Registration Dossier

Data platform availability banner - registered substances factsheets

Please be aware that this old REACH registration data factsheet is no longer maintained; it remains frozen as of 19th May 2023.

The new ECHA CHEM database has been released by ECHA, and it now contains all REACH registration data. There are more details on the transition of ECHA's published data to ECHA CHEM here.

Diss Factsheets

Administrative data

Key value for chemical safety assessment

Effects on fertility

Description of key information

An oral NOAEL for parental animals and for offsprings was found to be > 1000 mg/kg bw/d in a three-generation reproduction toxicity study of male and female rats, DePass (1986).

NTP (1984) reported a study on fertility assessment of continous breeding (FACB) and found a NOEL of 1000 mg/kg bw/d for parental and F1 male and female mice, administering EG in drinking water.

Link to relevant study records

Referenceopen allclose all

Endpoint:
three-generation reproductive toxicity
Type of information:
experimental study
Adequacy of study:
weight of evidence
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
study well documented, meets generally accepted scientific principles, acceptable for assessment
Principles of method if other than guideline:
Assessing the effect of EG on fertility and general reproductive performance in male and female rats.
GLP compliance:
no
Limit test:
no
Species:
rat
Strain:
Fischer 344
Sex:
male/female
Details on test animals or test system and environmental conditions:
TEST ANIMALS
- Age at study initiation: young adult nulliparous Fischer 344 rats
- Housing: two per cage in stainless-steel wire cages; during mating, each male was housed with 2 females; after mating and during lactation, the females were housed individually in plastic showbox cages with hardwood chips for nesting.
- Diet: Purina Formulab, ad libitum
- Water: city water, ad libitum

ENVIRONMENTAL CONDITIONS
- Temperature (°C): 20-24°C
- Photoperiod (hrs dark / hrs light): 12 /12
Route of administration:
oral: feed
Vehicle:
unchanged (no vehicle)
Details on exposure:
Administration of EG to the F0 rats of both sexes started at approximately 7 weeks of age.
Fresh diet was prepared every 2 weeks with the percentage of test item (EG) adjusted, based on the group mean body weight and food consumption, so as to maintain a relatively constant dosage level. However, the concentration of EG in the diet was not changed during gestation or during the first week of lactation, but was reduced two- and three-fold during the second and third weeks of lactation, respectively, to adjust for increased food consumption by the dams. This change in concentration was based on earlier unpublished results from the laboratory. Increased food consumption during lactation has since been reported in another study performed at the laboratory.
Details on mating procedure:
At approximately 100 days of age, 10 males were added to 20 females in each dosage group. The F1 and F2 rats were treated as described for the F0 animals until approximately 100 days of age, at which time the animals were cohabited. Brother and sister matings were avoided for each generation.
Analytical verification of doses or concentrations:
not specified
Duration of treatment / exposure:
3 generations
Frequency of treatment:
daily
Details on study schedule:
The date of parturition and the number of live and dead newborn were recorded for each litter. The appearance and behavior of dams and pups were observed daily. Litter size was randomly reduced to 10, if necessary, on Day 4 postpartum. Offspring were weighed as litters at 4 and 14 days and individually at 21 days postpartum, the day they were weaned. F1 rats were randomly selected within each dosage group for the next mating. Each litter was represented except for those conceived very late in the mating period.
Dose / conc.:
40 mg/kg bw/day
Dose / conc.:
200 mg/kg bw/day
Dose / conc.:
1 000 mg/kg bw/day
No. of animals per sex per dose:
30
Control animals:
yes, concurrent no treatment
Details on study design:
Two untreated diet control groups, designated 0.0A and 0.0B, were included to estimate the variation between 2 groups treated alike.
Parental animals: Observations and examinations:
Body weights and diet consumption were recorded weekly except during gestation and lactation.
Litter observations:
Offspring were weighed as litters at 4 and 14 days and individually at 21 days post partum, the day they were weaned. F1 rats were randomly selected within each dosage group for the next mating.
Postmortem examinations (parental animals):
Necropsies were performed on five males and five females randomly selected from each dosage level of the F2 parents and the F3 weanlings. Microscopic examinations were performed on sections of liver, kidneys, lung, heart, adrenals, thyroid, trachea, accessory sex glands, adipose tissue, lymph nodes, pituitary, thymus, and testes and epididymis, or uterus and ovaries.
Statistics:
Continuous data such as body weights were compared by analysis of variance validated by Bartlett's test for homogeneity of variance. Duncan's multiple range test was used to identify individual mean differences when indicated by a significant F value. Where Bartlett's test indicated heterogeneous variances, t tests for equal or unequal variances were used to delineate differences between groups. Pup weights were compared by the method of Weil (Weil, 1970). Discontinuous data such as implantations and reproductive indices were compared by a multiple sum of ranks test. Frequency data were compared by the X2 test and by Fisher's exact test. The following reproductive indices were calculated and evaluated statistically by the previously described non parametric methods: fertility index (male and female), days from first mating to parturition, gestation index (fraction of pregnancies that resulted in litters with live pups), gestation survival index (fraction of newborn pups alive at birth), 0 to 4-day survival index, 4 to 14-day survival index, 4 to 21day survival index. The last four indices are summarized in the tables as means for ease of understanding and presentation, although the nonparametric statistical methods did not include a comparison of means.
Reproductive indices:
yes
Clinical signs:
no effects observed
Description (incidence and severity):
Throughout the study there was no effect of EG treatment on body weight gain or diet consumption, nor was there any mortality among parental rats.
Dermal irritation (if dermal study):
not examined
Mortality:
no mortality observed
Body weight and weight changes:
no effects observed
Food consumption and compound intake (if feeding study):
no effects observed
Food efficiency:
not examined
Water consumption and compound intake (if drinking water study):
not examined
Ophthalmological findings:
not examined
Haematological findings:
not examined
Clinical biochemistry findings:
not examined
Urinalysis findings:
not examined
Behaviour (functional findings):
not examined
Immunological findings:
not examined
Organ weight findings including organ / body weight ratios:
no effects observed
Histopathological findings: non-neoplastic:
no effects observed
Histopathological findings: neoplastic:
not examined
Other effects:
no effects observed
Reproductive function: oestrous cycle:
not examined
Reproductive function: sperm measures:
not examined
Reproductive performance:
not examined
Key result
Dose descriptor:
NOAEL
Effect level:
> 1 000 mg/kg bw/day
Based on:
test mat.
Sex:
male/female
Basis for effect level:
other: There were no reproductive effects associated with EG treatment doses up to 1000 mg/kg bw/d via diet.
Key result
Critical effects observed:
no
Reproductive performance:
no effects observed
Description (incidence and severity):
No treatment-related effect was observed for any of the indices. Also, EG treatment did not affect neonatal body weight at days 4, 14, or 21 post partum.
Clinical signs:
no effects observed
Dermal irritation (if dermal study):
not examined
Mortality / viability:
not examined
Body weight and weight changes:
not examined
Food consumption and compound intake (if feeding study):
not examined
Food efficiency:
not examined
Water consumption and compound intake (if drinking water study):
not examined
Ophthalmological findings:
not examined
Haematological findings:
not examined
Clinical biochemistry findings:
not examined
Urinalysis findings:
not examined
Sexual maturation:
not examined
Organ weight findings including organ / body weight ratios:
not examined
Gross pathological findings:
effects observed, non-treatment-related
Histopathological findings:
effects observed, non-treatment-related
Description (incidence and severity):
There were no treatment-related histopathologic findings in F2 parents or in F3 weanlings. Although the kidney has been shown to be the primary target organ for EG-induced toxicity, there was no increase in the incidence or severity of kidney lesions in this study. One high dose F2 animal of each sex had mild focal interstitial nephritis. However, this condition was also seen in a control male and a control female. Unilateral hydronephrosis occurred in another high-dose F2 male. In addition, mild focal tubular hyperplasia was observed in one high-dose male F3 pup but was also diagnosed in two control male pups.
Other effects:
no effects observed
Key result
Dose descriptor:
NOAEL
Effect level:
> 1 000 mg/kg bw/day
Based on:
test mat.
Sex:
male/female
Remarks on result:
other: there were no treatment-related effects observed
Key result
Critical effects observed:
no
Key result
Reproductive effects observed:
no

