50% of the original animals were alive. The survival of male rats was affected in a dose-related manner: thus, at 18 months, about 80% of male control rats but only about 50% of those at the highest dose were still alive. By 24 months, about 10% of the animals in all groups were alive. Four transitional-cell tumours of the bladder were found, one in a male and one in a female given 0.09 g/kg bw and two in males fed 0.81 g/kg bw; an angiosarcoma of the bladder was found in a male control. Bladder calculi were recorded, but there was no association between the presence of calculi, saccharin treatment and/or bladder tumours. The animals were free from bladder parasites. The combined incidences of lymphomas and leukaemias were 7/54 in males at the highest dose of saccharin and 2/57 in untreated male controls (Munro et al., 1975).

Groups of 25 male Charles River CD-1 rats [age unspecified] received sodium saccharin (containing 345 mg/kg (ppm) ortho-toluenesulfonamide) in the diet at concentrations of 0, 1 or 5% for up to two years. Animals that died before six months were not examined, and the survival times were not reported. Animals were killed when obvious tumours were seen or when they were moribund; all survivors were killed at two years. All animals that survived six months or longer were examined grossly, and any tissues with abnormal changes were examined histologically; in addition, all vital organs from at least 12 animals in each group were examined histologically. Tumours of the urinary bladder, pituitary, breast and subcutaneous tissue were seen with equal incidence in all groups (Homburger, 1978). [The Working Group noted the inadequate reporting of the experiment].

A group of 75 male and 50 female Wistar SPF rats, eight weeks of age, received sodium saccharin made by the Remsen-Fahlberg method (containing 698 mg/kg (ppm) ortho-toluenesulfonamide) in the drinking-water, to give a daily intake of 2 g/kg bw saccharin. Another group of 75 males and 75 females received 4 g/kg bw per day saccharin in the diet. A group of 55 males and 50 females served as controls. The males receiving saccharin in the drinking-water were also given 1% ammonium chloride for four weeks and then 0.5% for life, in order to correct a treatment-associated rise in urinary pH. Of the male controls, 25 were given ammonium chloride at the same concentrations. No treatment-associated change in urinary pH occurred in either of the treated groups of females or in males receiving saccharin in the diet. The experiment was terminated after two years. Survival at 18 months was 49/55 male and 43/50 female untreated controls, 65/75 males and 44/50 females that received saccharin in the drinking-water and 55/75 males and 52/75 females fed saccharin in the diet. At 100 weeks, 37/55 male and 13/50 female controls, 49/75 males and 29/50 females receiving saccharin in the drinking-water and 12/75 males and 16/75 females fed saccharin in the diet were still alive. In control animals, the total tumour incidence was 1/52 males and 9/46 females. In rats receiving saccharin in the drinking-water at 2 g/kg bw per day, the incidence was 11/71 in males and 10/44 in females; while in rats fed saccharin at 4 g/kg bw per day it was 10/70 in males and 7/68 in females. Transitional-cell carcinomas of the urothelium were not seen in male or female controls, but accounted for 1/71 in males (in the ureter) and 1/44 in females (in the renal pelvis) in rats receiving saccharin in the drinking-water and 3/70 in males (all in the bladder) and 0/68 in females fed saccharin. The incidence of lymphosarcomas and/or leukaemia was 0/52 in male and 0/46 in female controls, 4/71 in males and 1/44 in females given saccharin in the drinking-water, and 2/70 in male and 1/68 in female saccharin-fed rats. One Leydig-cell tumour was found in each of the saccharin-treated groups of males, but none occurred in the testes of untreated male controls. There was a treatment-associated increase in the number of microcalculi within the renal tubules of male (but not female) saccharin-treated rats, with an incidence of 2/52 in controls, 30/71 in males given saccharin in the drinking-water and 16/70 in saccharin-fed males. The animals were free from bladder parasites (Chowaniec & Hicks, 1979). [The Working Group noted the incomplete histopathological examination and the lack of lymphomas and/or leukaemia, a common neoplasm in controls.]

As part of a two-generation study (see section 3.1.2), groups of 50 male and 50 female Charles-River CD (Sprague-Dawley) rats, 30 days of age, were fed either a control diet or a diet containing 5% sodium saccharin prepared by the Maumee process and free of ortho-toluenesulfonamide. Survival was not affected by treatment. Bladder tumours (benign and malignant) were observed in 1/36 control males and in 7/38 male rats (p < 0.03) fed saccharin which survived 87 weeks or more (the time at which the first tumour was observed). In addition, one treated male and two treated females had urothelial tumours of the renal pelvis, and one treated male had a urethral tumour; no other urothelial tumours were observed in controls. The incidence of bladder calculi was not related to treatment or to tumour incidence. The animals were free of bladder parasites (Arnold et al., 1980).

As part of a two-generation study, 50 female Wistar rats [age unspecified] were given 2 g/kg bw per day sodium saccharin made by the Maumee process in the diet for two years. A group of 63 animals served as controls. At week 84, 50/63 controls and 37/50 saccharin-fed rats were still alive. The overall tumour incidences were similar in the two groups, and no bladder neoplasm occurred. Mild focal urothelial hyperplasia was seen in one rat fed saccharin. The animals were free from bladder parasites (Hooson et al., 1980). [The Working Group noted that the animals were not started on the test at weaning but had been fed a normal diet for several weeks before the start of the study.]

Groups of 40–48 male ACI, Wistar, Fischer 344 and Sprague-Dawley rats, six weeks of age, were fed 5% sodium saccharin (food additive grade; purity, 99.5%, with 7 ppm ortho-toluenesulfonmide) in powdered diet, and surviving rats were killed at the end of 52 weeks of treatment. Interim sacrifices were also performed on five rats of each strain at 12, 24 and 36 weeks. Corresponding control groups of 40–45 rats were fed untreated diet. The treated groups had significant growth retardation, with average body weights at week 52 of treated versus control groups as follows: ACI, 299 versus 327 g; Wistar, 400 versus 447 g; Fischer 344, 403 versus 427 g; and Sprague-Dawley, 593 versus 716 g. There was no apparent effect on survival. At the end of 52 weeks, no urinary bladder lesions were seen in the Wistar, Fischer 344 or Sprague-Dawley rats; of the ACI rats, 1/28 controls had simple hyperplasia and 25/32 (78%) rats treated with sodium saccharin had simple hyperplasia, 20 (62.5%) had papillary/nodular hyperplasia, nine (28.1%) had papillomas and three (9.4%) had carcinomas. It was also reported that one of the ACI rats had a bladder calculus, but more than half of the control and test ACI rats bore the bladder nematode, Trichosomoides crassicauda (Fukushima et al., 1983a). [The Working Group noted the limited duration of treatment.]

A group of 68 male Fischer 344 rats, seven weeks of age, were fed 5% sodium saccharin in Oriental MF diet; a control group of 31 rats were fed the basal diet alone. Interim sacrifices were carried out during the course of the experiment, and the remaining rats were killed at the end of 112 weeks of treatment. The body-weight gain was similar in the two groups up to 15 weeks, whereas after 20 weeks the body weight increased more slowly in treated rats than in controls. Simple hyperplasia was seen in the bladders of treated rats as early as eight weeks of treatment, and about two-thirds of rats at all times had simple hyperplasia. Papillary or nodular hyperplasia was observed in the bladders of approximately one-third of rats killed after 8, 12, 20, 80 and 112 weeks of treatment. At 4, 16, 60, 90 and 100 weeks, no rats had papillary or nodular hyperplasia. Simple hyperplasia was occasionally seen in control rats: 2/6 rats at four weeks, 1/5 at 20 weeks, and 1/7 at 100 weeks. No papillary or nodular hyperplasia was seen in the control group, and no papilloma or transitional-cell carcinoma was seen in either group. Trichosomoides crassicauda were not present in the bladders. When the stomachs of 20 treated and 11 control rats killed after 80 weeks were examined, all of the sodium saccharin-treated rats had hyperkeratosis at the limiting ridge of the forestomach, and five papillomas of the limiting ridge of the forestomach were reported. Ulcers were seen in the glandular stomach in four animals. No squamous-cell carcinomas or adenocarcinomas were observed in either group (Hibino et al., 1985). [The Working Group noted the incomplete sampling of the stomach between groups. The Group could not agree on the diagnosis of papillomas from the illustrations provided.]

A group of 36 male Sprague-Dawley and 35 male analbuminaemic (a mutant strain derived from Sprague-Dawley rats) rats, six weeks of age, were fed 5% sodium saccharin in powdered CE-2 diet for 80 weeks. Fourteen Sprague-Dawley and 12 analbuminaemic rats served as controls. There was no apparent effect on body weight or on survival. No bladder tumours were present in any of the rats. Simple hyperplasia of the urinary bladder was observed in 2/35 analbuminaemic and 2/36 Sprague-Dawley rats given sodium saccharin (Homma et al., 1991). [The Working Group noted that the treatment period was less than two years.]

Hamster: Groups of 30 male and 30 female random-bred Syrian golden hamsters, eight weeks of age, received saccharin made by the Maumee process at concentrations of 0, 0.156, 0.312, 0.625 or 1.25% in the drinking-water for their natural lifespan. The highest dose used in this study was the maximum tolerated, as determined in an eight-week study. The average daily consumption ranged from 44 mg/animal given 0.156% to 353 mg/animal given 1.25%. The mean survival time was 50–60 weeks in all groups. The pathological changes observed and the distribution and histological types of neoplasms were within the range of tumours that occur commonly in hamsters in this colony (Althoff et al., 1975).

Fifty male Syrian golden hamsters, six weeks of age, were fed 5% sodium saccharin (purity, 99.5%, with 7 ppm ortho-toluenesulfonamide) in the diet for 52 weeks. Five animals were killed at 0, 4, 8, 16 or 20 weeks after the beginning of the experiment; 20 hamsters were available at the terminal sacrifice. A group of 35 hamsters served as untreated controls. There was no effect on growth or survival, and no bladder lesions were detected by autoradiography, histology or scanning electron microscopy (Fukushima et al., 1983a). [The Working Group noted the small number of animals and the short duration of the experiment.]

Guinea-pig: A group of 30 male Hartley guinea-pigs, six weeks of age, were fed 5% sodium saccharin (purity, 99.5%, with 7 ppm ortho-toluenesulfonamide) in Oriental RC diet for 52 weeks with interim sacrifices of three guinea-pigs at 0, 4, 12, 16 or 20 weeks from the beginning of the experiment; 12 guinea-pigs were available at the terminal sacrifice. A group of 20 guinea-pigs served as untreated controls. Treated guinea-pigs had lower body-weight gain than the controls, but no bladder lesions were detected by autoradiography, histology or scanning electron microscopy (Fukushima et al., 1983a). [The Working Group noted the small number of animals and the short duration of the experiment.]

Monkey: In a study reported in an abstract, one of two batches of sodium saccharin made by the Remsen-Fahlberg method (containing 2.4 and 3.2 mg/kg ortho-toluenesulfonamide) (Coulston et al., 1975) was given orally at doses of 20, 100 or 500 mg/kg bw per day on six days a week to groups of two, two and three Macaca mulatta (rhesus) monkeys of each sex, respectively. Three animals of each sex served as controls. After 79 months on this regimen, 11 monkeys remained in the treated groups, and all monkeys were killed and autopsied. Histopathological examination revealed no abnormal lesions in the urinary bladder, kidneys or testis in monkeys that survived the treatment or in those that died during the test (McChesney et al., 1977).

Groups of five female and one male African green (Cercopithecus aethiops), two female and five male rhesus (Macaca mulatta), three female and three male cynomolgus (Macaca fascicularis) and one hybrid (rhesus male cross cynomolgus female) monkey, 0–10 days of age, were given 25 mg/kg bw sodium saccharin (purity, > 99%) in the diet on five days a week for up to 283 months. Eight monkeys died during the course of the experiment, after 103, 128, 157, 168, 170, 192, 214 and 282 months of feeding. The remainder were killed at the end of the experiment, after 207–283 months. Five male and four female cynomolgus and five male and two females rhesus monkeys were available for comparison and were killed at 206–301 months of age. No bladder tumours were detected, and there was no evidence of hyperplasia by light or scanning electron microscopy (Takayama et al., 1998). (Preliminary reports of this experiment were published by Sieber & Adamson, 1978; Thorgeirsson et al., 1994). [The Working Group noted the relatively small dose of saccharin administered, the relatively small number of monkeys used and the multiplicity of species.]

3.1.2. Multigeneration exposure

In these studies, animals of each sex of the parent (F₀) generation were fed saccharin or sodium saccharin from weaning (or very soon after weaning) throughout both pregnancy and before weaning of their offspring. As the offspring were then placed on the same diet as their parents for their lifespan, their exposure to saccharin was greater than that of the F₀ generation by the length of the gestation and suckling periods.

Mouse: Saccharin containing 0.5% ortho-toluenesulfonamide was fed to groups of Swiss mice in a multigeneration study for life at concentrations of 0, 0.2 or 0.5% in the diet. The F₀, F_3b and F_6a generations, consisting of 50 males and 50 females, were used to test the compound for carcinogenicity. The experiments were terminated at 21 months. The survival rates at 18 months were 66, 62, 64 (F₀), 61, 54, 53 (F_3b) and 67, 48, 54 (F_6a) at 0, 0.2 and 0.5% respectively. Histopathological examination showed that lesions were equally distributed in the control and experimental groups. Two male mice, one of the F₀ generation receiving 0.2% saccharin and one of the F_3b generation receiving 0.5% saccharin, developed transitional-cell carcinomas of the bladder at 20.5 months. One female control of the F₀ generation had an anaplastic carcinoma of the bladder at 20.5 months (Kroes et al., 1977).

Rat: Groups of 20 male and 20 female weanling Sprague-Dawley rats of the F₁ generation were fed sodium saccharin made by the Remsen-Fahlberg method [purity unspecified] at concentrations of 0, 0.05, 0.5 or 5% of the basal diet for up to 100 weeks. Of the F₁ animals, 12, 10, 11 and 15 males and 16, 14, 14 and 19 females, respectively, survived to 80 weeks. Seven transitional-cell carcinomas of the urinary bladder developed, all in F₁ males on the 5% saccharin diet (p = 0.001). The presence or absence of bladder parasites was not recorded. The total numbers of tumour-bearing animals were two males and eight females at 0%, one male and six females at 0.05%, one male and five females at 0.5%, and seven males and 13 females at 5% (Tisdel et al., 1974).

Groups of 50 male and 50 female Charles River CD (Sprague-Dawley) rats, 30 days of age, were fed either a control diet or a diet containing 5% sodium saccharin continuously for life. The saccharin was prepared by the Maumee process and was free of ortho-toluenesulfonamide. After three months on test, the animals were mated on a one-to-one basis. All litters were culled to eight pups (four males and four females) four days post partum in a random manner. The pups were weaned onto their parents’ diet, and 50 males and 50 females from each group were randomly selected to constitute the second generation. The survival of the offspring (F₁ generation) was not affected by treatment. Of the F₁ animals surviving 67 weeks or longer, at which time the first tumour was observed, none of the 42 male controls but eight of the 45 saccharin-treated males had developed transitional-cell carcinomas of the bladder [p = 0.002], and four had transitional-cell papillomas; two of the 49 surviving females fed 5% sodium saccharin also had bladder cancers. Although urinary bladder calculi were noted occasionally, their incidence was not related to treatment, nor were they associated with the tumours. The animals were free of bladder parasites (Arnold et al., 1980).

