Illuminating the secretory pathway

Proteins destined for secretion, the plasma membrane or membranous cellular organelles enter the secretory pathway in the endoplasmic reticulum (ER) and move via the Golgi complex to the trans-Golgi network, where they are sorted and transported to their final destination. Malfunction in this pathway may lead to the breakdown of cellular organelles, cell polarity, overall cellular architecture and ultimately cell death. A fundamental question in the understanding of the secretory pathway is how the identity of the membrane-bounded structures involved is maintained. Each transport step between adjacent membranes must be tightly controlled at at least four basic levels: (i) sorting of secretory cargo from residents, which have to remain behind at the donor membrane; (ii) formation and transport of the cargo carriers; (iii) delivery of the cargo at the donor membrane; and (iv) recycling of the transport machinery to the donor membrane, which is essential for subsequent rounds of transport. Most of our knowledge of these processes is based either on ultrastructural studies that aimed to localise molecules precisely to individual membrane structures of the secretory pathway, or on in vitro studies using simplified systems to identify the key players and characterise their biochemistry. Together these approaches have given rise to widely accepted transport models in which small (60-80 nm), coated vesicles act as carriers that mediate unior bi-directional transport between two adjacent membranes in the secretory pathway (Schekman and Orci, 1996; Rothman and Wieland, 1996). Transmembrane proteins, together with cytosolic coat protein complexes and additional regulating factors, are thought to mediate sorting of cargo from residents and the formation of transport vesicles. Vesicle targeting and docking with acceptor membranes is mediated by a number of proteins, including fibrous ‘tethering’ molecules such as p115 and giantin, which, through specific interactions, provide a first level of specificity for these transport steps (Pfeffer, 1999). v-SNARES on the vesicles specifically bind t-SNARES at the target membrane and have been proposed to facilitate delivery of cargo by fusion (Jahn and Sudhof, 1999; Scales et al., 2000). Here, we discuss recent in vivo work that has challenged the vesicular transport model. Visualisation of different molecules and transport steps in the secretory pathway in living cells shows that larger (300-500 nm) membrane aggregates or tubular membrane structures rather than small vesicles function as the long-range transport intermediates between the ER and Golgi complex or between the TGN and plasma membrane. This raises the question of the precise role of vesicles in these transport steps.