A systematic review of the evaluation of endocrine-disrupting chemicals in the Japanese medaka (Oryzias latipes) fish

Japanese medaka (Oryzias latipes) is an acceptable small laboratory fish model for the evaluation and assessment of endocrine-disrupting chemicals (EDCs) found in the environment. In this research, we used this fish as a potential tool for the identification of EDCs that have a significant impact on human health. We conducted an electronic search in PubMed (http://www.ncbi.nlm.nih.gov/pubmed) and Google Scholar (https://scholar.google.com/) using the search terms, Japanese medaka, Oryzias latipes, and endocrine disruptions, and sorted 205 articles consisting of 128 chemicals that showed potential effects on estrogen–androgen–thyroid–steroidogenesis (EATS) pathways of Japanese medaka. From these chemicals, 14 compounds, namely, 17β-estradiol (E2), ethinylestradiol (EE2), tamoxifen (TAM), 11-ketotestosterone (11-KT), 17β-trenbolone (TRB), flutamide (FLU), vinclozolin (VIN), triiodothyronine (T3), perfluorooctanoic acid (PFOA), tetrabromobisphenol A (TBBPA), terephthalic acid (TPA), trifloxystrobin (TRF), ketoconazole (KTC), and prochloraz (PCZ), were selected as references and used for the identification of apical endpoints within the EATS modalities. Among these endpoints, during classification, priorities are given to sex reversal (masculinization of females and feminization of males), gonad histology (testis–ova or ovotestis), secondary sex characteristics (anal fin papillae of males), plasma and liver vitellogenin (VTG) contents in males, swim bladder inflation during larval development, hepatic vitellogenin (vtg) and choriogenin (chg) genes in the liver of males, and several genes, including estrogen–androgen–thyroid receptors in the hypothalamus–pituitary–gonad/thyroid axis (HPG/T). After reviewing 205 articles, we identified 108 (52.68%), 46 (22.43%), 19 (9.26%), 22 (17.18%), and 26 (12.68%) papers that represented studies on estrogen endocrine disruptors (EEDs), androgen endocrine disruptors (AEDs), thyroid endocrine disruptors (TEDs), and/or steroidogenesis modulators (MOS), respectively. Most importantly, among 128 EDCs, 32 (25%), 22 (17.18%), 15 (11.8%), and 14 (10.93%) chemicals were classified as EEDs, AEDs, TEDs, and MOS, respectively. We also identified 43 (33.59%) chemicals as high-priority candidates for tier 2 tests, and 13 chemicals (10.15%) show enough potential to be considered EDCs without any further tier-based studies. Although our literature search was unable to identify the EATS targets of 45 chemicals (35%) studied in 60 (29.26%) of the 205 articles, our approach has sufficient potential to further move the laboratory-based research data on Japanese medaka for applications in regulatory risk assessments in humans.


