Registration Dossier

Administrative data

Description of key information

Oral: Several subacute oral toxicity studies were conducted to assess kidney toxicity in rat (BASF 1997/1998/2003, Merski 1982, Anderson and Kanerva 1978/79). The reversibility of NTA-associated nephrotoxicity was investigated in a subchronic feeding study in SD rats (Myers, 1982). Results indicate a complete recovery of vacuolation and hyperplasia. In some rats hydronephrotic response was clearly sufficiently severe to preclude the restoration of normal renal structure. In subchronic and chronic oral toxicity studies in rat and dog, haematology and histology of a broad spectrum of organs/tissues were performed (Nixon 1971/1972, Budny 1973). In addition there are information from a 2 y repeated dose/carcinogenicity study (NCI, 1977).
Inhalation: A subacute inhalation study assessed the effect of Na3NTA on 6 rats, 6 guinea pigs, and 2 monkeys/sex/dose when exposed to graded aerosol for 4 weeks (6 h/d, 5d/w) (P&G, 1971).
Dermal: A combined subacute/subchronic dermal toxicity study assessed the effect of 50 mg Na3NTA/kg/d applied to the abraded (28 d) and intact (91 d) skin of 6 New Zealand White rabbits.

Key value for chemical safety assessment

Repeated dose toxicity: via oral route - systemic effects

Endpoint conclusion
Dose descriptor:
92 mg/kg bw/day
Study duration:

Additional information

Studies with administration of Na3NTA and H3NTA were considered as relevant information because both compounds dissociate into the nitrilotriacetate anions and the respective cations under physiological conditions.


A 3-week gavage study (GLP) similar to OECD 407 was conducted in 5 Wistar rats/sex/dose at dose levels of 0, 150, 500, and 1500 mg/kg/d (BASF 1997). Study duration was only 3 weeks, blood chemistry was limited to urea and creatinine, haematology parameters were not examined, and histomorphologic examination was conducted on the kidney only. One animal/sex/dose was perfused with formaldehyde solution and the kidneys were examined by electron microscopy to characterise tubular degeneration. Significantly elevated relative kidney weights were found in high dose males and high and mid dose females. Light microscopy of the kidneys revealed tubular vacuolation, often multifocal and bilateral, in most males and females of the mid and high dose groups. Males were more affected than females, and mean severity of this lesion increased with dosage. Vacuolated cells appeared to have fewer cell organelles such as mitochondria and lysosomes, the content of strands of the rough endoplasmatic reticulum was markedly reduced. The observed vacuolar degeneration was regarded a consequence of acute cellular swelling which finally results in cell death. Diffuse and focal degeneration of the pars recta of the proximal tubule was present in high dose males and females and one mid dose male. The incidence of low graded basophilic (regenerating) tubules was similar between treatment and control groups, whereas the incidence of this lesion with higher severity grades increased with dosage. Mononuclear cell infiltration, interstitial fibrosis, proteinaceous casts, focal glomerulopathy and calcification at the cortico-medullary junction were similarly distributed between the treatment and control groups. Ultrastructural examinations did not show alterations of the pars recta of the proximal tubule in animals of the low and mid dose groups. No adverse effect was observed at the dose of 150 mg/kg H3NTA (NOAEL).

The mechanism underlying kidney toxicity and tumorgenic action of NTA was assessed in a 30 day gavage study in male SD rats (Merski 1982). Two animals of each group receiving 0, 0.73 or 7.3 mmol Na3NTA/kg/d (appr. 0, 187.6 or 1876 mg/kg/d) were killed 24 h after dosing on day 9, 13, 16, 20, 23, 27, or 30. Cytoplasmic vacuolation of cells of the proximal convoluted tubules (PCT) occurred within a few hours after administration of 0.73 mmol/kg and remains the only alteration to develop in kidney of rats at both dose levels for up to day 9. This initial change is followed in the course of the study by development of an increased incidence of various other lesions most notably by hyperplasia of the PCT epithelium. After 13 days high dose animals develop focal haemorrhage, necrosis, erosion, and hyperplasia of the epithelium of the renal pelvis. These lesions were not observed in the low dose group. The LOAEL for adverse effects on the kidney was 0.73 mmol/kg bw/d (appr. 187 mg/kg bw/d). Data demonstrate a close association between vacuolation and hyperplasia and strongly suggest that the structural alterations that develop in the renal tubules occur in a sequential pattern beginning with cytoplasmic vacuolation.

