2,2'-iminodiethanol

Endpoint summary

Key value for chemical safety assessment

Effects on fertility

Description of key information

For DEA CAS 111-42-2, an extended One-Generation Reproductive Toxicity Study (EOGRTS) according to OECD TG 443 has been performed and finalized on January 29^th2018 (Ethanolamine Reach Consortium). Based on the final decision of the substance evaluation under CORAP cohorts 2A and B as well as cohort 3 were included for an investigation of the developmental neurotoxicity module (DNT) and the developmental immunotoxicity module (DIT), but without the extension to an F2-generation. Additionally, measurements of the essential nutrient choline were performed in different tissues of the F0 and F1 pups (plasma and liver, respectively). Furthermore, additional investigation of parameters such as platelet-activating factor (PAF) have been included as another modification to follow-up on a mode-of-action.

2,2’-iminodiethanol was administered to groups of 30 male and 30 female healthy young Wistar rats (F0 parental generation) as a solution to the drinking water in different concentrations (0, 100, 300 and 1000 ppm). At least 16 days after the beginning of treatment, F0 animals were mated to produce a litter (F1 generation). Mating pairs were from the same dose group. Pups of the F1 litter were selected (F1 rearing animals) and assigned to 5 different cohorts which were continued in dose groups 10 - 13 in the same fashion as their parents and which were subjected to specific post weaning examinations. The study terminated with the terminal sacrifice of the male and female animals of cohort 1B. Test drinking water containing 2,2’-iminodiethanol were offered continuously throughout the study.

Intake of test substance: the overall mean dose of 2,2’-iminodiethanol throughout all study phase and across all cohorts was approx. 12.75 mg/kg body weight/day (mg/kg bw/d) in the 100 ppm group, approx. 37.68 mg/kg bw/d in the 300 ppm group and approx. 128.35 mg/kg bw/d in the 1000 ppm group.

Under the conditions of the present modified extended 1-generation reproduction toxicity study the NOAEL (no observed adverse effect level) for general toxicity is 100 ppm for the F0 parental animals, based on evidence for distinct kidney toxicity and stomach irritation, as well as corresponding effects on water consumption, food consumption, body weights and clinicopathological parameters, which were observed at the LOAEL (Lowest Observed Adverse Effect Level) of 300 ppm. Similar toxicity was noted in the adolescent F1 animals, which had no stomach irritation but liver toxicity in addition.

The NOAEL for fertility and reproductive performance for the F0 and F1 rats is 300 ppm, based on a lower number of implants, prolonged/irregular estrous cycles as well as pathological changes in sexual organs, pituitary and mammary glands of both genders at the LOAEL (Lowest Observed Adverse Effect Level) of 1000 ppm. Most of the reported effects on reproduction and reproductive organs occurred in the range of general and systemic toxicity and have been assessed to be secondary in nature. Please compare with the section “Justification for classification or non classification” for a detailed discussion. However, eosinophilic cysts in the pituitary gland were present in the F1 animals of cohort 1A down to the 100 ppm dose level, but no assessment on adversity of this finding is possible at present. Therefore, no NOEL can be established for this particular effect.

The increased mean dose-group T4 values observed in some of the dose groups were within historical control data and there were no significant changes of TSH observed in any dose groups.

A dose dependent statistically significant choline depletion was observed in plasma and liver tissue of both male and female F1B animals starting already from the lowest dose tested.

The NOAEL for developmental toxicity in the F1 progeny is 100 ppm, based on impaired pup survival at 1000 ppm as well as reduced pup body weights in the F1 offspring, which were observed at the LOAEL (Lowest Observed Adverse Effect Level) of 300 ppm. As these weight reductions were only observed in the presence of maternal toxicity, including lower weight gain during pregnancy, they are not regarded as independent effect of the treatment.

The NOAEL for developmental neurotoxicity for the F1 progeny is 300 ppm, based on adverse clinical observations, impaired auditory startle response and corresponding neuropathological findings at the LOAEL (Lowest Observed Adverse Effect Level) of 1000 ppm.

The NOAEL for developmental immunotoxicity for the F1 progeny is 300 ppm, based on effects on the T-helper cells and cytotoxic T-cells in the spleen in the F1 females at the LOAEL (Lowest Observed Adverse Effect Level) of 1000 ppm. Lower mean and median anti-SRBC IgM antibody titers of the positive control group (4.5 mg/kg bw/d cyclophosphamide, oral) demonstrated that the test system worked properly.

 

In addition to the EOGRTS / OECD 443 that is available for DEA, there is a 3-month inhalation study (BASF AG, 2002) in rats reporting an influence on the male reproductive system at the high concentration. The NOAEC for male fertility parameters was 0.15 mg/l. When DEA was orally administered to rats via the drinking water for 13 weeks, decreases in testis and epididymis weights, testicular degeneration, atrophy of the seminal vesicles and prostate glands and associated effects on spermatology were observed. The NOAEL for fertility effects in males was 48 mg/kg bw. In all of these studies no histopathological effects were observed in female reproductive organs.