Introduction
Due to the increase in industrial and agricultural activities, endocrine-disrupting chemicals (EDCs), defined by the World Health Organization (WHO) as "Exogeneous substances that alter function(s) of the endocrine system and consequently cause adverse health effects in an intact organism or its progeny, or (sub)populations," are accumulated in the environment.A strategic approach to identify EDCs would be utilized by the existing knowledge to prioritize and focus on the screening and environmental monitoring efforts of these chemicals.The European Commission also set criteria for the identification of EDCs that require regulatory action.Currently, endocrine disruptors (EDs) are identified on a case-by-case basis using the available guidance provided in the OECD Guidance Document 150 (2018).The OECD Conceptual Framework for Testing and Assessment of EDs provided a tiered framework for the organization of study information to assess endocrine activity.This framework provides guidance for prioritizing relevant data streams and methods according to the type and level of information needed for a regulatory assessment.In the USA, EPA's EDSP has developed the requirements for the prioritization, screening, and testing of environmental contaminants, including pesticides, commercial chemicals, and agricultural products, for their potential to impact the endocrine system, especially in relation to estrogen, androgen, and thyroid (EAT) hormones and their nuclear receptors (NIEHS, 2018).Moreover, the perturbation of the enzymes of steroidogenesis by EDCs has potential effects on EAT pathways.Therefore, a two-tier testing approach was designed by EDSP.Tier 1 assays detect the potential effects of a chemical by various modes of action (Tier 1: screening) on EATS pathways.The results of the Tier 1 assays are evaluated by using a "weight of evidence" approach to determine whether the potential of the chemical is to interact with EATS and whether a Tier 2 assay is necessary.The purpose of Tier 2 studies is to use in vivo testing to further characterize the EATS effects and establish a dose-response relationship for adverse effects produced by the chemicals.Tier 2 tests are much longer-term studies that include exposure during critical life stages and have a broad range of more tightly spaced treatment than Tier 1.Moreover, Tier 2 tests can encompass multiple generations, covering effects on fecundity and fertility, development, growth, and sexual maturity.The successful completion of Tier 2 testing provided information to establish exposure and effect relationships, and assessed relevant endpoints across most life stages.
In aquatic environments, fish are considered one of the primary risk organisms for EDCs, especially those interacting with reproductive hormones.Sex determination in fish is very labile and can be disrupted or functionally reversed by external agents at critical developmental stages (Francis, 1992).Fish populations are directly exposed to a wide variety of EDCs, originating from industrial, agricultural, or municipal effluents (Ternes et al., 1999;Chen et al., 2007;Kim et al., 2014a).Evidence shows that EDCs can have long-term effects on reproduction and subsequent population development in natural fish populations (Kidd et al., 2007).The effects of EDCs on nuclear receptors have been studied extensively in small fish models like zebrafish, Japanese medaka, stickleback, and roach (Rutilus rutilus) (Iguchi et al., 2006;Lange et al., 2009;Tohyama et al., 2015).Since endocrine disruptions are linked to the receptor level, to predict ED effects, the identification of appropriate biomarkers at molecular levels is necessary.
Japanese medaka (Oryzias latipes) fish are small, freshwater teleost fish that inhabit gently flowing rivers and waterways.Like zebrafish (Danio rerio) and fathead minnows (Pimephales promelas), it is one of the small fish models (vertebrate) used in EDC studies (OECD, 2018).The sex determination locus has been identified in this fish species, and external sex-specific markers (chromatophores, shape of the anal and dorsal fins, anal fin papillae) can be used to easily differentiate males from females both from phenotypic and genotypic standpoints (Scholz and Mayer, 2008).Several OECD test guidelines (OECD TG 229; OECD TG 240) were used during the evaluation of EDCs in Japanese medaka, following tier-based approaches (Tier 1 and Tier 2).Moreover, the effects of endocrine active chemicals on Japanese medaka were reviewed previously (Urushitani et al., 2007;Flynn et al., 2017;Onishi et al., 2021;Kawashima et al., 2022).Based on the available publications found in public databases, we hypothesized that a literature search can identify the number and sources of EDCs that disrupted the EATS-related pathways of Japanese medaka (O.latipes) and correlate the effects with specific receptors at the molecular level.
In this review, we summarized the data on EDCs available in public databases, highlighting the links between molecular, phenotypic, and physiological endpoints using Japanese medaka as a single fish species.Although majority of the data refer to interfering with reproductive and thyroid hormone signaling pathways (EATS), limited information about the disruption of other endocrine organs, like the endocrine pancreas and interrenal gland (fish homolog of the adrenal gland), is also available (Dasmahapatra and Tchounwou, 2022a;b, 2023a;b).We evaluated the selective effects of 128 EDCs reported in 205 articles.As a result, we believed that 43 of them (EDCs) show potential to proceed to Tier 2 tests, and 13 chemicals should be considered EDCs without any further tier-based studies.
2 Materials and methods

Literature search strategy
The objectives of the literature search were to identify the relevant studies published in peer-reviewed journals that focused on the endocrine disruption of Japanese medaka (O.latipes) induced by various chemicals detected in the aquatic environments.The search was performed in PubMed (http://www.ncbi.nlm.nih.gov/pubmed) and Google Scholar (https://scholar.google.com/).PubMed was considered the main and reliable source of information; however, Google Scholar was used if the full text article was not available in PubMed.We initiated our search in PubMed using the search term Japanese medaka (Oryzias latipes), which provided 3,747 results (until 30 June 2023).We narrowed down the search by adding the term "endocrine" (Japanese medaka, Oryzias latipes, and endocrine), which reduced the number to 646, and finally, the addition of the term "endocrine disruption" reduced the results to 239 (Figure 1).We finally sorted 205 articles for review that focused on EATS pathways of Japanese medaka (Figure 1).We identified 128 chemicals that have potential ED effects on this fish (Japanese medaka, O. latipes) (Figure 2).After a literature search, we assembled ED-related information in Supplementary Table S1, which was also deposited in Figshare (doi/10.6084/mg.figshare.22598068).For classification of these compounds as selective disruptors of EATS pathways, 14 chemicals from 128 searched chemicals were selected as reference chemicals (Figure 2; Table 1).For estrogen endocrine disruptors (EEDs), E2 and EE2 were used as reference chemicals for agonists, and TAM was used for antagonists.For androgen endocrine disruptors (AEDs), 11-KT and TRB were used for agonists, and FLU and VIN were used for antagonists.For thyroid endocrine disruptors, (TEDs), T3 was used for agonists, and PFOA and TBBPA were used for antagonists.For steroidogenesis, TPA and TRF were used for stimulators, and KTC and PCZ were used for inhibitors (Table 1).After critical evaluation of the ED effects of these reference chemicals, the criteria of evaluation of endocrine disruption induced by an EDC on Japanese medaka are determined (Table 1).The chemicals which were unable to fulfill the criteria were considered unclassified.

Genes sensitive to EDCs within the EATS modalities of Japanese medaka
Within the EATS modalities, most of the EDCs function via the hormone-responsive element of a target gene by binding to the ligands of nuclear receptors (NRs), including ESRs (esr1, esr2a, and esr2b), ARs (arα and arβ), or TRs (trα and trβ).The effects of EDCs

FIGURE 1
Strategies for the selection of literature reports from peerreviewed articles published on Japanese medaka (Oryzias latipes).

