Green tea extract disrupts gonadal differentiation in a model avian system

Abstract

In the poultry industry, green tea extract (GTE) is becoming an increasingly popular dietary supplement due to its observed beneficial effects on performance and egg quality. However, work in mammals indicates that the active ingredients in green tea extract, catechins, act as endocrine disruptors, interfering with the actions of estrogen. Given that GTE is often supplemented to hens during egg-laying, there is potential for the endocrine-disrupting properties of catechins to interfere with the sensitive process of sexual differentiation in embryos. We tested whether green tea extract (GTE) alters gonadal differentiation in Japanese quail (Coturnix japonica) embryos. We injected eggs with either GTE or a Control vehicle before the onset of differentiation on the 4th day of incubation. On day 15, we evaluated phenotypic sex (gonadal morphology) and genetic sex (via PCR). Embryos from eggs that were injected with GTE had increased genotype-phenotype mismatches, which occurred exclusively in genetic females (ZW) who developed bilateral gonads resembling testes. Among genetic females, GTE significantly increased the probability of a male phenotype. Treatment with GTE did not increase embryonic mortality compared to Controls. We conclude that GTE modifies the developmental trajectory of embryonic gonads, specifically for female embryos. We propose that GTE disrupts the critical estrogen-dependent window of female sex differentiation, leading to sex reversal.

Introduction

Plant-derived compounds intended to support health and performance are popular additions to the diets of captive birds. These phytochemicals, such as carotenoids and polyphenols, are bioactive plant metabolites that frequently exhibit antioxidant, antimicrobial, and antiviral properties (reviewed in Kumar et al., 2023). In poultry, phytogenic compounds are increasingly used to improve production traits or to provide alternatives to antibiotics, and this has been reviewed considerably in recent years (e.g., as in Abdelli et al., 2021; Aminullah et al., 2025). Notably, a subset of phytochemicals can interact with estrogen pathways, raising the possibility of endocrine effects in the birds that consume them. For example, phytoestrogenic food additives can influence endocrine and reproductive parameters in both broilers and layers (Heng et al., 2017; Saleh et al., 2019; Qiang et al., 2023), as well as in other production birds such as quail, ducks, and geese (Zhao et al., 2005; Rochester et al., 2009; Zhao et al., 2013). Even beyond managed systems, wild birds consume phytoestrogens under natural conditions, and this can modulate their reproductive physiology (reviewed in Rochester and Millam, 2009). Both the rising use of phytogenic compounds in production animals and natural exposure in free-living birds warrant closer evaluation of compounds with known endocrine activity. One such compound that reflects this is green tea extract (GTE). GTE has drawn interest as a feed additive for layers, broilers, and quail alike, due to positive effects on health, laying performance, and egg yolk antioxidant and lipid profiles (reviewed in Seidavi et al., 2020; Mahlake et al., 2022), however its active ingredients may also exert endocrine-disrupting effects.

The effects of green tea are often attributed to its polyphenol constituents, catechins, which are present in high amounts in its leaves (reviewed in Zhao et al., 2022; Radeva-Ilieva et al., 2025). The principal catechin in green tea is epigallocatechin gallate (EGCG), but epigallocatechin (EGC), epicatechin gallate (ECG), and epicatechin (EC) are also abundant (Graham, 1992). In addition to the other biological effects of these catechins, some of their structures allow them to interact with estrogen receptors and influence estrogenic activity (reviewed in Kiyama, 2020). Mechanistically, catechins can act in many ways to influence endocrine pathways, with either agonistic or antagonistic effects, depending on dose, tissue type, presence of endogenous hormones, and other factors (Kuruto-Niwa et al., 2000). One way that catechins may disrupt estrogen signaling is through direct actions on estrogen receptors, as well as on the transcriptional and epigenetic factors that regulate their expression and function (Kuruto-Niwa et al., 2000; Goodin et al., 2002; Belguise et al., 2007; Li et al., 2010). Another way is indirectly through their inhibitory effects on steroidogenic enzymes, such as aromatase (CYP19A1), a cytochrome P450 enzyme that converts androgens to estrogens (Satoh et al., 2002; Monteiro et al., 2006; aromatase reviewed in Simpson et al., 1994).

