Insulin/insulin-like Growth Factor Signaling in C. Elegans *

The C. elegans insulin/IGF-1 signaling (IIS) pathway connects nutrient levels to metabolism, growth, development, longevity, and behavior. This fundamental pathway is regulated by insulin-like peptide ligands that bind to the insulin/IGF-1 transmembrane receptor (IGFR) ortholog DAF-2. DAF-2/IGFR controls the activity of a conserved phosphoinositide 3-kinase (PI3K)/Akt kinase cascade, culminating in the regulation of a FoxO transcription factor, DAF-16, that governs most of the functions of this pathway. In light of the evolutionary conservation of the IIS pathway, its study in C. elegans is likely to shed light on its functions and regulation in higher organisms, including humans. Originally identified based on its role in the regulation of larval development and aging, IIS also controls a host of other biological processes. Here we review what is currently known about the biological functions and the molecular components of C. elegans IIS.


Introduction
Insulin and insulin-like growth factor-1 (IGF-1) exert their biological effects by binding to and activating cell surface transmembrane receptors with intrinsic tyrosine kinase activity (Ullrich et al., 1985;Ullrich et al., 1986).Activated insulin and IGF-1 receptors phosphorylate a variety of substrates, among which are the IRS (Insulin Receptor Substrate) family of scaffold proteins (White, 1998).Tyrosine-phosphorylated IRS proteins promote the recruitment and activation of components of downstream cascades such as the phosphoinositide 3-kinase Insulin/insulin-like growth factor signaling in C. elegans (PI3K)/Akt, Ras/MAPK (mitogen-activated protein kinase), and mTOR pathways (Taniguchi et al., 2006).These and other insulin and IGF-1 signaling (IIS) pathway components are widely conserved in metazoans (Pinero Gonzalez et al., 2009).Moreover, the role of the IIS in connecting system-wide functions, such as growth, reproduction, and aging, to nutritional status is both conserved and universally important.The use of the genetically tractable system, C. elegans, to study the function of IIS has been valuable, yielding many important insights into the functions, regulation, and outputs of IIS.
Here we review the discovery of the C. elegans IIS pathway and its components (Section 2), phenotypes of IIS mutants (Section 3), molecular characterization of IIS pathway components (Section 4), the FoxO transcription factor DAF-16 (Section 5), DAF-16/FoxO co-factors and regulators (Section 6), other transcription factors regulated by IIS (Section 7), components of IIS in other organisms that are conserved in C. elegans (Section 8), and non-autonomous regulation and integration of insulin signaling (Section 9).We conclude by exploring some prominent remaining questions in the field (Section 10).

Discovery of the C. elegans IIS pathway
The discovery of the IIS pathway in C. elegans emerged from converging lines of investigation in several laboratories studying dauer regulation and the genetic control of longevity.
In genetic screens pioneered by the Riddle lab and subsequently performed in the Thomas and Ruvkun labs, dauer-constitutive (Daf-c) and dauer-defective (Daf-d) mutants were isolated (Riddle et al., 1981;Dauer).Among these were mutants that defined the Daf-c genes daf-2 and daf-23 and the Daf-d gene daf-16.The dauer-constitutive phenotype caused by mutations in daf-2 and daf-23 required the activity of the dauer-defective genes daf-16 and daf-18 (Vowels and Thomas, 1992;Gottlieb and Ruvkun, 1994;Ogg and Ruvkun, 1998;Gil et al., 1999;Mihaylova et al., 1999;Rouault et al., 1999), indicating that DAF-2 and DAF-23 act to prevent dauer arrest in non-dauer-inducing conditions by antagonizing DAF-16 and DAF-18.Genetic analysis indicated that DAF-2 and DAF-23 likely acted in a single pathway to inhibit DAF-16 activity.
In parallel, the labs of Klass and Johnson isolated and characterized the first long-lived C. elegans mutant, age-1.age-1 was first identified by Klass in a genetic screen for long-lived animals (Klass, 1983), and subsequent analysis by the Johnson lab suggested that the longevity phenotype in this mutant strain was due to a recessive mutation in a single gene (Friedman and Johnson, 1988b;Friedman and Johnson, 1988a;Johnson, 1990).Later, Kenyon, et al. found that daf-2 mutants also live twice as long as wild-type animals, and that daf-16 is required for the extended longevity of daf-2 mutants (Kenyon et al., 1993).Molecular analysis demonstrated that age-1 and daf-23 are allelic and that daf-2, age-1/daf-23, and daf-16 encode components of a conserved insulin-like signaling pathway (Morris et al., 1996;Kimura et al., 1997;Lin et al., 1997;Ogg et al., 1997).The extension of life span by mutations in age-1 also required wild-type daf-16 function (Murakami and Johnson, 1996), and the discovery that the life span extension caused by daf-23 mutations also requires daf-16 and daf-18 suggested that the DAF-2/AGE-1/DAF-23/DAF-16 pathway functioned at a crossroads between dauer regulation and longevity (Dorman et al., 1995;Larsen et al., 1995).Intriguingly, despite the natural link between dauer longevity (Klass and Hirsh, 1976) and adult longevity, the long lifespan of daf-2 mutants does not require passing through the dauer stage (Kenyon et al., 1993).Taken together, these results defined a genetic pathway in which DAF-2 and DAF-23/AGE-1 control dauer arrest and life span by acting in opposition to DAF-16 and DAF-18.

Conservation of IIS components
Prior to the availability of genome sequences for multiple organisms, the discovery that the longevity and dauer formation mutants daf-2 and age-1/daf-23 encode C. elegans orthologs of mammalian insulin/IGF-1 receptors (IGFRs) and PI 3-kinase (PI3K), respectively (Morris et al., 1996;Kimura et al., 1997), established the evolutionary conservation of IIS in metazoans and provided seminal insights into the role of IIS in controlling dauer arrest and adult life span.Furthermore, this finding suggested that mobilizing the power of C. elegans as a experimental system to study IIS could lead to new discoveries that would inform investigations into mammalian IIS -a prediction borne out shortly thereafter by the revelation that the C. elegans FoxO transcription factor DAF-16 is a major target of IIS (Lin et al., 1997;Ogg et al., 1997).Subsequent studies have revealed new regulatory components and mechanisms involved in IIS regulation of a variety of biological processes.Moreover, many of the downstream transcriptional targets that control these processes have been identified.

IIS pathway loss-of-function phenotypes
IIS mutants exhibit a wide array of phenotypes, including developmental, reproductive, life history, and behavioral changes, with varying penetrance and expressivity.While the vast majority of early studies on IIS in C. elegans focused on its roles in the control of dauer arrest and aging, IIS has since been implicated in a wide variety of biological functions.daf-2/IGFR mutants, in addition to their dauer-constitutive and increased adult longevity (Age) phenotypes, are more resistant to stress and exhibit increased intrinsic thermotolerance (Itt) (Lithgow et al., 1994).Some alleles exhibit additional defects in development, behavior, fat storage, and reproduction (Gems et al., 1998), as described in greater detail below.
Insulin/insulin-like growth factor signaling in C. elegans

IIS regulation of dauer arrest
Components of C. elegans IIS first emerged from forward genetic screens designed to identify mutants with abnormal regulation of dauer arrest (Riddle et al., 1981).A detailed description of the role of IIS in dauer regulation is available in the Dauer chapter of WormBook.
In dauers, the vulval precursor cells (VPCs) that generate the hermaphrodite vulva are reprogrammed to pluripotency, allowing normal vulval development to proceed upon dauer exit.DAF-16/FoxO is specifically required in dauers for VPC reprogramming, acting cell autonomously to prevent VPC specification and maintain pluripotency by blocking inductive (EGFR) and lateral (LIN-12/Notch) signaling (Karp and Greenwald, 2013).

IIS regulation of L1 arrest
Besides its critical function in dauer regulation, DAF-2/IGFR activity also controls development and starvation-induced non-dauer larval arrest, largely in a daf-16/FoxO-dependent fashion.daf-2/IGFR mutants develop at a slower rate than wild-type animals (Ruaud et al., 2011), and overexpression of DAF-16/FoxO slows developmental rates (J.Ashraf and C. T. Murphy, unpublished), suggesting that the larval developmental process is regulated by DAF-16/FoxO and its targets even under normal conditions.IIS also plays a role in regulating the developmental response to limiting conditions in non-dauer larvae.When C. elegans eggs hatch, first-stage (L1) larvae undergo developmental arrest if food is absent.This phenomenon is known as "L1 arrest."daf-2/IGFR mutants have a partially penetrant L1 arrest phenotype even in the presence of food.In the absence of food, L1-arrested daf-2/IGFR mutant larvae survive starvation more readily than wild-type L1 larvae.daf-16/FoxO mutation suppresses both the L1 arrest phenotype of daf-2/IGFR mutants in the presence of food as well as the prolonged survival of daf-2/IGFR mutant L1 larvae in the absence of food, and DAF-16/FoxO is also required for the survival of L1-arrested wild-type larvae.Therefore, the role of DAF-16/FoxO in promoting survival in starvation-induced L1 arrest is in accord with its role in promoting dauer arrest in response to environmental stressors.