Reproductive Indices

     1.0 g/kg bw/d  0.2 g/kg bw/d  0.04 g/kg bw/d  0.0A  0.0B
 F0 -> F1  Fertility index (%)  100  90  100  90  90
   Male  95  90  90  75  90
   Female  100  100  100  100  100
 F1 -> F2  Fertility index (%)  100  100  90  90  90
   Male  85  95  85  90  85
   Female  100  100  100  100  94
 F2 -> F3  Fertility index (%)  100  90  100  80  80
   Male  90  75  85  80  70
   Female  100  100  100  100  100

Neonatal body weight at day 21

     1.0 g/kg bw/d  0.2 g/kg bw/d  0.04 g/kg bw/d  0.0A  0.0B
 F1 pups  males  30.6 +/- 4.5  30.9 +/- 4.9  30.7 +/- 6.4  30.6 +/- 3.6  27.9 +/- 4.3
   females  29.0 +/- 4.5  29.2 +/- 4.5  29.5 +/- 4.7  27.9 +/- 3.3  27.0 +/- 3.5
 F2 pups  males  32.8 +/- 3.5  30.9 +/- 5.8  29.3 +/- 4.7  30.0 +/- 4.0  28.8 +/- 4.3
   females  30.8 +/- 3.4  30.2 +/- 4.9  28.8 +/- 3.8  28.5 +/- 3.1  27.5 +/- 3.4
 F3 pups  males  30.2 +/- 4.0  30.9 +/- 4.0  30.9 +/- 4.0  32.0 +/- 3.9  30.2 +/- 4.6
   females  28.6 +/- 3.8  28.2 +/- 3.4  29.7 +/- 4.0  30.1 +/- 3.5  27.7 +/- 3.9
Conclusions:
In conclusion, there were no reproductive effects associated with the inclusion of as much as 1000 mg/kg bw/d of test item in the diet.
Endpoint:
reproductive toxicity, other
Remarks:
based on test type
Type of information:
experimental study
Adequacy of study:
weight of evidence
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
study well documented, meets generally accepted scientific principles, acceptable for assessment
Qualifier:
no guideline followed
Principles of method if other than guideline:
"Fertility Assessment by Continuous Breeding (FACB)" assay.
GLP compliance:
yes
Limit test:
no
Species:
mouse
Strain:
CD-1
Sex:
male/female
Details on test animals or test system and environmental conditions:
TEST ANIMALS
- Age: CD-1 mice were purchased at 6 weeks of age
- Housing: Two animals were housed per cage in polycarbonate shoebox type cages with stainless steel wire bar lids. Cages were rotated (relative placement) at least once a week.
- Diet: Purina certified rodent chow animal diet, ad libitum
- Water: ad libitum

ENVIRONMENTAL CONDITIONS
- Temperature (°C): 20-25°C
- Humidity (%): 20 to 70%
- Photoperiod (hrs dark / hrs light): 14-hour light cycle
Route of administration:
oral: drinking water
Vehicle:
water
Details on exposure:
PREPARATION OF DOSING SOLUTIONS:

WATER PREPARATION
- Rate of preparation of water: fresh preparation once every 2 weeks
- Storage temperature of water: room temperature, under yellow light

VEHICLE
- distilled water
Details on mating procedure:
Animals were randomly paired within each treatment group. Treatment was initiated at 11 weeks of age and was continued for 18 weeks (1 week of premating, 14 weeks of cohabitation, and 3 weeks thereafter). During this period, the following parameters were evaluated: body weight, number of litters produced, number of live pups per litter, minimum number of dead pups per litter, group body weight of live pups (males and females recorded separately), percent of infertile pairs and abnormal pups and a brief description of the deformity, if any.
Analytical verification of doses or concentrations:
yes
Details on analytical verification of doses or concentrations:
Aliquots of various dosage formulations were sent to Midwest Research Institute (MRI), Kansas City, MO, for analysis at week 1 of Task 1 and weeks 1, 5, 11, and 17 of Task 2.
Duration of treatment / exposure:
Task 1: 14 consecutive days
Task 2: 18 weeks (1 week prior to cohabitation, 14 weeks of cohabitation, and 3 weeks thereafter)
Task 4: 10 weeks
Frequency of treatment:
dosed/undosed water ad libitum, daily
Details on study schedule:
Fertility Assessment by Continuous Breeding (FACB). It consists of four related tasks, not all of which are necessarily performed for a given compound. These tasks include Task 1 - dose finding; Task 2 - cohabitation phase; Task 3 - identification of the affected sex and Task 4 - offspring assessment. This test protocol is designed to provide an alternative to multigeneration studies which produce similar comprehensive reproductive data but in a considerably shorter time and at a lower cost.
Task 1 is conducted to determine suitable doses for the continuous breeding phase. The test chemical is administered for 14 consecutive days and them maximum tolerated dose (MTD) is computed. Task 2 is designed to determine the effect of the MTD and two lower dose levels on fertility and reproduction. In this phase, treatment is continued for 18 weeks (1 week prior to cohabitation, 14 weeks of cohabitation, and 3 weeks thereafter). If the fertility is significantly affected, Task 3 is conducted to determine whether the male, female or both sexes are affected. If the overall response in Task 2 is negative, Task 4 is conducted. It is designed to evaluate reproductive performance in the offspring from the final and generally the fifth litter of the control and high dose groups. If the fertility in the first generation offspring is significantly affected, a Task 3 may be performed using these animals to determine the affected sex. At the conclusion of either Task 3 or 4, experimental animals may be necropsied.
During necropsy the liver, brain, pituitary, female reproductive tract (ovaries, oviduct, uterus, and vagina), testes, epididymis, prostate, and seminal vesicles with coagulating glands are weighed and fixed for histopathology. Based on the overall response during Task 3 or 4, vaginal smears are prepared to check the effect on oestrous cycle and sperm studied in detail to evaluate the effect on sperm density, sperm motility, and sperm head morphology.
Dose / conc.:
0 other: % in water (w/v)
Remarks:
dose range-finding, main study
Dose / conc.:
0.25 other: % in water (w/v)
Remarks:
dose range-finding, main study; corresponding to a dose of approx. 410 mg/kg bw
Dose / conc.:
0.5 other: % in water (w/v)
Remarks:
dose range-finding, main study; corresponding to a dose of approx. 840 mg/kg bw
Dose / conc.:
1 other: % in water (w/v)
Remarks:
dose range-finding, main study; corresponding to a dose of approx. 1640 mg/kg bw
Dose / conc.:
2.5 other: % in water (w/v)
Remarks:
dose range-finding
Dose / conc.:
5 other: % in water (w/v)
Remarks:
dose range-finding
No. of animals per sex per dose:
Task 1: 8/8 male/female
Task 2: 20/20 male/female treatment; 40/40 male/female vehicle control
Task 4: 20/20 male/female treatment and control
Control animals:
yes, concurrent vehicle
Details on study design:
Based on the information available in the literature, the following dose levels were selected: 0, 0.25, 0.5, 1.0, 2.5, and 5.0% administered in drinking water.
Parental animals: Observations and examinations:
CAGE SIDE OBSERVATIONS: Yes
- time schedule: twice daily

