Lifespan Regulation by Insulin Signaling Through Phosphorylation of Proteins Beyond FOXO

Insulin/IGF-1 Signaling (IIS) constrains longevity by inhibiting the transcription factor FOXO. Beyond FOXO, little is known about how phosphorylation—as mediated by IIS kinases— regulates lifespan. Here, we proled IIS-dependent phosphorylation changes in a large-scale quantitative phosphoproteomic analysis of wild-type and three IIS mutant C. elegans strains. Our state-of-the-art analysis experimentally identied more than 15,000 phosphosites, among which 448 were differentially phosphorylated in the long-lived daf-2/insulin receptor mutant. We developed a machine-learning-based tool for systematically ranking the likely functional importance of phosphosites to guide candidate selection for follow-up validation. We show that AKT-1 pT492 inhibits DAF-16/FOXO and compensates the loss of daf-2 function, that EIF-2α pS49 potently regulates protein synthesis and daf-2 longevity, and that reduced phosphorylation of multiple germline proteins (e.g., CDK-1) apparently transmits a signal representing reduced DAF-2 signaling to the soma. Finally, kinase-substrate analysis and subsequent experimental validation conrm that casein kinase 2 negatively regulates lifespan. Our new benchmark data resource and machine-learning tool enables unprecedented access to detailed functional insights for studies of longevity.


Introduction
Despite the great diversity of lifespan in the animal kingdom, ancient genetic pathways have been found to regulate lifespan across species 1,2 . The best-known example is Insulin/Insulin Growth Factor 1 (IGF-1) signaling (IIS). Polymorphisms of the component genes of this pathway are tightly associated with human longevity 2 . Disrupted IIS can extend lifespan up to ten fold in C. elegans 3 . The canonical IIS pathway of the worm comprises insulin-like ligands, the insulin/IGF-1 receptor tyrosine kinase DAF-2, the phosphatidylinositol-3-OH kinase (PI3K) AGE-1, the serine/threonine kinases PDK-1, AKT-1, and AKT-2, and a downstream transcription factor (TF) DAF-16, the C. elegans homolog of human FOXO 4 . Inhibiting the IIS kinases leads to nuclear translocation of DAF-16, subsequent activation of target gene transcription, and ultimately lifespan extension. While DAF-16 and other TFs are required for IIS-mediated lifespan extension 4 , it has not been demonstrated that activation of TFs and the subsequent regulation at transcriptional level is su cient for daf-2 longevity.
Deep pro ling of the C. elegans transcriptomes and proteomes made clear that age-dependent protein abundance changes correlated poorly with mRNA abundance changes 5,6 . Similarly, for a signi cant subset of proteins that are up-or down-regulated in the long-lived daf-2 mutant worms, there were no corresponding changes in the abundance of their mRNA templates. These changes affect known lifespan modulators such as components of translational machinery 7,8,9 , indicating that post-transcriptional regulation does impact lifespan control.
Phosphorylation modi cation, among various post-transcriptional regulation, is understood as the fundamental regulatory mechanism underlying insulin/IGF-1 signal transduction. The IIS kinases AKT-1 and AKT-2 prevent lifespan extension by sequestering DAF-16 in the cytoplasm. Other kinases such as JNK-1/JNK, CST-1/MST1, and AAK-2/AMPK contribute to daf-2 longevity partly by promoting nuclear translocation of DAF-16 4 . All those kinases directly phosphorylate DAF-16 in vitro, yet such phosphorylation-based regulation has not been con rmed in vivo. daf-2 longevity is also modulated by protein phosphatases: PPTR-1, a regulatory subunit of PP2A, reduces the phosphorylation of AKT-1 T350, and renders AKT-1 less active 10 ; PP4 SMK − 1 dephosphorylates the transcriptional regulator SPT-5/SUPT5H, which facilitates DAF-16 activity in daf-2 worms 11 . To our knowledge, very few studies have reported phosphosites which regulate lifespan 12,13,14,15,16 , and there are no reports of large-scale survey studies of C. elegans IIS-related phosphorylation.
To date, 119,809 phosphorylation sites on human proteins have been identi ed; the number for C. elegans is only 10,767 17,18 . The paucity of data in C. elegans re ects the fact that C. eleganstraditionally a genetic system -has fallen seriously behind, even though mass spectrometry (MS) -based phosphoproteomics technologies have been leaping forward. Here, using the state-of-the-art phosphoproteomics technology, we surveyed the landscape of protein phosphorylation in C. elegans and compared through a 15 N-labled reference sample the long-lived daf-2 mutant worms to the wild type, the daf-16 mutant, and the daf-16; daf-2 double mutant. This led to the identi cation of over 15,000 phosphosites, a doubling of the C. elegans phosphorylation database, and the discovery of 448 phosphosites regulated by IIS. Further, we developed a machine learning based algorithm to identify phosphosites that likely exert biological impacts. This tool guided our functional investigations of three phosphosites, all of which do impact worm lifespan. Brie y, we added a new element -phosphorylation of AKT-1 T492 -to the negative feedback regulation mechanism of IIS. We also uncovered two branches of signaling downstream of DAF-2 that extend lifespan: inhibition of translation through phosphorylation of EIF-2α S49 by GCN-2 and signaling in the germline through phosphorylation of CDK-1 T179. Lastly, global enrichment analysis and subsequent validation experiments highlighted the germline as a target tissue of IIS and revealed a role for casein kinase 2 in lifespan determination.

Results
Pro ling the C. elegans Phosphoproteome by MS in Advanced Approach The phosphoproteome of C. elegans has not been surveyed rigorously; seeking to increase the coverage of the C. elegans phosphoproteome while aiming for high accuracy in both identi cation and quanti cation of phosphopeptides, we combined a number of technical elements and optimized the analytical work ow (Fig. 1A). These technical elements include extensive high-pH reverse phase fractionation coupled with interval pooling 19,20 , polyMAC-Ti enrichment of phosphopeptides 21,22 , highspeed and accurate-mass mass spectrometry, and stable isotope ( 15 N) metabolic labeling, a highly accurate quantitative proteomics strategy (Fig. 1A).
From wild type (WT) C. elegans and the insulin signaling mutants (daf-2, daf-16, and the daf-16; daf-2 double mutant) -each analyzed in three or four biological replicates and two technical replicates -we identi ed a total of 15,443 phosphorylation sites with > 0.75 PhosphoRS site probability 23 . These phosphosites are represented by 22,536 phosphopeptides or 15,723 phosphoisoforms that belong to 4,418 proteins (Supplementary Fig. 1A-C). 9,949 phosphosites identi ed in this study are not covered by dbPAF, which is a comprehensive database dedicated to collecting phosphosites in humans, animals, and fungi 17 . Notably, the addition of these newly identi ed phosphosites close to doubles the current collection for C. elegans (Fig. 1B).
Although it is well established that phosphorylation is the primary means by which the IIS pathway transmits signals, very little is known about which sites are phosphorylated, even for the core components of C. elegans IIS. For example, dbPAF presently contains no phosphosites for the PI-3 kinase AGE-1. Here, our phosphoproteomics analysis uncovered 32 phosphosites in ten C. elegans IIS proteins, 17 of which have not been previously reported (Fig. 1C). These new, high con dence phosphosites ( Fig. 1C, dark blue) are distributed throughout the pathway, from the upstream insulin-like ligands to the downstream FOXO transcription factor DAF-16, and for every kinase in between.
More than 15,000 phosphopeptides were quanti ed against their 15 N-labeled cognate peptides, which were introduced as an internal reference standard by feeding C. elegans entirely on 15 N-labeled bacteria ( Supplementary Fig. 1B, see Methods). These peptides represent 10,705 quanti able phosphoisoforms (Fig. 1D, Supplementary Fig. 1B-C, and Supplementary Table 1), about a quarter of which carry combinatorial information for two or more phosphosites. Clustering analysis of their abundance levels (relative to 15 N-labeled peptides) across WT and IIS mutants indicated that the daf-2 mutant samples are clearly different from the WT, the daf-16, and daf-16; daf-2 samples. In other words, the quantitative phosphoproteomics data clustered according to the lifespan phenotype, not by batch ( Supplementary  Fig. 1E). The Spearman correlation coe cient between biological replicates of the same worm strain is in the range of 0.70-0.91. These results are indicative of high data quality for phosphopeptide identi cation and quanti cation.