IIS regulation of cell proliferation in arrested states
Developmental arrest in L1 or dauer larvae is accompanied by cessation of developmentally programmed cell divisions.These cell divisions are regulated by IIS through at least two mechanisms.In starvation-induced L1 arrest, cells in the V and M lineages that stop dividing in wild-type larvae continue to divide in daf-16/FoxO mutant larvae, indicating that DAF-16/FoxO promotes cell cycle arrest (Baugh and Sternberg, 2006).DAF-16/FoxO may inhibit cell division by upregulating the p27 kip1 cyclin-dependent kinase inhibitor cki-1, as induction of cki-1 transcription in response to starvation is abrogated by daf-16/FoxO mutation (Baugh and Sternberg, 2006).DAF-16/FoxO may also contribute to the starvation-induced transcriptional repression of lin-4, a microRNA induced during the L1 larval stage that orchestrates the temporal program of larval cell divisions in replete conditions by repressing LIN-14 and LIN-28 translation (Wightman et al., 1993;Moss et al., 1997;Baugh and Sternberg, 2006).It is likely that DAF-16/FoxO regulates the expression of other genes that in turn regulate cell division, development, and growth (Baugh and Sternberg, 2006).
By contrast, cell cycle arrest of the Z2 and Z3 primordial germ cells in starved L1 larvae is controlled by a distinct mechanism.Whereas most somatic cells that arrest in starved L1 larvae do so prior to replicating their DNA, Z2 and Z3 arrest in the G2 phase of the cell cycle (Fukuyama et al., 2006).Mutations that inactivate DAF-18/PTEN cause continued proliferation of Z2 and Z3 in arrested L1 larvae, and arrest was restored in age-1; daf-18 and daf-18; akt-1 double mutants.Therefore, PI3K/Akt signaling controls starvation-induced Z2/Z3 arrest.Surprisingly, Insulin/insulin-like growth factor signaling in C. elegans DAF-16/FoxO was not required for Z2/Z3 arrest under these conditions, indicating that AGE-1/PI3K and AKT-1 control Z2/Z3 proliferation independently of DAF-16/FoxO (Fukuyama et al., 2006).Whether DAF-2/IGFR is the activating input to AGE-1/PI3K in this context is not known.

IIS regulation of germline proliferation in adulthood
In the adult, the only mitotically proliferating cells are in the germline, and IIS plays a critical role in regulating the development and maintenance of the proliferative process.Excessive proliferation can result in germline "tumors" in mutants in which meiotic germ cells re-enter the mitotic cell cycle and overproliferate (Francis et al., 1995).Such germline "tumors" are reduced in daf-2/IGFR mutants in a daf-16/FoxO-dependent manner, and downstream targets of DAF-16/FoxO influence germline cell proliferation and p53-dependent apoptosis (Pinkston-Gosse and Kenyon, 2007).This effect is likely due to the role of IIS in promoting Notch-independent germline cell proliferation and cell cycle activity during normal development in reproductive larval stages (Michaelson et al., 2010).
daf-2/IGFR-dependent mitotic proliferation in the germline is regulated by the ILPs INS-3 and INS-33, which are expressed primarily in the nervous system and hypodermis, respectively.These ILPs do not affect dauer formation or intestinal DAF-16/FoxO activity (Michaelson et al., 2010), suggesting that the activity of INS-3 and INS-33 is restricted to daf-2/IGFR-mediated regulation of germline proliferation.Notably, INS-33 is also a direct target of LIN-14 (Hristova et al., 2005), the L1 transcriptional repressor that is itself a target of the lin-4 microRNA (Arasu et al., 1991;Lee et al., 1993;Olsen and Ambros, 1999), perhaps suggesting the presence of a regulatory loop.

IIS regulation of longevity and aging
The isolation and characterization of age-1(hx546) as a long-lived mutant (Klass, 1983;Friedman and Johnson, 1988b;Friedman and Johnson, 1988a;Johnson, 1990) and the subsequent demonstration that daf-2/IGFR mutants live twice as long as wild-type animals (Kenyon et al., 1993) established the importance of IIS in life span control.The daf-2/IGFR mutant longevity phenotype does not require passing through the dauer state, as animals are long-lived when shifted to the restrictive temperature at the L4 larval stage, after both the dauer stage (L3 equivalent) and the dauer decision point late in the L1 larval stage (Kenyon et al., 1993).Cynthia Kenyon's lab also showed that RNAi inactivation of daf-2/IGFR in early to mid-adulthood could extend life span, while treatment in the L3 larval stage affected progeny number, clearly delineating the temporal regulation of dauer formation, reproduction, and longevity by DAF-2/IGFR (Dillin et al., 2002).
IIS regulation of longevity is not restricted to nematodes; similar effects are seen in flies (Clancy et al., 2001;Tatar et al., 2001) and mice (Bluher et al., 2003;Holzenberger et al., 2003).The role of IIS in life span control may also extend to humans.Studies in cohorts of Ashkenazi Jewish centenarians have identified non-synonymous polymorphisms in the gene encoding the IGF-I receptor that are associated with longevity (Suh et al., 2008).Importantly, cells expressing IGF-1 receptors harboring these polymorphisms exhibit reductions in ligand-dependent Akt phosphorylation, gene regulation, and cell cycle progression compared to cells expressing wild-type IGF-1 receptors, indicating that a reduction in IGF-1 signaling is correlated with longevity in humans (Tazearslan et al., 2011).Thus, important downstream functions of IIS may be conserved.Moreover, longevity and phenotypes associated with healthy aging in humans, including cognitive function and metabolic profiles, have been reported to be associated with non-coding genetic variation in FOXO3A (Willcox et al., 2008).The impact of these non-coding polymorphisms on FoxO3A function is not known.
daf-2/IGFR mutants are also more immune to pathogenic bacterial infections than are wild-type animals (Garsin et al., 2003;Kerry et al., 2006), again due to the regulation of DAF-16/FoxO target genes (Section 5.4).Intriguingly, daf-2/IGFR mutants also exhibit enhanced RNAi (Wang and Ruvkun, 2004); as the C. elegans RNAi machinery is required for antiviral defense (Felix et al., 2011), daf-2/IGFR mutants might also be more resistant to viral infection (although this has not yet been shown).

IIS regulation of fat metabolism
daf-2/IGFR mutants have increased fat content compared to wild-type animals (Kimura et al., 1997;Ogg et al., 1997;Ashrafi et al., 2003;Perez and Van Gilst, 2008;O'Rourke et al., 2009).Imaging of live animals using coherent anti-Stokes Raman scattering (CARS) indicates that this increase is due at least in part to greater fat accumulation in the hypodermis (Hellerer et al., 2007).In vivo labeling of fat stores with 13 C-enriched E. coli revealed that increased de novo lipogenesis largely accounts for the increased fat phenotype of daf-2/IGFR mutants (Perez and Van Gilst, 2008).The increased fat storage phenotype of daf-2/IGFR mutant animals is dependent on daf-16/FoxO (Ogg et al., 1997;Ashrafi et al., 2003;Perez and Van Gilst, 2008).Several downstream targets of DAF-16/FoxO likely contribute to changes in fat metabolism, including fat-genes that act in fatty acid desaturation and lipolysis, acyl-CoA and alcohol dehydrogenases, glyoxylate cycle regulators, and autophagy regulators (Melendez et al., 2003;Murphy et al., 2003;Lapierre et al., 2011;Lapierre et al., 2012) (see Section 5.4 for details).An analysis of several independent long-lived daf-2/IGFR mutants suggests that alterations in lipogenesis may be somewhat uncoupled from longevity (Perez and Van Gilst, 2008), suggesting that factors other than fat levels may ultimately determine life span.

IIS regulation of reproductive aging
The extended germline maintenance of daf-2/IGFR mutants, coupled with their extension of oocyte quality maintenance (Luo et al., 2010), results in the greatly extended reproductive span of daf-2/IGFR mutants (Hughes et al., 2007;Luo et al., 2009;Luo et al., 2010), again highlighting the role of IIS in regulating mitotic cell maintenance as well as post-mitotic survival and stress responses.The cell autonomy of IIS activity in reproductive maintenance differs from that of germline proliferation during development: while DAF-2/IGFR-mediated proliferation requires DAF-16/FoxO activity in the germline (Michaelson et al., 2010), maintenance of reproduction with age requires DAF-16/FoxO activity in muscle and intestine (Luo et al., 2010), the same tissue that provides the bulk of the longevity-promoting activity of DAF-16/FoxO (Libina et al., 2003).The contribution of muscle DAF-16/FoxO activity to reproductive longevity was a surprise, as this tissue plays no role in somatic longevity (Libina et al., 2003).Interestingly, Michaelson et al. also report a minor contribution from DAF-16/FoxO activity in muscle in regulation of germline proliferation during development (Michaelson et al., 2010).
Insulin/insulin-like growth factor signaling in C. elegans
Intriguingly, although both IIS and the germline control life span by regulating DAF-16/FoxO activity, reducing IIS in animals lacking a germline extends life span substantially (Arantes-Oliveira et al., 2003).Thus, the mechanisms by which IIS and the germline regulate DAF-16/FoxO activity might be distinct to some degree.This is underscored by the observation that KRI-1 and TCER-1 are required for life span extension caused by germline ablation, but are dispensable for longevity in the context of reduced IIS (Berman and Kenyon, 2006;Ghazi et al., 2009).Germline proliferation longevity signals regulate the expression of genes both dependent on and independent or upstream of DAF-16/FoxO activity (McCormick et al., 2011).The dafachronic acid pathway makes up one arm of this regulation: in animals lacking a germline, daf-36 expression is induced, generating DA, in turn activating DAF-12.The let-7 family miRNAs mir-84 and mir-241 are induced by DAF-12, and in turn down regulate AKT-1 and LIN-14, resulting in the activation of DAF-16/FoxO (Shen et al., 2012).The miRNA mir-71, which is upregulated with age and promotes longevity (de Lencastre et al., 2010), is required for the long lifespan of animals lacking a germline; mir-71 acts in the nervous system to promote longevity, upstream of DAF-16/FoxO activity and independently of daf-12 (Boulias and Horvitz, 2012).The mir-71 targets that are important for life span control are not known.
Germline stem cell loss in Drosophila also increases life span through IIS (Flatt et al., 2008), suggesting that the germline/gonad communication via IIS to regulate somatic longevity is evolutionarily conserved.Such communication from the soma to the germline, and from the germline to the soma (Hsin and Kenyon, 1999;Luo et al., 2010), might enable an organism to tune its reproductive span and life span properly to adapt to changing nutrient availability (Luo and Murphy, 2011).This system-wide signaling integration is discussed in more detail in Section 9.

