Animal models of social stress: effects on behavior and brain neurochemical systems

Abstract

Social interactions serve as an evolutionarily important source of stress, and one that is virtually ubiquitous among mammalian species. Animal models of social stress are varied, ranging from a focus on acute, intermittent, or chronic exposure involving agonistic behavior, to social isolation. The relative stressfulness of these experiences may depend on the species, sex, and age of the subjects, and subject sex also appears to influence the value of hypothalamic--pituitary--adrenal (HPA) axis activity as a general criterion for stress response: higher glucocorticoid levels are typically found in dominant females in some species. Social stress models often produce victorious and defeated, or dominant and subordinate, animals that may be compared to each other or to controls, but the appropriateness of specific types of comparisons and the interpretations of their differences may vary for the different models. Social stress strongly impacts behavior, generally reducing aggression and enhancing defensiveness, both inside and outside the stress situation. Social and sexual behaviors may be reduced in subordinate animals, as is activity and responsivity to normally rewarding events. However, some components of these changes may be dependent on the presence of a dominant, rather than representing a longer-term and general alteration in behavior. Social stress effects on brain neurotransmitter systems have been most extensively investigated, and most often found in serotonin and noradrenergic systems, with changes also reported for other monoamine and for peptidergic systems. Morphological changes and alterations of neogenesis and of cell survival particularly involving the hippocampus and dentate gyrus have been reported with severe social stress, as have longer-term changes in HPA axis functioning. These findings indicate that social stress models can provide high magnitude and appropriate stressors for research, but additionally suggest a need for caution in interpretation of the findings of these models and care in analysis of their underlying mechanisms.