Groups of five to seven female Sprague-Dawley were treated by oral gavage with saccharin [form not specified] (containing < 10 ppm ortho-toluenesulfonamide) at doses of 0.2, 1 or 5 g/kg bw on days 14, 17 and 20 of pregnancy. Solutions containing 2.5 g/mL of the test compound were given by gavage to the rats, which had received no food or drinking-water overnight. A control group consisted of the offspring of five untreated rats. The numbers of offspring at the low, intermediate and high doses which survived to 28 days of age were 24 males and 27 females, 35 males and 23 females and 32 males and 25 females, respectively; the untreated controls had 25 male and 33 female offspring. Nine of 69 of the offspring of dams treated with the highest dose of saccharin died within the first four days after birth. The F₁ generation offspring were observed for life [time not specified] or killed when found moribund. The mean survival times of males tended to be higher with the higher doses of saccharin but were lower than those of controls. No significant differences in survival were found between groups. There were no lesions of the bladders in any of the treated groups, and there was no increase in the incidence of tumours at other sites. The tumours seen in the treated groups were the same as those seen in controls (Schmähl & Habs, 1980). [The Working Group noted the incomplete reporting of this experiment.]

Groups of 48 male and 48 female Charies River CD (Sprague-Dawley) rats of the F₁ generation were fed dietary levels of 0, 0.01, 0.1, 1, 5 or 7.5% sodium saccharin [method of production and purity unspecified] for 28 months after their parents had been fed the same diet from weaning. There were no significant differences in survival between treated and control animals. Although no difference in bladder tumour incidence was found between F₁ males fed 5% saccharin (1/21) and the F₁ controls (1/25) that survived beyond 18 months, 6/23 F₁ male rats fed 7.5% saccharin developed transitional-cell papillomas or carcinomas of the bladder. This result was significantly different from that in controls. There was no apparent correlation between tumour incidence and presence of bladder stones. The bladders were reported to be ‘free of visible parasites’ (Taylor et al., 1980).

Male Charles River CD rats of the F₁ generation were treated as follows: group 1 (350 rats), untreated controls; group 2 (700 rats), 1% sodium saccharin (produced by the Maumee process; purity, > 99%) in the diet; group 3 (500 rats), 3% sodium saccharin in the diet; group 4 (200 rats), 4% sodium saccharin in the diet; group 5 (125 rats), 5% sodium saccharin in the diet; group 6 (125 rats), 6.25% sodium saccharin in the diet; group 7 (125 rats), 7.5% sodium saccharin in the diet; group 8 (125 rats), 5% sodium saccharin administered through gestation period and then the F₁ rats fed control diet from birth until the end of the experiment; group 9 (125 rats), 5% sodium saccharin in the diet from the time of birth and continuing until the end of the experiment; group 10 (125 rats), 5% sodium hippurate (purity > 98%) in the diet until the age of eight weeks and then as 3% of the diet until the end of the experiment. The parents (the F₀ generation) were given the same diets from six weeks of age and continuing for nine weeks before the beginning of mating; they were then given the same diets through gestation and lactation (except as noted for groups 8 and 9). Group 8 received sodium saccharin throughout gestation and then received control diet. In group 9, the F₀ generation was given control diet until the birth of the F₁ generation, at which time the dams were given 5% sodium saccharin. There was significant growth retardation in all groups given sodium saccharin at doses of greater than 3% of the diet, including group 9. There were minimal changes in the growth of the rats in group 8, given sodium saccharin until parturition. Survival was comparable in the nine groups, except for a significant increase in survival in groups 5 and 7. The numbers of survivors of the F₁ generation in the nine groups at the time of killing at 30 months of age were: 80/350 (23%), 172/700 (25%), 114/500 (23%), 38/200 (19%), 46/125 (37%), 33/125 (26%), 42/125 (34%), 25/125 (20%), 33/125 (26%) and 38/125 (30%), respectively. The effective numbers of rats were those alive at 15 months of age when the first bladder tumour was detected. The incidences of transitional-cell papillomas were 0/324, 4/658 (0.6%), 4/472 (0.8%), 4/189 (2.1%), 4/120 (3.3%), 12/120 (10%), 18/118 (15%), 0/122, 4/120 (3.3%) and 0/118, respectively. The incidences of transitional-cell carcinomas were 0, 1 (0.2%), 4 (0.8%), 8 (4.2%), 11 (9.2%), 8 (6.7%), 19 (16%), 0, 8 (6.7%) and 0, respectively. It was noted that the background incidence of bladder neoplasia at the study laboratory was 0.8%. In this study, the incidences of bladder neoplasia were significantly increased in groups 3–7 and in group 9. The tumour incidences when 1% sodium saccharin and 5% were fed throughout gestation were not significantly greater than those of controls. There was no difference in the incidence of bladder neoplasms between groups 5 and 9 (5% sodium saccharin throughout gestation and after parturition and 5% sodium saccharin after parturition only) (Schoenig et al., 1985). [The Working Group noted that re-examination of the histopathology of the bladders was performed by Squire (1985), who found a significant increase in the incidence of bladder tumours only at doses of 4, 5, 6.25 and 7.25% sodium saccharin. There were no compound-related effects at 1%, and there was no statistically significant increase in the incidence of transitional-cell carcinomas (p = 0.25) at the 3% dose.]

3.3.1. Skin application

A total dose of 0.24 g saccharin [form not specified], made by the Remson-Fahlberg method, was applied as an 8% solution in acetone thrice weekly to the skin of ‘S’ strain mice. Twenty-five days after the start of the treatment, the animals were given 18 weekly applications of 0.17% croton oil in acetone. At the end of the croton oil treatment, 15 skin papillomas were observed in seven of the 20 saccharin-treated animals, by comparison with four papillomas in four of 19 controls treated with croton oil only. The increase was not statistically significant (Salaman & Roe, 1956).

3.3.2. Bladder insertion (implantation)

Saccharin [source and purity unspecified] was mixed with four times its weight of cholesterol, and pellets (9–11 mg) containing 2 mg saccharin were then inserted into the urinary bladder lumina of 20 ‘stock’ mice [sex and age unspecified]. An identical group composed of 28 mice received 9–11-mg pellets of cholesterol. The experiment lasted 52 weeks. Of mice that lived 30 weeks, 4/13 saccharin-treated and 1/24 control animals developed bladder tumours (p = 0.01) (Allen et al., 1957). [The Working Group noted serious difficulties in interpreting the results of studies conducted with pellet implantation in the bladder because of the significant carcinogenic effect of the pellet itself (Clayson, 1974; Jull, 1979; DeSesso, 1989).]

Sodium saccharin (analytically pure) was mixed at 4?5 mg with four times its weight of cholesterol; pellets weighing 20–24 mg containing sodium saccharin were then inserted into the urinary bladder lumina in two trials with groups each composed of 100 female Swiss mice aged 60–90 days. Ninety-nine percent of the sodium saccharin had disappeared from the pellet within 1.5 days. Identical groups received 20–24 mg pellets of pure cholesterol for 56 weeks. Only the bladders of animals surviving more than 25 weeks were examined microscopically. The first urinary bladder carcinoma was seen in a saccharin-treated animal 42 weeks after surgical insertion. The overall incidences of bladder carcinomas were 31/66 (trial 1) and 33/64 (trial 2) in saccharin-treated mice as compared with 8/63 (trial 1) and 5/43 (trial 2) in animals exposed to pure cholesterol pellets (p < 0.001). The carcinomas in saccharin-exposed mice were more frequently multiple and invasive (p < 0.009). They were composed of cells with a high mitotic index and exhibited more squamous or glandular metaplasia than was found in tumours in control animals. The incidence of tumours in other tissues were not different from those in control mice (Bryan et al., 1970). [The Working Group noted the paucity of details regarding this experiment and the difficulty in interpreting the results of studies involving pellet implantation technique in mice (Clayson, 1974; Jull, 1979; DeSesso, 1989)].

3.4.1. Benzo[a]pyrene

Groups of 50 female Swiss mice, 9–14 weeks of age, received an initial single gastric instillation of 0.2 mL polyethylene glycol either alone or containing 50 µg benzo[a]pyrene [purities unspecified]. Seven days later, the test diet containing 5% saccharin [purity unspecified] was fed for 72 weeks. The average survival rates were not different from those of controls. Although mice treated with benzo[a]pyrene had an increased incidence of squamous papillomas of the forestomach (20/61), saccharin did not enhance the occurrence (10/32). Hepatocellular adenomas, pulmonary neoplasms and malignant lymphomas occurred at similar frequencies in all groups. No pathological alterations were observed macroscopically in the urinary bladder (Roe et al., 1970). [The Working Group noted that benzo[a]pyrene is not organotropic for the bladder and that the urinary bladders were not examined histologically.]

3.4.2. 2-Acetylaminofluorene

Two groups of 12 female Horton Sprague-Dawley rats [age unspecified] were fed a diet supplemented with 300 mg/kg 2-acetylaminofluorene (AAF) for 40 weeks and/or 5% sodium saccharin. Eleven of the controls given AAF and 6/12 rats fed AAF plus saccharin developed palpable mammary and ear-duct tumours. Liver tumours were observed in both groups, but they were smaller and less malignant in the saccharin-fed animals. Microscopic examination of the urinary bladders indicated that the mucosal lining was hyperplastic in all rats fed AAF but more particularly so in those fed AAF plus saccharin; one animal fed AAF plus saccharin had squamous metaplasia and precancerous changes of the mucosal epithelium. No malignant lesions of the urinary bladder were observed (Ershoff & Bajwa, 1974). [The Working Group noted the inadequate number of animals and that the intake of AAF and saccharin could not be assessed since food consumption was not measured.]

Ten groups of weanling female BALB/c StCrlfCrlfC3H/Nctr mice, 21–26 days old, were fed control diet (groups 1–5) or were pretreated with 200 mg/kg diet (ppm) AAF (purity > 97%) (groups 6–10) for 13 weeks, followed by two weeks of control diet, and were then fed 0 (groups 1 and 6), 0.1% (groups 2 and 7), 0.5% (groups 3 and 8), 1% (groups 4 and 9) or 5% (groups 5 and 10) sodium saccharin (purity, > 98%). The saccharin was administered when the mice were 19–135 weeks old. There were 192 mice in each of groups 1–3 and 6–8, 144 mice in each of groups 4 and 9, and 96 mice in each of groups 5 and 10. The mortality rate of mice pretreated with AAF followed by control diet was significantly higher than that of mice that received only the control diet. Addition of sodium saccharin to the diet of AAF-pretreated animals increased their longevity in a dose-related fashion (p = 0.011). A less pronounced effect was observed in non-pretreated mice (p = 0.099). The incidences of bladder tumours were 2/164 (1%) for the group given AAF plus control diet and 3/165 (2%) for the groups given AAF plus 0.1% saccharin, respectively; no bladder tumours were seen in any of the other groups. The incidences of bladder hyperplasia were similar in the five groups pretreated with AAF and in the groups not pretreated with AAF. There was no dose–related increase in the incidence of neoplasms in any other tissues, including the Harderian gland, but a decreased incidence of lymphomas was seen in the group pretreated with AAF and given 5% saccharin (50% versus 62% in the controls). There was a statistically significant, dose-related reduction in the time to onset of lymphomas in AAF-pretreated and non-pretreated mice, correlated with the decreased mortality rate observed in saccharin-treated mice. An increase in the incidence of liver neoplasms was seen in AAF-pretreated when compared with non-pretreated mice, but there were no differences between the incidences at the different doses of saccharin and the corresponding controls (Frederick et al., 1989; see also section 3.1.1).

3.4.3. N-Methyl-N-nitrosourea

A group of 50 female Wistar SPF rats, six to eight weeks of age, were pretreated with a preparation of 1.5 mg N-methyl-N-nitrosourea (MNU) dissolved in 0.9% sodium chloride (pH 7.0) and instilled into the bladder, and then two days later received 4 g/kg bw per day sodium saccharin (Remsen-Fahlberg, containing an average of 698 mg/kg (ppm) ortho-toluenesulfonamide) in the diet for life or up to two years; a further group of 50 females was pretreated with 2 mg MNU and then given 2 g/kg bw sodium saccharin per day in the drinking-water. The control groups consisted of 55 male and 50 female untreated rats, 75 males and 50 females given 2 g/kg bw sodium saccharin per day in drinking-water, 75 males and 75 females fed 4 g/kg bw sodium saccharin per day in the diet, 85 males and females given 1.5 mg MNU, and 50 given 2 mg MNU and maintained on a saccharin-free diet for two years. The incidences of transitional-cell neoplasms of the bladder in surviving animals whose bladders were examined histologically were: 0/52 male and 0/46 female untreated controls, 0/71 males and 0/44 females at 2 g/kg bw per day sodium saccharin in drinking-water, 3/70 males and 0/68 females at 4 g/kg bw day sodium saccharin in the diet, 0/124 in MNU-treated animals, 23/49 females (47%; p < 0.0005) given MNU followed by 2 g/kg bw per day of sodium saccharin in drinking-water and 27/47 females (52%; p < 0.0005) given MNU followed by 4 g/kg bw per day sodium saccharin in the diet. The first bladder tumour was seen after 95 weeks in the saccharin-fed control group and after eight weeks in the MNU-initiated and sodium saccharin-treated test groups. The animals were free from bladder parasites (Hicks et al., 1978; Chowaniec & Hicks, 1979). [The Working Group noted the difficulty in determining the actual dose of MNU because of possible decay in solution over time.]

A single dose of 2 mg MNU was instilled into the urinary bladders of female Wistar rats (AF-Han strain) weighing 195 g. Thereafter, 50 animals were given 2% saccharin [purity unspecified] in the diet, the dose being increased after 10 weeks to 4%, for life, resulting in doses of 1.4–2.5 g/kg bw per day. The control groups consisted of 100 untreated female rats, 50 females receiving MNU alone and 50 females receiving distilled water. A further group of 50 female rats treated with MNU were given 3% calcium carbonate in the diet instead of saccharin. The survival rates at two years were 59/100 untreated controls, 28/50 water controls, 13/50 MNU-treated animals, 15/50 receiving MNU plus calcium and 14/50 given MNU plus saccharin. In the MNU-treated groups, the first tumour of the urinary bladder was found after 14 weeks. Benign and malignant urothelial neoplasms occurred in the renal pelvis, ureter and urinary bladder, the overall incidences of urinary tract tumours being 57% with MNU only (survival, 76 ± 29 weeks), 65% with MNU plus saccharin (survival, 78 ± 25 weeks) and 65% with MNU plus calcium carbonate (survival, 86 ± 23 weeks). The frequencies were 28, 57 and 43% in the renal pelvis, 17, 12 and 11% in the ureter and 39, 31 and 39% in the urinary bladder, respectively. The frequency of calcification in the urinary tract, including stone formation, was similar in all treated groups, including controls receiving water, and the occurrence did not correlate with that of tumours. One tumour of the urinary tract was found in an untreated control and one in a control receiving a water instillation into the urinary bladder. The presence or absence of bladder parasites was not reported (Mohr et al., 1978). [The Working Group noted that many bladder tumours were found and that the animals were heavier than those used in the experiment by Hicks et al. (1978).]