IIS and neuronal/behavioral phenotypes
IIS mutants also exhibit reduced rates of neural decline compared to wild-type animals, as indicated by a number of behavioral phenotypes (see Stein and Murphy 2012 for greater detail).They exhibit extended isothermal tracking (Murakami et al., 2005), motor activity (Duhon and Johnson, 1995;Huang et al., 2004;Hsu et al., 2009), and positive butanone associative learning with age (Kauffman et al., 2010).Touch receptor neurons and VNC cholinergic axons of daf-2/IGFR mutants also show delayed progression of age-dependent morphological defects, while such defects are accelerated in daf-16/FoxO mutants (Pan et al., 2011).daf-2/IGFR mutant animals also exhibit delayed ectopic neurite branching compared to age-matched wild-type animals, again in a daf-16/FoxO-dependent manner (Tank et al., 2011).Additionally, daf-2/IGFR mutants exhibit improved short-and long-term associative memory early in adulthood compared to wild-type animals (Kauffman et al., 2010).While the latter function is due to elevated CREB levels in daf-2/IGFR mutants (Kauffman et al., 2010), the molecular basis for increased short-term associative memory and extended learning ability in daf-2/IGFR mutant animals is not yet known.
By contrast, other learning behaviors are impaired in IIS mutants.Specifically, salt chemotaxis learning appears to require functional IIS, as calcium signaling in the ASER neuron, which senses salt, is altered in IIS mutants (Tomioka et al., 2006;Vellai et al., 2006;Lin et al., 2010;Oda et al., 2011).daf-2/IGFR mutant chemotaxis Insulin/insulin-like growth factor signaling in C. elegans to salt prior to training (naïve) is higher than that of wild-type as well (Tomioka et al., 2006;Oda et al., 2011).Cell-specific rescue experiments suggest that DAF-2/IGFR and AGE-1/PI3K act in the ASER, while the ILP INS-1 acts in the AIA interneurons, which have synaptic outputs to ASER (Tomioka et al., 2006;Oda et al., 2011).IIS mutants are also defective in a paradigm that pairs the positive chemoattractant benzaldehyde with starvation to create a negative association (Lin et al., 2010).ins-1 and daf-2/IGFR mutants, which display normal positive "butanone enhancement" learning (Torayama et al., 2007;Kauffman et al., 2010), are defective in benzaldehyde-starvation avoidance learning (Lin et al., 2010), an effect that is dependent upon daf-18/PTEN but only partially dependent on daf-16/FoxO.It is possible that insulin signals are required in these negative associative paradigms to properly signal the starved state; if the responses of IIS mutants to starvation and salt stress are muted, then their ability to form associations with these negative states might be hindered (Stein and Murphy, 2012).

DAF-2, the C. elegans insulin/IGF-1 receptor ortholog
Cloning of daf-2 revealed that it encodes a receptor tyrosine kinase that shares greater than 30% amino acid identity with the human insulin, IGF-1, and insulin receptor-related receptors (Kimura et al., 1997).Mammalian insulin receptors form α2/β2 heterotetramers that contain extracellular ligand-binding domains in the?α-subunit and transmembrane and intracellular tyrosine kinase domains in the β-subunit.Cysteines in both subunits form disulfide bonds that link each α-subunit to the extracellular domain of a β-subunit as well as to the other α-subunit of the heterotetramer (Lawrence et al., 2007).Insulin binding causes a conformational change that induces autophosphorylation, activating intrinsic receptor tyrosine kinase activity (Luo et al., 1999;Kido et al., 2001).
daf-2 encodes the sole C. elegans insulin receptor family member (Kimura et al., 1997).The daf-2 locus encodes two alternatively spliced transcripts; nothing is known about the specific functions of these DAF-2/IGFR isoforms.DAF-2/IGFR is highly expressed in the nervous system and the XXX neuroendocrine cells, with weaker expression in hypodermis, intestine, and other tissues (Hunt-Newbury et al., 2007;Kimura et al., 2011), and its abundance is modulated by nutritional status (Kimura et al., 2011).Although a family of putative receptors with weak homology to the DAF-2/IGFR extracellular domain has been identified bioinformatically (Dlakic, 2002), their biological significance is not yet known.
Although most ILPs lack many of the sequence characteristics of human insulin (Pierce et al., 2001), they may still function as receptor ligands.For example, the divergent ILP INS-6, which is predicted to lack a C-peptide, can bind to and activate the human insulin receptor (Hua et al., 2003).Another surprising feature of C. elegans ILPs is their apparent ability to act either as antagonists  et al., 2001;Hristova et al., 2005;Murphy et al., 2007;Cornils et al., 2011).The mechanistic basis for these differences in function is not yet understood.One hypothetical model is that alternate DAF-2/IGFR-like co-receptors (Dlakic, 2002) might integrate responses to ILPs in a combinatorial fashion, but such a role for these putative co-receptors has not been uncovered.Additionally, whether ILPs that antagonize DAF-2/IGFR signaling do so through direct as opposed to indirect effects on DAF-2/IGFR activity (e.g., by influencing the processing and maturation of other ILPs) is not yet known.Genes encoding agonist/antagonist pairs (e.g., ins-7 and ins-18), can also be regulated in opposite manners at the expression level (Murphy et al., 2003;Liu et al., 2004;Murphy et al., 2007;Shaw et al., 2007), perhaps to coordinate the downstream IIS response between tissues and pathways (Murphy et al., 2007;Shaw et al., 2007).For example, the feed-forward mechanism of ins-7 repression and ins-18 induction in the Insulin/insulin-like growth factor signaling in C. elegans intestine activates DAF-16/FoxO in other tissues, thus aligning IIS activity in all tissues to coordinate somatic aging, called "FoxO-to-FoxO" signaling (Murphy et al., 2007).The ins-7/ins-18 insulin pair is also regulated by the DAF-7/TGF-β dauer regulatory pathway, linking the two pathways (Liu et al., 2004;Shaw et al., 2007;Narasimhan et al., 2011); this is discussed in greater detail in Section 9.3.In another form of communication called "ILP-to-ILP" signaling, INS-6 in the ASI neuron represses INS-7 expression in the URX neuron; INS-7 antagonizes DAF-2/IGFR in the RIA neuron, disrupting aversive olfactory learning (Chen et al., 2013b).
While most experiments involving ILPs suggest that they signal through DAF-2/IGFR, there is at least one report of a daf-2/IGFR-independent insulin effect, which is ins-1 regulation of AWC sensory neuron activity (Chalasani et al., 2010).The receptor that mediates INS-1 action in this context is not yet known.The observation that mammalian G-protein-coupled receptors can act as receptors for the relaxin family of insulin-related peptide hormones (Hsu et al., 2002) suggests that non-daf-2/IGFR receptors could mediate some of the biological functions of ILPs.

Regulation of ILP secretion
Recent studies have identified both positive and negative regulators of ILP secretion.ASNA-1 is a conserved ATPase that positively regulates DAF-2/IGFR signaling by promoting the secretion of the ILP DAF-28 (Section 4.2) and possibly other ILPs (Kao et al., 2007).Human ASNA1 is expressed specifically in the insulin-producing β-cells of pancreatic islets and enhances insulin secretion in β-cell-derived cell lines, suggesting that the role of ASNA-1 in regulating ILP secretion is evolutionarily conserved (Kao et al., 2007).DAF-28 secretion also requires intact mitochondrial function, as knockdown of components of the TOMM mitochondrial protein import complex reduces DAF-28 secretion (Billing et al., 2011).Whether mitochondria communicate with ASNA-1 to promote DAF-28 secretion is not known.
An analysis of C. elegans orthologs of Bardet-Biedl syndrome (BBS) disease genes has revealed a role for BBS proteins in attenuating ILP secretion.BBS is an autosomal recessive genetic disorder of pleiotropic phenotype caused by mutations in any of fourteen genes encoding proteins that are required for the normal function of cilia (Zaghloul and Katsanis, 2009).Strikingly, C. elegans orthologs of BBS disease genes are expressed specifically in ciliated neurons (Ansley et al., 2003).In contrast to mutations in many cilia genes, which reduce DAF-28 secretion, bbs mutations enhance the secretion of DAF-28 and other neuropeptides (Lee et al., 2011).RNAi of murine BBS genes in the pancreatic β-cell-derived MIN6 cell line induced insulin secretion ~2-fold, suggesting that the role of BBS proteins in limiting ILP secretion is evolutionarily conserved.
The PI3K superfamily of enzymes phosphorylates phosphoinositides at the 3' position of the inositol ring.Activation of Class I PI3Ks by insulin and IGFs increases 3-phosphoinositide (PIP 3 ) concentrations at the plasma membrane, thus promoting the recruitment and activation of PIP 3 -dependent signaling complexes (Cantrell, 2001).Mammalian Class IA PI3Ks are heterodimers consisting of a p110 catalytic subunit and an adaptor subunit that contains two SH2 domains.Upon binding of ligand to insulin and IGF-1 receptors, IRS (insulin receptor substrate) scaffold proteins are phosphorylated on multiple tyrosine residues, creating binding sites for the SH2 domains of the PI3K adaptor subunit that promote the recruitment of PI3K to the plasma membrane (Taniguchi et al., 2006).
Insulin/insulin-like growth factor signaling in C. elegans The C. elegans genome encodes a single PI3K adaptor subunit (AAP-1; AGE-1 adaptor protein) and a putative IRS homolog (IST-1; insulin receptor substrate homolog) (Wolkow et al., 2002).An aap-1 mutant, aap-1(m889), was isolated in a genetic screen for thermotolerant mutants (Munoz and Riddle, 2003).Similar to age-1/PI3K mutants, aap-1(m889) mutants live long in a daf-16/FoxO-dependent manner (Munoz and Riddle, 2003).Furthermore, AAP-1 protein synthesized in E. coli can bind to an N-terminal fragment of AGE-1/PI3K (Wolkow et al., 2002).However, in contrast to the non-conditional dauer arrest phenotype of age-1/PI3K null mutants, aap-1(m889) animals develop reproductively when cultured at 25°C and only undergo dauer arrest at 27°C (Munoz and Riddle, 2003).This may be a consequence of the fact that the m889 allele is a nonsense mutation that is predicted to create a truncated AAP-1 protein that still contains the N-terminal SH2 domain as well as the inter-SH2 domain that mediates the association of mammalian p85 with p110 (Munoz and Riddle, 2003).Therefore, the mutant AAP-1 protein synthesized in aap-1(m889) may still be able to interact with AGE-1/PI3K.The aap-1 null phenotype has not been described.Notably, no aap-1 mutant alleles have emerged from genetic screens for dauer-constitutive mutants in which multiple alleles of daf-2/IGFR and age-1/PI3K were isolated.Thus, the possibility exists that DAF-2/IGFR may be able to activate AGE-1/PI3K in the absence of AAP-1.

