Stress Models of Depression: A Question of Bad Timing

Learned Immobility Produces Enduring Impairment of the HPA Axis Reactivity in Mice without Replicating the Broad Spectrum of Depressive-Like Phenotype

The forced swim stress test (FST) is widely used for screening pharmacological or non-pharmacological strategies with potential antidepressant activities. Recent data have suggested that repeated FST for five consecutive days (i.e., 5d-RFSS) could be used to generate a robust depressive-like phenotype in mice. However, the face, construct, and predictive validities of 5d-RFSS have been recently challenged. This study took advantage of recent findings showing that mice vulnerability to anxiety is enhanced when animals are stressed during the dark phase, to provide new insight into the relevance of this model. Our results showed a progressive increase in time of immobility in 5d-RFSS mice relative to control non-stressed animals (sham). Three weeks later, we noticed that 5d-RFSS mice injected with the vehicle compound (Veh) still exhibited a high level of immobility in the FST whereas this behavior was reversed by the antidepressant drug amitriptyline (AMI). However, 5d-RFSS/Veh and 5d-RFSS mice/AMI mice showed normal performances in the open field, the novelty suppressed feeding and the tail suspension tests. Despite this lack of generalized behavioral deficits, an impairment of different parameters characterizing the hypothalamic-pituitary-adrenal (HPA) axis reactivity was evidenced in 5d-RFSS mice/Veh but not in 5d-RFSS mice/AMI. Despite anomalies in the HPA axis, the activity of the central serotonergic system remained unaffected in 5d-RFSS mice relative to controls. From our results, it is suggested that learned immobility does not replicate the broad spectrum of depressive symptoms observed in other chronic models of depression such as the unpredictable chronic mild stress (UCMS) model, the chronic social defeat stress (CSDS) model or chronic corticosterone (CORT) exposure but its influence on the HPA axis is remarkable. Further experiments are warranted to makes this model suitable for modelling depression and therefore refine its translational applicability.