Ethynylestradiol feminizes gene expression partly in testis developing as ovotestis and disrupts asymmetric Müllerian duct development by eliminating asymmetric gene expression in Japanese quail embryos

Abstract

In avian embryos, xenoestrogens induce abnormalities in reproductive organs, particularly the testes and Müllerian ducts (MDs). However, the molecular mechanisms remain poorly understood. We investigated the effects of ethynylestradiol (EE₂) exposure on gene expression associated with reproductive organ development in Japanese quail embryos. Reverse transcription quantitative polymerase chain reaction (RT-qPCR) analysis revealed that the left testis containing ovary-like tissues following EE₂ exposure highly expressed the genes for steroidogenic enzymes (P450_scc, P450_{17α, lyase}, and 3β-HSD) and estrogen receptor-β, compared to the right testis. No asymmetry was found in these gene expression without EE₂. EE₂ induced hypertrophy in female MDs and suppressed atrophy in male MDs on both sides. RNA sequencing analysis of female MDs showed 1,366 differentially expressed genes between developing left MD and atrophied right MD in the absence of EE₂, and these genes were enriched in Gene Ontology terms related to organogenesis, including cell proliferation, migration and differentiation, and angiogenesis. However, EE₂ reduced asymmetrically expressed genes to 21. RT-qPCR analysis indicated that genes promoting cell cycle progression and oncogenesis were more highly expressed in the left MD than in the right MD, but EE₂ eliminated such asymmetric gene expression by increasing levels on the right side. EE₂-exposed males showed overexpression of these genes in both MDs. This study reveals part of the molecular basis of xenoestrogen-induced abnormalities in avian reproductive organs, where EE₂ may partly feminize gene expression in the left testis, developing as the ovotestis, and induce bilateral MD malformation by canceling asymmetric gene expression underlying MD development.

Japanese quail, ethynylestradiol, Müllerian ducts, testes, asymmetric gene expression

The avian species population is rapidly declining, partly due to harmful chemicals released into the environment (Ceballos et al., 2017; Maxwell et al., 2016). Evaluating chemical toxicity in avian species is crucial to mitigate the risk. Fertilized eggs serve as a valuable model for assessing chemical toxicity since chemicals accumulated in egg yolks from adult female birds’ exposure may pose risks to subsequent generations (Lin et al., 2004; Ottinger et al., 2008). Endocrine-disrupting chemicals mimicking or blocking estrogen actions can induce teratogenic anomalies and morphological changes in embryonic reproductive organs. These chemicals may also lead to suppressed egg formation, eggshell thinning, and reduced fertility in adults (Marlatt et al., 2022; Ottinger et al., 2013, 2008).

The Japanese quail (Coturnix japonica) is a widely used avian model in endocrinology (Brunström et al., 2009; Ottinger et al., 2013, 2008; Panzica et al., 2007). Female embryos exhibit higher blood estradiol concentrations than males during the embryonic period (Ottinger et al., 2001; Schumacher et al., 1988), resulting in sexually different process of reproductive organ development. Administering estradiol to male embryos can result in ovotestis formation in the left gonad and retention of Müllerian ducts (MDs), which may develop into oviducts with continued estradiol administration posthatching (Brunström et al., 2009; Rahil and Narbaitz, 1972). In females, excess estradiol during the embryonic period can impede oviduct development, causing left MD malformation and right MD retention (Rissman et al., 1984). Hence, appropriate estrogen levels during embryonic development are critical for sexual differentiation of reproductive organs. Endocrine-disrupting chemicals, as demonstrated by abnormalities induced by in ovo exposure to xenoestrogens and estrogenic chemicals, pose a serious risk to abnormal reproductive organ development (Berg et al., 1999, 2001a,b; Halldin et al., 2003; Nishijima et al., 2003).

Quail embryos express steroidogenic enzyme genes for estrogen synthesis highly in ovaries, with lower levels in testes, indicating sex differences (Mattsson et al., 2008b; Tsukahara et al., 2021). During late embryonic stages, estrogen receptor-α (ERα) expression is higher in ovaries and left MD compared to testes and right MD, respectively (Tsukahara et al., 2021). Estrogen, acting via ERα, influences left gonad and MD development, leading to ovary and oviduct formation, while estrogen receptor-β (ERβ) plays a minor role (Mattsson and Brunström, 2010, 2017; Mattsson et al., 2008a,b). Androgens through the androgen receptor (AR) are crucial for left ovary development (Katoh et al., 2006), and anti-Müllerian hormone (AMH) via AMH receptor type 2 (AMHR2) is vital for gonadal growth and MD regression (Lambeth et al., 2015). Thus, reproductive organ development requires genes involved in steroidogenesis, sex steroid actions, and AMH actions. Chemical exposure may induce abnormalities through changes in gene expression, although this hypothesis lacks current support. Additionally, endocrine-disrupting chemicals mimicking or blocking estrogen actions have the potential to alter unknown gene expression related to reproductive organ formation, as ligand-binding ERs function as transcription factors (Ikeda et al., 2015; Klinge, 2001).

Ethynylestradiol (EE₂), a predominant estrogen in birth control pills, is widely used. EE₂ in wastewater is incompletely broken during treatment (Kidd et al., 2007) and exhibits toxic effects on avian reproductive organs, as seen in Japanese quail studies (Berg et al., 1999, 2001b). EE₂ induces MD malformation in female embryos, right oviduct retention, left oviduct contraction in adult females, and persistence of MDs with left gonad development as ovotestis in male embryos. However, the mechanisms behind EE₂ effects remain unclear. In this study, we explored the impact of EE₂ on gene expression in steroidogenesis, sex steroid actions, and AMH actions during the embryonic period. In addition, to unravel the molecular mechanisms responsible for asymmetric MD development and EE₂-induced MD abnormalities, we conducted RNA sequencing (RNA-seq) and reverse transcription quantitative polymerase chain reaction (RT-qPCR) analyses of the MDs.

Materials and methods

Preparation and incubation of fertilized eggs

Animal breeding and the subsequent experiments were approved by the Animal Care and Use Committee at the National Institute for Environmental Studies (NIES; approval nos. AE-21-13-R1 and AE-22-16) and conducted in accordance with NIES guidelines. Fertilized eggs from NIES-L Japanese quail, established through rotation breeding in the NIES closed colony in Tsukuba, Japan, were used. To obtain fertilized eggs, adult male and female NIES-L Japanese quail were paired and housed in cages (14.0 cm  × 40.5 cm  × 18.0 cm) under standard breeding conditions (photoperiod: 14/10 h light/dark; temperature: 23°C ± 2°C; humidity: 50% ± 20%) with free access to food and tap water. Fertilized eggs were incubated at 37.8°C and 60%–70% humidity, turned every 60 min. The day of placing the eggs in an incubator was considered embryonic day 0 (E0).