FIGURE 2
Strategies for the selection of chemicals from peer-reviewed articles published on Japanese medaka (O.latipes).
The androgenic effects of EDCs are mediated via direct binding to ARs (arα and arβ) with distinctive binding properties or transactivation activity (Onishi et al., 2021;Kawashima et al., 2022).Molecular effects of AEDs could be identified from secondary sex characteristics (anal fin papillae of males) or indirectly by analysis of the induction/suppression of VTG (vtg1 and vtg2), LH, FSH, aromatase, ESRs, or T-hormone levels (Scholz and Mayers, 2008).The formation of papillary processes in the anal fin of Japanese medaka (males) is augmented by the bone morphogenic protein (bmp7) and lymphoid enhancer-binding factor (lef1), along with arα and arβ which can be used as markers for AEDs during evaluation (Ogino et al., 2014).
EDCs having TH-disrupting potential inhibit or accelerate THdependent processes, either directly or indirectly, including THdependent gene expression.The HPT axis is highly conserved among vertebrates, and the TH and receptors (trα and trβ) play crucial roles in the regulation of development, growth, and energy metabolism.A number of high-profile environmental pollutants adversely affect the TH system of Japanese medaka, including development, visual performance, malformation of the swim bladder, and TH-dependent gene (tshβ, trα, trβ, dio1, and dio2) expression (Godfrey et al., 2019;Dang et al., 2021;Horie et al., 2022c).Therefore, to evaluate the thyroid-disrupting potential of EDCs, in Japanese medaka, these genes (tshβ, trα, trβ, dio1, and dio2) can be used as markers for TEDs.
Moreover, within the estrogen-androgen-steroidogenesis (EAS) modalities, the steroid hormones, estrogen (E2) and androgen (A), are derived from cholesterol and secreted from the gonads (testis or ovary).The production, conversion, and breakdown of E2 and A in the endocrine glands and target tissues are carefully controlled by a range of steroidogenic enzymes (steroidogenesis), many of which belong to the cytochrome P450 family (CYP11, CYP17, and CYP19).Many EDCs have the abilities to disrupt the synthesis and function of steroidogenic enzymes, resulting in inappropriate concentrations of E2 or A, which impacts the reproduction, development, growth, and metabolism of fish (Japanese medaka).The enzyme aromatase (CYP19) converts testosterone/androgen (A) into estradiol/estrogen (E) and controls the fine balance between these two potent sex steroids.Therefore, the genes that show potential to regulate steroidogenesis in Japanese medaka are used as markers during EDC evaluation.

Results
Depending on the ED effects, we sorted 205 articles (Table 2) consisting of 128 chemicals (1.6 articles/chemicals or the approximate ratio is 8 articles:5 chemicals) that showed potential effects on Japanese medaka.Furthermore, based on the apical endpoints selected from 14 reference chemicals (Table 1) and after reviewing 165 articles, we identified 83 chemicals that target EATS pathways of Japanese medaka (Tables 3-10), and due to the lack of sufficient information, 45 chemicals reviewed from 60 articles remained unclassified (Table 11).Moreover, among the 83 chemicals that target EATS pathways, 43 chemicals were recommended for Tier 2 tests, and 13 chemicals show enough potential to be considered EDCs without any further tier-based studies (Tables 3-10).The rest of the EATS chemicals need further studies on Tier 1 screening.Moreover, with regard to the apical endpoints, the EATS chemicals were further classified as agonists and antagonists of EEDs, AEDs, and TEDs, and stimulators or inhibitors of steroidogenesis (Figure 2; Tables 3-10).

EEDs
For the identification and classification of EEDs from the searched chemicals, we considered three chemicals as references, E2 and EE2 as agonists, and TAM as antagonists (Table 1).Based on these reference chemicals, several endpoints, such as the femalebiased sex ratio, induction of serum VTG (protein) in male fish, alteration of the secondary sex characteristics (anal fin papillae in the male fish), and up-or downregulation of vtg and chg genes/mRNAs in the liver of male fish, as well as the estrogen receptors (ERs) of the HPG axis in both sexes, were considered (Table 1).Using these strategies, we reviewed 108 articles, which is 52.68% of the searched articles, consisting of 25 chemicals as agonists and 4 chemicals as antagonists (Tables 3, 4).Adding three reference chemicals to the list, the number of EED agonists increased to 27 (21.09% of 128 chemicals) and antagonists to 5 (3.9% of 128 chemicals), altogether 32, which is 25% of the total (128 chemicals) chemicals searched by the literature survey.Alternatively, it appears that for every 100 EDCs, ~21 of them are identified as EED agonists and ~4 of them are identified as EED antagonists.Moreover, considering the 108 articles that studied EEDs, every EED chemical was studied in 3.375 articles (27 articles: 8 EEDs).Moreover, among EED agonists, other than two reference chemicals (E2 was reviewed in 37 articles and EE2 in 27 articles), 4-nonylphenol (4-NP; 23 articles), bisphenol A (BPA; 21 articles), and 4-tert-octylphenol (4-t-OP; 8 articles) are the most studied EED agonist chemicals in Japanese medaka (Table 3).Among others, o,p′-DDT (4 articles), 4t-PP (4 articles), E1 (3 articles), PPB (3 articles), and TBCO (3 articles) have drawn significant interest among investigators.The remaining 17 estrogen agonists were studied either twice (8 chemicals) or once (9 chemicals).For EED antagonists, the reference chemical TAM was studied in five articles, whereas ATZ was studied twice (Ritcher et al., 2016), MET once (Lee et al., 2019), and TPhP in two articles (Li et al., 2019;Kawashima et al., 2022).Moreover, 16 of the EEDs as agonists and 4 as antagonists were recommended for Tier 2 tests.Therefore, based on the literature search, we recommend that eight chemicals (E1, E2, EE2, BPA, o,p′-DDT, 4-NP, 4-t-OP, and TAM) showed enough potential to be considered EEDs in Japanese medaka and did not require any further Tier 2 tests for estrogen signaling mechanisms.Furthermore, except PPB (Gonzalez-Doncel et al., 2014a), 4t-OP (Gray et al., 1999b), and MET (Lee et al., 2019), in most of the EED chemicals, whether agonists or antagonists, the thyroid-related endpoints remained uninvestigated, even though the reference agonists (E2 and EE2) have the potential to inhibit swim bladder inflation (a thyroid-related endpoint) in a concentration-dependent manner in larvae if the embryos were exposed either to E2 or EE2 during development (Pandelides et al., 2021).