Kidney toxicity of Na3NTA.H2O after repeated oral administration was assessed in a 28 day feeding study (GLP) similar to OECD 407 (BASF, 1998). Results were compared to that of FeNTA i.p. injections to determine whether effects are mediated by an internal cation exchange (formation of FeNTA). Na3NTA.H2O (92.4 %) was administered to male Wistar rats. Subgroups of 5 animals received 150 ppm (2 groups), 20000 ppm (4 groups) or served as controls (4 groups). Animals were examined for clinical observations, food consumption, body weight (change), serum transferrin, iron, total iron-binding capacity, urinalysis, histopathology of several organs, and ultrastructural pathology of the kidneys, determinations of 8-HO-deoxyguanosine and of lipid peroxidation in kidneys, as well as determination of DNA-synthesis in the kidneys. At 20000 ppm food consumption was decreased up to 48 % during the entire study. The body weights were significantly lower (-26 %) than in the control groups. Clinical chemistry examinations did not reveal changes of the levels of iron concentrations, transferrin levels and total iron binding capacity. Increased concentrations of zinc were excreted with the urine at the end of the study, but calcium and iron excretion remained unaffected. Standard urinalysis revealed macrohaematuria on day 8, increased blood and in the sediment elevated number of erythrocytes. All animals of this dose group showed enlarged kidneys and dilation of the renal pelvis. Microscopic lesions were seen in the kidney of all high dose animals consisting of tubular hyperplasia. These hyperplasias were characterised by tubules with large, vacuolized cells. Other changes in most or all animals of this group were basophilia, vacuolation (without hyperplasia), dilation, and calcification of tubules, interstitial nephritis, inflammation, necrosis, fibrosis and urothelial hyperplasia of the renal papilla and pelvic dilation. The severity of hyperplastic lesions was associated with the occurrence of pelvic dilation and loss/necrosis of the renal papilla. Ultrastructurally, the histomorphological vacuolation of tubular epithelial cells were confirmed to consist of different stages of vesiculation and dilation of the rough endoplasmic endothelium, occasionally accompanied by cytoplasmic blebbing into the tubular lumen, and of ballooning degeneration of mitochondria. These changes were found in samples of the cortex and outer medulla, but not in the inner medullar regions. They are characteristic for different stages of cell swelling and vacuolar degeneration up to lysis of cells. No renal lesions except a single animal with tubular calcification and two animals with mononuclear cell infiltration were observed in the low dose group. No lesions related to the Na3NTA treatment were observed in the liver, pancreas, and spleen. The comparison with FeNTA showed that the extent, pattern and mechanism of inducing kidney toxicity differ. The result indicates that Na3NTA related effects are not mediated by an internal formation of FeNTA. Due to the missing haematology examinations, the incomplete clinical biochemistry, and the incomplete list of organs examined for histopathology, this study did not fulfil the requirements of current standard test protocols of the OECD 407/EEC B.7 method. The NOAEL of this study was 150 ppm Na3NTA.H2O (9 mg/kg bw/d).

In a subchronic toxicity study (GLP) Na3NTA (92.4 %) was administered to 10 male Wistar rats at dietary concentrations of 0 and 150 ppm (appr. 9 mg/kg bw/d) for 4 weeks (BASF, 2003). Animals were examined for clinical effects, food consumption, body weight change, and pathology of the kidneys. In addition 8-HO-deoxyguanosine levels and lipid peroxidation in kidneys (being measured as the amount of TBA-reactive material) was determined as indicators of oxidative stress. Kidney weights and lipid peroxidation were not altered. The 8-HO-dG-level in the kidney was statistically significantly lower in treated rats than the control group. This is considered to be biologically relevant suggesting a protective effect of Na3NTA at low doses with regards to oxidative stress in the kidneys. The data confirmed the analysis of a previous study (BASF, 1998).