It is well-known that chemicals that inhibit estrogen signaling at least partially masculinize female avian embryos (reviewed in Brunström et al., 2003; Major and Smith, 2016; Zhang et al., 2023). Unlike mammals, where gonadal differentiation proceeds largely independent of sex steroids, estrogens are an absolute requirement for these processes in birds (reviewed in Nicol et al., 2022). In female embryos, high aromatase expression leads to increased concentrations of estrogens that then bind to estrogen receptors expressed asymmetrically in the gonads of both males and females (Scheib et al., 1985; Andrews et al., 1997; Nakabayashi et al., 1998; Tsukahara et al., 2021). This triggers differentiation of the left gonad into the ovary, whereas in males, low aromatase activity limits estrogen synthesis, leading to the development of paired testes (Yoshida et al., 1996; Ishimaru et al., 2008; Guioli et al., 2014). As a result, chemicals that disrupt estrogenic activity, such as aromatase inhibitors, can prevent ovarian development, leading to the development of bilateral testes instead (Elbrecht and Smith, 1992; Wartenberg et al., 1992; Vaillant et al., 2001; Koba et al., 2008).

There is now limited evidence that plant and food-based chemicals that disrupt estrogen signaling can also interfere with sexual differentiation during embryogenesis. For example, tomato and garlic extracts increased the proportion of male chicks at hatch, in some cases with efficacy similar to synthetic aromatase inhibitors like fadrozole, and often without adverse effects on hatchability (Fazli et al., 2015; Jamshasb and Mottaghitalab, 2019; Abdulateef et al., 2021). One study indicates that GTE may have similar effects in avian embryos: chicken eggs injected with GTE produced 80% males at hatch; however, whether this bias reflected sex reversal versus sex-specific mortality was unclear (Hadibigloo et al., 2018). To explore further explore this, we tested whether in ovo exposure to GTE altered gonadal sex differentiation in Japanese quail (Coturnix japonica). The effects of estrogen and aromatase inhibitors on sexual differentiation are well-documented in Japanese quail embryos (Koba et al., 2008). Yet, while studies are exploring the use of GTE in quail (Kara et al., 2016; AL-Hamed, 2020; Kamil et al., 2021; Ismael and Ameen, 2022), its impacts on sexual differentiation processes in this species have never been tested. To test whether GTE interferes with the process of gonadal differentiation, we injected eggs with either a GTE or Control solution on the 4th day of incubation, prior to when gonadal differentiation begins. We then incubated eggs for an additional 11 days, after which we quantified phenotypic and genotypic sexes of the embryos. We predicted that injections of GTE into quail eggs would disrupt sexual differentiation within female embryos, resulting in a higher proportion of embryos with male gonads, but that genetic sex ratios would be equivalent between the two treatment groups.

