DAF-16/FoxO and DAF-12/VDR control cellular plasticity both cell-autonomously and via interorgan signaling

Abstract

Many cell types display the remarkable ability to alter their cellular phenotype in response to specific external or internal signals. Such phenotypic plasticity is apparent in the nematode Caenorhabditis elegans when adverse environmental conditions trigger entry into the dauer diapause stage. This entry is accompanied by structural, molecular, and functional remodeling of a number of distinct tissue types of the animal, including its nervous system. The transcription factor (TF) effectors of 3 different hormonal signaling systems, the insulin-responsive DAF-16/FoxO TF, the TGFβ-responsive DAF-3/SMAD TF, and the steroid nuclear hormone receptor, DAF-12/VDR, a homolog of the vitamin D receptor (VDR), were previously shown to be required for entering the dauer arrest stage, but their cellular and temporal focus of action for the underlying cellular remodeling processes remained incompletely understood. Through the generation of conditional alleles that allowed us to spatially and temporally control gene activity, we show here that all 3 TFs are not only required to initiate tissue remodeling upon entry into the dauer stage, as shown before, but are also continuously required to maintain the remodeled state. We show that DAF-3/SMAD is required in sensory neurons to promote and then maintain animal-wide tissue remodeling events. In contrast, DAF-16/FoxO or DAF-12/VDR act cell-autonomously to control anatomical, molecular, and behavioral remodeling events in specific cell types. Intriguingly, we also uncover non-cell autonomous function of DAF-16/FoxO and DAF-12/VDR in nervous system remodeling, indicating the presence of several insulin-dependent interorgan signaling axes. Our findings provide novel perspectives into how hormonal systems control tissue remodeling.

Fig 1. Reagents generated for this study.

(A) Much simplified overview of the dauer pathways, with DAF-3/SMAD, DAF-16/FoxO, and DAF-12/VDR as transcriptional effectors [25]. (B) Genomic loci of daf-16, daf-12, and daf-3. The insertions sites and schematics of the expression reporters and of the AID tag are shown. (C) Schematic of the AID system. Skp1, Cul1, Rbx1, and E2 are phylogenetically conserved components of the E3 ligase complex. TIR1 is a plant-specific substrate-recognizing subunit of the E3 ligase complex. AID, fused to a protein of interest, is bound by TIR1 in the presence of auxin, which leads to ubiquitination and proteasomal degradation of the protein of interest. (D) Summary of TIR1 transgenes utilized in this study. See Materials and methods for details on TIR constructs [,–40]. AID, auxin-inducible degron.

Fig 2. Expression pattern of daf-16/FoxO, daf-12/VDR, and daf-3/SMAD.

(A) Expression of the daf-3::GFP CRISPR allele at different stages in development (in well-fed conditions). Anterior is to the left on all images. Scale bar, 20 μm (same for all images). (B) Expression of the daf-16::mKate2 CRISPR allele at different stages in development and nutritional states. Different tissue types are indicated with color-coded circles (or an arrow, in the case of neurons). (C) Expression of the daf-12::GFP CRISPR allele at different stages in development (in well-fed conditions). AID, auxin-inducible degron.

Fig 3. Requirement of DAF-16/FoxO, DAF-12/VDR, and DAF-3/SMAD for initiation and maintenance of the dauer state.

(A) Schematic of the experimental design for testing conditional alleles of daf-16, daf-12, and daf-3 for the requirement of the respective proteins in initiation of dauer formation (also applies to experiments described in later figures). (B) Dauer formation is suppressed upon ubiquitous depletion of DAF-16/FoxO, DAF-12/VDR, and DAF-3/SMAD, which serves as a positive control for subsequent experiments with tissue-specific protein depletion. (C) Schematic of the experimental design for testing conditional alleles of daf-16, daf-12, and daf-3 for the requirement of the respective proteins in maintenance of the dauer state. (D) Upon ubiquitous depletion of DAF-16/FoxO, DAF-12/VDR, and DAF-3/SMAD after dauer formation, worms exit the dauer state and initiate post-dauer development, indicating the requirement of the 3 proteins in the active maintenance of the dauer state. The data underlying this figure can be found in S1 Data. AID, auxin-inducible degron.

Fig 4. Neuronal depletion of DAF-3/SMAD, but not of DAF-16/FoxO or DAF-12/VDR, suppresses dauer formation.

