Chris A. Kaiser
Vesicle Assembly and LoadingFor secretion and endocytosis, protein and membrane constituents are carried from one intracellular compartment to another by small, coated, cytoplasmic vesicles. These vesicles are coated by particular sets of structural proteins which not only set the curvature of the vesicle membrane but also help to select the cargo molecules that will be accepted by the vesicle. Our lab is using the yeast Saccharomyces cerevisiae to identify the genes and proteins that are necessary for assembly of vesicles and for the loading of protein cargo into vesicles.
Vesicle Assembly: Genetic screens have identified a set of proteins that are necessary for budding of transport vesicles from the membrane of the endoplasmic reticulum (ER). A subset of these proteins, collectively known as COPII, have been shown by biochemical experiments to be recruited from the cytosol to the membrane where they assemble into the vesicle coat. We have been investigating the role of membrane proteins in the recruitment process. One of the membrane proteins necessary for coat assembly is Sec16p, a 240 kD protein that is tightly associated with the ER membrane and with the vesicles that have budded from the ER. We have recently found that three of the five COPII proteins bind to Sec16p, and a fourth COPII gene encoding the small GTPase SAR1 exhibits extensive genetic interactions with SEC16. Given that Sec16p resides on the ER membrane prior to formation of a coat and given the extensive contacts between Sec16p and other coat components we think it is likely that Sec16p is part of a membrane scaffold onto which the soluble COPII proteins assemble to construct a coat.
To understand the events that trigger assembly of the coat we have focused on connections between Sec16p and Sar1p which is the best candidate for the regulator of coat assembly. Sec12p is an integral ER protein whose cytosolic domain accelerates exchange of GTP for GDP by Sar1p. We have detected direct association between cytosolic domain of Sec12p and the C-terminal domain of Sec16p. In binding studies in vitro we find that the cytosolic domain of Sec12p competes with the COPII protein Sec23p for binding to Sec16p. An attractive possibility for initiation of coat assembly that we are now testing is that Sec16p is normally bound to Sec12p on the ER membrane then, by the action of Sar1p, Sec12p is displaced opening a binding site for Sec23p and subsequent assembly of the other soluble CPOII proteins onto Sec16p.
Cargo Loading: Mutations that affect the fidelity of protein transport have been more difficult to isolate than mutations that block protein transport altogether. Using a genetic suppression screen we have identified three genes (BST1, BST2 and BST3) that when mutated increase the rate at which the ER resident proteins BiP and PDI leak from the ER and at the same time decrease the rate at which the secretory protein invertase is transported from the ER. These properties imply that loss of BST function reduces the capacity of the ER to discriminate between secreted and retained proteins. The BST1 and BST2 genes were cloned and both encode integral ER membrane proteins. We are currently investigating how this novel class of proteins regulates the cargo content of ER-derived vesicles.
The major branch point in the secretory pathway occurs in a late-Golgi compartment where proteins that will eventually reside in the vacuole are segregated from proteins that are destined to the plasma membrane. We have discovered two novel features of this segregation process with implications for how different kinds of proteins are loaded into vesicles. Certain alleles of SEC13, a gene originally thought to act only in ER to Golgi transport, display a specific defect in transport of the general amino acid permease (GAP) from the Golgi to the cell surface. This blocade of GAP transport is highly selective since cell growth and the transport of other membrane proteins are not affected by SEC13 mutations. This selectivity implies that Sec13p is a component of a special class of transport vesicle that carry GAP to the plasma membrane or that GAP is carried to the plasma membrane in a general class of secretory vesicles and Sec13p is required to load GAP into these vesicles. We are currently purifying GAP-containing vesicles to distinguish between these possibilities.
In another set of experiments we developed a general method to examine the fate of misfolded polypeptides in the yeast secretory pathway by appending the DNA-binding domain of phage l repressor to the C-terminus of the secretory protein invertase. Most of this hybrid protein appears at the cell surface demonstrating that the repressor domain does not interfere with invertase transport through the secretory pathway. In contrast, hybrids composed of invertase fused to mutants of l repressor that are thermodynamically unstable are retained within the cell and the repressor sequences are degraded. A combination of physiological and genetic tests shows that the hybrid protein is targeted to the vacuole by the Golgi receptor protein Vps10p. This finding indicates that Vps10p has a general capacity to bind to nonnative protein structures and may be the basis of a salvage pathway for nonfunctional proteins in the lumen of the secretory pathway.