Materials and methods

Egg injections

For this experiment, we obtained (n = 110) eggs from a breeding flock of Japanese quail (Coturnix japonica) maintained by the Department of Poultry Science at the University of Georgia. We collected eggs daily and stored them in an egg cooler at ∼18.9°C, which was within a common temperature range for storing hatching eggs (Hassan and Alsattar, 2015; Hester, 2017). Eggs were kept in the cooler for no longer than 7 days and incubation was initiated simultaneously for all eggs. We incubated eggs in a NatureForm incubator (cat. # NMC-2000, Pas Reform North America LLC [formerly NatureForm Hatchery Technologies], Jacksonville, FL) at 37.2°C and 40% relative humidity. The Japanese quail incubation period is around 16 to 17 days (Ainsworth et al., 2010) and detailed embryonic staging of this species allowed for precise timing of our treatment (Fig. 1). Genital ridges appear by the 3rd day of incubation, bipotential gonads are formed on the 4th day, gonads begin to differentiate on the 5th day, and early sex differences in the gonads are observable by the 7th day (Rong et al., 2011; Chang et al., 2012; Intarapat and Satayalai, 2014). On the 4th day of incubation, prior to the onset of gonadal differentiation in quail embryos, we divided eggs into two groups: GTE and Control. We injected GTE eggs with 50 uL of a 50% glycerol:water solution containing 300 ug of GTE. Control eggs were injected with 50 uL of the 50% glycerol:water vehicle only. We injected into the small ends of eggs using 0.5 mL insulin syringes, then sealed the injection hole with ethyl cyanoacrylate gel (cat. # KG867, Krazy Glue, West Jefferson, OH). The quantity of GTE was determined by adjusting the dose that triggered sex ratio skews in chicken eggs (500 ug, Hadibigloo et al., 2018) for the smaller size of quail eggs. We created the GTE solution on the day of injections by first dissolving the GTE (cat. # 00954, Decaffeinated Mega Green Tea Extract, Life Extension, Fort Lauderdale, FL) in glycerol, then adding an equal volume of water and heating the solution for ∼10 minutes. This protocol prevented the GTE from precipitating out of solution. In total, we injected 53 eggs with GTE and 57 eggs with the Control vehicle. We then placed eggs back into the incubator until day 15 of incubation. At that point, we euthanized embryos via CO₂ asphyxiation then performed dual sex diagnostics to evaluate phenotypic and genotypic sex. Of the injected eggs, 43 in each group were fertile and fully developed by the 15th day of incubation.

Fig. 1.

Theoretical framework and experimental design. We hypothesized that embryonic exposure to GTE would impair normal ovarian development in genetic females and bias development toward testes. We tested this by injecting eggs (n = 110) with either GTE or a Control vehicle on day 4 of incubation, when gonads were still bipotential and just before the onset of differentiation. In Japanese quail, early sex differences are detectable by day 7 and incubation lasts ∼16 to 17 days. We sexed embryos on day 15 both phenotypically (using gonadal morphology) and genotypically (using molecular analysis of sex chromosomes). We then compared sex ratios and phenotype:genotype mismatches between the GTE and the Control group.

Embryonic sexing

We visually inspected embryonic gonads via dissection and determined phenotypic sex based on the presence of one (female) or two (male) gonads (Fig. 2a), then collected liver tissue from each embryo. Of the 43 developed embryos in each group, we successfully dissected and identified gonads for 35 embryos from the GTE group and 37 embryos from the Control group. We performed molecular sexing using the liver samples from embryos to determine genotypic sex. We extracted DNA from embryonic tissue using a standard salt extraction method (Lambert et al., 2000). We then amplified portions of the CHD1-Z and CHD1-W alleles from sex chromosomes using polymerase chain reaction (PCR) with three sexing primers: ZF (5ʹ-CTCTGGGTTTTGACTGTATTG-3ʹ), WF (5ʹ-CATCTGTTTTCCCCCCCAAA-3ʹ), and P2 (5ʹ-TCTGCATCGCTAAATCCTTT-3ʹ). This methodology was adapted from Coustham et al. (2017), which discriminates males and females based on the presence of one or two amplicons, respectively (Fig. 2b). Each reaction contained 4 uL of 5 × Platinum II PCR Buffer (supplied with Platinum II Taq Hot-Start DNA Polymerase, cat. # 14966-001, Invitrogen, Thermo Fisher Scientific Inc., Waltham, MA), 0.4 uL of dNTP mix 10 mM each (cat. # R0192, Thermo Fisher Scientific Inc., Waltham, MA.), 0.4 uL of each 10 uM primer (custom DNA oligonucleotides, Integrated DNA Technologies Inc., Coralville, IA), 0.16 uL of Platinum II Taq Hot-Start DNA Polymerase (cat. # 14966-001, Invitrogen, Thermo Fisher Scientific Inc., Waltham, MA), at least 100 ng of template DNA, and nuclease-free water (cat. # 351-068-131, Quality Biological Inc., Gaithersburg, MD) to bring the reaction up to a total volume of 20 uL. The PCR program was run in an iCycler Thermal Cycler (cat. # 170-8720, Bio-Rad Laboratories Inc., Hercules, CA) and included 1 cycle of initial denaturation at 95°C (30 s), 35 cycles at 94°C (45 s), 58°C (45 s), and 72°C (45 s), a final extension cycle at 72°C for 5 min, and a holding period at 4°C to keep the samples cold until they were used for gel electrophoresis. We visualized PCR products on a 3% agarose gel stained with ethidium bromide in a Sub-Cell GT horizontal electrophoresis system (Bio-Rad Laboratories Inc., Hercules, CA) hooked up to a PowerPac Basic power supply (cat # 164-5050, Bio-Rad Laboratories Inc., Hercules, CA). We were unable to molecularly sex all embryos that we had previously phenotypically sexed due to variations in sample tissue quantity or quality. In total, we successfully PCR-sexed 33 embryos from the GTE group and 34 embryos from the Control group, as well as one adult male and female quail for validation and comparison.