In two subacute feeding study trisodium nitrilotriacetate as Na3NTA and H3NTA was administered to Sprague Dawley and Fischer-344 rats for 28 days at dietary levels of 0.15, 0.75 and 1.5 % for both Na3NTA and H3NTA and at 2 % for Na3NTA (Anderson and Kanerva, 1978 and 1979). Effects on growth, urine pH and urine volume were assessed. No mortality occurred and no clinical signs of toxicity were recorded. Urines from animals consuming 0.75 % NTA contain insoluble material. Haematuria was observed in the urines of rats of both strains at the highest dose level. X-ray analyses showed that the insoluble material contained crystalline CaNaNTA. All doses of Na3NTA reduced growth and increased urine pH above the controls. Body weight reduction and decreased urine pH-values were seen at 0.75 % and above. Kidney:body weight ratios were increased in Na3NTA treated animals at the highest dose level, but increase was only minimal in H3NTA groups. The urinary concentrations of Ca varied considerably but all high dose groups had elevated Ca levels. Renal excretions of NTA showed two-fold variations, values were in the 30-40 % range of the ingested dose for all groups. At necropsy, only one F344 female of the Na3NTA high dose groups developed hydronephrosis, whereas Sprague Dawley rats presented hydronephrosis in 40 % of females and 100 % of males of the Na3NTA groups and in 60 % of males receiving H3NTA. Overall, both strains gave similar quantitative responses to high dose of both forms of NTA, such as reduced growth, increased urinary Ca, increased kidney/body weight ratio, haematuria and the presence of crystaline CaNaNTA in the urine. Due to the limited parameters under investigation, no NOAEL was estimated for both Na3NTA and H3NTA. 0.5 % (350 mg/kg bw/d) NTA was considered as LOAEL.

The reversibility of NTA-assiciated nephrotoxicity was investigated in a subchronic feeding study in SD rats (Myers et al., 1982). Nitrilotriacetate was applied to 6-8 males at 2 % Na3NTA (1309 mg/kg/d), and 1.5 % H3NTA (1050 mg/kg/d), respectively. One group was sacrificed immediately after the 7 week administration period while a second group was allowed to recover for 5 weeks. Throughout the study, weekly weight gains and feed consumption were measured. At necropsy longitudinal midsagittal sections from each kidney were assessed for pathological alterations. In comparison to the control group, both forms of NTA resulted in comparable decreases in growth rate during the exposure phase of the study. Both forms induced tubular vacuolation and hyperplasia often composed of vacuolated cells in the epithelium of the proximal convoluted tubules. Vacuolation and nodular basophic cell hyperplasia was evident in each animal of the NTA groups, but was absent in the control groups and in the recovery groups indicating complete recovery. Pelvic epithelial and subepithelial inflammation and fibrosis were observed in the groups given either form of NTA for 49 days. Mild to severe hyperplasia of the transitional epithelium (TE) in the renal pelvis occurred in 4/7 animals given diets containing Na3NTA and moderate TE hyperplasia was observed in 1/8 rats given H3NTA in the diet. Animals with TE hyperplasia also exhibited pelvic dilation and TE erosion, ulceration and haemorrhage. TE dysplasia, intracytoplasmic globules, and mitotic figures were also noted. Tissues from recovery animals showed no abnormal cellular morphology in the pelvic urothelium in the presence of persisting hydronephrosis. In conclusion, NTA treatment was related to vacuolation, simple and nodular hyperplasia of the vacuolated cell type and the basophilic cell type in the proximal convoluted tubules, and transitional cell erosion, ulceration, hyperplasia and dysplasia associated with hydronephrosis in the renal pelvis. Nodular hyperplasias and urothelium lesions were shown to be reversible within this model. Hydronephrotic response in the Na3NTA-treated group was clearly sufficiently severe to preclude the restoration of normal renal structure in a few rats. Vacuolation and basophilia of tubular cells with or without increased cell size is known to be a response to tubular injury. Lesions observed at the pelvis area are also characteristic for degenerative and regenerative response of urothelial cells. The authors discussed the higher incidence of basophilic cell hyperplasia being nonspecific and related to spontaneous age related nephrosis. Although no data were reported for the acclimatisation period, it is assumed that rats were about 6-13 weeks during the course of the study. Therefore the role of rat typical spontaneous chronic progressive nephropathy of the old rat is debatable. With respect to the renal damage, no NOAEL was established, the LOAEL was 2 % Na3NTA respectively 1.5 % H3NTA. Due to the restricted set of examinations, this study does not fulfill the requirements of current standard test protocols of the OECD 408 guidelines.