AEDs
For AEDs, four chemicals, 11-KT and TRB as agonists and FLU and TRB as antagonists, were considered references (Table 1).Based on these reference chemicals, the apical endpoints, such as masculinization of females (development of anal fin papillae), male-biased sex ratio, upregulation of gsdf mRNA in XX  references, the ED effects of three compounds, DHT, LNG, and P4, were evaluated together (Onishi et al., 2021).Furthermore, among the androgen agonists, the AED effects of MT were peer-reviewed in eight articles, followed by GEN (four articles) (Table 4).Among the other agonists, LNG was reviewed in three articles, and the remaining seven chemicals were studied only once (Table 5).Among the apical endpoints, masculinization was induced by BF, GEN, LNG, P4, and SPR, while downregulation of hepatic vtg in females was observed in MT and SPR (Table 5).Among the eight chemicals identified as potential antagonists, the most studied chemical was DZ, which was studied in three articles (Hamm and Hinton, 2000;Flynn et al., 2018;Kawashima et al., 2022), followed by LD-BP and FNT, which were studied in two articles each (Li et al., 2016;2017;Horie et al., 2017;2022a).Other than these chemicals, the remaining five chemicals were studied once (one article/chemical).Moreover, based on the targeted apical endpoints related to AED and the literature review, we recommend that nine chemicals showed enough potential to proceed to Tier 2 tests, and five chemicals (FLU, 11-KT, MT, TRB, and VIN) did not require Tier 2 tests for the evaluation of androgen signaling mechanisms.In addition, similar to EEDs, the thyroid-related apical endpoints, such as hypertrophy of thyroid follicular cells, were induced by 11-KT (reference agonist) and FLU (reference antagonist) in Japanese medaka (Leon et al., 2007).Other than the references, LNG (agonist) and DZ (antagonist) showed the potential to modulate swim bladder inflation in Japanese medaka larvae during development (Hamm and Hinton, 2000;Pandelides et al., 2021).Furthermore, GEN (agonist) shows potential to regulate the expression of dio2 mRNAs in larvae if the embryos were exposed to GEN during development (Schiller et al., 2013;2014).Therefore, during the classification of EDCs as AED, the thyroid-related apical endpoints should not be ignored.

TEDs
For TEDs, three chemicals, T3 as the agonist and PFOA and TBBPA as antagonists, were considered references (Table 1).The apical endpoints, such as swim bladder inflation in larvae, disruption of thyroid histopathology, and up-or downregulation of TH receptor genes (trα and trβ) and deiodinases (dio1 and dio2), were considered during TED evaluations.Our literature search found only 19 articles, which is 9.26% of the total articles (205 articles) sorted are focused on TED.From these articles, 12 chemicals, one as agonist (Table 7), and 11 chemicals as antagonists, were identified as TEDs (Table 8).Considering three references, 15 chemicals, 2 as agonists (1.56% of 128 EDCs) and 13 as antagonists (10.16% of 128 EDCs), which is only 11.72% of the screened chemicals (128 chemicals), showed TED effects on Japanese medaka.Alternatively, for 100 EDCs, 1.56 chemicals are identified as TED agonists, and ~10 chemicals are identified as TED antagonists.Moreover, 19 articles identified 15 chemicals, which indicated that one TED was reviewed in 1.266 articles (approximately 5 articles:4 chemicals).Moreover, 5 chemicals, including three references and two antagonists (DIC and EHMC), were recommended to proceed to Tier 2 tests.The reference agonist T3 was studied in two articles, and the reference antagonists PFOA and TBBPA were included in four articles and 1 article, respectively (Table 1).Other than the references, the most studied chemical as a TED antagonist in Japanese medaka was DIC, which was peer-reviewed in five articles (Hong et al., 2007;Lee et al., 2011;Yokota et al., 2017;2018;Pandelides et al., 2021).Other chemicals, such as ATBC, PFBA, PFOS/PFOSA, and TDCPP, were studied in two articles each.The remaining four antagonists were studied only once (Table 8).Although ATBC and TDCPP were evaluated as TED antagonists, the downregulation of liver vtg1 and vtg2 genes in XX fish by ATBC (Horie et al., 2023b) and upregulation of vtg mRNA in both male and female larvae by TDCPP (Godfrey et al., 2019) indicated that TED chemicals have the potential to regulate EAS pathways, which need further verifications.

