Cyril Herry inNature Neuroscience

Encoding of fear learning and memory in distributed neuronal circuits.
Herry C, Johansen JP. Nat Neurosci. 2014 Dec;17(12):1644-1654. doi: 10.1038/nn.3869. Epub 2014 Nov 21. Review.Cyril Herry dans Nature neuroscience

Anxiety disorders are a major health problem in developed western societies. For instance, in any given year, it has been estimated that 16% of the adult population in the US and 14% of the adult population in Europe exhibit an anxiety disorder. In particular, post-traumatic stress disorder (PTSD), an anxiety disorder associated with pathological fear memories, represents one of the most frequent anxiety disorders that can develops following the experience of a traumatic event and results in characteristic symptoms such as persistent fear reactions and systematic avoidance of traumatic reminders. It is largely accepted that associative processes are involved in the etiology and maintenance of PTSD and anxiety-related disorders. For instance, it is thought that environmental-contextual stimuli associated with the traumatic event can acquire the ability to elicit conditioned fear responses that can be triggered by subsequent encounters of these or similar stimuli during the course of normal life. The understanding of the precise neuronal circuits and mechanisms involved in pathological fear behavior are therefore of strong clinical importance. In this review we provide an extended review of recent studies evaluating the contribution of long-range neuronal circuits to the formation and expression of conditioned fear memories.

In the laboratory, fear memories are usually established using a simple and robust auditory fear conditioning learning paradigm during which a neutral stimulus such as a tone (the conditioned stimulus or CS) is repeatedly associated with a mild electrical footshock (the unconditioned stimulus or US). Over the past years, several structures including the basolateral amygdala (BLA), hippocampus, prefrontal cortex (mPFC) and periaqueductal grey matter (PAG) have been consistently involved in the formation, storage and extinction of conditioned fear memories. Recent technical developments such as optogenetic identification and manipulation of specific neuronal elements, genetic rodent models and large-scale recordings of neuronal populations have consider¬ably increased our capacity to dissect and understand the function of dedicated neuronal circuits regulating fear behavior. The emerging model of the neuronal circuits involved in fear behavior suggests the existence of parallel collaborative neuronal circuits and mechanisms involved in the acquisition or expression of learned fear behaviors. First, in addition to the activity-dependent plasticity that develops in the BLA during fear conditioning, recent studies have demonstrated a potentiation of BLA to central amygdala (CE) synapses dur¬ing fear learning, suggesting a potential CE gating mechanism for fear behavior. Moreover, it appears that thalamic and cortical sensory regions display activity-dependent plasticity during fear learning that could lead to the sharpening of frequency tuning curves toward fear-conditioned tones, a potential mechanism allowing fear discrimina¬tion.

Other studies have revealed that a nociceptive pathway through the PAG to the BLA supports an aversive teaching signal critical for fear learning that could be regulated by long-range amygdala-PAG circuit interactions, BLA local interneurons, and/or neuro¬modulatory mechanisms. Second, recent studies have revealed that fear expression could depend on multiple parallel neuronal circuits. One circuit directly modulates fear behavior through connections between the BLA and CE output neurons. In the CE, somatostatin-expressing interneurons (SOM+) also project to the PAG, where they can directly regulate conditioned fear responses. Another circuit relies on the projections of distinct sets of BLA neurons to the mPFC, and possibly the development of long-term synaptic plastic-ity or intrinsic plasticity mechanisms at BLA inputs to this re¬gion. Finally, in mPFC-BLA circuits, the recruitment of specialized neuronal populations such as parvalbumin-expressing interneurons (PV+), the development of neuronal oscillations and the synchronization of prefrontal output neurons contacting the BLA are potential neuronal mechanisms that could allow for the precise regulation of fear expression.

The conditions in which the different neural circuits and mech¬anisms mediating fear acquisition and expression are selected are still largely unknown, but could depend on the complexity of the behavioral task, the strength of the CS and US inputs activated during conditioning, internal states, or environmental situations that may impose the selection of distinct neuronal circuits to produce an appropriate behavioral output. From a clinical standpoint, it is clear that dysfunction in associative processing in amygdala and prefrontal neuronal circuits are at the core of pathological fear behav¬ior occurring in anxiety disorders such as post-traumatic stress disor¬der. Understanding the precise plasticity and neuronal mechanisms occurring in dedicated neuronal elements and across distributed cir¬cuits during fear behavior will be instrumental for the development of new therapeutic strategies for these psychiatric conditions.