Dousing the flames of fear

Abstract

Remembering a dangerous and thus frightful situation can be a lifesaver, but unchecked fearful memories can seriously affect the quality of life. Both the formation of fear memories and their extinction depend on the amygdala, a set of nuclei located deep within the temporal lobe. The way in which we and other mammals acquire and extinguish fear memories has been extensively studied over the last few years (Ehrlich et al. 2009). Importantly, instead of deleting the old trace, fear extinction forms a new memory that keeps the old, fearful association in check. What are the circuits that provide this control? Good evidence points to the involvement of inhibitory GABAergic neurons (Harris & Westbrook, 1998; Chhatwal et al. 2005; Heldt & Ressler, 2007). In the current issue of The Journal of Physiology, Manko et al. (2011) significantly add to our understanding of this process. They investigated intercalated neurons – a special class of projection GABAergic neuron that is found in several compact clusters around the basolateral nucleus of the amygdala and whose selective destruction causes a deficit in fear extinction (Likhtik et al. 2008; Amano et al. 2010). Manko and colleagues have studied the ‘main intercalated nucleus’ (Im), the largest, most ventrally located cluster of intercalated neurons – and curiously the least studied so far. On the one hand, they provide evidence that these intercalated neurons of the Im are less homogeneous than previously thought. They form three distinct groups of cells according to their intrinsic physiological properties. Studying their dendritic and axonal aspects and, physiologically, their synaptic inputs, Manko and colleagues mapped their local connectivity. Their results show that Im neurons receive synaptic inputs within the Im from diverse sources, including the basolateral nucleus. An important observation is that the influence of the intercalated network is more widespread than previously thought. Cells with the soma in the Im send functional projections to the central nucleus and basolateral amygdale as well as to extra-amygdaloid areas. On the other hand, they confirm findings that have been obtained studying other intercalated clusters (Marowsky et al. 2005) – in particular a prominent modulation of their activity by dopamine and the NMDA receptor-mediated component of their synaptic responses. Taken together, these results suggest that Im cells are subjected to activity-dependent changes with major behavioural consequences. A lot of work remains to be done to describe the circuitry that controls fear memory in detail. We know that intercalated neurons are crucial for fear extinction, but we do not know whether this function is mediated by all of them or only by specific subtypes. Characterising their baseline activity in vivo, and its changes upon emotional behaviour and the underlying mechanisms will be key to our understanding of how they fulfill their role. Finally and perhaps more challenging will be a complete mapping of their connectivity. The amygdala and the intercalated neurons receive many and diverse long-range inputs and project to the basal forebrain. These cannot be fully preserved in slice preparations and thus other methods, such as optogenetics or in vivo single-cell labelling may have to provide the necessary specificity. Overall, in spite of recent breakthroughs (Ciocchi et al. 2010; Haubensak et al. 2010), and of the potential behavioural and clinical importance of inhibition in the amygdala, our knowledge of its GABAergic circuitry remains surprisingly limited, and more work is necessary. We do not want to forget that something is frightening, rather we learn that it may – depending on the circumstances – actually be harmless. Change the context and the fear may be back. This intriguing and fragile regulation may save our lives, but it may also be crippling us with unnecessary fear. Miroslawa Manko et al. have just provided us with an important piece in the puzzle that describes this mechanism.