RNA helicase HEL-1 promotes longevity by specifically activating DAF-16/FOXO transcription factor signaling in Caenorhabditis elegans

Significance

RNA helicases are a large family of enzymes that regulate the generation and maintenance of RNA. However, the physiologic roles of RNA helicases in animal aging remained unknown. Here we show that an RNA helicase, helicase 1 (HEL-1), extends the lifespan of the roundworm Caenorhabditis elegans by up-regulating the longevity transcription factor forkhead box O (FOXO). Our finding suggests that an RNA helicase can have rather specific roles in animal longevity. A number of studies show that variants of FOXO are linked to human aging and longevity. In addition, the mammalian HEL-1 homolog has been implicated in cellular aging. Thus, our work may have direct implications in mammalian aging, and the human HEL-1 homolog may work with FOXO to increase lifespan.

Abstract

The homeostatic maintenance of the genomic DNA is crucial for regulating aging processes. However, the role of RNA homeostasis in aging processes remains unknown. RNA helicases are a large family of enzymes that regulate the biogenesis and homeostasis of RNA. However, the functional significance of RNA helicases in aging has not been explored. Here, we report that a large fraction of RNA helicases regulate the lifespan of Caenorhabditis elegans. In particular, we show that a DEAD-box RNA helicase, helicase 1 (HEL-1), promotes longevity by specifically activating the DAF-16/forkhead box O (FOXO) transcription factor signaling pathway. We find that HEL-1 is required for the longevity conferred by reduced insulin/insulin-like growth factor 1 (IGF-1) signaling (IIS) and is sufficient for extending lifespan. We further show that the expression of HEL-1 in the intestine and neurons contributes to longevity. HEL-1 enhances the induction of a large fraction of DAF-16 target genes. Thus, the RNA helicase HEL-1 appears to promote longevity in response to decreased IIS as a transcription coregulator of DAF-16. Because HEL-1 and IIS are evolutionarily well conserved, a similar mechanism for longevity regulation via an RNA helicase-dependent regulation of FOXO signaling may operate in mammals, including humans.

Various genetic and environmental factors, including insulin/insulin-like growth factor 1 (IGF-1) signaling (IIS), target of rapamycin signaling, dietary restriction, mitochondrial respiration, and reproductive systems, influence aging across phyla (reviewed in refs. 1–3). IIS is one of the most evolutionarily conserved pathways involved in the regulation of aging. Mutations in Caenorhabditis elegans daf-2, which encodes an insulin/IGF-1 receptor homolog, double the lifespan of C. elegans (4). Inhibition of DAF-2 reduces phosphatidylinositide 3 (PI3)-kinase signaling, which up-regulates several transcription factors, including DAF-16/forkhead box O (FOXO), heat shock factor-1 (HSF-1), and SKN-1/NRF2 (reviewed in refs. 1, 3, and 5). DAF-16 is one of the best characterized of these longevity factors; reduced PI3-kinase signaling leads to the dephosphorylation and nuclear translocation of DAF-16, which up-regulates the expression of various target genes. DAF-16 target genes contribute to longevity by promoting stress resistance, reproductive span, innate immunity, and protein homeostasis.

RNA helicases are essential for biogenesis, maturation, processing, and homeostasis of various types of RNAs (reviewed in ref. 6). In addition, RNA helicases work with factors, such as a CREB-binding protein, RNA polymerases I and II, and histone deacetylases, to regulate transcriptional activity (7, 8). For example, DDX5 (p68), a DEAD-box RNA helicase, interacts with Smad3, a transcriptional activator and intracellular effector of TGF-β (9), and with RNA polymerase II to modulate transcriptional activity (10). Because RNA helicases comprise a large family of housekeeping proteins essential for RNA biogenesis and homeostasis, their importance in aging and lifespan regulation is well expected. However, RNA helicases that regulate longevity remain largely unknown.

In this report we identified RNA helicases that influenced the lifespan of C. elegans by performing a large-scale RNAi screen. We found that several RNA helicases modulated lifespan via the IIS pathway. In particular, we showed that a DEAD-box RNA helicase, helicase 1 (HEL-1), promoted longevity by up-regulating DAF-16 as a transcription coregulator. Given the evolutionarily conserved nature of RNA helicases, our study raises the possibility that RNA helicases influence aging in other species, including mammals.

Results

Knockdown of Several RNA Helicases Influences Lifespan Through the IIS Pathway.

RNA helicases are a large family of evolutionarily conserved enzymes essential for RNA biology. Because the identities of RNA helicases are not well defined in many species, including C. elegans, we first established 96 genes that encode proteins containing RNA helicase domains (SI Appendix, Fig. S1A and Table S1). To identify RNA helicases that are crucial for the regulation of aging, we measured the lifespan of animals treated with commercially available RNAi clones (11, 12) targeting each of 78 RNA helicases (SI Appendix, Fig. S1B). The IIS pathway is one of the most evolutionarily conserved pathways involved in the regulation of aging. Thus, we used daf-2(e1370) insulin/IGF-1 receptor mutants [daf-2(-)] and wild-type worms for our screen. We also hypothesized that the effects of RNAi on lifespan would be more pronounced in the long-lived daf-2(-) mutants than in wild-type worms because of the larger window for lifespan changes. To eliminate the effects of RNAi on development and to focus on changes in adult lifespan, we treated animals with RNAi only during adulthood.

We noticed that the knockdown of more than 30 RNA helicase genes significantly decreased lifespan, indicating a general role for RNA helicases in the maintenance of normal lifespan (Fig. 1 A and C and SI Appendix, Table S2). We used an arbitrary 10% differential effect of RNAi on the lifespan of daf-2 mutants versus wild-type animals as the cutoff value for specificity. We found that RNAi targeting 11 RNA helicases had rather specific effects on the lifespan of daf-2 mutants or wild-type animals. RNAi targeting each of eight RNA helicases [hel-1, suppressor with morphological effect on genitalia 2 (smg-2), ZK512.2, R03H10.6, replication protein A homolog 1 (rpa-1), masculinization of germline 5 (mog-5), synthetic mutator 1 (smut-1), and suppressor of acy-4 sterility 1 (sacy-1)] decreased lifespan in daf-2(-) mutants to a greater degree than in wild-type worms (Fig. 1C, Table 1, and SI Appendix, Fig. S2 A–H and Table S2). This effect was similar to the effects of RNAi targeting daf-16/FOXO, a transcription factor required for the longevity of daf-2(-) (Fig. 1 A–C). Knocking down F53H1.1 or conserved germline helicase (cgh-1) significantly decreased the lifespan of wild-type but not of daf-2(-) animals (Table 1 and SI Appendix, Fig. S2 I and J and Table S2). RNAi targeting ZK686.1 significantly increased the lifespan of wild-type but not of daf-2(-) animals (Table 1 and SI Appendix, Fig. S2K and Table S2). Thus, a targeted RNAi screen allowed us to identify many RNA helicases that may function as regulators of lifespan.

Fig. 1.

