Sodium tungstate dihydrate (ST) is present naturally in the environment and can enter waterways through the weathering of rocks and soils. ST also is a high-production volume compound that is used in a variety of commercial applications including fire- and waterproofing fabrics, in the preparation of complex compounds (e.g., phosphotungstate and silicotungstate), as a reagent for biological products, and as a precipitant for alkaloids. Tungsten was nominated to the National Toxicology Program (NTP) by the Centers for Disease Control and Prevention to evaluate its potential to cause chronic toxicity and carcinogenicity because of concern about potential human exposure via contaminated drinking water (e.g., in the form of salts like tungstate) and inadequate data to assess human health implications of elevated exposures. ST was selected for study because it is the most prevalent water-soluble form of tungsten. In these studies, Sprague Dawley (Hsd:Sprague Dawley SD) rat dams were exposed to ST in drinking water from gestation day (GD) 6 through lactation day (LD) 20. Their pups were exposed to the same exposure concentrations in drinking water from postnatal day (PND) 12 through 3 months or 2 years. Adult male and female B6C3F1/N mice were exposed to ST in drinking water for 3 months or 2 years. Genetic toxicology studies were conducted in Salmonella typhimurium, Escherichia coli, rat and mouse peripheral blood erythrocytes, and cells from liver, kidney, and ileum; peripheral blood leukocytes from rats and mice also were assessed for DNA damage.
Perinatal and three-month study in rats
Beginning on GD 6, groups of eight F0 time-mated female rats were exposed to ST in drinking water throughout gestation and lactation at one of five exposure concentrations (125, 250, 500, 1,000, or 2,000 mg/L) or were provided the vehicle control (deionized water). Groups of 10 F1 rats per sex continued on in the study after weaning and were given drinking water containing the same respective ST concentrations for 3 months. There were no significant effects of ST exposure on pregnancy status, maternal survival, or littering parameters. By the end of lactation, dams in the 1,000 and 2,000 mg/L groups showed significant decreases in group mean body weight of approximately 10% and 18%, respectively, and water consumption was significantly decreased for the 500, 1,000, and 2,000 mg/L groups relative to the vehicle control group over the LD 17 to LD 21 interval. When adjusted for litter size, the mean body weight of male and female pups in the 2,000 mg/L group on PND 21 was significantly decreased by approximately 16% and 11%, respectively, compared to the corresponding vehicle control groups.
There were no early deaths during the 3-month study. When compared to the vehicle control group, final mean body weights were lower for the 1,000 and 2,000 mg/L males and 2,000 mg/L females. Water consumption was lower for the 1,000 and 2,000 mg/L males and females. The urine xanthine/creatinine ratios were significantly increased in all male and female exposed groups. Serum insulin concentrations were significantly decreased in the 2,000 mg/L males relative to the vehicle control males. Significantly decreased absolute weights were observed in several organs but were considered secondary to body weights reductions. Exposure-related histological lesions were limited to the kidneys and included increased incidences of renal tubule regeneration in the 1,000 and 2,000 mg/L males and females; the increases in the 2,000 mg/L groups were significant relative to the vehicle control group.
Perinatal and two-year study in rats
Beginning on GD 6, F0 time-mated females were exposed to ST in drinking water throughout gestation and lactation at one of three exposure concentrations (250, 500, or 1,000 mg/L) or were provided the vehicle control (deionized water). Groups of 50 F1 rats/sex/group continued on in the study after weaning and were provided drinking water containing the same respective ST concentration as their dam for 2 years. An additional 40 F1 rats/sex/exposure group were used for interim evaluations and were provided dosed drinking water or the vehicle control for 3, 6, 12, or 18 months. There were no significant effects on reproductive performance, including the percentage of mated females producing pups. During gestation and lactation, the mean body weight of dams in the 1,000 mg/L group was lower than that of the vehicle control group. There were no exposure-related differences between the vehicle control group and the ST‑exposed groups in the number of litters, litter size, mean litter weights, sex ratio, or the pup mean weights of males and females.
Interim evaluations were performed on male and female rats at 3, 6, 12, or 18 months for organ weights and tungsten concentrations in plasma, kidney, and urine. Although there was no consistent pattern of changes in kidney weights across sex or over time, kidney tungsten concentrations increased with exposure concentration, and the kidney/plasma ratios were higher than 1 at all exposure concentrations and time points demonstrating retention of tungsten in the kidney. This finding was consistent with the nephrotoxicity observed in the 2-year study.
Survival to study termination was significantly increased in all groups of exposed male rats compared to the vehicle control males, with survival of the vehicle control males being lower than that typically seen in groups of control male Sprague Dawley rats in previous 2-year NTP studies. There were no significant differences in the survival of female groups. At study termination, mean body weights of all groups of exposed males were within 10% of the vehicle control group. In females, mean body weights of the 500 mg/L and 1,000 mg/L groups at study termination were approximately 11% and 21% less than those of the vehicle control group, respectively. Over the course of the 2-year study, mean water consumption for the 250, 500, and 1,000 mg/L groups averaged 93%, 99%, and 84% of the vehicle control males and averaged 95%, 100%, and 91% of the vehicle control females.
