Community Research and Development Information Service - CORDIS



Project ID: 311701
Funded under: FP7-IDEAS-ERC
Country: Austria

Final Report Summary - EMOTIONCIRCUITS (Circuit mechanics of emotions in the limbic system)

Most, if not all, of what we consider relevant has at least some positive or negative emotional component. As such, emotions make the core of our mental self, guide our decisions and behaviors, and - given the prevalence of psychological conditions like depression and posttraumatic stress disorder (PTSD) - are of considerable biomedical interest. Here, we explored how neuronal circuits control emotional behaviors and how genes and drugs modulate these circuit motifs on a molecular level. Ultimately, this research helps to understand the neuronal basis of emotions, as well as the neuronal principles underlying how the perception of basic stimuli transforms into behavior.
One straightforward approach into the complexity of circuitry, genetics and pharmacology of emotions is to study, in exemplary fashion, how basic emotions, like fear and reward, are processed in key hubs of the brain emotion pathways. To this end, we chose the central amygdala (CE) as an entry point and employed deep brain calcium imaging, chemo- and optogenetic manipulations and optogenetic fMRI to study how the CE and interconnected networks process fear and reward related information. Overall, our results suggest that the three major CE neuronal populations differentially process negatively and positively valenced stimuli and behaviors through recruiting different forebrain and brainstem functional networks. CEl SST+, CEm neurons are more tuned towards aversive and CEl PKCd+ cells towards appetitive states.
Interestingly, the CE is embedded in three interregional networks that control CE information processing and behavioral decisions. The first is an interconnected system between the CE, basal forebrain (BFB) and insular cortex (IC), where CE-to-BFB projection signals attention and drive IC activity. IC-to-CE feedback relays interoceptive states from IC to control behavioral decisions in CE. Perturbation of any of these connections increases errors in stimulus discrimination and behavioral decisions. This design links graded interoceptive representations (pain or reward representations) in the IC to binary behavioral responses in the CE and might be necessary to compute fast unambiguous decisions in Pavlovian learning tasks. The second connects the paraventricular thalamus (PVT) with the CE and reacts to (non-Pavlovian) stress experiences. The PVT releases stress peptides in the CE and sensitizes amygdala fear circuits. This, in turn, shifts active to passive coping strategies, a phenomenon shared with generalized anxiety disorder in humans. Interestingly, we find that benzodiazepines (BZDs) antagonize the relay of stress signals through the CE (Griessner et al., unpublished). This provides a long sought-after circuit level mechanism for the BZD anxiolytic effect. Pre-clinical CRHR1 antagonists share this mechanism (Pliota at al., 2018). Novel in silico methods for predicting genetic modulation of circuitry revealed that the thalamo-amygdalar fear circuitry modulating passivity and benzodiazepine anxiolytic effects identified above, is a significant target of natural genetic variation underlying anxiety traits.
In contrast, Pavlovian memories are predominately processed by a third network. At its core, we discovered a novel dopaminergic (DAergic) circuit module that connects the dorsal tegmental area to the amygdala fear circuitry. It operates as minimal learning circuit that senses unpredicted Pavlovian events and writes fear experiences into amygdala synaptic memory by DA dependent long-term potentiation. This module implements the brain’s minimal solution to a learning problem. It contains all major components of associative learning (CS-US integration, reinforcement, memory trace, prediction error coupled teaching signals) assigns a previously unrecognized function to an understudied class of DAergic neurons in basic associative learning and provides a circuit framework for DA in negative reinforcement learning. These results suggest that the brain multiplexes Pavlovian fear and non-Pavlovian stress experience in the same circuitry and similar dynamic states to express aversive experiences. However, while non-Pavlovian stimuli trigger stress peptide release in the CE (CRH) to form short-term memories (Pliota et al., 2018), Pavlovian experiences form long-term memory via DA dependent LTP (Grössl et al., 2018), to facilitate amygdala signalling and the likelihood of freezing.
Taken together, our work contributed workflows for resolving the neural circuit organization of the brain, identified key circuit modules controlling emotional behavior and provided mechanistic frameworks for the pathophysiology of several clinically relevant psychiatric conditions and their treatments.

Reported by

Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top