The effects of traumatic experience on behavior, neuronal activity, and functional connections in the conscious mouse brain resting state networks  Ksenia Toropova , Olga Ivashkina, Anna Ivanova, Elena Konovalova, Aleksey Ivanitsky, Konstantin Anokhin (Lomonosov Moscow State University; Institute of Higher Nervous Activity and Neur, Moscow, Russian Federation)   C15

It is known that the brains of animals and humans is active at resting state in absence of external stimulation or functional tasks. In this paper we investigate how past experience affects characteristics of such resting state networks in animals. To do this we subjected mice to single traumatic experience that induced posttraumatic stress disorder (PTSD) and then analyzed activity of their brain (42 regions including associative, sensory and motor cortices; hippocampus; parahippocampal regions; amygdala; basal ganglia; thalamus; hypothalamus and midbrain) by c-Fos cellular mapping during traumatic memory retrieval and at rest in comparison with non-stressed animals. PTSD development led to global changes in brain activity: number of c-Fos-active neurons was significantly increased in different areas of the brain during traumatic memory retrieval. Similarly, animals at rest with PTSD showed increased activity in 11 brain regions participating in fear networks. Using graph theory approaches, we identified functional connections in resting state networks of PTSD and in control and found the main clusters of these networks. To do this, experimentally obtained networks were compared with model networks: random, scale-free and small-world. In both groups of mice, level of clustezation was the same as in the scale-free network: number of clusters exceeded random level. At the same time, these clusters did not interact well with each other: global efficiency of experimental networks was at the level of random network. The resting state networks of PTSD and control mice differed: PTSD network was less clustered and longer paths linked the clusters. Functional connectivity analysis showed that induction of PTSD led to global changes in the structure of resting state networks affecting almost all brain areas. In naive animals, cortical regions had the most connections, whereas PTSD animals thalamus, striatum and amygdala had. PTSD destroyed virtually all functional connections present in naive mice; only fully connected cluster of auditory and visual cortices remained. In addition, if in naive animals the main hubs of resting state network were cingulate and retrosplenial cortices, in PTSD animals these areas almost completely lost their connections, and paraventricular thalamic nucleus became the hub. In contrast, amygdala functional connectivity was virtually zero in naive animals, whereas in PTSD significant number of connections between amygdala, associative cortices, and striatum were observed. In addition, we have shown that PTSD induction changes spontaneous behavior of animals, causing elevated anxiety and decreased research activity in safe conditions of home cages. Behavior in conditioned fear, EPM and sensitization tests also changed, and these changes could be disrupted by protein synthesis inhibition during traumatic experience. Protein synthesis inhibition also returned brain activity and structure of resting networks to normal in PTSD animals. Our findings show that stressful experiences can alter spontaneous behavior, induced and spontaneous brain activity and patterns of functional connections in resting state neuronal networks long after traumatic episode. We assume that these changes reflect replay of neuronal ensembles of the animal's past subjective experience. This assumption was tested by disrupting the development of PTSD. Supported by RSF 16-15-00300, 20-15-00187 and RFBR 19-015-00534, 20-015-00427.