SEC14p is the yeast phosphatidylinositol (PI)/phosphatidylcholine (PC) transfer protein, and it effects an essential stimulation of yeast Golgi secretory function. We now report that the SEC14p localizes to the yeast Golgi and that the SEC14p requirement can be specifically and efficiently bypassed by mutations in any one of at least six genes. One of these suppressor genes was the structural gene for yeast choline kinase (CKI), disruption of which rendered the cell independent of the normally essential SEC14p requirement. The antagonistic action of the CKI gene product on SEC14p function revealed a previously unsuspected influence of biosynthetic activities of the CDP-choline pathway for PC biosynthesis on yeast Golgi function and indicated that SEC14p controls the phospholipid content of yeast Golgi membranes in vivo.

Progression of proteins through the secretory pathway of eukaryotic cells involves a continuous rearrangement of macromolecular structures made up of proteins and phospholipids. The protein SEC14p is essential for transport of proteins from the yeast Golgi complex. Independent characterization of the SEC14 gene and the PIT1 gene, which encodes a phosphatidylinositol/phosphatidylcholine transfer protein in yeast, indicated that these two genes are identical. Phospholipid transfer proteins are a class of cytosolic proteins that are ubiquitous among eukaryotic cells and are distinguished by their ability to catalyse the exchange of phospholipids between membranes in vitro. We show here that the SEC14 and PIT1 genes are indeed identical and that the growth phenotype of a sec14-1ts mutant extends to the inability of its transfer protein to effect phospholipid transfer in vitro. These results therefore establish for the first time an in vivo function for a phospholipid transfer protein, namely a role in the compartment-specific stimulation of protein secretion.

Abstract. The budding mode of Saccharomyces cerevisiae cell growth demands that a high degree of secretory polarity be established and directed toward the emerging bud. We report here our demonstration that mutations in SAC1, a gene identified by virtue of its allele-specific genetic interactions with yeast actin defects, were also capable of suppressing secl4 lethalities associated with yeast Golgi defects. Moreover, these sad suppressor properties also extended to sec6 and sec9 secretory vesicle defects. The genetic data are consistent with the notion that SAClp modulates both secretory pathway and actin cytoskeleton function. On this basis, we suggest that SAClp may represent one aspect of the mechanism whereby secretory and cytoskeletal activities are coordinated, so that proper spatial regulation of secretion might be achieved.

Abstract. Mutations in the SAC1 gene exhibit allelespecific genetic interactions with yeast actin structural gene defects and effect a bypass of the cellular requirement for the yeast phosphatidylinositol/phosphatidylcholine transfer protein (SEC14p), a protein whose function is essential for sustained Golgi secretory function. We report that SAClp is an integral membrane protein that localizes to the yeast Golgi complex and to the yeast ER, but does not exhibit a detectable association with the bulk of the yeast F-actin cytoskeleton. The data also indicate that the profound in vivo effects on Golgi secretory function and the organization of the actin cytoskeleton observed in sad mutants result from loss of SAClp function. This cosuppression of actin and SEC14p defects is a unique feature of sad alleles as mutations in other SAC genes that result in a suppression of actin defects do not result in phenotypic suppression of SEC14p defects. Finally, we report that sac1 mutants also exhibit a specific inositol auxotrophy that is not exhibited by the other sac mutant strains. This sad-associated inositol auxotrophy is not manifested by measurable defects in de novo inositol biosynthesis, nor is it the result of some obvious defect in the ability of sad mutants to utilize inositol for phosphatidylinositol biosynthesis. Thus, sac1 mutants represent a novel class of inositol auxotroph in that these mutants appear to require elevated levels of inositol for growth. On the basis of the collective data, we suggest that SAClp dysfunction exerts its pleiotropic effects on yeast Golgi function, the organization of the actin cytoskeleton, and the cellular requirement for inositol, through altered metabolism of inositol glycerophospholipids.

Abstract. Several physiologically important proteins lack a classical secretory signal sequence, yet they are secreted from cells. To investigate the secretion mechanism of such proteins, a representative mammalian protein that is exported by a nonclassical mechanism, galectin-1, has been expressed in yeast. Galectin-1 is exported across the yeast plasma membrane, and this export does not require the classical secretory pathway nor the yeast multidrug resistance-like protein Ste6p, the transporter for the peptide a factor. A screen for components of the export machinery has identified genes that are involved in nonclassical export. These findings demonstrate a new pathway for protein export that is distinct from the classical secretory pathway in yeast.

Inductive bias is the set of assumptions that a person or procedure makes in making a prediction based on data. Different methods for ligand-based predictive modeling have different inductive biases, with a particularly sharp contrast between 2D and 3D similarity methods. A unique aspect of ligand design is that the data that exist to test methodology have been largely man-made, and that this process of design involves prediction. By analyzing the molecular similarities of known drugs, we show that the inductive bias of the historic drug discovery process has a very strong 2D bias. In studying the performance of ligand-based modeling methods, it is critical to account for this issue in dataset preparation, use of computational controls, and in the interpretation of results. We propose specific strategies to explicitly address the problems posed by inductive bias considerations.

No abstract

Systematic annotation of the primary targets of roughly 1000 known therapeutics reveals that over 700 of these modulate approximately 85 biological targets. We report the results of three analyses. In the first analysis, drug/drug similarities and target/target similarities were computed on the basis of three-dimensional ligand structures. Drug pairs sharing a target had significantly higher similarity than drug pairs sharing no target. Also, target pairs with no overlap in annotated drug specificity shared lower similarity than target pairs with increasing overlap. Two-way agglomerative clusterings of drugs and targets were consistent with known pharmacology and suggestive that side effects and drug-drug interactions might be revealed by modeling many targets. In the second analysis, we constructed and tested ligand-based models of 22 diverse targets in virtual screens using a background of screening molecules. Greater than 100-fold enrichment of cognate versus random molecules was observed in 20/22 cases. In the third analysis, selectivity of the models was tested using a background of drug molecules, with selectivity of greater than 80-fold observed in 17/22 cases. Predicted activities derived from crossing drugs against modeled targets identified a number of known side effects, drug specificities, and drug-drug interactions that have a rational basis in molecular structure.