Model Sex: How Animal Research Informs Human Biology

Animal models have been revealing sex-based differences in disease, drug response, and basic biology for decades, and those discoveries increasingly shape how medicine treats real people. The trouble is that for most of biomedical history, researchers defaulted to male animals, leaving an enormous blind spot in our understanding of female biology. A landmark review covering ten biological disciplines found that single-sex studies of male animals outnumbered those of females by 5.5 to 1 in neuroscience alone, and that male bias was evident in eight of the ten fields surveyed.¹ The consequences of that bias are still being untangled, but the picture emerging from animal research that does include both sexes is striking: males and females can use entirely different cellular machinery to produce the same symptom, metabolize the same drug at different rates, and develop the same disease through distinct molecular routes.

How Research Came to Run on Males

The preference for male animals was rarely stated as policy; it simply became convention. Researchers worried that the estrous cycle in female rodents would introduce too much hormonal fluctuation and inflate their data variability, making experiments messier and more expensive. The rationale went largely unchallenged for decades, even as the proportion of women enrolled in human clinical trials slowly improved. By the time researchers quantified the problem in 2009 data, male bias in non-human studies had actually grown over the preceding half-century, moving in the opposite direction from human-subject research.¹

In 2016, the U.S. National Institutes of Health responded with its “Sex as a Biological Variable” (SABV) policy, requiring grant applicants to account for sex in vertebrate animal research or justify why a single-sex design was appropriate. A national survey of biomedical researchers found that while almost all had heard of the policy, and over a third had changed study designs to comply, the likelihood of actually analyzing results by sex did not differ based on whether researchers had recent NIH funding.² In other words, awareness is high but follow-through is inconsistent. A more recent analysis of NIH R01 grantees confirmed this gap: about 61 percent of funded studies now include both sexes, yet only 44 percent of those conduct sex-based analyses of their data.³ You can put both sexes in the cage and still not look at whether they differ in the results.

The problem persists in clinical domains too. A systematic review of cardiac resynchronization therapy studies found that only about 14 percent considered sex in their study design and only 13 percent performed sex-stratified analyses, though there has been some improvement over time.⁴ In neuroscience specifically, as of 2017, roughly 16 percent of published articles still did not report the sex of their animals at all, and just 15 percent of those using both sexes analyzed sex as an experimental variable.⁵

The Variability Myth

The single most persistent justification for excluding female animals was the belief that hormonal cycling makes them inherently noisier data points. That belief turns out to be wrong. A meta-analysis of neuroscience studies in rats found no sex differences in trait variability across any category, even when male and female mean values were significantly different. Female rats were not more variable at any stage of the estrous cycle than males were.⁶ Broader analyses across multiple species confirmed this and went further: male mice actually showed slightly but significantly greater mean variability than females, with higher coefficients of variation for hormone measures, metabolism-related traits, and body morphology. Females did not show significantly greater variability in any trait category.⁷

This is worth dwelling on because the myth had real consequences. Generations of researchers avoided female subjects to keep their data “clean,” and the data they ended up with was not actually cleaner. It was just incomplete.

Separating Hormones from Chromosomes

One of the most powerful tools for understanding sex differences comes from a genetically engineered mouse called the “four core genotypes” (FCG) model. In these mice, the gene that normally triggers testis development has been moved off the Y chromosome, so researchers can create four types of animals: XX mice with ovaries, XX mice with testes, XY mice with ovaries, and XY mice with testes. This lets scientists tease apart which differences are driven by gonadal hormones and which come from having XX versus XY chromosomes, something impossible to do in normal animals where the two are always linked.⁸

The distinction matters more than you might expect. Some sex differences that look hormonal are actually chromosomal, and vice versa. The FCG model and related tools have shown that both categories contribute, often interacting in ways that neither alone would predict.⁹ The practical takeaway for medicine is that sex differences cannot be reduced to “estrogen does X and testosterone does Y.” The chromosomes themselves, independent of the hormones they typically come with, change how cells behave.

Pain Processing Uses Different Cellular Hardware

Perhaps the most startling animal-model finding in recent years came from pain research. When researchers investigated how chronic pain hypersensitivity develops in mice, they discovered that male and female mice achieve the same behavioral outcome through fundamentally different immune pathways. In males, microglia (the brain’s resident immune cells) are essential for driving mechanical pain hypersensitivity. In females, microglia are not required. Female mice instead use adaptive immune cells, likely T lymphocytes, to reach similar levels of pain sensitivity.¹⁰

The implications for drug development are obvious. A painkiller designed to target microglial signaling might work in men and fail in women, not because women do not experience pain, but because their pain runs on different cellular hardware. And all those years of testing analgesics exclusively in male rodents? The results may have systematically overestimated effectiveness for half the population.