DETAILED CLINICAL OBSERVATIONS: Yes
- time schedule: twice daily

BODY WEIGHT: Yes
- time schedule for examinations: weekly

WATER CONSUMPTION AND COMPOUND INTAKE: Yes
- time schedule for examintations: weekly
- estimate of substance intake by multiplying water consumption by chemical concentration divided by the sum weight of the pair
Litter observations:
The following parameters were examined in [F1 / F2 / F3] offspring: number and sex of pups, stillbirths, live births, postnatal mortality, presence of gross anomalies, weight gain, physical or behavioural abnormalities

GROSS EXAMINATION OF DEAD PUPS: yes
Postmortem examinations (parental animals):
SACRIFICE
- Male animals: All surviving animals
- Maternal animals: All surviving animals, after litter was weaned

GROSS NECROPSY
- Gross necropsy consisted of external and internal examinations including the cervical, thoracic, and abdominal viscera.
Postmortem examinations (offspring):
SACRIFICE
- the F1 offspring was sacrificed after mating trial was completed
- these animals were subjected to postmortem examinsations as follows:
body weight,
organ weight: liver, brain, pituitary
reproductive tract: females: cranial half of the vagina, cervix, uterus, and ovaries; males: testes, epididymis, seminal vesicles and prostate
Statistics:
yes
Clinical signs:
no effects observed
Description (incidence and severity):
Dose range-finding study: severe respiratory distress; at 2.5 and 5 % treatment groups
Main study: no clinical signy in all treatment groups
Dermal irritation (if dermal study):
not examined
Mortality:
mortality observed, treatment-related
Description (incidence):
Dose range-finding study: treament related deaths in the 2.5 (2/8 males) and 5 % (3/8 males; 1/8 females) groups
Main study: 2 females died in 0.5% treatment group; 1 males and 2 females died in control group
Body weight and weight changes:
no effects observed
Food consumption and compound intake (if feeding study):
not examined
Food efficiency:
not examined
Water consumption and compound intake (if drinking water study):
no effects observed
Ophthalmological findings:
not examined
Haematological findings:
not examined
Clinical biochemistry findings:
not examined
Urinalysis findings:
not examined
Behaviour (functional findings):
not examined
Immunological findings:
not examined
Organ weight findings including organ / body weight ratios:
no effects observed
Histopathological findings: non-neoplastic:
no effects observed
Description (incidence and severity):
Dose range-finding study: dilated renal tubules, tubular nephrosis related to oxalate crystal accumulation at 2.5 and 5 % treatment groups
Main study: moderate number of oxalate crystals in renal tubuli of one animal (0.5% treatment group)
Histopathological findings: neoplastic:
not examined
Other effects:
no effects observed
Reproductive function: oestrous cycle:
not examined
Reproductive function: sperm measures:
not examined
Reproductive performance:
effects observed, treatment-related
Description (incidence and severity):
Main study: statistically significant decrease in number of litters per fertile pair, mean number of live pups, and mean live pup weight in 1 % treatment group. As well as significant reduced average total litter size; average number of male pups per litter; and both male and female pup weights.
Key result
Dose descriptor:
NOEL
Effect level:
0.5 other: percent in drinking water (w/v)
Based on:
test mat.
Sex:
male/female
Basis for effect level:
reproductive performance
Remarks on result:
other: equivalent to approx. 840 mg/kg bw/d
Key result
Critical effects observed:
no
Clinical signs:
no effects observed
Dermal irritation (if dermal study):
not examined
Mortality / viability:
no mortality observed
Body weight and weight changes:
no effects observed
Food consumption and compound intake (if feeding study):
not examined
Food efficiency:
not examined
Water consumption and compound intake (if drinking water study):
no effects observed
Ophthalmological findings:
not examined
Haematological findings:
not examined
Clinical biochemistry findings:
not examined
Urinalysis findings:
not examined
Sexual maturation:
not examined
Organ weight findings including organ / body weight ratios:
not examined
Gross pathological findings:
effects observed, treatment-related
Description (incidence and severity):
1 % treatment group: facial defects, pattern of skeletal defects, affected skull, cleft lip, abnormally shaped or missing sternbrae, fused ribs and abnormally shaped vertebrae, twisting of spine.
Neither the 0.25 nor 0.5% dose groups were significantly affected.
Histopathological findings:
no effects observed
Other effects:
no effects observed
Behaviour (functional findings):
not examined
Developmental immunotoxicity:
not examined
Exposure to ethylene glycol resulted in a small but significant decrease in the number of litters per breeding pair, in the number of live pups per pair and in the live pup weight. A significant number of pups in the 1.0% dose group were born with distinct facial deformities. In the retained litters at this dose, the facial deformities were more obvious with age. These malformed animals also exhibited fused ribs and shortened nasal, parietal, and/or frontal bones of the skull. When pups from the high dose group were raised to adulthood (with continued exposure to ethylene glycol) and mated, they exhibited decreased mating and fertility indices relative to controls handled in the same manner, but there were no effects on litter size, pup weight or sex ratio.
Key result
Dose descriptor:
NOEL
Generation:
F1
Effect level:
0.5 other: percent in drinking water (w/v)
Based on:
test mat.
Sex:
male/female
Basis for effect level:
gross pathology
Remarks on result:
other: equivalent to approx. 840 mg/kg bw/d
Key result
Critical effects observed:
yes
Lowest effective dose / conc.:
1 other: percent in drinking water (w/v); equivalent to approx. 1640 mg/kg bw/d
System:
musculoskeletal system
Organ:
other: facial deformities
Treatment related:
yes
Clinical signs:
no effects observed
Dermal irritation (if dermal study):
not examined
Mortality / viability:
no mortality observed
Body weight and weight changes:
not examined
Food consumption and compound intake (if feeding study):
not examined
Food efficiency:
not examined
Water consumption and compound intake (if drinking water study):
not examined
Ophthalmological findings:
not examined
Haematological findings:
not examined
Clinical biochemistry findings:
not examined
Urinalysis findings:
not examined
Sexual maturation:
not examined
Organ weight findings including organ / body weight ratios:
not examined
Gross pathological findings:
effects observed, treatment-related
Description (incidence and severity):
gross malformations; eight animals with distinct facial deformities in the 1% EG group were not necropsied and later utilized for skeletal examinations.

