BackgroundChronic social defeat stress is a widely used depression model in male mice. Several proposed adaptations extend this model to females with variable, often marginal effects. We examine if the widely used male-defined metrics of stress are suboptimal in females witnessing defeat. MethodsUsing a data-driven method we comprehensively classified social interaction behavior in 761 male and female mice after chronic social witness/defeat stress, examining social modulation of behavior and associations with conventional metrics (i.e., social interaction (SI) ratio). ResultsSocial stress induces distinct behavioral adaptation patterns in defeated males and witness females. SI ratio leads to underpowered analyses in witness females with limited utility to differentiate susceptibility/resilience. Data-driven analyses reveal changes in social adaptation in witness females that are captured in attenuated velocity change from no target to target tests (ΔVelocity). We explore the utility of this metric in four female social stress models and in male witnesses. Combining SI ratio and ΔVelocity optimally differentiates susceptibility/resilience in witness females and reveals resilient-specific adaptation in a resilience-associated neural circuit in female mice. ConclusionsWe demonstrate that chronic witness stress induces behavioral changes in females that are qualitatively distinct from those observed in defeated males and not adequately sampled by standard male-defined metrics. We identify modulation of locomotion as a robust and easily implementable metric for rigorous research in witness female mice. Overall, our findings highlight the need to critically evaluate sex differences in behavior and implement sex-based considerations in preclinical model design.
Read full abstract