Identification of RNA helicases whose knockdown influences lifespan in C. elegans. (A and B) Volcano plots that show percent changes in lifespan and P values upon knockdown of RNA helicases in wild-type (A) and daf-2(e1370) [daf-2(-)] (B). Black circles indicate the effects of candidate RNA helicase RNAi clones on the percent of mean lifespan changes in wild-type (A) and daf-2(-) (B) animals. Red triangles indicate the effects of daf-16 RNAi that was used as a positive control. All circles and triangles in A and B represent the effects of individual lifespan assay experiments. (C) Average percent changes in the mean lifespan of wild-type and daf-2(-) animals treated with each of 78 RNAi clones that target RNA helicases. Black circles indicate RNAi clones that indiscriminately affected lifespan. Green circles indicate RNAi clones that rather specifically decreased the lifespan of daf-2(-) mutants. Magenta circles show RNAi clones that decreased lifespan less in wild-type than in daf-2(-) animals. The orange circle indicates an RNAi clone that increased lifespan in wild-type but not in daf-2(-) animals. The red triangle indicates the effects of daf-16 RNAi, which was used as a positive control. Circles/triangle and error bars indicate averages values and standard error of mean (SEM) of two independent lifespan experiments for RNAi clones targeting individual RNA helicases, except for daf-16 RNAi [wild-type: n (number of repeats) = 14; daf-2(-): n = 18], and ZK512.2 and Y54G11A.3 RNAi [daf-2(-): n = 1] because of bacterial contamination. See SI Appendix, Fig. S2 for individual lifespan curves and SI Appendix, Table S2 for values and statistical analysis.

Table 1.

RNA helicase genes whose knockdown specifically influenced the lifespan of wild-type or daf-2 mutant animals

Gene targeted by RNAi	Description	% lifespan change
		Wild-type	daf-2(-)
daf-16#	FOXO transcription factor	−22.7 ± 8.4**	−46.9 ± 0.0**
hel-1	DEAD-box helicase	−9.3 ± 2.3**	−26.9 ± 3.6**
mog-5	Masculinization of germ line	−16.1 ± 4.0**	−32.8 ± 3.3**
rpa-1	Replication protein A homolog	−2.5 ± 1.4	−19.4 ± 0.2**
R03H10.6	Homology to replication protein A (RPA)	1.7 ± 1.6	−20.2 ± 3.6*
sacy-1	Probable ATP-dependent RNA helicase DDX41	−13.3 ± 1.3**	−24.7 ± 0.0**
smg-2	Suppressor with morphological effect on genitalia	11.8 ± 4.2**	−23.1 ± 4.5**
smut-1	Homology to probable ATP-dependent RNA helicase DDX4 isoform 4	−30.1 ± 9.8**	−43.2 ± 3.8**
ZK512.2	Homology to ATP-dependent RNA helicase DDX55	−0.4 ± 8.8*	−22.0*
F53H1.1	Homology to DEAD (Asp-Glu-Ala-Asp) box polypeptide 46, isoform CRA_a	−21.1 ± 4.6**	3.4 ± 2.9
cgh-1	Conserved germline helicase	−31.9 ± 7.4**	−0.4 ± 3.0
ZK686.1	DEAD/DEAH-box helicase	25.2 ± 4.0**	4.1 ± 2.7

^*P < 0.05; **P < 0.001.

^#Positive control.

RNA Helicase HEL-1 Influences Longevity and the Reproductive Span of daf-2 Mutants.

Among the candidate RNA helicases, hel-1 RNAi had a robust and specific effect on the longevity of daf-2 mutants: −26.9% on the lifespan of daf-2(-) versus −9.3% on the lifespan of wild-type worms (Fig. 1C, Table 1, and SI Appendix, Fig. S2A). Furthermore, HEL-1, a DEAD-box helicase homologous to mammalian DDX39A/UAP56/BAT1, is evolutionarily conserved across phyla (SI Appendix, Fig. S3 A and B). We confirmed the RNAi screening results for hel-1 through different genetic methods. Knockdown of hel-1 specifically decreased daf-2(-)–mediated longevity in an RNAi-hypersensitive rrf-3(pk1426) mutant background (Fig. 2C). We also found that hel-1(gk148684) [hel-1(-)] mutations, which caused a conserved glutamate-to-lysine (E155K) change in the DEAD-box domain (Fig. 2 A and B, and SI Appendix, Fig. S3 A and C), specifically reduced the long lifespan of daf-2(-) mutants (Fig. 2D and SI Appendix, Fig. S4A) without affecting development (SI Appendix, Fig. S4C). In addition, the longevity-suppressing effects of hel-1 RNAi and the mutations were reproduced in the absence of 5-fluoro-2′-deoxyuridine (FUdR) (Fig. 2D and SI Appendix, Fig. S4B); FUdR is used for the standard adult lifespan assay in C. elegans to prevent progeny from hatching. These results established that hel-1 is required for the long lifespan conferred by daf-2 mutations.

Fig. 2.

hel-1 is required for the long lifespan and the reproductive span of daf-2 mutants. (A) A schematic diagram of HEL-1 showing functional domains and the amino acid change caused by hel-1(gk148684) [hel-1(-)] mutation. DEAD-box and helicase domain-encoding regions in the hel-1 gene are indicated. The hel-1(-) mutants contain a point mutation that results in an E155K change in the DEAD-box domain. (B) Modeling of C. elegans HEL-1 protein based on the crystal structure of a human HEL-1 homolog DDX39A (78). See SI Appendix, Fig. S3C for the predicted structure of whole HEL-1. Red and blue indicate negative and positive charges, respectively. The 155th amino acid, glutamate in wild-type and lysine in hel-1(-), is shown in the dotted circle. (C) In an RNAi-hypersensitive rrf-3(pk1426) [rrf-3(-)] mutant background, hel-1 RNAi specifically decreased the lifespan of daf-2(e1370) [daf-2(-)] mutants but not that of control animals. (D) hel-1(-) mutations partly suppressed the longevity of daf-2(-) mutants without FUdR treatment. See SI Appendix, Fig. S4 A and B for other data showing the requirement of hel-1 for the long lifespan of daf-2 mutants. (E) The increased reproductive span in daf-2 mutants was partly suppressed by hel-1 mutations. (F) hel-1 mutations had no effect on the thermotolerance of daf-2(e1370) or wild-type animals at 35 °C. The survival experiments shown in this figure were repeated twice independently. See SI Appendix, Table S3 for additional repeats and statistical analysis of survival data other than lifespan shown in this and other figures. (G and H) Dauer formation was not influenced by hel-1(-) mutations in daf-2(e1370) (G) or daf-2(e1368) (H) mutant background at 25 °C. daf-16(mu86); daf-2(-) double mutants were used as a positive control for dauer suppressor. Please note that hel-1 RNAi caused a developmental arrest phenotype (SI Appendix, Fig. S4D); therefore we used hel-1 mutants for dauer assays (n = 4).