Phosphorylation Changes Resulting from Genetic Disruption of IIS
Disrupting the activity of IIS induced abundance changes on 501 phosphoisoforms (> 1.5-fold in at least one of the IIS mutants relative to WT, Supplementary Fig. 1F). As expected, clustering and pathway enrichment analysis show that phosphorylation on proteins involved in FOXO signaling and longevity regulation were down-regulated in the daf-2 mutant. We also found that proteins related to protein synthesis or degradation had lower phosphorylation levels in the daf-2 mutant, but the changes were not preserved in the daf-16 or daf-16; daf-2 mutants. Notably, glycerolipid metabolism and glycerophospholipid metabolism were enriched for proteins with up-regulated phosphorylation in the daf-2 mutant. Up-regulation of lipid metabolism is a major phenotype of daf-2 mutants 4,24 . Underlying this phenotype are gene expression changes 25,26 and protein abundance changes 8 . Thus, our results should be highly informative in supporting characterization of biological processes regulated by IIS and are likely to extend the existing mechanistic understanding of this eld to encompass PTM-level regulation.

Development of a Computational Strategy to Prioritize Putative Functional Phosphosites
A bottleneck in present-day biomedical research is a lack of e cient methods for extracting useful information from omics data 27,28 . To facilitate the translation of phosphoproteomics data into biological insights, we developed a machine learning based method named iFPS (Inference of Functional Phosphorylation Sites) to predict whether a given phosphosite likely exerts a biological impact ( Fig. 2A).
Although the lack of suitable training data has to date prevented development of such a tool in C. elegans, note that a tool for similar predictive analysis of phosphorylation sites recently became available for Homo sapiens 18 . Our iFPS tool assesses six types of constraints for each phosphosite under examination (Supplementary Fig. 2A-H, see Methods), including 1) how many kinase families have consensus substrate sites that match its sequence context?; 2) how evolutionarily conserved a phosphosite is; 3) how many interacting domains are predicted to be in uenced by phosphorylation at this site, 4) occurrence of a predicted acetylation site near the phosphosite (which could engage in PTM crosstalk); 5) relative surface accessibility; and 6) predicted secondary structure of the peptide containing the phosphosite.
For the initial iFPS training data, we searched in the literature for C. elegans phosphosites whose functions have been experimentally validated: the resulting 121 functional phosphosites served as the original positive training data set (Supplementary Table 2). The negative data set contained 605 (121 × 5) randomly selected phosphosites from dbPAF. Multinomial logistic regression (MLR), a widely used machine learning algorithm, was adopted for training the computational models, and 10-fold crossvalidations were performed. The nal model was determined automatically, and the highest area under the curve (AUC) value was 0.88 ( Supplementary Fig. 2I). iFPS was applied to score all phosphosites identi ed in this study, which contain 31 known functional phosphosites from the positive data set (Supplementary Table 2). Distributions of iFPS scores show that functionally impactful phosphosites ranked higher than other phosphosites (Fig. 2B). Half of the functional phosphosites were among the top 5% iFPS scoring list.
Next, we focused on putative functional phosphosites regulated by daf-2. From the quantitation data reliability measured at least three times in both the daf-2 mutant and a control (WT or WT plus daf-16 mutants, see Methods), we found 222 down-and 226 up-regulated phosphoisoforms upon reduction of daf-2 activity (Fig. 2C). By overlapping the phosphosites regulated by daf-2 and the top 5% highest scoring iFPS phosphosites (Supplementary Table 2-3), we identi ed 27 high-priority phosphosites (i.e., with a high probability of being functionally impactful) (Fig. 2D). Notably, these sites do not represent a random set: the majority of the parent proteins harboring these sites function in either AMPK/insulin signaling, translation initiation/ribosome biogenesis, or cell cycle regulation. Further, 13 out of the 27 high-priority phosphosites are from eight proteins (AAK-2, AKT-1, CDK-1, DAF-16, EGL-45, MLT-3, MVK-1 and PDHA-1) known to regulate lifespan (phenotypic data from WormBase release WS275). Of note, the phosphorylation state of three conserved S/T residues of AAK-2, the catalytic subunit of C. elegans AMPK, was differentially regulated in the daf-2 mutant: phosphorylation of T597 and S601 increased whereas S553 decreased.
Any function(s) for most of the high-priority phosphorylation sites remain uncharacterized. In lifespan regulation, only S345 of DAF-16, a conserved AKT site, has been implicated: simultaneous mutation of S345 and other three predicted AKT sites induced nuclear accumulation of DAF-16, much like in the daf-2 mutant but without the extraordinary longevity phenotype 13 . To experimentally test the performance of iFPS and to esh out the mechanism of lifespan extension by protein phosphorylation in response to reduced insulin signaling, we focused on several phosphosites for in-depth functional analysis (colored red in Fig. 2D). These are AKT-1 T492, EIF-2α S49, and CDK-1 T179, one in each of the three prominent protein function groups.
Constitutive Phosphorylation of AKT-1 T492 Promotes AKT-1 Activity iFPS prioritized pT492 of worm AKT-1 (corresponding to pT450 of human AKT-1) (Fig. 3A). This site is positioned in a highly conserved turn motif near the AKT-1 C-terminus, and work in mammalian cells has shown that this site is co-translationally phosphorylated by mTORC2, supporting that this site may stabilize newly synthesized AKT 29,30 . However, the functional impact of this site has not been con rmed.
Verifying the earlier suggestion, we found that phosphorylation of C. elegans AKT-1 on T492 is constitutive. The AKT-1 protein and T492 phosphorylation levels both doubled in the long-lived daf-2 mutant (FC = 2.2-2.4, daf-2/WT) as measured by shotgun proteomics (Fig. 2D and Supplementary Table 3) and by targeted quantitation assays using synthesized peptides bearing isotope labels (Fig. 3B). Whereas the T492-containing peptide of AKT-1 was undetectable in any of the four strains analyzed (Fig. 3B), the pT492-containing peptide of AKT-1 was readily detectable, and its abundance change followed that of the AKT-1 protein very closely. Thus, the T492 site is apparently constitutively phosphorylated following ATK-1 translation.
To determine whether the loss-of-function phenotypes resulting from the T492A mutation are caused by destabilization of AKT-1, we used a knock-in approach to fuse a GFP reporter C-terminal to AKT-1. AKT-1::GFP and AKT-1-T492A::GFP were present in nearly all examined tissues, with no discernable difference in GFP intensity ( Supplementary Fig. 3B-C), suggesting that T492A imparts no or little destabilizing effect on AKT-1. However, we did observe an effect related to the subcellular localization of AKT-1. Compared with AKT-1::GFP, there is more AKT-1-T492A::GFP in the nuclei of oocytes ( Supplementary Fig. 3D). This T492A-induced localization change for AKT-1 was limited to the germline, and had high penetrance (84%). Moreover, this phenotype does not result from an overexpression artifact, because both AKT-1::GFP and AKT-1-T492A::GFP are expressed from the edited endogenous akt-1 gene locus. Since AKT-1 is normally recruited to the plasma membrane -where it transmits signals from receptor tyrosine kinases such as DAF-2-the nuclear translocation of AKT-1 may partially account for the observed loss-offunction effect of the T492A mutation. These results support that mutation of T492 to alanine impairs the activity of AKT-1, weakening AKT-1's inhibition of DAF-16 and leading to both longer lifespan and a higher propensity for dauer formation. Thus, in WT animals, constitutive phosphorylation of T492 promotes the kinase activity of AKT-1.
AKT-1 is controlled by a negative feedback loop at the gene transcription level; that is, expression of the akt-1 gene is positively regulated by DAF-16 32 , while DAF-16 itself is negatively regulated by AKT-1. In the long-lived daf-2 mutant, activated DAF-16 induces transcription of akt-1, although the akt-1 mRNA level is elevated by only 10% 33 . However, this elevation is strikingly higher when examined at the protein level: the AKT protein level is elevated by around 140% as measured by quantitative proteomics 6 , a nding validated by our data for AKT-1::GFP in the present study ( Supplementary Fig. 3E). Our phosphoproteomics analysis thus reveals T492 phosphorylation as a previously unknown layer of regulation in a complex regulatory network. Recalling that AKT-1 is phosphorylated at T492 immediately following its translation and that this PTM promotes AKT-1's activity, our work at the phosphoproteomics level underscores how a negative IIS feedback loop is intricately controlled at multiple regulatory layers, including gene transcription, protein synthesis, and post-translational modi cation (Fig. 3E).