 

 

 Other data on Diethanolamine (DEA)

The US National Toxicology Program conducted a prenatal developmental toxicity study by gavage administration of DEA to groups of 12 pregnant CD rats at doses of 0, 50, 125, 200, 250, or 300 mg/kg bw/d during GD 6–19 (Price et al., 2005). The test substance was administered in water, and the pH of the dosing solutions was adjusted to 7.4±0.2 with hydrochloric acid. Dams were allowed to litter and raise their offspring to PND 21.

Dams in the high dose group showed signs of excessive toxicity and were euthanised by GD 15; the following discussion omits further mention of this group. The principal toxicity noted in the remaining dams was of a dose-related reduction in bodyweight (gain) during gestation compared to the control group. This was most noticeable at 250 mg/kg BW/d, in which animals had lost weight by GD 12 and didn’t start gaining weight until GD 15, after which weight gain maintained parity with the control. Animals in the 200 mg/kg bw/d group lost weight by GD 12, but weight gain maintained parity with the control thereafter. There were no differences in bodyweight (gain) between control and the other treatment groups.

There was a statistically significant increase in post-implantation loss in the groups that were administered 200 or 250 mg/kg bw/d, which in the higher dose group was also manifested as total loss of four litters in dams that survived until term, one of which consisted of all dead pups at PND 0 and the other three which consisted only of implantation sites. It is not clear from the publication as to whether the implantations were lost at an early or later stage. Two other dams in this group were either found dead on GD 15 or euthanised moribund on GD 21. Both had litters. One dam in the 200 mg/kg bw/d dose group was euthanised on GD 22 and had a litter of dead foetuses. There was a statistically significant increase in postnatal mortality during PND 0–4 in groups administered ≥125 mg DEA/kg bw/d.

In general, Ethanolamines inhibit the uptake of [3H]-choline in cultured CHO cells, with estimated EC50values of ~0.2 mM for DEA and ~1.1 mM for the structural analogue TEA (Lehmann-McKeeman and Gamsky, 1999; Stottet al., 2004).

 

For DEA various mechanistic in vitro and in vivo studies identified that choline depletion is the key event in hepatic carcinogenicity. DEA decreased gap junctional intracellular communication in primary cultured mouse and rat hepatocytes; induced DNA hypomethylation in mouse hepatocytes; decreased phosphatidylcholine synthesis; and increased S-phase DNA synthesis in mouse hepatocytes, but had no effect on apoptosis. All of these effects were mediated by the inhibition of choline sequestration, and were prevented with choline supplementation. No such effects were noted in human hepatocytes in vitro. Apparent differences in the susceptibility of two different mice strains (B6C3F1 > C57BL) were noted. B6C3F1 mice are extremely sensitive to non-genotoxic effects and are susceptible to spontaneous liver tumors. Moreover, chronic stimulation and compensatory adaptive changes of hepatocyte hypertrophy and proliferation are able to enhance the incidence of common spontaneous liver tumors in the mouse by mechanisms not relevant to humans (adapted from the DEA OECD SIAR, 2009).

Effects of DEA on pre- and post-implantation may be mediated by effects on choline homeostasis (as described above) rather than through a direct embryo toxicity. These effects may be inhibition of choline-uptake in the liver, subsequent perturbation of choline-homeostasis, with subsequent impairment of C1-metabolism, DNA-methylation, lipid metabolism, and intercellular communication. Choline metabolism is connected to Phosphatidylcholine and Betaine. The latter is reported to be central for the synthesis of SAM (S-Adenosyl-Methionine), a principle methylating agent for biosynthetic pathways and maintenance of critical gene methylation patterns (Stott et al. 2004; Zeisel and Blusztajn, 1994). 

It has been accepted for DEA tumorigenicity, that the effects observed are caused by a non-genotoxic modulation of DNA-methylation. Such effects may also explain the observed effects on implantation. Significantly, this is important for the final evaluation of DEA as this potential mode-of-action may display a species-specific effect with humans being resistant towards choline-deficiency and its consequences in rodents.

Recently, it could be demonstrated, that Monoethanolamine (MEA) reduced implantation success in a two-generation reproduction toxicity study (ACC and CEFIC, 2009). When administered to pregnant rats during gestation days (GD) 1–3, 4–5, or 6–7, MEA had no effect upon implantation success. In a second experiment, MEA was administered either in the diet or by oral gavage from two weeks prior to mating through to GD 8. Parallel groups also received a diet supplemented with choline. In the absence of supplementary choline, MEA induced early resorptions, statistically significant only when administered in the diet. A slight reduction in implantation success was ameliorated by supplementary choline. We conclude that implantation is affected by MEA only when exposure starts before mating; that dietary administration is more effective than gavage dosing; and that interference with choline homeostasis may play a role in the etiology of this lesion (Moore et al 2018, Reprod. Tox. 78, 102 -110).

