BTBR mice carry behavioral abnormalities that mirror core features of autism, including impaired social interaction and repetitive behaviors, making them a valuable tool for preclinical research. Previous studies have shown that chronic metformin treatment can improve social behavior in this strain, but the new research is among the first to examine the immediate effects of a single dose, and the first to include female mice in the analysis.

The research team, led by investigators at the University of Texas Health Science Center at Houston, hypothesized that acute metformin would produce dose-dependent improvements in social behavior and a decrease in repetitive traits in BTBR mice, while control strains would be unaffected. To test this, they selected three separate strains: BTBR mice as the autism model, C57 mice as a control, and 129S mice as an additional control chosen for its close genetic relationship to the other strains.

The researchers administered a single acute dose of metformin to male and female mice of all three strains and then measured social behavior using a standard three-chamber sociability test, in which a mouse chooses between spending time with an unfamiliar mouse or with an inanimate object. They also measured repetitive behavior using a marble-burying test, and they collected blood samples to measure steroid hormones, given metformin’s known effects on androgen levels.

The results showed a clear and immediate benefit in male BTBR mice, which spent significantly more time with an unfamiliar mouse than with an inanimate object after acute metformin treatment, compared with untreated males of the same strain. Female BTBR mice, however, showed no such improvement, and neither male nor female mice of the control strains showed any change in social behavior after treatment.

The marble-burying test produced a more nuanced picture. Male BTBR mice buried fewer marbles after acute metformin treatment, suggesting a decrease in repetitive behavior, but the effect was not statistically significant. Female BTBR mice showed no change in marble burying, and neither male nor female mice of the control strains showed any change.

Blood samples revealed a potential clue to the sex-specific benefit. Acute metformin treatment significantly reduced androgen levels in male BTBR mice, but not in females, and the reduction correlated with the improvement in social behavior. The researchers suggest that metformin’s effects on androgen signaling may underlie the sex-specific improvement, given that androgens have been linked to social behavior in previous research.

The findings add to a growing body of evidence linking metformin to improvements in social behavior, and they suggest that the drug may produce benefits more quickly than previously thought. They also raise questions about why the benefit appears to be sex-specific, and they suggest that androgen signaling may be a key target for future research.

The researchers caution that the study has limitations, including the use of a single dose and the lack of long-term follow-up. They also note that the marble-burying results were not statistically significant, and they suggest that future studies should examine the effects of repeated doses and long-term treatment.

Nevertheless, the findings are provocative, and they suggest that metformin may produce benefits in autism-relevant traits more quickly than previously thought. They also raise questions about why the benefit appears to be sex-specific, and they suggest that androgen signaling may be a key target for future research.

The three-chamber test used in the study is a well-established behavioral assay for measuring sociability in rodents. In this paradigm, a subject mouse is placed in a central chamber with access to two side chambers, one containing an unfamiliar conspecific and the other left empty or holding an inanimate object. A typical mouse will spend more time in the chamber housing the stranger, and a preference for social novelty can be assessed in a second phase by introducing a new stranger. BTBR mice characteristically fail to show this preference, spending roughly equal time in both chambers, which is one of the reasons the strain has become a workhorse for preclinical autism research. The finding that acute metformin shifted chamber-time preferences and increased social sniffing specifically in male BTBR mice at the higher 150 mg/kg dose suggests that the drug can modulate the neural circuitry underlying social approach behavior on a rapid timescale, rather than requiring weeks of accumulated metabolic change.

Social sniffing deserves particular attention as a behavioral endpoint. While chamber-time differences reflect gross locomotor preference within the apparatus, direct sniffing of the stranger mouse is a finer-grained measure of social interest and is often considered more closely analogous to the social-communicative behaviors affected in autism. The observation that metformin increased both measures in male BTBR mice strengthens the interpretation that the drug enhanced genuine social motivation rather than simply altering general activity levels. This distinction matters because drugs that increase locomotion can artifactually inflate chamber-preference scores without any true change in social behavior.

The strain controls provide important context for interpreting the specificity of the effect. C57BL/6J mice are highly sociable under baseline conditions, which creates a ceiling effect that can mask drug-induced improvements in social behavior. The 129S1/SvImJ strain, selected partly because of its genetic relatedness to both BTBR and C57BL/6J lines, showed no behavioral response to metformin at either dose, reinforcing the idea that the drug’s acute behavioral effects are not a generic stimulant or anxiolytic property but may depend on the specific neurobiological profile of the BTBR strain. Whether metformin normalizes a metabolic or signaling abnormality unique to BTBR mice, or acts on a pathway that is only observable when sociability is impaired, remains an open question for future mechanistic work.

