We report here the first genome-wide functional genomic screen for longevity genes. We systematically surveyed Caenorhabditis elegans genes using large-scale RNA interference (RNAi), and found that RNAi inactivation of 89 genes extend C. elegans lifespan. Components of the daf-2/insulin-like signaling pathway are recovered, as well as genes that regulate metabolism, signal transduction, protein turnover, and gene expression. Many of these candidate longevity genes are conserved across animal phylogeny. Genetic interaction analyses with the new longevity genes indicate that some act upstream of the daf-16/FOXO transcription factor or the sir2.1 protein deacetylase, and others function independently of daf-16/FOXO and sir2.1, and might define new pathways to regulate lifespan.[Keywords: Longevity; aging; C. elegans; insulin signaling; RNAi screen; genomic] Supplemental material is available at http://www.genesdev.org.

Cell polarity, as reflected by polarized growth and organelle segregation during cell division in yeast, appears to follow a simple hierarchy. On the basis of physical cues from previous cell cycles or stochastic processes, yeast cells select a site for bud emergence that also defines the axis of cell division. Once polarity is established, rho protein-based signal pathways set up a polarized cytoskeleton by activating localized formins to nucleate and assemble polarized actin cables. These serve as tracks for the transport of secretory vesicles, the segregation of the trans Golgi network, the vacuole, peroxisomes, endoplasmic reticulum, mRNAs for cell fate determination, and microtubules that orient the nucleus in preparation for mitosis, all by myosin-Vs encoded by the MYO2 and MYO4 genes. Most of the proteins participating in these processes in yeast are conserved throughout the kingdoms of life, so the emerging models are likely to be generally applicable. Indeed, several parallels to cellular organization in animals are evident.

HetR plays a key role in regulation of heterocyst differentiation. When the Cys-48 residue of the HetR from Anabaena sp. PCC 7120 was replaced with an Ala residue, the mutant HetR (HetR C48A) could not dimerize, indicating that HetR forms a homodimer through a disulfide bond. The Anabaena strain C48, containing the hetRc48a gene, could not produce HetR homodimer and failed to form heterocyst. We show that HetR is a DNA-binding protein and that its homodimerization is required for the DNA binding. HetR binds the promoter regions of hetR, hepA, and patS, suggesting a direct control of the expression of these genes by HetR. We present evidence that shows that the up-regulation of patS and hetR depends on DNA binding by HetR dimer. The pentapeptide RGSGR, which is present at the C terminus of PatS and blocks heterocyst formation, inhibits the DNA binding of HetR and prevents hetR up-regulation. The heterocystous cyanobacteria such as Anabaena sp. PCC strain 7120 contain specialized cells called heterocysts for nitrogen fixation when they are grown in the absence of combined nitrogen (1, 2). Many structural and metabolic changes occur during heterocyst differentiation (2-5). In cyanobacteria with long filaments, the spacing of heterocysts along the filaments is often regular, so there is a pattern formation (1). According to fossil records, the heterocyst pattern was one of the earliest pattern forms in evolution (6, 7).The pattern formation of cyanobacteria depends on cell-cell communication and molecular interactions (1, 8). Several important genes, such as ntcA (9, 10), hetR (11), hetC (12), hetF (13), hetN (14, 15), and patS (16, 17), play important roles in heterocyst differentiation and pattern formation. Recent evidence suggests that ␣-ketoglutarate could play an important role in the initiation of heterocyst differentiation (18).hetR is the master gene in controlling heterocyst differentiation and pattern formation (3,4,19), and it may also control other cellular processes in nonheterocystous cyanobacteria (20). By putting hetR under the control of the copper-inducible promoter PpetE, it was shown that heterocyst frequency was controlled by the expression level of hetR (19). hetR is autoregulatory (21) and regulates the expression of some other genes involved in heterocyst differentiation (8,21). HetR is a Ser-type protease required for heterocyst differentiation (22, 23). The mechanism for regulation of heterocyst differentiation by HetR is not clearly understood. The study by Buikema and Haselkorn (19) raised a very interesting question: Why does HetR as a protease function in a dose-dependent fashion?patS encodes a small peptide, and a diffusible shorter peptide could be generated from the full gene product (16,17). Like hetR, the expression of patS is localized primarily in proheterocysts and heterocysts (16). However, little is known about the regulation of the patS expression and the molecular mechanism for inhibition of heterocyst differentiation by PatS.In this communication, we report that HetR funct...

Formins are actin filament nucleators regulated by Rho-GTPases. In budding yeast, the formins Bni1p and Bnr1p direct the assembly of actin cables, which guide polarized secretion and growth. From the six yeast Rho proteins (Cdc42p and Rho1–5p), we have determined that four participate in the regulation of formin activity. We show that the essential function of Rho3p and Rho4p is to activate the formins Bni1p and Bnr1p, and that activated alleles of either formin are able to bypass the requirement for these Rho proteins. Through a separate signaling pathway, Rho1p is necessary for formin activation at elevated temperatures, acting through protein kinase C (Pkc1p), the major effector for Rho1p signaling to the actin cytoskeleton. Although Pkc1p also activates a MAPK pathway, this pathway does not function in formin activation. Formin-dependent cable assembly does not require Cdc42p, but in the absence of Cdc42p function, cable assembly is not properly organized during initiation of bud growth. These results show that formin function is under the control of three distinct, essential Rho signaling pathways.

The transcription factor DAF-16/forkhead box O (FOXO) is a critical longevity determinant in diverse organisms, however the molecular basis of how its transcriptional activity is regulated remains largely unknown. We report that the Caenorhabditis elegans homolog of host cell factor 1 (HCF-1) represents a new longevity modulator and functions as a negative regulator of DAF-16. In C. elegans, hcf-1 inactivation caused a daf-16-dependent lifespan extension of up to 40% and heightened resistance to specific stress stimuli. HCF-1 showed ubiquitous nuclear localization and physically associated with DAF-16. Furthermore, loss of hcf-1 resulted in elevated DAF-16 recruitment to the promoters of its target genes and altered expression of a subset of DAF-16-regulated genes. We propose that HCF-1 modulates C. elegans longevity and stress response by forming a complex with DAF-16 and limiting a fraction of DAF-16 from accessing its target gene promoters, and thereby regulates DAF-16-mediated transcription of selective target genes. As HCF-1 is highly conserved, our findings have important implications for aging and FOXO regulation in mammals.