Skeletal Examination: A series of skeletal deformities were apparent in offspring delivered by the high dose (1% EG) group. Briefly, (a) frontal and nasal passages were significantly shortened and sometimes curved; (b) one or more pairs of ribs were fused; (c) one or more ribs were branched; (d) one or more centra were abnormal; and (e) parietals were smaller than the normal width. None of these abnormalities were noted in the untreated CD-1 mice.
Histopathological findings:
no effects observed
Other effects:
effects observed, treatment-related
Description (incidence and severity):
facial and skeletal defects
Behaviour (functional findings):
not examined
Developmental immunotoxicity:
not examined
Fertility was 80% for the control group compared to 61% to those animals receiving 1% EG, but the difference was not statistically significant. The number of live pups per litter and the live pup weight were lower in the EG-treated group, as they were in task 2, but differences were not statistically significant in either case. Although no clinical signs of EG toxicity were reported in the study, unusual facial features were noticed in some of the offspring of the treated mice but not in the controls. The affected offspring generally had a shorter snout with wide-set eyes compared to the control CD- 1 mice. The examination revealed a pattern of skeletal defects in the treated mice affecting the skull, sternebrae, ribs, and vertebrae in both males and females. The defects included shortened frontal, nasal, and parietal bones; one or more pairs of fused ribs; abnormally shaped or missing sternebrae; abnormally shaped vertebrae; and twisting of the spine. The untreated mice showed no such defects. It was apparent from low-magnification examination of the histological sections that the size and shape of the bones in treated mice differed from the controls. Bones from treated mice were smaller and had altered shapes. However, histologic alterations were not evident when bones from treated mice were examined by light microscopy. Formation of lamellar bone, numbers and activity of osteoblasts and osteoclasts, and cartilage structure were identical in treated and control tissues. Tooth structure was normal in all treated mice. No alterations were seen in nasal turbinates, skeletal muscle, various salivary glands, eyes, or in those portions of the olfactory lobes of the brain which were frequently present in the sections. Deposition of calcium oxalate crystals were not found during careful examination of the microvasculature of bony or adjacent soft tissues. No other treatment-related effects were noted in the EG treated mice.
Key result
Dose descriptor:
NOAEL
Generation:
F2
Effect level:
0.5 other: percent in drinking water (w/v)
Based on:
test mat.
Sex:
male/female
Basis for effect level:
other: skeletal deformities
Remarks on result:
other: equivalent to approx. 840 mg/kg bw/d
Key result
Critical effects observed:
yes
Lowest effective dose / conc.:
1 other: percent in drinking water (w/v); equivalent to approx. 1640 mg/kg bw/d
System:
other: facial deformities
Organ:
other: gross malformations
Treatment related:
yes
Key result
Reproductive effects observed:
yes
Lowest effective dose / conc.:
1 other: percent in drinking water (w/v); equivalent to approx. 1640 mg/kg bw/d
Treatment related:
yes
Relation to other toxic effects:
not specified
Dose response relationship:
not specified

TASK 1:

EG administered in drinking water for 2 weeks at 0.25, 0.5, 1.0, and 2.5% (w/v) dose levels (equivalent to approx. 410, 840, and 1640 mg/kg bw/d) had no significant effect on body weights of both male and female mice. Mice exposed to 5% EG lost weight. The group mean body weights for control male mice at the beginning and the end of Task 1 were 34.4 and 34.6 g, respectively. Corresponding values for the 5% EG group were 34.9 and 31.4 g.

Certain animals in the 2.5 and 5% EG groups were sluggish and appeared to have respiratory problems within one week of treatment. Hair coat changed from normal to rough. During the later part of the first week or the early part of the second week, a number of animals in these two dose groups were lethargic and hunched. A significant number of these mice died. No such symptoms were noted in the control group or in animals in the 0.25, 0.5, and 1% EG groups.

Two of the eight male mice exposed to 2.5% EG died during Task 1. None of the female mice died in this dose group. One female and three male mice died in the 5% EG group. These animals exhibited severe respiratory distress prior to death. Pathologic examination showed distinct lesions in the kidneys and dilated tubule with moderate numbers of oxalate crystals in the lumina. The lungs were moderately congested. The cause of death was diagnosed as tubular nephrosis due to oxalate crystals.

Daily consumption of distilled water by the control mice or dosed water by the animals in the different treatment groups was essentially the same except for the male mice in the 5% EG group (p<0 .001).

 

TASK 2:

EG administered in drinking water at 0.25, 0.5, and 1% (w/v) dose level (equivalent to approx. 410, 840, and 1640 mg/kg bw/d) had no apparent effect on male or female body weights. The group mean body weights of male mice in the control and different treatment groups varied between 36.8 to 37.3 g at the beginning of Task 2. After 18 weeks of treatment, the group mean body weights for animals exposed to 0, 0.25, 0.5, and 1% EG were 40.3, 40.6, 40.2, and 40.3 g, respectively. Body weights for female mice varied considerably during Task 2 and the weight depended on the gestation phase.

Only five (5) mice died during 18 weeks of Task 2; one male and two females in the control group and two females in the 0.5% EG group. The cause of death for the control male mouse was mild myodegeneration of the heart and pulmonary congestion. One of the two control female mice revealed gravid uterus. Only one of the two female mice in the 0.5% EG group was necropsied. Pathologic examination revealed dilated tubule with moderate numbers of oxalate crystals in the lumina. The cause of death was suspected due to tubular nephrosis due to oxalate crystals.

The presence of ethylene glycol at a concentration of 1% or less did not significantly interfere with daily water consumption during Task 2 for both male and female mice.