In two sub-chronic toxicity study (pre-test and main study) Na3NTA was administered to 40 SD rats/sex/dose in diet at dose levels of 0, 7500, 10000 ppm (0, 750, 1000 mg/kg bw/day) and 10 SD rats/sex/dose at dose levels of 0, 2000, 20000 ppm (0, 200, 2000 mg/kg bw/day) (Nixon 1971). Haematologic examination and histologic examination of 15 organs/tissues were performed in both studies. Retarded body weight gain resulting in increased relative weights of the liver and kidney was observed at 20,000 ppm. Kidney lesions in the form of hydronephrosis and tubule cell damage, which was probably related to the increased hydrostatic pressure of the glomerular filtrate, were symptoms noted generally at the higher levels, with severity being in proportion to the level of Na3NTA in the diet. At 20,000 ppm hydronephrosis was observed in 9/10 male and 3/9 female rats. At 7500 ppm histologic examination showed mild hydropic degeneration of the tubular cells in 4/10 male rats, and tubular atrophy and dilatation in two other males. More extensive kidney changes of the same nature were seen in animals receiving 10,000 ppm. None of the other organs or blood levels examined in this study were adversely affected by the ingestion of Na3NTA. Changes of kidney and liver weights were thought to relate to toxicity of NTA rather than being a primary effect related to lower body weight gain, because no effect was seen on the weight of other organs, and kidneys and liver were affected at the mid dosage without any change of body weight gain in these animals. The LOAEL is 750 mg/kg/d, based on increased kidney/body weight ratio and mild hydropic degeneration of the tubule cells and tubular atrophy/dilation, both in males. The NOAEL is 200 mg/kg/d. This sub-chronic toxicity study in the rat is acceptable and satisfies the main requirements for a sub-chronic oral study in rats.

In a chronic toxicity study Na3NTA was administered in their diets to groups of 50 SD rats/sex/dose at dose levels of 0.03 %, 0.15 % and 0.5 % (0, 19, 97, 322 mg/kg/d) for a period of 24 months (Nixon, 1972). At 6, 12, 19 and 24 months, 5 animals of either sex were randomly selected for metabolic and histological studies of kidney lesions. Additional tissues taken for histologic examination after 19 and 24 months were spleen, liver, lung, heart, stomach, esophagus, small intestine, adrenal, trachea, urinary bladder, gonads, thyroid and bone (however, no data were reported on the absence of evidence of abnormalities in these organs). The femurs were analysed for Zn, P, Co, Mg, Ca and NTA content. There was a male dose-related decrease in survival rate documented in the study. Furthermore, the liver weight:bw ratio was reported to be significantly higher in females receiving 0.5 % Na3NTA.H20 at 12 months. Significant increases in urinary zinc level in animals dosed at 0.15 % and 0.5 % were also reported. The NTA deposition in the bone increased dose-related without any difference between sexes. No differences were noted microscopically in bones from test and control animal; the breaking strength of the bones from treated rats was not significantly altered from that of the control animals. In mid and high dose animals, kidney lesions started at 6 months in 1 or 2 animals/group consisting of hydropic degeneration of the tubule and minor tubule dilatation. At 12 months they were still considered to be mild. At 24 months, moderate to severe chronic interstitial nephritis and nephrosis were observed with a dose-related increase of incidence and severity in the mid and high dose groups. Chronic renal lesions observed in 32 % of the 0.03 % Na 3 NTA groups did not show significant differences from that of control animals (28 %). Tumors of various types were reported to be similar for all groups and were considered not to be related to the treatment groups. The NOAEL is 0.03 % based on the lack of apparent adverse effects on the health of SD rats over the course of the 2 year exposure period. This chronic study in the rat reveals significant reporting deficiencies, including a lack of reported data for both haematology and clinical chemistry parameters. It is pre-guideline and does not satisfy the guideline requirement for a chronic oral study OECD 452 in rats. Hence, it cannot be used as a standalone study to assess chronic toxicity of Na3NTA.

In a subchronic feeding study, 4 doses of Na3NTA (0, 0.03, 0.14, 0.5 %) were administered to groups of 4 beagle dogs/sex/dose for 90 days (Budny, 1973). No unscheduled death occured during the study, no substance-induced physical, ophthalmologic or behavioral changes were observed. Food consumption and an increase in body weight in the treated groups were similar to the control group. The necropsy didn't show any significant differences in organs nor in tissue between the tested animals and the control groups. The results of haematologic parameters and serum contents were comparable among all animals. In the mid and high dose group, there were significantly high zinc excretions in the animals urine (with a high interindividual variability and without dose-dependency in females). Because of increased urinary zinc excretion and unchanged fecal zinc excretion, a depletion of zinc was suggested by the authors. However, no signs of zinc deficiency (emaciation, emesis, conjunctivitis, keratitis, general debility and retardation of growth) were manifested in this study. Although urine zinc excretion was elevated, this was not accompanied by any renal lesions like in some rat studies. An examination in bones showed a dose-related increased NTA content. Histological examination of bone showed no differences in the amount of osteoid tissue or widths of the epiphyseal plates between samples from control animals and animals of the high dose group. This subacute toxicity study in the Beagle dog is acceptable and satisfies the main requirement for a subchronical oral study in non rodents. Based on altered zinc excretion at doses from 0.14 % (42.5 mg/kg/d), the NOAEL of this feeding study in beagle dogs was estimated to be 0.03 % Na 3 NTA (11 mg/kg bw/d).