We then examined whether HEL-1 contributed to phenotypes other than long lifespan in the daf-2 mutants. The daf-2 mutations extend reproductive span (13, 14), which is defined as the time during which progeny are produced in the whole lifetime of the organism. The hel-1 mutations decreased the reproductive span of daf-2 mutants but not that of wild-type animals (Fig. 2E). The daf-2(-) mutants also display increased resistance against diverse stresses, including heat stress, pathogenic bacteria, and oxidative stress (reviewed in refs. 1 and 3). Resistance to heat stress was not affected by hel-1 mutations or RNAi (Fig. 2F and SI Appendix, Fig. S5A). The effects of hel-1 RNAi and hel-1 mutations on resistance to pathogenic bacteria (Pseudomonas aeruginosa, PA14) and oxidative stress were variable (SI Appendix, Fig. S5 B–G; see legends of SI Appendix, Fig. S5 for details and discussion). In harsh environmental conditions, daf-2 mutations increase the formation of dauer, a hibernation-like, alternative larval stage (reviewed in ref. 15). The hel-1 mutations did not affect the formation of dauer caused by two different alleles of daf-2 mutations, e1370 and e1368 (Fig. 2 G and H and SI Appendix, Fig. S5H). These data suggest that inhibition of hel-1 affects the long lifespan and reproductive span of daf-2 mutants independently of resistance to heat stress and dauer formation.

Genetic Inhibition of hel-1 Specifically Affects Lifespan via IIS.

We next examined whether hel-1 affects the longevity caused by genetic mutations in various aging-regulatory pathways. We measured the lifespan of sensory-defective osm-5(p813), germline-deficient glp-1(e2141), HIF-1 gain-of-function vhl-1(ok161), dietary restriction mimetic eat-2(ad1116), and mitochondrial respiration-defective isp-1(qm150) mutants upon genetic inhibition of hel-1. We found that hel-1 RNAi or mutations had little or no effect on the lifespan of glp-1(e2141) or vhl-1(ok161) mutants (Fig. 3 A and B and SI Appendix, Fig. S6A). The hel-1 mutation or RNAi slightly shortened the lifespan of eat-2(ad1116) and isp-1(qm150) mutants, but the lifespan-decreasing effect on these two mutants was smaller than that on daf-2(-) mutants (Fig. 3 C and D and SI Appendix, Fig. S6 B and C). The sensory osm-5(-) mutants’ longevity, which is dependent largely on the IIS pathway (16), was decreased by hel-1 mutation, although hel-1 RNAi had a small effect (Fig. 3E and SI Appendix, Fig. S6D). Together, these data suggest that hel-1 influences the lifespan primarily through effects on the IIS pathway.

Fig. 3.

The effects of hel-1 RNAi or mutations on the lifespan of various long-lived mutants. (A and B) Long lifespan in glp-1(e2141) [glp-1(-)] (A) or vhl-1(ok161) [vhl-1(-)] (B) mutants was not significantly reduced by knockdown of hel-1. (C and D) hel-1 RNAi slightly decreased longevity in eat-2(ad1116) [eat-2(-)] (C) and isp-1(qm150) [isp-1(-)] (D) mutants, but the effect was much smaller than that in daf-2(e1370) (daf-2(-)) mutants. (E) Long lifespan in osm-5(p813) [osm-5(-)] was decreased by hel-1(gk148684) [hel-1(-)] mutation. Please see SI Appendix, Fig. S6 A–D and the legends for lifespan data regarding hel-1(-); vhl-1(-), hel-1(-) eat-2(-) and hel-1(-); isp-1(-) double mutants and osm-5(-) treated with hel-1 RNAi. Lifespan curves regarding glp-1(-) treated with daf-16 RNAi and vhl-1(-) treated with hif-1 RNAi, which were used as positive controls for the RNAi lifespan experiments, are also shown in SI Appendix, Fig. S6 E and F.

Overexpression of hel-1 Is Sufficient for Long Lifespan.

We generated transgenic animals that expressed HEL-1::GFP fusion proteins controlled by a hel-1 promoter (hel-1p::hel-1::gfp). We found that HEL-1::GFP localized to the nuclei of cells in various tissues, including neurons, hypodermis, and intestine, throughout development (Fig. 4A and SI Appendix, Fig. S7 A–E). The hel-1::gfp transgene appeared to be functional, because it completely rescued longevity in hel-1(-); daf-2(-) animals (Fig. 4B). We also found that the hel-1::gfp transgene slightly but significantly extends lifespan in the wild-type and hel-1(-) backgrounds (Fig. 4 C and D). In contrast, the hel-1::gfp transgene did not further increase the longevity of daf-2 mutants (Fig. 4D). This finding, together with the hel-1 RNAi and mutant data, suggests that HEL-1 is necessary and sufficient for extending lifespan.

Fig. 4.

The expression patterns and the requirement of hel-1 for longevity in various tissues. (A) HEL-1::GFP fusion protein was predominantly expressed in the nuclei of cells in many tissues, including neurons, hypodermis, and intestine in adult animals. Arrows indicate nuclei of the neuronal, hypodermal, and intestinal cells. (Scale bars, 100 μm.) Consistent with previous proteomics (36) and SAGE data (79), we also found that expression of hel-1 was reduced in the daf-2 mutant background using hel-1::gfp transgenic worms (SI Appendix, Fig. S7 F and G) and qRT-PCR (SI Appendix, Fig. S7H). (B) Transgenic expression of hel-1::gfp (hel-1p::hel-1::gfp) (hel-1 OE) in hel-1(gk148684); daf-2(e1370) [hel-1(-); daf-2(-)] mutants restored the longevity of daf-2 mutants. (C) Two independent lines (line 1: yhEx92; line 2: yhEx93) of hel-1p::hel-1::gfp transgenes extended lifespan slightly but significantly. (D) hel-1p::hel-1::gfp did not influence the long lifespan of daf-2 mutants but extended the lifespan of hel-1 mutants. Please note that the lifespan assays for animals with daf-2(+) and daf-2(-) backgrounds were performed separately but are shown together in this panel for comparison. (E and F) Intestine-specific [rde-1(ne219); kbls7(nhx-2p::rde-1; rol-6D)] (E) and neuron-specific [sid-1(pk3321); uIs69(myo-2p::mCherry; unc-119p::sid-1)] (F) hel-1 RNAi decreased the long lifespan of daf-2(e1370) mutants but did not influence the lifespan of control [daf-2(+)] animals. (G) Hypodermis-specific [rde-1(ne219); kzls9(lin-26p::nls::gfp; lin-26p::rde-1; rol-6D)] hel-1 RNAi did not decrease the lifespan of animals with the control or daf-2(-) background. Please note that the requirement of intestinal and neuronal HEL-1 for the longevity of daf-2 mutants is consistent with the tissue-specific requirement of DAF-16 for longevity (80).

HEL-1 in the Intestine and Neurons Plays a Key Role in the Longevity of daf-2 Mutants.