Experimental design

Effects of EE₂ exposure on reproductive organs in quail embryos

Quail embryos were cultured in a surrogate eggshell culture system and exposed to sesame oil (control; n = 48) or sesame oil with EE₂ at 7, 20, and 60 ng/egg (n = 92, 60, and 77, respectively) from E3. The experimental groups exposed to EE₂ at 7, 20, and 60 ng/egg were termed EE7, EE20, and EE60, respectively. EE₂ doses were chosen based on a prior study indicating that in ovo exposure exceeding 2 ng/g egg, but not at 0.7 ng/g egg, from E3 to E15 induces ovotestis formation, inhibits MD regression in male embryos, and causes MD malformation in female embryos (Berg et al., 1999). These doses were administered to NIES-L quail eggs (weight ≈10 g), and embryo survival rates were monitored at E9, E12, and E15. At E15, gonadal, MD, and Wolffian duct (WD) morphology was observed (n = 15 in control female, 22 in EE7 female, 24 in EE20 female, 21 in EE60 female, 15 in control male, 19 in EE7 male, n = 12 in EE20 male, and 29 in EE60 male). Samples were then taken from the gonads and MDs from E15 embryos for RT-qPCR or histological analysis. Gonads (n = 6 in each female group, 7 in control male, 6 in EE7 male, n = 4–6 in EE20 male, and 6 in EE60 male) underwent gene expression analyses via RT-qPCR to determine mRNA levels of steroidogenic enzymes [cholesterol side chain cleavage enzyme (P450_scc), 3β-hydroxysteroid dehydrogenase (3β-HSD), 17α-hydroxylase/C17-20 lyase (P450_{17α, lyase}), aromatase (P450arom), and 17β-hydroxysteroid dehydrogenase (17β-HSD)], sex steroid receptors (ERα, ERβ, and AR), AMH, and AMHR2. Additionally, RT-qPCR analysis measured mRNA levels of sex steroid receptors and AMHR2 in the MDs (n = 4–6 in control female, 3–7 in EE7 female, 5–6 in EE20 female, 5–6 in EE60 female, and 6 in EE60 male). Gonads (n = 4 in control female, 3 in EE60 female, 3 in control male, and 5 in EE60 male) and MDs (n = 3 in control female, 4 in EE60 female, and 6 in EE60 male) were processed for histological analysis. Plasma estradiol and androgen concentrations were measured using an enzyme-linked immunosorbent assay (ELISA) with blood samples collected from the chorioallantoic vein (n = 6 in each group for each sex). Genetic sex was determined in embryos using polymerase chain reaction (PCR) with DNA extracted from the leg.

Effects of EE₂ exposure on the transcriptome of MDs in quail embryos

Quail embryos were cultured using a surrogate eggshell system and exposed to either sesame oil (control) or sesame oil with EE₂ (60 ng/egg) from E3. The reproductive organs of the control and EE60 embryos were observed (n = 14 in control female, 14 in EE60 female, 12 in control male, and 14 in EE60 male), and the left and right MDs were sampled at E12. After extracting and purifying total RNA from the MDs and determining their sex via genomic PCR, RNA-seq analysis was performed on the total RNA samples from control and EE60 female embryos (n = 3 in each group). The samples were prepared by combining equal amounts of total RNA from four embryos in each group (see “RNA-seq analysis”). Then, RT-qPCR analysis was conducted on total RNA samples from control and EE60 female embryos, as well as EE60 male embryos (n = 6 in each group), to validate the RNA-seq analysis and investigate the effects of EE₂ on gene expression in both sexes.

General procedures

Surrogate eggshell culture and EE₂ exposure

Embryos were cultured using a slightly modified version of the system previously reported by Perry (1988). Fertilized eggs from NIES-L Japanese quail were utilized 3 days after the start of incubation, and the eggshells for culturing were obtained from NIES-Brn Japanese quail, a strain established at NIES. Upon transferring the contents of the fertilized eggs, 50 µl of sesame oil, with or without EE₂ (7, 20, or 60 ng), was added to the embryo surface. The open part of the eggshells was covered with a polyvinylidene chloride film sheet, and the embryos within the eggshells were cultured at 37.8°C and 60%–70% humidity, with turning performed every 30 min.

Sexing

The genetic sex of each embryo was determined via PCR amplification of leg DNA using a specific primer set (forward primer: 5′-GTTACTGATTCGTCTACGAGA-3′; reverse primer: 5′-ATTGAAATGATCCAGTGCTTG-3′) designed for the genes coding chromo-helicase-DNA binding protein 1, with loci in the Z chromosome (CHD1Z) and W chromosome (CHD1W). DNA samples (3 µl) were amplified in a reaction mixture (10 µl) containing 1 µl of the forward primer (10 µM), 1 µl of the reverse primer (10 µM), and 5 µl of GoTaq Green Master Mix (Promega Corp., Madison, WI, USA). The PCR conditions were as follows: an initial step to activate Taq polymerase at 95°C for 5 min, followed by 35 cycles of denaturation at 95°C for 30 s, annealing at 55°C for 30 s, and extension at 72°C for 45 s, with a final extension at 72°C for 5 min. PCR products were electrophoresed with 2% agarose in tris-borate-EDTA buffer and stained with ethidium bromide (1 µg/ml) in tris-borate-EDTA buffer to detect PCR products from either the CHD1Z or CHD1W gene.

Histology

The gonads and MDs from quail embryos were fixed in 10% formalin–saline, dehydrated using a graded series of ethanol, immersed in xylene, and embedded in paraffin. The organs were sliced into 5-μm-thick sections, mounted on glass slides, and stained with hematoxylin and eosin. These sections were observed and imaged using a microscope (BZ-X800, Keyence Corp., Osaka, Japan).

ELISA

Plasma androgen and estradiol concentrations were determined based on a slightly modified version of a previously described procedure (Tsukahara et al., 2021) using testosterone ELISA (Cayman Chemical, Ann Arbor, MI, USA) and estradiol ELISA (Cayman Chemical) kits, respectively. The testosterone ELISA kit exhibited 100% cross-reactivity with testosterone and 27.4% cross-reactivity with 5α-dihydrotestosterone, the concentration of which in adult male quails is approximately 75% of the concentration of testosterone (Ottinger and Mahlke, 1984). The estradiol ELISA kit demonstrated 100% cross-reactivity with estradiol and 0.14% with EE₂.

RT-qPCR analysis

RT-qPCR analysis was conducted following our previously reported procedure (Tsukahara et al., 2021). In brief, first-strand cDNA was synthesized from total RNA extracted and purified from sampled tissues. Subsequently, qPCR was performed using specific primers for each target gene (Table 1). Target gene mRNA levels were quantified using the 2^-ΔΔCt method, with gene normalization performed using a housekeeping gene, and calibration conducted using a pooled cDNA sample. Selection of suitable housekeeping genes for normalization included the evaluation of five commonly used genes [β-actin, glyceraldehyde-3-phosphate dehydrogenase (GAPDH), TATA box binding protein (TBP), phosphoglycerate kinase 1 (PGK1), and tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein zeta (YWHAZ)]. Based on these results, GAPDH and YWHAZ were chosen as normalizers for gonads and MDs, respectively. TBP was used as the normalizer for MDs in the analysis to validate RNA-seq.

Table 1.

