Ethyl 4-hydroxybenzoate

Specific investigations: other studies

Link to relevant study record(s)

Reference

Endpoint:

endocrine system modulation

Type of information:

other: Publicataion

Adequacy of study:

weight of evidence

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

data from handbook or collection of data

Details on results:

Fayyay et al evaluated all available and reliable in vivo data (especially level 4 and 5 of accordig to OECD CF for Testing and Assessment of Endocrine
Disrupting Properties (OECD 2012)) to address human health concerns with respect to 90-day repeated-dose toxicity, developmental and reproductive toxicity (DART), and endocrine disrupting potential for methyl, ethyl and propyl paraben.
For methyl paraben and propyl paraben, all higher-tier studies of relevance for the determination of repeated-dose toxicity, DART and endocrine disrupting potential have been requested under REACH. For both methyl paraben and propyl paraben, the NOAEL with regard to repeated-dose toxicity and DART was set at 1000 mg/kg bw/day.For ethyl paraben, a NOAEL of 1000 mg/kg bw/day for repeated-dose toxicity and DART was estimated by interpolation from methyl paraben and propyl paraben,
i.e. the two category members on both sides of the target substance (ECHA 2008; OECD 2014).

Conclusions:

This research article presents the outcomes of higher-tier studies using Wistar rats requested for methyl paraben and propyl paraben under the European Union chemicals legislation. No adverse effects were recorded up to 1000 mg/kg body weight/day, which was set as no-observed adverse effect level (NOAEL) for repeated-dose toxicity and developmental and reproductive toxicity (DART). These findings for methyl paraben and propyl paraben were then used to interpolate the corresponding hazard profile for ethyl paraben, further considering available data for butyl paraben. The underlying read-across hypothesis (all shorter-chained linear n-alkyl parabens are transformed to a common compound and can be considered a ‘category’ on account of their very high structural similarity) was confirmed by in vivo toxicokinetics screening studies. All four parabens were taken up systemically and eliminated within one hour after oral gavage administration to Wistar rats, and they were all metabolised to p-hydroxybenzoic acid, which was also cleared rapidly within 4-8 hours. Accordingly, for ethyl paraben, the NOAEL for repeated-dose toxicity and DART was interpolated to be 1000 mg/kg body weight/day. Finally, all findings were further evaluated to address concerns expressed in the literature that parabens might be endocrine disruptors. The higher-tier studies do not provide any indication for any endocrine disrupting property. This is the first time that a comprehensive dataset from higher-tier studies, conducted following internationally agreed test protocols, has become available for shorter-chained linear n-alkyl parabens. Consistently, the dataset shows that these parabens are devoid of repeated-dose toxicity, DART and endocrine disrupting properties.

Executive summary:

As discussed by Barlow et al. (2015), it is an essential aspect of the EU REACH Regulation to use the collated data for hazard-based substance classification and labelling following the provisions ofRegulation(EC) No 1272/2008 onClassification, Labelling and Packaging of substances and mixtures(EP and Council 2008). For substances with endocrine disrupting properties, EU legislation specifically mandates regulation via a hazard-based approach (Brescia 2020). In line with these provisions, the present research article has focussed on hazard identification of methyl paraben, ethyl paraben and propyl paraben (while also considering the further category member butyl paraben and briefly referring to the sodium salts of methyl paraben, ethyl paraben and propyl paraben).For methyl paraben and propyl paraben, all higher-tier studies of relevance for the determination of repeated-dose toxicity, DART and endocrine disrupting potential have been requested under REACH, and the findings from these studies have been presented herein.For bothmethyl paraben and propyl paraben, the NOAEL with regard to repeated-dose toxicity and DART was set at 1000 mg/kg bw/day.

For ethyl paraben, a NOAEL of 1000 mg/kg bw/day for repeated-dose toxicity and DART was estimated by interpolation from methyl paraben and propyl paraben, i.e. the two category members on both sides of the target substance (ECHA 2008; OECD 2014).The chemical category of shorter-chained linearn-alkyl parabens is founded on their high structural similarity (the only difference between these parabens istheir increasing chain length) and, importantly, on their common metabolic pathway. All categorymembers exhibit similarphysico-chemical properties and similar acute toxicity, local toxicity and genotoxicity potential.The rattoxicokinetics screening studies consistently showed that due to the very rapid elimination of these parabens and their major metabolite (that is further non-toxic), systemic target organs and tissues are not exposed to these compounds for a sufficiently long period of time for effects to evolve.Due to the consistency of the findings, the interpolation of a NOAEL of 1000 mg/kg bw/day for repeated-dose toxicity and DART for ethyl paraben is assessed as acceptable with high confidence. Performing the corresponding higher-tier studies, that encompass large numbers of animals (Table 4), for the hazard assessment of ethyl paraben would breach the 3Rs principle implemented in Directive 2010/63/EU (EP and Council 2010) and Article 25(1) of the REACH Regulation (EP and Council 2006) thatrequires that testing on vertebrate animals shall be undertaken only as a last resort.