EIF-2α pS49 Potently Regulates Protein Synthesis and Lifespan in the daf-2 Mutant
Down-regulation of the processes that support protein synthesis (e.g., translation initiation and ribosome biogenesis) has been associated with longevity in previous studies 34,35,36 . The same down-regulation trend was evident in our phosphoproteomics data: phosphorylation of multiple eukaryotic initiation factors (EIF) was generally reduced in the long-lived daf-2 mutant ( Supplementary Fig. 4A). The only exception to this trend was EIF-2α. Phosphorylation of EIF-2α at S49, which is an iFPS prioritized site, nearly doubled in the daf-2 mutant relative to WT worms ( Fig. 2D and 4A), and this was veri ed by western blotting (Fig. 4B).
C. elegans EIF-2α S49 is a highly conserved site and is equivalent to human eIF2α S51, whose phosphorylation is known to block global mRNA translation 37,38 . We thus asked whether mRNA translation is suppressed in the daf-2 mutant through hyper-phosphorylation of EIF-2α S49. We engineered an EIF-2α S49A mutation in the C. elegans genome using a CRISPR/Cas9 mediated gene editing method. Indeed, the S49A mutation, which locks EIF-2α in the dephosphorylation state, markedly increased the poly-ribosome fraction in the daf-2 mutant, albeit short of restoring it to the WT level (Fig. 4C). Further, the EIF-2α S49A mutation, which had no effect on WT lifespan, suppressed daf-2 longevity by 30% (Fig. 4D). These results suggest that enhanced phosphorylation of EIF-2α S49 in the daf-2 mutant may promote longevity by suppressing protein synthesis.
Next, we asked which kinase is responsible for hyper-phosphorylation of EIF-2α S49 in the daf-2 mutant.
Of note, two lines of evidence suggest that phospho-EIF-2α has a potent effect. First, a tiny amount of EIF-2α pS49, so low that it was undetectable by MS unless the phosphopeptides were enriched beforehand, is su cient to generate the protein synthesis and lifespan phenotype. The S49 containing peptide generated by trypsin digestion from endogenous EIF-2α was only detectable and quanti able by MS in the non-phosphorylated form in whole worm lysate samples (Supplementary Fig. 4C-E). Second, overexpression or knock-in mutation of the phospho-mimic EIF-2α S49D/E is lethal, suggesting a strong dominant effect of EIF-2α S49 phosphorylation. These target quantitation and genetics results both support that EIF-2α S49 phosphorylation has a potent inhibitory effect on protein synthesis and contributes substantially to daf-2 longevity (Fig. 4F).
Notably, our quantitative phosphoproteomics data also suggest that the observed EIF-2α pS49 increase of the daf-2 mutant (daf-2/WT = 1.83) may occur independently of daf-16: the EIF-2α pS49 increase was still observed upon deletion of daf-16 (daf-2; daf-16/ WT = 1.91) ( Supplementary Fig. 4F). Pursuing this, it was surprising when we found that among the 448 phosphoisoforms which were differentially regulated in the daf-2 mutant (Fig. 2B), 124 apparently require daf-16, while 123 do not (Supplementary Table 3). This apparently very-well-balanced distribution of daf-16 dependent vs. daf-16 independent phosphorylation changes seems quite unique; to our knowledge, most of the documented changes in daf-2(lf) worms are dependent on daf-16. For example, two thirds or more of the protein abundance changes seen in the daf-2 mutant were suppressed by daf-16(lf) 9 .
Beyond EIF-2α, we characterized another EIF protein C37C3.2 (C. elegans eIF5). iFPS did not rank EIF-5 pT376 and pS380 among the top 5% ( Supplementary Fig. 4G). The phosphorylation level of pS380 or pT376 pS380 either decreased or had no change, respectively, in the daf-2 mutant ( Supplementary  Fig. 4G). Simultaneous mutation of EIF-5 T376 and S380 to T375A S380A (2A) or T375E S380E (2E) by CRISPR/Cas9 had no or little effect on WT lifespan, and did not alter the lifespan of daf-2 (e1370 or RNAi) worms ( Supplementary Fig. 4H-I). These ndings indicate that, at least in the context of insulinsignaling-mediated lifespan extension, the two phosphosites of EIF-5 are not functionally impactful. At minimum, this result helps validate the utility of iFPS ranking as a hypothesis-generating tool to e ciently inform prioritization of candidates for functional studies.
Since the inactive form of CDK-1 (pT32 pY33) is not dominant negative, a reduced level of pT179 can be interpreted as a reduction of CDK-1 activity in the daf-2 mutant. Note that interpretation is supported by elaborate study of the daf-2 germline which reported a cell cycle delay in G2 in the proliferative zone; that is, proliferating daf-2 germ cells are slow to enter the M phase 41 . Importantly, all of our phosphoproteomics samples were synchronized to adult day one-a stage at which germ cells are the only dividing cells-so we can con dently assume that any detected CDK-1 activity must come from the germline.
We then asked whether the reduction of CDK-1 pT179 or CDK-1 activity in the daf-2 germline contributes to longevity. Mutating CDK-1 T179 to either A or E by gene editing was predictably unsuccessful: experimentally locking CDK-1 into either a completely inactive or a constitutively active state prevents cell cycle progression, causing lethality. We then took advantage of a temperature sensitive allele of cdk-1(ne2257ts) harboring an I173F mutation ve amino acids away from T179 in the activation loop. We found that shifting cdk-1(ne2257ts) worms from the permissive temperature of 15 °C to the restrictive temperature 22.5 °C on adult day one signi cantly extended WT lifespan (by 11-30%), and noted that this extension was daf-16 dependent (Fig. 5C). We also found that temperature-shift-induced inactivation of CDK-1(I173F) at earlier time points extended WT lifespan ( Supplementary Fig. 5A). Likewise, we observed an extended lifespan of 20-30% upon knockdown of cdk-1 starting from adult day one in the rrf-1(pk1417) mutant (in which RNAi is restricted in the germline, intestine, and some hypodermal cells 42 ), whereas no extended lifespan phenotype resulted from intestine-or hypodermis-restricted cdk-1 RNAi in these animals (Supplementary Table 4). These results support that reduced CDK-1 activity in the adult germline is su cient to promote a moderate lifespan extension.
Next, we investigated whether reduced CDK-1 pT179 in the adult germline is necessary for lifespan extension upon DAF-2 depletion. Using both gene editing and auxin-induced protein degradation (AID) technologies 43 , we were able to selectively degrade DAF-2 or WEE-1.3, or both, in the adult germline with high spatiotemporal precision. Degradation of WEE-1.3, the C. elegans ortholog of human WEE1/MYT1 40 , should eliminate inhibitory phosphorylation of CDK1 on T32 and Y33 to drive an elevation of CDK-1 activity. Indeed, degrading WEE-1.3 speci cally in the adult germline signi cantly shortened the lifespan of worms lacking germline DAF-2 (Fig. 5D). Moreover, both adult-speci c and germline-speci c degradation of DAF-2 slightly increased the mean lifespan and the maximal lifespan in two independent experiments, but not in a statistically signi cant manner. We thus conclude that reduced CDK-1 pT179 in the adult germline may confer a small contribution to daf-2 longevity.
It was highly striking that germline expression was predicted for the parent proteins of more than 70% of the iFPS-prioritized phosphosites ( Supplementary Fig. 5B). Further, it was conspicuous that proteins of the reproductive system were highly enriched among the hypo-phosphorylated proteins detected in the long-lived daf-2 mutant (Fig. 5E). These ndings motivated us to conduct a small-scale RNAi screen in the rrf-1(pk1417) mutant background to explore how germline phosphoproteins may affect ageing of the soma (Supplementary Fig. 5C). Interestingly, we found that adult onset RNAi of genes that promote mitosis or meiosis generally extended lifespan, whereas RNAi of genes that limit the genesis of germ cells or gametes shortened lifespan ( Fig. 5F and Supplementary Fig. 5C). These results are in line with reports of lifespan extension through germline ablation 44 , and echo with the antagonistic pleiotropy theory of ageing. They also suggest that, although reduced CDK-1 pT179 alone contributes marginally to daf-2 longevity, the phosphorylation changes among all germline proteins may collectively confer a sizable contribution to lifespan extension (Fig. 5G).