MOS
For identification of the MOS chemicals in Japanese medaka, four chemicals, TPA and TRF as stimulators and KTC and PCZ as inhibitors, were used as reference chemicals (Table 1).The apical endpoints selected for steroidogenesis are either the up-or downregulation of cyp19 genes that show potential to regulate the aromatase enzyme activity and lead to an increase or decrease in the circulating estrogen level in Japanese medaka.Our literature search selected 26 articles, which is 12.68% of the sorted articles, for the evaluation of steroidogenesis in Japanese medaka (Table 2).After reviewing these literature reports, four chemicals were considered stimulators of steroidogenesis and six chemicals were considered inhibitors (Table 9).Including the references, the total number of chemicals that interrupt steroidogenesis is 14, 6 stimulators (~5%), and 8 inhibitors (~6%), which is 10.93% of the identified chemicals that showed potential ED activities in Japanese medaka.Alternatively, for Frontiers in Toxicology frontiersin.orgevery 100 EDCs, 5 chemicals show potential to stimulate steroidogenesis and 8 chemicals inhibit steroidogenesis.Moreover, 14 MOS were identified after reviewing 26 articles, which indicated that for the identification of a chemical as MOS, 1.857 articles/MOS are reviewed (approximately 9 articles: 5 chemicals).Moreover, although the thyroid-related endpoints were not considered in these chemicals, including two references (TPA and TRF as agonists), nine chemicals (six as agonists and three as antagonists) were recommended for Tier 2 tests (Tables 9, 10).Among the stimulators, the ED activities of FPN, an insecticide, and TRI, a pharmaceutical product, were studied together (Sun et al., 2014).However, FPN was included separately in two articles (Sun et al., 2014;Wagner et al., 2017); the remaining three chemicals, OCL, RCT, and TRI, were investigated once (Sun et al., 2014;2016a;Yan et al., 2020) (Table 9).Among inhibitors, the most studied chemical is FAD, a nonsteroidal aromatase inhibitor, which was studied in five articles (Suzuki et al., 2004;Kuhl and Brower, 2006;Zhang et al., 2008b;Park et al., 2008;Thresher et al., 2011).Moreover, LET, a nonsteroidal triazole, was included in four articles    Frontiers in Toxicology frontiersin.org(Sun et al., 2007b;2009;2011a;Liao et al., 2014).BP, a UV filter used in cosmetics, was evaluated in three articles (Coronado et al., 2008;Kim et al., 2014;Kawashima et al., 2022), while the herbicide LNR and the organotin compound TPT-Cl were studied in two articles each (Table 10), and PRN was studied only once (Schiller et al., 2014).
Although the apical endpoints of MOS are mainly concentrated on aromatase enzyme genes and enzyme activities, the ED effects of these compounds on Japanese medaka either as an EED or AED can also be observed in TRI (Yan et al., 2020), RCT (Sun et al., 2016a), BP (Coronado et al., 2008;Kawashima et al., 2022), FAD (Zhang et al., 2008b), and LET (Sun et al., 2007b).

