Single-nucleus map reveals estradiol’s regulatory effects in obese mouse livers

The liver is not a uniform organ, and the hormone estradiol does not act on it uniformly either. That is the central message of a new study published in Biology of Sex Differences, in which researchers used single-nucleus RNA sequencing to map, cell by cell, how chronic estradiol treatment reshapes gene activity in the obese liver. Working with gonadectomized male and female mice rendered obese through a high-fat diet, the team—led by Yutian Zhao and Xia Yang at the University of California, Los Angeles, together with John M. Stafford of Vanderbilt University Medical Center and The Ohio State University—set out to answer a deceptively simple question: when estradiol protects the liver against the metabolic ravages of obesity, does it do the same things in males and females, or does biology write two different rulebooks?

The motivation for the work stems from a long-standing clinical puzzle. Estradiol, the primary female sex hormone, is known to be hepatoprotective in both sexes: it supports glucose and lipid handling, guards against hepatic fat accumulation, and helps maintain insulin sensitivity. Yet obesity-driven liver disease—manifesting as metabolic dysfunction-associated fatty liver disease, dyslipidemia, and insulin resistance—progresses differently in men and women, and the role of estradiol in those differences has remained murky. Most prior studies averaged gene expression across whole liver tissue, a technique that flattens the organ’s remarkable internal architecture into a single measurement. The liver, however, is functionally zoned. Hepatocytes near the central veins, the pericentral zone, specialize in lipid synthesis, glycolysis, and xenobiotic metabolism, while hepatocytes near the portal triads, the periportal zone, where oxygenated, nutrient-rich blood first arrives, specialize in oxidative metabolism, gluconeogenesis, and cholesterol processing. Estradiol’s effects on these zones, in the context of obesity, in each sex, had never been systematically cataloged.

To build that catalog, the researchers performed single-nucleus RNA sequencing on liver tissue from obese, gonadectomized mice of both sexes treated with chronic estradiol. Gonadectomy was a crucial design choice: by surgically removing the ovaries or testes, the team eliminated endogenous sex hormone production, ensuring that any transcriptional changes they observed could be attributed to the administered estradiol rather than to fluctuating native hormones. Sequencing at the level of individual nuclei allowed the investigators to distinguish hepatocyte subtypes and non-parenchymal cell populations, and to quantify thousands of genes in each cell. On top of differential gene expression analysis, the team layered pathway enrichment analysis, computational modeling of intracellular and intercellular regulatory networks, and—critically—an association analysis linking the E2-responsive mouse genes to human genetic signals for metabolic disease and traits, thereby connecting the mouse data to human cardiometabolic risk.

The headline finding is one of surprising harmony amid striking divergence. Among hepatocyte subtypes, the transcriptional response to estradiol was largely sex-concordant, meaning that in the majority of genes, males and females responded in the same direction and to a similar degree. In pericentral hepatocytes, estradiol consistently suppressed expression of genes involved in insulin resistance, FOXO signaling—the forkhead transcription factor pathway central to insulin action and fasting metabolism—and lipid biosynthesis. In periportal hepatocytes, the hormone produced an equally consistent double effect: it enhanced genes governing cholesterol efflux, the process by which cholesterol is exported from cells for transport to the liver and eventual excretion, while suppressing genes of the oxidative stress response, the cellular machinery that counteracts damage from reactive oxygen species. These shared effects suggest a common core of estradiol action that operates regardless of chromosomal sex, reinforcing the hormone’s reputation as a broadly protective metabolic agent.

But the concordance is only part of the story. Superimposed on the shared response, the researchers documented clearly sex-specific effects. In XY mice, estradiol additionally suppressed lipid and energy metabolic pathways in pericentral hepatocytes and dampened oxidative phosphorylation—the mitochondrial energy-generating machinery—in periportal hepatocytes, deepening the metabolic reprogramming beyond what was seen in females. In XX mice, by contrast, estradiol uniquely regulated programs of glucose and lipid utilization, rewiring how fuel is burned and stored in ways not observed in males. Because the animals were gonadectomized, these differences cannot be attributed to differing levels of circulating sex hormones; instead, they point to sex chromosome complement itself—the presence of XX versus XY chromosomes—as an intrinsic modifier of how liver cells interpret the estradiol signal. This distinction, long debated in the field of sex-differences biology, carries direct implications for how estrogen-based therapies might behave in men versus women.