Fig. 2.

Embryonic sexing on the 15th day of incubation.(a) Phenotypic sex was determined using the visual presence of either a pair of testes (left, male) or a single ovary (right, female). (b) Genotypic sex was determined molecularly via PCR using the presence of either one amplicon (male) or two amplicons (female).

Statistical analysis

We conducted our analyses using R Statistical Software (v 4.3.3; R Core Team, 2024) in RStudio (v 2025.09.0-387; Posit Team, 2025). We used packages from tidyverse (Wickham et al., 2019) for data wrangling. Figures were created using ggplot2 (Wickham, 2016), ggpattern (Mike and Davis, 2025), and scales (Wickham et al., 2023). Wilson CIs for observed proportions from the data were computed using binom (Dorai-Raj, 2022). To address our main hypothesis, we tested whether the following differed between the GTE and Control groups: [I] the proportion of embryos that were phenotypically male (i.e. had visually male testes during dissection), [II] the proportion of embryos that were genotypically male, [III] the incidence of gonad:genotype mismatch, and [IV] the likelihood that genetic females contained male gonads. For each test, we fit generalized linear models (binomial logistic) to binary outcomes of either sex (male vs. female) or match-mismatch (phenotypic sex = genotypic sex vs. phenotypic sex ≠ genotypic sex). We used likelihood-ratio tests (LRTs) for fixed-effect (injection treatment) inference. We fit individual models to test each of the four outcomes. We only included embryos in our analyses that were PCR-sexed since those individuals had both genotypic and phenotypic sexes identified. Additionally, we tested whether injection type affected embryonic mortality. Statistical significance was set at the 0.05 alpha level.

Results

When comparing phenotypic sexes (gonadal sex), significantly more of the embryos in eggs injected with GTE had male gonads (80%) compared to Controls (54%) (χ²₁ = 5.58, p = 0.018); however, when comparing genotypic sex, there was no difference between injection groups (χ²₁ = 0.14, p = 0.710; Fig. 3). The GTE group had significantly more mismatches between genotypic and phenotypic sex, and in all cases, the mismatches were a male phenotype but a female genotype (χ²₁ = 7.29, p = 0.007; Fig. 4). When we restricted our analysis to only genetically female embryos, we found that GTE injections were associated with increased probability of being phenotypically male (χ²₁ = 9.69, p = 0.002; Fig. 5). Finally, we found no evidence that injection treatment impacted embryonic mortality (χ²₁ = 0.52, p = 0.469): 43 of 57 Control-injected eggs (∼75%) and 43 of 53 GTE-injected eggs (81%) contained developed embryos.

Fig. 3.