Why Women Get More Autoimmune Disease

Women are far more likely than men to develop autoimmune diseases like lupus. Animal and human cell research is converging on a chromosomal explanation that has nothing to do with hormones. Normally, one of the two X chromosomes in female cells is supposed to be silenced to equalize gene dosage between XX and XY individuals. But some genes escape that silencing. One key escapee is TLR7, a gene encoding a receptor that detects viral RNA and activates the immune system. Single-cell analyses have shown that substantial fractions of female B lymphocytes, monocytes, and plasmacytoid dendritic cells express TLR7 from both X chromosomes, and biallelic cells show higher TLR7 protein expression than monoallelic cells.¹¹

This is not just a curiosity. The same pattern appears in men with Klinefelter syndrome (who carry an extra X chromosome), and they too have elevated lupus risk. Research in plasmacytoid dendritic cells confirmed that TLR7 escape from X inactivation leads to higher mRNA expression and greater interferon-alpha production, fueling the inflammatory cascades that characterize autoimmune disease.¹² These inflammatory cell subsets are disproportionately enriched in patients with autoimmunity or inflammatory diseases, conditions already known to be far more common in females.¹³

Drug Metabolism and Sex-Specific Liver Enzymes

Your liver breaks down drugs using a family of enzymes, and which enzymes you express depends partly on whether you are male or female. In rats, this is dramatic: males produce certain cytochrome P450 enzymes (like CYP2C11 and CYP3A2) that females do not, while females express CYP2C12 instead. Most known sex differences in drug toxicity in rats trace back to these liver enzyme differences.¹⁴ Humans share this pattern in broad outline, though the specific enzymes differ. Sex-based differences in drug metabolism are considered the primary cause of sex-dependent pharmacokinetics in people, driven by differences in cytochromes P450, sulfotransferases, and other enzyme families.¹⁵

A well-known real-world example is zolpidem, a sleep aid. The FDA cut the recommended dose for women in 2013 after data showed they cleared the drug more slowly, leading to dangerously high morning blood levels. Pharmacokinetic modeling has since confirmed that clearance is lower in females, though the picture is complicated: women actually show higher activity in one of the key metabolizing enzymes, CYP3A4.¹⁶ The net effect on drug exposure is modest, but it was enough to cause measurable impairment. That kind of nuance is exactly what you miss when you only test drugs in male animals.

Hearts, Estrogen, and Protection That Disappears

Animal studies have repeatedly shown that female hearts tend to tolerate oxygen deprivation better than male hearts. In rat models, females often show less damage after blood flow is cut off and then restored (ischemia-reperfusion injury), though the effect is not universal across all experimental conditions.¹⁷ Estrogen appears to be a major player. In mouse models where ovaries are removed, supplementing estrogen back reduces the size of heart damage and lowers inflammation-related cell death.¹⁸

What makes this particularly interesting is that estrogen receptor signaling protects hearts in both sexes. A study using male mice engineered to lack estrogen receptor-alpha found that their hearts recovered worse after ischemia, with more calcium buildup, less nitric oxide production, and more cell death than normal males. It was the first demonstration that this receptor has a cardioprotective role in males too.¹⁹ The implication is that estrogen-receptor-based therapies might benefit both sexes after a heart attack, something that would never have been investigated if researchers had stuck to the assumption that estrogen is “only” a female hormone.

Sex Differences in Alzheimer’s and Neurodegeneration Models

Women develop Alzheimer’s disease at higher rates than men, and animal models are starting to explain why, although the answer is not simple. In the triple-transgenic Alzheimer’s mouse model, female mice develop more amyloid plaques, more neurofibrillary tangles, more neuroinflammation, and greater spatial memory deficits than males.²⁰ A different Alzheimer’s model (the 5XFAD mouse) shows a related but distinct pattern: females accumulate more amyloid and have heightened inflammation, but males express higher levels of stress markers and neurotrophic factors regardless of whether they carry the disease genes.²¹

In a model of Parkinson’s-like pathology, the pattern flips. When mice are injected with alpha-synuclein fibrils (the protein aggregates associated with Parkinson’s), males show faster neurodegeneration, with accelerated volume loss in the striatum, substantia nigra, and motor cortex. Female mice showed resistance to this disease progression despite having early pathology markers.²² These findings complicate the narrative that “women are more vulnerable to neurodegeneration” and suggest instead that vulnerability depends on which disease, which brain region, and which molecular pathway you are looking at.