Reproductive Performance and Fertility
Ethylene glycol administered continuously in drinking water at 0.25, 0 .5, and 1% dose levels had no effect on fertility in CD-1 mice. The fertility index for control and all three dose levels was 100%, i.e. every experimental pair delivered at least one litter. EG treatment at 1% dose level (approx. 1640 mg/kg bw/d) significantly reduced (p <0.05): (1) the average number of litters, 4.45 vs. 4.89 in the control group; (2) the average total litter size; (3) the average number of male pups per litter; and (4) both male and female pup weights. This is regarded as a sequel of developmental toxicity. No significant (p >0.05) differences existed between the 1% EG and the control group with respect to: (1 ) the proportion of live pups; and (2) sex ratio. EG treatment at 0.25 and 0.5% dose levels did not significantly (p >0.05) affect reproductive performace of CD-1 mice with respect to any of the above parameters. Interestingly, a significant number of pups delivered by breeding pairs in the 1% EG group showed distinct facial deformities. Six pups representing three different litters revealed a possible cleft lip/palate.

Gestation Period: EG exposure had no apparent effect on the gestation period. Cumulative days to 1st, 2nd, 3rd, 4th, and 5th litters were 28, 51, 69, 91, and 112 (days of the study), respectively. Corresponding values for the 1% EG group were 30, 55, 75, 96, and 111, respectively. It must be added that these values (days of the study) include 7 days of the premating period.

TASK 4:

The average pup weight at weaning ranged between 14.0 to 18.4 g. Offspring from both the control and 1% EG (approx. 1640 mg/kg bw/d) groups gained weight at essentially the same rate.

At least four male and four female mice among the offspring selected for Task 4 matings showed distinct facial deformities. These mice were later used for detailed skeletal examinations.

There was no mortality among the first generation offspring in the control group. Three offspring died in the 1% EG group; one male and two females. The cause of death was not treatment related.

Water consumption by first generation offspring from the control and 1% EG group was monitored on a weekly basis. There were no apparent difference in the average intake of distilled/dosed water by the control/1% EG group offspring.

Reproductive Performance and Fertility:
Twenty (20) pairs of first generation offspring were randomly selected from the control as well as 1% EG groups to assess their reproductive performance. The breeding pairs were mated until a copulatory plug was detected or for a maximum of seven days. The percent of plug positive/No. cohabited (mating index) for the control and treated pups was 90 and 74, respectively. Fertility was also affected by EG treatment. The percent of No. fertile/No. cohabited (fertility index) in the control and EG treated pups was 80 and 61, respectively. Other reproductive parameters were not significantly different (p>0.05) from the control values, i.e. the number of live pups per litter (males, females, or combined), proportion of pups born alive, and sex ratio.

Gross Necropsy:

No significant differences existed in terms of body weight, reproductive tract, and pituitary weight (p>0.05). Average brain weight of treated animals was lower than the control value (p<0.05). Organ weights were then adjusted for body weight by analysis of covariance. Male offspring body and organ weights for control and the treated animals were essentially the same except for the brain and right cauda. The brain weight was decreased by approximately 8 percent of the control value and right cauda weight by 14 percent. Male organ weights were also adjusted for body weight by analysis of covariance. It must be added here that eight animals with distinct facial deformities in the 1% EG group were not necropsied and later utilized for skeletal examinations.

Skeletal Examination:

A series of skeletal deformities were apparent in offspring delivered by the high dose (1% EG) group. Briefly, (a) frontal and nasal passages were significantly shortened and sometimes curved; (b) one or more pairs of ribs were fused; (c) one or more ribs were branched; (d) one or more centra were abnormal; and (e) parietals were smaller than the normal width. None of these abnormalities were noted in the untreated CD-1 mice.

Additional information

Toxicity to reproduction:

DePass et al. (1986) reported a three-generation reproduction toxicity study. Male and female rats were given orally (feed) daily doses of 40, 200 and 1000 mg/kg bw/d. No reproductive effects associated with the inclusion of as much as 1000 mg/kg bw/d of EG in the diet were found. The NOAEL for parental animals and for offsprings was found to be > 1000 mg/kg bw/d.

NTP (1984) reported a FACB study and found a NOEL of 840 mg/kg bw/d for parental and F1 male and female mice. Ethylene glycol was administered in drinking water in concentrations of approx. 410, 840 and 1640 mg/kg bw/d. Exposure to ethylene glycol resulted in a small but significant decrease in the number of litters per breeding pair, in the number of live pups per pair and in the live pup weight. A significant number of pups in the 1.0% dose group (1640 mg/kg bw/d) were born with distinct facial deformities. In the retained litters at this dose, the facial deformities were more obvious with age. These malformed animals also exhibited fused ribs and shortened nasal, parietal, and/or frontal bones of the skull. When pups from the high dose group were raised to adulthood (with continued exposure to ethylene glycol) and mated, they exhibited decreased mating and fertility indices relative to controls handled in the same manner, but there were no effects on litter size, pup weight or sex ratio.

Classification concerning toxicity to reproduction is not warranted.

Effects on developmental toxicity

Description of key information

EG has been evaluated for effects upon pre-natal development following exposure during the period of organogenesis in both rodent (mouse, rat) and non-rodent (rabbit) species.


Distinct differences exist between rodents and non-rodents in their susceptibility to teratogenicity induced by ethylene glycol. Oral administration of EG to pregnant mice and rats during organogenesis induced malformations at dose levels of 500 mg/kg bw/day and above. In contrast, administration to pregnant rabbits at doses up to 2000 mg/kg bw/day had no effect upon development, while  causing substantial (42%) maternal mortality.


Subsequent investigations, both in vivo and in vitro, have shown that the developmental toxicity of EG in rats is related to the accumulation of glycolic acid (GA) in the embryo. The embryotoxicity of GA, both in vivo and in vitro, is exacerbated under acidic conditions, and species differences are related to differences in disposition of GA. When EG was administered to rats and rabbits at a dose of 1000 mg/kg, oral gavage, the concentration of GA in the rat embryo compared to maternal blood was significantly higher (embryo/blood concentration ratio: 1.54) whereas this was not the case in the rabbit (embryo/blood  concentration ratio: 0.31).


Recent investigations demonstrated that GA uptake into the rat embryo occurs predominantly by a specific, pH-dependent, active uptake transporter protein, consistent with the proton-linked monocarboxylate transporter family (MCT); passive disposition is a minor component. Two isoforms of the MCT exist in the placenta, a high-affinity isoform (MCT1) and a low affinity isoform (MCT4). The polarity of these isoforms in the mouse and rat placenta syncytio-trophoblast is opposite to that in the rabbit and human placenta. In the rodent, MCT1 is located on the side of maternal blood, while MCT4 is located on the side of embryonic blood; in rabbits and humans MCT1 can be found on the side of embryonic blood while MCT4 is located on the side of maternal blood.


Given that GA disposition between maternal blood and the embryo is driven by the polarity of MCT1 and MCT4 in the placenta; that GA is sequestered in the rat embryo and not the rabbit embryo; and that the rabbit and human placenta show similar polarity, opposite to that of the rat, it is concluded that the rabbit is the appropriate species for human hazard characterisation for ethylene glycol. As EG is not a developmental toxicant in the rabbit, classification for developmental toxicity is not warranted.