Impacts of treatment on proportions of phenotypic and genotypic males. A significantly higher proportion of eggs injected with GTE contained embryos with male gonads compared to Controls, but the proportion of embryos that were genetically male did not differ between the treatment groups. We fit separate binomial logistic models to phenotypic sex and genotypic sex to test whether injection treatment differs within each. Bars are the observed proportion of males and error bars are Wilson 95% CIs. Proportions listed inside each bar are n males / n total. Asterisks ** indicate statistically significant differences (p < 0.05).

Fig. 4.

Mismatch between genotypic and phenotypic sex. GTE injections were associated with increased mismatches of phenotypic-genotypic sex. All mismatches were genetic females that contained male gonads. The 0 to 100% stacked bar chart shows the observed proportion of matches vs mismatches. Proportions listed inside each bar are n mismatches / n total. Asterisks ** indicate statistically significant differences (p < 0.05).

Fig. 5.

Proportion of genetic females presenting as males. GTE injections significantly masculinized genetic females, evidenced by the development of embryonic testes. Points are the observed proportions of phenotypic male presentation among genotypic females (n = 17 Control, n = 15 GTE). Error bars are Wilson 95% CIs. Asterisks ** indicate statistically significant differences (p < 0.05).

Discussion

The purpose of this experiment was to test whether green tea extract (GTE) influences gonadal development in Japanese quail. In line with our hypothesis, GTE shifted phenotypic sex ratios toward males; 80% of the embryos in the GTE group, but only 54% in the Control group, had testes. This difference was not due to differences in genotypic sex ratios between groups; 55% of embryos in the GTE group were genetic males, compared to 50% of Control embryos. Additionally, there were no differences in embryonic survival between groups, indicating the repression of ovarian development in genetic females rather than sex-biased mortality. In further support of this, GTE increased the incidence of mismatch between genotypic and phenotypic sex in embryos. In fact, the vast majority (8 out of 9) of mismatches between gonadal and genotypic sex were in the GTE embryos, and the mismatches were always of the same combination (male gonad, female genotype). These data indicate that genetic females (ZW) developed bilateral gonads consistent with the appearance of testes. Given these results, we restricted our analysis to only PCR-verified genetic females. Among just these individuals, GTE significantly increased the probability of presenting phenotypically as males. This supports our prediction that GTE masculinized genetic females.

There is abundant evidence that experimental manipulation of estrogen concentrations or aromatase activity can induce a level of sex reversal in galliform embryos (Elbrecht and Smith, 1992; Vaillant et al., 2001; Guioli et al., 2020). Most studies used synthetic aromatase inhibitors or exogenous estrogens to induce sex reversal. Relatively few studies have tested the impacts of natural substances with anti-aromatase or phytoestrogenic activity on avian sexual differentiation, though there is some evidence that they can induce similar effects. For example, garlic, tomato, and green tea extracts lead to female-to-male sex reversal in chickens when injected prior to the period of sexual differentiation (Fazli et al., 2015; Hadibigloo et al., 2018; Jamshasb and Mottaghitalab, 2019; Abdulateef et al., 2021). Likewise, in our study with quail, treatment with GTE induced the masculinization of genetic females with no change in genetic sex ratios, similar to studies using aromatase inhibitors. By using dual sex diagnostics (phenotype and genotype), we lend support to existing evidence that GTE influences the estrogen-dependent pathway that controls the fate of the avian female gonad. While anti-aromatase activity is the leading explanation for these results (Satoh et al., 2002; Monteiro et al., 2006), GTE contains various bioactive components that could potentially engage other endocrine routes (Kuruto-Niwa et al., 2000; Goodin et al., 2002; Belguise et al., 2007; Li et al., 2010). We did not directly quantify aromatase expression, circulating steroids, receptor signaling, or perform histological staging of cortical and medullary components of the gonads. As a result, we cannot definitively point to a mechanism in which GTE influences sexual differentiation. Nonetheless, the sex-specific directionality (female to male) and pre-differentiation timing of our treatment argue strongly for interference with ovarian developmental pathways. Further, it’s well known that estrogen signaling is central to the development of ovaries (reviewed in Nicol et al., 2022), and GTE has documented effects on estrogens and aromatase in vitro in cells and in vivo in rodents (Goodin et al., 2002; Satoh et al., 2002).