IRS (insulin receptor substrate) proteins
IST-1 was identified based on comparative sequence analysis with mammalian IRS proteins.IST-1 has limited amino acid similarity to mammalian IRS proteins; the greatest sequence identity lies in the pleckstrin homology and phosphotyrosine binding motifs (Wolkow et al., 2002).Unlike mammalian IRS proteins, which contain several YxxM motifs that can bind to Class IA PI3K adaptor subunits when phosphorylated (Sun et al., 1991;Sun et al., 1995), IST-1 contains a single YxxM motif.ist-1 RNAi does not induce dauer arrest in wild-type animals at any temperature but enhances dauer arrest to a comparable extent in animals harboring either partial loss-of-function or null age-1/PI3K mutants (Wolkow et al., 2002).Although the phenotype of an ist-1 null mutant has not been reported, these results suggest that, in contrast to Drosophila chico (Bohni et al., 1999) and mammalian IRS proteins (Shaw, 2011), IST-1 may function primarily to couple DAF-2/IGFR to signaling outputs that act in parallel to AGE-1/PI3K.ist-1 mutants have also not emerged from dauer-constitutive screens, suggesting that IST-1 may be dispensable for DAF-2/IGFR-dependent activation of AGE-1/PI3K.

AKT-1 and AKT-2
In mammals, Akt/Protein Kinase B (PKB) family serine/threonine kinases mediate many of the cellular responses to insulin and IGF-1.Ligand-dependent activation of Class I PI3Ks increases PIP 3 levels, resulting in the recruitment of Akt/PKB to the plasma membrane via binding of its N-terminal pleckstrin homology (PH) domain to PIP 3 .Akt/PKB is subsequently activated by phosphorylation at two conserved residues (T308 and S473 in human AKT1) (Taniguchi et al., 2006).Mammalian Akt/PKB directly phosphorylates several substrates (Manning and Cantley, 2007).

PDK-1
Full activation of Akt/PKB requires phosphorylation of both T308 in the activation segment and S473 in the C-terminal hydrophobic motif (Datta et al., 1999).Both of these sites are conserved in AKT-1 and AKT-2 (T350/S517 in AKT-1A and T337/S505 in AKT-2A, respectively) (Paradis and Ruvkun, 1998), suggesting that the underlying mechanisms governing Akt/PKB activation are conserved.In both mammalian cells and C. elegans, T308 is phosphorylated by the phosphoinositide-dependent kinase Pdk, a serine/threonine kinase that, similar to Akt/PKB, is recruited to the plasma membrane through binding of its PH domain to PIP 3 (Toker and Newton, 2000).A gain-of-function mutation in the C. elegans Pdk ortholog pdk-1, mg142, was isolated in the same screen in which the akt-1(mg144) gain-of-function mutant was isolated (Paradis et al., 1999).pdk-1(mg142) strongly suppresses the dauer-constitutive phenotype of age-1/PI3K null mutants but does not suppress dauer arrest in animals with reduced akt-1 and akt-2 activity, suggesting that PDK-1 acts downstream of AGE-1/PI3K and upstream of AKT-1 and AKT-2 (Paradis et al., 1999).The suppression of age-1(null) dauer arrest by pdk-1(mg142) is abrogated by RNAi of either akt-1 or (to a lesser extent) akt-2, indicating that AKT-1 and AKT-2 are the major targets of PDK-1 in dauer regulation (Paradis et al., 1999).The pdk-1(sa680) loss-of-function mutant has a dauer-constitutive phenotype at 25°C and extended longevity; both of these phenotypes require DAF-16/FoxO activity (Paradis et al., 1999).The pdk-1(null) phenotype has not been reported; therefore, it is not known whether AGE-1/PI3K has other outputs in addition to PDK-1 that contribute to the activation of AKT-1 and AKT-2.
Multiple mammalian kinases, including Rictor/mTORC2 and DNA-dependent protein kinase, have recently been shown to activate Akt/PKB through phosphorylation of its C-terminal hydrophobic motif site (Bozulic and Hemmings, 2009).The C. elegans kinase(s) that phosphorylates this site is not known.To date, no candidates have emerged from genetic screens.Loss-of-function mutations in the C. elegans Rictor ortholog rict-1, identified in screens for mutants with reduced fat content, shorten life span and do not promote dauer arrest under standard conditions (Jones et al., 2009;Soukas et al., 2009), suggesting that, in contrast to mammalian Rictor/mTORC2 (Sarbassov et al., 2005), RICT-1 does not contribute to Akt/PKB activation in dauer regulation or life span control.

The FoxO transcription factor DAF-16
daf-16/FoxO null mutations fully suppress most phenotypes associated with reduced IIS, suggesting that these phenotypes are a consequence of increased DAF-16/FoxO activity.Accordingly, DAF-16/FoxO is likely the primary mediator of the dauer-constitutive, increased longevity, and stress resistance phenotypes observed in mutants with reduced IIS.DAF-16 is a member of the FoxO family of Forkhead transcription factors, which are key regulators of growth, metabolism, stress responses, cell cycle control, and longevity in many organisms (Accili and Arden, 2004).The Forkhead DNA binding domain, DNA binding specificity, RxRxxS/T phosphorylation motifs, and 14-3-3 binding sites are conserved between C. elegans DAF-16 and mammalian FoxOs (Obsil and Obsilova, 2008).DAF-16/FoxO is inhibited by the IIS pathway through phosphorylation at its RxRxxS/T motifs, which results in its cytoplasmic retention (Lee et al., 2001;Lin et al., 2001).Mutation of daf-2/IGFR, age-1/PI3K, pdk-1 or akt-1/2 reduces DAF-16/FoxO phosphorylation, resulting in its nuclear translocation and subsequent transcriptional regulation of DAF-16/FoxO target genes (see Section 5.4) that results in IIS loss-of-function phenotypes (see Section 3).
The three groups of DAF-16/FoxO isoforms share carboxy-terminal sequences but diverge at their amino-termini (Figure 2B).Human FoxO1, FoxO3, and FoxO4 proteins are most similar in sequence to DAF-16A and DAF-16D/F/H, which share their 428 carboxy-terminal amino acids (including the entire Forkhead DNA binding domain) (Kwon et al., 2010).Importantly, whereas all three RxRxxS/T motifs in mammalian FoxO transcription factors that are phosphorylated in response to Akt/PKB activation are conserved in DAF-16A, the divergent amino-terminus of DAF-16D/F/H lacks the amino-terminal RxRxxT motif (Kwon et al., 2010).Instead, DAF-16D/F/H isoforms are phosphorylated in vivo at an amino-terminal site within a QxRxxS motif (Bodenmiller et al., 2008;Kwon et al., 2010).Since the -5 arginine in the canonical RxRxxS/T motif of the Akt/PKB substrate GSK3 makes important molecular contacts with conserved acidic residues in the kinase domain of Akt/PKB and other highly related kinases (Yang et al., 2002), the QxRxxS motif in DAF-16D/F/H and the amino-terminal RxRxxT motif in DAF-16A may be phosphorylated and regulated by distinct kinases.
The 319 C-terminal amino acids of DAF-16B, which include part of but not the entire Forkhead domain, are identical to those of DAF-16A and DAF-16D/F/H.The amino-terminal 211 amino acids of DAF-16B, including the amino-terminal portion of its Forkhead domain, are encoded by a unique exon (Ogg et al., 1997).The fact that the DAF-16B DNA binding domain differs in amino acid sequence from that of DAF-16A and DAF-16D/F/H suggests that it may bind to DNA sequences distinct from those recognized by DAF-16A and DAF-16D/F/H.However, a daf-16b cDNA expressed under the control of the daf-16a promoter extends the life span of daf-16/FoxO null mutant animals to nearly the same degree as daf-16a expressed from the daf-16a promoter, and expression of either cDNA under the control of the daf-16b promoter suffices to induce pharyngeal remodeling in dauers (Lee et al., 2001).Thus, DAF-16A and DAF-16B may have similar functions in spite of some amino acid sequence divergence in their Forkhead domains.Although the amino-terminus of DAF-16B is unique, it does contain an RxRxxT motif, suggesting that DAF-16A and DAF-16B may be regulated by the same or similar protein kinases.

Functions of specific DAF-16/FoxO isoforms
daf-16/FoxO null mutations and mutations predicted to only affect daf-16a and daf-16d/f/h isoforms suppress the dauer-constitutive and extended life span phenotypes of daf-2/IGFR mutants to similar degrees (Lee et al., 2001), and RNAi knockdown of daf-16a and daf-16d/f/h together suppresses most of the life span extension caused by daf-2/IGFR mutation (Kwon et al., 2010).Furthermore, expression of daf-16a-specific and daf-16d/f/h-specific transgenes suffices to rescue dauer-constitutive and longevity phenotypes in daf-16(null); daf-2(e1370) double mutants (Lin et al., 2001;Kwon et al., 2010).By contrast, daf-16b-specific transgenes have a small effect on longevity and dauer arrest (Lee et al., 2001;Kwon et al., 2010).Taken together, these results suggest that DAF-16A and DAF-16D/F/H are the major isoforms that control dauer arrest and life span.Studies with isoform-specific transgenes suggest that AKT-1 influences life span primarily by controlling DAF-16D/F/H activity, whereas AKT-2 controls life span mainly by regulating DAF-16A (Kwon et al., 2010).A recent study examining the contributions of specific mTOR complexes to life span control revealed that C. elegans TORC1 controls life span by regulating the subcellular localization of DAF-16D/F/H without influencing DAF-16A activity (Robida-Stubbs et al., 2012).Thus, a picture is emerging whereby specific DAF-16/FoxO isoforms influence life span by responding to distinct regulatory inputs.The unique transcriptional outputs that may distinguish the biological activities of the different isoforms are not yet known.
No daf-16b-specific mutations have been reported, and the observation that daf-16b is mainly expressed in tissues that may be refractory to RNAi, such as the pharynx and the nervous system, complicates the interpretation of negative daf-16b RNAi results.Expression of daf-16a transgenes under the control of the daf-16a promoter in daf-16/FoxO null mutants is sufficient to rescue the dauer-defective phenotype; however, dauer formation in these animals is incomplete, as these dauers do not have fully remodeled pharynxes that are typically seen in natural dauers (Henderson and Johnson, 2001;Lee et al., 2001;Lin et al., 2001;Kwon et al., 2010).Expression of a daf-16b transgene under the control of its native promoter is sufficient for partial rescue of the dauer-defective phenotype of Insulin/insulin-like growth factor signaling in C. elegans daf-16/FoxO null mutants, and dauers formed in this context have fully remodeled pharynxes (Lee et al., 2001;Kwon et al., 2010).Therefore, DAF-16B plays an important role in pharyngeal remodeling during dauer formation.A recent study also implicates DAF-16B in promoting neurite outgrowth cell autonomously in the AIY interneuron (Christensen et al., 2011).