The toxicokinetic screening studies also indicate that butyl paraben is not systemically bioavailable for a sufficiently long period of time to elicit DART or endocrine disrupting effects. This supported by the recent findings from Hubbard et al. (2020). Finally, for the sodium salts of methyl paraben, ethyl paraben and propyl paraben, read-across of the NOAEL of 1000 mg/kg bw/day for repeated-dose toxicity and DART also appears justifiable on account of their very high similarities (T_c0.98) as compared to the respective paraben.

To the best of the authors’ knowledge, this is the first time that a comprehensive dataset from higher-tier studiesconducted following internationally agreed OECD TG test protocols and in full compliance with the GLP principles has become available for linearn-alkyl parabens. The data also enable a comprehensive evaluation of the endocrine disrupting potential of these parabens according to all five levels of theOECD CFfor Testing and Assessment of Endocrine Disrupting Properties(OECD 2012).The higher-tier (OECD CF Level 4 and 5) studies do not provide any indication that any endocrine activity (if it were presentin vivo) would be sufficiently pronounced to overwhelm the physiological adaptive capacities of endocrine systems leading to adversity as indispensable prerequisite for endocrine disruption (WHO IPCS 2002).

As compared to the hazard-based approach pursued under the REACH Regulationfor the assessment of endocrine disruptors, a risk-based approach is applied inother jurisdictions, such as USA, Canada, Australia and Japan (Brescia 2020).Risk-based approaches include exposure assessment in addition to hazard assessment to derive a conclusion on the safety of the respective substance. As highlighted by Brescia (2020), the application of a risk-based approachfor the assessment of endocrine disruptors is scientifically justified since the available scientific evidence indicates that endocrine disruption exhibits a concentration threshold below which no effects will occur (Brescia 2020). Accordingly, the setting of concentration limits provides an additional safeguard to ensure consumer protection.For example, for all four shorter-chained linearn-alkyl parabens, EFSA (2004), SCCS (2013), and EMA (2015) have concluded that they are safe to the consumer when used in food, cosmetics and human medicinal products for oral use, respectively, provided that specific concentrations are not exceeded. This estimation has been re-confirmed by the findings from the present research article.

Description of key information

Ethylparaben does not show any ED properties (based on level 4 and 5 of OECD CF for Testing and Assessment of Endocrine Disrupting Properties (OECD 2012)).

Additional information

Ethylparaben was investigated for estrogenic activity in several in vitro and in vivo studies. In estrogen receptor (ER) competitive binding assays using the human MCF-7 breast cancer cell line and isolated ER from rat uteri, ethylparaben showed a weak ER binding affinity with IC50 values of 46 - 270 µM (Okubo et al., 2001; Byford et al., 2002; Blair et al., 2000). In comparison, for 17b-estradiol an IC50 value of 0.0009 µM was determined (Blair et al., 2000). 

For cell proliferation in MCF-7 cells, an EC50 value of 3.2 µM ethylparaben was determined (Okubo et al., 2001). In comparison, for 17b-estradiol, an EC50 value of 0.0000016 µM was determined for cell proliferation.

Uterotrophic assays according to OECD 440 were conducted in rats and mice. Immature rats were treated subcutaneously with 6-180 mg/kg bw/d ethylparaben for 3 consecutive days (Lemini et al., 2003). Significantly increased uterine weights (wet and dry) were noted only at the highest dose. In the same study, immature and adult ovariectomised mice were treated subcutaneously with up to 180 mg/kg bw/d for 3 consecutive days. Uterine weights in immature and ovariectomised mice increased dose-dependently from 0.6 to 180 mg/kg bw/d and 6 to 180 mg/kg bw/d, respectively. Statistically significant uterus weights were observed at 6, 60 and 180 mg/kg bw/d for immature mice and at 18, 60 and 180 mg/kg bw/d for ovariectomised mice. Histopathological examination of uterine horns of treated ovariectomised mice resulted in significantly increased luminal epithelium height, glandular epithelium height and myometrium width at doses of 60 and 180 mg/kg bw/d. No lower doses were tested for histopathological abnormalities.

In a further uterotrophic assayaccording to OECD 440, mice were orally treated with 1000 mg/kg bw/d ethylparaben for 3 consecutive days (Hossaini et al., 2000). Ethylparaben did not influence the uterus wet weight and the uterine weight to body weight ratio after repeated oral doses of 1000 mg/kg bw/d.