The regulatory logic behind these responses emerged from network modeling. Estradiol-responsive genes in hepatocyte subtypes of both sexes converged on transcription factors associated with estrogen receptor signaling and with lipid sensing and metabolism, indicating a shared upstream architecture. Alongside this common backbone, the analysis revealed sex-specific regulators—transcriptional drivers that were recruited only in males or only in females—providing a mechanistic explanation for the discordant pathway effects. The sex differences extended beyond individual cells into the social network of the liver. Intercellular signaling analysis, which models the ligand–receptor conversations between cell populations, showed that estradiol induced more changes in cell–cell communication in XY mice than in XX mice across hepatocyte subtypes. In other words, chronic estradiol did not merely reprogram what hepatocytes do internally; it also reshaped how they talk to their neighbors, and it did so more extensively in males.

Perhaps the most translational aspect of the study is its bridge to human genetics. When the team interrogated whether the E2-responsive genes in mouse hepatocytes overlap with human genetic signals for metabolic disease, the results were compelling. Genes responsive to estradiol in hepatocyte subtypes of both sexes were significantly enriched for human genetic associations with lipid profiles and coronary artery disease, suggesting that the pathways estradiol modulates in the mouse liver are the same pathways that shape cardiovascular and lipid risk in people. The analysis also surfaced sex-specific associations: notably, genes downregulated by estradiol in XY hepatocytes were linked to body mass index in men, hinting that estrogen’s influence on male hepatic metabolism may have direct relevance to obesity susceptibility. Additional cell type–specific and sex-specific associations with diabetes were observed, underscoring that a single hormone’s metabolic footprint is filtered differently through each cell type and each sex.

Taken together, the findings reframe how scientists should think about estrogen therapy for metabolic disease. The dominant narrative in the field has emphasized that premenopausal women are relatively protected against fatty liver disease and cardiovascular risk, and that this protection erodes after menopause as estradiol declines—fueling interest in hormone-based interventions for older women and even for men. This study adds a crucial layer of nuance: estradiol is not a single lever but a master regulator whose effects branch at the level of liver zonation, cell subtype, and chromosome complement. A therapeutic strategy calibrated on the shared, sex-concordant actions of the hormone would leverage its protective effects on insulin resistance, lipid synthesis, cholesterol efflux, and oxidative stress in both sexes. But the sex-specific programs—additional lipid and energy suppression in males, unique glucose and lipid utilization changes in females—suggest that dosing, timing, and patient selection may need to be sex-aware to maximize benefit and minimize unintended metabolic consequences.

The technical achievement underlying these conclusions deserves emphasis. Single-nucleus RNA sequencing overcomes a key obstacle in liver genomics: hepatocytes are large, metabolically active, and difficult to dissociate intact, so capturing their transcriptomes nucleus by nucleus preserves cell-type identity without the dissociation artifacts that plague whole-cell protocols. By combining this resolution with a rigorous factorial design—sex, obesity, and estradiol treatment—and then validating the relevance of the mouse findings against human genome-wide association data, the team has produced a resource that will serve as a reference map for anyone studying hepatic sex differences, estrogen pharmacology, or zonal liver physiology. The public availability of the dataset, released as an open-access article, invites the broader community to mine it for additional cell types, pathways, and disease links.

The work also raises questions that will shape the next generation of experiments. Do the sex-specific estradiol responses in hepatocytes translate into different physiological outcomes—different degrees of fatty liver resolution, insulin sensitization, or cholesterol lowering—in males versus females? Which of the sex-specific transcription factors identified by network analysis are causal drivers rather than downstream markers? And how do these programs change with age, diet composition, or the presence of a functioning gonadal axis? The authors note that their study, supported by National Institutes of Health grants R01DK109102 and R01HL144846, establishes the regulatory landscape but that functional follow-up will be needed to convert the map into medicine.

What is already clear is the study’s central lesson for precision medicine: sex is not merely a demographic variable in metabolic disease, it is a biological variable encoded in how every liver cell reads its hormonal environment. As obesity continues to drive an epidemic of fatty liver disease, dyslipidemia, and insulin resistance worldwide, and as interest in estrogen-based and estrogen-mimicking therapies grows, this single-nucleus atlas of the estradiol-treated obese liver provides both a caution and an opportunity. The hormone’s benefits are real and largely shared—but woven through them are strands of sex-specific biology that, if understood and exploited, could finally allow clinicians to prescribe not just the right drug, but the right drug for the right patient.