As supporting evidence, in the EOGRTS / OECD443 which is available for DEA, a statistically significant decrease in the choline levels at all doses was observed in F1B animals. The analytical results demonstrated the presence of choline in all plasma samples from the animals dosed with the test substance DEA (100 ppm, 300 ppm and 1000 ppm dosed animals) and in those from control, non-dosed animals. In general, it can be stated that the administration of 2,2’-iminodiethanol led to a reduction in the content of choline in the plasma samples analyzed. The mean plasma choline level decreased dose dependently and statistically significant through the dose groups. Furthermore, also in the offspring the analytical results demonstrated the clear presence of choline in all liver samples from the animals dosed with the test substance DEA (100 ppm,300 ppm and 1000 ppm dosed animals) and in those from control, non-dosed animals. This was true from all time points investigated (4-day old pups, 22-day old descendants and ~90-day old adolescents). In general, it can be stated that the presence of the test substance DEA led to a reduction in the content of choline in the liver samples analyzed. This effect appears to be dose-dependent, in that higher dose levels were associated with greater choline reduction. At higher dosing levels, no further dramatic liver choline content reduction was observed, indicating that liver choline levels of all non-control animals have reached an approximate minimum and distinct choline depletion can be induced already at diet concentrations as low as 100 ppm DEA.

 

 

 

References 

Lehman-McKeeman LD, Gamsky EA (1999). Diethanolamine inhibits choline uptake and phosphatidylcholine synthesis in Chinese hamster ovary cells. Biochem Biophys Res Commun262:600–604.

OECD SIDS (2009). Diethanolamine.

Smyth et al., (1951).Range-finding Toxicity Data: List IV.Arch.Hyg. Occup. Med.4: 119 - 122

Stott WT, Kleinert KM (2008). Effect of diisopropanolamine upon choline uptake and phospholipid synthesis in Chinese hamster ovary cells. Fd Chem Toxicol46:761-766.

Stott WT, Radtke BJ, Linscombe VA, Mar M-H, Zeisel SH (2004). Evaluation of the potential of triethanolamine to alter hepatic choline levels in female B6C3F1 mice.Toxicol Sci79:242-247.

Zeisel SH and Blasztajn JK (1994). Cholin and human nutrition. Ann. Rev. Nutr.14: 269-296

 

 

 

 

 

 

 

 

 

Effects on developmental toxicity

Description of key information

Based on the available developmental toxicity studies (via the inhalation, dermal and oral route of exposure) with rats and rabbits, DEA caused only developmental toxicity in the presence of clear maternal toxicity and at dose levels considered as high. Therefore no DNEL has to be derived based on developmental toxicity. Furthermore, maternal toxicity was observed at levels higher/comparable to general toxic effects in the repeated dose toxicity studies.

Additional information

The exposure of pregnant female Wistar rats to an aerosol of DEA in a head/nose exposure systems for 6 h/day on day 6 through day 15 post coitum at concentrations of 0; 0.01; 0.05; 0.2 mg/l (0; 10, 50, 200 mg/m³) led to signs of maternal toxicity at the highest concentration (0.2 mg/l) (BASF AG, 1993b, OECD TG 414 study). Maternal toxicity was substantiated by adverse clinical symptoms (vaginal haemorrhages) in 8 of the 21 pregnant rats on day 14 p.c. At this dose level a markedly increased number of fetuses with skeletal variations (mainly cervical rib(s)) were also recorded but substance-related teratogenic effects were not detected in any foetus. Thus, signs of prenatal developmental toxicity did only occur at a maternal toxic concentration. There were no adverse effects on dams or foetuses at the low or mid concentrations (0.01 or 0.05 mg/l). The NOAEC for maternal and prenatal developmental toxicity was 0.05 mg/l (50 mg/m³), the NOAEC for teratogenicity was >0.2 mg/l (200 mg/m³).

 

DEA was administered dermally to pregnant CD rats from gestation day 6 through day 15 at doses of 0, 150, 500 and 1500 mg/kg bw/day (Marty et al, 1999, a protocol equivalent/similar to OECD TG 414). At 500 and 1500 mg/kg bw/day moderate and severe skin irritation was caused, respectively. Maternal body weight gain was decreased in the 1500 mg/kg bw. Absolute and relative kidney weights were increased at 500 and 1500 mg/kg bw/day. Haematological effects including anaemia, abnormal red cell morphology (poikilocytosis, anisocytosis, polychromasia), and decreased platelet count were observed in all treatment groups. The 1500 mg/kg bw/day group also had increased lymphocytes and total leukocytes. In the fetuses, there were no effects of treatment on body weight or on incidence of external, visceral, or skeletal malformations/abnormalities. Increased incidences of six skeletal variations involving the axial skeleton and distal appendages were observed in litters from the 1500 mg/kg bw/day group. The skeletal variations included poor ossification in the parietal bones, cervical centrum #5, and thoracic centrum #10, lack of ossification in all proximal hindlimb phalanges and some forelimb metacarpals, and callused ribs. Consequently, the LOAEL for maternal toxicity was 150 mg/kg bw/day, while the NOAEL for prenatal developmental toxicity was adjusted to 380 mg/kg bw due to dosing discrepancy at the 500 mg/kg bw/day group. The NOAEL for teratogenicity was >1500 mg/kg bw/day. Thus, signs of prenatal developmental toxicity did only occur at clearly maternal toxic dose levels.