Future work should address the following limitations of our study: First, sample sizes were modest, and although they were adequate to detect the reported effects, larger sample sizes in future studies would be beneficial. Second, we detected effects of GTE on sexual differentiation in genetic females, but as gonads were only inspected macroscopically, we cannot conclude whether the gonadal tissue was fully masculinized, nor can we definitively conclude that GTE did not affect the gonads of genetic male embryos. In future studies, histological analysis of the gonadal tissue should accompany the visual and genetic analyses. In our study, we used a single effective dose of GTE to test whether that dose can lead to sex reversal in quail embryos. It is now necessary to test whether the impacts of GTE on sex determination vary in a dose-dependent manner, since the effects of GTE and its catechins on health and physiology are frequently reported to be dose-specific (e.g., in both birds and mammals: Satoh et al., 2002; Lee et al., 2012; Baker and Bauer, 2015; Jelveh et al., 2022). Additionally, it would be useful to examine the impacts of GTE’s different components (for example, the individual catechins) to specifically test the purified constituents on sexual differentiation. Finally, we measured the effects of GTE on gonadal sex in embryos. Influences of GTE on sex phenotypes after hatching have yet to be explored. This includes longitudinal assessments through sexual maturity, especially the persistence of sex-reversed male phenotypes in terms of gonadal morphology, reproductive function, fertility, endocrine status, and the development of secondary sexual traits. The most important mechanistic probing will come from a combination of the above, with additional emphasis on the histology of gonads and ducts, immunohistochemistry and in situ hybridization for important genes and receptors (e.g., DMRT, SOX9, AMG, FOXL2, CYP19A1, ESR1), and steroid profiling throughout embryonic development.

Our work, along with future investigations into GTE’s impacts on sex differentiation, has important implications for birds that consume phytogenic compounds with estrogen-disrupting capacity either in the wild (reviewed in Rochester and Millam, 2009) or in production systems (Zhao et al., 2005; Opalka et al., 2008; Saleh et al., 2019; Qiang et al., 2023). GTE is increasingly being used as a dietary supplement to benefit poultry production in both chickens and quail (reviewed in Seidavi et al., 2020; Mahlake et al., 2022). Importantly, some studies specifically supplemented green tea products to hens or breeder quail to provide antioxidants to eggs or to reduce lipid content (Uuganbayar et al., 2005; Kara et al., 2016; Ling et al., 2022). If sex reversal is an unintended consequence of GTE, supplementation could shift sex ratios in suboptimal ways or reduce the production of reproductively competent adults. We also need to test whether GTE fed to hens is deposited in their eggs, and if so, how much is then available to the embryo at the time of gonadal differentiation to produce effects. When evaluating post-hatch effects in later studies, one should examine the impacts of GTE supplementation on young birds to determine whether chronic exposure triggers negative impacts on reproductive development. In conclusion, we have shown that GTE can act as a potent endocrine disruptor in birds. Given this, and the increasing interest in using GTE to improve poultry health, welfare, and productivity, we recommend practicing care with the use of GTE in industrial settings until we have a better understanding of the full range of its effects.

Funding information

This study was supported by the University of Georgia College of Agricultural & Environmental Sciences.

Ethics statement

This research was approved by the UGA Institutional Animal Care and Use Committee (A2021 09-003-Y3-A0).

CRediT authorship contribution statement

Taylor M. Miller: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Project administration, Methodology, Investigation, Formal analysis, Data curation. Sara-Belle F. Ozburn: Writing – review & editing, Investigation, Data curation. Sara J. Hoover: Writing – review & editing, Investigation. Kristen J. Navara: Writing – review & editing, Validation, Supervision, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization.

Disclosures

The authors declare no conflicts of interest.