A fundamental neuronal microcircuit for learning

Imaging the activity of amygdala interneurons: Individual VIP+ interneurons labe

Imaging the activity of amygdala interneurons: Individual VIP+ interneurons labelled with the indicator for calcium imaging (left) and corresponding antibody staining of the peptide VIP (right).



How does the brain control the mechanisms of memory so that it only remembers major events in a constantly changing environment? The group of Andreas Lüthi has now described a fundamental neuronal microcircuit that allows mice to learn about unexpected important events and adapt their behavior accordingly.

In daily life, our brain constantly receives sensory information from our environment, yet we will only remember a small fraction of it. Why do we recall certain experiences of our lives but forget about others’ Powerful triggers for memory formation are both pleasant and aversive events, which produce powerful emotional responses. For example, some of the earliest memories from our childhood might include our first ice cream on a sunny day at the beach, or the neighbor’s vicious dog frightening us through the fence. But how exactly can our brains translate important experiences into neural signals to trigger memory formation?

A key structure for emotional memory in the brain is the amygdala. This brain region mainly consists of principal neurons, which are considered to be the main site of plasticity during learning. However, a minority of interneurons can powerfully inhibit the activity of principal neurons and therefore play an important role in local information processing in the brain. As the Lüthi group showed in a 2014 study , two types of interneurons are key mediators of the learning process: PV+ and SOM+ interneurons provide constant inhibition to principal cells, preventing that cellular learning mechanisms can be triggered in absence of important sensory stimuli.

More recently, Sabine Krabbe, a postdoc in the Lüthi group, together with Enrica Paradiso, a visiting scientist from the Medical University of Innsbruck, investigated the role of interneurons in the learning process in more detail; they just published their results.

Using auditory fear conditioning as a model for learning, the researchers trained mice to associate a sound with an unexpected unpleasant stimulus (a light foot shock), so that the auditory cue became predictive of the aversive event. By monitoring the activity of amygdala neurons during fear learning with a miniaturized microscope, they identified another interneuron type called VIP+ cell, which got strongly activated during the unpleasant sensory stimulus thereby inhibiting other interneurons. By combining miniature microscopy with optogenetics (to selectively deactivate VIP+ cells), they showed that the aversive stimulus can indeed ‘open a disinhibitory gate’ for principal neurons during which plasticity and learning is made possible by relieving inhibition onto principal neurons.

“The brain has to control the mechanisms of memory so that it only remembers important events. These various types of interneurons highlighted in our study can be described as gatekeepers that are only going to activate the principal neurons when something important, that’s worth learning, happened,” Lüthi explains. “Our study elucidates how the gatekeepers work and highlights a key mechanism by which important events control local microcircuit computations and plasticity. This allows us to learn about aversive experiences, so that we can adapt our behavioral strategy to avoid these dangers in the future.” Lüthi concludes: “These findings are of interest in the context of conditions where learning processes are impaired or dysfunctional, as in the case of anxiety disorders.”

Original publication:
Sabine Krabbe*, Enrica Paradiso*, Simon d’Aquin, Yael Bitterman, Julien Courtin, Chun Xu, Keisuke Yonehara, Milica Markovic, Christian Müller, Tobias Eichlisberger, Jan Gründemann, Francesco Ferraguti and Andreas Lüthi (2019). Adaptive disinhibitory gating by VIP interneurons permits associative learning. Nature Neuroscience
* These authors contributed equally