Primer sequences used for RT-qPCR analysis

Gene	Species	Forward primer sequence (5′–3′)	Reverse primer sequence (5′–3′)	Amplicon size (bp)	Accession number
P450scc	Gallus gallus	CATCCTCTTCAGCCTCCTTG	TCTTCAGCATCTTCACCCTGT	218	NM_001001756.1
P45017⍺,lyase	Coturnix coturnix	CTACCTGGAGGCTACCATCA	TATCCCACTCCTTCTCATCG	170	FJ267698.1
3β-HSD	Coturnix coturnix	TGTGACCTGTGCCCTGAGAT	ATGCGTCTGGCTTTCTGTG	210	FJ607242.1
17β-HSD	Coturnix coturnix	ACCTGGATGTGCTGGTGTG	CAGGAAGGCCTGAATGGTG	131	XM_015885716.1
P450arom	Coturnix japonica	CTGCCACAGTGCCTATCATC	CCGCTGAACCAAACTCTCAC	205	AF533667.2
ER⍺	Coturnix japonica	CTCACCCTTCATCCATCACC	AGGAGCAGTTTCCCTCATCC	88	AF442965.1
ERβ	Coturnix japonica	ATTCCAGCATGTTTCCCTTG	TTGCAATAACCCACACCAGA	102	AF045149.2
AR	Coturnix japonica	GCCTGAAGACACAAGACGAAG	CTCAGCAGGTTGGAGAAAGAG	199	AB188828.1
AMH	Coturnix japonica	AATGCGGCTGTCGGTAGA	TGTAGATGGGCAGAGTGAGC	178	AY904049.1
AMHR2	Coturnix japonica	CTGGCCTATGAGGCTGAGTT	GGGATCAATGGTCTCCTTCT	89	KU715092.1
CCNE2	Coturnix japonica	CCGCACTCTGCCACTATACC	AAGGGAACCATCCACTCCAC	97	XM_032442802.1
MCM2	Coturnix japonica	ATCTCCCATCTCCCTCTGGT	GGGTTTCACCTCCTGGTTCT	195	XM_015874982.2
MCM3	Coturnix japonica	AGACCGAGAAGGAGGAGGAG	TCAGGTGTCTTGGGAGTGTG	166	XM_032443739.1
MCM4	Coturnix japonica	CGTCTTGCTCGCCATTTAGT	TGCCACTTCCAATCTTCCTC	190	XM_015855556.2
MCM6	Coturnix japonica	TCGCATTCAAGAGACACAGG	CCACATCAGGCACAACAATC	146	XM_015868882.2
POLA1	Coturnix japonica	TTCCTCCTCCTCCTCCTCTT	AGCTGCCATAGCCACAATCT	95	XM_015885864.2
CCNA2	Coturnix japonica	GAAGCAGAAGAGCGACCAAA	CCAGCCAGTCCACCAGAATA	163	XM_015861915.1
CCNB2	Coturnix japonica	CAGTTGGTGGTGTTGGTGTT	GGCGACCTCAAAGAAGACC	129	XM_015872411.1
CCNB3	Coturnix japonica	TGGCACTCACCATGAAGAAC	GGCGACCTCAAAGAAGACC	182	XM_015860251.2
CENPH	Coturnix japonica	AACAGGACAGAGGTGAGAGGAG	GTTAGGCAGGTAGGCGTCAG	84	XM_015848599.2
NUF2	Coturnix japonica	TGGCACTTAGTTTGCCACAG	TGAGAGAGCAGGGAGGAGAA	133	XM_015869691.2
SMC2	Coturnix japonica	GGATGCTGCTCTTGACCTTT	ACCACAAGGAACTGGGATTG	84	XM_015849837.2
GAPDH	Coturnix coturnix	GGGAACGCCATCACTATCTT	TACTCAGCACCAGCATCTCC	155	Z19086.1
β-actin	Coturnix japonica	GTCCCTGTATGCCTCTGGTC	ATTTCTCTCTCGGCTGTGGT	204	AF199488.1
PGK1	Coturnix japonica	GAAGAAGGGAAAGGGAAGGA	CTACCATGGAGCTGTGAGCA	151	XM_015860450.1
TBP	Coturnix japonica	CTGCAACACCAGACTTGGAC	AGCCCAACAAAGCAACTCAC	116	XM_015857924.1
YWHAZ	Coturnix japonica	TGATGTGCTGTCTCTGTTGG	GCTTCTTGATACGCCTGTTG	171	XM_015856086.1

Abbreviation: RT-qPCR, reverse transcription quantitative polymerase chain reaction.

RNA-seq analysis

Total RNA was extracted and purified from embryos’ MDs using an RNeasy Plus Micro Kit (Qiagen, Venlo, Netherlands) or an RNeasy Plus Mini Kit (Qiagen) following the manufacturer’s protocol. For RNA-seq analysis, a total RNA sample was prepared by combining equal amounts of total RNA from four embryos [RIN value determined using Agilent TapeStation (Agilent Technologies, Santa Clara, CA, USA): ≥9.2], and three total RNA samples were analyzed in each group. Library preparation involved isolating poly(A) mRNA from the total RNA samples, followed by mRNA fragmentation using divalent cations and high temperature. First-strand and second-strand cDNA were synthesized using random primers. Purified double-stranded cDNA underwent end repair, dA-tailing, and T-A ligation to add adaptors to both ends. After size selection of adaptor-ligated DNA using DNA clean beads, each sample was PCR amplified with P5 and P7 primers, and the PCR products were validated. Libraries with different indexes were multiplexed and loaded on a NovaSeq 6000 system (Illumina, San Diego, CA, USA) for sequencing in a 2 × 150 paired-end configuration per the manufacturer’s instructions. Sequence data are publicly available at the Gene Expression Omnibus under the accession number GSE228535.

Cutadapt (v1.9.1) was employed to eliminate technical sequences, including adapters, PCR primers, fragments, and low-quality bases (<20) from the RNA-seq data. Hisat2 (v2.0.1) indexed the reference genome sequence of the Japanese quail (Coturnix japonica 2.1, GCF_001577835.2) and aligned clean data to the reference genome. Gene expression levels were estimated using HT-seq (v0.6.1), and fragments per kilobase per million reads (FPKM) were calculated based on read counts from HT-seq results. Differential gene expression analysis was executed using DESeq2 (V1.6.3), with differentially expressed genes (DEGs) identified using criteria: fold change >2 and p-value adjusted for multiple testing using Benjamini–Hochberg to estimate the false discovery rate (p_adj) <.05. Gene Ontology (GO) enrichment analysis of the DEGs was conducted using GOSeq (v1.34.1), and significantly enriched GO terms (p < .05) were determined. Results were represented using acyclic graphs plotted via TopGO (v2.18.0), with the top five enriched GO terms selected as primary nodes.

Statistical analysis

For statistical analyses, GraphPad Prism software version 9.4 (GraphPad, San Diego, CA, USA) was utilized. A chi-square test compared the survival rate of embryos and the prevalence of abnormal organ morphology between the control and EE₂-exposed groups. One-way analysis of variance (ANOVA) determined differences in the mRNA levels of the ovary and plasma hormone levels among experimental groups. Two-way ANOVA assessed the effects of laterality and treatment and the interaction between the main factors in the mRNA levels of the testes and female MDs of E15 embryos. Two-way ANOVA was also employed to determine the main effects of laterality on the experimental groups and the interaction between the main factors in the mRNA levels of the MDs of E12 embryos. When significant overall effects were found in one-way ANOVA and when the effects of the main factor with more than three levels or the effects of the interactions between main factors were significant in two-way ANOVA, Tukey’s test was performed as a post hoc analysis. An unpaired t-test evaluated differences in MD mRNA levels between EE60 males and females.

