Since the human genome contains only ~30,000 genes, it is unlikely that such complex synaptic connections can be determined by the pattern of gene expression alone. Furthermore, connections need to be modified in response to external stimulation to establish a meaningful link between the external environment and internal sensory representations. Storage of information by the nervous system relies on the continuous refinement of connections within the hippocampus and the cerebral cortex to shape the pattern of neural activity. Understanding the principles governing this refinement process and the processes that provide synaptic organization is therefore of considerable interest to elucidating how information is maintained in the brain.

This fundamental premise has been studied extensively. Hubel and Wiesel's pioneering work established that the pattern of neural activity plays a crucial role in shaping neural connections during early visual system development. Substantial work in the field of synaptic plasticity provided a greater understanding of how patterns of neural activity can modify synaptic connections. The central questions remaining are how neural networks form and are “tuned” by modifying the synaptic connections within neural networks. The answer to these questions can provide the bridge between knowledge at the synaptic level and the phenomena at the level of the neural network. To address these questions, one needs to determine the elementary properties of individual synapses and then identify the properties that play a critical role during the formation of neural networks. To test and refine hypotheses, one needs to monitor the dynamic process of neural network formation. We have carried out a series of experiments to address these questions in cultured hippocampal neurons. This reduced preparation allows us to study the dynamic processes of synapse formation and maturation. It also provides the accessibility necessary to carry out biophysical experiments on a single synapse. To check whether the conclusions derived from results obtained from reduced preparations are applicable to intact animals, we also carry out the experiments in intact animals.

Recently, we have focused on:

Organization of excitatory and inhibitory synapses in hippocampal dendrites. Each neuron in the central nervous system has several thousand excitatory and inhibitory synapses distributed throughout its extensive dendritic tree. Dynamic interactions of excitatory and inhibitory synaptic inputs play an important role in controlling the temporal pattern of neural activity, an essential feature of neural computation. We have developed extensive knowledge of the physiological properties of individual synapses. However, the modes of structural organization and functional interaction of excitatory and inhibitory synapses in the dendritic tree are still largely unknown. An understanding of these principles is important, because neural computation in functional neural networks involves the activation of hundreds of excitatory and inhibitory synapses. Furthermore, growing numbers of studies suggest that neural computation occurs within dendritic branches. To address these fundamental questions, we studied the patterns of organization of excitatory and inhibitory synapses on the dendritic tree. We found that excitatory and inhibitory synapses are balanced structurally and functionally in individual dendritic branches. This excitatory and inhibitory balance is established and maintained by a powerful “push-pull” regulatory mechanism, resulting in an optimal level of synaptic inputs in each dendritic tree. To explore the functional implications of this synaptic arrangement, we studied the functional interactions of excitatory and inhibitory inputs on dendritic trees and found that inhibitory synapses can determine the impact of adjacent excitatory synapses only if they are co-localized on the same dendritic branch and are activated coincidentally. This is the first experimental work that demonstrates the local interaction of excitatory and inhibitory synapses on the dendritic tree and supports early theoretic predictions of the properties of excitatory and inhibitory (E/I) interaction. Based on these novel findings, we propose that the even distribution of E/I synapses and the local nature of their functional interactions in dendrite branches set the structural and functional foundations for neural computation in dendritic branches.

Identifying the parameters of synaptic strength. A continuing goal of synaptic physiology is to identify the parameters of synaptic structure that most dramatically influence synaptic function. Collecting a list of such attributes is the first natural step in identifying the actual physiological switches used by networks of neurons to refine their connectivity. While work by many groups has successfully identified a host of postsynaptic mechanisms that determine synaptic strength, comparatively less progress has been made in discovering presynaptic mechanisms that contribute to synaptic strength, even though such mechanisms could influence postsynaptic activation by controlling the profile of transmitter release. Recent work in our lab has identified one of the first presynaptic sites of regulating specification of quantal synaptic transmission. We experimented with small transporter molecules that work to fill synaptic vesicles with neurotransmitter prior to transmission. By genetically enhancing the number of these transport molecules, we were able to boost excitatory transmission by presynaptically increasing the amount of transmitter loaded and released. This exciting result demonstrates that presynaptic factors, in addition to postsynaptic ones, can be important for determining the efficacy of synaptic transmission. Work in the lab continues to move forward in identifying new regulatory sites that neural networks can use to control the strength of their connections.

Selected Publications
Liu, G. Presynaptic control of quantal size: kinetic mechanisms and its implications in synaptic transmission and plasticity. Current Opinion in Neurobiology. 13: 324-331 (2003).

Liu, G. Local structural balance and functional interaction of excitatory and inhibitory synapses in hippocampal dendrites. (submitted) (2003).

Wilson, N.R., Kang, J., Varoqui, H., Leung, T., Murnick, J.G., Erickson, J.D., and Liu, G. Enhanced excitatory transmission via VGLUT1 overexpression. (submitted) (2003).

Murnick, J., Dubé, G.R., Krupa, B., and Liu, G. High-resolution iontophoresis for single-synapse stimulation. J. Neurosci. Methods. 116:65-75 (2002).

Renger, J.J., Egles, C., and Liu, G. A developmental switch in neurotransmitter flux enhances synaptic efficacy by affecting AMPA receptor activation. Neuron 29: 469-484 (2001).