Discussion
Japanese medaka (Oryzias latipes) is one of the small laboratory fish models used for the evaluation of EDCs found in the environment (OECD, 2018).Like all other vertebrates, EATS pathways and their associated hypothalamus pituitary-releasing and -stimulating hormones are targeted by EDCs and disrupt the normal development and reproductive processes of this fish.For the identification of EDCs that specifically affect the endocrine systems of Japanese medaka (O.latipes), we searched the research articles in PubMed (http://www.ncbi.nlm.nih.gov/pubmed) and Google Scholar (https://scholar.google.com/)databases with the search terms, Japanese medaka, O. latipes, and endocrine disruptions.We hypothesized that literature search and evaluation can identify the number and sources of EDCs that disrupted the EATS-related pathways of Japanese medaka (Oryzias latipes) and provide additional evidence for the selection of a chemical as to whether to proceed to Tier 2 tests or not.
We sorted 205 articles that involved 128 chemicals for review (Figures 1, 2; Tables 1-11).Due to wide variations in experimental protocols and methodologies described in the research articles (n = 205), especially in non-TG studies, interpretation of the data from the literature survey became more complex.Moreover, the use of different life stages (embryos/larvae/adults), diversity in the modes of exposure (injection, immersion, and feeding), or in the duration Frontiers in Toxicology frontiersin.org27 Dasmahapatra et al. 10.3389/ftox.2023.1272368 of exposure (restricted either only in one generation or continued through multiple generations) made the problem even more complex.Therefore, to maintain consistency in the apical endpoints associated with ED effects, among the 128 identified chemicals, we selected 14 chemicals as the reference (Table 1).These chemicals (references) are either evaluated in this model (Japanese medaka) as reference chemicals by other investigators or screened through Tier 2 tests, following OECD guidelines (Flynn et al., 2017;Onishi et al., 2021;Kawashima et al., 2022;Myosho et al., 2022).Among these chemicals, E2 and EE2 (estrogen agonists), TAM (estrogen antagonist), 11-KT and TRB (androgen agonists), FLU and VIN (androgen antagonists), and KTC and PCZ (steroidogenesis inhibitors) were verified as agonists or antagonists for esr1 (for estrogen) and arβ (androgen) genes of Japanese medaka in vitro by RGA (Onishi et al., 2021;Kawashima et al., 2022).Additionally, the potential of E2, TAM, TRB, VIN, KTC, and PCZ as an EDC was evaluated in medaka through Tier 2 tests, following the MEOGRT protocol (Flynn et al., 2017).For stimulators of steroidogenesis, we considered TPA and TRF as reference chemicals (Jang and Ji, 2015;Zhu et al., 2015).For the thyroid, T3 as the agonist and PFOA and TBBPA as antagonists were considered, which were recently referenced by Godfrey et al. (2019) and Horie et al. (2023a) in Japanese medaka.Therefore, we think that the selection of reference chemicals for the identification of EATS-related apical endpoints and to set up guidelines is very reasonable and acceptable.Our approach identified 69 chemicals that show potential to target the EATS pathways of Japanese medaka, and 45 chemicals remained unclassified due to limited information, even though these unclassified chemicals induced ED-like effects in Japanese medaka (Table 11).Taken together, considering 14 references, 83 (69 identified +14 references = 83) chemicals are identified as EDCs (~65%) that disrupt EATS pathways of Japanese medaka (O.latipes), and 45 EDCs (~35%) remain unclassified due to the lack of sufficient information.
We further classified the EATS chemicals as agonists/ stimulators and antagonists/inhibitors of EEDs, AEDs, and TEDs, and MOS.The apical endpoints selected for agonists should be in contrast with antagonists, and in many cases, these borderlines cannot be maintained.For example, one of the significant apical endpoints of an EED as an agonist is the upregulation of VTG in the liver of male (XY) medaka (Flynn et al., 2017); however, TAM, which was used as a reference chemical of the EED antagonist, increased the liver VTG content in male fish (Flynn et al., 2017).To avoid complicacy, during analysis, we ignored the classification of EATS chemicals as agonists and antagonists, and simply included all the agonists and antagonists together and expressed them as EEDs, AEDs, TEDs, and MOS where applicable (Table 2).
As mentioned previously, 128 EDCs were identified after reviewing 205 individual articles, which indicates that for the identification of a chemical as an EDC in Japanese medaka, more than one article was reviewed (1.60 articles/chemical, or the approximate ratio is 8 chemicals: 13 articles).Our studies also showed that after reviewing 165 articles, 83 EDCs were identified that targeted EATS pathways (Tables 3-10), and 45 chemicals remained unidentified after reviewing 60 articles (Table 11).Accordingly, approximately 65% of the EDCs were identified with their specific EATS targets after reviewing 80% of the searched articles and 35% of the EDCs remained unclassified after reviewing 20% of the searched articles (Table 2).Therefore, it appears that the databases consist of more articles as classified EDCs (related to EATS) than unclassified EDCs (Table 2).Moreover, as the EATS pathways are interdependent on each other through the common hypothalamus-pituitary axis (HP axis), it is very difficult to classify the EDCs on the basis of apical endpoints specific to the EATS pathways.However, our studies showed that more than 65% of the articles identified EDCs as EED, 28% of the articles identified EDCs as AED, 12% of the articles identified EDCs as TED, and 16% of the articles identified EDCs as MOS (Table 2), which can be arranged in the order of TED < MOS < AED < EED.Furthermore, among 83 EDCs that targeted EATS pathways, 39% of them are identified as EEDs, 27% are AEDs, 18% are TEDs, and 17% are MOS (Tables 3-10), and the order of arrangement appears to be MOS < TED < AED < EED.Therefore, the potential of literature searching to identify EATS-targeted chemicals in Japanese medaka partially supports the concept that the more the number of articles in the databases, the more the number of EDCs should be identified.
