To reduce expression of gene products not required under stress conditions, eukaryotic cells form large and complex cytoplasmic aggregates of RNA and proteins (stress granules; SGs), where transcripts are kept translationally inert. The overall composition of SGs, as well as their assembly requirements and regulation through stress-activated signaling pathways remain largely unknown.We have performed a genome-wide screen of S. cerevisiae gene deletion mutants for defects in SG formation upon glucose starvation stress. The screen revealed numerous genes not previously implicated in SG formation. Most mutants with strong phenotypes are equally SG defective when challenged with other stresses, but a considerable fraction is stress-specific. Proteins associated with SG defects are enriched in low-complexity regions, indicating that multiple weak macromolecule interactions are responsible for the structural integrity of SGs. Certain SG-defective mutants, but not all, display an enhanced heat-induced mutation rate. We found several mutations affecting the Ran GTPase, regulating nucleocytoplasmic transport of RNA and proteins, to confer SG defects. Unexpectedly, we found stress-regulated transcripts to reach more extreme levels in mutants unable to form SGs: stress-induced mRNAs accumulate to higher levels than in the wild-type, whereas stress-repressed mRNAs are reduced further in such mutants.Our findings are consistent with the view that, not only are SGs being regulated by stress signaling pathways, but SGs also modulate the extent of stress responses. We speculate that nucleocytoplasmic shuttling of RNA-binding proteins is required for gene expression regulation during stress, and that SGs modulate this traffic. The absence of SGs thus leads the cell to excessive, and potentially deleterious, reactions to stress.

eThe phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K) catalytic subunit p110␣ is the most frequently mutated kinase in human cancer, and the hot spot mutations E542K, E545K, and H1047R are the most common mutations in p110␣. Very little is known about the metabolic consequences of the hot spot mutations of p110␣ in vivo. In this study, we used adenoviral gene transfer in mice to investigate the effects of the E545K and H1047R mutations on hepatic and whole-body glucose metabolism. We show that hepatic expression of these hot spot mutations results in rapid hepatic steatosis, paradoxically accompanied by increased glucose tolerance, and marked glycogen accumulation. In contrast, wild-type p110␣ expression does not lead to hepatic accumulation of lipids or glycogen despite similar degrees of upregulated glycolysis and expression of lipogenic genes. The reprogrammed metabolism of the E545K and H1047R p110␣ mutants was surprisingly not dependent on altered p110␣ lipid kinase activity. Class IA phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K) is an integral mediator of several signaling pathways, including insulin-like growth factor 1 (IGF-1) and insulin signaling (1). Upon activation by IGF-1, insulin, or other growth factors, it catalyzes the formation of phosphatidylinositol (3,4,5)-trisphosphate (PIP 3 ), which has a high affinity for the pleckstrin homology (PH) domain of the downstream target Akt/protein kinase B (Akt/ PKB). The interaction of PIP 3 with the PH domain enables phosphorylation of Akt/PKB by phosphoinositide-dependent kinase 1 (PDK 1) and PDK 2. PI3K and Akt/PKB are both critical nodes in many intracellular signaling pathways and are imperative for the actions of IGF-1, insulin, and other growth factors (2).Class IA PI3Ks consist of two subunits. The catalytic subunit, p110, contains the kinase domain responsible for the formation of PIP 3 . The regulatory subunit, usually designated p85, binds to phosphorylated tyrosine residues in other proteins via its SH2 domain, leading to subsequent activation of the kinase. Both subunits exist as several different isoforms. There are four known human catalytic subunit isoforms: p110␣, p110␤, p110␦, and p37␦. p110␣, p110␤, and p110␦ are encoded by three different genes, i.e., PIK3CA, PIK3CB, and PIK3CD, respectively (reviewed in reference 1), whereas p37␦ (PIK3CD_v2) is a splice variant of p110␦ (3). We and others have shown that among the catalytic subunits, p110␣ is the major contributor to transmission of the insulin signal (4-6), whereas p110␤ becomes active primarily in response to G-protein-coupled receptor signaling and plays a role in proliferation (7,8). p110␦ is more cell specific than p110␣ and p110␤ and plays an important role in immune cells and the embryonic nervous system (9-11).p110␣ is the most frequently mutated kinase in human cancer; three mutations in particular, often referred to as the hot spot mutations, namely, E542K, E545K, and H1047R, are the most common in p110␣ (PIK3CA) and comprise 80% of all mutations (reviewed in ref...