 

Additional information

Developmental toxicity of ethylene glycol


Ethylene glycol has been evaluated for effects on development following exposure during the period of organogenesis in both rodent (mouse, rat) and non-rodent (rabbit) species. EG caused malformations after oral administration in mice and rats, and specifically gavage administration in rats, but not after dermal or inhalation exposure when oral ingestion was precluded. Effects on development were absent in rabbits.


Mouse


Administration of EG (0, 750, 1500, or 3000 mg/kg/day; via oral gavage) to pregnant CD‑1 mice during organogenesis caused increased incidences of skeletal malformations at all doses, and increased incidences of external and visceral malformations at the high-dose level. External malformations were predominantly neural tube defects (meningoencephalocele; exencephaly) and facial malformations (cleft lip and palate; facial cleft), while skeletal malformations involved the ribs, arches, and centra. Foetal body weight was reduced at all doses, but live litter size was only reduced at the high-dose level. This was observed in the presence of dose-related decreases in maternal weight gain and gravid uterine weight at the mid and high dose (Price et al., 1985). A follow-up study was performed to define the NOAEL for developmental effects. Pregnant CD-1 mice were administered EG (0, 50, 150, 500, or 1500 mg/kg/day; oral gavage) resulting in skeletal malformations and variations and reduced foetal weight at the highest dose only, and an increased incidence of one skeletal variation, namely an extra 14th rib on the first lumbar arch (bilateral), at 500 mg/kg/day. The authors identified a NOAEL of 150 mg/kg/day (Neeper-Bradley et al., 1995).


Whole body exposure of CD-1 mice to EG aerosol (0, 150, 1000, 2500 mg/m³ 6 hr/day) during organogenesis was embryotoxic, eliciting malformations and reduced foetal body weight at the mid- and high-exposure concentrations (Tyl et al., 1995a). The potential for oral exposure to EG by grooming of the fur was investigated by exposing CD-1 mice to EG aerosol either nose-only (0, 500, 1000, 2500 mg/m³ 6 hr/day) or whole body (0, 2100 mg/m³ 6 hr/day). Foetal body weight was reduced at the highest concentration by both exposure regimens leading to a NOAEL of 1000 mg/m³ for the nose-only group. Several skeletal malformations were found in the whole body exposure group compared to controls; in the high concentration nose-only group skeletal malformation was limited to fused ribs (Tyl et al., 1995b). Furthermore, dermal exposure of CD-1 mice to EG (0, 404, 1677, or 3549 mg/kg/day, the latter equivalent to 100% undiluted EG) during gestation days (GD) 6-15 was without effect on the fetuses (Tyl et al., 1995c).


In a breeding study in which CD-1 mice were continuously administered 1% EG in the drinking water (equivalent to approx. 1640 mg/kg/day) and repeatedly bred, adult offspring of EG-treated dams had unusual facial features, characterised by shortened snout and wide-set eyes. Skeletal examination revealed defects of the skull, sternebrae, ribs, and vertebrae. Gross malformations, such as cleft lip, were also noted in neonates from both the F0 and F1 generations. No findings were noted in animals that received 0.25% or 0.5% EG in the drinking water (Lamb et al., 1985).


Rat


Administration of EG (0, 1250, 2500, or 5000 mg/kg/day; oral gavage) to pregnant CD rats during organogenesis resulted in increased incidences of visceral malformations at all doses, increases in the incidences of skeletal malformations at the mid- and high-dose levels, and increases in the incidence of external malformations at the high-dose level. As for mice, external malformations were predominantly neural tube defects (meningoencephalocele; exencephaly; meningocele) and facial malformations (cleft lip and palate; anophthalmia; micrognathia), while skeletal malformations involved the ribs, arches, and centra, and visceral anomalies were predominantly of the great vessels. Foetal body weight and live litter size were also reduced at the mid- and high-dose levels (Price et al., 1985). In a subsequent study to define the NOAEL for developmental effects, pregnant CD rats were administered EG (0, 150, 500, 1000, or 2500 mg/kg/day; oral gavage) resulting in external and skeletal malformations at the highest dose. Skeletal malformations and variations (reduced ossification) were noted at 1000 mg/kg/day, while the only statistically significant finding at 500 mg/kg/day was poor ossification of the supraoccipital bone (Neeper-Bradley et al., 1995).


Maronpot et al. (1983) administered EG in the diet to F344 rats during GD 6-15 at doses of up to 1000 mg/kg/day. This was without effect on the fetuses, indicating that the effects of EG are attenuated by non-bolus dosing regimens. This conclusion was supported by the results of a study in which the same overall dose of EG (0, 1000, 2000 mg/kg/day) was administered during organogenesis as a bolus dose or by subcutaneous infusion. Bolus dosing resulted in skeletal malformations and variations at both dose levels, whereas infusion did not (Carney et al., 2011).


Whole body exposure of CD rats to EG aerosol (0, 150, 1000, 2500 mg/m³ 6 hr/day) during organogenesis was without effect, other than an increased incidence of skeletal variations at 1000 and 2500 mg/m³ (Tyl et al., 1995a).


Rabbit


Administration of EG to New Zealand White rabbits (0, 100, 500, 1000, or 2000 mg/kg/day; oral gavage) during organogenesis caused mortality in 42% of the does in the highest-dose group, as a result of renal failure. In the offspring, no effects on survival, foetal weight, or on the incidences of external, visceral, or skeletal malformations or variations were observed at any dose level, including the foetuses of surviving does treated with 2000 mg/kg/day EG (Tyl et al., 1993).


 


 


Identifying the proximate toxicant


Following absorption, EG is rapidly metabolised by oxidative pathways. At low doses, metabolism is extensive, and the major elimination products are exhaled carbon dioxide and the urinary parent ethylene glycol and metabolites. However, in rats, the oxidation of glycolic acid (GA) to glyoxylic acid becomes saturated at doses in the range 125–500 mg/kg, resulting in accumulation of GA in the blood and increased excretion in the urine, with no significant difference between pregnant and non-pregnant female CD rats (Corley et al., 2005; Pottenger et al., 2001). The accumulation of GA in the blood is associated with metabolic acidosis.


EG is of low inherent toxicity to the rat embryo. The NOAEC for morphological changes in rat embryos exposed to EG in vitro (whole embryo culture, WEC) is in excess of 50 mM (Carney et al., 1996; Grafton & Hansen, 1987; Klug et al, 2001). In contrast, Cmax for EG in the  maternal blood following administration of EG (1000 mg/kg/day; oral gavage) to pregnant rats is around 12–14mM (Carney et al., 2011; Triskelion, 2018a). Klug et al. (2001) evaluated the toxicity of EG metabolites as well as the parent EG in rat WEC, finding that all metabolites were more potent than EG in inducing anomalies and growth retardation. Comparison of the NOAEC/LOAEC with measured levels of metabolites in blood following administration of EG to female rats (Frantz et al., 1996a, 1996b; Pottenger et al., 2001) suggested that GA might be the proximate toxicant (Carney et al., 1996; Klug et al., 2001).  The NOAEC and LOAEC for GA were established in two separate rat WEC studies, and were found to be 1–2.5 mM and 3–12.5 mM, respectively (Carney et al., 1996; Klug et al., 2001). In comparison, Cmax for GA in the maternal blood following administration of EG (1000 mg/kg/day; oral gavage) is 2.8–4.8mM (Carney et al., 2011; Triskelion, 2018a).