Groups of 50 female Wistar/AF-Han rats, average body weight 195 g, were treated with a single intravesicular instillation of water, a single intravesicular instillation of 2 mg MNU in 0.5 mL water with no further treatment, a single intravesicular instillation of 2 mg MNU followed by treatment with 3% calcium carbonate in the diet, a single intravesicular instillation of 2 mg MNU followed by 2% sodium saccharin in the diet for 10 weeks and then increased to 4% of the diet for the remainder of the experiment or a single intravesicular instillation of 2 mg MNU followed by 2% sodium cyclamate in the diet for 10 weeks and then increased to 4% of the diet for the remainder of the experiment. Satellite groups of 12, 15 and 12 rats for the last three groups, respectively, received the same treatments as the other rats in the group but were examined periodically by radiography for the presence of stones. A group of 100 rats was untreated. The incidences of bladder tumours were 0/100 in untreated controls, 1/50 in water controls, 19/49 with MNU, 24/49 with MNU plus calcium carbonate and 25/50 with MNU plus sodium saccharin, indicating no significant increase in tumorigenic activity due to saccharin. Bladder stones were noted in 0, 0, 6, 3 and 10 rats, respectively. No correlation between the presence of stones and the induction of bladder tumours was found (Green & Rippel, 1979). [The Working Group noted the incomplete reporting of this experiment.]

Three groups of 63 female Wistar rats [age unspecified] were pretreated with 0.15 mL of a saturated solution of MNU in saline (10 mg/mL) instilled into the bladder. Two weeks later, rats were given 0 or 2 g/kg bw per day sodium saccharin in the drinking-water for two years; one group received saccharin prepared by the Maumee process and the second group received saccharin prepared by the Remsen-Fahlberg method (containing 40 mg/kg (ppm) ortho-toluenesulfonamide). By week 84, 22 controls, 43 animals given ‘Maumee’ sodium saccharin and 37 rats given ‘Remsen-Fahlberg’ sodium saccharin had died. An increase in the number of proliferative bladder lesions occurred in animals treated with MNU plus saccharin. The incidence of bladder neoplasia was not significantly different in the saccharin-treated groups, but the mean latent period was shorter (55 and 52 weeks versus 87 weeks). The animals were free from bladder parasites (Hooson et al., 1980). [The Working Group noted that the animals were not started on the test at weaning but had been fed a normal diet for several weeks before the start of the study.]

Male Fischer 344 rats weighing 130 g were kept for one week on basal diet and then given twice weekly intraperitoneal injections of 20 mg/kg bw MNU for four weeks followed by either 32 weeks of 0.05% phenobarbital or 5% sodium saccharin in the diet; given the same four-week treatment with MNU followed by 32 weeks of control diet; or given twice weekly injections of 5 mg/kg bw citrate buffer for four weeks and then either 0.05% phenobarbital or 5% sodium saccharin in the diet. The rats were killed at the end of 36 weeks. No information was provided on growth or survival. MNU followed by sodium saccharin produced papillary or nodular hyperplasia of the bladder in 9/19 (47%) rats, whereas no hyperplasia was seen in the bladders of other animals. Sodium saccharin did not increase the incidences of adenoma or adenocarcinoma of the thyroid, γ-glutamyl transferase-positive foci in the liver or lesions of the forestomach or glandular stomach when administered alone or after MNU. Phenobarbital increased the incidences of thyroid tumours and liver hyperplastic foci but not of neoplasia in the bladder. MNU by itself induced a high incidence of forestomach and glandular stomach foci and tumours and thyroid adenomas (Tsuda et al., 1983). [The Working Group noted the relatively short period of administration and lack of bladder tumour induction with this protocol.]

Female Sprague-Dawley rats, eight weeks of age, were treated with a single intravesicular instillation of 0.5 mg MNU (synthesized at the institute and quantified on the day of use) in 300 µL saline or with an intravesicular instillation of 300 µL of saline. Two days after dosing, the rats were fed diets containing either 0 (120 rats), 0.1% (150 rats), 0.5% (120 rats), 1% (90 rats), 2.5% (60 rats) or 5% (60 rats) sodium saccharin in the diet. Groups of 60 rats were also given 5% saccharin in the diet or sodium saccharin in the drinking-water at a concentration of 2, 3 or 4% to achieve maximum consumption and yet preserve the animals’ health. The 2% dose was then used for the remainder of the study. An additional group of 60 rats received four doses of MNU at weekly intervals and were then fed untreated diet for the remainder of the experiment. The experiment was terminated at the end of two years. Administration of MNU did not affect body weight. The average weight of rats receiving saccharin was lower than that of control animals, but the decrease exceeded 10% of the control weight only with 5% sodium saccharin and 5% saccharin in the diet and with 2% sodium saccharin in water. The mortality rates in these groups were not significantly different from each other or from those of the controls. Animals not treated with MNU lived longer than did those treated with MNU, and saccharin appeared to lower the mortality rate substantially throughout much of the study. Animals receiving sodium saccharin in the drinking-water had high, early morbidity and mortality. Two of the 233 rats given sodium saccharin in the diet but not pretreated with MNU developed bladder tumours [specific groups not specified]. The incidences in the group receiving sodium saccharin in the diet after MNU pretreatment were 17/106 (16%) with no sodium saccharin, 15/140 (11%) with 0.1%, 21/107 (20%) with 0.5%, 15/78 (19%) with 1%, 20/50 (40%) with 2.5% and 8/49 (16%) with 5%. The incidence in the group given sodium saccharin in the drinking-water was 9/46 (20%), that in the group given saccharin was 8/49 (16%), and that in the group given four intravesicular instillations of MNU was 24/57 (42%). A significantly increased incidence was seen only in the group given 2.5% sodium saccharin after MNU pretreatment (West et al., 1986). [The Working Group noted the unconventional statistical analysis of the incidence of bladder tumours, whereby animals dying during the course of the experiment were analysed separately from those killed at the end. The Working Group also noted the lack of an increased incidence of bladder with 5% sodium saccharin, in contrast to other reports with the MNU model. The authors attributed the lack of activity at the highest dose to toxicity, but the degree of toxicity was similar to that seen in other experiments with MNU at a dose of 5% sodium saccharin.]

Groups of 60 female Sprague-Dawley rats, eight weeks of age, were treated with a single instillation of 0.5 mg MNU in 300 µL saline or with 300 µL of saline alone. The MNU was synthesized freshly and quantified on the day of use. Sodium saccharin was administered in the diet by a variety of schedules: 0, 1, 2.5 or 5% was administered in the diet either four weeks immediately preceding, following or on the day of bladder instillation with MNU, and the animals were then maintained on control diet until the end of the experiment at 106 weeks of age; additionally, one group of rats were exposed neonatally through the milk of their mothers, which were given sodium saccharin in the diet for the three weeks of lactation, starting on the day of parturition, and were then fed control diet from weaning until the end of the experiment. An interim sacrifice was conducted at approximately 85 weeks of age, and the experiment was terminated when the rats were 112 weeks of age (106 weeks for the neonatal treatment). There was no apparent effect on body weight or on survival, and deaths were due primarily to the development of mammary tumours. The mortality rate was significantly lower than that of controls for animals given MNU and 1% sodium saccharin at 4–8 and 6–10 weeks, 2.5% sodium saccharin at 8–12 weeks or 5% sodium saccharin at 4–8 and 6–10 weeks. In the groups not given MNU, there were no deaths due to sodium saccharin treatment. Transitional-cell neoplasms of the urinary bladder were diagnosed only in animals treated with MNU; the incidences of bladder tumours ranged from 20 to 41% in the various groups, but there was no statistically significant increase in the incidence of bladder tumours between groups for any treatment period or for any dose of sodium saccharin (West et al., 1994).

3.4.4. N-[4-(5-Nitro-2-furyl)-2-thiazolyl]formamide

Groups of 20 male Fischer rats, four weeks of age received 0.2% N-[4-(5-nitro-2-furyl)-2-thiazolyl]formamide (FANFT) in powdered diet for six weeks, followed immediately by 5% sodium saccharin (containing < 0.03 mg/kg ortho-toluenesulfonamide) in powdered diet for up to 83 weeks, then standard diet up to 104 weeks; pretreatment with FANFT as above, followed by six weeks on standard diet, 5% sodium saccharin in the diet for 77 weeks, then standard diet up to 104 weeks; normal diet for six weeks, followed by 5% sodium saccharin in the diet for 83 weeks, then standard diet up to 104 weeks; or pretreatment with FANFT, followed by standard diet for 98 weeks. Forty-two untreated controls received standard diet for 104 weeks. The experiment was terminated after 104 weeks, at which time 6/20, 9/20, 19/20, 16/20 and 27/42 animals were still alive. The incidences of urothelial carcinomas were 18/19 and 13/18 in the groups given FANFT plus saccharin, 0/20 in the group receiving saccharin alone, 4/20 in the group receiving FANFT alone and 0/42 in the untreated controls. In addition, 1/19 and 1/18 animals given FANFT plus saccharin had urinary bladder sarcomas, and 1/20 given FANFT only had a bladder papilloma. The presence or absence of bladder parasites was not reported (Cohen et al., 1979).

Male Fischer rats, five weeks of age, were given 0.2% FANFT in the diet for four weeks followed by 5% sodium saccharin in the diet for 100 weeks; control diet for four weeks followed by sodium saccharin for 100 weeks; FANFT followed by 2% l-tryptophan; control diet followed by l-tryptophan; FANFT followed by control diet; FANFT for the entire course of the experiment; or control diet for the entire experiment. The incidences of bladder carcinoma were 5/26, 0/26, 2/26, 0/26, 0/25, 8/8 and 0/27, respectively, and the incidences of bladder papilloma were 2, 0, 3, 0, 1, 0 and 0, respectively (Fukushima et al., 1981).

Male Fischer 344 rats, five weeks of age, were treated as follows: 5% sodium saccharin plus 2% l-tryptophan in the diet (13 rats); 5% sodium saccharin plus 0.005% FANFT in the diet (16 rats); 5% sodium saccharin in the diet (20 rats); 2% l-tryptophan in the diet (16 rats); FANFT in the diet (11 rats); or were untreated (23 rats). The respective diets were fed for two years, at which time the remaining rats were killed. A significant decrease in weight gain was seen only in the group given sodium saccharin plus l-tryptophan. Bladder tumours were seen only in the group given sodium saccharin plus FANFT, in which three papillomas and two carcinomas were observed; five rats had nodular or papillary hyperplasia of the bladder. Two of 11 rats in the group given FANFT alone also had nodular or papillary hyperplasia (Murasaki & Cohen, 1983a). [The Working Group noted the small numbers of animals.]

Groups of 30 male Fischer rats, five weeks of age, were treated in two phases, the first lasting for six weeks and the second for 61 weeks with a one-week interval between them. The treatments were as follows: 0.5% aspirin for two days, 0.2% FANFT and 0.5% aspirin in the diet for six weeks, one week of aspirin followed by 5% sodium saccharin up to 68 weeks; 0.2% FANFT, one week of control diet followed by aspirin plus sodium saccharin; 0.2% FANFT, one week of control diet followed by sodium saccharin; 0.5% aspirin for two days, 0.2% FANFT plus aspirin, one week of aspirin followed by control diet; 0.2% FANFT followed by control diet; control diet for seven weeks followed by aspirin plus sodium saccharin; control diet for seven weeks followed by sodium saccharin; aspirin for the entire course of the experiment; or control diet for the entire course of the experiment (40 rats). The experiment was terminated at the end of 68 weeks. The incidences of transitional-cell carcinoma of the urinary bladder were 6 (20%), 8 (28%), 24 (83%), 3 (10%) and 3 (10%) in the five groups given FANFT, respectively, with none in the other groups. The incidences of papillomas were 1, 2, 1, 0 and 3 in the groups given FANFT, respectively, and 1 in the group given only aspirin. The total bladder tumour incidence was significantly lower in the groups given aspirin with FANFT when compared with that given sodium saccharin and FANFT, indicating inhibition by aspirin of the bladder tumorigenic effects of both FANFT and sodium saccharin (Sakata et al., 1986). Johansson et al. (1986) reported the effects on the kidneys of the rats in this study separately. The lesions reported included renal pelvic hyperplasia, metaplasia and atypia, calcification, renal papillary necrosis, vascular changes and rat nephropathy. It was noted that administration of sodium saccharin significantly inhibited the incidence and severity of nephropathy due to ageing but significantly increased the incidence and severity of hyperplasia and calcification in the renal pelvis. Aspirin was noted to increase the incidence and severity of renal papillary necrosis and vascular changes. [The Working Group noted the short period of administration of sodium saccharin.]

Groups of weanling male Fischer rats were treated for five-week pretreatment periods as follows: groups 1–4 (42 rats per group) were fed untreated diet and untreated water; groups 5–8 (42 rats per group) were given 0.2% FANFT in the diet with untreated drinking-water; and groups 9–12 (43, 41, 42 and 42 rats per group) were given untreated diet with 0.005% N-nitrosodibutylamine (NDBA) in the drinking-water. For the remaining 95 weeks of the experiment, the rats were all given untreated drinking-water and fed diets containing either no additional chemical (groups 1, 5 and 9), 0.05% phenobarbital (groups 2, 6, and 10), 0.15% phenobarbital (groups 3, 7 and 11) or 5% sodium saccharin (groups 4, 8 and 12). Body-weight gain was decreased by ≥ 10% in groups 3, 4, 5, 6, 7, 11 and 12 relative to their respective controls. There was slightly decreased survival in rats given phenobarbital, but survival was not affected by treatment with FANFT, NDBA or sodium saccharin. The incidences of bladder carcinoma were 0, 0, 0, 0, 8/40 (20%), 15/40 (38%), 19/39 (49%), 15/39 (38%), 5/43 (12%), 2/37 (5%), 9/40 (23%) and 27/40 (18%) for the 12 respective groups. The only group that showed a statistically significant increase in the incidence of bladder carcinomas relative to the respective controls was the group given FANFT followed by 0.15% phenobarbital. Benign papillomas were seen in 0, 0, 0, 0, 10, 10, 6, 3, 2, 2, 2 and 4 rats in the 12 respective groups; again, the only group that showed a significantly increased incidence of papillomas plus carcinomas when compared with the respective controls was group 7. Hepatocellular carcinomas were found in 0, 0, 0, 0, 0, 0, 2.5%, 2.6%, 0, 7%, 27%, 43% and 2.5%, respectively, and oesophageal squamous-cell carcinomas were observed in no animals in groups 1–8 and in 46, 32, 45 and 45% of the rats in groups 9–12, respectively (Imaida & Wang, 1986).

