Take a guess: how many calcium channels are there in a dendritic spine? 50? 100? 500? Well, guess again. By using calcium imaging and relatively simple statistical methods, Sabatini and Svoboda have obtained evidence that there are no more than 20, typically around five channels per spine. This amazing observation is likely to put a new perspective on our sense of scale when thinking about the molecular organization of the synapse.

Sabatini and Svoboda measured action-potential-mediated calcium transients in individual spines, making sure that the rise in the intracellular cation was due to opening of spine channels, and not to dendritic calcium channels. They measured the trial-to-trial fluctuations in the amplitude of the transients and analysed them using the same statistical tools used for conventional quantal analysis of synaptic transmission. They calculated the coefficient of variation for the calcium increases and the probability of failure to elicit a transient to then obtain estimates of the number of calcium channels and their probability of opening at individual spines. Their methods, which make it possible to image the function of single channels, allowed the authors to show that calcium channels had a 50% chance of opening on any given trial. In addition, they found that the number of channels was between one and 20, depending on the size of the spine.

Perhaps this paucity of channels is not so surprising if we consider that, based on structural considerations, few glutamate receptors have similarly been predicted to exist in the postsynaptic density. What is more surprising is that such a handful of molecules can endow the synapse with its proverbial capacity for plasticity. The fact that there are so few molecules of any given type implies that changes in one or two might be enough to alter significantly the efficacy of synaptic transmission, and implies the existence of regulatory mechanisms that are capable of maintaining the delicate balance among the few components of the synapse. Is there an intimate, almost 'personal', recognition between a channel or a receptor and the proteins in charge of regulating its activity? What mechanisms ensure that the right number of channels will be present at a spine of a given size? The observations of Sabatini and Svoboda raise profound questions about the workings of the synapse and highlight the potential of optical methods to tackle questions that have remained elusive to our powerful arsenal of electrophysiological techniques.