Figure 5

The stepwise evolution of eukaryotic vesicle trafficking. From left to right the figure depicts a simple hypothesis for the evolution of the eukaryotic secretion and vesicle trafficking systems. Initially, proteins (black dots) would have been secreted from ribosomes bound to rough endoplasmic reticulum (ER) into the space at the bases of blebs by the Sec translocase and signal recognition particles (SRP) [50]. Secreted proteins could then undergo stepwise processing using machinery adapted from that used to process glycoproteins in the archaeal S-layer (that is, through N-linked glycosylation of asparagine-X-serine or asparagine-X-threonine-containing proteins, and proteolysis [99]). The elaboration of ER tubules and local membrane bending regulated by the Sar1 GTPase, in the presence of generic SNAP Receptors (SNAREs) (blue bars), would have enabled the transient fusion of ER to the outer cell membrane, releasing these glycosylated proteins into the extracellular space. These transient openings would have been closed by Dynamin-mediated fission. Specialized SNARE proteins (differently colored bars) and Dynamin (triple diagonal lines), would then have generated vesicular intermediates to better regulate secretion. The intercalation of additional processing steps and the diversification of these protein families would have yielded compartment-specific paralogs, together with the evolution of regulatory Arf and Rab GTPases, and a Golgi compartment. Finally, membrane bending machinery together with Dynamin, actin, and Rho family GTPases would have been co-opted to drive endocytosis, phagocytosis, and the development of the modern retrograde trafficking pathway.