The organ system that was mainly affected by NTA treatment was the urinary tract. A NOAEL of 92 mg/kg/d (cytotoxicity) was established in a 2 year carcinogenicity study (NCI, 1977) [IUCLID section 7.7]. This value is close to the lowest LOAEL of 97 mg/kg/d established in a less reliable 2 year feeding study (Nixon, 1972). Urinary tract toxicity was considered as a precondition in the development of preneoplastic changes. Hyperplasia was observed at lower doses than cytotoxicity and is therefore considered the most sensitive effect. Hence, it is proposed to use the NOAEL for carcinogenicity (9 mg/kg/d) rather than to apply the NOAEL for cytotoxicity as a solitary effect for risk assessment on repeated dose toxicity.


In a subacute inhalation study, 6 rats, 6 guinea pigs, and 2 monkeys/sex/dose were exposed to graded aerosol concentrations of Na3NTA for 4 weeks (6 h/d, 5 d/w) (P&G, 1971). Immediately after exposure one-half of the animals in each group were sacrificed and necropsies were performed. After a 14 day post-exposure period, the remaining animals were sacrificed. All animals were observed daily of pharmacotoxic signs and weighed weekly. Respiratory physiology, including measurements of total respiratory flow resistance, diffusion and ventilation, was performed on all monkeys. Total respiratory resistance and diffusion were measured in all guinea pigs. Blood was obtained from all monkeys and rats and haematological and biochemical determinations were performed. Urine samples were obtained from all monkeys and rats and analyzed for trace metals. Slight to severe respiratory distress was experienced by guinea pigs exposed to 0.34 mg/l Na3NTA. Monkeys and rats were not affected. Growth curves were not affected to any major extent in any of the animals. Haematological determination did not reveal consistent abnormal values in any of the experimental groups. Blood biochemistry revealed increases in SGOT, albumin, globulin, total protein, and gamma globulin in monkeys exposed to 0.34 mg/l at the 28 day interval. Analyses of trace metals in the urine of rats and monkeys did not reveal consistent abnormal values. Pulmonary function studies on guinea pigs and monkeys indicate various abnormalities which did not correlate to the results of histopathological examinations of the lungs. Observations recorded at necropsy of rats, guinea pigs and monkeys did not reveal consistent lesions which could be classified compound related. Under consideration of the incompleteness of the study comparing to standard test protocols, the NOAEC for local effects was 0.21 mg/l for rats and guinea pigs and 0.34 mg/l for monkeys. The NOAEC for systemic effects was 0.21 mg/l for monkeys and rats, and 0.34 mg/l for the guinea pigs. The study does satisfy main requirements of an OECD 412 guideline study but the concentration levels are not adequate to assess high dose exposure. Hence, this study is considered inacceptable for regulatory purposes.


In a combined subacute/subchronic dermal toxicity study 6 New Zealand White rabbits were exposed to 2 ml/kg/day of a 2.5 % aqueous solution of Na3NTA (50 mg/kg/day) either to clipped (28 days, 20 treatments) or abraded skin (91 days, 65 treatments) (Nixon, 1971). No adverse gross effects were observed. Growth, organ to body weight ratios, and haematologic values were within normal limits. Microscopic examination revealed only mild skin irritation; no treatment associated change was registered in any of the 15 internal organs examined. Although no indication on systemic effects was observed, the dose regime in this study is considered inacceptable for regulatory purposes. The NOAEL for local effects on the intact or abraded skin was 50 mg/kg bw/d.

Repeated dose toxicity: via oral route - systemic effects (target organ) urogenital: kidneys

Justification for classification or non-classification

No full OECD guideline study is available for the assessment of repeated dose toxicity. However, the available studies on oral toxicity provide a sound base for a weight of evidence assessment. At lower concentrations nodular hyperplasia and urothelium lesions were shown to be reversible. Only at high doses (2 % Na3NTA) hydronephrotic response was sufficiently severe to preclude the restoration of normal renal structure in a few rats. A NOAEL of 92 mg/kg/d for cytotoxic effects was established in a 2 year carcinogenicity study (NCI, 1977). Na3NTA does not need to be classified as causing specific target organ toxicity.