 

DEA was administered dermally to pregnant New Zealand White rabbits from gestation day 6 through day 18 at doses of 0, 35, 100 and 350 mg/kg bw/day (Marty et al, 1999, a protocol equivalent/similar to OECD TG 414). Rabbit dams at 350 mg/kg bw/day showed several signs of marked skin irritation, reduced food consumption, and colour changes in the kidneys but no haematological changes. Body weight gain was reduced at 100 mg/kg bw/day. There was no impairment of gestational parameters. No evidence of developmental toxicity was observed at any dose level, especially, there were no apparent effects of treatment on the incidences of external, visceral, or skeletal abnormalities. Consequently, the NOAEL for maternal toxicity was 35 mg/kg bw/day, the NOAEL for prenatal developmental toxicity including teratogenicity was >350 mg/kg bw/day.

 

Treatment of time-mated pregnant rats with DEA dose levels of 0, 50, 125, 200, 250, 300 mg/kg bw/day by oral gavage from gestation day 6-19 (Price, 1999) led to maternal morbidity or mortality occurred at 200 and 250 mg/kg bw/day and all females at 300 mg/kg bw/day had to be terminated early due to excessive toxicity. Maternal water intake was transiently affected during early gestation (125 and 250 mg/kg bw/day) but was comparable to controls for all measurement periods after GD 12. Maternal absolute kidney weight was increased on PND 21 (>= 125 mg/kg bw/day) indicating persistence of DEA-induced toxicity for up to about 3 weeks after cessation of exposure. Reduced maternal body weight and weight change, as well as reduced feed intake, were noted at >= 200 mg/kg bw/day. Exposure to 50 mg/kg bw/day was not associated with any significant maternal toxicity during or after the treatment period. Developmental toxicity was observed specifically in form of an increase in post implantation mortality at >= 200 mg/kg bw/day on PND 0, and early postnatal mortality (PND 0-4) was increased at >= 125 mg/kg bw/day. Pup body weight was reduced at >= 200 mg/kg bw/day, with females more affected than males. When expressed as a percentage of control weight, pup body weight reductions were most pronounced during the early postnatal period. Statistically significant differences were also evident at the end of the lactation period. Consequently, the NOAEL for maternal and developmental toxicity was 50 mg/kg bw/day. Signs of developmental toxicity did only occur at maternal toxic dose levels.

Mode of Action Analysis / Human Relevance Framework

For details cf. "Justification for classification or non-classification"

Justification for classification or non-classification

Based on the available data, diethanolamine is classified and labelled for effects on fertility and developmental toxicity according to EU Classification, Labelling and Packaging of Substances and Mixtures (CLP) Regulation (EC) No. 1272/2008 (cat. 2; H371).

 

Reasoning for classification:

Reproductive toxicity was substance- and dose related but occurred in the presence of distinct general systemic toxicity in the dams and in the offspring. Administration of DEA in the drinking water resulted in reduced water and food consumption (particularly marked during lactation), with accompanying reductions in body weight, body weight gain, and terminal body weight. These findings were consistent across sexes, generations, and cohorts. The severity of findings increased with dose. In most cases, by study termination the body weights of treated groups were within 10% of the control group mean value, except in the case of HD F1 males and females, where mean body weight values were 80±5% of control. However, more critical reductions in body weight were seen during the study. Effects have been reported on kidneys where treatment was associated with tubular degeneration/regeneration in both F0 and F1 animals of the MD and HD groups. This was accompanied by signs of impaired renal function (increased urine volume, decreased urine specific gravity). Liver effects were associated with hypertrophy in both F0 and F1 animals of the HD group. F1 males also exhibited this change and fatty change also at the MD level. Additionally, a microcytic anemia was evident in the HD groups. DEA can be transformed into phosphoglyceride and sphingomyelin analoga leading to disturbances in phospholipid metabolism and a multitude of physiological functions, supporting a secondary mode of action (cf. appendix for further details).

Thus, the observed reproductive toxicity cannot be regarded as an isolated effect and may be attributed to general systemic toxicity.

Furthermore, as a possible mode-of-action (MOA) with quantitative differences in species specificity, DEA caused a clear reduction of choline levels in plasma and liver of all dose groups. DEA is already labelled for specific target organ toxicity (STOT RE cat. 2) for effects on blood, liver, kidney and nervous system which may be attributed to a disturbed choline-homeostasis. Choline is an essential nutrient; however, rodents appear to be more susceptible towards an impaired choline-homeostasis than humans. Leung et al. (2005) summarized the evidence why humans are less susceptible for choline-deficiency than rodents in the context of the carcinogenicity endpoint (further references given within the original article):“…choline is an essential nutrient in all mammals, the proposedmechanism of DEA-induced choline deficiency is qualitatively applicable to humans. However, there are marked species differences in susceptibility to choline deficiency, with rats and mice being far more susceptible than other species including humans. These differences are attributed to quantitative differences in the enzyme kinetics controlling choline metabolism. Rats and mice rapidly metabolize choline to betaine in the liver and it is likely that choline oxidase activity determines choline requirements and controls species sensitivity to choline deficiency. For example, choline oxidase activity is much lower in primates than rodents and primates are less sensitive to choline deficiency. Humans have the lowest choline oxidase activity of all species and are generally refractory to choline deficiency, with evidence of choline deficiency observed only after prolonged fasting, significantly depressed liver function or deficient parenteral feeding. It is noteworthy that there was no evidence of GJIC inhibition in human hepatocytes treated with DEA or cultured in choline-deficient media.”