In a later report (Masui et al., 1990), all rats given 5 or 7.5% sodium saccharin in the diet were reported to have squamous-cell hyperplasia and hyperkeratosis at the limiting ridge of the forestomach. Further experiments indicated that administration of saccharin in various diets (AIN-76, Prolab, Purina and NIH-07), either as the acid or the sodium or calcium salt, nearly always produced squamous-cell hyperplasia at the limiting ridge but without progression to papilloma or carcinoma in both Fischer 344 and Sprague-Dawley rats. Control diets, in particular AIN-76 diet, also caused some hyperplasia of the limiting ridge.

Groups of 30 male Fischer 344 rats, five weeks of age, were treated as follows: 0.2% FANFT in pelleted AIN-76A diet (prepared on site) was fed for four weeks followed by an interval of one week of control diet, followed by 5% sodium saccharin in AIN-76A diet up to 104 weeks; 0.2% in AIN-76A diet for four weeks followed by AIN-76A diet up to 104 weeks; group 3 (only 25 rats), AIN-76A control diet for five weeks followed by 5% sodium saccharin in AIN-76A diet up to 104 weeks; group 4, 0.2% FANFT in Prolab 3200 diet for four weeks followed by one week of control Prolab diet followed by 5% sodium saccharin in Prolab diet up to 104 weeks; group 5, 0.2% FANFT in Prolab diet for four weeks followed by control Prolab diet up to 104 weeks; group 6, 0.2% FANFT in AIN-76A diet for four weeks followed by one week of Prolab control diet followed by 5% sodium saccharin in Prolab diet up to 104 weeks; group 7, 0.2% FANFT in AIN-76A diet for four weeks followed by Prolab control diet up to 104 weeks; group 8, AIN-76A control diet for four weeks followed by one week of Prolab control diet followed by 5% sodium saccharin in Prolab diet up to 104 weeks; group 9, 0.2% FANFT in AIN-76A diet for two weeks followed by one week of control Prolab diet followed by 5% sodium saccharin in Prolab diet up to 104 weeks; group 10, 0.2% FANFT in AIN-76A diet for one week followed by one week of control Prolab diet followed by 5% sodium saccharin in Prolab diet up to 104 weeks; group 11, 0.2% FANFT in AIN-76A diet for two weeks followed by one week of control Prolab diet followed by control Prolab diet up to 104 weeks; group 12, 0.1% FANFT in AIN-76A diet for two weeks followed by one week of Prolab control diet followed by 5% sodium saccharin in Prolab diet up to 104 weeks; group 13, 0.1% FANFT in AIN-76A diet for one week followed by one week of control Prolab diet followed by 5% sodium saccharin in Prolab diet up to 104 weeks; and group 14, 0.1% FANFT in AIN-76A diet for two weeks followed by control Prolab diet up to 104 weeks. The number of rats alive after 50 weeks of the experiment, when the first bladder tumour was detected, was considered to be the effective number of rats. A significant decrease in body weight and a significant effect on survival were seen only in group 3. Transitional-cell carcinomas of the urinary bladder were observed in 2/27 (7.4%), 5/30 (17%), 0/22, 9/30 (30%), 3/30 (10%), 9/30 (30%), 3/29 (10%), 1/30 (3.3%), 0/30, 0/29, 0/30, 2/30 (6.7%), 0 and 1/29 (3.5%) rats, respectively. In addition, papillomas were observed in 1, 2, 0, 3, 2, 1, 0, 1, 0, 0, 1, 0, 0 and 1 rats, respectively. The authors concluded that sodium saccharin promotes bladder cancer when administered to male rats in Prolab diet but not when administered in AIN-76A diet (Okamura et al., 1991). [The Working Group noted the absence of a group administered AIN-76A diet only throughout the experiment. They also noted the lack of bladder tumours in rats given sodium saccharin in AIN-76A diet (group 3), whereas two rats given sodium saccharin in Prolab diet (group 8) had one papilloma and one carcinoma of the bladder.]

Groups of 40 male Fischer 344 rats, six weeks of age, were treated in two phases, the first phase lasting six weeks and the second phase the subsequent 72 weeks. Chemicals were administered in Prolab 3200 diet in groups 1–18 and in NIH-07 diet in groups 19–20. Groups 1–14, 19 and 20 were treated with 0.2% FANFT in the diet and groups 15–18 with control diet in the first phase. The dietary treatments in the second phase were as follows: group 1, 5% sodium saccharin; group 2, 3% sodium saccharin; group 3, 5.2% calcium saccharin; group 4, 3.12% calcium saccharin; group 5, 4.21% saccharin; group 6, 2.53% saccharin; group 7, 5% sodium ascorbate; group 8, 4.4% ascorbic acid; group 9, 5% sodium saccharin plus 1.15% calcium carbonate; group 10, 5.2% calcium saccharin plus 1.34% sodium chloride; group 11, 5% sodium saccharin plus 1.23% ammonium chloride; group 12, 1.15% calcium carbonate; group 13, 1.34% sodium chloride; group 14, control diet; group 15, 5% sodium saccharin; group 16, 5.2% calcium saccharin; group 17, 4.21% acid saccharin; group 18, control diet (untreated controls); group 19, 5% sodium saccharin; and group 20, control diet. The experiment was terminated at the end of 78 weeks. The effective number of rats in each group was considered to be 40, but there were 39 in groups 1, 14, 16 and 18. In comparison with the untreated controls (group 18), all groups except groups 12, 14 and 20 showed decreased body-weight gain. The numbers of rats with urothelial bladder carcinomas were 38, 29, 21, 20, 19, 13, 30, 16, 37, 34, 11, 16, 27, 12, 1, 0, 0, 2, 38 and 21, respectively. The numbers of rats with carcinomas plus papillomas in the various groups were 39, 34, 27, 22, 24, 19, 32, 21, 38, 37, 16, 21, 32, 17, 1, 0, 0, 2, 38 and 26, respectively. Statistically significant enhancement of the incidence of bladder cancer was seen with sodium saccharin, calcium saccharin, sodium ascorbate, sodium saccharin plus calcium carbonate, calcium saccharin plus sodium chloride and sodium chloride. Sodium saccharin enhanced the bladder tumour incidence when given in either the Prolab or the NIH-07 diet. The authors noted that the increased urinary pH resulting from the administration of 0.2% FANFT in the diet remained after FANFT treatment was discontinued. Sodium saccharin, calcium saccharin and acid saccharin by themselves did not significantly increase the incidence of bladder tumours or other lesions, except for simple hyperplasia in the group given sodium saccharin (Cohen et al., 1991a). [The Working Group noted the limited period of administration of the test compound for evaluating single-generation carcinogenic effects.]

3.4.5. Combinations of N-[4-(5-nitro-2-furyl)-2-thiazolyl]formamide, cyclophosphamide, freeze-ulceration and sodium saccharin

Male Fischer 344 rats, five weeks of age, were treated as follows: group 1 (23 rats), freeze-ulceration of the urinary bladder on day 0, followed immediately by administration of 0.2% FANFT in powdered diet for two weeks and then 5% sodium saccharin for 102 weeks; group 2 (23 rats), freeze-ulceration of the urinary bladder on day 0, followed immediately by 0.2% FANFT in powdered diet for two weeks and then control diet for 102 weeks; group 3 (23 rats), freeze-ulceration of the urinary bladder on day 0, two weeks of control diet and then 5% sodium saccharin in the diet for 102 weeks; group 4 (23 rats), freeze-ulceration on day 0 followed by 104 weeks of control diet; group 5 (21 rats), 0.2% FANFT in powdered diet for two weeks followed by 102 weeks of 5% sodium saccharin in the diet; group 6 (22 rats), 0.2% FANFT in powdered diet for two weeks followed by control diet for 102 weeks; group 7 (21 rats), control diet for two weeks followed by 5% sodium saccharin in the diet for 102 weeks; group 8 (23 rats), 0.2% FANFT in powdered diet for two weeks followed immediately by freeze-ulceration of the urinary bladder and 5% sodium saccharin in the diet for 102 weeks; group 9 (23 rats), 0.2% FANFT in the diet for two weeks followed by freeze-ulceration of the urinary bladder and control diet for 102 weeks; group 10 (23 rats), freeze-ulceration on day 0 followed by 104 weeks of 5% sodium saccharin in the diet; group 11 (26 rats), intraperitoneal injection of 100 mg/kg bw cyclophosphamide in distilled water on day 0 followed immediately by 0.2% FANFT in the diet for two weeks and then 102 weeks of 5% sodium saccharin in the diet; group 12 (26 rats), intraperitoneal injection of 100 mg/kg bw cyclophosphamide followed immediately by 0.2% FANFT in the diet for two weeks and then control diet for 102 weeks; group 13 (25 rats), intraperitoneal injection of 100 mg/kg bw cyclophosphamide followed by two weeks of control diet and then 5% sodium saccharin for 102 weeks; group 14 (26 rats), intraperitoneal injection of 100 mg/kg bw cyclophosphamide followed immediately by 104 weeks of control diet; group 15 (25 rats), intraperitoneal injection of 100 mg/kg bw cyclophosphamide followed immediately by 104 weeks of 5% sodium saccharin; group 16 (37 rats), 104 weeks of control diet (untreated control group); group 17 (21 rats), 5% sodium saccharin for 104 weeks. The experiment was terminated after 104 weeks. The survival rates of groups 12, 14 and 15 were significantly decreased, with 19, 16, 16, 20, 18, 16, 16, 19, 16, 16, 17, 0, 14, 4, 9, 28 and 14 rats alive at the end of the experiment in the various groups, respectively. The effective number of rats was considered to be the number alive after 76 weeks of the experiment, when the first bladder tumour was detected. The incidences of bladder carcinomas were 4/23 (17%), 0/22, 5/20 (25%), 0/23, 1/21 (5%), 0/19, 0/17, 8/22 (36%), 2/21 (10%), 2/21 (10%), 4/22 (18%), 0/9, 3/17 (18%), 0/7, 3/17 (18%), 0/32 and 0/20, respectively. The numbers of rats with urinary bladder papillomas were 1, 1, 2, 3, 2, 1 and 3, in groups 3, 8, 10, 11, 13, 14 and 15, respectively. Sodium saccharin enhanced the incidence of bladder tumours when administered after FANFT, cyclophosphamide or freeze-ulceration, but sodium saccharin alone administered at a concentration of 5% in the diet to rats for 102 or 104 weeks did not produce benign or malignant tumours of the urinary bladder (Cohen et al., 1982).

3.4.6. N-Nitrosobutyl-N-(4-hydroxybutyl)amine or N-nitrosodibutylamine

Groups of 29–32 male and 29–32 female Fischer 344 rats, six weeks of age, were treated with 0.01% N-nitrosobutyl-N-4-(hydroxybutyl)amine (NBHBA) in the drinking-water for four weeks (groups 1–5) or received no treatment (groups 6–10), followed by 32 weeks of sodium saccharin in the diet (purity, > 99.5%, containing 7 ppm ortho-toluenesulfonamide) at concentrations of 5% (groups 1 and 6), 1% (groups 2 and 7), 0.2% (groups 3 and 8) or 0.04% (groups 4 and 9) or untreated diet (groups 5 and 10), respectively. All rats were killed at the end of 36 weeks. No bladder lesions were seen in the groups without pretreatment with NBHBA. In males, the incidences of papillomas were 6, 10, 8, 11 and 8, respectively, for groups 1–5. The incidences of papillary or nodular hyperplasia were 24, 19, 14, 12 and 11, respectively, which was significant (p < 0.001) only at the 5% dose when compared with NBHBA only. The number of hyperplastic lesions per 10 cm of basement membrane was also significantly increased at the 5% dose. Papillomas were seen in two females in group 2 and one in each of groups 1 and 3–5. The incidences of papillary or nodular hyperplasia were 23, 17, 14, 11 and 9 in groups 1–5, respectively; the increases were statistically significant only at 1 and 5%. The authors concluded that sodium saccharin caused dose-related promotion in both males and females, only for hyperplastic lesions and not for bladder papillomas (Nakanishi et al., 1980a). [The Working Group noted the lack of effect on bladder tumours as an end-point.]

Male Wistar rats, eight weeks of age, were treated with 0.01% NBHBA in the drinking-water (groups 1–4) or with untreated drinking-water (groups 5–8) for four weeks and then received the following treatments for 32 weeks: 5% sodium saccharin (purity, > 99.5% with 7 ppm ortho-toluenesulfonamide) in the diet (groups 1 and 5); 0.1% caffeine in the drinking-water (groups 2 and 6); 5% sodium saccharin in the diet plus 0.1% caffeine in the drinking-water (groups 3 and 7); or untreated drinking-water and diet (groups 4 and 8). The authors stated that the water consumption was increased by approximately 50% in the groups treated with sodium saccharin when compared with other groups. No difference in growth or survival in the various groups was described. Rats that were not pretreated with NBHBA had no bladder tumours, but 5/26 rats treated with sodium saccharin (group 5) and 7/23 given sodium saccharin plus caffeine (group 7) had papillary or nodular hyperplasia compared with 0/12 in group 8. In group 1, 1/31 rats had a bladder carcinoma; and the incidences of bladder papillomas in groups 1–4 were 9/31 (29%), 6/28 (21%), 4/30 (13%) and 3/23 (13%), respectively; these were not statistically significantly different from each other. There were significant increases in the incidences and numbers of papillary/nodular hyperplasia in groups 1 and 3 compared with group 4. In a second experiment with 0.001% NBHBA, combinations of NBHBA, sodium saccharin and caffeine were administered simultaneously, with NBHBA and caffeine in the drinking-water and sodium saccharin in the diet. The treatments were as follows: group 1, NBHBA plus sodium saccharin plus caffeine; group 2, NBHBA plus sodium saccharin; group 3, NBHBA plus caffeine; group 4, NBHBA; group 5, sodium saccharin plus caffeine; group 6, sodium saccharin; group 7, caffeine; and group 8, no added chemical (untreated controls). The incidences of bladder papillomas were 9/32 (28%) and 10/24 (42%) in groups 1 and 2, respectively, compared with 0/22 in BBN controls, and the incidences of bladder carcinoma were 1/32 (3.1%), 2/24 (8.3%) and 0/22, respectively. No bladder tumours were induced in the other groups (Nakanishi et al., 1980b). [The Working Group noted that the increased water consumption associated with sodium saccharin in the diet resulted in increased exposure of groups 1 and 2 to NBHBA, and this might have accounted for the increased incidences of tumours and other bladder proliferative lesions in these groups.]

