Silent synaptic connections and their modifiability.
Comparison of the two afferent systems illustrates certain features common to synaptic transmission as well as differences that might be important for synaptic plasticity. Transmission at both the inhibitory and excitatory connections is satisfactorily described by a simple binomial model that considers the average probability of release to be the same at each active site, although it should be stressed that the best evidence derives from the first set of afferents. Another similarity between the two systems is that short-term changes in synaptic efficacy, namely, facilitation and depression, appear to be due to changes in p. We previously suggested that both phenomena occur during repetitive stimulation, with the dominant effect depending upon the initial probability of release. It remains to be seen if depression dominates at other inhibitory connections, although it is already clear that one cannot generalize about excitation, because some excitatory junctions have an initial high p and exhibit a marked depression rather than the facilitation described here. We have found no evidence for the notion that some synapses within a connection may be silent. That idea has been proposed, but not proven, for other synaptic connections in the vertebrate central nervous system. Indeed, it will be difficult to assess as long as quantal release cannot be reliably detected at these junctions, and morphological confirmation at the ultrastructural level will also be required. On the other hand, evidence from a few peripheral junctions where one presynaptic afferent establishes hundreds of contacts with its target cell, does suggest the possibility of silent synapses, or at least an extremely low probability of release in those cases. These situations may correspond to extremes of our finding that as the number of release sites increases, p decreases. Regardless, the inverse relation between n and p suggests caution should be exercised in interpreting data indicating that synaptic plasticity is associated with increased numbers of synapses between two cells. Although we have not detected silent synapses within a transmitting connection, we have observed chemically silent connections between neurons, and the evidence reviewed here suggests transmission may be blocked postsynaptically, as with the inhibitory connections, or presynaptically, as with the excitatory ones. Although the underlying mechanisms are only partially elucidated, it is also clear that such connections can be switched into a transmitting mode. Consequently, they may provide a significant reserve that might well become functional in different behavioral states or in response to certain patterns of activity.(ABSTRACT TRUNCATED AT 400 WORDS)