Reduction of Casein Kinase 2 (CK2) Activity Prolongs Lifespan
Based on the hyper-and hypo-phosphorylated sites we detected in the daf-2 mutant, and in light of kinase-substrate relationships predicted with the iGPS algorithm 45 , we explored which kinases are likely to be more or less active upon reduced insulin signaling. Speci cally, we used hypergeometric tests followed by Benjamini-Hochberg adjustment to assess whether the predicted or potential substrate sites of a given kinase were enriched among the differentially regulated phosphoisoforms of the daf-2 mutant. The hypo-phosphorylated sites displayed signi cant enrichment for putative substrate motifs (of 22 kinases), whereas no enrichment for kinase binding motifs was evident among the hyper-phosphorylated sites (Fig. 6A). There are studies for 5 of the 22 kinases reporting that RNAi or loss-of-function mutation extend lifespan (Fig. 6A), including investigations of C. elegans mTOR kinase LET-363 and the MAPK activated kinase MAK-2 and MNK-1.
Casein kinase 2 (aka CK2), was among the 22 kinases with a predicted activity decrease in the daf-2 mutant. In fact, there were two kinase binding motifs in CK2 which were signi cantly overrepresented among the hypo-phosphorylated sites found in the daf-2 mutant (Fig. 6B). Further implicating the likely impact of CK2 phospho-status in daf-2 longevity, an motif-x analysis 46 detected CK2 but none of the 21 other kinases from our initial iGPS analysis. The C. elegans CK2 holoenzyme is composed of KIN-3, the catalytic subunit, and KIN-10, the regulatory subunit. CK2 has been shown to slow down ageing in C. elegans 47 , but studies in yeast revealed an opposite effect, reporting that the Saccharomyces cerevisiae CK2 accelerates both chronological and replicative ageing 48,49 . We examined the lifespans of worms treated variously with kin-3 RNAi, kin-10 RNAi, or the CK2 inhibitor TBB. kin-3 or kin-10 knockdown during adulthood moderately but signi cantly extended WT lifespan in four independent trials (Fig. 6C and Supplementary Table 4). More strikingly, 24 and 48-hour TBB treatment (from adult day one) extended WT lifespan by 21-27% ( Fig. 6D and Supplementary Table 4). These results demonstrate that inhibition of CK2 in young adults promotes longevity in C. elegans.

Discussion
Reducing the activity of IIS signi cantly extends lifespan and mobilizes deeply conserved lifespan modulators, primarily through phosphorylation. However, the in-depth mechanisms of lifespan regulation by IIS-related phosphorylation have been largely neglected. Also, technical challenges for large-scale characterization of functional phosphorylation sites have hindered the gathering of experimentally con rmed phosphosites. In the present study, we conducted a large-scale quantitative phosphoproteomics survey to address these issues. Our results dramatically increase the total number of in vivo phosphorylation sites in C. elegans. Moreover, by developing a machine learning based prioritization tool for prioritizing phosphosites for functional con rmation with extensive phosphositespeci c mutagenesis experiments, we offer multiple demonstrations for how functional phosphorylation events modulate signaling pathways to control lifespan regulation.