In a further study in rats, pregnant dams were treated between gestation day (GD) 7 and 21 with 400 mg/kg bw/d ethylparaben (Taxvig et al., 2008). After caesarian section on GD 21, no effects were observed on the fetal body weight and the anogenital distance. No statistically and biologically significant effect was noted on the plasma levels of progesterone, 17α-hydroxyprogesterone, T3 and T4 in the dams on GD 21. Moreover, no significant effect on fetal hormones (T4, progesterone, testosterone, cortisol) was observed in plasma, testes and adrenal glands. Ethylparaben did also not influence the level of estradiol that was analyzed in fetal ovaries on GD21. Testes, adrenals and ovaries from fetuses as well as thyroids from dams were histologically similar to controls. In summary, no effects of ethylparaben were observed on vaginal opening day, estrous cycle, body and organ weights of peripubertal rats. In this study, a dose-dependent reduction in estradiol level and an increase in prolactin level were noticed in serum of the treatment groups. After treatment of pregnant rats, no effects were noticed on the fetal body weight and testes, adrenal and ovary weights, the anogenital distance and hormones in plasma, testes, ovaries and adrenal glands as well as plasma hormone level of dams. In an uterotrophic assay, ethylparaben did not influence the uterus wet weight and the uterine weight to body weight ratio after repeated oral doses of 1000 mg/kg bw/d. However, in further uterotrophic assays in rats and mice, ethylparaben was shown to increase the uterus weights significantlyafter subcutaneous injection. In several ER competitive binding assays, ethylparaben was shown to have a weak binding affinity. However, as disscussed by Fayyaz et al. It has to be considered that across the entirety of linear n-alkyl parabens,a U-shaped association between chain length and in vitro interaction with oestrogen receptors is observable: amongst 12 parabens with linear n-alkyl chains ranging in length from C1 to C12, heptyl paraben (C14H20O3, i.e. C7 - alkyl chain) and pentyl paraben (C12H16O3, i.e. C5 -alkyl chain) showed the highest potency in activating human oestrogen receptors α and β; at 10–7 M and 10–8 M, respectively (Watanabeet al. 2013). The potency of oestrogen receptor activation decreased in a stepwise manner (and the lowest concentrationinducing receptor effects increased) as the alkyl chain was shortened to C1(methyl paraben) or lengthened to C12 (dodecyl paraben) (Watanabe et al. 2013). Importantly, all receptor effects (also those of the most active heptyl paraben and pentyl paraben) only occurred at concentrations that were many orders of magnitude higher than that of the natural oestrogen 17β-estradiol (2.5 × 10–12) (Watanabe et al. 2013), i.e. the commonly used positive control in in vitro oestrogen receptor transactivation studies (Soni et al. 2005;Watanabe et al. 2013; US EPA 2015).

Fayyay et al evaluated all  available and reliable  in vivo data (especially level 4 and 5 of  accordig to OECD CF for Testing and Assessment of Endocrine Disrupting Properties (OECD 2012)) to  address human health concerns with respect to 90-day repeated-dose toxicity, developmental and reproductive toxicity (DART), and endocrine disrupting potential for methyl, ethyl and propyl paraben. For methyl paraben and propyl paraben, all higher-tier studies of relevance for the determination of repeated-dose toxicity, DART and endocrine disrupting potential have been requested under REACH. For both methyl paraben and propyl paraben, the NOAEL with regard to repeated-dose toxicity and DART was set at 1000 mg/kg bw/day.For ethyl paraben, a NOAEL of 1000 mg/kg bw/day for repeated-dose toxicity and DART was estimated by interpolation from methyl paraben and propyl paraben, i.e. the two category members on both sides of the target substance (ECHA 2008; OECD 2014).

Conclusion:

