Over the last 50 years, mechanisms controlling the release of neurotransmitter have been studied intensively (see reviews by Parnas et al. 1994, 1997; Zucker, 1996). Despite these efforts, the molecular mechanisms that transduce the physiological signal, the action potential, into release of neurotransmitter remain a mystery. To understand these mechanisms, it is useful to study two variables: the amount of release (quantal content) and the time course of release (reflected by the synaptic delay histogram; Katz & Miledi, 1965), especially if (as argued below) different mechanisms control these two variables.

It is widely accepted that entry of Ca2+ and its accumulation to high levels near release sites triggers the initiation of release. Furthermore, the rapid removal of Ca2+ from the release sites has been hypothesized to be responsible for cessation of release (the Ca2+ microdomain hypothesis; see review by Zucker, 1996). Accordingly, a change in the kinetics of entry and removal of Ca2+ is expected to affect both the quantal content and the time course of release. For example, a smaller quantal content should be associated with a shorter period of release. However, this straightforward prediction is not met experimentally. For example, Andreu & Barret (1980) showed that in the frog neuromuscular junction (NMJ), the time at which release starts and stops - the time course of release - remained the same even with a many-fold change in the quantal content. In other studies, synaptic delay histograms were found to be independent of experimental manipulations known to affect Ca2+ entry or removal. For example, when a fast Ca2+ buffer was injected into presynaptic terminals (Hochner et al. 1991), the quantal content declined markedly, indicating a faster removal of Ca2+, but the reduction in quantal content was not accompanied by a change in the time course of release. Taken together, these findings suggest that different processes control quantal content and the time course of release.

In most studies concerned with modulation of release (e.g. facilitation, depression and post-tetanic potentiation), only the quantal content is measured, either directly or as a change in the amplitude of synaptic potential or synaptic current. Generally, the effect of the experimental manipulation on the time course of release has not been investigated.

The article by Bukcharaeva et al. (1999) in this issue of The Journal of Physiology is a rare example of a study in which quantal content and the time course of release were measured simultaneously. Using the frog NMJ in controls and after treatment with noradrenaline (NA) these authors showed that after application of NA, quantal content remained unaltered whereas the duration of release became shorter. In the studies cited above (Andreu & Barret, 1980; Hochner et al. 1991), the experimental manipulation affected the quantal content but did not alter the time course of release. The study of Bukcharaeva et al. (1999) is the first case where the reverse happens: the time course of release changes without any effect on the quantal content. This finding suggests, once again, that different mechanisms are involved in the control of quantal content and the time course of release.

It is well accepted that quantal content (but not the time course of release) is very sensitive to even small changes in extracellular or intracellular concentrations of Ca2+. It is therefore reasonable to assume that the acceleration in termination of evoked release, which was not accompanied by a change in quantal content, did not result from a lower level of Ca2+ entry or from faster Ca2+ removal (Hochner et al. 1991). Understanding the molecular mechanisms behind this important finding may shed some light on the mechanisms that control the time course of evoked release, namely synchronization of the quanta released after a depolarizing impulse.

Since NA affected the time course of ACh release it is possible that some heteroreceptors are involved, not only in modulation but also in control of neurotransmitter release. Recently, it has been found that the muscarinic autoreceptor of the m2 subtype physically interacts with syntaxin, synaptosomal-associated protein of 25 kDa (SNAP-25), vesicle-associated membrane protein (VAMP) and synaptotagmin, all proteins of the exocytotic machinery (Linial et al. 1997). To fully appreciate the possible involvement of presynaptic heteroreceptors such as the β-adrenergic receptor in the process of ACh release, experiments to examine whether there is a physical interaction (direct or via a G-protein) between this receptor and the soluble NSF-attachment protein receptor (SNARE) proteins will be beneficial.

Another physiological phenomenon is also elucidated by the work of Bukcharaeva et al. (1999). It is known that NA increases muscle contraction, and this has been attributed to an increase in quantal content. However, Bukcharaeva et al. (1999) show that the larger synaptic potential actually results from a more synchronized release, even though quantal content is the same. Thus, more than one presynaptic mechanism can control synaptic efficacy.