Table 1. Calculated kinetic parameters for EG and GA in maternal and conceptus matrices of the rat at the commencement of placentation (Triskelion, 2018a).



































































  

Rat GD11


 
  

Blood



EEF



Embryo


 

Cmax (μg/mL)



EG



11.6



12.2



6.3


 
 

GA



2.9



4.5



4.2


 

AUC0–last (h.μg/mL)



EG



52.0



56.8



37.3


 
 

GA



28.3



47.8



45.8


 

AUC0–∞ (h.μg/mL)



EG



52.3



57.1



37.5


 
 

GA



28.3



47.9



45.9


 

When administered to pregnant CD rats, GA does induce a spectrum of skeletal malformations similar to that of EG (Carney et al., 1999; Munley et al., 1999). In order to distinguish the roles of glycolate anion and acidosis in the aetiology of EG-induced developmental toxicity, EG (2500 mg/kg; oral gavage), GA (650 mg/kg; oral gavage), or sodium glycolate (833 mg/kg; s.c.; pH 7.4) were administered to time-mated pregnant CD rats to deliver high, equimolar blood glycolate concentrations in the presence (EG and GA administration) or absence (sodium glycolate administration) of acidosis (Carney et al., 1999). Compared to the vehicle group, EG and GA induced a range of skeletal malformations and variations. Sodium glycolate only induced fetal alterations with reduced incidence and severity compared to EG and GA. EG also induced craniofacial defects that were not observed in GA-treated animals, but this was attributed to the almost threefold higher integrated systemic dose (area under the curve) of blood glycolate following EG administration compared to administration of GA itself.


Carney et al. (1996) further evaluated the roles of glycolate and acidosis in vitro, using rat WEC. Embryos (GD 10.5) were exposed for 46 hours to GA (12.5 mM; pH 6.7), sodium glycolate (12.5 mM; pH 7.4), or control medium (pH 6.7 or pH 7.4), after which time they were evaluated for growth and morphological parameters. In comparison to the pH 7.4 control, acidification alone (i.e. pH 6.7) caused only a mild impairment of embryo growth, determined by protein content and head length, but no other measures of growth of morphogenesis were significantly altered. Sodium glycolate (pH 7.4) caused a greater reduction of embryonic growth over a greater number of parameters, while also reducing the overall morphological score and increasing dysmorphogenesis, primarily craniofacial lesions. GA (pH 6.7) induced an even more severe impairment of growth and morphological development.


Disposition of GA into the rat embryo


With the roles of GA and acidosis in the aetiology of EG-induced developmental toxicity established, it is also important to evaluate the disposition of GA into the embryo.


Following administration of EG (1000 mg/kg; oral gavage) to time-mated pregnant CD rats on GD 11, Cmax for GD in maternal blood was 4.1 mM, while in the embryos and surrounding EEF it was 6.3 mM and 6.7 mM, respectively. Subcutaneous infusion of the same dose of EG resulted in peak GA concentrations of 0.07 mM, 0.13 mM, and 0.13 mM, respectively (Carney et al., 2011). Similar findings were made when EG was administered by oral gavage on consecutive days during GD 6–11: blood, 2.9mM; embryo, 4.3mM; EEF, 4.5mM (Triskelion, 2018a). This means that, following a teratogenic dose of EG, the maximum concentration of GA in the rat embryo exceeds that in maternal blood by 50%.


The disposition of GA into rat WEC was established by Ellis-Hutchings et al. (2014) as being predominantly mediated by a specific, energy- and pH-dependent carrier. Uptake of 1mM 14C-labelled GA from the culture medium into the embryo and yolk sac occurred against a concentration gradient over a six-hour incubation period, although it reached a plateau in the embryo within two hours, such that the concentration of GA in the embryo was threefold higher than in the culture medium. The uptake of GA was substantially and significantly reduced by incubation at low temperature (4°C vs. 37°C), thereby establishing energy-dependence, and enhanced under acidic conditions (pH 7.1 vs. pH 7.8), thereby establishing proton-dependence. Uptake of 1 mM and 12 mM GA was inhibited by up to 80% by co-incubation with d-lactic acid (12–120 mM), suggesting that a substantial component of the uptake is mediated by a carrier protein with cross-affinity for small carboxylic acids. These characteristics have been widely described for the monocarboxylate transporter (MCT) protein family, present in the placenta of humans, rabbits and rodents.


Placental MCT determines species differences in GA disposition


The MCT is believed to mediate the transfer of lactic acid between maternal and embryofoetal compartments, although there are species differences in the characteristics of transplacental transfer. Nagai et al. (2010) examined the expression and subcellular localisation of MCT isoforms in the ddY mouse placenta during gestation. Although six isoforms were initially studied, only two of them, MCT1 and MCT4, are known to be involved in the transport of monocarboxylates; the others were not extensively evaluated and are not discussed here. mRNA for MCT1 and MCT4 was expressed at GD 11, 14, 17, and 18, but not at GD 9, prior to the development of a functional placenta. Expression was greatest on GD 11 and 14 and reduced towards term. Immunohistochemical staining for the two protein isoforms showed that MCT1, which is a high-affinity transporter, is localised in the apical membrane of the synctytio-trophoblast facing the maternal blood; MCT4, which is a low-affinity transporter, is localised in the basal plasma membrane close to the embryofoetal blood. A similar partitioning of MCT1 and MCT4 has been identified in the CD rat placenta (Moore et al., 2016).


In contrast to the polarity of MCT1 and MCT4 in the rodent placenta, these isoforms show the opposite polarity in placenta from non-rodent species (Settle et al., 2004; Nagai et al., 2010). Settle et al. (2004) described the localisation of MCT1 and MCT4 in term human placenta. MCT1 was localised predominantly to the basal membrane (embryofoetal side of the syncytio-trophoblast) while MCT4 was found on both plasma membranes with more intense staining on the microvillous membrane (maternal side). The localisation of these two isoforms has also been evaluated in the New Zealand White rabbit placenta. Again, MCT1 was localised on the side apposed to the embryofoetal circulation while MCT4 was localised to the side of the maternal blood (Moore et al., 2016). Semi-quantitative evaluation of MCT, through intensity of staining under light microscopy using isoform-specific antibodies, indicates that the level of MCT1 decreased in both rat and rabbit placenta during placentation, while that of MCT4 remains essentially unchanged (Moore et al., 2016).