(A) Quantification of dauer formation upon depletion of DAF-3/SMAD with a panneuronal (otIs730[UPN::TIR1::mTurquoise2]) and a pansensory TIR1 (otEx[Pift-20::TIR1::mRuby]) drivers. We note that Greer et al. observed that expression of the TGFβ receptor DAF-1 in the RIM/RIC interneurons is sufficient to rescue the dauer phenotype of daf-1 mutants [46], but these sufficiency experiments are conceptually distinct from our necessity experiments. (B) Quantification of dauer formation upon tissue-specific depletion of DAF-16/FoxO. (C) Quantification of dauer formation upon tissue-specific depletion of DAF-12/VDR. (D) Overall appearance of dauers with panneuronal depletion of DAF-16/FoxO and DAF-12/VDR is superficially similar to that of the background Daf-c strains, daf-2(e1370), and daf-7(e1372), respectively. Scale bars, 60 μm. The data underlying this figure can be found in S1 Data.

Fig 5. Neuronal removal of DAF-16/FoxO and DAF-12/VDR affects dauer-specific sensory receptor expression changes.

(A) sri-9::GFP (otIs732) expression induced in multiple neurons in dauer (red). The dauer-specific expression is largely dependent on the neuronal function of DAF-16/FoxO. Left, head images of control and auxin-treated dauers with panneuronal depletion of DAF-16/FoxO. Right, quantification of sri-9::GFP expression in each neuronal type. n = 22 (control) and n = 10 (auxin treated). Scale bars, 10 μm. (B) sra-25::GFP (sIs12199) expression induced in ADL neurons in dauer (red). The dauer-specific expression is not dependent on neuronal function of DAF-16/FoxO. Left, head images of control and auxin-treated dauers with panneuronal depletion of DAF-16/FoxO. Right, quantification of sra-25::GFP expression. n = 22 (control) and n = 10 (auxin-treated). Scale bars, 10 μm. (C) gcy-6::NLS::GFP (otIs586) expression is gained in AWC^OFF neuron in dauer. Above, images of AWC in control and auxin-treated dauers with panneuronal depletion of DAF-16/FoxO. Below, quantification of gcy-6::NLS::GFP expression. n = 12. Scale bars, 5 μm. (D) The dauer-specific expression of sri-9::GFP (otIs732) also depends on the neuronal function of DAF-12/VDR. Above, image of an auxin-treated dauer with panneuronal depletion of DAF-12/VDR. Below, quantification of sri-9::GFP expression in each neuronal type. Scale bar, 10 μm. The data underlying this figure can be found in S1 Data. AID, auxin-inducible degron.

Fig 6. Neuronal removal of DAF-16/FoxO controls dauer-specific expression changes of electrical synapse components.

(A) The dauer-specific expression gain of an inx-6 reporter allele (ot804) in AIB is affected upon continuous AIB-specific depletion of DAF-16/FoxO, as well as in animals where DAF-16/FoxO is depleted in AIB after dauer remodeling. (B) A che-7 reporter (otEx7112) expression is gained in BDU neurons in dauer. This dauer-specific che-7 expression is lost upon panneuronal depletion of DAF-16/FoxO in auxin-treated dauers. (C) A che-7 reporter (otEx7112) expression is gained in AIM neurons in dauer. This dauer-specific che-7 expression is lost upon panneuronal depletion of DAF-16/FoxO in auxin-treated dauers. (D) A che-7 reporter (otEx7112) expression is down-regulated in PHA neurons in dauer. This dauer-specific down-regulation of che-7 expression is affected upon panneuronal depletion of DAF-16/FoxO in auxin-treated dauers. (E) Expression of an inx-2 reporter allele (ot906) is down-regulated in multiple neurons in dauer. This dauer-specific down-regulation of inx-6 expression is unaffected upon panneuronal depletion of DAF-16/FoxO in auxin-treated dauers. In all the animals scored, AID-tagged DAF-16/FoxO was also completely depleted in the presence of auxin. The data underlying this figure can be found in S1 Data. AID, auxin-inducible degron.

Fig 7. DAF-16/FoxO and DAF-12/VDR act in the nervous system to affect dauer-specific locomotory behavior.

(A) A heat map of p-values associated with locomotory features of panneuronally DAF-16/FoxO-depleted dauers. (B) Representative locomotory features, which distinguish dauers with panneuronal depletion of DAF-16/FoxO from controls. (C) A heat map of p-values associated with locomotory features of panneuronally DAF-12/VDR-depleted dauers. (D) Representative locomotory features which distinguish dauers with panneuronal depletion of DAF-12/VDR from controls. Each circle represents the experimental mean of a single animal. Red lines indicate the mean of means, and rectangles indicate SEM. Wilcoxon rank-sum tests and false discovery rate q values for each comparison: n.s., nonsignificant, *q < 0.05, **q < 0.01, ***q < 0.001, ****q < 0.0001. The data underlying this figure can be found in S2 Data.