Groups of 30 male Fischer 344 rats, six weeks of age, were treated with 0.01% NBHBA in the drinking-water for four weeks followed by either 5% sodium saccharin or 0.5% sodium saccharin in the diet for 34 weeks; a further group received NBHBA followed by basal diet, and additional groups received the same treatments after four weeks of untreated drinking-water. The body-weight gain of rats treated with 5% sodium saccharin with or without NBHBA was slightly decreased, but there was no apparent effect on survival. No bladder lesions were seen in rats that were not pretreated with NBHBA. The incidences of bladder papillomas were 20% with NBHBA plus 5% sodium saccharin, 33% with NBHBA plus 0.5% sodium saccharin and 33% with NBHBA alone, and the incidences of papillary or nodular hyperplasia were 83%, 47% and 45%, respectively (Fukushima et al., 1983b). [The Working Group noted the lack of effect on bladder tumour incidence and the relatively short duration of the experiment.]

Male Fischer 344 rats, six weeks old, were treated in two phases, the first phase consisting of administration of drinking-water with or without NBHBA for four weeks, followed by treatment with 5% sodium saccharin in the diet or untreated control diet for 32 weeks. The first group received 0.05% NBHBA followed by 5% sodium saccharin; group 2 received 0.05% NBHBA followed by untreated control diet; group 3 received 0.01% NBHBA followed by 5% sodium saccharin in the diet; group 4 received 0.01% NBHBA in the drinking-water followed by untreated control diet; and group 5 received untreated water followed by 5% sodium saccharin in the diet. No bladder tumours were observed in groups 3–5. Twenty of 25 rats in group 1 (80%) and 1/26 (3.8%) rats in group 2 developed bladder carcinomas. The number of papillary or nodular hyperplasias was increased in the group given 5% sodium saccharin after 0.05% NBHBA over that with the same dose of NBHBA alone (100% versus 46%); there were also significant increases in the incidences and numbers of hyperplasias in the rats pretreated with 0.01% NBHBA followed by sodium saccharin (83% versus 39%) (Ito et al., 1983). [The Working Group noted the limited details reported for this experiment.]

Male Fischer 344 rats were treated as follows: group 1, 0.02% AAF in the diet for four weeks followed by 0.05% phenobarbital in the diet for 32 weeks; group 2, AAF followed by 5% sodium saccharin in the diet; group 3, AAF followed by untreated control diet; group 4, 0.01% NBHBA in the drinking-water for four weeks followed by 32 weeks of phenobarbital; group 5, NBHBA followed by 5% sodium saccharin; group 6, NBHBA followed by control diet; group 7, untreated water and diet followed by phenobarbital; group 8, untreated water and diet followed by 5% sodium saccharin in the diet; and group 9, untreated controls. No bladder papillomas were seen in groups 1–3 or 7–9. The incidences of bladder papillomas in groups 4–6 were similar: 9/30 (30%), 6/29 (21%) and 8/28 (29%), respectively. The incidences of papillary or nodular hyperplasia were 14 (47%), 24 (83%) and 11 (39%), respectively, in these groups; additionally, 4/29 (14%) rats in group 2 had papillary or nodular hyperplasia. Liver hyperplastic nodules were observed in 24/24, 27/29, 27/28 and 8/30 rats, respectively, in groups 1–4, with no hyperplastic nodules in groups 5–9. Five rats in group 1 had hepatocellular carcinoma, whereas none were seen in other groups. The authors concluded that phenobarbital enhances liver hyperplasia and sodium saccharin enhances bladder hyperplasia (Ito et al., 1983). [The Working Group noted the lack of effect on bladder tumour incidences and the limited description of the details of this experiment.]

Male Fischer 344 rats, seven to eight weeks of age, received untreated water and control diet (group 1), control diet and 0.02% NDBA in the drinking-water (group 2), 5% sodium saccharin in the diet and untreated drinking-water (group 3) or 5% sodium saccharin in the diet and 0.02% NDBA in the drinking-water (group 4) for a total of 26 weeks with interim sacrifices at four and eight weeks. The end-points of the study were effects on γ-glutamyl transferase-positive foci, hyperplastic nodules and hepatocellular carcinomas in the liver. The number of foci at 26 weeks were 0.31 ± 0.31 (10 rats), 5.03 ± 1.06 (29 rats), 0.23 ± 0.23 (36 rats) and 31.3 ± 4.65 (21 rats), respectively. Similar trends were seen at four and eight weeks. The incidences of hyperplastic nodules were 0/11, 5/29 (17%), 0/36 and 17/21 (81%), respectively, and those of carcinomas were 0, 1 (3%), 0 and 17 (81%), respectively. A major confounding factor in the interpretation of these data is the marked increase in water consumption of rats given sodium saccharin. In group 4, this would have led to a marked increase in the consumption of NDBA, which was administered in the drinking-water, in comparison with group 2; however, the water consumption of these animals was not determined (Pereira et al., 1983). [The Working Group noted the confounding issue of water consumption and therefore the dose of NDBA, suggesting that the effect on liver lesions might have been due to increased NDBA consumption rather than a direct effect of sodium saccharin.]

Groups of 25 male Fischer 344 rats, six weeks of age, were treated for four weeks with 0.01% NBHBA in the drinking-water followed by 32 weeks of 5% sodium saccharin in the diet or control diet for 32 weeks. A third group received no treatment for the first four weeks of the experiment followed by 32 weeks of 5% sodium saccharin in the diet. Rats treated with 5% sodium saccharin had a slight decrease in body-weight gain, but their survival was not affected. Papillomas were seen in five and six rats in groups 1 and 2, respectively, and none in group 3. The incidences of papillary or nodular hyperplasia were 23, 9 and 0 in groups 1–3, respectively (Hagiwara et al., 1984). [The Working Group noted the lack of effect on bladder tumours but the increased incidence of nodular or papillary hyperplasia.]

Male Fischer 344 rats, six weeks old, were treated as follows: groups 1–4 were treated with 0.01% NBHBA in the drinking-water for four weeks; group 5 received no treatment during this period. Group 1 then received consecutive 10-week courses of 5% sodium saccharin, 2% dl-tryptophan and 5% sodium ascorbate in the diet, followed by two weeks of control diet; group 2 received consecutive 10-week courses of sodium saccharin and dl-tryptophan followed by basal diet; group 3 received one 10-week course of sodium saccharin followed by basal diet; and group 5 received consecutive 10-week courses of sodium saccharin, dl-tryptophan and sodium ascorbate followed by basal diet. All rats were killed at the end of 36 weeks. There was an increased incidence of bladder carcinomas, papillomas and papillary or nodular hyperplasia in group 1 but not in the other groups. No bladder lesions were seen in group 5. The incidences of carcinoma were 7/25 (28%), 2/24 (8.3%), 2/25 (8%), 0 and 0, respectively (Sakata et al., 1984). [The Working Group noted the short period of administration of each chemical and the short duration of the experiment.]

Male Fischer 344 rats, six weeks of age, were pretreated with 0.05% NBHBA in the drinking-water for four weeks and then treated with chemicals in the diet for 32 weeks: group 1, untreated controls; group 2, 5% sodium saccharin in the diet; group 3, 5% sodium ascorbate in the diet; group 4, sodium saccharin plus sodium ascorbate in the diet; group 5, 5% ascorbic acid in the diet; and group 6, sodium saccharin plus ascorbic acid in the diet. All animals were killed at the end of 36 weeks. There was significant (> 10%) reduction in body-weight gain in all treated animals compared with controls. This was particularly marked in groups 4 and 6 in which the body-weight gain was decreased by approximately 20%. The incidences of bladder carcinomas were 0/15, 5/15 (33%), 11/16 (69%), 13/13 (100%), 2/16 (12%) and 1/16 (6.3%), respectively. The authors concluded that there was significant synergy between sodium saccharin and sodium ascorbate and that the significant inhibition seen with ascorbic acid might have been related to the decreased urinary pH (Fukushima et al., 1990). [The Working Group noted the short period of administration and the small number of animals in each group.]

Male Fischer 344 rats, six weeks of age, were treated in two phases. The first phase lasted four weeks and consisted of treatment with either control drinking water (groups 1–4) or drinking-water containing 0.05% NBHBA (groups 5–8). The second phase consisted of 36 weeks of treatment with either control diet (groups 1 and 5), 5% sodium saccharin in the diet (groups 2 and 6), 5% sodium saccharin plus 0.1% nordihydroguaiaretic acid in the diet (groups 3 and 7) or 0.1% nordihydroguaiaretic acid in the diet (groups 4 and 8). The groups fed sodium saccharin plus nordihydroguaiaretic acid had significant growth retardation, but survival was not affected. No bladder hyperplastic lesions or tumours were seen in groups 1–4 (11 rats per group), which were not pretreated with NBHBA. In groups 5–8, the incidences of bladder carcinoma were 1/20 (5%), 2/23 (8.7%), 1/22 (4.5%) and 4/20 (20%), respectively; the incidences of papillomas were 2, 5, 0 and 3, respectively; and the incidences of papillary or nodular hyperplasia were 6, 19, 12 and 7, respectively. The authors concluded that nordihydroguaiaretic acid had anti-tumour promoting activity (Yu et al., 1992). [The Working Group noted the inadequate number of animals, the short duration of the experiment and the lack of effect based on incidences of carcinoma.]

3.4.7. Urethane

In a bioassay to screen for lung adenomas, groups of 25 male A/St mice, eight weeks of age, were treated with one of four commercial saccharin preparations (Sweeta tablets, pharmaceutical powder, Sweet 10 liquid or Sweet-n-Low powder). The saccharin preparations were made up freshly each week in distilled water and administered at a dose of 1 g/kg bw daily by gavage on five days a week for a total of 17 weeks. Three groups of controls received distilled water. In the first series, no treatment other than saccharin or water was administered. In a second series, the mice also received an intraperitoneal injection of 0.1 mg/g bw urethane in 0.9% saline; the third group also received an intraperitoneal injection of 1 mg/g bw urethane. The urethane injections were given one week after initiation of saccharin administration. The incidences and numbers of adenomas per lung were determined; a few lungs from each group of mice were examined microscopically to confirm that the adenomas were being counted. The animals generally survived the treatment well, most of the deaths being due to trauma induced by gavage. The animals exposed to saccharin preparations gained somewhat less weight than the control animals during the course of this experiment, but their weight gain was substantial. The commercial saccharin preparations did not elicit a significant lung tumour response when given alone, with respect to either the incidence or number of lung tumours per mouse. At the low dose of urethane, the only sweetener that produced a lung tumour response greater than that in animals treated with urethane alone was Sweet-n-Low, which produced an approximately twofold increase in the number of lung tumours, although the incidence (47%) was the same as in the control group (52%) and was not statistically significant. At the high dose of urethane, all of the saccharin preparations significantly increased the number of lung tumours per mouse over that in animals treated with urethane alone: 2.13 ± 0.44, 4.36 ± 0.70, 5.95 ± 1.29, 6.08 ± 1.29 and 9.86 ± 3.65 in mice treated with 1 mg/g urethane plus either distilled water, Sweeta tablets, pharmaceutical powder, Sweet 10 liquid or Sweet-n-Low powder, respectively. The tumour incidences, however, were nearly the same and were not statistically different between the groups (Theiss et al., 1980). [The Working Group noted the complex nature of the protocol, the lack of effect at the low dose of urethane and the lack of effect on incidence at the high dose of urethane and that there was no effect when the sweetener preparations were used without administration of urethane.]

3.4.8. N-Methyl-N′-nitro-N-nitrosoguanidine

Groups of 20 treated and 40 control male Wistar rats, seven weeks of age, were given 10 mg/L N-methyl-N′-nitro-N-nitrosoguanidine (MNNG) in the drinking-water and 10% sodium chloride in the diet for eight weeks (groups 1–5); groups 6–10 received neither MNNG nor sodium chloride during the first eight weeks. After these treatments, the rats in groups 1 and 6 were continued as the respective control groups without any further dietary supplementation for the remaining 32 weeks of the experiment (total experimental period of 40 weeks), animals in groups 2 and 7 were fed 10% sodium chloride in the diet, animals in groups 3 and 8 were fed 5% saccharin [form not specified] in the diet, groups 4 and 9 were fed 0.05% phenobarbital in the diet, and groups 5 and 10 received 1% aspirin in the diet. The growth of the animals was significantly retarded by the administration of MNNG plus sodium chloride; these animals showed a compensatory increase in growth rates once the treatments ended, although their weights at the end of the experiment were still lower than those in the groups not treated with MNNG. During the second phase of the experiment, the growth was similar, except for rats receiving the aspirin-supplemented diet, which showed a marked retardation of growth compared with the other groups. No gastric or duodenal tumours occurred in the rats that were not pretreated with MNNG plus sodium chloride. The incidences of gastric adenocarcinomas were 3/39 (7.7%), 4/20 (20%), 3/20 (15%), 1/19 (5.3%) and 0/20, respectively, for groups 1–5, and 3, 3, 1, 1 and 1 rats had duodenal adenocarcinomas, respectively. Only sodium chloride produced a significantly increased incidence in the total number of rats bearing tumours, including those of the stomach plus duodenum. In saccharin-treated rats, diffuse alterations in the fundic mucosa were observed, in which the thickness of the zone of proliferating cells was increased (Takahashi et al., 1984).

3.4.9. Betel nut

Male and female C17 mice, 10–12 weeks of age, were either untreated (22 females and 12 males), received a diet containing 10% of a commercial preparation of saccharin-coated betel nut [no information on form or amount of saccharin] (18 females and 14 males), received 0.2 mL of a 0.1% solution of 1, 4-dinitrosopiperazine by oral intubation (14 females and 15 males) or received saccharin-coated betel nut powder in the diet plus 1,4-dinitrosopiperazine by gavage (12 females and 12 males). The mice were treated for 40 weeks and then allowed to live a normal life span on standard diet. There was no significant difference in mortality rate between the groups [overall survival was not indicated]. The incidences of squamous-cell carcinoma of the stomach were 0, 2, 7 and 6 in males and 0, 0, 4 and 3 in females. Reticular-cell neoplasms-type A of the uterus occurred in 2, 3, 5 and 3 female mice in groups 1–4, respectively (Pai et al., 1981). [The Working Group noted the inadequate numbers of animals, the incomplete reporting of the experiment, the lack of information on exposure to saccharin, the lack of a group given saccharin only and the lack of information about the form of saccharin administered.]

3.4.10. Freeze-ulceration

Eight groups of 40 male Fischer rats, five weeks of age, were treated as follows: group 1, freeze-ulceration of the urinary bladder followed immediately by 5% sodium saccharin in powdered diet for 104 weeks; group 2, freeze-ulceration of the urinary bladder followed by two weeks of control diet and then 102 weeks of 5% sodium saccharin in the diet; group 3, freeze-ulceration of the urinary bladder followed by four weeks of control diet and then 100 weeks of 5% sodium saccharin in the diet; group 4, freeze-ulceration of the urinary bladder followed by six weeks of control diet and then 98 weeks of 5% sodium saccharin in the diet; group 5, freeze-ulceration of the urinary bladder followed by 18 weeks of control diet and then 86 weeks of 5% sodium saccharin in the diet; group 6, 104 weeks of 5% sodium saccharin in the diet; group 7, freeze-ulceration of the urinary bladder followed by 104 weeks of control diet; and group 8, 104 weeks of control diet (untreated control group). None of the treatments had a significant effect on survival or on body weight. The experiment was terminated after 104 weeks. The incidences of urothelial carcinomas were 11/36, 6/36, 12/40, 7/36, 9/39, 0/39, 1/39 and 0/39 in the eight groups, respectively. In addition, papillomas were found in two rats in group 2 and one rat in group 6. There were no significant differences in the incidences of bladder tumours between groups 1–5. Only one benign papilloma developed in the urinary bladder of rats fed 5% sodium saccharin alone throughout the experiment (group 6) (Hasegawa et al., 1985).