The concerns relating to reproductive toxicity and endocrine activity of common parabens expressed by ECHA (2017a, 2018) were predominantly associated with results from in vitro and in vivo screening assays that correspond to OECD CF Level 2 and Level 3 assays according to the Guidance for the identification of endocrine disruptors in the context of Regulations (EU) No 528/2012 and (EC) No 1107/2009 (EFSA and ECHA 2018). These findings mainly relate to oestrogen agonistic activity (Routledge et al. 1998; Blair et al. 2000; Byford et al. 2002 etc). The ability of parabens to transactivate oestrogen receptors in in vitro OECD CF Level 2 assays increases with alkyl chain length. From amongst the shorter-chained linear n-alkyl parabens, methyl paraben elicits the least and butyl paraben the most pronounced oestrogenicity. By contrast, across the entirety of linear n-alkyl parabens, a U-shaped association between chain length and in vitro oestrogenicity is observable: Amongst twelve parabens with linear n-alkyl chains ranging in length from C1 to C12, heptyl paraben (C14H20O3, i.e. C7-alkyl chain) and pentyl paraben (C12H16O3, i.e. C5-alkyl chain) showed the most potent agonistic activity on human oestrogen receptors α and β in the order of 10-7 M and 10-8 M, respectively (Watanabe et al. 2013). Oestrogen agonistic activities decreased in a stepwise manneras the alkyl chain was shortened to C₁(methyl paraben) or lengthened to C₁₂(dodecyl paraben) (Watanabe et al. 2013). Importantly, all oestrogenicity (also that of the most active parabens heptyl paraben and pentyl paraben) still remained many orders of magnitude lower than that of the natural oestrogen 17ß-estradiol, which is commonlyused as positive control in suchin vitrostudies (Soni et al. 2005; Watanabe et al. 2013; US EPA 2015). Focus of the read-across/analogy approach used here is on the shorter-chained linearn-alkyl parabens and an interpolation of missing data for ethyl paraben (as target substance) using measured values from other members on both sidesof that member (i.e. methyl paraben and propyl paraben as source substances) as requested in ECHA (2008) and OECD (2014).As ECHA (2008) clearly denotes that interpolation is preferred to extrapolation and that the extrapolation of missing data requires special considerations. Thus data on propyl paraben with regard to ED and reproductive toxicity can be considered as worst case.

Assessments of shorter-chained linearn-alkyl parabens in OECD CF Level 3 rodent uterotrophic assays yielded equivocal results (Routledge et al. 1998; Soni et al. 2001, 2002; CIR 2008, 2019; Ohta et al. 2012). A variety of mostly non-TG-conforming studies have investigated whether parabens have the potential to affect the reproductive system as important target organ system for endocrine disruptors. Conflicting reports are available on effects of parabens on the male or female reproductive system (Section 4.4). As the U.S.-based Cosmetic Ingredients Review (CIR) Expert Paneldenoted:

“(1) Many of these reports are irrelevant to the routes of exposure associated with intended cosmetic use, or otherwise did not account for the extensive metabolism of parabens (to metabolites with no known developmental and reproductive toxicity activity);

(2) are the result of poorly designed studies; and

(3) were not verified by other methods” (CIR 2019).

The higher-tier studies requested under REACH Annexes IX-X for methyl paraben and propyl paraben correspond to Level 4 and Level 5 studies of the OECD CF (OECD 2012). While OECD CF Level 2 and Level 3 assays inform on a substance’s potential to exhibit endocrine activity (e.g. oestrogenicity), only OECD CF Level 4 and Level 5 studies also inform on a substance’s potential to elicit adverse effects in an intact organism as a consequence of the endocrine activity. Such adversity is an indispensable prerequisite to meet the definition of an endocrine disruptor (WHO IPCS 2002).

Since the OECD CF Level 4 and Level 5 studies served regulatory purposes, they were performed following internationally agreed, standardised test protocols (i.e. OECD TGs). Thereby, the relevance, reliability and repeatability of findings is ensured, and it is ascertained that animal group sizes are adequate for the statistical analysis of findings. Adherence to OECD TGs also facilitates the mutual acceptance of data on an international level (OECD 2020). Study relevance was further enhanced by conducting all studies in full compliance with the principles of Good Laboratory Practice (GLP; OECD 1998).

The EOGRTS (OECD TG 443) is the most sensitive and most comprehensive study for detecting DART and/or endocrine disrupting effects that may occur as a result of pre‐ and postnatal substance exposure. This study provides information on gonadal function, the oestrus cycle, epididymal sperm maturation, mating, conception, gestation, parturition, lactation, weaning, and growth and development of the offspring. Further, the assessment included breeding up to the second-generation offspring and the two optional cohorts to investigate potential for DNT and DIT. Methyl paraben and propyl paraben elicited no toxicologically relevant alterations of any of the parameters addressed in the EOGRTS. In this regard, they also did not exhibit DNT (evaluated by neurobehavioural testing, neurohistopathology, learning and memory testing) or DIT (evaluated by an integrated analysis of all immunologically relevant data including a T-dependent antibody response of a functional immune system (data not shown)). Since no adverse effects were observed for either methyl paraben or propyl paraben in the EOGRTS up to the limit dose, the NOAEL for both substances was set at 1000 mg/kg bw/day. Due to structural/ function similarities and a comparable metabolism of methyl, ethyl and propyl paraben (for more details see attached read-across justification – chapter 13) a NOAEL of 1000 mg/kg bw/day for systemic repeated-dose toxicity and DART is interpolated for ethyl paraben.