Plasticity of excitatory synapses in the central nervous system has been reported to involve a host of mechanisms including alterations in neurotransmitters (vesicle size, concentration, release rate, probability of release), receptors (types, kinetics, numbers and conductance) and neuronal shape (postsynaptic density, size and spine morphology). Determining the contribution of each of these properties is limited by the recording techniques employed. We expanded a cable model of a neuron with many of these mechanisms (Benke et al. 1998) to include variations in synaptic vesicle volume (coefficient of variation = 0.4), a kinetic scheme for both AMPA- and NMDA-type glutamate receptors, multiple synapses, added background noise and branching dendrites. Assuming a typical background noise level of 1.8 pA root mean square, simulation of synaptic events recorded with a dendritically placed patch-clamp electrode (35 M¢ access resistance) revealed that synapses with as few as 3-4 AMPA receptors could be reliably separated from background noise. This assumes the presence of the saturating (3 ± 1.2 mM (mean ± S.D.)) peak glutamate concentrations used. Under these conditions the peak open probability for AMPA receptors is 0.79. If sub-saturating concentrations (0.5 ± 0.2 mM) of peak glutamate are used, 8-9 AMPA receptors could be reliably separated from background noise. These data compare with previous estimates from dendritic recordings that excitatory postsynaptic currents (EPSCs) generated by fewer than four activated AMPA receptors are detectable (Isaac et al. 1998). Thus electrically detectable synapses could contain no more than 2-3 AMPA receptors under dendritic recording conditions.This work was supported by the NIH and the MRC.

Benke, T.A., Luthi, A., Isaac, J.T.R. & Collingridge, G.L. (1998). Nature 395, 793-797.

Isaac, J.T.R., Luthi, A., Palmer, M.J., Anderson, W.W., Benke, T.A. & Collingridge, G.L (1998). Neuropharmacology 37, 1399-1410.