According to chapter 3.7.2.2.1. of Regulation EC 1272/2008 (CLP), "Classification as a reproductive toxicant is intended to be used for substances which have an intrinsic, specific property to produce an adverse effect on reproduction ... .".

According to chapter 3.7.2.4.3 of Regulation EC 1272/2008 (CLP), "Classification should not automatically be discounted for chemicals that produce developmental toxicity only in association with maternal toxicity, even if a specific maternally-mediated mechanism has been demonstrated. In such a case, classification in Category 2 may be considered more appropriate than Category 1.".

According to chapter 3.7.2.3.2 and 3.7.2.5.5 of Regulation EC 1272/2008 (CLP), ". . . when the toxicokinetic differences are so marked that it is certain that the hazardous property will not be expressed in humans then a substance which produces an adverse effect on reproduction in experimental animals should not be classified.".

In conclusion, classification with category 2 for reproductive toxicity (H361) is considered the most appropriate in line with the criteria laid down in Regulation EC 1272/2008 (CLP).

 

 

 

 

Appendix

Statistically significant findings by dose level

In most cases, cohorts 1A and 1B are combined. Derived from report summary findings. Detailed explanation for the annotations (1), (2), (3) and (4) are given below the table. The critical effects for the endpoint evaluation (marked in bold in the table) and the interpretation thereof are being discussed in the text below the table.

 

Dose	Group	Findings
HD	F0(M)	↓ food consumption, body weight, body weight gain
		↑ liver weight, kidney weight
		Microcytic anaemia, ↓ prothrombin time, ↑ platelet count
		↑ urea, albumin, AST, ALP
		↓ urinary specific gravity, ↑ urine volume
		Liver: centrilobular hypertrophy (minimal to slight)
		Kidney: tubular degeneration/regeneration (minimal to slight)
	F0(F)	↑gestation length (4)
		↓ implantation sites, litter size (4)
		↓ water and food consumption, body weight, body weight gain
		↑ liver weight, kidney weight
		Microcytic anaemia,↓ prothrombin time
		↑ urea, albumin, ↓ platelet activating factor
		Liver: centrilobular hypertrophy (minimal to slight)
		Kidney: tubular degeneration/regeneration (minimal to moderate)
		Glandular stomach: erosion/ulcer, oedema, inflammation (minimal to slight)
	F1	↑ dead and cannibalised offspring (4)
		↓ body weight from PND 4 to weaning
		↑ T4 in female weanlings (2)
	1A/1B (M)	High-stepping gait, piloerection
		↓ water and food consumption, body weight, body weight gain
		Small and immature testes (1)
		Microcytic anaemia, ↑ platelet count
		↑ urea, AST, ALP
		↑ T4
		↑ liver weight, kidney weight
		Kidney: tubular degeneration/regeneration (slight to moderate)
		Liver: centrilobular hypertrophy (minimal to slight), peripheral hypertrophy (slight to moderate), peripheral fatty change (minimal to slight)
		Mammary gland: feminisation (slight to severe), diffuse hyperplasia in one animal (moderate) (1)
		Pituitary gland: eosinophilic cysts (minimal to moderate) (2)
		Testis: immature epithelium (severe), tubular degeneration (slight to moderate) (1)
	1A/1B (F)	High-stepping gait, piloerection
		↓ water and food consumption, body weight, body weight gain
		Prolonged/irregular oestrus cycle (4)
		Microcytic anaemia, ↓ prothrombin time, ↓ monocyte count
		↑ urea, AST
		↓ T-cell helper count and CD4/CD8 ratio
		↑ T-cell count
		↑ T4 (2)
		↑ liver weight, kidney weight
		Kidney: tubular degeneration/regeneration (slight to moderate)
		Liver: centrilobular hypertrophy (minimal to slight), peripheral fatty change (minimal to slight)
		Mammary gland: increased secretion
		Ovary: ↓ size, diffuse atrophy, luteal cysts, (1) ↓ follicles (primordial and growing), absent corpora lutea (two females of 1B) (4)
		Pituitary gland: eosinophilic cysts (minimal to moderate) (2)
	2A (M)	↓ water and food consumption, body weight, body weight gain
		↓ maximum amplitude of auditory startle response, no habituation to test environment
		CNS: degeneration of nerve fibres of medulla oblongata and spinal cord (3)
		Pituitary gland: eosinophilic cysts (2)
	2A (F)	High-stepping gait, piloerection (one animal)
		↓ water and food consumption, body weight, body weight gain
		↓ maximum amplitude of auditory startle response, no habituation to test environment and
		CNS: degeneration of nerve fibres of medulla oblongata and spinal cord (3)
		Pituitary gland: eosinophilic cysts (2)
	2B (M/F)	↓ terminal body weight
		Pituitary gland: eosinophilic cysts (2)
	3 (M)	High-stepping gait, piloerection (one animal)
		↓ water and food consumption, body weight, body weight gain
		Small testes, epididymides, prostate, seminal vesicle (one animal) (1)
	3 (F)	High-stepping gait, piloerection (one animal)
		↓ water and food consumption, body weight, body weight gain
MD	F0(M)	↓ body weight, body weight gain
		↑ kidney weight
		↓ prothrombin time,↑ platelet count
		↓ urinary specific gravity, ↑ urine volume
		Kidney: tubular degeneration/regeneration (minimal)
	F0(F)	↓ water and food consumption (postnatal), body weight, body weight gain
		↓ prothrombin time
		Glandular stomach: erosion/ulcer, oedema, inflammation (minimal to moderate)
	F1	↓ body weight from PND 14 to weaning
		↑ platelet count in males
		↑ T4 in female weanlings (2)
	1A/1B (M)	↓ water and food consumption, body weight, body weight gain
		↑ liver weight, kidney weight
		Kidney: tubular degeneration/regeneration (minimal to slight)
		Liver: centrilobular hypertrophy (minimal), peripheral fatty change (minimal to slight)
		Pituitary gland: eosinophilic cysts (minimal to slight) (2)
	1A/1B (F)	↓ water consumption, body weight, body weight gain
		↑ kidney weight
		Kidney: tubular degeneration/regeneration (minimal to slight)
		Pituitary gland: eosinophilic cysts (minimal to slight) (2)
	2A (M)	Pituitary gland: eosinophilic cysts (2)
	2A (F)	↓ water consumption, body weight, body weight gain
		Pituitarygland: eosinophilic cysts (2)
	2B (F)	↓ terminal body weight
	3	–
LD	F0(M/F)	–
	F1	↑ T4 in females PND 4 (2)
	1A (M/F)	Pituitary gland: eosinophilic cysts (2)
	1B (M/F)	–
	2A (M/F)	Pituitary gland: eosinophilic cysts (2)
	2B (M/F)	–
	3	–

