Apparently, the molecular mechanisms underlying docking and release are largely identical between different types of synapses. While we cannot exclude that major proteins have remained undetected, several lines of evidence suggest that we have achieved a high coverage of the docking site proteome. Foremost, all known active zone proteins (with the exception of Munc13) were identified in our mass spectrometric approach. Second, all proteins of the exocytotic machinery were recovered including the SNAREs, Munc18, complexin, and synaptotagmins. Moreover, a high coverage

of the protein inventory is supported when comparing the proteins identified GSK2118436 here with those found in the previous studies. For instance, the list of the presynaptic find more proteome reported by Morciano and colleagues (Morciano et al., 2009) contained 135 proteins (excluding mitochondrial proteins), 62 of which were also identified by us. Of the remaining

73 proteins, only few can be assigned to a specific presynaptic function (such as additional isoforms of membrane transporters) whereas most others are soluble proteins with general cellular functions. Similarly, the proteins that were identified by Abul-Husn and colleagues but not in our study (52 of 99 proteins) are also mostly general cellular proteins, with the exception of a group of proteins involved in clathrin-mediated endocytosis (Abul-Husn et al., 2009). We assume that soluble or only loosely

membrane-associated proteins were washed off during our isolation procedure. It needs to be borne in mind that the mild proteolysis required for separating pre- from postsynaptic membranes constitutes Linifanib (ABT-869) an inherent limitation for proteomic analysis. Thus it is not surprising that our recovery of cell adhesion molecules is somewhat lower than in the other studies (Abul-Husn et al., 2009; Morciano et al., 2009). These proteins possess only small intracellular but large extracellular domains that are expected to be degraded during the protease treatment of the synaptosomes. On the other hand, we identified a large number of plasma membrane residents documenting that the remaining intracellular regions are generally sufficient for protein identification. In this context it is notable that in neither our nor in any of the previous studies were receptors for neurotransmitters or neuromodulators found. While the function of such receptors in regulating presynaptic function is well established, many of these receptors likely function only in subsets of synapses and others may be expressed in low copy numbers, explaining why they may have escaped detection. Intriguingly, substantial overlap was also found with the proteome of protein complexes associated with presynaptic calcium channels that were isolated by immunoprecipitation of Cav2 after detergent extraction (Müller et al., 2010).