Evolution of Estrogen Receptors Explains Why Species Respond Differently to Chemicals

One of the most persistent puzzles in environmental toxicology is why a chemical that disrupts the hormones of one mammal may leave a closely related species largely unaffected. Regulatory risk assessment has long struggled with this problem, because safety thresholds established in laboratory rodents cannot simply be extrapolated to whales, bats, elephants, or the thousands of other mammals that share the planet. A new peer-reviewed study published in Environmental Science & Technology by researchers at Ehime University, led by Hisato Iwata, now offers a molecular explanation rooted in evolution itself. By tracing how the estrogen receptor alpha protein has diversified across mammalian lineages, and by linking that diversification to the ecological lifestyles of different species, the team has taken a substantial step toward predicting, rather than merely observing, species-specific chemical sensitivity.

The centerpiece of the research is the estrogen receptor alpha, or ERα, a ligand-activated transcription factor that binds estrogenic molecules and switches on suites of genes governing reproduction, development, and metabolism. Because many environmental contaminants, from natural phytoestrogens to synthetic industrial compounds, can interact with this receptor, ERα sits at the heart of endocrine disruption biology. The ligand-binding domain of the receptor contains a pocket whose shape and chemical properties determine which molecules fit and how strongly they bind. Even a single amino acid substitution in this region can alter the geometry of the pocket, changing the receptor’s affinity for particular chemicals. The Ehime University team reasoned that if evolutionary pressures have reshaped ERα differently in different mammalian lineages, those structural changes should translate into measurable differences in chemical responsiveness, and that these differences could be mapped and predicted systematically.

To test this idea, the researchers assembled an unusually broad comparative dataset. They collected ERα protein sequences from 169 mammalian species and paired this genetic information with 107 ecological traits for each species, encompassing diet, habitat, reproductive strategy, and other life-history characteristics. The scale of this comparison matters. Most earlier studies of receptor evolution focused on a handful of model organisms, which limited their ability to detect patterns tied to ecology. By casting the net across nearly the entire mammalian class, the team could ask whether species that share similar lifestyles also share similar receptor architectures, a pattern that would suggest common evolutionary pressures acting on hormone signaling.

The bioinformatic analysis revealed exactly such a pattern. When ERα sequences were clustered according to their amino acid composition, the resulting groupings aligned with the ecological profiles of the animals rather than simply with their phylogenetic relatedness alone. Species occupying comparable ecological niches, facing comparable diets and environments, tended to carry receptors with comparable sequence features. This finding implies that ecological factors have exerted selective pressure on the structure of the estrogen receptor, fine-tuning its sensitivity to the endogenous and exogenous molecules each species encounters in its environment. In other words, the receptor is not a static, universal sensor of estrogen; it is an evolving interface between an animal’s biology and its world.

Sequence comparisons alone cannot show how amino acid differences translate into three-dimensional structure, so the researchers turned to in silico modeling. Using computer-based structural analysis of the ERα ligand-binding domain, they examined how the binding pocket, the cavity where estrogenic chemicals dock, varies across species with different diets. The results were striking: omnivorous mammals tended to possess larger binding pockets than either carnivores or herbivores. A broader pocket plausibly accommodates a wider chemical repertoire, which fits the ecological reality that omnivores ingest a more chemically diverse diet than specialists. This structural observation provides a mechanistic bridge between what an animal eats and how its hormone receptor responds to chemicals, connecting ecology to molecular pharmacology in a single framework.

Structural modeling, however, remains a prediction until it is tested in living systems, and this is where the study’s experimental component becomes decisive. The team conducted cell-based in vitro assays using ERα from six mammalian species, exposing the receptors to both natural and synthetic estrogenic chemicals and measuring the resulting activation responses. The experiments demonstrated marked interspecies differences in sensitivity. Receptors that appeared structurally similar in some regions behaved distinctly when challenged with the same compounds, confirming that the amino acid variations documented in the sequence analysis have functional consequences. These assays transform the study from a purely comparative exercise into an empirical demonstration that evolution has produced receptors with genuinely different chemical sensitivities across mammals.

Among the most notable discoveries was a unique mutation found in cetaceans, the group that includes whales and dolphins. In human ERα, the 349th amino acid position is occupied by asparagine; in cetaceans, the team found that this position carries serine instead. This substitution alters the structure of the receptor and is linked to aquatic adaptation as well as changes in chemical responsiveness. The finding is particularly significant for wildlife risk assessment, because cetaceans are apex predators in marine ecosystems and are known to accumulate persistent organic pollutants at high concentrations. If their estrogen receptors have evolved structural properties that change how they respond to estrogenic chemicals, then risk models built on terrestrial mammal data may systematically misjudge the vulnerability of these ocean giants.

The final analytical layer brought artificial intelligence into the framework. Using machine learning, the researchers built models designed to identify which factors best predict variation in ERα structure and, by extension, differences in chemical sensitivity among species. Among the variables examined, one emerged as especially promising: the brain-to-body mass ratio, the proportion of an animal’s total body weight accounted for by brain tissue. That a neurological and life-history trait should correlate with receptor structure and chemical responsiveness is an intriguing result, suggesting that the evolution of endocrine signaling is entangled with broader physiological and ecological adaptations. Machine learning proved capable of integrating these multidimensional data, sequence features, structural parameters, and ecological variables, into predictive models, which is precisely the kind of capability needed for forward-looking risk assessment.

The methodological architecture of the study is as important as its individual findings. The work exemplifies New Approach Methodologies, or NAMs, an advanced framework that combines bioinformatic sequence analysis, in silico structural modeling, artificial intelligence and machine learning, and in vitro cell-based assays into a single integrated pipeline. NAMs are gaining momentum in toxicology because they promise to reduce reliance on animal testing while increasing the breadth and speed of chemical safety evaluation. This study demonstrates the power of that promise: no single method could have connected 169 species’ genomes, 107 ecological traits, protein structures, and functional receptor responses, but the integrated approach made those connections visible. Each layer of analysis constrained and validated the others, producing conclusions more robust than any component could deliver alone.

The implications reach well beyond academic evolutionary biology. Environmental regulators currently depend heavily on data from a small set of laboratory species, and the study shows why that dependence is hazardous: receptor evolution has embedded species-specific sensitivities that no single-model extrapolation can capture. With the predictive factors identified here, including binding pocket architecture, key amino acid positions, and traits such as brain-to-body mass ratio, it becomes conceivable to estimate the chemical vulnerability of poorly studied wildlife before contamination occurs, rather than after populations show harm. For conservation biologists, the cetacean mutation offers a concrete example of how adaptation to one environment, in this case aquatic life, can reshape sensitivity to human-made chemicals in another. For the broader field of endocrine disruption research, the message is that the molecular targets of chemical action are themselves products of evolution, and understanding that evolutionary history is not optional but essential. By revealing the ecological and structural determinants of species-specific chemical sensitivity, the Ehime University team has outlined the molecular mechanisms required to predict chemical risks to wildlife with high precision, marking a meaningful advance toward a new generation of ecologically informed toxicology.

Subject of Research: Evolutionary determinants of species-specific chemical sensitivity in mammalian estrogen receptor alpha

Article Title: Uncovering species-specific chemical sensitivity: Insights from mammalian hormone receptor evolution and new approach methodologies

Article References: Uncovering species-specific chemical sensitivity: Insights from mammalian hormone receptor evolution and new approach methodologies. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: estrogen receptor alpha, species-specific sensitivity, New Approach Methodologies, endocrine disruption, mammalian evolution, bioinformatics, in silico modeling, machine learning, in vitro assays, cetaceans, wildlife toxicology, binding pocket