 

(1) Secondary Effects to maternal toxicity

Effects marked with (1) can be attributed to be secondary to reductions in body weight in consequence of general systemic toxicity. References are given within the final report. Therefore, as already identified in the summary of effects, the following changes in reproductive organs can be attributed to the reduced body weight and probably reduced choline-levels in the F1 generation

Testis: Macroscopically small testes were identified in three HD males from cohort 1A (476, 478, 479), three from cohort 1B (693, 694, 697), and one from cohort 3 (1240). The size of the secondary sex organs (epididymis, prostate, seminal vesicle) was also reduced. Small testis was correlated with immature or degenerative epithelium. Further evaluation shows that the affected individuals were overall much smaller than their counterparts, and the effects on the reproductive tract organs are considered secondary to retarded growth.

The mean terminal body weights of the animals in Cohort 1A and 1B were 24 % and 20 % lower compared to the controls. However, the body weights of animals for which above mentioned findings and effects in the reproductive organs have been observed were even more reduced (45 – 70 % lower compared to controls).

Ovary:Four HD animals in cohort 1B (777, 785, 797, 798) had small, atrophied ovaries with reduced or absent corpora lutea. These animals were of overall retarded growth, and the ovary changes compared to control are considered a secondary consequence.

Differential ovarian follicle count (both primordial and growing follicles) was reduced in the cohort 1A and 1B (combined) HD group, but not in the F₀HD group.

Mammary gland:Feminisation of the mammary gland (grade 2–4) was identified in four HD males of cohort 1A, one of which also showed diffuse hyperplasia (grade 3). It should be noted that there was insufficient mammary tissue for another four of the 25 males in this cohort.

There was an increase in mammary gland secretion in six females of the HD cohort 1A group.

The above-mentioned findings can be clearly attributed to be a consequence of general systemic toxicity caused by an impaired choline homeostasis. The body weights were 16 % lower in Cohort 1B (where findings occurred) respectively and the animals with above mentioned findings were 18 - 50 % lower compared to controls.

Additionally, animals showed clinical signs (e.g. high-stepping gait, piloerection etc.) and mortality occurred in the F1 HD groups (2 males and 1 female in two rearing cohorts from cohorts 1A and 1B were either sacrificed moribund).

 

Further literature research revealed (Lindern-Moore et al., 1981) (cite references) that similar effects can be expected in case of starvation experiments in rats, especially in weanling and growing rats which appears not to be surprising assuming a higher energy demand. Unfortunately, similar reports could not be retrieved for an induced choline-deficiency, where focus has been laid on organ effects and CNS effects, but not on reproductive organs and outcome.