From these physiological differences, a model for GA disposition in rodents and non-rodents can be proposed. In this model, GA is sequestered from maternal blood and transported across the syncytio-trophoblast to embryonic blood. In the rodent, the high affinity MCT1 apposed to the maternal blood effectively transports GA to the embryo, while MCT4 does not sequester GA from the embryonic blood as effectively. The result is a net influx of GA from the maternal blood to the conceptus. In the non-rodent, the opposite is the case. It is proposed that this qualitative difference in MCT polarity explains the quantitative differences in maternal-embryonal disposition and response to GA exposure. Because the polarity of the MCT in the rabbit placenta is the same as that in humans, the rabbit is the appropriate model for evaluating the intrinsic hazard relevant to humans.


Since the expression of MCT1 appears to diminish during placentation (Moore et al., 2016), and if GA disposition is mediated by MCT, then disposition should be attenuated with age of gestation. To address this question, groups of timed-pregnant rats and rabbits were administered EG (1000 mg/kg/day; oral gavage) daily from GD 6 until either the commencement or completion of placentation. After the final dose was administered, subgroups of animals were sacrificed at selected timepoints over the following 24 hours, and GA was analysed in maternal blood and conceptus compartments (extraembryonic fluid, embryo).


At GD11 in the rat, the embryo:blood ratio was significantly above unity at 3 and 6 hours post-dose, but was significantly below unity at all time points on GD16. The ratio was significantly below unity at all time points in the rabbit, except the 6-hour timepoint on GD10, at which it was not significantly different to unity. These data support the hypothesis that GA disposition into the embryo is predominantly influenced by MCT; however passive transport remains a possibility and is consistent with pioneering investigations of GA disposition.


Embryonic acidosis as mode of action


An hypothesised mode of action has been investigated in cultured rat embryos. The sequestration of GA into the embryo via MCT would result in acidosis, since the MCT is a proton-dependent symporter. In general, acidosis would, in turn, inhibit hERG cardiac K+ channels (Van Slyke et al., 2012), resulting in bradycardia, reperfusion hypoxia, and production of reactive oxygen species (Danielsson et al., 2007); a common teratogenic mechanism across rat, rabbit, and human (Danielsson et al., 2013).


Rat embryos, harvested at GD 11 and GD 13, were exposed to graded GA concentrations under graded acidic conditions, and the heart rate was measured prior to, and after one hour of incubation. In vitro exposure to acidic pH alone, or to GA under neutral pH conditions had little effect on heart rate. Combined exposure, however, caused a significant bradycardia and increased incidences of arrhythmic or dead embryos at high exposure concentrations. Interestingly, the effects were not mediated by GA specifically, since similar bradycardia was also induced following incubation with l-lactic acid under the same conditions. Thus, the critical component in the aetiology of bradycardia is the transport of protons, culminating in embryonic acidosis (Ritchie et al., 2017).


These findings lend support to the hypothesis for rodent embryonic acidosis underlying the mode of action for EG.


Conclusion


The proposed mode of action underlying EG teratogenicity, and the similarities and differences between rodent, rabbit, and humans, are outlined as follows.



  • EG is metabolised to GA (demonstrated across all species)

  • GA is a substrate for the MCT, a symporter that binds GA and a proton and transports them down respective concentration gradients (relevant to all species);

  • GA partitions preferentially into the conceptus from the maternal blood (demonstrated in the rat, the opposite is the case in the rabbit)

  • The partitioning of GA between maternal blood and the embryo is determined by the ‘polarity’ of MCT1 and MCT4 across the placenta (high affinity MCT1 on the maternal side and low affinity MCT4 on the embryonic side of the rodent trophoblast, low affinity MCT4 on the maternal side and high affinity MCT1 on the embryonic side of the rabbit and human trophoblast)

  • Sequestration of GA into the rodent embryo resulted in acidosis of the rodent embryonic tissue (relevant to rat and mouse, not to rabbit and human)

  • Embryonic acidosis in general causes K+ channel blockade, bradycardia, reperfusion hypoxia, and formation of reactive oxygen species (relevant across species)


Therefore, the reason that the rabbit does not respond in the same way as the rodent does, is due to the lack of sequestration into the embryo, and this is in turn related to the ‘polarity’ of the MCT isoforms across the placenta. Since the ‘polarity’ of MCT1 and MCT4 in the human placenta is the same as that in the rabbit, the rabbit is the appropriate species for hazard characterisation and risk assessment.


 

Justification for classification or non-classification

Classification, Labelling, and Packaging Regulation (EC) No 1272/2008


 


Ethylene glycol (EG; CAS RN 107-21-1) is not classified in the EU as a reproductive toxicant.


According to Regulation (EC) No 1272/2008 classification of reproductive toxicants encompasses effects upon fertility and on the developing offspring (‘developmental toxicity’). EG does not affect fertility in animals, but developmental toxicity has been noted in some species at high doses.




Distinct differences exist between rodents and non-rodents in their susceptibility to teratogenicity induced by ethylene glycol. Oral administration of EG to pregnant mice and rats during organogenesis induced malformations at dose levels of 500 mg/kg bw/day and above. In contrast, administration to pregnant rabbits at doses up to 2000 mg/kg bw/day had no effect upon development, while  causing substantial (42%) maternal mortality.


Subsequent investigations, both in vivo and in vitro, have shown that the developmental toxicity of EG in rats is related to the accumulation of glycolic acid (GA) in the embryo. The embryotoxicity of GA, both in vivo and in vitro, is exacerbated under acidic conditions, and species differences are related to differences in disposition of GA. When EG was administered to rats and rabbits at a dose of 1000 mg/kg, oral gavage, the concentration of GA in the rat embryo compared to maternal blood was significantly higher (embryo/blood concentration ratio: 1.54) whereas this was not the case in the rabbit (embryo/blood  concentration ratio: 0.31).


Recent investigations demonstrated that GA uptake into the rat embryo occurs predominantly by a specific, pH-dependent, active uptake transporter protein, consistent with the proton-linked monocarboxylate transporter family (MCT); passive disposition is a minor component. Two isoforms of the MCT exist in the placenta, a high-affinity isoform (MCT1) and a low affinity isoform (MCT4). The polarity of these isoforms in the mouse and rat placenta syncytio-trophoblast is opposite to that in the rabbit and human placenta. In the rodent, MCT1 is located on the side of maternal blood, while MCT4 is located on the side of embryonic blood; in rabbits and humans MCT1 can be found on the side of embryonic blood while MCT4 is located on the side of maternal blood.


Given that GA disposition between maternal blood and the embryo is driven by the polarity of MCT1 and MCT4 in the placenta; that GA is sequestered in the rat embryo and not the rabbit embryo; and that the rabbit and human placenta show similar polarity, opposite to that of the rat, it is concluded that the rabbit is the appropriate species for human hazard characterisation for ethylene glycol. As EG is not a developmental toxicant in the rabbit, classification for developmental toxicity is not warranted under Regulation (EC) No 1272/2008.



Additional information