Extensive Phosphorylation Regulation Orchestrated by IIS Kinases Controls Lifespan Regulation
Previous studies investigating how reduced IIS extends lifespan have focused on transcriptional regulation by FOXO. While it is clear that DAF-16/FOXO activation is required for the long lifespan of the daf-2 mutant, it is an open question whether DAF-16 activation is su cient for daf-2 longevity. Here, we looked into this issue by analyzing the phosphorylation changes against previously documented protein and mRNA abundance changes 6,33 . We found lifespan-affecting phosphorylation changes at the three phosphorylation we characterized in-depth: AKT-1 pT492, EIF-2α pS49, and CDK-1 pT179 (Fig. 7). In particular, increased EIF-2α pS49 contributes signi cantly to daf-2 longevity, and this regulation occurs speci cally at the PTM level, not at the mRNA or protein level. Besides, hyper-phosphorylation of EIF-2α at S49 persisted in the daf-16; daf-2 double mutant (Supplementary Fig. 4F and Supplementary Table 3), suggesting a daf-16 independent change. GO terms related to cell cycle and translation were signi cantly enriched from phosphosites regulated by daf-2 but independent of daf-16, whereas no enrichment was evident for daf-16 dependent phosphosites. Importantly, our present study as well as previous evidences con rm that retarding the cell cycle or mRNA translation results in lifespan extension, indicating daf-16independent phosphorylation events as IIS-related lifespan regulation mechanisms. Taken together, our phosphoproteomics and functional studies suggest that DAF-16 mediated transcriptional regulation alone may be insu cient for daf-2 longevity.
Functional analysis results from this study indicate that the phosphoproteins from the reproductive system do affect lifespan; indeed, there are often negative correlations between the reproduction-and lifespan-related phenotypes upon mutating these phosphoproteins. Consider that adult-speci c knockdown of genes including cdk-1, chk-2, hoe-1, hsr-9, and htp-3 promotes cell cycle progression and results in lifespan extension, whereas early death phenotypes result from knockdown of the germline hyperproliferation suppressor gld-1 or the oogenesis-restricting puf-3. These data endorse the antagonistic pleiotropy theory, which proposes that ageing is an adaptation to natural selection of pleiotropic genes that bene t tness in early life, but are detrimental later 50 . Assayed individually, these hypo-phosphorylated proteins of the reproductive system exhibit only small effects; however, we speculate that they may exert larger effects when combined. Notably, we did not detect clear patterns for mRNA-or protein-level changes for these phosphoproteins in the daf-2 mutant (Fig. 7).
Regarding kinases with activities predicted to be reduced in the daf-2 mutant, a survey of their lifespan phenotypes reported in the literature and our target analysis of the CK2 kinase KIN-3/KIN-10 suggests that, for the most part, their reduced activities contribute positively to daf-2 longevity, with CST-1 being the only exception among seven studied kinases. As no clear pattern of changes is evident for these kinases in the daf-2 mutant at the mRNA or protein-abundance levels (Fig. 7), our study underscores the utility of phosphoprotein-based surveys for elucidating the impacts of insulin signaling on longevity speci cally and for deepening biological understanding generally.

EIF-2α Phosphorylation Links IIS-Modulated Amino Acid Metabolism to Translation and Longevity
The daf-2(lf) mutation led to reduced levels of ribosomal proteins and poly-ribosome associated RNAs, suggesting a repression of global mRNA translation. Previous studies reported that that tts-1, a long noncoding RNA, was required for the reduction of ribosome-associated RNA in daf-2 worms 51 . We found here that GCN-2/EIF-2α phosphorylation signaling bridges translation and longevity in daf-2 worms ( Fig. 4F and 7). In eukaryotes, eIF2α phosphorylation converts eIF2-GDP into a competitive inhibitor of eIF2B, thus dominantly reducing pre-initiation complex assembly and general translation initiation 38 . Paradoxically, EIF-2α phosphorylation stimulates translation of atf-5, a transcription factor homologous to yeast GCN4 and mammalian ATF4 52 ; phospho-eIF2α-induced ATF4 expression is required for tumor cell survival and embryonic stem cell proliferation 53,54 . However, knocking out atf-5 had no effect on the lifespan of WT or daf-2 worms (Supplementary Table 4), indicating that the pro-longevity effect of EIF-2α phosphorylation mainly results from retarding general translation, rather than from promotion of ATF-5 translation. It remains to be clari ed whether phospho-EIF-2α targets speci c mRNA translation to promote lifespan. It is notable that EIF-2α S49A did not fully restore the peak heights of ribosomal fractions in daf-2 (Fig. 4C), suggesting additional forms of regulation, for example tts-1 associated ribosome reduction, is apparently involved.
The eIF2α kinases GCN2 and PERK are activated by amino acid depletion and ER stress, respectively. Our data suggest that GCN-2, not PEK-1, mediates the hyper-phosphorylation of EIF-2α S49 and promotes daf-2 longevity. This is consistent with previous reports that pek-1 null status neither abrogated ER stress resistance nor shortened the long lifespan of daf-2 worms 55 . The next question is how the daf-2 mutation activates GCN-2: several lines of evidence suggest that GCN-2 activity may respond to the IIS-modulated amino acid metabolism. First, uncharged tRNAs or ribosomal stalling directly stimulate GCN2 upon amino acid starvation 56,57 . Also, knockdown of worm tRNA synthetases induces EIF-2α S49 phosphorylation through GCN-2 52 . Additionally, reduction of daf-2 activity lowers the abundance of tRNA synthetases and amino acid pools in young adult worms 7,9,58 . It therefore seems plausible that a shortage of tRNA synthetases and amino acids could lead to loss of tRNA charging and/or ribosome pausing, therefore stimulating GCN-2, a scenario that would support a role for GCN-2/EIF-2α signaling in altering amino acid metabolism to ameliorate translation and longevity in daf-2 animals.