As recommended by USEPA, the effects of an EDC should be evaluated using a tier-based approach.In Tier-1 studies, the endpoints are focused mainly on lethal concentrations (LC/LD/ IC 50 , NOEC, and LOEC), reproductive activity (fecundity, fertility, breeding behavior, and hatching of the embryos), sex reversal, secondary sexual features (the number of papillae in the anal fin rays which are present in juvenile/adult males and absent in female Japanese medaka), VTG (the egg yolk precursor protein), and choriogenins (the eggshell protein), which are absent in the liver of male fish, and histopathology of the gonad, liver, and kidney.The Tier 2 approach is multigenerational, consisting mostly of the same features evaluated in Tier 1 (fecundity, fertility, hatching, VTG content of the male fish liver, secondary sexual features, sex reversal, survivability of embryos, larvae, and adults, and histopathology of the gonad, liver, and kidney).Even though the Tier 2 tests are time-consuming, expensive, and need proper validation of the chemicals as an EDC through Tier 1 screening, for proper classification of the EDCs and their respective target endocrine organs or hormones in fish (Japanese medaka), multigenerational studies (Tier 2) are necessary (Kawashima et al., 2022).Accordingly, among 83 EATS (69 classified and 14 references), we recommend that six of the references (11-KT, T3, PFOA, TBBPA, TPA, and TRF), due to the limited number of articles (studies in Japanese medaka), should be considered high-priority candidate substances for Tier 2 testing.The eight other references (E2, EE2, TAM, TRB, FLU, VIN, KTC, and PCZ) were already verified either as reference chemicals during the evaluation of other EDCs or through multigenerational MEOGRT tests (OECD TG 240) (Flynn et al., 2017;Onishi et al., 2021;Kawashima et al., 2022).Therefore, these eight reference chemicals did not need any further Tier 2 tests for potential EAS-related effects; however, evaluation of thyroid-dependent mechanisms of these chemicals may require investigation (Myosho et al., 2019(Myosho et al., , 2021;;Pandelides et al., 2021).
During screening of EEDs, among the identified chemicals, we recommend 16 (AR-1260, BZP, BMT, CFR, CTC, p,p′-DDE, DES, EDS, EQ, E3, 4-MBC, MPB, OMC, PPB, 4t-PP, and TBCO) as agonists, and 4 chemicals (ATZ, MET, PFAA, and TPhP) as antagonists were high-priority chemicals for Tier 2 tests.Among the rest, EED potentials of o,p′-DDT and 4-t-OP were evaluated by multigenerational MEOGRT tests (Flynn et al., 2017) and probably did not require any further Tier 2 tests as well (Flynn et al., 2017).In addition, BPA, E1, 4-NP, and 4-t-OP were recommended for Tier 2 tests after successful evaluation through the OECD TG 229 protocol (Kawashima et al., 2022).Moreover, our literature search found that BPA was reviewed in 21 articles, 4-NP in 23 articles, 4-t-OP in 8 articles, and E1 in 3 articles (Table 3).Therefore, we believe that these EEDs (E1, BPA, o,p′-DDT, 4-NP, and 4-t-OP) showed enough potential to be considered EED agonists without performing any further Tier 2 tests.In AEDs, six chemicals (BF, CFD, GEN, LNG, P4, and SPR) as agonists and three chemicals (CPA, DZ, and 2-EHHB) as antagonists were recommended for Tier 2 tests (Tables 5, 6).Moreover, our literature search showed that MT was studied in eight articles and probably did not require Tier 2 tests anymore.However, P4 and LNG, as progestins, induced secondary sexual features in female Japanese medaka (XX) (Onishi et al., 2021), and further evaluation by Tier 2 tests is necessary.Among TEDs, two antagonists (DIC and EHMC) were recommended for Tier 2 tests and for MOS, four chemicals (OCL, FPN, RCT, and TRI), as stimulators, and three chemicals (BP, FAD and LET), as inhibitors, were recommended for Tier 2 tests.Taken together, among the 83 EDCs that targeted EATS pathways, 43 chemicals were recommended for Tier 2 tests, and 13 chemicals can be considered potential EDCs without any further Tier 2 tests in Japanese medaka.
Our literature search did not classify the EATS pathways of 45 chemicals (35%), even though several of them induced specific EATS-related apical endpoints (Table 11).Generally, in in vivo studies, probably due to the HPG and HPT axes, the overlapping effects of the chemicals within the EATS pathways cannot be ruled out; therefore, many of these unclassified chemicals demonstrated effects on endocrine-related apical endpoints, such as alteration in the liver VTG content (upregulated by CMP, CHDM, IBP, MTZ, NPX, 1NT, PDM, and RLX, and downregulated by DIBP), upregulation of chg in the liver of male fish (OXY and SRF), impaired reproductive activity and gonad histology (Cd, LD-BP, and nZnO), histopathological changes in the thyroid (BDE-47), inhibition of aromatase (TBT), and regulation of the E2 concentration in the blood of fish (Cd and SFT) (Table 11).In addition, several of the unclassified EDCs have potential as ESR agonists (CMP, DIBP, and FU) or antagonists (CYN, PHT, and RLX), and the ESR agonist and ARβ agonist (INT) and ESR agonist and ARβ antagonist (CMP) were observed in in vitro RGA with medaka esr1 and arβ genes (Onishi et al., 2021;Kawashima et al., 2022).Moreover, the nanocarbon, GO, was evaluated in 10 articles targeting the gonads, thyroid, interrenal glands, and pancreas in adults; and gonads, thyroids, and interrenal glands in larvae (Dasmahapatra et al., 2020a;Dasmahapatra et al., 2020b;Dasmahapatra and Tchounwou, 2022a;Dasmahapatra and Tchounwou, 2022b;Dasmahapatra and Tchounwou, 2023a;Dasmahapatra and Tchounwou, 2023b;Asala et al., 2021;2022;Myla et al., 2021a;b).Despite the histopathological alterations and cellular disruptions induced in the gonads, liver, kidneys, thyroid, interrenal glands, and pancreas of the adults and larvae of Japanese medaka by GO, due to the lack of specific Tier 1 and Tier 2 tests, GO remained unclassified without identifying any EATS-specific pathways (Dasmahapatra et al., 2020a;Dasmahapatra et al., 2020b;Dasmahapatra and Tchounwou, 2022a;Dasmahapatra and Tchounwou, 2022b;Dasmahapatra and Tchounwou, 2023a;Dasmahapatra and Tchounwou, 2023b;Asala et al., 2021;2022;Myla et al., 2021a;b).Therefore, we think that, before excluding the potential of these unclassified chemicals as an ED, further validations using tier-based approaches are necessary.Alternatively, the effects should be considered nonspecific, mediated through oxidative stress, or not related to EATS-specific mechanisms.
Although the effects of 128 EDCs in Japanese medaka are classified based on EATS modalities, the disruptions of non-EATS pathways by these chemicals need to be investigated carefully (Martyniuk et al., 2022).Moreover, compared to EATS, less attention has been given to other endocrine organs, including the endocrine pancreas and the interrenal gland (adrenal gland), which should belong to non-EATS pathways of Japanese medaka.Due to the lack of validated in vivo or in vitro methods and the availability of the appropriate literature in the public databases, the evaluation of EDCs targeting non-EATS modalities of Japanese medaka is not properly focused on this review.Our literature search on the effects of EDCs on the endocrine pancreas and interrenal glands of Japanese medaka found only four articles, two for pancreas (Dasmahapatra and Tchounwou, 2023a;Dasmahapatra and Tchounwou, 2023b), and two for interrenal glands (Dasmahapatra and Tchounwou, 2022a;Dasmahapatra and Tchounwou, 2022b) (Tables 2, 11) in PubMed (www.ncbi.gov).Therefore, despite the significant importance of non-EATS modalities in Japanese medaka, due to the lack of sufficient literature and standard methods, the evaluation of EDCs mediated through non-EATS pathways is not appropriately described in this review article.

