Fig. 3. Major transitions in evolution of the endomembrane system. In prokaryotes, secretion is a comparatively simple matter of translocation of polypeptides across the plasma membrane. Although there are several distinct mechanisms for achieving this, all appear to require an unfolded substrate for translocation. The type I, SRP/SecY-mediated, pathway is homologous to the co-translational ER import pathways of eukaryotes. In the hypothesized LCEA, comparative genomic evidence suggests that the major structures and pathways constituting the endomembrane system were already present, including the ER, Golgi complex and the main features of the endocytosis and recycling systems. The paralogous relationship of the families of SNAREs, Rabs, SM proteins and GTPases, as well as the homology of many coat components to each other and also to components of the nuclear pore complex, provide a potential mechanism for how this system arose. Compartmentalization and gene family expansion led to establishment of multiple protein systems capable of deforming membranes (i.e. transport steps). Elaboration of this basic pattern has been a major driving force for subsequent diversification of the endomembrane system, giving rise to the array of systems present in extant taxa. Subsequent evolution yielded multiple modes of endocytosis, specialized exocytic pathways and increased complexity of post-Golgi pathways. The red ovals indicate a trans-membrane translocation system. Lys, lysosome or vacuole; LCEA, last common eukaryotic ancestor. Small numbers in red indicate an associated pathway-specific Rab protein. The grey structure in the prokaryote indicates the non-compartmentalized genome.