Germline Phosphoproteins Mediate the Effects of IIS on Reproduction and Lifespan Regulation
daf-2 and daf-16 mainly function in neurons, the hypodermis, and intestine to regulate lifespan 26,59,60 . On the one hand, ablation of germline precursor cells further extends the lifespan of daf-2 worms, suggesting that IIS acts in parallel with germline signaling to regulate lifespan 44 . On the other hand, IIS is required for germline cell cycle progression to ensure robust germline proliferation 41 . Recent data are starting to indicate a link (probably orchestrated by IIS-regulated SUMOylation) between lifespan regulation by germline signaling and IIS 61 .
Here, we found that IIS actively induces phosphorylation on hundreds of germline proteins involved in the cell cycle, apoptosis, translation, etc., indicating pronounced functional impacts from phosphomodulation of germline proteins. As an example, we show that hypo-phosphorylated CDK-1 T179, which inactivates CDK-1 and thereby delays the cell cycle, potentially thereby contributing to daf-2 longevity by transmitting a signal representing reduced DAF-2 pathway activity from the germline to the soma. Furthermore, results from our initial RNAi screen suggest dual roles of germline phosphoproteins in mediating the effects of IIS on reproduction and lifespan regulation. Notice that germline phosphoproteins are not necessarily germline-speci c proteins. Techniques for tissue-speci c mutagenesis of phosphosites are currently unavailable. Still, as a start, targeted protein degradation technologies, such as the AID method 43 used in the present study, are clearly helpful for testing hypotheses about tissue-speci c signal transduction of IIS.
iFPS Together with Quantitative Phosphoproteomics as a General Approach to Study Functional

Phosphorylation in vivo
We thoroughly mapped and quanti ed the IIS-regulated phosphorylation changes in C. elegans. Our phosphoproteomic work ow yielded the largest-to-date phosphoproteome dataset, almost doubling the size of previously available C. elegans database entries. Further, we designed and implemented a machine learning based tool -iFPS -for systematically ranking the likely functional importance of worm phosphosites. Such a tool was not previously available for the C. elegans research community. iFPS scores, together with the in vivo phosphosites identi ed here, constitute a major resource for studies of phosphorylation and insulin signal transduction.
iFPS scored 15,266 phosphosites in total. 50% of the true positive phosphosites were successfully recalled from the top 5% iFPS ranking, which is a ten-fold increase the propensity to hit a functional phosphosites from the overall background. Therefore, we considered a top 5% ranking as an initial cutoff for prioritizing the phosphosites regulated by daf-2. This prioritization strategy successfully guided our follow-up experiments, which successfully demonstrated multiple mechanisms of lifespan regulation by IIS at the level of phosphorylation. Notably, we also tested two phosphosites that were among top 10-13%. Our ndings that they were not obviously impactful for regulating the lifespan of WT or daf-2 worms offers another form of validation for the utility of iFPS-based prioritization. Conventionally, these sites would almost certainly have been selected as candidates for functional studies. First, they are from translation initiation factor EIF-5 protein, which is in line with the hypothesis that translational repression contributes to lifespan extension. Second, they are close neighbors, and one of them is regulated by daf-2. It is thus clear that our new resources can help overcome the frequently encountered struggle with misleading "negative results" in attempts to validate ndings from phosphoproteomic analyses.
Together with powerful prioritization tools and sophisticated validation experiments, phosphoproteomescale functional elucidation of phosphosites sets a new benchmark for studies of longevity. The current iFPS scoring system is an encouraging start. The prediction power has not been fully explored owing to a paucity of experimentally con rmed functional phosphosites, kinase-substrate relations, protein-protein interactions, and other PTMs in C. elegans. Thus, additional experimental data, further computational resources, and innovative research strategies will still be needed to better comprehend which data signals actually convey informative hints to support identi cation of functional phosphosites in the future.
Nevertheless, our study illustrates how our present strategy can be generally applied for a potentially wide range of other studies examining complicated biological phenomena to inform both basic mechanistic research and rational drug design.

C. elegans and E. coli strains
The genotypes, sources, and generating methods of C. elegans strains are listed in Supplementary  Table 6. Worms were maintained on Nematode Growth Media (NGM) agar plates seeded with E.coli strain OP50 at 20 °C using standard protocols 62 , unless otherwise indicated.
The 15 N-labeled food source was prepared by growing E. coli MG1655 in M9 minimal media ( 15 NH 4 Cl as the unique nitrogen source) until OD600 value reached to 1.0 at 37 °C 63 . MG1655 were concentrated and seeded on nitrogen-free worm plates. E. coli HT115 transferred with RNAi plasmids or the empty vector control were cultured overnight in LB plus ampicillin (100 µg/ml) and tetracycline (10 µg/ml) and then seed on NGM plates containing ampicillin (100 µg/ml), tetracycline (10 µg/ml), and IPTG (1 mM). RNAi clones made in this paper were constructed by inserting the cDNA of genes into the L4440 vector. Other RNAi bacteria were derived from Ahringer RNAi library or RCE1181 C. elegans RNAi Feeding library.

Mass spectrometry data acquisition
Samples was analyzed twice though a Q-Exactive mass spectrometer (Thermo Fisher Scienti c) interfaced with an Easy-nLC1000 reversed-phase chromatography system (Thermo Fisher Scienti c). 5 µl sample was loaded on a 75 µm × 4 cm trap column packed with 10 µm, 120 Å ODS-AQ C18 resin (YMC Co.) and separated by a 75 µm × 10 cm analytical column packed with 1.8 µm, 120 Å UHPLC XB-C18 resin (Welch Materials) at a ow rate of 200 nl/min using a linear gradient of 0-28% ACN (0.1% formic acid) over 80 min, followed by raising the ACN concentration to 80% within 15 min and maintaining for another 15 min.
The FTMS full scan between 350-2000 m/z were acquired from the Orbitrap at 70,000 resolution in pro le data type with 1e6 AGC target, 60 ms maximum injection time. Ion 445.12003 was used for internal calibration. Top ten most abundant precursor ions were selected using a 2.0 m/z isolation window for HCD fragmentation (27% collision energy). MS2 scans were acquired from the Orbitrap at 17,500 resolution in pro le data type with 5e4 AGC target, 250 ms maximum injection time. The intensity threshold for MS2 scan was 4e3. Precursors with + 1, > +6, and unassigned charge state were excluded.
Peptide match was set as preferred. Dynamic exclusion was 60 s.

Phosphopeptide identi cation and phosphosite localization
Raw mass spectrometry les were searched against a composite target/decoy database using ProLuCID 64 . The C. elegans protein database (WS233) was used as target while the corresponding reversed sequences was generated as decoy. Spectra were searched with ± 50 ppm for both precursor ion and fragment ion accuracy, peptide length above 7 residues, fully tryptic restriction.
Carbamidomethylation of cysteines was included as a xed modi cation. Phosphorylation of serine, threonine and tyrosine residues were included as variable modi cations. The peptide spectrum matches were ltered using DTASelect2 65 . The estimated false discovery rate (FDR) was no more than 1.07% for phosphopeptides. The 14 N and 15 N-labeled peptides were identi ed in paralleled pipelines.
For each 14 N peptide, phosphosites with phosphoRS site probability 23 above 0.75 were assigned as con dent modi cation. The phosphosite localization on the 15 N-labeled peptide was corrected corresponding to its 14 N-isotopic version. The residual number of phosphosites was determined by mapping the phosphopeptides to the longest transcripts of C. elegans genes (UniProt 201501). Phosphosites from single-phosphorylated peptides and multi-phosphorylated peptides were extracted as different phosphoisoforms (see examples in Supplementary Fig. 1C).