We examined the tissues in which HEL-1 contributed to longevity. We performed lifespan assays using tissue-specific hel-1 RNAi methods by exploiting tissue-specific rde-1 (17) and sid-1 (18) transgenic systems. As expected, hel-1 RNAi did not influence lifespan in RNAi-defective rde-1 or sid-1 mutant backgrounds (SI Appendix, Fig. S8 A and B). We found that intestine- or neuron-specific hel-1 RNAi partly suppressed the long lifespan conferred by daf-2 mutations (Fig. 4 E and F). In contrast, hel-1 RNAi specific for hypodermis or muscle did not reduce the long lifespan (Fig. 4G and SI Appendix, Fig. S8C). Thus, intestinal and neuronal HEL-1 appears to be crucial for the long lifespan of daf-2 mutants.

Up-Regulation of a Large Subset of Genes in daf-2 Mutants Is Affected by hel-1.

To understand the mechanism by which HEL-1 regulated the longevity of daf-2 mutants, we performed an mRNA sequencing using wild-type, hel-1(-), daf-2(-), and hel-1(-); daf-2(-) animals. We analyzed differentially expressed genes (DEGs) in daf-2 mutant and wild-type animals. We found increased expression of 1,914 genes and decreased expression of 243 genes in daf-2 mutants (fold change >1.5, P < 0.05) (Fig. 5A and Dataset S1). The gene list had significant overlaps with those of previous microarray and RNA sequencing studies using daf-2 mutants (Table 2), validating our RNA sequencing data. We then compiled a list of genes that were differentially expressed in daf-2(-) and hel-1(-); daf-2(-) animals. We found that 779 of the 1,914 genes that were up-regulated in daf-2(-) mutants compared with wild-type animals also were up-regulated in daf-2(-) mutants compared with hel-1(-); daf-2(-) animals (Fig. 5 A and B and Dataset S1). In contrast, 14 of 243 genes that were down-regulated in daf-2(-) compared with wild-type animals overlapped with genes that were down-regulated in daf-2(-) compared with hel-1(-); daf-2(-) animals (Fig. 5 A and B). These data indicate that HEL-1 affects the induction of a significant portion of genes in daf-2 mutants but has a minor effect on repression (Fig. 5 A and B). Consistently, more gene ontology (GO) terms overlapped in DEGs that were up-regulated in daf-2 mutants than those that were down-regulated (SI Appendix, Fig. S9A). These results suggest that HEL-1 contributes to changes in gene expression, particularly induction, caused by daf-2 mutations.

Fig. 5.

mRNA sequencing data indicate that hel-1 is required for the induction of DAF-16/FOXO target genes in daf-2 mutants. (A) Venn diagrams show significant overlap between genes up- or down-regulated in daf-2(e1370) [daf-2(-)] animals compared with wild-type and genes up- or down-regulated in daf-2(-) animals compared with hel-1(gk148684); daf-2(e1370) [hel-1(-); daf-2(-)] as indicated (RF, representation factor. ***P < 0.001). The DEGs were selected as the ones with fold changes >1.5 and P < 0.05. (B) Scatterplots showing gene-expression changes in daf-2 mutants by hel-1 mutations. Genes whose expression was significantly induced (Left) or repressed (Right) in daf-2(-) compared with wild-type animals are shown as black dots. Red dots indicate gene-expression changes in hel-1(-); daf-2(-) compared with wild-type animals. (C) A heatmap shows representative overlap between genes up-regulated in daf-2(-) compared with hel-1(-); daf-2(-) animals (hel-1–dependent genes) and genes up-regulated in daf-2(-) compared with daf-16(-); daf-2(-) animals (daf-16–dependent genes). Please see SI Appendix, Fig. S9 B and C for comparison of down-regulated genes. (D) Motif search analysis for upstream sequences of hel-1–dependent up-regulated genes in daf-2(-) animals indicates that DAE, which is also known as PQM-1/GATA-like binding element, was enriched (P value = 1.40E-08, e value = 1.10E-04). See SI Appendix, Fig. S9D for other enriched motifs.

Table 2.

Overlapping DEGs from this study and from published genomics data

Comparison used in our study	No. of DEGs	Comparison used in previous publications	No. of DEGs	No. of overlapping genes	Representation factor	P value
Up in daf-2(-) vs. WT	1,914	Up in daf-2(-) vs. WT (65)	259	132	4.7	<10−58
Up in daf-2(-) vs. WT	1,914	Up in daf-2(-) vs. WT (19)	3223	810	2.7	<10−191
Down in daf-2(-) vs. WT	243	Down in daf-2(-) vs. WT (65)	250	17	4.9	<10−7
Down in daf-2(-) vs. WT	243	Down in daf-2(-) vs. WT (19)	577	63	9.2	<10−41
Up in daf-2(-) vs. hel-1(-); daf-2(-)	779	Up in daf-2(-) vs. daf-16(-); daf-2(-) (22)	774	174	5.9	<10−86
Up in daf-2(-) vs. hel-1(-); daf-2(-)	779	Up in daf-2(-) vs. daf-16(-); daf-2(-) (19)	1,794	176	2.6	<10−32
Up in daf-2(-) vs. hel-1(-); daf-2(-)	779	Up in daf-2(-) vs. swsn-1(-); daf-2(-) (19)	1,113	88	2.1	<10−10
Up in daf-2(-) vs. hel-1(-); daf-2(-)	779	Up in daf-2(-) vs. daf-2(-); skn-1(-) (20)	428	84	5.2	<10−35

The number of up- or down-regulated genes indicates genes with altered expression for each comparison. The specified numbers of genes in this table were used for calculating overlaps among the datasets. The raw RNA sequencing data generated in ref. 19 were reanalyzed, and DEGs were identified using the method described in Materials and Methods. DEGs in the genomewide microarray data in ref. 22 were selected using an arbitrary cutoff [daf-16–dependent up-regulated genes in daf-2(-): t value >4; daf-16–dependent down-regulated genes in daf-2(-): t value ≤4]. DEGs in refs. 20 and 65 were used based on the information in the supplemental tables of those papers. Representation factors are ratios of actual overlaps divided by expected overlaps calculated from random independent datasets. P values were calculated using exact hypergeometric probability. WT: wild-type, (-) indicates reduction of function by mutations or RNAi.

HEL-1 Extends Lifespan by the Induction of DAF-16/FOXO–Dependent Genes in daf-2 Mutants.

How does HEL-1 affect mRNA levels of genes induced in daf-2 mutants? DAF-16/FOXO is an essential longevity transcription factor in the IIS pathway. The longevity effect of DAF-16 results mostly from its function as a transcriptional activator in animals with reduced IIS (19, 20). Therefore, we examined whether HEL-1 contributes to the induction of genes in daf-2 mutants by influencing gene expression regulated by DAF-16. We compared the DEGs that were up-regulated in daf-2 mutants versus hel-1; daf-2 mutants with those that were previously published as being up-regulated in daf-2 mutants versus daf-16; daf-2 mutants. Importantly, we found significant overlap between our data and published data (Fig. 5C, Table 2, and SI Appendix, Fig. S9 B and C). We also found that the DAF-16–associated element (DAE) was one of the most significantly enriched (P value = 1.40E-08) motifs in the upstream sequences of DEGs that were up-regulated in daf-2 mutants versus hel-1; daf-2 mutants (Fig. 5D and SI Appendix, Fig. S9D). SWI/SNF chromatin remodeler and SKN-1/NRF2 are longevity factors that work together with DAF-16/FOXO (20–22). We identified significant overlaps between our hel-1–dependent up-regulated genes and genes up-regulated by SWI/SNF, DAF-16, or SKN-1 (Table 2) (20, 21, 23). Together these data suggest that HEL-1 regulates the induction of many DAF-16 target genes in daf-2 mutants.