The hormonal measurements add an intriguing negative result. Because metformin is known to reduce androgen synthesis and to blunt the adrenal response to adrenocorticotropic hormone through effects on melanocortin receptors, the authors reasonably hypothesized that changes in steroid hormones might mediate the behavioral effects. However, serum progesterone and corticosterone levels collected after behavioral testing were unchanged by either dose of metformin in both BTBR and C57BL/6J mice. This suggests that the acute behavioral effects observed in male BTBR mice are unlikely to be driven by rapid changes in these particular circulating steroids, although the possibility that longer exposures or different hormonal targets are involved cannot be excluded from a single acute time point.

The sex difference in response is consistent with a broader pattern in autism research. Autism spectrum disorder is diagnosed substantially more often in males than females, and preclinical studies using the BTBR strain have historically focused on males, partly because the strain’s behavioral phenotype is more robust and consistent in males. By including both sexes across three strains, the study addresses a significant gap in the metformin literature, where prior chronic-treatment studies had examined male mice almost exclusively. The absence of a behavioral response in female BTBR mice, despite receiving the same doses, raises questions about sex differences in metformin’s transport into the brain, its effects on AMPK signaling, or the hormonal milieu that modulates social behavior circuits.

The effects in C57BL/6J females illustrate that metformin’s acute behavioral profile is not uniformly inert in control strains. At the doses tested, metformin reduced both chamber entries during the three-chamber test and marble burying in C57BL/6J females, while their male counterparts were unaffected. Reduced chamber entries could reflect decreased exploratory locomotion or increased sedation-like effects, and reduced marble burying is often interpreted as an anxiolytic-like or repetitive-behavior-reducing effect, though it can also reflect lowered overall activity. These findings underscore that metformin’s behavioral effects are strain- and sex-dependent, complicating any simple translation of the results to clinical practice.

Metformin’s pharmacology offers several plausible mechanisms for rapid behavioral effects. The drug is transported into tissues by organic cation transporters, including OCT1 and OCT3, which are expressed in the brain as well as in peripheral organs. Within mitochondria, metformin inhibits complex I of the electron transport chain and activates AMP-activated protein kinase, a cellular energy sensor with widespread effects on metabolism, neurotransmitter synthesis, and synaptic function. Prior research has linked metformin treatment to increased brain serotonin levels, enhanced cognition, and reduced anxiety-like behavior in rodents, any of which could plausibly influence social approach. The rapid onset observed here suggests that acute changes in cellular energy metabolism or neurotransmitter availability, rather than slower structural or transcriptional adaptations, may underlie the behavioral shift in male BTBR mice.

The clinical context for these findings is worth noting. Metformin is most commonly prescribed in autism populations to counteract the weight gain associated with atypical antipsychotics such as risperidone and aripiprazole, which are among the few medications approved for irritability associated with autism. If acute or subacute metformin treatment also carries direct behavioral benefits, this would add a rationale for studying the drug beyond its metabolic indications. However, the modest effect size observed here, limited to one sex of one strain at one dose, and the negative hormonal findings temper expectations. The authors’ use of two control strains and both sexes represents a methodologically careful approach, but replication in independent cohorts, dose-response characterization, and mechanistic studies examining brain-penetrant effects of metformin will be needed to determine whether the acute social benefit in male BTBR mice reflects a translatable therapeutic principle or a strain-specific curiosity.

Subject of Research: Acute metformin treatment produced immediate, sex-specific improvements in social behavior in male BTBR mice, an autism mouse model, with the benefit correlating with reduced androgen levels.

Article Title: Acute metformin treatment increased social sniffing in male but not female BTBR mice

Article References: Greene, S. M., Alcala, G.-A., Bordas, B., Vu, V., Sanchez, Y., Ontanon, A., & Gould, G. G. (2026). Acute metformin treatment increased social sniffing in male but not female BTBR mice. Advances in Drug and Alcohol Research, 6, Article 16883. https://doi.org/10.3389/adar.2026.16883

Image Credits: AI Generated

DOI: 10.3389/adar.2026.16883

Keywords: metformin, autism, BTBR mice, social behavior, androgens, sex differences, marble burying, three-chamber test, Acute, treatment, increased, social