TABLE 1
The apical endpoints of the reference chemicals related to EATS pathways in Japanese medaka.

TABLE 1 (
Continued) The apical endpoints of the reference chemicals related to EATS pathways in Japanese medaka.

TABLE 1 (
Continued) The apical endpoints of the reference chemicals related to EATS pathways in Japanese medaka.

TABLE 1 (
Continued) The apical endpoints of the reference chemicals related to EATS pathways in Japanese medaka.

TABLE 1 (
Continued) The apical endpoints of the reference chemicals related to EATS pathways in Japanese medaka.

TABLE 1 (
Continued) The apical endpoints of the reference chemicals related to EATS pathways in Japanese medaka.

TABLE 2
Literature reports sorted for the evaluation of the effects of EDCs on Japanese medaka (Oryzias latipes).

TABLE 2 (
Continued) Literature reports sorted for the evaluation of the effects of EDCs on Japanese medaka (Oryzias latipes).

TABLE 2 (
Continued) Literature reports sorted for the evaluation of the effects of EDCs on Japanese medaka (Oryzias latipes).

TABLE 2 (
Continued) Literature reports sorted for the evaluation of the effects of EDCs on Japanese medaka (Oryzias latipes).

TABLE 2 (
Continued) Literature reports sorted for the evaluation of the effects of EDCs on Japanese medaka (Oryzias latipes).TABLE Potential EED agonist chemicals identified from the literature search.TABLE (Continued) Potential EED agonist chemicals identified from the literature search.
Only the names of the authors are listed in the first column of the table.The cells filled in colors (red = EED; blue = AED; green = TED; yellow = MOS; black = unidentified pathways) represent the specific endocrine pathways/organs disrupted by EDCs.EDCs, endocrine-disrupting chemicals; EED, estrogen endocrine disruptor; AED, androgen endocrine disruptor; TED, thyroid endocrine disruptor; MOS, modulators of steroidogenesis.

TABLE 3 (
Continued) Potential EED agonist chemicals identified from the literature search.

TABLE 4
Potential EED antagonist chemicals identified by the literature search.

Table 1 )
, were mostly considered during the evaluation of AEDs.With these efforts, from 46 articles, which is 22.43% of the sorted articles (Table2), we identified 10 chemicals as agonists (Table5) and 8 chemicals as antagonists (Table6).With the addition of four reference chemicals, the number of AEDs increased to 22 (~9% agonists and ~8% antagonists), which is 17.18% of the 128 chemicals screened through the literature search.Alternatively, for every 100 EDCs, ~9 chemicals are identified as AED agonists and ~8 chemicals are identified as AED antagonists.Moreover, with regard to 46 articles that studied 22 AEDs, it appears that one AED chemical was studied in 2.09 articles (approximately 2 articles: 1 AED).Moreover, among the reference chemicals, effects of TRB were observed in 14 articles, FLU was in 8 articles, and 11-KT and VIN were included in 5 articles (Table1).Other than the

TABLE 5
Potential androgen endocrine-disrupting agonist chemicals identified from the literature search.

TABLE 6
Potential AED antagonist chemicals identified by the literature search.

TABLE 8
Potential TED antagonist chemicals identified from the literature search.

TABLE 9
Potential steroidogenesis stimulating EDCs identified from the literature search.
2. Upregulation of vtg1 and vtg2 in males and downregulation of vtg1 and vtg2 in females 3. Upregulation of esr1 and arα in males

TABLE 10
Potential steroidogenesis inhibitory EDCs identified by the literature search.

TABLE (
Continued) Potential EDCs with unidentified EATS pathways.