We examined the expression of 12 of the hel-1–dependent DAF-16 target genes, including sod-3, mtl-1, and dod-11, by quantitative RT-PCR (qRT-PCR). We found that hel-1 mutations partly suppressed the induction of these 12 DAF-16 target genes in daf-2 mutants (Fig. 6 A–H and SI Appendix, Fig. S10 A–D). In contrast, hel-1 mutations had little effect on the mRNA expression of other genes tested in daf-2(-) mutants; these genes included chaperone genes, hsp-16.1/2, hsp-70, aip-1, and hsp-12.6, which are targets of both HSF-1 and DAF-16 (SI Appendix, Fig. S10 E–J). We also found that overexpression of hel-1 increased the expression of three (mtl-1, sqrd-1, and B0218.8) of the five HEL-1–dependent DAF-16 target genes that we tested (Fig. 6 I–M). Thus, HEL-1 contributes to the induction of a subset of DAF-16 target genes in daf-2 mutants.

Fig. 6.

hel-1 appears to increase the lifespan of daf-2 mutants by up-regulating DAF-16/FOXO. (A–H) mRNA expression levels of mtl-1 (A), sod-3 (B), dod-11 (C), sqrd-1 (D), B0218.8 (E), cpr-2 (F), F35E8.6 (G), and sdz-24 (H) measured using qRT-PCR were increased in daf-2(-) mutants. The hel-1 mutation appeared to suppress the induction of the genes by daf-2 mutation. n = 9 for A–C and n = 8 for D–H. (I–K) qRT-PCR results show that mRNA expression levels of mtl-1 (I), sqrd-1(J), and B0218.8 (K) were increased by hel-1 overexpression (hel-1 OE, yhIs45). (L and M) sod-3 (L) and dod-11 (M) were not induced by hel-1 overexpression. n = 4 for I–M. Error bars indicate SEM.

Physical Interaction Between HEL-1 and DAF-16/FOXO Appears to Regulate Longevity.

We examined the mechanism by which HEL-1 influenced the expression of DAF-16 target genes by assessing known modes of activation, including physical interaction with cofactors (20), nuclear translocation (24–26), and induction at the transcriptional level (27). We performed coimmunoprecipitation assays in cultured mammalian cells and found that HEL-1 bound to DAF-16 (Fig. 7 A and B). In contrast, hel-1 mutations did not affect nuclear localization or mRNA levels of DAF-16 (Fig. 7 C–E). Thus, HEL-1 appears to regulate the transcriptional activity of DAF-16 through physical interaction. We next tested whether daf-16 was involved in the requirement of hel-1 for the longevity of daf-2 mutants. We found that hel-1 mutations had little or no effect on the shortened lifespan of daf-16(RNAi) or daf-16(RNAi); daf-2(-) animals (Fig. 7 F and G). In contrast, hel-1 mutations further decreased the lifespan of hsf-1(RNAi) and hsf-1(RNAi); daf-2(-) animals (SI Appendix, Fig. S11 A and B). These results are consistent with the concept that HEL-1 and DAF-16 act together to influence longevity in daf-2 mutants. Taken together, our data suggest that HEL-1 promotes longevity conferred by reduced IIS by specifically activating DAF-16/FOXO via physical interaction.

Fig. 7.

HEL-1 appears to promote the longevity of daf-2 mutants through physical interaction with DAF-16/FOXO. (A and B) Coimmunoprecipitation (Co-IP) assays for the detection of a physical interaction between HEL-1 and DAF-16 in HEK 293 cells. (A) HA-tagged DAF-16 (HA-DAF-16) was coimmunoprecipitated with FLAG-tagged HEL-1 (FLAG-HEL-1) using anti-FLAG antibody (two of four trials). (B) HA-tagged HEL-1 (HA-HEL-1) was coimmunoprecipitated with FLAG-tagged DAF-16 (FLAG-DAF-16) using anti-FLAG antibody (five of seven trials). Proteins were immunoblotted (IB) using anti-HA or anti-FLAG antibodies as indicated. For inputs, five percent (A) and 0.2% (B) of total proteins were used. (C) hel-1 mutations did not influence the nuclear localization of DAF-16::GFP in daf-2(e1368) [daf-2(-)] mutants. (Scale bars, 100 μm.) (D) Quantification of data in C (n > 20, triplicate). (E) mRNA expression levels of daf-16 were not influenced by hel-1 mutations in daf-2(e1370) [daf-2(-)] mutants. n = 3. Error bars indicate SEM. (F and G) hel-1(gk14868) [hel-1(-)] mutations did not further decrease the lifespan of daf-16(RNAi) (F) or daf-16(RNAi); daf-2(e1370) [daf-2(-)] (G) animals.

Discussion

RNA helicases are evolutionarily conserved enzymes that help processing various kinds of RNA. However, the functions of RNA helicases in aging are poorly understood. Here we exploited the molecular genetics of C. elegans to identify RNA helicases that influence lifespan, in particular via the IIS pathway. We revealed the mechanisms and functions of a DEAD-box helicase, HEL-1, which influenced lifespan through IIS by regulating DAF-16/FOXO activity. The connection between RNA helicases and the regulation of organismal lifespan has not been previously reported. Therefore, our study provides insights into mechanisms through which longevity is achieved at the level of RNA regulation.

In addition to HEL-1, we identified seven other RNA helicases that play rather specific roles in the regulation of lifespan in daf-2 mutants. The smg-2 gene encodes one of the subunits of the nonsense-mediated decay complex (28). The rpa-1 gene is a homolog of the replication protein A large subunit, which functions in double-stand break repair (29), and R03H10.6 is a telomeric DNA-binding protein (30). The mog-5 gene is a homolog of yeast PRP-22 (31), which regulates premessenger RNA splicing (32), a process that combines exons through removing introns. The sacy-1 gene functions as a negative regulator of meiotic maturation (33). smut-1 plays roles in the formation of endogenous siRNA (34). The function of ZK512.2 is not known in either C. elegans or mammals. Overall, RNA helicases that influence lifespan in IIS regulate various processes of RNA maintenance and biogenesis. Repair and splicing of RNAs may be particularly important for longevity by contributing to the quality control of RNAs.

