Estudio conductual y neurobiológico del estrés crónico social en ratón

Abstract

Major depression is a mental disorder often preceded by exposure to chronic stress or stressful life events. Recently, animal models based on social conflict such as chronic social defeat stress (CSDS) are proposed to be more relevant to stress-induced human psychopathology compared to environmental models like the chronic mild stress (CMS). Although CSDS has been proposed as a model of depression, most CSDS studies rely only on analysis of stress-induced social avoidance. Moreover, the predictive validity of the model has been poorly analyzed let alone the possible alterations in dendritic and synaptic structures of brain regions involved in emotional processing.

Therefore, the aims of this work were to study comparatively social versus environmental stress using the CMS and CSDS models. Secondly we have fully characterized the validity of the CSDS model as a possible model of depression.

Comparing both stress models we observed that CMS induced a clear depressive-like profile including anhedonia, helplessness and memory impairment. CSDS induced anhedonia, hyperactivity, anxiety, social avoidance and freezing, signs also common to anxiety and posttraumatic stress disorders. Both models disrupted the excitatory inhibitory balance in the prefrontal cortex. Moreover, CSDS decreased dopamine in the prefrontal cortex and brainstem. We suggests that while depressive-like behaviours might be associated to altered aminoacid neurotransmission in cortical areas, CSDS induced anxiety behaviours might be linked to specific alteration of dopaminergic pathways involved in rewarding processes.

We next explored the predictive validity of the CSDS model using different classes of antidepressants and we observed that this validity was limited. Firstly, repeated venlafaxine showed antidepressant-like activity and both venlafaxine and tianeptine behaved as effective anxiolytics. However CSDS-induced social avoidance was only partially reverted by tianeptine. On the other hand, fluoxetine failed to revert most of the behavioral alterations. In addition, defeated mice showed a downregulation of VGLUT1 mRNA in amygdale and VGLUT1+/- mice showed an enhanced vulnerability to stress-induced social avoidance. However, unlike the environmental model of CMS, these mice did not show enhanced vulnerability to depressive behaviour when exposed to CSDS.

Moreover, CSDS induced an opposite regulation of the spine density and different synaptic plasticity markers in the prefrontal cortex and the amygdale. Specifically while in the prefrontal cortex spine density, GluR1 AMPA receptor subunit, PSD95 and Synaptophysin were increased, a decrease was found in the amygdale. In addition, in the amygdale a downregulation of NR1 NMDA subunit, mGlur4, brain derived neurotrophic factor (BDNF) and CREB was also found. In addition, stressed mice exposed to the social interaction test with a dominant mouse experienced marked freezing and showed increased in pCREB/CREB ratio, proBDNF and mature BDNF in the amygdale.

Taken together these results suggest that CSDS is not a pure model of depression. Indeed it addresses relevant aspects of anxiety-related disorders. Firstly, CSDS induced anhedonia and social avoidance is not associated in this model. Moreover, CSDS might be affecting brain areas mainly involved in the processing of social behaviour, such as the amygdale, where glutamatergic mechanism could play a key role. In addition, these results suggest a persistent maladaptive function of the prefrontal cortex and amygdale of socially defeated mice. Moreover, this would be the first evidence that late reencounter with an aversive social stimuli can triggers fear conditioning in socially defeated mice and support the validity of CSDS to model clinical aspects of post-traumatic stress disorders.