Short Term Plasticity of Synaptic Transmission

calyx of Held schematic In collaboration with Prof. Ian Forsythe (MRC Toxicology Unit, University of Leicester), we are developing a variety of mathematical models of synaptic transmission at the calyx of Held, a giant synapse in the mammalian auditory system. The aim is to illuminate the pre- and postsynaptic biophysics underlying short-term modulation of the synaptic response. We are using the models to explore the information carrying capacity and signal filtering characteristics of this synapse in an effort to understand its functionality.

calyx of Held in Lucifer Yellow

Mathematical Models
Experimental Data
Signal Transmission
Volume Transmission
Funding
Key References
brainstem auditory nuclei

Mathematical Models of the Calyx [top]

A number of variations on the model illustrated above have been developed as an aid to understanding the contributions of different molecular mechanisms (including vesicle recycling and release, calcium channel inactivation, metabotropic glutamate receptor activation and AMPA receptor desensitization) to short term plasticity (depression and facilitation) at the calyx of Held (Hennig et al., 2008; Hennig et al., 2007; Graham et al., 2004; Graham et al., 2001).

These models include stochastic versions that track and release individual vesicles of neurotransmitter (Yang et al., 2009; Wong et al., 2003; Graham et al., 2001) and simpler average response (mean field) models that use deterministic ODEs to represent the average size of the readily releasable vesicle pool and average amount of transmitter release (Hennig et al., 2008; Hennig et al., 2007; Billups et al., 2005; Graham et al., 2004).

The models have sufficiently few parameters that mathematical optimization can be used to match the model with experimental data. Tight fits have been obtained to a variety of experimental data showing depression in EPSC amplitude at various frequencies of stimulation.

Matches to Experimental Data [top]

Detailed modelling of transmitter release and AMPA receptor kinetics predicts that acceleration in EPSCs with increasing temperature may purely be due to acceleration of the AMPA receptor kinetics (Postlethwaite et al., 2007).

The models have also been used to explore the possible effects of presynaptic metabotropic glutamate receptors on EPSC amplitude during depression and recovery (Billups et al., 2005). In line with the data, the models demonstrate that the effects of mGluRs are only observed during recovery from depression.

Experimental data from the Forsythe laboratory indicate that desensitization plays a significant role in short-term depression. They have developed a novel assay for desensitization using low-affinity competitive glutamate antagonists, the efficacy of which has been confirmed by the model (Wong et al., 2003).

Information Transmission and Signal Filtering [top]

We have used the models to calculate the amount of information the postsynaptic EPSC amplitude carries about the previous interspike interval of action potentials arriving at the synapse. We have shown that the EPSC amplitude carries significant information about the presynaptic ISIs even at high (>100Hz) mean frequencies of stimulation (Yang et al., 2009; Graham, 2002). The main information carrying mechanism changes with stimulation frequency.

Short term synaptic plasticity results in a synapse acting as a filter of the incoming spike train. Both depression and particularly the rate of recovery of the EPSC amplitude are stimulus-history-dependent (Hennig et al., 2008). The recovery time course depends on both the loudness and duration of a sound stimulus that generates the incoming spike train.

In a collaboration with Dr Christian Stricker (Neuronal Network Laboratory, John Curtin School of Medical Research, Australian National University), I have investigated the filtering properties of synapses in response to more complex dynamic signals (Graham and Stricker, 2008).

Volume Transmission [top]

NO spread in the MNTB Synaptic and cellular properties in the MNTB are modulated by volume transmission signals generated through activity-dependent synthesis of nitric oxide (NO). Work in the Forsythe laboratory has demonstrated the NO is produced in the MNTB under physiological conditions and subsequently modulates cell excitability and synaptic efficacy (Steinert et al, 2008). I have modelled the production and diffusion of NO in response to stimulation of MNTB neurons through input to the calyx (Steinert et al, 2008).