Trends in Neurosciences
TechniquesUnveiling synaptic plasticity: a new graphical and analytical approach
Section snippets
Construction and visual interpretation of a variance–mean plot
Postsynaptic currents (PSCs) are recorded under several different release probability conditions (typically 20-200 PSCs under each condition; Box 1). r can be adjusted by altering the Ca2+ to Mg2+ ratio or by adding cadmium (Cd2+) to the extracellular solution. The variance and the mean of the PSCs are calculated during a stable recording epoch after wash-in of each solution, and the variance is plotted against the mean (Box 1). The general form of the variance–mean (V–M) plot is parabolic.
Quantitative analysis of a V–M plot
The graphical approach outlined above can be made more quantitative by fitting a V–M plot with an equation that incorporates one or more of the synaptic parameters, , r and N (Box 1, Box 2). The locus of expression of a modulation can be determined by comparing synaptic parameters obtained before and after the modulation. This quantitative approach also permits a comparison to be made between the functional and morphological parameters of a particular synapse, or between the functional
Analysis of V–M when r is low or moderate
When a V–M plot is approximately linear, r is in the low range (<0.3) and the plot can be analysed by fitting the equation for a straight line8 (Box 1). This permits an estimate of average quantal amplitude, w, but r and N cannot be estimated under these conditions. The ‘w’ subscript indicates that w is a weighted average that emphasizes terminals with higher release probabilities and larger postsynaptic amplitudes. The weighted average will be slightly larger than the arithmetic average, .
Analysis of V–M when r is high
Further insights into synaptic behaviour are possible if data are available over a wider range of mean release probabilities. A full description of the synaptic behaviour can be made at nonuniform synaptic connections, with mean quantal parameters (, r and N) being estimated, rather than the weighted means. This estimation can be achieved by fitting the V–M plot with an equation for a modified parabola (Box 2). However, a larger number of experimental observations are required to estimate
Practical considerations
When constructing a V–M plot, the variance and mean must be calculated over an epoch where the recording conditions are stable. Unfortunately, a number of processes can cause the synaptic variance to be overestimated. These include depression or rundown of the synaptic response, a change in the series resistance, activation of active dendritic conductances and the occasional failure of the stimulus to excite a presynaptic axon. Standard precautions can be taken to guard against these
Concluding remarks
A new method for identifying the locus of expression of synaptic plasticity and for estimating functional synaptic parameters has been developed. The method is easy to apply, because it requires little or no mathematical analysis and will work even when synaptic properties are nonuniform. It is hoped that V–M analysis will prove to be a useful tool for understanding the functional basis of synaptic behaviour and diversity.
Acknowledgements
The authors’ research is supported by an ARC Senior Fellowship from the Australian Research Council (J.D.C.) and an RCDF from the Wellcome Trust, UK (R.A.S.). The authors thank Mark Farrant and Sharon Oleskevich for useful comments on the manuscript.
References (34)
The statistics of transmitter release at chemical synapses
- et al.
Postsynaptic expression of long-term potentiation in the rat dentate demonstrated by variance-mean analysis
J. Physiol.
(1999) Is maintenance of LTP presynaptic?
Nature
(1991)- et al.
Presynaptic enhancement shown by whole-cell recordings of long-term potentiation in hippocampal slices
Nature
(1990) - et al.
Getting the most out of noise in the central nervous system
Trends Neurosci.
(1998) Calcium channel types with distinct presynaptic localization couple differentially to transmitter release in single calyx-type synapses
J. Neurosci.
(1999)Heterogeneous release properties of visualized individual hippocampal synapses
Neuron
(1997)- et al.
Quantal components of the end plate potential
J. Physiol.
(1954) - et al.
Spontaneous subthreshold activity at motor nerve endings
J. Physiol.
(1952) Quantal analysis of synaptic potentials in neurons of the central nervous system
Physiol. Rev.
(1990)