Perinatal SSRI exposure impacts innate fear circuit activation and behavior in mice and humans

Abstract

Before assuming its role in the mature brain, serotonin modulates early brain development across phylogenetically diverse species. In mice and humans, early-life SSRI exposure alters the offspring's brain structure and is associated with anxiety and depression-related behaviors beginning in puberty. However, the impact of early-life SSRI exposure on brain circuit function is unknown. To address this question, we examined how developmental SSRI exposure changes fear-related brain activation and behavior in mice and humans. SSRI-exposed mice showed increased defense responses to a predator odor, and stronger fMRI amygdala and extended fear-circuit activation. Likewise, adolescents exposed to SSRIs in utero exhibited higher anxiety and depression symptoms than unexposed adolescents and also had greater activation of the amygdala and other limbic structures when processing fearful faces. These findings demonstrate that increases in anxiety and fear-related behaviors as well as brain circuit activation following developmental SSRI exposure are conserved between mice and humans. These findings have potential implications for the clinical use of SSRIs during human pregnancy and for designing interventions that protect fetal brain development.

Fig. 1. Postnatal fluoxetine (PND2-11) increases freezing to a predator.

A Timeline of the adult mouse behavioral study. Created in BioRender. Zanni, G. (2025) https://BioRender.com/m34p441. B Manual freezing score during habituation early and late phases. C Manual freezing score during context re-exposure and during predator odor. D Latency to freeze expressed as percent change of habituation freezing latency. E Labeling of different animal body parts: nose (purple), tail base (orange), right ear (electric blue), left ear (cyan), right hip (blue), left hip (light blue), mid back (yellow). Top picture shows animals exploring and the bottom picture shows an animal freezing. F Deeplab Cut freezing score during context re-exposure and during predator odor. G Deeplab Cut speed scoring during context re-exposure and during predator odor. Data are presented as mean values +/- SEM. Statistical tests used: ANOVA and Fisher’s LSD posthoc. PNSAL N = 13, PNFLX N = 11. *p < 0.05; **p < 0.01; ***p < 0.001. Source data are provided as a Source Data file.

Fig. 2. Postnatal fluoxetine (PND2-11) increases innate fear circuit activation.

A Timeline of the adult mouse fMRI study. Created in BioRender. Zanni, G. (2025) https://BioRender.com/s15z123. B Shown are the different components of the mouse imaging system. Below are sagittal and axial views of an awake mouse brain. Note the linearity along the Z-axis. The axial images taken from a 22-slice RARE sequence (0.6 mm thickness) demonstrate complete brain coverage from the olfactory bulbs to the brainstem. C Statistical map displaying relative positive BOLD signal in response to predator urine odor in awake PNFLX compared to PNSAL mice. PNSAL N = 18, PNFLX N = 26. D List of significant brain areas, such as thalamus, putamen, brainstem and cerebellum, central amygdala, substantia nigra, periaqueductal gray, raphe, and habenula ranked in order of their significance in red for change in positive BOLD volume of activation (number of voxels) in PNFLX compared to PNSAL (false discovery rate p = 0.0401). Statistical test used: Wilcox rank sum test corrected for multiple (134) comparisons.

Fig. 3. Effects of in utero SSRI exposure in human adolescents.

A The experimental timeline. At baseline mothers reported on their pregnancy and demographics. Children underwent an MRI scan approximately two years post-baseline, and mothers filled out the Child Behavioral Checklist (CBCL) to assess symptoms at that time and one year post-MRI. Created in BioRender. Zanni, G. (2025) https://BioRender.com/e26b462. B Left. In utero SSRI exposure is associated with increased CBCL symptoms, accounting for maternal lifetime depression and maternal anxiety and depressive symptoms. N = 97 SSRI⁺, N = 3876 SSRI^-. Right. Maternal lifetime depression is also associated with increased offspring symptoms. C Left. In utero SSRI exposure is associated with increased amygdala, hippocampus, insula, putamen and thalamus activation to fearful-neutral faces, depicted as change in standard deviation of the BOLD response accounting for maternal lifetime depression and maternal anxiety and depressive symptoms. Right. Exposure to maternal lifetime depression does not result in significant differences in BOLD response to fearful-neutral faces. D Same analysis as (C) depicted as predicted means by group. N = 70 SSRI⁺, N = 2989 SSRI^-. Left. In utero SSRI exposure is associated with increased BOLD activation to fearful-neutral faces. Right. Exposure to maternal lifetime depression does not result in differential BOLD response. Linear mixed effects models with propensity score weighting, 2-tailed, FDR-corrected. Error bars signify 95% confidence intervals. Precise p-values are reported in the main text. *p < 0.05; ***p < 0.005.