The DAF-16 Binding Element (DBE)
The DNA binding domain of DAF-16/FoxO, like that of other members of the FoxO transcription factor family, binds in vitro to a core consensus DNA sequence, TTGTTTAC, known as the DBE (DAF-16 Binding Element), as shown through in vitro SELEX studies (Furuyama et al., 2000).Variations of the DBE are overrepresented in the promoters of genes that are upregulated in daf-2/IGFR mutants in a DAF-16/FoxO-dependent manner (Class 1 DAF-16/FoxO targets) (Murphy et al., 2003), and recent DamID (chromatin profiling by DNA adenine methyltransferase identification) experiments suggest that DAF-16/FoxO binds directly to these target genes (Schuster et al., 2010).
By contrast, DAF-16/FoxO appears not to bind the promoters of genes that are downregulated in daf-2/IGFR mutants in a DAF-16/FoxO-dependent manner (Class 2 DAF-16/FoxO targets).These promoters contain a different overrepresented sequence, CTTATCA (Murphy et al., 2003), a reverse GATA site that was later dubbed the "DAE" for "DAF-16 Associated Element" (McElwee et al., 2004).Given its lack of similarity with the DBE, it seems unlikely that FoxO transcription factors would bind to the DAE; thus a transcription factor distinct from DAF-16/FoxO may be involved in the regulation of Class 2 (downregulated) DAF-16/FoxO targets.In fact, ChIP-Seq data implicates the zinc finger transcription factor PQM-1 as the DAE-binding factor, acting antagonistically with DAF-16/FoxO (Tepper et al., 2013) (Section 6.12).
Insulin/insulin-like growth factor signaling in C. elegans Conversely, "Class 2" genes are downregulated in daf-2/IGFR mutants in a DAF-16/FoxO-dependent manner (Murphy et al., 2003).The functions of most of these genes and their roles in normal development and life span control are not yet understood, but recent GO analysis suggests that many of these genes regulate growth and development (Tepper et al., 2013), and reducing expression of some of these genes by RNAi extends life span (Murphy et al., 2003).Many Class 2 genes are upregulated in response to pathogen exposure (Troemel et al., 2006), suggesting a specific role in pathogen response that may normally "cost" longevity, but this has not yet been shown.Most of the Class 2 genes, and some of the Class 1 genes, are regulated by the zinc finger transcription factor PQM-1, which acts in opposition to DAF-16/FoxO (Tepper et al., 2013) (Section 6.12).
The regulation of Class 1 and Class 2 gene expression by DAF-16/FoxO may function in part to coordinate the state of IIS at the organismal level.When IIS is reduced, DAF-16/FoxO represses the Class 2 ILP gene ins-7 (which encodes an agonist of DAF-2/IGFR) and promotes the expression of the Class 1 ILP gene ins-18 (which encodes a DAF-2/IGFR antagonist), resulting in feed-forward inhibition of IIS throughout the animal (Murphy et al., 2003;Murphy et al., 2007).

Other DAF-16/FoxO co-factors and regulators
DAF-16/FoxO is regulated by a host of additional proteins that directly or indirectly modulate its activity (Figure 3) (Landis and Murphy, 2010).Like the IIS pathway, DAF-16/FoxO co-factors and regulators and the nature of their interactions with DAF-16/FoxO are remarkably conserved throughout metazoan phylogeny.

DAF-16/FoxO regulation by kinases other than Akt
While IIS-regulated DAF-16/FoxO phosphorylation by AKT-1 and AKT-2 at conserved RxRxxS/T motifs promotes the cytoplasmic sequestration of DAF-16/FoxO, phosphorylation of DAF-16/FoxO by other kinases may also regulate DAF-16/FoxO activity.For example, the mammalian MST1 Ste20-like kinase activates FoxO3 by phosphorylating it at a conserved residue within its DNA binding domain (S207), thus promoting its nuclear translocation by disrupting interactions with 14-3-3 proteins.Mammalian MST1 can also phosphorylate C. elegans DAF-16/FoxO at the site corresponding to S207 in FoxO3 (S196).Overexpression of the C. elegans MST-1 homolog CST-1 extends life span in a DAF-16/FoxO-dependent manner, suggesting that MST1 activates FoxO transcription factors via a conserved mechanism (Lehtinen, 2006).However, phosphorylation of the FoxO1 DNA binding domain by MST1 inhibits its binding to cognate DNA sites in vitro (Brent et al., 2008).Therefore, our understanding of how MST1 regulates FoxO activity appears to be incomplete.
In mammalian cells, oxidative stress activates FoxO4 through c-Jun N-terminal kinase (Jnk)-dependent phosphorylation of two threonine residues near its C-terminus (T447 and T451) (Essers et al., 2004).Jnk also activates FoxO in invertebrates, as Jnk orthologs promote life span extension in a FoxO-dependent manner in C. elegans and Drosophila (Oh et al., 2005;Wang et al., 2005).However, whether the underlying mechanisms of FoxO Insulin/insulin-like growth factor signaling in C. elegans activation by Jnk are conserved is unclear; whereas Jnk phosphorylation sites lie near the C-terminus of mammalian FoxO4 (Essers et al., 2004), C. elegans JNK-1 appears to phosphorylate residues in an N-terminal fragment of DAF-16/FoxO in vitro (Oh et al., 2005).

EAK proteins
Although Akt inhibition of FoxO transcription factors by direct phosphorylation is well established as a conserved mechanism of FoxO regulation by IIS, data from mammalian cell culture as well as C. elegans experiments suggest that nuclear translocation of FoxO is necessary but not sufficient for full activation of FoxO.For example, in mammalian cells, insulin inhibits transcriptional activation by a FoxO1 mutant that is not exported from the nucleus (Tsai et al., 2003).Similarly, a DAF-16/FoxO mutant with all RxRxxS/T phosphoacceptor sites mutated to alanine localizes to nuclei constitutively but does not promote dauer arrest or extend life span unless IIS is reduced (Lin et al., 2001).These data are consistent with the existence of a conserved pathway that acts in parallel to Akt to inhibit FoxO activity after its translocation to the nucleus.
In contrast to the other five eak genes, which are specifically expressed in the XXX cells, the conserved gene eak-7 is expressed in the same tissues as daf-16/FoxO.EAK-7 and its human ortholog KIAA1609 both contain N-myristoylation motifs and a TLDc domain of unknown function (Alam et al., 2010).How EAK-7 controls nuclear DAF-16/FoxO activity is not known.Since EAK-7 localizes to the plasma membrane (Alam et al., 2010), it likely acts through intermediary proteins to regulate DAF-16/FoxO in the nucleus.
Insulin/insulin-like growth factor signaling in C. elegans

E3 ubiquitin ligases
In C. elegans, the RLE-1 ubiquitin ligase controls dauer arrest and life span by promoting the polyubiquitination and proteasomal degradation of DAF-16/FoxO (Li et al., 2007b).Interestingly, although a number of E3 ubiquitin ligases have been shown to promote FoxO polyubiquitination and degradation in mammalian cell culture, the RLE-1 ortholog Roquin is not one of them (Matsuzaki et al., 2003;Huang et al., 2005;Kato et al., 2008;Fu et al., 2009).This suggests that distinct E3 ligases may be capable of regulating FoxO turnover in different biological contexts.E3 ligases can also regulate DAF-16/FoxO indirectly: components of an SCF E3 ligase complex containing the cullin CUL-1 are required for life span extension in daf-2/IGFR mutants, and CUL-1 promotes DAF-16/FoxO target gene expression (Ghazi et al., 2007).

BAR-1/β-catenin
β-catenin has a conserved function as a FoxO regulator.A C. elegans β-catenin, BAR-1, promotes DAF-16/FoxO activity in response to oxidative stress.bar-1 mutation reduced expression of a DAF-16/FoxO-dependent reporter gene in the presence of paraquat, but not through the canonical β-catenin/TCF pathway.bar-1 is also required for dauer arrest in daf-2(m41) mutants.BAR-1 physically interacts with DAF-16/FoxO when both proteins are expressed in mammalian cell culture, and this interaction is enhanced in the presence of oxidative stress.Accordingly, mammalian β-catenin binds to FoxO transcription factors directly and promotes transcription from a FoxO-dependent promoter, supporting a conserved role for β-catenin in FOXO activation (Essers et al., 2005).
Recently, Chiang et al. characterized the C. elegans homolog of mammalian SIRT6 and SIRT7 proteins, SIR-2.4,and found that it is required for nuclear localization of DAF-16/FoxO in response to stress but dispensable for DAF-16/FoxO nuclear localization in the context of reduced IIS.Although DAF-16/FoxO is hyperacetylated in sir-2.4mutants, a catalytically inactive mutant SIR-2.4 protein rescued the DAF-16/FoxO localization defect caused by sir-2.4mutation, suggesting that SIR-2.4 controls the acetylation state of DAF-16/FoxO indirectly (Chiang et al., 2012b).

