Microcircuits in action – from CPGs to neocortex

To understand the interface between global brain function and molecular neuroscience – that is, the microcircuit level – a major challenge. Such understanding is prerequisite if we are to account for neural function in cellular terms. Very few vertebrate microcircuits are yet understood because their analysis is demanding technically. In this review of the TINS Microcircuits Special Feature, we attempt to shed light on the problem by comparing the operation of four types of microcircuit, to identify common molecular and cellular components. Central pattern generator (CPG) networks underlying rhythmic movements and hippocampal microcircuits that generate gamma and theta rhythms are compared with the neocortical microcircuits used in cognitive tasks and a cerebellar network. The long-term goal is to identify the components of a molecular and synaptic tool kit for the design of different microcircuits.

Introduction

Recent progress in neuroscience has been phenomenal on two extreme levels: the molecular biology of neurons and synapses on the one hand, and brain-imaging techniques on the other. The latter have provided important new knowledge about which parts of the brain are involved in different cognitive, emotional and motor functions. The maximal resolution for functional maps is the collective activity of thousands of neurons, whereas for neuronal and synaptic function it is the activity of individual molecules. The vast conceptual gap between these two levels is wide, bridged by only few and feeble strands. To understand the progressive emergence of complex function from the molecular level, it is essential to bridge this gap. Fortunately, the brain is modular: it consists of various networks that serve different functions in different brain regions, from the forebrain to the spinal cord. These functional modules, whether cortical columns or networks serving motor functions, form the interface between the molecular and higher-brain levels. One major challenge of current neuroscience is to unravel the designs and modes of operation of these modules, which we refer to as ‘microcircuits’^*.

The reason why until recently so little has been learnt about the dynamic processing that goes on in microcircuits, such as those of the neocortex and hippocampus, is that it has been technically demanding to study microcircuits systematically. We need to be able to identify the main types of neurons involved, draw a detailed circuit diagram, determine the forms of transmission and plasticity between different neurons, and study the dynamics of the microcircuit at cellular and synaptic resolution. Simulations are ultimately essential to explore whether a given set of data actually can account for a given function (De Schutter et al., in this issue). A multitude of demanding techniques have to be combined to address these issues, such as simultaneous intracellular recording from two or more interacting nerve cells. These techniques have now been successfully applied to several vertebrate circuits – the topic of this review – and previously also to certain invertebrate ‘simple systems’ circuits in crustaceans and molluscs 1, 2, 3, 4.

Although the neurons, synapses and circuit diagrams of different vertebrate microcircuits can differ considerably, it is remarkable that all microcircuits identified so far in the brain display rhythmic behaviours, each at different preferred frequencies. In this TINS Microcircuits Special Feature, we explore the molecular, cellular, synaptic and circuit designs for the different forms of rhythmic behaviour observed in these microcircuits. This is crucial in a more general framework because it enables us to bridge the gulf from the single-cell level to integrated functions of the whole nervous system.