Caenorhabditis elegans EAK-3 inhibits dauer arrest via nonautonomous regulation of nuclear DAF-16/FoxO activity

Abstract

Insulin regulates development, metabolism, and lifespan via a conserved PI3K/Akt pathway that promotes cytoplasmic sequestration of FoxO transcription factors. The regulation of nuclear FoxO is poorly understood. In the nematode Caenorhabditis elegans, insulin-like signaling functions in larvae to inhibit dauer arrest and acts during adulthood to regulate lifespan. In a screen for genes that modulate C. elegans insulin-like signaling, we identified eak-3, which encodes a novel protein that is specifically expressed in the two endocrine XXX cells. The dauer arrest phenotype of eak-3 mutants is fully suppressed by mutations in daf-16/FoxO, which encodes the major target of C. elegans insulin-like signaling, and daf-12, which encodes a nuclear receptor regulated by steroid hormones known as dafachronic acids. eak-3 mutation does not affect DAF-16/FoxO subcellular localization but enhances expression of the direct DAF-16/FoxO target sod-3 in a daf-16/FoxO- and daf-12-dependent manner. eak-3 mutants have normal lifespans, suggesting that EAK-3 decouples insulin-like regulation of development and longevity. We propose that EAK-3 activity in the XXX cells promotes the synthesis and/or secretion of a hormone that acts in parallel to AKT-1 to inhibit the expression of DAF-16/FoxO target genes. Similar hormonal pathways may regulate FoxO target gene expression in mammals.

Fig. 1. eak-3 interacts with the daf-2 insulin-like signaling pathway

Animals were assayed for dauer arrest phenotypes at 25°C (A. and B.) or 27°C (C. and D.). A. eak-3 mutation enhances the dauer arrest phenotype of akt-1 mutants but not of eak-4 or sdf-9 mutants. B. Dauer arrest of an eak-3;akt-1 double mutant is suppressed by a mutation in daf-16/FoxO but not by mutations in daf-3/SMAD or osm-5/Tg737. C. Dauer arrest of an eak-3 mutant at 27°C is suppressed by a mutation in daf-16/FoxO. D. Dauer arrest of an eak-3 mutant at 27°C is suppressed by daf-18/PTEN loss-of-function and akt-1 gain-of-function mutations. Data are represented as mean + s.d. The number of animals scored is documented in Table S1.

Fig. 2. eak-3 interacts with the daf-9/CYP27A1 hormonal pathway

Animals were assayed for dauer arrest phenotypes at 25°C (A., B., D., and E.) or 27°C (C.). A. eak-3 mutation enhances the dauer arrest phenotype of a partial loss-of-function allele of daf-9. B. Dauer arrest of an eak-3;akt-1 double mutant is suppressed by a mutation in daf-12. C. Dauer arrest of an eak-3 mutant at 27°C is suppressed by a mutation in daf-12. D. eak-3 mutation enhances the dauer arrest phenotype of a daf-36 null mutation. E. akt-1 RNAi enhances the dauer arrest phenotype of eak-3 and daf-36 mutants. Data are represented as mean + s.d. The number of animals scored is documented in Table S1.

Fig. 3. Effects of eak-3 mutation on lifespan

Lifespans were assayed at 25°C (A.) or 20°C (B. and C.). The number of animals scored is documented in Table S1. Mean lifespans (days) +/− standard deviation are as follows: A. (p-value vs. wild-type): wild-type, 12.56 +/− 1.99; daf-2(e1370), 22.54 +/− 2.89 (p = 4.53 × 10⁻³⁵); akt-1(mg306), 13.49 +/− 2.92 (p = 0.06); eak-3(mg344), 11.84 +/− 2.97 (p = 0.16); eak-3;akt-1, 12.64 +/− 1.79 (p = 0.83). For akt-1 vs. eak-3, p = 0.01; akt-1 vs. eak-3;akt-1, p = 0.08; eak-3 vs. eak-3;akt-1, p = 0.11. B. daf-2, 45.82 +/− 4.06; eak-3 daf-2, 50.85 +/− 4.58 (p = 3.67 × 10⁻¹¹). C. wild-type, 25.47 +/− 4.61; eak-3, 24.54 +/− 5.22 (p = 0.27); daf-16, 15.97 +/− 2.26; daf-16;eak-3, 16.24 +/− 2.25 (p = 0.42 vs. daf-16).

Fig. 4. Expression pattern of eak-3::GFP fusion constructs

A. Fluorescent protein expression in animals harboring eak-3p::GFP and sdf-9p::RFP promoter fusions. B. Fluorescent protein expression in animals harboring EAK-3::GFP translational and sdf-9p::RFP promoter fusions. C. GFP localization in animals expressing wild-type EAK-3::GFP or EAK-3::GFP containing a mutation at the conserved glycine of the N-myristoylation motif (G2A). Representative images are shown.

Fig. 5. Effects of eak-3 mutation on DAF-16/FoxO subcellular localization and activity in vivo

A. Wild-type and mutant animals (late L1 to early L2 larvae) harboring a DAF-16::GFP translational fusion were assayed for DAF-16::GFP subcellular localization. B. Wild-type and mutant animals (late L1 to early L2 larvae) harboring a sod-3p::GFP promoter fusion were assayed for GFP expression. Identical exposure times were used to photograph all animals harboring a specific GFP reporter. Representative images are shown.

Fig. 6. Quantification of endogenous sod-3 mRNA levels in wild-type and mutant L2 stage animals by real-time reverse transcription PCR

A. sod-3 mRNA levels in single and double mutants. B. sod-3 mRNA levels in eak-3;akt-1 double mutants after RNAi. C. sod-3 mRNA levels after akt-1 RNAi in wild-type, eak-3 mutant, and daf-36 mutant animals. Relative expression units are shown. Data are represented as mean + s.d.

Fig. 7. Model of EAK-3 regulation of DAF-16/FoxO target gene expression in A. wild-type and B. eak-3;akt-1 double mutant animals

Schematics of an XXX cell (top of figure) and a target cell (bottom of figure) are shown. Dashed arrows and lines denote hypothesized relationships between molecules. See text for details.

Fig. 7. Model of EAK-3 regulation of DAF-16/FoxO target gene expression in A. wild-type and B. eak-3;akt-1 double mutant animals

Schematics of an XXX cell (top of figure) and a target cell (bottom of figure) are shown. Dashed arrows and lines denote hypothesized relationships between molecules. See text for details.