After many years of work that had given only limited insights into the proteins involved in regulated exocytosis, several soluble and membrane proteins that are essential or regulatory components of the exocytotic machinery have been identified recently. The picture that is emerging suggests that several proteins are likely to act in discrete steps in exocytosis or act together to form some kind of fusion machine. Further components are likely to be identified in the near future and a major area for investigation will be the nature of the interactions between these proteins. Work on enveloped viruses has identified proteins with specific fusion peptide sequences that mediate viral fusion with the plasma membrane. These proteins act in binding the virus to the membrane as well as in membrane fusion. The fusion peptides from various viruses do not possess any marked sequence similarity, but all of them can potentially fold as a helix with a hydrophobic domain on one side. A protein that mediates sperm fusion with the egg has structural similarities to these viral proteins So far, no protein has been identified as being involved in exocytosis that has the properties of these fusion peptides. It will be interesting to see if intracellular bilayer fusion turns out to involve mechanisms related to or distinct from that in extracellular fusion, although one important point in these considerations will be the different lipid compositions of the inner and outer leaflets of the plasma membrane bilayer. It is interesting that viruses have evolved quite distinct fusion peptide sequences to solve the same biological problem and so it is conceivable that exocytosis could involve quite different proteins. One important consideration is the relationship between constitutive and regulated exocytosis and between exocytosis and other intracellular transport vesicle steps. As described above, certain proteins known to be required for various intracellular vesicular transport steps (NSF and SNAPs) have been implicated in neurotransmitter release and in constitutive exocytosis In addition, annexins have been implicated in various membrane fusion events in the endocytic pathway It is not yet clear to what extent the same or similar proteins are involved in membrane fusion of both constitutive and regulated secretory vesicles, but homology has been noted between nerve terminal proteins involved in Golgi-to-plasma-membrane transport in yeast Since constitutive secretory vesicles can readily fuse with the plasma membrane immediately after their formation this suggests that fusion of regulated vesicles must be inhibited at resting [Ca2+](i). One possibility is that Ca2+](i). acts primarily to disinhibit the fusion mechanism. The inhibition could be due to the cytoskeleton acting as a cortical barrier, or holding regulated vesicles in a network (e.g. via synapsin I) to prevent exocytosis, or to additional mechanisms. Recent work on the biogenesis of small synaptic-like microvesicles in PC12 cells has suggested that synaptic vesicle proteins may be exported from the trans-Golgi network to the plasma membrane in constitutive secretory vesicles, recycled via the endosome pathway and only then become segregated to the regulated secretory vesicle. The implications of this is that the synaptic vesicle proteins begin their life in vesicles (constitutive and recycling endocytic) that can spontaneously fuse with the plasma membrane but then become restricted to vesicles that can only fuse in response to a [Ca2+(i)] rise.