Results

Survival of the embryos

The survival of quail embryos cultured in surrogate eggshells remained unchanged after exposure to any EE₂ doses but decreased over time in all groups (Table 2). The survival rate in the EE₂-exposed groups did not significantly differ from that in the control group on the same day.

Table 2.

Survival of quail embryos cultured in a surrogate eggshell system

Embryonic day	Survival rate (%)
	Control (n = 48)	EE7 (n = 92)	EE20 (n = 60)	EE60 (n = 77)
E3	100.0	100.0	100.0	100.0
E9	75.0	77.2	78.3	80.5
E12	68.8	72.8	66.7	74.0
E15	62.5	57.6	58.3	61.0

The day on which egg incubation began was considered E0. Cultured embryos were treated with sesame oil (control) or sesame oil with ethynylestradiol at 7, 20, and 60 ng/egg (EE7, EE20, and EE60, respectively) from E3.

Reproductive organ morphology of E15 embryos

In the control group (vehicle-treated), female quail embryos exhibited a left ovary and MD, a right atrophic MD, and WDs on both sides (Figure 1A). Male embryos in the control group had testes and WDs on both sides but no MDs (Figure 1B). Abnormal development of reproductive organs was observed in both sexes in the EE₂-exposed groups. EE₂-exposed female embryos displayed hypertrophic MDs on both sides (Figure 1C) and MDs with torose structures (Figure 1D). In EE₂-exposed male embryos, an asymmetry in testis size was observed (Figure 1E), and residual MDs with cystic (Figure 1E), torose, and/or dropsical structures (Figs. 1F and 1G) were found. No abnormalities were found in the ovaries and WDs of both EE₂-exposed groups and the control group.

Figure 1.

Ventral view of internal reproductive organs of quail embryos. Photomicrographs of internal reproductive organs in a control female (A), control male (B) ethynylestradiol (EE₂)-exposed females (C and D), and EE₂-exposed males (E and F). Embryos were cultured in a surrogate eggshell culture system, exposed to vehicle (sesame oil) or sesame oil with EE₂ from E3, and scarified on E15 (day egg incubation began: E0). EE₂-induced hypertrophy (C) and torose structures (asterisks [*] in D) in the Müllerian ducts (MDs) of female embryos and the asymmetry of the testes (E) and persistence of MDs with cystic (E), dropsical (F), and/or torose structures in male embryos. Scale bars: 1 mm. Schematic illustrations of internal reproductive organs of female and male embryos treated with sesame oil (control) or sesame oil with EE₂ at 7, 20, or 60 ng/egg (EE7, EE20, and EE60, respectively) (G). 1, left ovary; 2, right mesonephros; 3, left mesonephros; 4, right Wolffian duct; 5, left Wolffian duct; 6, right MD; 7, left MD; 8, cloaca; 9, right testis; 10, left testis.

EE₂ exposure resulted in dose-dependent abnormal morphogenesis in the MDs and testes (Figure 1G and Tables 3 and 4), as shown by a significant increase in the prevalence of hypertrophy and torose structures in the left MD of EE60 female embryos (p < .01 and .05, respectively) and the prevalence of hypertrophy in the right MD of EE20 and EE60 female embryos (both p < .01) (Table 3). EE20 and EE60 male embryos also had significantly higher left-biased asymmetry in testis size (p < .05 and .01, respectively) and residual MDs on the left (both p < .01) and right (p < .05 and .01, respectively) sides than the control male embryos (Table 4). The prevalence of cystic structures in residual MDs on the left (p < .01 and .05, respectively) and right (both p < .05) sides was also significantly higher in the EE20 and EE60 groups, and the prevalence of torose structures in residual MDs on the left side was significantly higher (p < .05) in the EE60 group.

Table 3.

Prevalence of abnormal reproductive organs in female quail embryos on E15

Experimental groups	Abnormality in the left Müllerian duct (%)				Abnormality in the right Müllerian duct (%)
	Hypertrophic	Cystic	Torose	Dropsical	Hypertrophic	Cystic	Torose	Dropsical
Control (n = 15)	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
EE7 (n = 22)	18.2	0.0	18.2	9.1	13.6	4.5	4.5	0.0
EE20 (n = 24)	4.2	4.2	16.7	0.0	45.8**	0.0	16.7	0.0
EE60 (n = 21)	71.4**	0.0	23.8*	0.0	81.0**	0.0	19.0	0.0

Embryos were treated with sesame oil (control) or sesame oil with ethynylestradiol at 7, 20, and 60 ng/egg (EE7, EE20, and EE60, respectively) from E3.

*

p < .05 versus control.

**

p < .01 versus control.

Table 4.

Prevalence of abnormal reproductive organs in male quail embryos on E15

Experimental groups	Asymmetry in testis size (left > right) (%)	Abnormality in the left Müllerian duct (%)				Abnormality in the right Müllerian duct (%)
		Residual	Cystic	Torose	Dropsical	Residual	Cystic	Torose	Dropsical
Control (n = 15)	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
EE7 (n = 19)	5.3	10.5	10.5	0.0	0.0	10.5	10.5	0.0	0.0
EE20 (n = 12)	25.0*	41.7**	41.7**	0.0	0.0	25.0*	25.0*	0.0	0.0
EE60 (n = 29)	69.0**	93.1**	34.5*	27.6*	10.3	93.1**	34.5*	17.2	13.8

Embryos were treated with sesame oil (control) or sesame oil with ethynylestradiol at 7, 20, and 60 ng/egg (EE7, EE20, and EE60, respectively) from E3.

*

p < .05 versus control.

**

p < .01 versus control.

Tissue structure of gonads and MDs in E15 embryos

In female embryos, the development of the left ovary cortex containing oocytes was not significantly different between the control and EE60 groups (Figs. 2A and 2B). The left testis in all EE60 male embryos developed a cortex containing oocyte-like germ cells in addition to the medulla containing testicular cords, whereas the left testis in all control male embryos only developed the medulla containing testicular cords (Figs. 2C and 2D). The right testis in both the control and EE60 males had a thin cortex, with no significant differences between the two groups (Figs. 2E and 2F). The rate of ovotestis determined histologically was 0% in the left and right testes of the control males and in the right testis of EE60 males, while 100% in the left testis of EE60 males.

Figure 2.

Representative photomicrographs of the left ovary in a control female embryo (A), female embryo exposed to ethynylestradiol at 60 ng/egg (EE60) (B), left and right testes in a control male embryo (C and E), and EE60 male embryo (D and F) on E15. Scale bars: 50 µm. Co, cortex; T, testicular cords.

The right MD of control female embryos was small, and the tissues were composed of single or stratified columnar epithelium, subepithelial cell layer, and muscle cell layer from inside to outside (Figure 3A). The left MD of control female embryos had the same tissue structures as those in the right MD, but it developed and became large compared to the right MD (Figure 3B). The hypertrophic MDs of EE60 female embryos had thick subepithelial cell layer, and the tissues were folded in the lumen (Figs. 3C and 3D). The residual MDs of EE60 male embryos were large in size, but the tissues were thin with the same cell components as those in the control EE60 female embryos (Figs. 3E and 3F).

Figure 3.

Representative photomicrographs of the right and left Müllerian ducts (MDs) in a control female embryo (A and B), female embryo exposed to ethynylestradiol at 60 ng/egg (EE60) (C and D), and EE60 male embryo (E and F) on E15. Scale bars: 50 µm.