Phosphoisoform quanti cation
Ratios of 14 N to 15 N-labeled phosphopeptide were determined by a modi ed version of pQuant software 66 . In brief, con dent quanti cation was accepted when both the least interfered isotopic ratio and the monoisotopic ratio of a 14 N and 15 N ion pair had the σ values below 0.5. The median of least interfered ratios was assigned as the 14 N/ 15 N ratios of quanti ed peptides. Ratios were normalized to the median value of all quanti ed peptides per technical replicates, and then assigned to their corresponding phosphoisoforms.

Phosphorylation changes in IIS mutants
To determine the daf-2 regulated phosphorylation, phosphoisoforms, which were quanti ed at least three times in both daf-2 and WT, were subjected to statistical comparison. We noticed that 1125 phosphoisoforms, which were reliably measured more than twice in daf-2 samples were only quanti ed once or twice in WT samples. In this scenario, if the quantitation values in daf-16 and/or daf-16; daf-2 samples were similar to those in WT, we manually imputed the values quanti ed in the WT-like short-lived controls to the WT samples. Log2 (median of daf-2/ median of control) distribution was plotted to estimate the median and s.iqr values. The 14 N/ 15 N ratios of each phosphoisoforms were subjected to Wilcoxon rank-sum test. A "strict" lter was applied with log2 (daf-2/control) fold change above a factor 1.5 (s.iqr, in either direction) and p < 0.05 from the Wilcoxon rank-sum test. A "loose" subset of daf-2 regulated phosphoisoforms met the criteria of either (log2 (daf-2/control) changed above a factor 1.5 while p ≤ 0.1) or (log2 (daf-2/WT) changed at least a factor 1 while p < 0.05).
Prioritization of the potentially important phosphosites iFPS (Inference of Functional Phosphorylation Sites) algorithm was developed to systematically rank the functional importance of C. elegans phosphosites. Constraints that may indicate the biological impacts of phosphosites were selected based on a survey of literature. The following six constraints were estimated on their potential to discriminate functional phosphosites in C. elegans, and were further integrated into one score by a machine learning method.

Kinase families
Functional phosphosites are often regulated by kinases. A previous developed software package iGPS 1.0 45 was used to predict kinases that recognize individual phosphosite. For a better coverage, low thresholds with false positive rates (FDRs) of 10% for serine/threonine kinases and 15% for tyrosine kinases were adopted. Potentially false-positive hits were ltered by using protein-protein interaction information integrated in iGPS. Numbers of kinase families targeting each phosphosite were evaluated and selected as a predictor of functional phosphorylation.

Conservation
Comparing to sequence conservation, phosphorylation conservation is a better predictor of functional phosphosites. Phosphorylation conservation was indicated by RCS value as described below.
Sequences of the C. elegans phosphoproteins were multi-aligned by MUSCLE 67

Crosstalk
Adjacent PTMs may co-regulate a biological event, for example histone acetylation and phosphorylation are synergistic in remodeling chromatin structure of target genes. GPS-PAIL 2.0, a software developed to predict the histone acetyltransferases (HATs)-speci c sites 74 was applied to predict acetylation sites close to phosphosites. The C.elegans orthologs of ve HATs (EP300, HAT1, KAT2B, KAT5 and KAT8) were considered as potential regulators. For a better coverage, low thresholds with speci city (Sp) values of 85% was adopted. The phosphosite with at least one acetylation site located within 15 amino acids was assigned with a crosstalk value of 1, otherwise the value was 0.

Relative surface accessibility (RSA) and secondary structure
The accessibility and structure environment around phosphosites affect their chance to be targeted by speci c kinases. Here, the surface accessibility and secondary structure of phosphosites were predicted by the webserver of NetSurfP v1.1 (http://www.cbs.dtu.dk/services/NetSurfP/) 75 . The RSA scores and the probabilities for Alpha-Helix, Beta-strand and Coil of each phosphosite were used for model training and prediction.
For model training, the manually compiled 121 known functional phosphosites of C. elegans were used as positive data set (Supplementary Table 2). Other C.elegans phosphosites in dbPAF database were selected as negative samples. The multinomial logistic regression model with a ridge estimator in Java package Weka 3.8 76 was adopted for training. Due to the much larger size of negative samples, the performance and robustness of prediction system were evaluated with different ratios (1:1, 1:2, 1:5 and 1:10) of positive and negative samples. Ratio of 1:5 won best. Therefore, 10 sets of 5 times negative samples were randomly selected for the training. The model with the best performance was chosen as the nal model. 10-fold cross validate model was used to avoid over tting during training.

Target quanti cation of phosphorylation and proteins
Endogenous levels of target phosphorylation and protein were quanti ed by LC-MS/MS analysis of the worm proteome using isotopically labeled peptides as a spike-in standard. Total proteins were extracted from adult day one worms by cryogenic grinding (mixer mill MM 400, Retsch) and resolved in lysis buffer (0.1 M Tris/HCl, pH 7.6, 4% SDS, 0.1 M DTT, protease inhibitor cocktail, and phosphatase inhibitor cocktail). The crude extract was incubated at 95 °C for 5 min followed by centrifugation at 14,000 rpm (10 min, room temperature). Supernatant was collected and sent for protein concentration measurement by 2-D quant. 100 µg of total proteins were subjected to buffer exchange, alkylation reaction and trypsin digestion by the lter-aided sample preparation (FASP) method 77 . The synthesized isotopically labeled peptides were simultaneously spiked in right before adding trypsin (see Supplementary Table 5).
Peptides were mobilized in the positive-ion mode by electrospray ionization with 2 kV spray voltage and 320 °C capillary temperature. Ion 445.120025 m/z was used for internal calibration. Full-scan mass spectra were acquired in the Orbitrap over the m/z range of 300 to 1500 at a resolution of 70,000. AGC target was set to 3e6. Maximum injection time was 60 ms. Precursor ions of target peptides were selected for Higher-energy Collisional Dissociation (HCD) fragmentation and Orbitrap detection during desired acquisition time. The operating parameters were: resolution 35,000; AGC targets 1e5; maximum IT auto; isolation window 2 m/z; normalized collision energy 27.
MS data were processed in Xcalibur (version 2.2 SP1.48, Thermo Fisher Scienti c) and pLabel (version 2.4) 78 . Isotopically labeled peptide ions were used to locate the endogenous targets across the elution. For each target precursor ion, at least three fragment ions with high abundance and low interference were selected for identi cation and quantitation. Peaks areas of precursor ions generating each desired fragment ion were determined in Xcalibur with default parameter and used for quanti cation.
Endogenous peptides (light ions) and their isotopically labeled counterparts (heavy ions) were quanti ed by extracting peak areas of each quanti able transition (precursor → fragment). For each transition, peak areas of light were divided by that of heavy. Relative abundance of individual phosphorylation or protein was determined by the mean value of light/heavy ratios. Two-tailed p value comparing quantitation in IIS mutants to WT was calculated by Student's t test.

