Dendro-somatic Coupling and global neuronal signaling

Introduction

The mechanism for anaesthesia has eluded neuroscience despite being used as a medical intervention for more than two centuries. Two facts are clear: loss of consciousness under anaesthesia correlates with a massive reduction in global neuronal signalling and a de-coupling of feedback information.

Findings

We recently found that the apical dendrites of layer 5 pyramidal neurons are effectively "cut off" from the cell body under anaesthesia. Higher order thalamic input maintains dendro-somatic coupling via the activation of metabotropic glutamate and acetylcholine receptors.

Hypothesis

We hypothesize that dendro-somatic coupling, the influence of the apical dendrite on the soma, regulates the flow of information around the brain. If true, this would mean that dendro-thalamic coupling is a ubiquitous mechanism by which the brain, via thalamo-cortical interactions, can regulate the degree to which feedback is reintegrated into any given cortical column.

Moreover, since human pyramidal neurons are twice as long as rodent neurons, the issue of dendro-somatic coupling in human neurons is even more profound.

Proposed Research

Here, we propose to comprehensively investigate this cortical coupling mechanism in vitro and in vivo using state-of-the-art approaches developed in my laboratory and specifically designed to investigate dendritic signalling.

1. We will perform experiments in resected cortical tissue from human patients using fast-viral expression of optogenetic constructs, an approach that we have also developed and tested in my laboratory.
2. Lastly, we will use modelling to probe the consequences of dendro-somatic coupling both for single-cell computation and the principles of multi-compartment neuronal networks.

Conclusion

Dendro-somatic coupling as a dynamic mechanism for global neuronal signalling is a bold new perspective for which we have clear evidence. This stands to revolutionize our understanding of the cortex and possibly also provide insights into more efficient neural network architectures.