Gene expression in ovary and sex steroid levels in female E15 embryos

Of the five steroidogenic enzymes tested, only P450_scc was affected by EE₂. The P450_scc mRNA level in the left ovary differed significantly among the groups (F_{3, 20} = 5.720, p = .0054) and decreased significantly in EE60 female embryos (p = .0050) (Figure 3A). Although the mRNA level of P450_17α,lyase in the left ovary was significantly different among the groups (F_{3, 20} = 4.200, p = .0185), no significant difference was found between the control and EE₂-exposed groups (Figure 4B). The mRNA levels of 3β-HSD, 17β-HSD, and P450_arom in the left ovary did not change following EE₂ exposure (Figs. 4C–E).

Figure 4.

Gene expression levels of synthesizing and receiving gonadal hormones in the left ovary of female quail embryos on E15. The mRNA levels of P450_scc (A), P450_17⍺,lyase (B), 3β-HSD (C), 17β-HSD (D), P450_arom (E), ER⍺ (F), ERβ (G), AR (H), AMH (I), and AMHR2 (J) in the left ovary of female embryos treated with sesame oil (control) or sesame oil with ethynylestradiol at 7 ng/egg (EE7), 20 ng/egg (EE20), and 60 ng/egg (EE60) were determined by RT-qPCR analysis. Plasma concentrations of androgens (K) and estradiol (L) in female embryos of the control, EE7, EE20, and EE60 groups. Data are presented as mean ± standard error. The dots in the graphs represent individual data points (n = 6 in each group). Abbreviation: RT-qPCR, reverse transcription quantitative polymerase chain reaction.

EE₂ significantly affected the mRNA levels of ERα (F_{3, 20} = 4.064, p = .0209) and ERβ (F_{3, 20} = 3.281, p = .0422) but not AR in the left ovary (Figs. 4F–H). The ERα mRNA level increased in an EE₂ dose-dependent manner and was significantly elevated in EE20 (p = .0471) and EE60 (p = .0225) females (Figure 4F). Although the ERβ mRNA level appeared to increase with EE₂ exposure at any dose, a significant increase was observed only in EE7 females (p = .0300; Figure 4G).

The mRNA level of AMH in the left ovary differed significantly among the groups (F_{3, 20} = 4.054, p = .0211). AMH mRNA levels increased in an EE₂ dose-dependent manner, with significantly higher levels observed in EE60 females than in control females (p = .0291; Figure 4I). The mRNA level of AMHR2 also differed significantly among the groups (F_{3, 20} = 3.221, p = .0446), but no significant differences were found between the groups (Figure 4J). Plasma concentrations of androgens (Figure 4K) and estradiol (Figure 4L) in female embryos were not significantly affected by EE₂ exposure at any dose.

Gene expression in testes and sex steroid levels in male E15 embryos

The mRNA levels of steroidogenic enzymes, except P450_arom, were affected by EE₂ exposure in the testes, with prominent effects in the left testis. P450_scc and P450_{17α, lyase} mRNA levels were significantly affected by laterality (P450_scc: F_{1, 42} = 5.123, p = .0288; P450_{17α, lyase}: F_{1, 42} = 6.389, p = .0153), EE₂ treatment (P450scc: F_{3, 42} = 7.956, p = .0003; P450_{17α, lyase}: F_{3, 42} = 3.274, p = .0303), and the interaction between the main factors (P450_scc: F_{3, 42} = 4.590, p = .0072; P450_{17α, lyase}: F_{3, 42} = 4.997, p = .0047). The mRNA levels of P450_scc and P450_{17α, lyase} did not significantly differ between the left and right testes in control males. EE₂ administered at 60 ng/egg significantly increased the mRNA levels of P450_scc and P450_{17α, lyase} in the left but not right testis (P450_scc: p = .0001; P450_{17α, lyase}: p = .0022), resulting in significant differences between the left and right sides (P450_scc: p = .0032; P450_{17α, lyase}: p = .0010) (Figs. 5A and 5B). The 3β-HSD mRNA level was significantly affected by the interaction between laterality and treatment (F_{3, 42} = 2.949, p = .0436), with no asymmetric difference in control males and a significant asymmetric difference in EE60 males (p = .0424; Figure 5C). The 17β-HSD mRNA level was significantly affected by treatment (F_{3, 42} = 4.173, p = .0113), with a significant difference between the control and EE20 groups (p = .0066; Figure 5D). Additionally, the 17β-HSD mRNA levels were significantly higher in the right testis than in the left testis (F_{1, 42} = 7.516, p = .0089). P450_arom mRNA levels in the testes did not change with EE₂ exposure and did not show asymmetry (Figure 5E).

Figure 5.

Gene expression levels of synthesizing and receiving gonadal hormones in the testes of male quail embryos on E15. The mRNA levels of P450_scc (A), P450_17⍺,lyase (B), 3β-HSD (C), 17β-HSD (D), P450_arom (E), ER⍺ (F), ERβ (G), AR (H), AMH (I), and AMHR2 (J) in the left and right testes of male embryos treated with sesame oil (control) or sesame oil with ethynylestradiol at 7 ng/egg (EE7), 20 ng/egg (EE20), and 60 ng/egg (EE60) were determined by RT-qPCR analysis. Plasma concentrations of androgens (K) and estradiol (L) in male embryos of the control, EE7, EE20, and EE60 groups. Data are presented as mean ± standard error. The dots in graphs represent individual data points (n = 7 in control; n = 6 in EE7, n = 4–6 in EE20; n = 6 in EE60). Abbreviation: RT-qPCR, reverse transcription quantitative polymerase chain reaction.

EE₂ significantly increased mRNA levels of ERα (F_{3, 42} = 4.145, p = .0116), ERβ (F_{3, 42} = 11.37, p < .0001), and AR (F_{3, 42} = 6.420, p = .0011) in male embryo testes. Though the effect on ERα mRNA levels was weaker than on ERβ or AR mRNA levels, ERα mRNA levels increased significantly in EE7 males (p = .0206; Figure 5F). The ERβ mRNA level was significantly affected by laterality (F_{1, 42} = 10.24, p = .0026) and the interaction between laterality and treatment (F_{3, 42} = 11.51, p < .0001). There was no asymmetric difference in ERβ mRNA level in control males, but the left testis of EE60 males exhibited a significantly higher ERβ mRNA level (p < .0001) than the right testis of EE60 males and the left testis of control males (Figure 5G). Additionally, AR mRNA levels were significantly increased in EE60 males (p = .0019; Figure 5H).

The AMH mRNA level was significantly higher in the right testis than in the left testis (F_{1, 42} = 15.70, p = .0003), but no significant effect of EE₂ was observed (Figure 5I). The AMHR2 mRNA level in the testes was significantly affected by EE₂ (F_{3, 42} = 5.141, p = .0041), and it increased significantly in EE60 males (p = .0019; Figure 5J). EE₂ exposure did not affect the plasma concentrations of androgens (Figure 5K) or estradiol (Figure 5L) in male embryos.