In addition, “secondary” evidence shows data from genetic knockout models can be used demonstrating that effects on implantation loss might be referred to an impaired choline-homeostasis and/or overall altered lipid metabolism. This before mentioned evidence being derived from the reproductive outcome of such transgenic animals with a knockout of an enzyme being involved in phospholipid-synthesis is only secondary and indirect. Therefore and as there is non direct evidence from e.g. a reproductive screening using choline-deficiency, certain effects reported in the EOGRTS for DEA have been categorized as being “critical effects” which deserve further discussion and mechanistic investigation (i.e. to establish the link as a secondary effects on a molecular level).

Moreover, effects on choline-homeostasis have been observed in all dose groups and levels (i.e. reduction of choline-levels.) Choline is an essential nutrient in humans and other mammalian species such as the rat. Wistar rats are assumed to be very sensitive towards choline-deficiency (Kirsch et al., 2003; Nocianitri and Aoyama, 2001).

 

Effects, that are covered by the existing “Specific target organ toxicity” STOT RE cat. 2 already:

Kidney: Treatment was associated with tubular degeneration/regeneration in both F0 and F1 animals of the MD and HD groups. This was accompanied by signs of impaired renal function (increased urine volume, decreased urine specific gravity). Similar findings have been previously reported in rats (please compare with the repeated dose toxicity section of the IUCLID6 – chapter 7.5.1.)

Liver: Treatment was associated with hypertrophy in both F0 and F1 animals of the HD group. F1 males also exhibited this change and fatty change also at the MD level. Similar findings have been previously reported in mice².

A STOT RE cat. 2 has a cut-off according to the CLP of mg/kg BW/day. Therefore, no additional label is warranted and the hazard is considered to be already covered by the existing classification and labeling.

 

(2) Effects of lower concern not triggering classification (adversity not clearly given)

In this section, effects are located which were observed at all dose levels.

The adversity of effects is however unclear and effects need to be discussed therefore.

Pituitary gland: eosinophilic cysts occurred in the pars distalis of F1 males and females from all dose levels. Pathologists could not determine the adversity of this finding.

Plasma hormones: The increased mean dose-group T4 values observed in some of the dose groups were within historical control data and thus not biologically significant. In addition, there were no significant changes of TSH observed in any dose groups; supporting the evidence that administration of DEA has no significant impact on thyroid hormone homeostasis. The production of thyroid hormones is primarily regulated by thyroid-stimulating hormone (TSH) released from the anterior pituitary gland. TSH release is in turn stimulated by the thyrotropin-releasing hormone (TRH) from the hypothalamus. The thyroid hormones provide negative feedback to TSH and TRH: when the thyroid hormones are high, TSH production is suppressed (ECHA, 2018; Joseph-Bravo et al., 2016).

 

(3) Effects already covered by the existing C&L - STOT RE cat.2

Central nervous system (cohorts 2A and 2): Degeneration of nerve fibres in the medulla oblongata and spinal cord was observed in HD males and females of cohort 2A, but not cohort 2B. Demyelination of the medulla and spinal cord has been reported at a similar dose in a 90-day repeated-dose toxicity study in F344 rats (NOAEC=630ppm, LOAEC=1250ppm), but not in mice. It is clear, therefore, that these findings represent a response to continuous postnatal exposure rather than an effect upon development per se.Therefore, these effects are thought to be covered by a STOT RE cat. 2.

The only functional deficits to be identified were a decrease in the maximum amplitude of the startle response, although latency was unaffected, and lack of habituation to the test environment. No other treatment-related changes were identified.

 

(4) Effects that may triggering further mechanistic research

Effects marked with (4) are not directly associated with the general and the systemic toxicity observed. However, as a “secondary” evidence, data from genetic knockout models can be used demonstrating that effects on implantation loss might be referred to an impaired choline-homeostasis and/or overall altered lipid metabolism.

Gestation and outcome: Parturition was slightly (+2%), but statistically significantly, delayed in the HD group. The number of implantations sites per dam was decreased, and the incidence of stillborn pups was increased, resulting in an overall reduction in litter size in this group. Survival through to weaning was unaffected in all treatment groups compared to control.

Gonads:Differential ovarian follicle count (both primordial and growing follicles) was reduced in the cohort 1A and 1B (combined) HD group, but not in the F0 HD group.

Secondary evidence comes from the literature. Vance and Vance reported (2009) on the physiological consequences of disruption of mammalian phospholipid biosynthetic genes:

“…several lines of mice with disrupted genes of phospholipid biosynthesis were generated. From this research, we have learned that embryonic lethality occurs in mice that lack choline kinase (CK) a, CTP:phosphocholine cytidylyltransferase a, CTP:phosphoethanolamine cytidylyltransferase, or phosphatidylserine decarboxylase. Whereas mice that lack CK b are viable but develop hindlimb muscular dystrophy and neonatal bone deformity. Mice that lack CTP:phosphocholine cytidylytransferase b have gonadal dysfunction and defective axon branching.”

