Analysis of synaptic transmission in the brain using purely electrophysiological techniques has proved problematic because of the large number of synapses on most neurons. The advent of confocal microscopy has made it possible to circumvent this problem by studying synaptically evoked calcium transients at single dendritic spines in organised brain tissue. We have examined two kinds of synapse in organotypic hippocampal cultures prepared from 8 day postnatal male Wistar rats, maintained for 14Ð21 days in vitro prior to recording: associational and Schaffer-collateral synapses at small spines on distal apical dendrites of CA3 and CA1 pyramidal cells, respectively, and mossy fibre synapses at complex dendritic spines (Cajal’s ‘thorny excrescences’) on proximal dendrites of CA3 pyramidal cells. Cells were filled with a high-affinity Ca²⁺ indicator (Oregon Green 488 BAPTA-1, Molecular Probes), and activated by stimuli to stratum radiatum (to activate Schaffer/associational fibres), or to granule cells (to activate mossy fibres). At Schaffer/associational synapses, the synaptically evoked calcium transient is NMDA-receptor dependent, and is abolished by drugs that suppress release of calcium from intracellular stores (Emptage et al. 1999). Consistent with this finding, in serial-section EM studies using immunogold labelling of ryanodine receptors (RyRs), we find that RyRs are expressed in over 60 % of spines in distal apical dendrites of CA1 pyramidal cells. A slightly lower proportion of spines displayed immunoreactivity for the RyR accessory protein, sorcin. In contrast, the main source of the synaptically evoked calcium transient in thorny excrescences is calcium entry through NMDA receptors and voltage-dependent calcium channels (Reid et al. 2001).

We have used synaptically evoked postsynaptic Ca²⁺ transients to analyse changes in the quantal components of synaptic transmission at individual Schaffer/associational and mossy fibre synapses following the induction of long-term potentiation (LTP). In 8/13 Schaffer/associational synapses, LTP was accompanied by an increase in the probability of a calcium transient (P_Ca) 30 min post-induction (mean pre-tetanus vs. post-tetanus values of P_Ca, ± S.E.M., for the group of 13 spines was 0.26 ± 0.07 pre-LTP vs. 0.56 ± 0.10 post-LTP, P < 0.007, 2-tailed paired t test); in a control population of 14 spines, imaged on cells in which LTP was blocked by D-AP5 (50 mM), or in which low-frequency trains were delivered that failed to induce persistent changes in the intracellularly recorded EPSP, there was no change in P_Ca over a similar time period (P_Ca was 0.33 ± 0.06 pre-treatment vs. 0.35 ± 0.06 post-treatment, n.s). An increase in the amplitude of the calcium transient was also observed in many cases 30 min after the induction of LTP. These observations indicate a presynaptic component to the early phase of LTP at Schaffer/associational synapses, as well as a novel postsynaptic biochemical component of LTP; they do not preclude additional postsynaptic components of LTP expression. At mossy fibre synapses, where expression of LTP is believed to be exclusively presynaptic, we found that an increase in P_Ca was also common; in some cases the increase in P_Ca was accompanied by the emergence of higher amplitude synaptically evoked calcium transients which at these multi-release site synapses can be explained by the recruitment of additional release sites following the induction of LTP. In some cases, mossy fibre LTP was associated with the activation of previously silent synapses.

We thank Dr F.A. Lai for pan-RyR antiserum, and Dr M. Meyer for sorcin antiserum. This work was supported by the MRC, and grants from the Human Frontier Science Program, European Union Framework 5, and Canadian Institutes of Health Research. C.A.R. was a Howard Florey Fellow of the Royal Society; R.F.-F. is a Killam Fellow (Canada).

All procedures accord with current local guidelines.