Gene expression in MDs of E15 embryos

Overall, EE₂ had no significant effect on the mRNA levels of ERα (Figure 6A), ERβ (Figure 6B), AR (Figure 6C), and AMHR2 (Figure 6D) in the MDs of female embryos. Although a two-way ANOVA showed a significant effect of EE₂ on the ERα mRNA level (F_{3, 41} = 3.025, p = .0403), post hoc analysis did not reveal any significant effect. The mRNA levels of ERβ and AR were significantly higher on the left side than on the right side of the MDs in female embryos (ERβ: F_{1, 34} = 8.447, p = .0064; AR: F_{1, 41} = 8.768, p = .0051). In some female embryos, the ERβ mRNA levels were below the minimum detection level in RT-qPCR analysis (n = 2 on the left of the control and on the right of EE20; n = 1 on the right of the control and EE20 and on the left of EE60) and were excluded from the statistical analysis.

Figure 6.

Gene expression levels of gonadal hormone receptors in the Müllerian ducts (MDs) of quail embryos on E15. The mRNA levels of ER⍺ (A and E), ERβ (B and F), AR (C and G), and AMHR2 (D and H) in the left and right MDs of female embryos treated with sesame oil (control) or sesame oil with ethynylestradiol at 7 ng/egg (EE7), 20 ng/egg (EE20), and 60 ng/egg (EE60) (A–D) and EE60 male embryos (E–H) were determined by RT-qPCR analysis. Data are presented as mean ± standard error. The dots in graphs represent individual data points (n = 4–6 in control females; n = 3–7 in EE7 females; n = 5–6 in EE20 females; n = 5–6 in EE60 females; and n = 6 in EE60 males). Abbreviation: RT-qPCR, reverse transcription quantitative polymerase chain reaction.

Most male embryos in the EE60 group (>90%) did not experience MD atrophy (Table 4). Gene expression in the MDs of EE60 male embryos was analyzed, revealing the expression of ERα (Figure 6E), ERβ (Figure 6F), AR (Figure 6G), and AMHR2 (Figure 6H), with levels not significantly different from those in EE60 females.

Reproductive organ morphology in E12 embryos

Female embryos exposed to EE₂ exhibited malformations in MDs at E12. Most EE60 female embryos showed hypertrophied MDs on both sides, with a significantly higher prevalence compared to control females (p < .01; Table 5). The prevalence of hypertrophied right-side MDs with torose structures significantly increased following EE₂ exposure (p < .01). However, the morphology of the left ovary and WDs on both sides did not differ between control and EE60 females.

Table 5.

Prevalence of abnormal reproductive organs in female quail embryos on E12

Experimental groups	Abnormality in the left Müllerian duct (%)				Abnormality in the right Müllerian duct (%)
	Hypertrophic	Cystic	Torose	Dropsical	Hypertrophic	Cystic	Torose	Dropsical
Control (n = 14)	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
EE60 (n = 14)	92.9**	7.1	21.4	0.0	85.7**	14.3	57.1**	14.3

Embryos were treated with sesame oil (control) or sesame oil with ethynylestradiol at 60 ng/egg (EE60) from E3.

**

p < .01 versus control.

Males in the control groups did not have any MDs, while all males in the EE60 group had MDs on both sides (Table 6). Half of the EE60 males had cystic MDs, and more than half exhibited left-biased asymmetry in testis size. The percentages of cystic structures in residual MDs and asymmetric testis size were significantly higher in EE60 males (p < .01). Meanwhile, no abnormality was observed in the morphology of the WDs.

Table 6.

Prevalence of abnormal reproductive organs in male quail embryos on E12

Experimental groups	Asymmetry in testis size (left > right) (%)	Abnormality in the left Müllerian duct (%)				Abnormality in the right Müllerian duct (%)
		Residual	Cystic	Torose	Dropsical	Residual	Cystic	Torose	Dropsical
Control (n = 12)	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
EE60 (n = 14)	64.3**	100.0**	50.0**	0.0	0.0	100.0**	50.0**	0.0	0.0

Embryos were treated with sesame oil (control) or sesame oil with ethynylestradiol at 60 ng/egg (EE60) from E3.

**

p < .01 versus control.

DEGs in MDs of female embryos

As mentioned before, the morphology of the MDs exhibited asymmetry in control female embryos. This asymmetry was completely disappeared in EE60 female embryos, which was due to MD hypertrophy on both sides, but not in EE7 and EE20 female embryos. Because of no effects of EE₂ on sex steroid receptor mRNA levels in the MDs, the molecular mechanisms responsible for asymmetric MD morphogenesis and EE₂-induced disruption of asymmetric MD morphology remained unclear. In this study, we then performed RNA-seq analysis of the MDs from control and EE60 female embryos. As a result, in control female embryos, 1,366 DEGs were identified by comparing gene expression in the left and right MDs (Figure 7 and Supplementary Table 1). Of these DEGs, 709 genes were upregulated, and 657 genes were downregulated in the left MD relative to the right MD. In contrast, in EE60 female embryos, only 21 DEGs were found between the left and right MDs (Figure 7 and Supplementary Table 2). Among these, five genes were upregulated, and 16 genes were downregulated in the left MD relative to the right MD.

Figure 7.

Asymmetric gene expression in the Müllerian ducts (MDs) of female embryos was eliminated by ethynylestradiol (EE₂) exposure. Differentially expressed genes between left and right MDs in female embryos treated with vehicle (control) or EE₂ at 60 ng/egg (EE60) were determined by RNA-seq analysis, and the number of the genes is illustrated. More details are shown in Supplementary Table 1 and Table 2.

GO enrichment of DEGs in MDs of female embryos

In control female embryos, DEGs in MDs were associated with 180 significantly enriched GO terms (Supplementary Table 3). Among the top 30 most enriched GO terms, more DEGs were assigned to biological process GO terms than to molecular function and cellular component GO terms (90, 13, and 74 DEGs, respectively) (Figure 8A). The 10 most enriched biological process GO terms included DNA replication initiation (GO:0006270), cell division (GO:0051301), positive regulation of phosphatidylinositol 3-kinase (PI 3-kinase) signaling (GO:0014068), mitotic chromosome condensation (GO:0007076), and microtubule-based movement (GO:0007018) (Figure 8B and Supplementary Table 4).

Figure 8.

Gene Ontology (GO) enrichment of differentially expressed genes (DEGs) between the left and right Müllerian ducts of control female embryos. The number of DEGs (A) and GO enrichment p-values (B) are shown for the top 30 GO terms. Additional details are shown in Supplementary Table 3 and Table 4.

Figure 9 shows a directed acyclic graph representing the hierarchical relationship among the enriched biological process GO terms of the DEGs in the MDs of control female embryos. The top five enriched GO terms, set as the primary nodes in the graph, were DNA replication initiation (GO:0006270), cell chemotaxis (GO:0060326), cell cycle process (GO:0022402), cell cycle (GO:0007049), and cell division (GO:0051301).

Figure 9.

Directed acyclic graph showing the hierarchical relationship among the enriched biological process Gene Ontology (GO) terms of differentially expressed genes in the Müllerian ducts of control female embryos. Arrows represent inclusion relationships, and the functional range is defined from the top to the bottom in descending order. The five enriched GO terms (GO:0007049, GO:0051301, GO:0022402, GO:0060326, and GO:0006270) were selected as primary nodes, and associated GO terms are shown through the inclusion relationship. The colors of boxes represent the relative significance of the GO enrichment, ranging from dark (most significant) to light (least significant).