The disruption of genes and enzymes in consequence involved in phosphatidyl-choline synthesis is thought to be similar to a state of choline-depletion or at least impaired choline-homeostasis. Accordingly, e.g. embryonic death has been reported as physiological consequence for some knockout models of certain enzymes being involved in phospholipid synthesis (Vance & Vance; 2009).

Additionally, it has been demonstrated by Jackowski and coworkers (2004) that disruption of CCTß2 (CTP:phosphocholine cytidylyltransferase, CCT) expression leads to gonadal dysfunction and degeneration in mice (i.e. defective ovarian follicle development and disruption of spermatogenesis).

With respect to the effects observed on DOFC both primordial and growing follicles (down to 89% and 72% compared to controls, respectively), similar effects have been described after starvation and semi-starvation of pre-pubertal Wistar rats (Lintern-Moore et al., 1981). Herein the authors describe that after starvation of young animal from PND 21 – 42 leads to a reduction growing follicles on day 42. Additionally, no corpora lutea were present. However, after re-feeding (ad libitum) up to day 66, growing follicle numbers and corpora lutea could be restored. The body weights caused by the starvation were reduced by starvation but recovered after refeeding up to day 66. This means, that the effects on DOFC might be a secondary consequence of an impaired choline-homeostasis (i.e. reduced PC content as reported to be DEA-induced in vitro and in vivo).

A secondary mode of action is further supported by the fact that DEA is incorporated into phosphoglyceride and sphingomyelin analoga. Phosphatidylcholines are a class of phospholipids that incorporate choline as a headgroup, a glycerophosphoric acid, with a variety of fatty acids. Phosphatidylcholine is formed through the methylation of phosphatidylethanolamine catalyzed by phosphatidylethanolamine N-methyltransferase (Li and Vance, 2008). Sphingomyelins are a group of phospholipids, which consist of phosphocholine (or phosphoethanolamine) as a head group and a ceramide. The formation typically involves the enzymatic transfer of a phosphocholine head group from phosphatidylcholine to ceramide (Tafesse et al., 2006). Due to its structural similarity to ethanolamine and choline, DEA is metabolized by very similar pathways undergoing O-phosphorylation and N-methylation making incorporation into phosphoglyceride and sphingomyelin analogs possible (Leung et al., 2005). Barbee and Hartung could show in vitro and in vivo that DEA is incorporated into various phospholipid derivates and thereby competing with choline and ethanolamine inhibiting their incorporation. DEA-derived phospholipids are characterized by an increased biological half life and DEA-derived sphingomyelines are not being a suitable substrate for sphingomyelinase, which hydrolyzes sphingomyelines leading to free phosphocholine and ceramide (Barbee and Hartung, 1979). Thus, incorporation of DEA into phospholipid derivates is leading to disturbances in phospholipid metabolism and a multitude of physiological functions.

  

References

Barbee, S. J., Hartung, R. (1979). The effect of diethanolamine on hepatic and renal phospholipid metabolism in the rat. Toxicology and Applied Pharmacology 47(3): 421-430.

ECHA Guidance for the identification of endocrine disruptors in biocides and pesticides; EFSA Journal 2018; 16 (6); 5311

Leung HW, Kamendulis LM, Stott WT. (2005). Review of the carcinogenic activity of diethanolamine and evidence of choline deficiency as a plausible mode of action. Reg Tox and Pharmacol, 43: 260-271

Li, Z., Vance, D.E. (2008). Phosphatidylcholine and choline homeostasis. J. Lipid Res. 49, 1187–1194.

 

Moore NP et al. (2013). Guidance on classification for reproductive toxicity under the globally harmonized system of classification and labelling of chemicals (GHS).Critical Reviews in

Toxicology, 43:10, 850-891

Tafesse F., Ternes P. and Holthuis J. (2006). The Multigenic Sphingomyelin Synthase Family. The Journal of Biological Chemistry. 281, 29421-29425

 

Vance VE and Vance JE. (2009). Physiological consequences of disruption of mammalian phospholipid biosynthetic genes. Journal of Lipid Research, April suppl. S132-137

 

Nocianitri KA and Aoyama Y. (2001). Note: Different response to choline deficiency of the serum ornithine carbymoyltransferase activity in four strains of rats. Biosci. Biotechnol. Biochem. 65(4): 935-38.

 

Kirsch et al. (2003). Rodent nutritional model of non-alcoholic steatohepatitis: Species, strain and sex difference studies. Journal of Gastroenterology and Hepatology 18: 1272–1282

 

Jackowski S et al (2004). Disruption of CCT2 Expression Leads to Gonadal Dysfunction. MOLECULAR AND CELLULAR BIOLOGY 24(11): 4720–4733

Joseph-Bravo P, Jaimes-Hoy L and Charli JL, 2016. Advances in TRH signaling. Reviews in Endocrine and Metabolic Disorders, 17, 545–558.

 

Lindern-Moore S et al. (1981). The effect of restricted food intake and refeeding on the ovarian follicle population of the pre-pubertal Wistar rat. Reprod. Nutr. Develop. 21(5A9): 611-620