Mechanisms underlying synapse-specific clustering of GABAA receptors

To recognise things around us, process information and respond in a useful, safe and socially acceptable way, the brain performs extremely complex computations. Our brains contain millions of nerve cells (neurones) which process information and transfer it to other neurones via synapses. Since there are many types of neurones, there are many different types of synapse. Even subtle changes at one type of synapse can produce behavioural, or emotional changes and contribute to neurological or psychiatric disease. This project focusses on inhibitory synapses which reduce activity in other neurones, blocking their responses to other inputs. They select precisely which information is processed and control inappropriate perceptions, responses and behaviour patterns. Many drugs affect their function, eg. anaesthetics, sedatives and anti-anxiety drugs, while changes at some of these synapses caused by changing hormone levels contribute to premenstrual tension, increased epileptic seizure susceptibility at some times of the month and to ?postpartum blues?.

At each synapse, a minute, highly specialised region of the output fibre of one neurone comes very close to the surface of another making a functional connection. On each side of the synapse so formed, proteins cluster into highly specific, complex functional units. These synaptic proteins are highly specialised components. We know something of their structures, their interactions with each other as they control information transfer and that subtly different components are used by different types of synapse. What we do not yet understand is how each of them is selected and inserted at just the right place, or precisely how each combination of components leads to one set of distinctive functional properties. A first requirement for this level of precision is for two neurones, one on either side of the synapse, to recognise each other. Neurone A might receive inputs from twenty different types of neurones and might generate output onto twenty different types of other neurones. It must therefore construct its own half of each of these synapses with enormous precision using just the right components at each one. The first question to be answered is therefore - how does it recognise the neurone on the other side ? The sheer complexity has, until recently, precluded a deeper understanding. The tools needed to probe further are however, becoming available and with them, new insight into the mechanisms that underlie this precision. We will combine these tools in two parallel, novel and complementary experimental approaches to the problem.

For two decades, neurophysiologists have documented selective expression of synaptic properties at different subclasses of central synapse. It is clear that postsynaptic neurones play a fundamental role in determining the properties of axon terminals innervating them and presynaptic terminals determine the receptors that cluster postsynaptically. This proposal asks how this is achieved.

Cleft-spanning proteins have recently become recognised as the elements mediating transynaptic recognition and with extensive alternative splicing, may provide the molecular diversity controlling functional diversity. Several proteins specific to glutamatergic, or GABAergic synapses have been identified and their importance in the development and stabilization of synapses demonstrated. However, few studies have attempted to go further to explore synapse specificity at subsets of excitatory or inhibitory synapses.

This project focusses on one key question; how is the specific clustering of only one subtype of GABA-A receptor at each class of GABAergic synapse achieved ? We propose two parallel approaches. First to explore the level of pre- and post-synaptic molecular complexity required to ensure selective clustering, by co-culturing neurones with non-neuronal cell lines expressing GABA-A receptors, neuroligin 2 and associated proteins. Immunofluorescence will indicate colocalisation of GABA-A receptors with synaptic proteins and dual whole cell recordings explore the functional integrity of synapses formed. Secondly, to identify receptor-associated proteins and splice variants that are selectively inserted at specific subsets of mature GABAergic synapses. Synaptosomes derived from GABAergic neurone subclasses will be purified, detergent-solubilised, affinity purified on a column decorated with GABA-A receptor extracellular domains and the proteins identified using Mass Spectroscopy.