How does HEL-1 influence the induction of DAF-16 target genes and regulate lifespan? The orthologs of HEL-1 have been shown to act as components of the mRNA export complex, which translocates mRNAs from the nucleus to the cytoplasm (35). Thus, HEL-1 may promote longevity by expediting the nuclear-to-cytoplasmic flow of mRNAs that are crucial for longevity, such as DAF-16 target gene mRNAs. Another possibility is that HEL-1 may help the processing of pre-mRNAs to generate mature mRNAs of DAF-16 target genes, because a well-known function of RNA helicases is the regulation of mRNA splicing. However, our qRT-PCR data did not support this possibility (SI Appendix, Fig. S12). Based on our data, we speculate that HEL-1 may act as a transcription coregulator, and there are precedents for the roles of RNA helicases as transcription regulators (7–10). In any case, in contrast to the expectation that RNA helicases have general housekeeping roles in RNA metabolism, our findings reveal that the RNA helicase HEL-1 has specific roles in a specific longevity pathway.

An intriguing finding in our study is that the level of HEL-1 was decreased in long-lived daf-2 mutants (SI Appendix, Fig. S7 F–H), whereas HEL-1 contributed to longevity. This finding is reminiscent of TAX-6/calcineurin levels in daf-2 mutants and the role of TAX-6 in lifespan; TAX-6 level is increased in daf-2(-) mutants, but knockdown of tax-6 extends lifespan (36). The authors suggested that TAX-6 level is increased in daf-2 mutants as a compensatory response (36). Likewise, the level of a longevity-promoting factor HEL-1 may be decreased as a compensatory response to maintain IIS at a normal level. Previous reports show that ATP enhances the activity of the HEL-1 homologs, human UAP56 (37) and yeast SUB2 (38). Interestingly, ATP levels are increased by genetic inhibition of daf-2 (39, 40). Therefore, we speculate that high ATP levels in daf-2 mutants may increase the activity of HEL-1, possibly counteracting the decreased expression of hel-1 in daf-2 mutants.

Mutation or dysregulation of RNA helicases has been linked to diseases and implicated in cellular senescence in mammals (41). The mammalian HEL-1 homolog DDX39A binds to telomeric repeat-binding factor 2 (TRF2) to modulate the maintenance of telomere length and genomic integrity, suggesting a possible role for DDX39A in cellular senescence (42). In cultured mammalian cells, DDX39A negatively regulates the expression of inflammation-associated proteins (43). Moreover, mRNA levels of DDX39A are increased in the frontal cortex of patients with Alzheimer’s disease (44). Further, a polymorphism of the DDX39A promoter is associated with increased risk of Alzheimer’s disease (44). Many aging-related genes discovered in C. elegans are evolutionarily conserved; thus, our study raises the possibility that RNA helicases, including DDX39A, may regulate organismal lifespan in mammals.

Materials and Methods

Strains.

All strains were maintained at 20 °C. Some C. elegans strains used in this study were provided by Caenorhabditis Genetics Center, which is funded by the NIH National Center for Research Resources. The following C. elegans strains were used: N2 wild-type, CF1041 daf-2(e1370) III outcrossed six times to N2, CF3152 rrf-3(pk1426) II, CF1814 rrf-3(pk1426) II; daf-2(e1370) III, IJ474 yhEx92[hel-1p::hel-1::gfp; odr-1p::rfp], IJ475 yhEx93[hel-1p::hel-1::gfp; odr-1p::rfp], IJ571 daf-2(e1370) III; yhEx92[hel-1p::hel-1::gfp; odr-1p::rfp], IJ640 hel-1(gk148684) II; yhEx92[hel-1p::hel-1::gfp; odr-1p::rfp], IJ883 hel-1(gk148684) II; daf-2(e1370) III; yhEx92[hel-1p::hel-1::gfp; odr-1p::rfp], IJ791 yhIs45[hel-1p::hel-1::gfp; odr-1p::rfp], IJ828 daf-2(e1370) III; yhIs45[hel-1p::hel-1::gfp; odr-1p::rfp], IJ501 hel-1(gk148684) II outcrossed eight times to N2, IJ527 hel-1(gk148684) II; daf-2(e1370) III, CF1085 daf-16(mu86) I; daf-2(e1370) III, IJ385 daf-2(e1368) III outcrossed nine times to N2, IJ407 daf-16(mu86) I; daf-2(e1368) III, IJ1121 hel-1(gk148684) II; daf-2(e1368) III, CF2553 osm-5(p813) X outcrossed three times to N2, IJ1116 hel-1(gk148684) II; osm-5(p813) X, CF1903 glp-1(e2141) III, IJ173 eat-2(ad1116) II outcrossed four times to N2, IJ1117 hel-1(gk148684) II eat-2(ad1116) II, CF2172 isp-1(qm150) IV outcrossed three times to N2, IJ1120 hel-1(gk148684) II; isp-1(qm150) IV, IJ7 vhl-1(ok161) X outcrossed three times to N2, IJ1119 hel-1(gk148684) II; vhl-1(ok161) X, WM27 rde-1(ne219) V, NR222 rde-1(ne219) V; kzls9[lin-26p::nls::gfp; lin-26p::rde-1; rol-6D], NR350 rde-1(ne219) V; kzls20[hlh-1p::rde-1; sur-5p::nls::gfp], VP303 rde-1(ne219) V; kbls7[nhx-2p::rde-1; rol-6D], IJ411 daf-2(e1370) III; rde-1(ne219) V, IJ415 daf-2(e1370) III; rde-1(ne219) V; kzls9[lin-26p::nls::gfp; lin-26p::rde-1; rol-6D], IJ416 daf-2(e1370) III; rde-1(ne219) V; kzls20[hlh-1p::rde-1; sur-5p::nls::gfp], IJ417 daf-2(e1370) III; rde-1(ne219) V; kbls7[nhx-2p::rde-1; rol-6D], CF2688 daf-16(mu86) I; daf-2(e1368) III; muIs112[daf-16p::daf-16::gfp; odr-1p::rfp], IJ817 daf-16(mu86) I; hel-1(gk148684) II; daf-2(e1368) III; muIs112[daf-16p::daf-16::gfp; odr-1p::rfp], NL3321 sid-1(pk3321) V, IJ807 daf-2(e1370) III; sid-1(pk3321) V, TU3401 sid-1(pk3321) V; uIs69[myo-2p::mCherry; unc-119p::sid-1], IJ898 daf-2(e1370) III; sid-1(pk3321) V; uIs69[myo-2p::mCherry; unc-119p::sid-1].

RNA Helicase Gene Search.

RNA helicases were identified based on the sequences for RNA helicase-containing domains, DEAD/DEAH-box helicases (PF00270), helicase conserved C-terminal domains (PF00271), and tRNA/helicase type (PF01336) domains. For stringent selection of RNA helicase genes, amino acid sequences of genes were aligned with reference domain sequences and were selected as positive if at least 95% of domain sequences were covered. In addition to the Pfam domain search methods (pfam.xfam.org/) (45), RNA helicase database (www.rnahelicase.org/) (46) and WormBase (www.wormbase.org/) (47) were used to manually identify additional RNA helicases.

Generation of the Phylogenic Tree.