The incidences of thyroid gland C-cell adenomas were higher in all exposed groups of female rats, and the increase was significant in the 500 mg/L group relative to the vehicle control group. Although not significant, the incidence of C-cell carcinomas was higher in the 1,000 mg/L females. The incidences of C cell adenoma or carcinoma (combined) exceeded the historical control range in the 250 and 500 mg/L females.
In the kidney, the incidences of suppurative inflammation of the renal tubules were significantly increased in the 1,000 mg/L males and females, and the incidence of renal tubule regeneration was significantly increased in the 1,000 mg/L females, relative to the respective vehicle control groups.
In the uterus, there was a significant increase in the incidence of atypical hyperplasia, relative to the vehicle control group, in the 500 mg/L females.
Three-month study in mice
Groups of 10 male and 10 female mice were exposed to ST in drinking water for 3 months at one of five exposure concentrations (125, 250, 500, 1,000, or 2,000 mg/L) or were provided the vehicle control (deionized water). All mice survived to the end of the study. Over the course of the study, mean body weights were below 90% of the vehicle control group for the 250, 1,000, and 2,000 mg/L females and the 2,000 mg/L males. At study termination, the mean body weights of all exposed groups of males and females were within 10% of the vehicle control groups. Weekly mean water consumption values were <90% of the vehicle control groups in the 1,000 and 2,000 mg/L males and the 2,000 mg/L females. Lower absolute organ weights were attributed to body weight reductions.
The only histological lesion associated with exposure was in the kidney. The incidences of renal tubule regeneration were higher in the 1,000 and 2,000 mg/L male and female groups compared to the respective vehicle control groups. The increases in the male groups were significant.
Two-year study in mice
Groups of 50 male and 50 female mice were exposed to ST in drinking water for 2 years at one of three exposure concentrations (500, 1,000, or 2,000 mg/L) or were provided the vehicle control (deionized water). An additional 40 mice/sex/exposure group were included for interim evaluations at 3, 6, 12, and 18 months.
More males in the ST-exposed groups survived to study termination than did the vehicle control males; however, the differences were not significant. Survival in females was similar across all groups. At study termination, the mean body weight of the 2,000 mg/L males was 88% of the vehicle control group, and water consumption was approximately 78% of the vehicle control group; all other groups of exposed males and all groups of exposed females had mean body weights within 10% of their respective vehicle control groups. Clinical observations included more occurrences of thinness and ruffled fur in exposed males compared to vehicle control males.
Kidney tungsten concentrations increased with exposure concentration and the kidney/plasma ratios were higher than 1 at all exposure concentrations and time points, demonstrating retention of tungsten in the kidney. Renal tubule neoplasms were only recorded in exposed males; one renal tubule adenoma was observed in the 1,000 mg/L males, and two renal tubule carcinomas were observed in the 2,000 mg/L males. Compared to the respective vehicle control groups, there were significantly increased incidences of renal tubule regeneration in all exposed groups of males and in the 1,000 and 2,000 mg/L groups of females.
In the large intestine, the incidences of pigment in the cecum were significantly increased in the 1,000 and 2,000 mg/L males and females.
In the testes, there was a significantly increased incidence of germinal epithelium degeneration in the 500 mg/L group relative to the vehicle control group; the incidences were increased, but not significant, in the 1,000 and 2,000 mg/L groups.
Genetic toxicology
ST was not mutagenic in any of several bacterial tester strains, with or without exogenous metabolic activation (S9 mix). No increases in micronucleated erythrocytes were seen in male and female rats and mice administered ST in drinking water for 3 months. An exposure concentration-related significant increase in the percent of circulating immature erythrocytes was seen in male and female rats and in male mice, whereas there were no changes in this population of cells in female mice. Significantly increased DNA damage, as measured by the comet assay following administration of ST in drinking water for 3 months, was seen in liver cells of male and female rats and male mice; it also was seen in cells from the ileum of male mice. No increases in the levels of DNA damage were observed in blood leukocytes from either species or in kidney cells from mice.
Conclusions
Under the conditions of these 2-year drinking water studies, there was no evidence of carcinogenic activity of sodium tungstate dihydrate (ST) in male Hsd:Sprague Dawley SD rats at exposure concentrations of 250, 500, or 1,000 mg/L. There was equivocal evidence of carcinogenic activity of ST in female Hsd:Sprague Dawley SD rats based on increased incidences of C-cell adenoma or carcinoma (combined) of the thyroid gland.
There was equivocal evidence of carcinogenic activity of ST in male B6C3F1/N mice based on the occurrences of renal tubule adenoma or carcinoma (combined) in exposed animals. There was no evidence of carcinogenic activity of ST in female B6C3F1/N mice at exposure concentrations of 500, 1,000, or 2,000 mg/L.
Exposure to ST in drinking water caused increased incidences of nonneoplastic lesions in the kidney of male and female rats and mice, in the uterus of female rats, in the large intestine of male and female mice, and in the testes of male mice.
National Toxicology Program (NTP). 2021. NTP technical report on the toxicology and carcinogenesis studies of sodium tungstate dihydrate (CASRN 10213-10-2) in Sprague Dawley (Hsd:Sprague Dawley SD) rats and B6C3F1/N mice (drinking water studies). Research Triangle Park, NC: National Toxicology Program. Technical Report 599. https://doi.org/10.22427/NTP-TR-599