CRISPR/Cas9 based mutagenesis
The CRISPR/Cas9-mediated mutagenesis of C. elegans endogenous genes was performed as described with little modi cation 79,80,81 . For the sgRNA-Cas9 expression plasmid-based mutagenesis, individual sgRNA was incorporated into the pDD162 plasmid (Addgene, #47549) at desired locus. Usually two different sgRNAs were used simultaneously. For the Cas9 ribonucleoprotein-based mutagenesis, a crRNA that contains the target sequence at the 5′ end was synthesized. Two or more alleles relating to each target were assayed in follow-up studies.

Western blot analysis
Synchronized worms of each strain were grown on OP50 plates at 20 °C and harvested at adult day one using M9 buffer, followed by liquid nitration freezing. Worm pellets were boiled in SDS loading buffer and loaded to replicate SDS-PAGE gels. The transferred uorescence PVDF membranes (Millipore) were probed overnight at 4 °C with anti-phospho-eIF2α (Ser51) and anti-eIF2α (kindly provided by Dr. Shin Takagi) primary antibody, respectively. The blots were visualized by one-hour incubation at room temperature with IRDye 800CW uorescent secondary antibodies (Odyssey), followed by scanning in the LI-COR Odyssey Infrared Imaging System according to the manufacturer's instruction. Images were quanti ed with Image Studio Lite Ver 4.0 (LI-COR). The signi cance of intensity difference was evaluated by paired t test. To examine the AKT-1::GFP level, nitrocellulose membranes were probed with anti-GFP or anti-tubulin antibodies. HRP conjugated secondary antibodies were used for detection.

Lifespan assays
Lifespan assays were modi ed based on previous description 15 . Worms were synchronized via collecting eggs laid within 4 hours by adult day two hermaphrodites. Worms were transferred to desired plates (25 to 35 worms per plate) when they reached adulthood, and continually transferred to fresh plates every other day. After they ceased laying egg, living worms were scored every two days and transferred to fresh plates every four to seven days. Statistical analysis of lifespan data was performed in SPSS software. Replicates as well as measuring conditions including temperature and supplements in plates are recorded in Supplementary Table 4.
Auxin treatment was performed as previous description 43 . Brie y, auxin, which was dissolved in ethanol was added to NGM agar before pouring plates. The nal concentration of auxin and ethanol per plate was 1 mM and 0.25%, respectively. 0.25% ethanol was used as control.
For TBB treatment, TBB was resolved in 80% DMSO: PBS solvent. NGM plates containing 50 ng/µl FUDR were seeded with OP50 and dried at room temperature for 12 hours. Then TBB solution was added to the surface of plates with the nal concentration of 15 uM or 45 uM TBB. The nal concentration of DMSO for each plate was adjusted to 0.36%, including the control. Supplied volume is based on the volume of media. The freshly prepared plates were left for drug diffusion overnight and used within two days. Synchronized adult day one WT were transferred to plates with TBB or control treatment at 20 °C. After 24 or 48 hours, worms were moved to fresh NGM plates (50 ng/µl FUDR).

Dauer formation
Parent worms were maintained on NGM plates seeded with OP50 at 15 °C for over three generations and allowed to lay eggs for 4-6 hours at 21 °C. Progeny were incubated at 21 °C. Dauer and non-dauer animals were scored and con rmed at the third, fourth, or fth day post egg-laying.
Polyribosome pro ling assays Polyribosome pro ling was performed as previous description with little modi cation 9 . 10 ml 7-50% (w/v) linear sucrose gradients in gradient buffer (110 mM KAc, 20 mM MgAc2 and 10 mM HEPES pH 7.6) were prepared in 13 ml polyallomer centrifuge tubes (Beckman-Coulter) just before use. Adult day one synchronized worms were lysed in lysis buffer (30 mM HEPES pH 7.6, 100 mM KCl, 10 mM MgCl 2 , 0.1% NP-40, 100 mg/ml cycloheximide, 2 mM DTT, 40 U/ml RNase inhibitor, and protease inhibitor cocktail) with a dounce homogenizer. Worm lysate was centrifuged at 14,000 g (10 min, 4 °C). The supernatant was immediately subject to protein content estimation by A 280 nm on NanoDrop 1000 (Thermo Fisher Scienti c). The same amount of A 280 nm units (3000-6100 units) of each sample was layered atop the 7-50% (w/v) linear sucrose gradient and centrifuged for two hours at 40,000 rpm in a SW41Ti rotor (Beckman-Coulter) at 4 °C. Gradients were analyzed with a density gradient fractionator coupled with the absorbance recording at an optical density of 254 nm (Teledyne Isco). Images were transferred into digital data using Adobe Photoshop and merged by aligning base line of absorbance using Adobe Illustrator.

Microscopy
Worms were cultured at 20 °C. To measure the cellular localization of DAF-16::GFP, L4 or young adult worms were picked on pad and visualized under uorescent microscopy within two minutes. DAF-16::GFP localization in intestinal cells was manually classi ed (Fig. 3C). The number of worms in each category was counted. Images were taken using a Zeiss Axio Imager M1 microscope at 400-fold magni cation. Images of worms expressing AKT-1::GFP or AKT-T492A::GFP were taken by the Zeiss Axio Imager M1 microscope at 100-fold or 200-fold magni cation. The penetrance of AKT-1-T492A::GFP nuclear localization in proximal gonad was calculated by scoring the number of worms under the same microscope at 1000-fold magni cation. Fluorescence images of AKT-1::GFP or AKT-T492A::GFP in the germline were acquired using a Spinning Disk microscope.

Tissue expression prediction
Proteins with decreased or increased phosphorylation on the daf-2 regulated phosphoisoform were de ned as hypo-or hyper-phosphorylated proteins respectively. 180 hypo-and 169 hyper-phosphorylated proteins were mapped, 8 of which were shared by both. Expression of the daf-2 regulated phosphoproteins across 76 tissues and cell types were implemented on an interactive webserver (http://worm.princeton.edu) 82 . The prediction scores were downloaded for statistical enrichment calculation using R software (v.3.5.0). The two-tailed p value per sub-tissue was calculated by Z-test.

Kinase substrate analysis
The iGPS-predicted kinase-substrate relations were adopted for kinase substrate analysis.
Hypergeometric test followed by Benjamini-Hochberg adjustment was used to determine whether targets of particular kinases were enriched in the daf-2 hyper-or hypo-phosphorylated dataset by setting all C. elegans phosphosites as background. Alternatively, Motif-X (http://motif-x.med.harvard.edu/motif-x.html) was performed to identify the signi cantly overrepresented motifs, with the default parameters (width = 13, occurrences = 20, and signi cance < 1E-6) and the total quanti ed phosphopeptides as background.

Data And Code Availability
The MS raw data for phosphoproteomics and target quantitation in this study are deposited to the ProteomeXchange Consortium via the iProX partner repository 83 with the dataset identi er PXD020440 (https://www.iprox.org/page/PSV023.html;?url=1595231837574FhIp, password "yXtx"). Code and functions used to generate iFPS are available online (https://github.com/CuckooWang/iFPS).

Competing Interests
The authors declare no competing interests.