A phylogenic tree for potential RNA helicases in C. elegans was built based on sequence similarity using the Dendroscope program (dendroscope.org) (48). A phylogenetic tree showing the sequence similarities among HEL-1 proteins in several organisms was generated using ClustalW2 (www.ebi.ac.uk/Tools/msa/clustalw2/) (49) and was revisualized by TreeView (50).

Sequence Homology Alignment.

Protein sequences were obtained from the National Center for Biotechnology Information. Sequence homology was aligned by using the ClustalW2 (www.ebi.ac.uk/Tools/msa/clustalw2/) (49) and ClustalX programs (51).

Schematic Domain Structure of HEL-1.

The HEL-1 protein domain structure was visualized by using Exon-Intron Graphic Maker (wormweb.org/exonintron).

Effects of Genetic Inhibition of hel-1 on C. elegans Development.

Effects of hel-1 mutations or RNAi on development were determined by observing animals under a dissecting stereomicroscope (SMZ645; Nikon) with a DIMIS-M camera (Siwon Optical Technology) 3 d (wild-type) or 4 d [daf-2(e1370)] after hatching.

Lifespan Assays.

Lifespan assays were performed as described previously, with some modifications (52). Gene-specific RNAi clones from the Julie Ahringer (11) or Marc Vidal libraries (12) were cultured overnight in LB with 50 μg/mL ampicillin (USB) at 37 °C and then were seeded onto nematode growth media (NGM) plates containing 50 μg/mL ampicillin. RNAi bacteria seeded on plates were treated with 1 mM isopropylthiogalactoside (IPTG; Gold Biotechnology) to induce dsRNA expression for 1 d at room temperature. Gravid adults were placed on plates containing control RNAi bacteria to synchronize the worms, and the progeny were allowed to develop to the L4 stage and then were transferred onto gene-specific RNAi plates. After 1 or 2 days, the worms were transferred again onto fresh RNAi plates containing 5 μM FUdR (Sigma) to prevent progeny from developing during lifespan measurements. For the lifespan experiments without FUdR treatment, worms were transferred every other day until they stopped producing progeny. For the lifespan experiments of mutants, worms were grown on OP50-seeded plates and then were transferred onto FUdR-treated plates at day 1 and again at day 2 of adulthood. All lifespan assays were performed at 20 °C, except for glp-1(e2141) mutants that display germline deficiency at 25 °C. The glp-1(-) mutants were allowed to develop to L4 stage at 25 °C, and the same lifespan-measuring methods were applied after L4 stage at 20 °C. Three plates with a total of >25 worms per plate were used for each condition. Dead worms were removed from plates and scored as dead. Worms that crawled off, displayed internal hatching or vulval extrusion were censored but were included in subsequent statistical analysis. All RNAi clones were verified by sequencing and BLAST analysis (53) before use. Online application of survival analysis (OASIS; sbi.postech.ac.kr/oasis) (54) was used for statistical analysis. P values were calculated by using the log-rank (Mantel–Cox) method. For the lifespan graphs shown in Fig. 4D and SI Appendix, Fig. S2, lifespan assays for animals with daf-2(+) and daf-2(-) backgrounds were performed separately but are shown in the same panels for comparisons.

Generation of Transgenic Worms and Fluorescence Observation.

hel-1p::hel-1::gfp transgenic animals were generated as described previously, with some modifications (52). A promoter and the coding region of hel-1 (3.6 kb) were cloned into the pDESTR4-R3 vector using the Gateway cloning system (Invitrogen). The construct (25 ng/μL) was microinjected into the gonad of day 1 adult worms with the coinjection marker odr-1p::rfp (75 ng/μL) (55–57). For the images of hel-1p::hel-1::gfp animals, animals were synchronized on OP50-seeded plates, and the fluorescence and brightfield images were captured using an AxioCam HRc CCD digital camera (Zeiss Corporation) with a Zeiss Axio Scope A1 compound microscope (Zeiss Corporation) or a Zeiss LSM 780 confocal microscope (Zeiss Corporation).

Determination of DAF-16 Subcellular Localization.

The subcellular localization of DAF-16::GFP was determined as described previously, with some modifications (26). Two mM levamisole (tetramisole; Sigma) was used to anesthetize worms on 2% (mass/vol) agar pads. Worms at L2 or L3 larval stage were used to observe the DAF-16 localization. Fluorescence images were captured using an AxioCam HRc CCD digital camera (Zeiss Corporation) with a Zeiss Axio Scope A1 compound microscope (Zeiss Corporation).

Reproductive Span.

Reproductive span assays were performed as described previously, with some modifications (13). Synchronized individual L4-stage worms were transferred to new plates until they had stopped laying eggs at least for 2 d.

Heat Stress Resistance Assays.

Heat stress resistance assays were performed as previously described, with minor modifications (58). Synchronized eggs were allowed to develop on control RNAi plates until they reached L4 stage. For RNAi experiments, ∼120 L4-stage worms were transferred onto each NGM plate containing RNAi bacteria. Synchronized day 2 adults were transferred from a 20 °C incubator into a 35 °C incubator for heat treatment. For experiments with mutants, animals were transferred to new OP50-seeded plates at day 1 adult stage before heat treatment. The survival of animals was examined first after 6 and 12 h for wild-type and daf-2 mutant animals, respectively, and then was determined every 2–3 h. The survival assays were repeated at least twice. OASIS was used for statistical analysis, and P values were calculated using the log-rank (Mantel–Cox method) test (54).

mRNA Sequencing.

Day 1 adult worms were obtained by culturing synchronized eggs from bleached gravid adults. Total RNAs were extracted from the day 1 adult worms using RNA Isoplus (Takara) extraction methods. For each condition [wild-type, daf-2(e1370), hel-1(gk148684), or hel-1(gk148684); daf-2(e1370)], two independent biological repeats were used and processed as duplicates until final analysis. The quality of total RNA samples was examined by measuring RNA integrity numbers (RINs) using the Agilent RNA 6000 Nano Kit (Agilent Technologies). For all samples, RINs were sufficiently high (>9.5) to perform mRNA sequencing. For the preparation of cDNA libraries, mRNAs were extracted and converted to cDNAs by reverse transcription. The cDNAs were sequenced using a HiSEq 2500 platform (Chunlab Inc.).

mRNA Sequencing Analysis.

The quality of generated sequencing reads was verified using FastQC (www.bioinformatics.babraham.ac.uk/projects/fastqc). The adapter sequences (TruSeq Universal Adapter and indexed TruSeqTM Adapters, provided by Illumina) then were trimmed using the Cutadapt software (59). These adapter-trimmed reads were aligned to the C. elegans genome (WS245) provided by WormBase using TopHat2 (60) with default options. Two mismatches were allowed in the seed region defined by Bowtie (61). Alignment results were submitted to HTsEq (62) to obtain the counts of the reads aligned onto each annotated gene. The locus information of all annotated genes was provided by WormBase (WS245). The obtained read counts then were normalized by the trimmed mean of m values method implemented in edgeR (63). Normalized counts for all annotated genes were transformed to log₂ values after adding one to avoid negative infinity. Genes were considered expressed if the normalized counts of the genes were larger than a cutoff value in at least one condition, and the first quartile of the normalized count (in this study, 0.3271) was determined as the cutoff value. To identify DEGs, Student’s t test and the log₂ median ratio test were performed to compute t values and median ratios for all the annotated genes (64–66). The adjusted P values from each test were computed using an empirical distribution of the null hypothesis, which was obtained from random permutations of the samples. Finally, the adjusted P values from the individual tests were combined to compute the overall P values using Stouffer's method (65, 66), and genes with overall P < 0.05 and fold change >1.5 were selected as DEGs.