Effects of EE₂ exposure on differential gene expression in MDs

To validate the results of RNA-seq analysis and investigate the effects of EE₂ on gene expression in MDs of female and male embryos, RT-qPCR analyses were conducted. Twelve genes associated with the top two enriched biological process GO terms, DNA replication initiation (GO:0006270) and cell division (GO:0051301), were selected as genes of interest (Supplementary Table 4): cyclin E2 (CCNE2), minichromosome maintenance complex (MCM) components 2, 3, 4, and 6 (MCM2, MCM3, MCM4, and MCM6), DNA polymerase alpha 1, catalytic subunit (POLA1), cyclin A2 (CCNA2), cyclin B2 (CCNB2), cyclin B3 (CCNB3), centromere protein H (CENPH), NUF2 component of NDC80 kinetochore complex (NUF2), and structural maintenance of chromosomes 2 (SMC2).

In RNA-seq analysis of control female embryos, the FPKM values of the 12 selected genes were higher in the left MD than in the right MD (Supplementary Table 1). RT-qPCR analyses confirmed higher mRNA levels of these genes in the left MD compared to the right MD in control female embryos (Figure 10). However, no left-biased asymmetry in mRNA levels was detected in EE60 female embryos. MDs of EE60 male embryos expressed the 12 selected genes at levels similar to those of EE60 female embryos. Interestingly, sex differences were observed in the mRNA levels of CCNA2, CCNB3, CENPH, NUF2, and SMC2, with higher expression in EE60 males than in EE60 females.

Figure 10.

Effects of ethynylestradiol (EE₂) exposure on the expression of genes assigned to the top two enriched biological process Gene Ontology (GO) terms: DNA replication initiation (GO:0006270) and cell division (GO:0051301). The mRNA levels of CCNE2 (A), MCM2 (B), MCM3 (C), MCM4 (D), MCM6 (E), POLA1 (F), CCNA2 (G), CCNB2 (H), CCNB3 (I), NUF2 (J), CENPH (K), and SMC2 (L) in the left and right Müllerian ducts of female embryos treated with vehicle (control female) or EE₂ at 60 ng/egg (EE60 female) and male embryos treated with EE₂ at 60 ng/egg (EE60 male) were determined by RT-qPCR analysis. Data are presented as mean ± standard error. The dots in the graphs represent individual data points (n = 6 for each group). Abbreviation: RT-qPCR, reverse transcription quantitative polymerase chain reaction.

Two-way ANOVA results indicated that the mRNA levels of the 12 selected genes, except CENPH, were significantly affected by the interaction of main factors (CCNE2: F_{2, 30} = 12.94, p < .0001; MCM2: F_{2, 30} = 11.74, p = .0002; MCM3: F_{2, 30} = 7.211, p = .028; MCM4: F_{2, 30} = 9.607, p = .0006; MCM6: F_{2, 30} = 9.341, p = .0007; POLA1: F_{2, 30} = 7.373, p = .0025; CCNA2: F_{2, 30} = 5.198, p = .0115; CCNB2: F_{2, 30} =6.309, p = .0052; CCNB3: F_{2, 30} = 5.503, p = .0092; NUF2: F_{2, 30} = 5.679, p = .0081; SMC2: F_{2, 30} = 3.465, p = .0443). In control females, post hoc tests revealed that the mRNA levels of the following genes were higher in the left MD than in the right MD: CCNE2 (p < .0001; Figure 10A), MCM2 (p < .0001; Figure 10B), MCM3 (p = .0002; Figure 10C), MCM4 (p < .0001; Figure 10D), MCM6 (p < .0001; Figure 10E), POLA1 (p = .0006; Figure 10F), CCNA2 (p = .0030; Figure 10G), CCNB2 (p = .0006; Figure 10H), CCNB3 (p = .0070; Figure 10I), and NUF2 (p = .0442; Figure 10J). Although the mRNA levels of CENPH (Figure 10K) and SMC2 (Figure 10L) were also higher in the left MD in control females, it was not significantly different to the right MD.

In EE60 female embryos, the mRNA levels of the 12 selected genes in the left and right MDs did not significantly differ. Compared with the right MD of control females, the right MD of EE60 females had significantly higher mRNA levels of CCNE2 (p = .0035; Figure 10A), MCM2 (p = .0021; Figure 10B), MCM3 (p = .0427; Figure 10C), MCM4 (p = .0116; Figure 10D), and CCNB2 (p = .0173; Figure 10H). Similar to EE60 females, EE60 males exhibited no asymmetric difference in the mRNA levels of the 12 selected genes between the left and right MDs. However, compared with the right MD of EE60 females, the right MD of EE60 males had significantly higher mRNA levels of CCNA2 (p = .0454; Figure 10G), CCNB3 (p = .0326; Figure 10I), and NUF2 (p = .0492; Figure 10J). In both MDs, EE60 males showed significantly higher mRNA levels of CENPH (p < .0001; Figure 10K) and SMC2 (left: p = .0250, right: p = .0006; Figure 10L) than EE60 females.

Discussion

EE₂ exposure resulted in ovary-like tissue formation in the left testis, MD hypertrophy in females, and retention of MDs in males of quail embryos. These phenomena have been demonstrated previously (Berg et al., 1999), although the molecular mechanisms remained unknown. We here showed that the genes for certain steroidogenic enzymes and ERβ are highly expressed in the left testis, but not in the right testis, suggesting partial feminization of the left testis at the molecular level, which is supported by previous findings that steroidogenic enzyme and sex steroid receptor expression is higher in the ovary than in the testis (Nishikimi et al., 2000; Tsukahara et al., 2021). We further observed that female MDs asymmetrically express over a thousand genes, a pattern significantly reduced by EE₂ exposure. The left MD of female embryos highly expressed genes involved in cell cycle progression and oncogenesis compared to the right MD, but EE₂ equalized gene expression levels between the two sides. In EE₂-exposed males, these genes were overexpressed in MDs on both sides. EE₂ exposure during the embryonic period induces the retention of the right oviduct and contraction of the left oviduct, impairing egg-laying performance in adulthood (Berg et al., 2001b). Our findings contribute to understanding the molecular mechanisms of xenoestrogen-induced abnormalities in avian reproductive organs and developing the adverse outcome pathways, where in ovo exposure to endocrine-disrupting chemicals impairs female avian reproduction.

In female quail embryos, MDs express ERα and ERβ (Mattsson et al., 2008b; Tsukahara et al., 2021). Quail embryos treated with an ERα agonist exhibit abnormally developed MDs, which can be reversed by an ERα antagonist (Mattsson et al., 2008b). However, an ERβ agonist does not induce MD malformation (Mattsson and Brunström, 2017). Therefore, xenoestrogens and estrogenic chemicals may exert their toxicity primarily by binding to ERα. In chicken embryos, estrogen-binding capacity is greater in the left MD during E8–9 (Ha et al., 2004; MacLaughlin et al., 1983), suggesting higher expression levels and/or ligand-binding affinity for ERs in the left MD. However, in quail embryos, ER expression in MDs does not show laterality during E9–12 (Tsukahara et al., 2021). Left-biased asymmetry in ER expression emerges at E15, possibly due to differences in cell activity between developing and atrophied ducts. Asymmetric transcriptional mechanisms driven by ERs may be responsible for asymmetric MD development in quail. Our RNA-seq analysis revealed that 1,366 genes were differently expressed in the left and right MDs of E12 female embryos. Following EE₂ exposure, the number of DEGs dramatically decreased to only 21, aligning with the morphological shift of MDs from asymmetric to symmetric patterns induced by EE₂. Among the 1,366 DEGs were genes involved in the molecular mechanisms underlying the asymmetric morphogenesis of MDs. However, what genes of the DEGs are required for asymmetric MD development remains undetermined in this study. Further studies are needed for identifying the genes and their roles in asymmetric MD development.