HCF-1/host cell factor
HCF-1 is the C. elegans ortholog of mammalian host cell factor, which is a nuclear protein that associates with the herpes simplex virus transcriptional activator VP16 to promote early viral gene expression during lytic infection (Wilson et al., 1993).hcf-1 mutation extends life span and enhances stress resistance in a DAF-16/FoxO-dependent manner.HCF-1 is a nuclear protein that binds to DAF-16/FoxO and inhibits its activity by preventing its association with target gene promoters (Li et al., 2008).Interestingly, SIR-2.1 may control DAF-16/FoxO activity via HCF-1, as SIR-2.1 overexpression does not enhance stress resistance or extend life span in animals lacking HCF-1.Moreover, expression profiling experiments demonstrate significant overlap in the sets of genes regulated by SIR-2.1 overexpression and loss of HCF-1, and HCF-1 physically associates with SIR-2.1 (Rizki et al., 2011).Mammalian host cell factors also regulate FoxO target gene expression and form a complex with FoxO3 and SIRT1, suggesting that the role of host cell factors in mediating sirtuin control of FoxO transcription factor activity may be evolutionarily conserved (Li et al., 2008;Rizki et al., 2011).

SMK-1
SMK-1, which is orthologous to the mammalian protein phosphatase 4 regulatory subunit 3A, was originally identified in a screen in Dictyostelium for suppressors of the MEK1 null phenotype (Mendoza et al., 2005).C. elegans SMK-1 is a nuclear protein that is expressed in the intestine, hypodermis, and nervous system.It is required for longevity in IIS loss-of-function mutants but does not influence dauer arrest (Wolff et al., 2006).Interestingly, SMK-1 promotes resistance to oxidative stress, ultraviolet radiation, and pathogens but does not appear to influence thermotolerance or reproduction in IIS mutants (Wolff et al., 2006).These data suggest that SMK-1 regulates a subset of DAF-16/FoxO functions.Although SMK-1 co-localizes and genetically interacts with DAF-16/FoxO, how it controls DAF-16/FoxO activity [specifically, whether it promotes the dephosphorylation of DAF-16/FoxO, and if so, at which site(s)] remains unclear.Furthermore, the full complement of DAF-16/FoxO targets that are regulated by SMK-1 is not known.

DPY-21
A genetic screen designed to identify new DAF-16/FoxO regulators that promote dauer arrest has revealed an unexpected role for the conserved dosage compensation protein DPY-21 in the regulation of IIS and DAF-16/FoxO activity.Three independent dpy-21 alleles emerged from a screen for suppressors of the eak-7; akt-1 dauer-constitutive phenotype.DPY-21 is a component of the C. elegans dosage compensation complex (DCC), a ten-protein complex that equalizes X-linked gene expression between XO males and XX hermaphrodites by binding to both hermaphrodite X chromosomes and reducing gene expression from each by approximately two-fold (see X-Chromosome dosage compensation; Meyer, 2010).DPY-21 activates DAF-16/FoxO by repressing X-linked genes encoding IIS components that promote the cytoplasmic sequestration of DAF-16/FoxO; accordingly, dpy-21 knockdown reduces DAF-16/FoxO target gene expression and enhances DAF-16/FoxO translocation to the cytoplasm (Dumas et al., 2013a).
Insulin/insulin-like growth factor signaling in C. elegans 6.12.PQM-1 The identification of the DAF-16 Associated Element as an overrepresented motif in the promoters of Class 2 (daf-2-downregulated) genes suggested that another factor might collaborate with DAF-16/FoxO in IIS-dependent gene regulation.By combining a ranking of genes by DAF-16/FoxO responsiveness with ChIP-Seq analysis, Tepper, et al. (2013) identified the zinc finger transcription factor PQM-1 as a candidate DAE-binding factor.PQM-1 binds to a Class 2 gene promoter in a DAE-dependent manner, and expression is reduced in the context of reduced IIS, as predicted for a Class 2 gene.More interestingly, PQM-1 and DAF-16/FoxO occupy the nucleus in a mutually exclusive and IIS-dependent manner.When IIS is intact, PQM-1 resides in the nucleus, activating growth and development genes, while DAF-16/FoxO is in the cytoplasm.Under stress or conditions of reduced IIS, DAF-16/FoxO enters the nucleus, inducing the expression of stress-response genes, and PQM-1 is primarily cytoplasmic.While PQM-1 is regulated by IIS, DAF-16 and PQM-1 also have an influence on the nuclear localization of one another, as the loss of each increases the nuclear occupancy of the other (Tepper et al., 2013).

FoxO regulators in other organisms
The conserved gene Melted was identified in a gain-of-function screen for genes that affect cell growth in Drosophila (Teleman et al., 2005).Melted mutants are smaller than wild-type flies; however, in contrast to Drosophila mutants with reduced IIS (Bohni et al., 1999), they have reduced triglyceride stores.The Melted protein contains a C-terminal PH domain that mediates plasma membrane localization.Melted binds to dFoxO and promotes insulin-dependent localization of dFoxO to the plasma membrane.Accordingly, transcript levels of the dFoxO target gene 4E-BP are elevated in Melted mutants.Both the increase in 4E-BP expression as well as the reduction in total body triglycerides observed in Melted mutants are suppressed by mutations in dFoxO.These observations suggest that Melted controls fat metabolism by regulating the expression of dFoxO target genes (Teleman et al., 2005).Although the C. elegans Melted ortholog K10B4.3 has not yet been implicated in IIS, RNAi of K10B4.3 extends C. elegans life span (Hamilton et al., 2005).Whether or not this phenotype requires daf-16/FoxO is not known.
A screen for modulators of Drosophila FoxO (dFoxO) activity has revealed the role of two conserved pathways in dFoxO regulation.RNAi of Drosophila ksr (kinase suppressor of Ras) enhances the dFoxO gain-of-function phenotype by reducing sensitivity to insulin, thus promoting dFoxO nuclear translocation (Zhang et al., 2011).ksr was first identified as a protein that is required for Ras pathway action in C. elegans and Drosophila (Kornfeld et al., 1995;Sundaram and Han, 1995;Therrien et al., 1995).Accordingly, RNAi of Drosophila MEK also reduced sensitivity to insulin.Thus, the Drosophila Ras/Raf/MEK pathway antagonizes FoxO action (Zhang et al., 2011).Whether the C. elegans Ras pathway (Canonical RTK-Ras-ERK signaling and related alternative pathways) regulates DAF-16/FoxO activity is not known.This same screen revealed a role for endoplasmic reticulum (ER) stress in dFoxO regulation.ER stress activates dFoxO by inducing dFoxO phosphorylation and subsequent nuclear translocation.dFoxO activation by ER stress requires phosphorylation of dFoxO at serine 243 by the ER-stress-dependent kinase PERK.PERK also phosphorylates human FoxO1 in vitro and potentiates FoxO1 activity in human cells, suggesting that FoxO regulation by ER stress is evolutionarily conserved (Zhang et al., 2013).A role for the C. elegans PERK ortholog PEK-1 in DAF-16/FoxO regulation and life span control has not yet been established, and the PERK phosphorylation site in dFoxO does not appear to be conserved in DAF-16/FoxO.However, ER stress likely influences DAF-16/FoxO activity, as the ER-stress-activated transcription factor XBP-1 is required for life span extension in daf-2/IGFR mutants and collaborates with DAF-16/FoxO to regulate gene expression (Henis-Korenblit et al., 2010).

Other transcription factors regulated by IIS
Although genetic analysis indicates that DAF-16/FoxO is the major transcriptional output of IIS, other transcription factors that participate in the regulation of DAF-16/FoxO target gene expression are also regulated by IIS.For example, HSF-1, the heat shock transcription factor that controls stress-induced gene expression (Morley and Morimoto, 2004), extends life span in a DAF-16/FoxO-dependent manner when overexpressed (Hsu et al., 2003) and delays the onset of protein aggregation in animals expressing a fluorescent protein containing polyglutamine repeats (Morley et al., 2002;Hsu et al., 2003).HSF-1 and DAF-16/FoxO promote longevity by binding to HSEs (heat shock elements) and DBEs (DAF-16/FoxO binding elements) in the promoters of genes encoding small heat shock proteins (Hsu et al., 2003).Notably, HSF-1 is required for longevity induced by reduced IIS.Recent work has shown that IIS regulates HSF-1 activity by promoting the formation of a multimolecular complex containing the conserved proteins DDL-1 and DDL-2 that sequesters HSF-1 in the cytoplasm (Chiang et al., 2012a) (Figure 1).
The C. elegans Nrf family transcription factor SKN-1 is an important component of the organismal response to oxidative stress and is required for life span extension by dietary restriction or inhibition of the TORC1 or TORC2 pathways (An and Blackwell, 2003;Bishop and Guarente, 2007;Robida-Stubbs et al., 2012).SKN-1 regulates genes that defend against oxidative and xenobiotic stress, along with many genes involved in protein homeostasis and metabolism (An et al., 2005;Kahn et al., 2008;Oliveira et al., 2009;Li et al., 2011;Paek et al., 2012).Some of its downstream targets appear to be shared with DAF-16/FoxO (An et al., 2005;Oliveira et al., 2009).As is the case for HSF-1, SKN-1 also contributes to life span extension in animals with reduced IIS (Tullet et al., 2008).IIS inhibits SKN-1 activity by promoting its cytoplasmic sequestration through an RxRxxS/T motif that can be directly phosphorylated by Akt/PKB and related kinases (Tullet et al., 2008) (Figure 1).

Components of IIS in other organisms that are conserved in C. elegans
Other than C. elegans, insulin signaling has been most intensively studied in mammals and Drosophila.This section highlights evolutionarily conserved proteins and pathways that have been implicated in IIS in mammals and/or flies but do not have clear functions in C. elegans IIS.

