On the basis of the information presented in this review, it is difficult to reach any firm decision regarding the role of cyclic AMP (or cyclic GMP) in synaptic transmission in the brain. While it is clear that cyclic nucleotide levels can be altered by the exposure of neural tissues to various neurotransmitters, it would be premature to claim that these nucleotides are, or are not, essential to the transmission process in the pre- or postsynaptic components of the synapse. In future experiments with cyclic AMP it will be necessary to consider more critically whether the extracellularly applied nucleotide merely provides a source of adenosine and is thus activating an extracellularly located adenosine receptor, or whether it is actually reaching the hypothetical sites at which it might act as a second messenger. The application of cyclic AMP by intracellular injection techniques should minimize this particular problem, although possibly at the expense of new difficulties. Prior blockade of the adenosine receptor with agents such as theophylline or adenine xylofuranoside may also assist in the categorization of responses to extracellularly applied cyclic AMP as being a result either of activation of the adenosine receptor or of some other mechanism. Ultimately, the development of highly specific inhibitors for adenylate cyclase should provide a firm basis from which to draw conclusions about the role of cyclic AMP in synaptic transmission. Similar considerations apply to the actions of cyclic GMP and the role of its synthesizing enzyme, guanylale cyclase.
The use of phosphodiesterase inhibitors in studies on cyclic nucleotides must also be approached with caution. The diverse actions of many of these compounds, which include calcium mobilization and block of adenosine uptake, could account for many of the results that have been reported in the literature.