Our GO analysis revealed that many DEGs in female MDs are significantly enriched in the GO terms “DNA replication initiation” (GO:0006270), “cell division” (GO:0051301), “mitotic chromosome condensation” (GO:0007076), “positive regulation of angiogenesis” (GO:0045766), “mitotic nuclear division” (GO:0007067), “positive regulation of cell migration” (GO:0030335), and “regulation of cyclin-dependent protein serine/threonine kinase activity” (GO:0000079). These GO terms are associated with critical events in organogenesis, including cell proliferation, cell migration, cell differentiation, and angiogenesis. Most genes assigned to these GO terms were more expressed in the left MD than in the right MD (see Supplementary Table 4), indicating active organization of the left MD with higher organogenesis-related gene expression. The hierarchical relationship among the enriched GO terms indicated that the biological processes involved in the DEGs eventuated in DNA replication initiation (GO:0006270), cell chemotaxis (GO:0060326), and cell cycle process (GO:0022402). Additionally, RT-qPCR analysis of DEGs assigned to the GO terms “DNA replication initiation” (GO:0006270) and “cell division” (GO:0051301) ensured the validity of RNA-seq data. Specifically, both RNA-seq and RT-qPCR analyses revealed that CCNA2, CCNB2, CCNB3, CCNE2, CENPH, MCM2, MCM3, MCM4, MCM6, POLA1, NUF2, and SMC2 were more highly expressed in the left MD than in the right MD. These genes encode proteins that play roles in promoting cell cycle progression and oncogenesis (Chu et al., 2021; Fu and Shao, 2016; Han et al., 2020; Parker et al., 2017; Roskoski, 2019; Shigeishi et al., 2006; Yam et al., 2002; Yoon et al., 2018), suggesting higher cell proliferation activity in the left MD. Consequently, the left MD grows, while the right MD becomes atrophic. Interestingly, EE₂ increased gene expression in the right MD, reaching levels comparable to those in the left MD. This supports the idea that the right MD, typically fated to regress during the embryonic period, develops with increased cell proliferation activity following EE₂ exposure, as indicated by our histological study. Similarly, male MDs may enlarge due to increased cell proliferation activity induced by EE₂, as it upregulates the expression of genes promoting cell cycle progression. For 50% of the identified genes (CCNA2, CCNB2, CCNB3, CENPH, NUF2, and SMC2), EE₂-induced expression levels were higher in males than in females, suggesting higher cell proliferation activity in MDs following EE₂ exposure in males. However, the residual MDs of EE₂-exposed male embryos did not have hypertrophic tissues. This paradox could be attributed to higher levels of AMH being produced in the testes than in the ovaries. Therefore, the increase in cell proliferation activity is higher in males, overcoming the inhibitory effects of AMH on cell growth. AMH mRNA levels are higher in the testes than in the ovaries in quail and chicken embryos (Nishikimi et al., 2000; Tsukahara et al., 2021), suggesting differing molecular mechanisms underlying EE₂-induced MD abnormalities between sexes.

The exact molecular mechanisms governing the development of the left MD and regression of the right MD, despite equal levels of ER expression on both sides, remain incompletely understood. In female embryos, MD morphology asymmetry is evident until E12, where the left duct elongates while the right duct shortens (Tsukahara et al., 2021). Our GO analysis highlighted that DEGs in the MDs at E12 were significantly enriched in the GO term “positive regulation of PI 3-kinase signaling” (GO:0014068). PI 3-kinase pathway activates ERα, even in the absence of ligands (Campbell et al., 2001), and the activated ERα regulates transcription by recruiting coactivator proteins and interacting with transcriptional machinery (Glass and Rosenfeld, 2000). Asymmetric ERα activation through the PI 3-kinase pathway may lead to asymmetric gene expression and morphogenesis. EE₂ exposure resulted in the exclusion of genes related to the PI 3-kinase signaling pathway from DEGs, suggesting that exogenously administered EE₂ disrupts this asymmetry. Further investigations are needed to elucidate the precise mechanisms by which EE₂ eliminates the asymmetry of gene expression in MDs.

This study reveals that the left testis in male quail embryos is more susceptible to the endocrine-disrupting actions of EE₂ than the right testis. Exposure to EE₂ at a dose of 60 ng/egg transformed testicular tissues partly into ovary-like tissues, increasing the mRNA levels of certain steroidogenic enzymes (P450_scc, P450_{17α, lyase}, and 3β-HSD) and ERβ in the left testis but not in the right testis, indicating partial feminization at the molecular expression level. The increased gene expression observed in this study may result from the effects of EE₂ on the earlier development of the left gonad. PITX2, encoding a homeobox transcription factor, is expressed in the left gonad but not in the right gonad, influencing the left gonad’s development into the ovary by affecting the expression of downstream genes like RALDH2, SF-1, cyclin D1, and ERα, all exhibiting an asymmetric expression pattern (Guioli et al., 2014; Ishimaru et al., 2008; Monsoro-Burq and Levin, 2018). These asymmetrically expressed genes may be targets of EE₂ toxicity. Further investigations are needed to clarify the molecular mechanisms of ovotestis formation.

We reported the effects of o,p'-dichloro-diphenyl-trichloroethane (o,p'-DDT) on reproductive organ development in quail embryos (Win-Shwe et al., 2023). Similar to EE₂, o,p'-DDT dose-dependently induced MD hypertrophy in both sexes and ovotestis formation in the left gonad of male embryos. However, the effects on gene expression in MDs and the left testis differed between o,p'-DDT and EE₂. o,p'-DDT exhibits higher estrogenic activity than p,p'-DDT and its metabolites (Bolger et al., 1998; Kanno et al., 2003; Kojima et al., 2004). However, DDT isomers and metabolites show lower estrogenic activity and ER-binding affinity than estradiol (Chen et al., 1997). The molecular mechanisms by which EE₂ induces abnormal organ development in quail embryos are likely different from those of o,p'-DDT.

The Organization for Economic Co-operation and Development (OECD) has established an avian reproduction testing method called Test Guideline 206 (TG206). Despite efforts to improve data quality and reduce the test duration, revisions to TG206 were discontinued in 2005 and the test has since remained unchanged. Given the 3Rs principle (replacement, reduction, and refinement) in animal experimentation, there is a need to revise animal tests and develop alternative methods. Our study using fertilized eggs would provide useful information to develop alternative methods to screen and estimate the impact of endocrine-disrupting chemicals on avian reproduction.

Supplementary data

Supplementary data are available at Toxicological Sciences online.

Declaration of conflicting interests

The authors declare no competing financial interests.

Funding

Environmental Research and Technology Development Fund (JPMEERF20225001) of the Environmental Restoration and Conservation Agency, Japan.

Acknowledgments

We thank Enago (https://www.enago.jp/) for the English language review of this manuscript.