(a) Saccharin

[³⁵S]Sodium saccharin instilled into the bladder of male rats was absorbed into the plasma (Colburn, 1978).

[¹⁴C]Saccharin administered by intravenous infusion to five rhesus monkeys at a dose of 4 µg/kg bw per min for 60 min during the last trimester of pregnancy crossed the placenta rapidly and was distributed in all fetal tissues except the central nervous system. During the infusion period, the fetal blood concentrations were approximately 30% those of the mothers. In contrast to the maternal organism, in which radiolabel disappeared quickly after infusion ended, the fetal compartment showed very slow clearance of saccharin and, 2 h after termination of the infusion, the fetal blood concentrations were higher than the maternal ones. The slow rate of fetal clearance suggests that considerable accumulation might result from repetitive maternal ingestion. No data were available on the penetration of saccharin into the embryonic compartment during organogenesis (Pitkin et al., 1971a).

In male Sprague-Dawley rats fed 5% saccharin [about 2500 mg/kg bw per day] in the diet for 66 days, the plasma concentration showed a twofold diurnal variation. In three male rats given diets containing up to 10% saccharin [about 5000 mg/kg bw per day] in the diet for 22 days and one group of females exposed to 5% saccharin, the concentrations of saccharin were higher in the gut wall, kidneys and urinary bladder than in the plasma. There was evidence of decreased clearance at the highest doses. Probenecid was found to inhibit clearance (Sweatman & Renwick, 1980).

The tissue distribution of saccharin was investigated in two-generation feeding studies in Sprague-Dawley rats. After a single oral dose of 50 mg/kg bw [³H]sodium saccharin dihydrate in late pregnancy, the concentrations of saccharin were lower in fetal than in maternal tissues after 6 and 12 h; however, concentrations were higher in fetal urinary bladder and liver after 24 and 48 h. Dams were exposed to 5% sodium saccharin [2500 mg/kg bw per day] in the diet beginning four weeks before mating and were either killed during late gestation or continued on treatment through the F₁ generation until 22 days of age. The concentration of saccharin in both maternal and fetal bladder walls showed wide interindividual variations, but the average concentration in fetal tissue was about twofold greater. The saccharin content of the stomachs of neonatal animals was very low (0.03%) and almost 200 times below the maternal dietary level. The tissue and urinary levels reflected these low dietary levels, but there was a large increase at 23 days, reflecting separation from the mother. In males, the urinary level increased over 10-fold from the neonatal period to 23 days of age, whereas the increase was somewhat less in females. During several time intervals up to 109 days of age, F₁ animals showed only equivocal evidence of accumulation of saccharin in the bladder when compared with other tissues (Sweatman & Renwick, 1982).

[³⁵S]Saccharin was observed to cross the placenta and enter the fetal circulation of Sprague-Dawley rats after treatment by gavage with 100 mg saccharin mixed with the radiolabelled compound (100 µCi, 266 mCi/mmol) on day 19 of gestation of animals that had been receiving 5% saccharin in the diet since gestation day 14. The concentrations in the fetal blood represented about 0.008% of the total dose up to 5 h after exposure. In contrast, the maternal blood levels during this time were 0.03–0.04% of the total dose (West, 1979).

The recovery of unmetabolized sodium saccharin was determined under conditions simulating one of the two-generation bioassays of carcinogenicity in which positive results were obtained (Arnold et al., 1980). Male rats that had been exposed to a 5% acid saccharin diet [about 2500 mg/kg bw per day] in utero and maintained on this diet throughout the experiment were given 5% [³H]sodium saccharin in the diet for 24 h when they weighed 290 g and were then returned to the unlabelled diet. Urine and faeces were collected at 24-h intervals and analysed. Almost all of the tritium was eliminated within 48 h, with 13–14% in the faeces and the remainder in the urine. Experiments with single low doses of [³H]sodium saccharin and dosing after administration of 3-methylcholanthrene gave similar results. All of the urinary tritium was found in saccharin by thin-layer chromatography, with a limit of detection of 0.4% of the administered dose (Sweatman & Renwick, 1979).

The renal excretion of sodium saccharin infused into adult male and female Sprague-Dawley rats was found to exceed that of inulin at all plasma concentrations. Maximal tubular secretion was clearly demonstrable at a concentration of 14–20 mg/100 mL, and there was no evidence of tubular reabsorption. Clearance was inhibited when sodium saccharin and para-aminohippurate were infused simultaneously, indicating that excretion occurs via the carrier-mediated tubular organic anion transport system. No differences between male and female rats were found in secretory patterns or urinary saccharin levels (Bourgoignie et al., 1980).

Purified [¹⁴C]sodium saccharin was administered to six-month-old male and female Osborne-Mendel rats as a single oral dose of 5, 50 or 500 mg/kg bw. Between 0.03 and 0.55% of the radiolabelled sodium saccharin was metabolized to carbon dioxide. In groups of Charles River CD rats fed diets containing 0.01, 0.1 or 1% sodium saccharin for one year before administration of a radiolabelled dose of 5, 50 or 500 mg/kg bw per day, recovery of radiolabel in urine and faeces was 80–96% after seven days, which was slightly lower than the recovery (95–100%) from rats that had not been fed any saccharin before the test dose. Identification of metabolites in the urine showed that > 99% of the sodium saccharin remained as saccharin; two metabolites were reported: ortho-sulfamoylbenzoic acid and (probably) benzenesulfonamide. [The Working Group noted that the possibility could not be excluded that these represented contaminants of the batch of sodium saccharin administered.] The highest counts of radiolabel after administration of [¹⁴C]saccharin were found in the kidney and urinary bladder, and the metabolic profiles in dogs, rabbits, guinea-pigs and hamsters were similar (Lethco & Wallace, 1975).

Two young male SIV50 Sprague-Dawley-derived rats were given single doses of 372 and 390 mg/kg bw [³⁵S]sodium saccharin by gavage and killed 50 h later. At least 99.6% of the radiolabel was found by thin-layer chromatography to correspond to unmetabolized saccharin. No binding of saccharin to DNA of liver or urinary bladder was found (Lutz & Schlatter, 1977).

Saccharin was not metabolized by liver microsomal preparations or by faecal homogenates taken from rats fed 1% hydrated sodium saccharin [approximately 0.2 g per day] in the diet for two years (Ball et al., 1977). The accumulation of saccharin by rat renal cortical tissue incubated in vitro was dependent on oxygen and was reduced by metabolic inhibitors, suggesting that saccharin is eliminated by active tubular secretion (Goldstein et al., 1978).

Saccharin was excreted rapidly and unchanged in the urine by rhesus monkeys (Pitkin et al., 1971b). More than 90% of an administered dose was found within 24 h after dosing in the urine of guinea-pigs; about 70% was found in rat urine and the remainder in the faeces (Minegishi et al., 1972). Although saccharin is rapidly excreted by rats, administration of multiple doses per day over several days resulted in accumulation in the bladder; however, it was completely cleared within three days after removal of saccharin from the diet (Matthews et al., 1973).

Another study showed that 90% of [¹⁴C]sodium saccharin was excreted unchanged in the urine by rats and monkeys of each sex. Pretreatment with phenobarbital to induce mixed-function oxidase activity or prior intake of daily doses of sodium saccharin for over two years did not influence the metabolism of a single oral dose of [¹⁴C]saccharin (Byard & Goldberg, 1973).

(b) Impurities of saccharin

The rates at which seven intragastrically administered impurities of saccharin (radiolabelled ortho-toluenesulfonamide, benz[d]isothiazoline-1,1-dioxide, 3-aminobenz[d]isothiazole-1,1-dioxide, 5-chlorosaccharin, toluene-2- and -4-sulfonamides and 2- and 4-sulfamoylbenzoic acid) were eliminated in rats were similar. At doses ranging from 20 to 80 mg/kg bw, 80–95% of the impurities were recovered within 24 h in urine and faeces; urinary metabolites of these impurities were identified (Ball et al., 1978; Renwick, 1978; Renwick & Williams, 1978; Renwick et al., 1978).

In female Wistar rats given single oral doses of 20, 125 or 200 mg/kg bw [¹⁴C]ortho-toluenesulfonamide, 79, 58 and 36% of the activity was recovered in 24-h urine samples; elimination at 24–48 h was 7, 14 and 33% of the dose, respectively. Within seven days, 4.5, 5.9 and 7% of the activity was recovered from the faeces. The main metabolites in the urine were 2-sulfamoylbenzyl alcohol and its sulfate or glucuronic acid conjugates (80%), N-acetyltoluene-2-sulfonamide (6%), saccharin (3%) and 2-sulfamoylbenzoic acid (2%) (Renwick et al., 1978).

In a similar study, 50% of administered ortho- and para-toluenesulfonamides excreted in urine had been metabolized to ortho- and para-sulfamoylbenzoic acids, respectively (Minegishi et al., 1972).

Low oral doses of 0.2–0.4 mg/kg bw [¹⁴C]ortho-toluenesulfonamide were excreted more slowly in humans than in rats, an average of 56% of the activity being excreted in the urine within 24 h and almost 90% within 48 h. Less than 1% of the activity was found in faeces. The main urinary metabolites were 2-sulfamoylbenzoyl alcohol (unconjugated, 7%; conjugated with glucuronic acid, 11%; conjugated with sulfate, 20%), saccharin (35%) 2-sulfanoylbenzoic acid (4%) and N-acetyltoluene-2-sulfonamide (2%) (Renwick et al., 1978).

(a) Cell proliferation

In a two-generation study, the parents of the F₀ generation of Charles River CD rats were fed diets containing 0, 0.01, 0.1, 1, 5 or 7.5% sodium saccharin [0, 5, 50, 500, 2500 or 3750 mg/kg bw per day] from the weanling stage through mating and gestation to weaning of their litters. Of the F₁ generation, 48 animals per dose per sex were exposed to the same concentrations of sodium saccharin for life. The authors reported a statistically significant increase in urinary bladder hyperplasia only in female rats of the F₁ generation exposed to 7.5% sodium saccharin. The hyperplasia found was focal rather than diffuse and not morphologically precancerous. The incidence of hyperplasia in males was not statistically significantly increased, possibly because of a high incidence of hyperplasia in the male controls. A statistically significant increase in the incidence of urinary bladder neoplasms was found in males fed 7.5% sodium saccharin (Taylor et al., 1980).

In a two-generation bioassay of sodium saccharin, 2500 F₁ male Charles River CD rats were divided into groups of 700 to 125 animals, the group at the lowest dose having the largest number. Dietary administration of 1, 3, 4, 5, 6.25 or 7.5% sodium saccharin [500, 1500, 2000, 2500, 3125 or 3750 mg/kg bw per day] was begun at six weeks for males and females of the F₀ generation; the F₁ generation was fed the same diet as the F₀ generation until the terminal kill at 30 months of age. The incidences of simple, papillary and nodular hyperplasia were analysed. Sodium saccharin increased the incidence of the combination of ‘any hyperplasia’ at 5, 6.25 and 7.5%, and the incidence in the last two groups achieved statistical significance (Schoenig et al., 1985; re-examined by Squire, 1985).

No simple, papillary or nodular hyperplasia or papilloma was found among groups of 10-week-old male and female inbred Charles River Fischer 344 rats fed 0.04, 0.2, 1 or 5% sodium saccharin [20, 100, 500 or 2500 mg/kg bw per day] in a stock diet for 32 weeks (Nakanishi et al., 1980a). In another experiment, two groups of 10-week-old male Wistar rats were exposed to 5% sodium saccharin in the diet for 36 weeks. In comparison with controls, which had no hyperplasia, sodium saccharin produced simple hyperplasia in 10/26 (38%) of one group and 11/21 (52%) of the other. The incidence of papillary or nodular hyperplasia was 5/26 (19%) and 9/21 (43%) in the two groups, respectively. The discrepancy between these results may be due to a strain difference in susceptibility (Nakanishi et al., 1980b).

In a time-course study, six-week-old male Charles River Fischer 344 rats were exposed to 5% sodium saccharin [2500 mg/kg bw per day] in the diet for up to 18 weeks. Three animals per group were killed for histopathological analysis at 1, 3, 5, 7, 9, 12, 15 and 18 weeks, and DNA synthesis was measured at 1, 5, 9 and 18 weeks by [³H]thymidine uptake and autoradiography. In exposed animals, vacuolar degeneration of the epithelial cells was found at three weeks, followed by simple hyperplasia at five weeks. By nine weeks, the degree of hyperplasia had increased, with a display of mitotic figures, hyperplastic foci and pleomorphic microvilli. Increased [³H]thymidine uptake was seen in the exposed rats at all times, and the increase was five- to eightfold the rate seen in controls (Fukushima & Cohen, 1980).

In a study of dose–response, cell proliferation was measured by autoradiography and scanning electron microscopy in groups of three to four five-week-old male Fischer 344 rats after 10 weeks of exposure to 0.1, 0.5, 1, 2.5 or 5% sodium saccharin [50, 250, 500, 1250 or 2500 mg/kg bw/day] in the diet. A dose-related increase in [³H]thymidine labelling index, hyperplasia and the presence of uniform and pleomorphic microvilli were found. At doses ≤ 1% sodium saccharin, there was no statistically significant increase in the incidence of hyperplasia or in the numbers of foci or cells exhibiting uniform and pleomorphic microvilli. At 0.1% sodium saccharin, there was no statistically significant change in the labelling index (Murasaki & Cohen, 1981). [The Working Group noted the small number of rats in each group and that subsequent studies showed labelling indices in occasional individual, untreated control rats that were similar to those seen in this study in rats treated with 1% sodium saccharin.]

Feeding of 7.5% sodium saccharin [3750 mg/kg bw per day] in the diets of male Sprague-Dawley rats from three weeks of age did not increase ³H-thymidine incorporation in the DNA of urinary bladder epithelium at 1, 2, 3, 6, 10, 15, 20, 30 or 50 weeks after the beginning of exposure. Autoradiographic analysis of the thymidine labelling index at 1, 15 and 50 weeks also did not show statistically significant increases in sodium saccharin-exposed animals when compared with controls (Lawson & Hertzog, 1981).

The effects of 5% sodium saccharin [2500 mg/kg bw/day] in the diet for 20 weeks from six weeks of age were compared in male Fischer 344 rats, B6C3F1 mice, Syrian golden hamsters and Hartley guinea-pigs. Rats developed hyperplasia of the urinary bladder and significantly increased DNA synthesis at 20 weeks, as determined by [³H]thymidine autoradiography, but the other species did not have similar changes. These results indicated to the investigators that the urothelial-cell proliferative response to high doses of sodium saccharin was specific to the rat. In a comparative study with males of four strains of rats (ACI, Wistar, Fischer 344 and Sprague-Dawley), the ACI strain was reported to be the most susceptible to the effects on the urinary bladder of 5% sodium saccharin in the diet for 52 weeks (Fukushima et al., 1983a).