Ras signaling
In mammalian cells, insulin and IGF-1 activate Ras signaling through the recruitment and phosphorylation of IRS and Shc adaptor proteins.Tyrosine phosphorylation of IRS proteins and Shc by ligand-activated insulin and IGF-1 receptors generates binding sites for the Grb2/Son-of-Sevenless (Sos) complex, thus promoting the translocation of the Sos guanine nucleotide exchange factor into proximity with membrane-associated Ras.Sos activates Ras by catalyzing the exchange of GDP bound to Ras for GTP.GTP-bound Ras activates the Raf/MEK/MAPK kinase cascade, leading to transcriptional regulatory events that promote mitogenesis (Siddle, 2011).
In C. elegans, the role of Ras in insulin signaling is less clear.C. elegans LET-60/Ras has a well-established role in vulval cell fate specification during larval development, where it is regulated by the epidermal growth factor receptor (EGFR) ortholog LET-23 (Canonical RTK-Ras-ERK signaling and alternative pathways).Based on the mammalian paradigm, DAF-2/IGFR activation would be predicted to activate LET-60/Ras.If activated LET-60/Ras transduces signals from DAF-2/IGFR, then reduction of LET-60/Ras signaling should promote dauer arrest and/or extend life span, whereas increases in LET-60/Ras activity should suppress the dauer-constitutive and life span extension phenotypes caused by reduced IIS.However, unlike age-1/PI3K, pdk-1, and akt-1 loss-of-function mutants, which have a dauer-constitutive phenotype, LET-60/Ras pathway loss-of-function mutations do not cause dauer arrest, suggesting that the PI3K/Akt pathway is the major output of DAF-2/IGFR.However, Ras pathway mutations do influence dauer arrest modestly in some (but not all) mutants with reduced IIS.For example, the let-60(n1046) gain-of-function (gf) mutation can suppress dauer arrest in daf-2(m577) mutants, and loss-of-function mutations in let-60/Ras and sem-5/Grb2 enhance daf-2(m577) mutant phenotypes (Nanji et al., 2005).These data suggest that Ras signaling has a weak modulatory effect on IIS in the control of dauer arrest.let-60(gf) also partially suppresses the dauer-constitutive phenotype of the age-1(mg44) null mutant, indicating that Ras pathway action in dauer regulation is independent of PI3K (Nanji et al., 2005).
Ras pathway mutations have complex effects on life span extension in the context of reduced IIS.In contrast to the enhancement of the dauer-constitutive phenotype of daf-2/IGFR mutants by LET-60/Ras signaling loss of function, these same mutations partially suppress life span extension of daf-2(m577).Thus, Ras signaling has opposite effects on dauer-constitutive and life span extension phenotypes in the context of reduced IIS.Although let-60(gf) extends maximum life span in daf-2/IGFR mutants, it has no effect on mean life span due to increases in early adult mortality in daf-2; let-60(gf) double mutants (Nanji et al., 2005).The interpretation of these life span experiments is further complicated by the pleiotropies of Ras pathway mutants (Canonical RTK-Ras-ERK signaling and related alternative pathways).
Insulin/insulin-like growth factor signaling in C. elegans Whether Ras acts downstream of or parallel to DAF-2/IGFR to influence dauer arrest and life span remains uncertain: there is no biochemical data directly implicating the LET-60/Ras pathway as a target of C. elegans IIS, and there are differences in the functions of mammalian and C. elegans Shc (Neumann-Haefelin et al., 2008) andIRS proteins (Wolkow et al., 2002).It is also not known whether LET-60/Ras influences the activity of DAF-16/FoxO.Although the Ras/MEK/MAPK pathway activated by insulin and IGF-1 in mammals appears to play a minor role in dauer regulation and life span control by IIS in C. elegans, DAF-2/IGFR does regulate this pathway in other contexts.For example, DAF-2/IGFR signaling potentiates Ras/MEK/MAPK signaling in vulval precursor cells during vulval development by promoting phosphorylation of the MAP kinase MPK-1 in an Akt-and DAF-16/FoxO-independent manner (Nakdimon et al., 2012), and DAF-2/IGFR signaling to LET-60/Ras may promote starvation-enhanced ribosome biogenesis in males (Tan et al., 2011).Conversely, in muscle, DAF-2/IGFR signals through AGE-1/PI3K and PDK-1 to inhibit Ras/MEK/MAPK-dependent protein degradation (Szewczyk et al., 2007).

Shc
The C. elegans genome encodes two Shc homologs, one of which (SHC-1) has been studied in some detail (Mizuno et al., 2008;Neumann-Haefelin et al., 2008).In contrast to mammalian Shc, which is a critical mediator of Ras pathway activation by insulin (Ravichandran, 2001), C. elegans SHC-1 does not appear to regulate Ras activity, as shc-1 mutants do not have phenotypes reminiscent of Ras pathway loss-of-function mutants (Mizuno et al., 2008).This may be a consequence of the fact that the major tyrosine phosphorylation sites on mammalian Shc that engage the GRB2/Sos complex are not conserved in SHC-1 (Luzi et al., 2000).This observation may also explain why Ras activation plays a more significant role in mammalian IIS than it appears to in C. elegans IIS.Although SHC-1 coprecipitates with DAF-2/IGFR when both are overexpressed in mammalian cell culture (Neumann-Haefelin et al., 2008), the biological significance of this interaction is not known.
SHC-1 has an established role in promoting stress resistance through a physical interaction with the MAP kinase kinase MEK-1 (Mizuno et al., 2008;Neumann-Haefelin et al., 2008).This interaction is required for resistance to heavy metals (Mizuno et al., 2008) and may also promote the nuclear translocation of DAF-16/FoxO in the context of environmental stress (Mizuno et al., 2008;Neumann-Haefelin et al., 2008).

TOR
The mammalian Target of Rapamycin (mTOR) is a conserved serine/threonine kinase that acts in distinct multiprotein complexes to control cell growth in response to nutrient availability and energy balance.mTOR functions primarily through two complexes, known as TORC1 (Target of Rapamycin Complex) and TORC2, which consist of common and distinct components.In mammalian cells, insulin and IGF-1 activate TORC1 through phosphorylation and inactivation of the TSC2 tumor suppressor by Akt (Avruch et al., 2006).By contrast, whether and how IIS couples to TOR in C. elegans is unclear.The long life span of let-363/TOR RNAi-treated animals is independent of daf-16/FoxO (Vellai et al., 2003).Instead, let-363/TOR appears to be part of the dietary restriction-mediated (DR) life span extension pathway (Hansen et al., 2007;Ching et al., 2010), which is genetically separable from IIS in C. elegans and requires the FoxA ortholog PHA-4 rather than DAF-16/FoxO (Panowski et al., 2007).The presence of nutrients promotes LET-363/TOR activity, which in turn represses PHA-4/FoxA activity, promoting development and normal life span (Sheaffer et al., 2008).Mutants of both let-363 and daf-15, the C. elegans ortholog of RAPTOR (regulatory associated protein of mTOR), form dauer-like larvae constitutively, and daf-15/RAPTOR heterozygotes are long-lived, like let-363/TOR knockdown animals (Jia et al., 2004).While there is no data to suggest that LET-363/TOR is regulated by IIS, daf-15/RAPTOR expression is negatively regulated by DAF-16/FoxO (Jia et al., 2004), providing a mechanism by which the IIS and DR pathways could be coordinated in growth vs. longevity decisions.
Although evidence is lacking for the regulation of TOR activity by IIS in C. elegans, TORC1 may regulate DAF-16/FoxO activity independently of IIS.The effect of TORC1 activation on DAF-16/FoxO appears to be context-dependent.The C. elegans ortholog of the TORC1 activator Rheb, RHEB-1, is required for life span extension in response to intermittent fasting (IF) (Honjoh et al., 2009).IF induces DAF-16/FoxO nuclear translocation and DAF-16/FoxO target gene expression, both of which require RHEB-1 and LET-363/TOR activity (Honjoh et al., 2009).By contrast, under ad lib feeding conditions, TORC1 inhibition extends life span in a DAF-16/FoxO-dependent manner and promotes DAF-16/FoxO target gene expression and nuclear translocation of Insulin/insulin-like growth factor signaling in C. elegans the D/F/H isoform of DAF-16/FoxO (Robida-Stubbs et al., 2012).Thus, TORC1 promotes DAF-16/FoxO activity in the context of IF but inhibits DAF-16/FoxO activity in well-fed animals.The molecular basis for DAF-16/FoxO regulation by TORC1 has not been established.

Glucose transport
In mammals, insulin regulates glucose uptake in muscle and adipose tissue by promoting the translocation of the facilitative glucose transporter GLUT4 to the plasma membrane.This process requires the activation of both canonical IRS/PI3K/Akt signaling as well as a PI3K-independent pathway involving the tyrosine phosphorylation of the adaptor proteins APS and Cbl and the subsequent nucleation of a protein complex that activates the small GTPase TC10 in plasma membrane lipid rafts (Chang et al., 2004).Although the C. elegans genome encodes members of the glucose transporter superfamily, there is no evidence to suggest that glucose transport is regulated by ILPs or DAF-2/IGFR.This may be due in part to the fact that, although the PI3K/Akt pathway and Cbl are conserved in C. elegans, the C. elegans genome appears to lack an ortholog of APS, the tyrosine phosphorylation of which facilitates insulin-dependent recruitment of Cbl to the insulin receptor and insulin-dependent translocation of GLUT4 to the plasma membrane (Liu et al., 2002).However, glucose does shorten C. elegans life span in an IIS-dependent manner, and ins-7 expression increases upon glucose exposure, suggesting that C. elegans IIS does respond to changes in glucose levels (Lee et al., 2009).

Cytohesins
The Drosophila cytohesin ortholog Steppke was identified as a gene required for larval growth (Fuss et al., 2006).It acts in the PI3K/Akt pathway to control dFoxO target gene expression.The triazole derivative SecinH3, which inhibits the guanine nucleotide exchange activity of cytohesins, reduces insulin-induced Akt phosphorylation and blocks insulin-dependent regulation of 4E-BP and insulin receptor transcript abundance in Drosophila S2 cells; accordingly, SecinH3 inhibits insulin signaling in human HepG2 cells and causes hepatic insulin resistance in mice (Fuss et al., 2006;Hafner et al., 2006).Thus, cytohesins have a conserved role in promoting insulin/PI3K/Akt signaling.Although the C. elegans genome contains a cytohesin ortholog (K06H7.4), a potential role in IIS has not been reported.

System-wide integration of IIS: non-autonomy and interaction with other pathways
Dauer entry, longevity, and reproduction in any individual are determined by the integration of information about nutrient availability, number of neighbors (or competitors), and temperature; these variables may be interpreted as predictors of future survival and reproductive conditions.Much of the C. elegans systemic response to these environmental factors is mediated through IIS and its interactions with other signaling pathways.