Enrichment Analysis of GO Biological Processes.

GO biological processes (GOBPs) represented by the DEGs were identified using DAVID (david.abcc.ncifcrf.gov/) (67). For the up- or down-regulated genes, only the GOBPs that have more than five genes and enrichment P values < 0.05 were included. The P values for all of the selected GOBPs were presented in an enrichment heatmap.

Motif Analysis.

Motif analysis was performed as previously described, with some modifications (21). Oligo analysis was performed for unbiased search using modified parameters for oligomer lengths (6–8), pseudofrequency (<0.01), and flanking residues (3) based on RSAT algorithms (rsat.ulb.ac.be) (68). Upstream sequences from −800 to −1 (default parameters) of HEL-1–dependent up-regulated genes were used as query sequences. Identified motifs were compared with the JASPAR nematode database in the TOMTOM tool (meme-suite.org/tools/meme) (69) to predict transcription factor-binding sites. Sequence motifs for specific transcription factor binding were based on previous reports: DAF-16–binding element sites (GTAAARA), DAF-16–associated sites (GAKAAG, also known as PQM-1/GATA-like binding sites) (23, 70, 71), HSF-1–binding sites (GGGTGTC and TTCTAGAA) (72), and SKN-1–binding sites (WWTRTCAT) (73). The χ² test was used to confirm enrichment of motifs.

Scatterplots, Venn Diagrams, and Heatmaps for RNA Sequencing Results.

Scatterplot (20), Venn diagram (74), and heatmap (21) analysis was performed as previously described, with some modifications. Scatterplots show gene expression in daf-2(e1370) and hel-1(gk1448684); daf-2(e1370) mutant strains compared with that in wild-type. Genes significantly induced or repressed by daf-2(e1370) mutations were shown. Scatterplots and heatmaps were drawn using MATLAB.

Dauer Formation.

Gravid adult worms were allowed to lay eggs at 22.5 °C or 25 °C for 6 h, and the progeny were grown for dauer formation assays for 3–4 d. Dauer formation was examined visually based on well-known dauer morphology (reviewed in ref. 15) under dissecting stereomicroscopes (SMZ645; Nikon).

qRT-PCR Analysis.

Day 1 adult worms synchronized by using a bleaching method were harvested, and RNA Isoplus (Takara) was used for RNA extraction. RNA was converted to cDNA using a reverse transcription system (Promega) with random primers. cDNA was used for qPCR to measure the expression of each specific gene with SYBR Green dye (Applied Biosystems) using the StepOne Real Time PCR System (Applied Biosystems) and was analyzed using the comparative C_T method. The ama-1 mRNA level was used as a control for normalization. The list of oligonucleotides used for the qRT-PCR is given in SI Appendix, Materials and Methods.

Modeling of HEL-1 Protein Structure.

Amino acid sequences of HEL-1 in wild-type and hel-1(gk148684) mutants animals were used to query SWISS-MODEL alignment (swissmodel.expasy.org/) (75). The PyMOL program was used to visualize the proteins (www.pymol.org/) (76). Transparency was established on 0.3.

Generation of Plasmids for Transfection in HEK 293 Cells.

The ORF of hel-1 (1.3 kb) or daf-16 (1.6 kb) cDNA was cloned into the HA or FLAG tag-containing pcDNA3.1(+) vector using the In-Fusion system (Clontech). Specifically, HEL-1 or DAF-16 cDNA was amplified using primers containing recombination regions in the vector. The vector was digested with two restriction enzymes (EcoRI/BamHI or BamHI/HindIII), and then homologous recombination was performed using the In-Fusion enzyme at 50 °C. Plasmids were verified by sequencing (Macrogen).

Coimmunoprecipitation and Western Blot Assays.

HEK 293 cells were grown in the presence of 5.0% (vol/vol) CO₂ in DMEM (HyClone) with penicillin/streptomycin (HyClone) and 10% (vol/vol) FBS (HyClone) at 37 °C. Transfection was performed using Lipofectamine 2000 reagent (Invitrogen) or METAFECTENE (Biontex Laboratories) by an optimized protocol provided by the manufacturer. Two micrograms of each plasmid expressing FLAG- or HA-tagged protein was used for transfection. Coimmunoprecipitation assays were performed as previously described (77), with some modifications. HEK 293 cell lysates were collected using a lysis buffer [20 mM Tris⋅HCl (pH 7.4), 10 mM KCl, 10 mM MgCl₂, 2 mM EDTA, 10% glycerol, 1% Triton X-100, 2.5 mM β-glycerophosphate, 1 mM NaF, 1 mM DTT, 1 mM PMSF, and protease inhibitor mixture (Biovison)] and were incubated on ice for 10 min. After sonication of the lysates (output setting 8, power output 30 W), 420 mM NaCl was added and then incubated at 4 °C for 1 h on an end-over-end mixer. The lysates were sonicated again to reduce viscosity and were centrifuged at 13,000 × g (5415R; Eppendorf) for 20 min at 4 °C. A total of 1 mg of proteins was incubated with 10 μL of anti-FLAG antibody-conjugated protein G agarose (Sigma) (Fig. 7A), or a total of 3 mg of protein was incubated overnight with FLAG antibody (Sigma) and then protein G-agarose was added (GE Healthcare) (Fig. 7B). The immunoprecipitated complex was incubated for 3 h at 4 °C on an end-over-end mixer and subsequently was washed three times with the lysis buffer by centrifugation at 4,000 × g for 1 min. The immunoprecipitated HEL-1 or DAF-16 complex then was denatured by using 2× SDS sample buffer at 95 °C and detected by using standard Western blot assays (52). Primary antibodies against HA (1:5,000, Sigma, or 1:500, Roche) and FLAG (1:5,000; Sigma) were used with anti-rabbit (1:15,000; Thermo) and anti-mouse (1:10,000; Thermo) secondary antibodies, respectively.

Preparation of Worm Extract and Western Blot Assays.

Synchronized day 1 adult worms were freshly frozen in liquid nitrogen with 2× SDS sample buffer. Frozen samples were boiled at 95 °C for 10 min and vortexed for 10 min. The extract was centrifuged at 14,000 × g for 30 min at 4 °C. The supernatant was used for the Western blot assays as described previously (52). Primary antibodies against GFP (1:1,000; made in house) and alpha tubulin (1:1,000; Santa Cruz) were used with anti-rabbit (1:15,000; Thermo) and anti-mouse (1:10,000; Thermo) secondary antibodies, respectively.