In male Fischer 344 rats, the degree of hyperplasia and microvillus formation induced by freeze-ulceration in the bladder epithelium was similar during the subsequent two-week period whether or not 5% sodium saccharin was administered immediately after the procedure. The group receiving sodium saccharin only did not show hyperplastic changes in this time interval. Two weeks after freeze-ulceration, continuation or initiation of sodium saccharin exposure prolonged the duration of hyperplasia until termination of the experiment at eight weeks. The labelling index in the bladder epithelium of sodium saccharin-exposed groups with or without prior freeze-ulceration was increased above the control value in the last six weeks of the experiment. In another experiment, sodium saccharin given two or eight weeks after freeze-ulceration produced a similar increase in hyperplasia, microvillus formation and labelling index (Murasaki & Cohen, 1983b).

The effect of cation-associated saccharin was studied in five-week-old male Fischer 344 rats given 5% of the diet [2500 mg/kg bw per day] as sodium saccharin, acid saccharin, potassium saccharin or calcium saccharin for 10 weeks. Cell proliferation was measured by [³H]thymidine incorporation, hyperplasia was determined by light microscopy and formation of microvilli by scanning electron microscopy. The labelling index was increased by approximately ninefold when compared with controls by sodium saccharin, threefold by potassium saccharin and twofold by calcium saccharin; there was no increase in labelling index with acid saccharin. Only the changes with the sodium and potassium salts reached statistical significance. Sodium saccharin also caused a statistically significant increase in the number of animals with simple hyperplasia when compared with controls. Although there was evidence of simple hyperplasia after treatment with potassium saccharin and calcium saccharin, these changes were not statistically significant. Pleomorphic microvilli were observed only in rats given sodium saccharin. Sodium saccharin in the diet increased the urinary excretion of sodium. The urinary concentration of saccharin was similar after administration of each of the four forms of saccharin (Hasegawa & Cohen, 1986).

When seven-week-old male Fischer rats were exposed to 5% sodium saccharin [2500 mg/kg bw per day] in the diet for 21 weeks, [³H]thymidine labelling index and the ornithine decarboxylase activity in the bladder were about fivefold greater than in controls (Tatematsu et al., 1986).

The proliferative effects of sodium saccharin were found to vary with diet in five-week-old male Fischer 344 and four-week-old Sprague-Dawley rats exposed to 5 or 7.5% [2500 and 3750 mg/kg bw per day] sodium saccharin. The diets containing sodium saccharin were continued for up to 10 weeks, at which time the animals were killed. The Fischer 344 rats showed statistically significant increases in the [³H]thymidine labelling index with Prolab and NIH-07 diets at four weeks and with the Prolab diet at 10 weeks with 7.5% sodium saccharin. At the other times and doses, increased labelling indices were found, but they were not statistically significant, and no increase was seen in rats fed AIN-76A diet. The hyperplastic and epithelial changes in the groups on the three diets, monitored by scanning electron microscopy, were not remarkably different, except for an increase in the severity of changes with the Prolab diet and 7.5% sodium saccharin fed for 10 weeks. The animals fed AIN-76A diet had lower urinary volumes and higher concentrations of sodium saccharin and calcium, but lower levels of potassium than those on the Prolab and NIH-07 diets. The pH of the urine was also lower in this group (Garland et al., 1989).

Weanling male Fischer 344 rats were fed 5% sodium saccharin [2500 mg/kg bw per day] in either a Wayne or an AIN-76A diet for 2, 4, 6, 10 and 16 weeks. Whereas the AIN-76A diet alone or with sodium saccharin resulted in a urinary pH of 5.5–6.5, the Wayne diet with sodium saccharin gave a urinary pH of 7.4. With both diets, sodium saccharin increased the [³H]thymidine labelling index by about fivefold when measured at 2, 4, 6, 10 or 16 weeks. Addition of 2% sodium bicarbonate increased the labelling index for the group on the AIN-76A diet by six- to ninefold; sodium saccharin and sodium bicarbonate in the diet appeared to be additive (Debiec-Rychter & Wang, 1990).

In a dose–response experiment, sodium saccharin was administered in Prolab 3200 diet at concentrations of 0, 3, 5 or 7.5% [0, 1500, 2500 and 3750 mg/kg bw per day] to four-week-old male Fischer 344 rats, which were killed after 4, 7 and 10 weeks of exposure. [³H]Thymidine was injected 1 h before killing, and hyperplasia was measured by light microscopy and scanning electron microscopy. No evidence of increased labelling index or hyperplasia was found with 3% sodium saccharin, but cell necrosis and exfoliation were evident after 10 weeks. A doubling of the labelling index was noted with exposure to 5% sodium saccharin at 10 weeks, with extensive cell damage. With 7.5% sodium saccharin, the labelling index was already increased by threefold after four weeks, with evidence of increasing hyperplasia over the 4- to 10-week period. In all three groups given sodium saccharin, there was an apparent progression of necrotic changes during treatment (Cohen et al., 1990).

In two-generation studies, weanling male and female Sprague-Dawley and Fischer 344 F₀ rats were exposed to either 5% sodium saccharin or acid saccharin [2500 mg/kg bw per day] in Prolab 3200 diet for two weeks and then mated. Offspring were killed 1 h after injection of [³H]thymidine in utero, during lactation or after weaning. In a study in which mostly male animals were evaluated, there was no effect of either sodium saccharin or acid saccharin through seven days of age; however, sodium saccharin increased the labelling indices above those of controls at 21, 63 and 91 days of age, while acid saccharin did not. In a study of both male and female Sprague-Dawley rats, increased bladder urothelial cell proliferation was found in animals of each sex at 21 but not at 7 days of age or on days 17 and 21 of gestation. There was also evidence of small foci of superficial necrosis and mild hyperplastic proliferation by 21 days of age in animals of each sex (Cohen et al., 1995a).

The effects of exposure to concentrations of up to 6.84% sodium ascorbate [3400 mg/kg bw per day] as a single agent or 5 or 7.5% sodium saccharin [2500 and 3750 mg/kg bw per day] alone or combined with 1.23 or 1.85% ammonium chloride [620 and 920 mg/kg bw per day] in the diet on the urinary bladder of male Fischer 344 rats were measured on the basis of various histopathological and chemical parameters. Rats were placed on the experimental diets and mated, and the offspring were kept on the same diets as the F₀ generation. Female offspring were killed at 21 days of age and males at 16 weeks of age. One hour before death, the male F₁ rats were injected intraperitoneally with bromodeoxyuridine. Sodium saccharin but not sodium ascorbate increased the water consumption significantly. Groups of the F₁ generation at the high dose of sodium saccharin had to be terminated early because of toxic effects, and most of the other experimental groups showed some growth retardation. The urinary pH of rats at 5% sodium saccharin was decreased to less than 6.5 by co-administration of 1.23% ammonium chloride; it remained decreased but above 6.5 on day 37 of exposure to 5% sodium saccharin alone and was similar to that of controls by day 100. All of the groups given sodium ascorbate had urinary pH values significantly greater than those of controls. An amorphous precipitate containing calcium phosphate was present in the urine of male F₁ rats given sodium saccharin or higher doses of sodium ascorbate, but not in the controls, after the lowest dose of ascorbate or in rats given sodium saccharin plus ammonium chloride. The labelling index was increased and there was evidence of hyperplastic changes in the groups receiving 5% sodium saccharin alone or the highest dose of sodium ascorbate; the second highest dose of ascorbate induced evidence of hyperplasia, but the increase in labelling index was not statistically significant. No hyperplasia was observed in the group fed sodium saccharin plus ammonium chloride. In a small, one-generation study, hyperplasia was observed in male rats that had been kept on a diet containing 6.84% sodium ascorbate for 10 weeks but was completely inhibited in a group fed 2.78 or 3.7% ammonium chloride plus 6.84% sodium ascorbate. The caeca of rats given sodium saccharin with or without ammonium chloride were enlarged, but this effect was not seen with sodium ascorbate (Cohen et al., 1995b).

Weanling male Fischer 344 male rats were exposed to 5% sodium saccharin [2500 mg/kg bw per day] in Prolab 3200 diet for up to 72 weeks, when their bladder epithelial cells were stained immunohistochemically for uroplakin expression and scored for hyperplastic changes. Simple hyperplastic lesions were induced by sodium saccharin and ascorbate, and the lesions showed identical uroplakin staining patterns (Ogawa et al., 1996).

Twenty male and female monkeys (six African green, seven rhesus, six cynomolgus and one rhesus/cynomolgus hybrid) were exposed to 25 mg/kg bw sodium saccharin in the diet on five days per week from 24 h after birth up to 24 years of age. Eight monkeys died between 103 and 282 months of age, and the remainder were killed at between 207 and 283 months of age. At autopsy, none of the monkeys had any abnormalities of the urothelium, including the renal pelvis, ureter, urinary bladder and urethra. Of the monkeys that died, many had multiple abnormalities: two had septic hepatitis, five had lung infections and other abnormalities, one had chronic ulcers of the stomach and oesophagus, two had myocardial degeneration or fibrosis and one had chronic ileitis. The monkeys that were killed had abnormalities similar to those seen in 16 monkeys that served as controls, of which two died. There was no evidence of urothelial hyperplasia in either sodium saccharin-exposed animals or controls. No obvious differences in the chemical composition of the urine were seen in two cynomolgus and two rhesus monkeys of each sex given saccharin and in that of controls less than two years before they were killed. Examination of urine by scanning electron microscopy showed no increase in the frequency of microcrystalluria and no evidence of abnormal microcrystals, precipitate or calculi (Takayama et al., 1998).

(b) Physiological and biochemical changes in urine

Sodium saccharin fed to weanling male Charles River rats at concentrations up to 7.5% [3750 mg/kg bw per day] for four weeks was found to increase the urine volume and decrease the urinary pH. It also increased faecal water content, faecal excretion of sodium and potassium and urinary excretion of calcium, magnesium and phosphorous (Anderson, 1979).

Administration to Wistar rats for two years of 2 g/kg bw per day sodium saccharin in the drinking-water or of 4 g/kg bw per day in the diet reduced the weight gain markedly; fluid intake was increased in the latter group and decreased in the former. The average urinary pH of males in the first group had risen to above 7.0 by 27 weeks, and some animals showed marked crystalluria. These pH changes could be reversed by the addition of ammonium chloride to the diet. The most important exposure-related findings were increased incidences of microcalculi and telangiectasia of the vasa recta in kidneys, of renal pelvic hyperplasia, of extramedullary haematopoiesis and of hepatic zonal necrosis. Hyperplasia of the bladder epithelium occurred earlier in animals of the second group (Chowaniec & Hicks, 1979).

Five-week-old male Fischer 344 rats were fed diets containing 5% sodium saccharin [2500 mg/kg bw/day] until week 104, beginning at week 0 or 4; further groups received FANFT or l-tryptophan in the diet. The urine was analysed at weeks 1, 2, 4, 5, 6, 7, 8 and 12 and monthly thereafter. Exposure to sodium saccharin increased the water intake, which produced diarrhoea and increased the urinary volume. The concentration of sodium in the urine was not changed, and the only abnormality observed was slightly increased urinary pH during the first three months. There was no increase in crystal formation (Demers et al., 1981).

Three groups of four-week-old male Crl:COBS^RCD^R(SD)BR rats were exposed for 16 weeks to 5% sodium saccharin [2500 mg/kg bw per day] in the diet, 4% sodium saccharin [2000 mg/kg bw per day] in the drinking-water or control diet. Rats receiving saccharin in the diet had increased food and water consumption, urine elimination, precipitate and crystalline urinary sediment and some evidence of mild urothelial hyperplasia. Rats receiving saccharin in drinking-water showed increased urine osmolality (West & Jackson, 1981).

In a two-generation study, seven-month-old Charles River CD rats (F₀) and F₁ rats were exposed to 1, 3, 5 or 7.5% sodium saccharin in the diet [500, 1500, 2500 or 3750 mg/kg bw per day]. Controls were either unexposed or given 5% sodium hippurate, the latter to determine effects attributable to sodium. Sodium saccharin produced several physiological changes in the urine, including increased volume and sodium concentration, decreased osmolality and decreased potassium and zinc concentrations. Increases in bladder weight, hydration and the mineral content of the bladder tissue were also seen. Sodium hippurate caused similar changes, but they were less severe than those produced by sodium saccharin. Similar effects were found in an additional group of animals exposed to sodium saccharin from birth. In a group of animals exposed in utero, but not after birth, the measured parameters were similar to those in controls. Some differences in response to sodium saccharin were noted between males and females; for example, caecal weight was increased more in the females, but the mineral and sodium saccharin concentrations in the urine were higher in the males. In addition, at dietary concentrations of sodium saccharin ≥ 5%, the concentrations of sodium, potassium, magnesium and zinc were significantly increased in the bladders of males but not females. The pattern of and dietary concentrations at which these changes occurred suggested to the authors that changes in these mineral constituents in the urine and/or urine volume may play an important role in the etiology of bladder tumours in rats fed high dietary concentrations of sodium saccharin (Schoenig & Anderson, 1985).

Adult male Charles Rivers CD rats were fed either control or a diet containing 7.5% [3750 mg/kg bw/day] sodium saccharin for one month. Increased daily water intake, total urine volume, frequency of urination and average volume per urination were found (Renwick & Sims, 1983).

Histological changes in the urinary bladder related to cell proliferation and urinary chemistry were evaluated in male Fischer 344 rats in relation to dietary exposure to 5% sodium saccharin, acid saccharin, ascorbic acid or sodium ascorbate [2500 mg/kg bw per day]. Urinary pH, sodium, urinary bladder hyperplasia and scanning electron microscopic changes were measured at 8, 16 and 24 weeks. The histological changes seen by light microscopy were ‘slight’ or ‘very slight’ at the earlier times, and none were noted at 24 weeks. Scanning electron microscopy showed slight pleomorphic microvilli at 24 weeks in one of five animals given sodium saccharin and at eight weeks in one of five given sodium ascorbate. Short, uniform microvilli and ropy or leafy microridges were moderately prevalent with sodium saccharin or sodium ascorbate, whereas they were absent in the groups given saccharin or ascorbic acid. The urinary pH was increased by sodium saccharin and sodium ascorbate, was decreased by saccharin and was the same as that in controls with ascorbic acid. Urinary sodium excretion was increased by the sodium salts but not by the acidic forms (Fukushima et al., 1986a).

Prolab 3200 or AIN-76A diets containing 5% sodium saccharin or calcium saccharin [2500 mg/kg bw per day] were administered to five-week-old male Fischer 344 rats for 10 weeks. Although the Prolab 3200 diet has a lower pH than the AIN-76 diet, the former contains larger amounts of sodium, calcium, potassium and most other ions than the latter. Exposure to either salt of saccharin resulted in a urinary pH above 6.5 with the Prolab diet and below 6.0 with the AIN-76A diet. The amount of sodium excreted in urine was lowest with the AIN-76A diet with or without calcium saccharin. With sodium saccharin administration, the sodium concentrations in urine were higher with the Prolab diet than with AIN-76A (Fisher et al., 1989).