IIS-mediated cell non-autonomous regulation of longevity, dauer formation, metabolism, and reproduction:
Larval and adult expression of a daf-2p::GFP promoter fusion has been reported in the intestine and nervous system (Hunt-Newbury et al., 2007), and recent immunofluorescence studies with affinity-purified anti-DAF-2 antibodies suggest that DAF-2/IGFR is most highly expressed in neurons and in the XXX secretory cells (which produce dafachronic acid and regulate dauer formation (Schaedel et al., 2012)), with weaker fluorescence in non-neuronal tissues, such as hypodermis (Kimura et al., 2011).Using mosaic analysis (Nomarski images for learning the anatomy, with tips for mosaic analysis), Apfeld and Kenyon found that DAF-2/IGFR acts cell non-autonomously to regulate dauer formation, longevity, and intestinal pigmentation associated with fat accumulation (Apfeld and Kenyon, 1998).To identify the site of IIS regulation of longevity and metabolism, Wolkow et al. (2000) used tissue-specific promoters to rescue daf-2/IGFR in specific tissues in daf-2(-) mutants.They found that daf-2/IGFR expression in neurons and intestine, but not in muscle, rescued longevity and dauer phenotypes, while restoration in muscle rescued metabolic defects (Wolkow et al., 2000).These results suggest that DAF-2/IGFR acts primarily in the neurons to control such system-wide phenotypes as longevity and dauer formation, but may have cell autonomous metabolic functions, at least when expressed at high levels in the intestine.The reason for the apparent discrepancy between the Apfeld and Kenyon and Wolkow et al. results might be due to technical differences (such as variation and non-endogenous levels of gene expression), but together the results do suggest that IIS in neurons is critical for non-autonomous regulation of longevity and dauer formation (Apfeld and Kenyon, 1998;Wolkow et al., 2000).Furthermore, the results imply that a downstream, secondary signal likely communicates with other tissues to coordinate the dauer decision and longevity response.
Insulin/insulin-like growth factor signaling in C. elegans The tissue-specific and cell autonomous roles of DAF-16/FoxO have also been investigated.The daf-16a and daf-16d/f/h isoforms are expressed in most tissues, whereas the daf-16b isoform has a more limited tissue distribution (Section 5.1) (Lin et al., 1997;Ogg et al., 1997;Lee et al., 2001;Kwon et al., 2010).Similar to the daf-2/IGFR mosaic and neuronal rescue results (Apfeld and Kenyon, 1998;Wolkow et al., 2000), expression of DAF-16A in the nervous system restored dauer formation in a significant fraction of daf-16; daf-2 double mutant animals (Libina et al., 2003), suggesting the existence of an organism-wide response to DAF-16/FoxO activation in neurons.Intriguingly, these animals arrest as dauers in the absence of DAF-16/FoxO activity in other important tissues such as the intestine and hypodermis.When DAF-16A is expressed solely in the intestine in a daf-16; daf-2 double mutant background, life span is increased up to 60%, while its expression in other tissues, including neurons, had little to no effect on longevity (Libina et al., 2003).These results imply that DAF-16/FoxO may have tissue-specific longevity targets, or that the intestine is particularly sensitive to age-related damage (Libina et al., 2003).Currently, there is no data on the role of DAF-16/FoxO in the hypodermis, a tissue that is increasingly recognized for its role in endocrine function.
In addition to its longevity phenotype, daf-16/FoxO restoration in the intestine also rescued reproductive aging, indicating a novel role for DAF-16/FoxO in this tissue (Luo et al., 2010).Surprisingly, DAF-16/FoxO expression in the muscle also rescued reproductive aging, suggesting that the muscle plays a previously unappreciated role in IIS/DAF-16 regulation of reproduction (Luo et al., 2010).
Signaling between tissues, employing both insulin-like peptides (Murphy et al., 2007) and lipophilic hormones (Gerisch et al., 2001;Jia et al., 2002;Mak and Ruvkun, 2004), likely coordinates IIS activity and longevity state throughout the organism.For example, DAF-16/FoxO activity in one tissue can regulate its activity in other tissues, suggesting that DAF-16/FoxO regulates IIS activity in neighboring cells through downstream signaling (Libina et al., 2003).This feed-forward activation of DAF-16/FoxO (FoxO-to-FoxO signaling) in the regulation of longevity is mediated largely by the activity of the INS-7 ILP (Murphy et al., 2007).DAF-16/FoxO activity in the intestine and muscle likely activate signaling to the germline, although the nature of this signal (or signals) and its receptor is unknown.Reproductive aging is also regulated by IIS in the intestine and muscles, but the signal for this non-autonomous activity is unknown (Luo et al., 2010).

IIS interactions with the dafachronic acid/DAF-12 pathway
Genetic analysis underscores the complexity of interactions between IIS and the dafachronic acid/DAF-12 pathway (for more details on this pathway see WormBook chapter Dauer).Mutants that lack dafachronic acid (in which the DAF-12 nuclear receptor is unliganded) undergo dauer arrest independently of DAF-16/FoxO (Gerisch et al., 2001;Jia et al., 2002), suggesting that unliganded DAF-12 acts downstream of IIS.However, studies with eak mutants have shown that the expression of some DAF-16/FoxO target genes requires DAF-12 activity (see Section 6.2), indicating that DAF-16/FoxO and DAF-12 may act in parallel in some contexts (Zhang et al., 2008;Alam et al., 2010;Dumas et al., 2010).Furthermore, whereas daf-12 null mutations fully suppress dauer arrest in most dauer-constitutive mutant backgrounds, including weaker daf-2/IGFR loss-of-function alleles such as e1368, daf-2(e1370); daf-12(null) double mutants have a synthetic phenotype, undergoing non-dauer larval arrest at 25°C (Dumas et al., 2013b).Recent work on the interactions of IIS and DA/DAF-12 pathways in life span control suggest that, whereas liganded DAF-12 promotes longevity in the context of reduced IIS, unliganded DAF-12 can either shorten or prolong life span in animals with reduced IIS (Dumas et al., 2013b).

IIS/TGF-β crosstalk
The TGF-β pathway that controls dauer arrest (see WormBook chapter Dauer) also regulates the expression of ins-7 and ins-18 (Liu et al., 2004;Shaw et al., 2007), linking TGF-β dauer regulatory activity to IIS, possibly in order to coordinate dauer and longevity decisions through multiple signaling inputs (Shaw et al., 2007).A recent study of the pyruvate dehydrogenase phosphatase homolog PDP-1, which acts in the TGF-β dauer pathway to affect insulin signaling upstream of IIS (Narasimhan et al., 2011), supports such a model of IIS/TGF-β signaling coordination through ILP regulation (Shaw et al., 2007).

IIS coordination of reproduction and longevity
Communication from the soma to the germline, and from the germline back to the soma (Hsin and Kenyon, 1999;Luo et al., 2010), might enable an organism to tune its reproductive span and life span properly to adapt to changing nutrient availability (Luo and Murphy, 2011).While the signals between the tissues are unknown, IIS plays a key role in each arm of this coordinating program, highlighting its importance in the regulation of important life history traits.
Insulin/insulin-like growth factor signaling in C. elegans

Conclusions and Future Directions
Here we have discussed the major components of C. elegans IIS and its impact on many phenotypes and behaviors, including development, longevity, stress resistance, and reproduction.Despite the vast amount of information that study of this pathway in C. elegans has revealed, important questions remain.For example, IIS has both cell autonomous and non-autonomous functions, and the activity of IIS in each tissue is not yet known.Tissue specificity has only been addressed in a few studies, and there is a disconnect between the daf-2/IGFR and daf-16/FoxO results regarding their roles in particular tissues.Furthermore, most genome-wide experiments (e.g., microarrays, SAGE, ChIP-Seq) performed on whole animals have not revealed insights into tissue-specific targets of IIS and DAF-16/FoxO.In fact, many of the organismal phenotypes studied in C. elegans IIS mutants (longevity, dauer formation, pathogen survival, stress resistance) are the result of system-wide signal integration.For most of these phenotypes, the communicating signals between tissues are still unknown.
IIS influences important life history phenotypes and integrates signaling information at least in part through interactions with other signaling pathways (including TGF-β, dafachronic acid/DAF-12, and Ras/Raf/MAP kinase pathways) and important transcription factors.Nonetheless, it is likely that the extent of these interactions is not yet fully appreciated.
ILPs may play a role in tissue-specific responses and coordination, but the sheer number of ILPs in C. elegans raises another set of questions.Where do they all act?What do they all do?Why are there so many?Do they all act through DAF-2/IGFR, or are there alternative receptors for ILPs?What is the cost of the increased robustness of IIS pathway mutants (said another way, why are all worms not IIS mutants?)? First and foremost, IIS mutants are slow to develop, they produce slightly fewer progeny (particularly early progeny), and they produce them over a longer period of time; together, these delays result in lower fitness when compared to wild-type worms, despite the fact that IIS mutants are more stress resistant (Walker et al., 2000;Jenkins et al., 2004).However, IIS mutants are better able to cope with stressful conditions in almost all stages of life.For example, under stressful conditions in early development, IIS controls the trigger for dauer formation, and starvation-induced L1 arrest is also IIS-dependent.With milder stresses, the same pathway can respond with protective transcriptional responses that extend somatic longevity, germline proliferation, and oocyte quality maintenance.Together, these data suggest that the responsiveness of the insulin/IGF-1-like signaling pathway may allow C. elegans to quickly respond to stressful conditions, allowing them to live to fight, or at least to reproduce and survive, another day.

Figure 2 .
Figure 2. daf-16/FoxO isoforms.A. Genomic structure of the daf-16/FoxO locus (not drawn to scale).a, b, and d/f/h isoforms and their promoters are depicted.Isoform-specific exons are colored green, orange, and purple.B. DAF-16/FoxO polypeptides.Amino acid sequences encoded by isoform-specific exons are colored green, orange, and purple.All isoforms share two conserved Akt-dependent phosphorylation sites that lie carboxyterminal to the Forkhead DNA binding domain but have unique sites in their aminotermini.See text for details.