The Gut Microbiome Has a Sex Too

The community of microbes living in the gut differs between males and females, both in animal models and in humans. These differences are not just incidental. They interact with hormones and the immune system in ways that can change susceptibility to inflammatory disease. Sex hormones shape which bacterial communities thrive, and those communities in turn influence local gut immunity and systemic inflammation.²³ Animal models have been critical here because you can control diet, housing, and genetics while varying sex, something impossible in humans whose microbiomes are shaped by a lifetime of different diets and environments.

A related finding from obesity research illustrates the broader metabolic picture. When male and female mice are fed the same high-fat diet, females show a greater capacity to enlarge their fat cells, which paradoxically seems to be protective. Female mice on this diet develop less macrophage infiltration into fat tissue, less fat deposited in the liver, and better insulin sensitivity than males, despite sometimes gaining more total weight.²⁴ The lesson for human medicine: body weight alone is a poor proxy for metabolic health, and the sex of the organism changes what weight gain means at the cellular level.

When Stress Echoes Across Generations

One of the more striking areas of animal research is how prenatal stress leaves biological marks that persist for multiple generations. In rats, stress during pregnancy alters the placenta’s epigenetic landscape, and these changes are not limited to the first generation of offspring. A systems-biology study found a moderate impact in the directly exposed offspring but drastic changes in the second and third generations, suggesting the effects compound rather than fade.²⁵ Critically, these effects are sex-specific: male and female offspring respond differently to the same prenatal stress exposure, with distinct changes in stress-pathway programming.²⁶ This kind of multigenerational experiment simply cannot be done in humans, making animal models indispensable for understanding how adversity gets biologically transmitted.

Stem Cell Therapies and Donor Sex

If sex matters for disease susceptibility and drug metabolism, it should not be surprising that it matters for regenerative medicine too. When muscle-derived stem cells from male and female donors are transplanted into mice with muscular dystrophy, female-derived cells regenerate skeletal muscle more efficiently.²⁷ More broadly, estrogen signaling has been shown to influence how musculoskeletal stem cells behave, affecting tissue repair processes in sex-dependent ways.²⁸ For clinicians developing stem cell therapies, this means the sex of the donor, the sex of the recipient, and the hormonal environment of both could affect outcomes. Ignoring any one of those variables risks designing treatments that work well in one group and underperform in another.

From Animal Cages to Chips on a Bench

A newer technology is starting to complement traditional animal models. Organ-on-a-chip systems use human cells grown on microengineered platforms that mimic the mechanical and biochemical environment of living organs. For sex-difference research, these platforms offer a unique advantage: you can take stem cells from male and female human donors, differentiate them into the same cell type (heart cells, liver cells, or others), and then add or withhold hormones independently. This separation of chromosomal sex from hormonal environment is something the four core genotypes mouse model achieves through genetic engineering, but organ-on-a-chip technology does it with human cells.²⁹

Cross-species validation also helps establish which animal findings translate to people. A study comparing brain gene expression between rhesus macaques and humans found that sex accounted for a similar fraction of gene-expression variation across both species, roughly half a percent on average. Sex differences on the X chromosome were particularly well-conserved: when a gene sat on the X chromosome in one species, it showed greater female-biased expression in that species, and estimated sex effects across thousands of genes were positively correlated between macaques and humans.³⁰ Findings like these give researchers more confidence that sex differences observed in animal brains are not just quirks of one species but reflect conserved biology with direct relevance to human health.

Behavioral Testing Has Its Own Sex Bias

Even behavioral assays, the standard tests researchers use to measure anxiety or risk-taking in rodents, can produce misleading results if sex is not carefully considered. In the elevated plus maze, a common anxiety test, male rats spend more time in exposed areas, which is conventionally interpreted as lower anxiety and greater risk-taking. But a finer-grained temporal analysis revealed something more interesting: while females spent more time in the enclosed safe areas for most of the test, they showed a larger increase in exploration of the risky open areas toward the end, potentially reflecting greater regulation of anxiety over time rather than simply higher baseline fear.³¹ Sex differences in zebrafish behavior and neuropharmacological responses add another layer, with mounting evidence that sex shapes central nervous system function across vertebrate species well beyond mammals.³² The broader point is that behavioral assays were largely developed and validated in male animals. When females are finally included, the tests themselves sometimes need rethinking.