Changes in gastrointestinal and urinary tract physiology in young rats were investigated by a protocol simulating the two-generation bioassay of Schoenig et al. (1985; p. 548). Sodium saccharin was administered at 1, 3 or 7.5% [500, 1500 and 3750 mg/kg bw/day] in the diet, beginning with the F₀ generation; males and females in the F₁ generation were killed at approximately 30 days. Rats at 7.5% had decreased urinary pH, potassium and calcium and increased urinary volume, sodium, magnesium, phosphate and ammonia; animals at other doses were not examined for these parameters. In addition, rats fed 7.5% sodium saccharin had anaemia, a 50% increase in serum cholesterol concentration, a 10-fold increase in serum triglyceride concentration and decreased serum and hepatic concentrations of vitamins. These changes were not found at lower doses of sodium saccharin. In another experiment, newborn animals were dosed with sodium saccharin for 90 days, while others were given only control diet after 30 days on sodium saccharin. Some of the observed changes were found to be reversible; however, histopathological examination of the bladder showed no changes at 30 days and only slight changes at 90 days. The authors considered that the observed biochemical and physiological effects were not involved in bladder tumorigenesis. Many of the effects were found to be associated with anaemia due to iron deficiency, and some of them—but not the urinary or bladder effects—could be prevented by dietary iron and/or folate supplementation. This supplementation by itself increased the frequency of bladder hyperplasia (Garland et al., 1991a,b, 1993).

In a study involving feeding of various sodium salts at doses equimolar to 7.5% sodium saccharin to male Fischer 344 rats for 10 weeks, urothelial-cell proliferation was studied by light microscopy (hyperplasia), by bromodeoxyuridine labelling and by scanning electron microscopy. The sodium salts of saccharin, ascorbate, glutamate, aspartate, citrate, erythorbate, bicarbonate and chloride caused various degrees of increase in urothelial proliferation and all resulted in formation of the calcium phosphate-containing urinary precipitate (amounts not quantified). A group of rats that received 1.85% ammonium chloride plus 7.5% sodium saccharin did not show increased proliferation or formation of the precipitate (Cohen et al., 1995c).

(c) Urinary precipitate formation and the role of urinary proteins

In a two-generation lifetime feeding study, groups of 50 male and 50 female Sprague-Dawley rats were exposed to 5% [2500 mg/kg bw per day] sodium saccharin or to various doses of ortho-toluenesulfonamide in the diet, beginning when the F₀ generation was 32 days old. Rats were mated on day 90 of the study, and the F₀ generation and F₁ progeny were exposed to the same diets throughout life. The urine of older animals exposed to 5% sodium saccharin, especially the males, contained a flocculent-type precipitate, which dissolved in acetic acid. In 21-day-old pups and in 105-day-old male and female rats receiving ortho-toluenesulfonamide in another experiment reported by these investigators, a significant dose–response relationship was found for the incidence of bladder calculi, while sodium saccharin free of ortho-toluenesulfonamide did not have a similar effect (Arnold et al., 1979, 1980).

Male Fischer 344 rats were fed 7.5% [3750 mg/kg bw/day] sodium saccharin in Prolab 3200 diet beginning at five weeks of age, and urine was collected on Millipore filters two and four weeks later. Whereas control rats had typical phosphate crystals, rats fed sodium saccharin had fewer crystals and approximately one-half contained silicate and were jagged in shape. These silicate crystals were found to result in occasional microabrasion of the urothelial surface. Gel filtration of urine with added radiolabelled sodium saccharin showed two fractions of saccharin-binding proteins: one corresponded to the size of α_2u-globulin, while the other corresponded to albumin. The authors postulated that saccharin–protein complexes form a precipitate with silicate in urine that has a pH greater than 6.5 (Cohen et al., 1991b).

Support for the role of α_2u-globulin in sodium saccharin-induced cell hyperplasia and cell proliferation rates comes from studies with male Fischer 344 and NCI-Black-Reiter (NBR) rats. The latter strain does not produce the male-specific, low-molecular-mass urinary protein α_2u-globulin and does not undergo sodium saccharin-enhanced cell proliferation. Male NBR, Fischer 344 and castrated Fischer 344 rats were fed 7.5% sodium saccharin [3750 mg/kg bw per day] in Prolab 3200 diet for 10 weeks. Examination of the bladders by light microscopy revealed that 7/10 sodium saccharin-exposed Fischer 344 rats and 1/10 NBR rats had hyperplastic changes, but none were seen in the unexposed Fischer 344 rats; however, 4/10 castrated Fischer 344 rats also had hyperplastic bladders, although the α_2u-globulin urinary content was only 10% of that in normal Fischer 344 males. Examination of the bladder by scanning electron microscopy showed that the most severe changes occurred in castrated and normal sodium saccharin-exposed Fischer 344 rats, whereas the exposed NBR rats showed less severe changes. The increases in caecal weights showed a pattern similar to the changes observed by scanning electron miscroscopy. The urinary volume of NBR controls was threefold higher than that of Fischer 344 controls, and exposure to sodium saccharin increased the volume by three-to fourfold in NBR, Fischer 344 and castrated Fischer 344 rats (Garland et al., 1994).

Six-week-old Fischer 344 and NBR rats were exposed to 5% sodium saccharin [2500 mg/kg bw per day], 5% sodium ascorbate or 3% uracil [1500 mg/kg bw per day] in the diet for eight weeks. Ascorbate produced simple hyperplasia in Fischer 344 rats but not in NBR rats, and uracil produced papillary hyperplasia in both strains. Scanning electron microscopy showed that uracil produced the most severe changes, and sodium saccharin and ascorbate produced some changes only in Fischer 344 rats. Large differences in the bromodeoxyuridine labelling index were found in Fischer 344 rats given sodium saccharin (20-fold) or ascorbate (36-fold) and in both strains exposed to uracil (over 50-fold). In Fischer 344 rats, the urinary pH and sodium concentration were increased by sodium saccharin and ascorbate, and the urine volume was decreased only by ascorbate. In ascorbate-treated NBR rats, a significant increase in urinary volume was observed (Uwagawa et al., 1994).

Studies on the appearance of the crystalline and amorphous material in sodium saccharin-exposed male rats are described above (Cohen et al., 1995b,c).

(d) Studies primarily on gastrointestinal changes

Dietary sodium saccharin was tested for its ability to alter caecal microflora in weanling male CD rats exposed to 7.5% sodium saccharin [3750 mg/kg bw/day] in the diet for 10 days. The weights of caecal tissue and its contents were higher than those of controls, which received 7.5% cellulose (Anderson & Kirkland, 1980).

Adult male Charles River (CD)-derived Sprague-Dawley rats (weighing about 300 g) were exposed to 0–10% sodium saccharin [0–5000 mg/kg bw per day], 2% tryptophan [1000 mg/kg bw per day] or 2% tryptophan plus 5% sodium saccharin [2500 mg/kg bw per day] in the diet for one to two months. The presence of sodium saccharin in the diet had a rapid, pronounced effect on the metabolism of dietary tryptophan. The excretion of indican, a metabolite of tryptophan, was increased 3.1-fold per 24 h in animals fed 10% sodium saccharin in the diet, and there was a linear relationship between the dietary concentration and indican excretion. High plasma concentrations of saccharin (200–300 µg/mL) reduced the renal clearance of both saccharin and indican; the authors suggested that this was due to saturation of renal tubular secretion. Indican concentrations in the urine, and the indole concentrations in the caecum were increased, possibly due to accumulation of protein and tryptophan in the caecum. Sodium saccharin produced a dose-related increase in both the weight of the caecal contents and the wall of the caecum. The effects of sodium saccharin and tryptophan were additive. These results are consistent with the concept that sodium saccharin has a major effect on protein digestion in the intestine, which favours increased microbial metabolism of dietary tryptophan to indole, a urinary bladder co-carcinogen (Sims & Renwick, 1983).

Indican and para-cresol excretion were found to be increased by three-to fourfold in male Charles River adult rats fed 7.5% sodium saccharin [3750 mg/kg bw per day] in the diet for 40 days, when compared with controls. As described above, indican is formed from indole, a microbial metabolite of tryptophan, whereas para-cresol is formed from tyrosine by the gut flora. The excretion of phenol, another microbial metabolite of tyrosine, was, however, abolished by the same regimen of sodium saccharin, indicating that there is an altered pattern of metabolism in addition to an increase in the amount of metabolites (Lawrie et al., 1985).

Three-week-old male Sprague-Dawley rats were fed a diet containing 5% sodium saccharin [2500 mg/kg bw per day] for 4 or 20 weeks. Marked caecal enlargement and decreased β-glucuronidase, nitrate reductase and sulfatase activities were seen in the caecal content; however, there was no change in bacterial concentrations at either exposure interval. Incubation of the caecal contents from control rats with 75 mmol/L sodium saccharin in vitro produced similar enzyme inhibitions. The authors surmised that such inhibition may decrease the rate of formation of toxic bacterial products in the hindgut (Mallett et al., 1985).

In a two-generation study, 7.5% [3750 mg/kg bw per day] sodium saccharin was fed in the diet to adult male and female Charles River (CD)-derived Sprague-Dawley rats from six weeks before mating. The pups of sodium saccharin-fed dams were found to be exposed to elevated levels of indican via the milk. Sodium saccharin-exposed pups were found to have increased caecal size and caecal protein, decreased caecal tryptophanase activity and increased urine volume and urinary indican excretion. Pups from dams fed sodium saccharin only from the time of parturition showed more variable responses than those from dams fed sodium saccharin from before conception; this was due to variations in tryptophanase activity. These changes were similar in males and females and were greatest during the first month after weaning (Sims & Renwick, 1985).

CD rat dams were fed increasing doses of sodium saccharin: 1% [500 mg/kg bw per day] beginning at parturition, 3% [1500 mg/kg bw] at day 8 and 5% [2500 mg/kg bw] at day 15. Pups were weaned at day 21 and continued on this diet. The pups of dams that had not received sodium saccharin were started on 5% sodium saccharin at weaning. When pups were exposed to sodium saccharin during nursing, no effect was seen on caecal or bladder mass at weaning but the caecal mass after weaning was almost doubled. The caecal mass of pups exposed from the time of weaning was intermediate between that of pups exposed from birth and that of controls. The changes in bladder mass followed a similar pattern but were not as extensive; the increased bladder mass correlated with increased urinary output (Anderson et al., 1988a).

Although the studies described above suggest that gastrointestinal changes are implicated in the effects of sodium saccharin, the effects of saccharin on the caecal mass were not found to depend on the chemical form. There was no difference in the mass of the caecum plus contents or caecal tissue after 10 weeks of feeding male weanling CD rats 5% sodium saccharin [2500 mg/kg bw per day] or the molar equivalent of acid saccharin, potassium saccharin or calcium saccharin. Bladder hyperplasia was found only in rats given sodium or potassium saccharin, and increased relative bladder mass was found only with sodium saccharin. Sodium and potassium saccharin were also responsible for an increased relative urinary mass. The authors concluded that the caecal effects were due to the saccharin anion but that the urinary changes were dependent on the cation. Consequently, caecal mass changes do not appear to be involved in sodium saccharin-induced bladder tumours (Anderson et al., 1988b).

The excretion of indican was found not to be responsible for the toxic effects on the bladder of dietary administration of 5% sodium saccharin. Male CD rats were exposed to 5% sodium saccharin [2500 mg/kg bw per day], 1.5% indole [750 mg/kg bw per day] or 5% sodium saccharin plus 1.5% indole added to diet for 10 weeks. Sodium saccharin and indole produced equivalent increases in bladder mass and epithelial hyperplasia; however, in rats exposed to sodium saccharin, the amount of indican excreted in the urine was less than one-tenth the amount produced after dietary exposure to indole. After exposure to indole and sodium saccharin together, the increase in bladder mass was additive, but there was no increase in hyperplasia over that observed with each compound alone (Anderson et al., 1989).

(e) Other mechanistic information

In an assay to detect tumour promoters in vitro, sodium saccharin inhibited metabolic cooperation between Chinese hamster V79 cells in culture. A small number of hprt⁻ (6-thioguanine resistant) cells were seeded in the presence of a large number of hprt⁺ cells with 0–5 mg/L sodium saccharin in the culture medium. Sodium saccharin concentrations > 2 mg/L resulted in a dose-related increase in the recovery of 6-thioguanine-resistant cells (Trosko et al., 1980).

Sodium saccharin inhibited binding of ¹²⁵I-labelled mouse epidermal growth factor to 18 cell lines, including HeLa (human carcinoma), MDCK (dog kidney), HTC (rat hepatoma), K22 (rat liver), HF (human foresekin), GM17 (human skin fibroblasts), XP (human xeroderma pigmentosum fibroblasts) and 3T3-L1 (mouse fibroblasts) (Lee, 1981).

Inhibition by sodium saccharin of urease and three proteases in vitro was not due to the sodium ion (Lok et al., 1982).

Saccharin [type or source not specified] inhibited binding of epidermal growth factor to cultured rat pituitary tumour cells and enhanced prolactin production (Brennessel & Keyes, 1985).

In two initiation-promotion studies, inhibitors of lipid peroxidation were found to inhibit the tumour-promoting effects of sodium saccharin. Both aspirin and the antioxidant nordihydroguaiaretic acid inhibited bladder tumour formation when co-administered with sodium saccharin. These studies indicate that oxidative damage may play a role in the cytotoxic effects of sodium saccharin (Sakata et al., 1986; Yu et al., 1992).

In cultured bladders from young female Fischer 344 rats, 12 mmol/L sodium saccharin produced urothelial hyperplasia and dysplasia, as detected by histology. Sodium saccharin increased the number of foci of cell proliferation induced by MNU in explant bladder epithelium cultures of young female Fischer 344 rats. Only transient hyperplasia was found with sodium saccharin alone (Knowles & Jani, 1986; Nicholson & Jani, 1988).

Rat bladder explants were exposed continuously to sodium saccharin for up to 28 days. Hyperplastic effects and increased DNA synthesis were reported after exposure to 0.5% (Norman et al., 1987), while explants treated with 0.1% resembled control cultures.

Saccharin [type not specified] was found to inhibit intercellular communication in a human urothelial cell-line (JTC-30), which was established from a well-differentiated transitional-cell carcinoma of the urinary bladder. Inhibition measured by dye transfer at 48 and 96 h occurred at a concentration of 3 mg/mL, which was relatively nontoxic. At lower concentrations, there was little or no effect, and higher concentrations were toxic (Morimoto, 1996).

The ionic structure of the saccharinate anion, as studied by two-dimensional nuclear magnetic resonance, was not altered by several ions at the wide ranges of concentrations found in urine after feeding of high doses of various salt forms of saccharin or its acid (Williamson et al., 1987).