Fig 8. Control of pharyngeal pumping by DAF-16/FoxO and DAF-12/VDR.

(A) Pharyngeal silencing in dauer depends on cell-autonomous and nonautonomous DAF-16 function. Quantification of pharyngeal pumping upon DAF-16/FoxO depletion in pharyngeal muscle (myo-2), intestine (ges-1), all neurons (UPN), pharyngeal neurons (ehs-1), body wall muscle (myo-3), and hypodermis (dpy-7). (B) Pharyngeal constriction depends on cell-autonomous and nonautonomous DAF-16/FoxO function. DIC images of control dauers and dauers with DAF-16/FoxO depleted from pharyngeal muscle, intestine, neurons, and body wall muscle. (C) Quantification of changes in pharynx volume upon depletion of DAF-16 from pharyngeal muscle, intestine, and neurons. (D) Pharyngeal silencing in dauer depends on the neuronal function of DAF-12. Quantification of pharyngeal pumping upon depletion of DAF-12/VDR in pharyngeal muscle and neurons, as well as upon depletion of DAF-3/Smad in pharyngeal muscle. (E) Pharyngeal constriction in dauer depends on neuronal function of DAF-12/VDR. DIC images of control dauers and dauers with DAF-12/VDR depleted from pharyngeal muscle and neurons as well as with DAF-3/Smad depleted from pharyngeal muscle. The data underlying this figure can be found in S1 Data. AID, auxin-inducible degron; DIC, differential interference contrast.

Fig 9. DAF-16/FoxO activity is distributed across multiple pharyngeal neurons to silence pharyngeal pumping in the dauer stage.

(A) DAF-16/FoxO depletion in pharyngeal neurons of daf-2(e1370) dauers using ehs-1prom4-driven TIR1. In 20/20 animals observed, AID-tagged DAF-16 was not detectable in TIR1-expressing neurons (all pharyngeal neurons except I1 and I4) after auxin treatment. Scale bars, 20 μm. (B) Pharyngeal pumping rate after DAF-16/FoxO depletion in daf-2(e1370) dauers using ehs-1prom4-driven TIR1 (otEx7628 and otEx7629). Horizontal lines represent median for 36 animals per condition. *** indicates p < 0.001 in Mann–Whitney test. (C) DAF-16/FoxO depletion in glutamatergic pharyngeal neurons of daf-2(e1370) dauers using eat-4prom7-driven TIR1. In 20/20 animals observed, AID-tagged DAF-16 was not detectable in TIR1-expressing pharyngeal neurons (I5 and M3) after auxin treatment. Scale bars, 20 μm. (D) Pharyngeal pumping rate after DAF-16/FoxO depletion in daf-2(e1370) dauers using eat-4prom7-driven TIR1 (otEx7670 and otEx7671). Horizontal lines represent median for 36 animals per condition. **** indicates p < 0.0001 in Mann–Whitney test. (E) DAF-16/FoxO depletion in cholinergic pharyngeal neurons of daf-2(e1370) dauers using unc-17prom4-driven TIR1. In 20/20 animals, AID-tagged DAF-16 was not detectable in TIR1-expressing pharyngeal neurons (I1, M1, M2, M4, M5, and MC) after auxin treatment. Scale bars, 20 μm. (F) Pharyngeal pumping rate after DAF-16/FoxO depletion in daf-2(e1370) dauers using unc-17prom4-driven TIR1 (otEx7668 and otEx7669). Horizontal lines represent median for 36 animals per condition. ** indicates p < 0.01 in Mann–Whitney test. (G) Neurotransmitter identity of all 14 pharyngeal neurons. Shaded boxes for each promoter fragment indicate the neurons in which TIR1 expression is strong enough to completely deplete AID-tagged DAF-16/FoxO in the presence of auxin (20 animals analyzed per strain). (H and I) Pharyngeal pumping rate in daf-2(e1370) dauers transferred from ethanol (solvent) to auxin or from auxin to ethanol after dauer entry (dauers were transferred on day 3 after hatching, and pumping rate was measured on day 6 after hatching). DAF-16/FoxO was depleted in the intestine (Pges1::TIR1) or in no tissue (no TIR1 control) in the presence of auxin. Horizontal red lines represent median for 18 animals per condition. p-Values are for Dunn’s multiple comparisons test performed after 1-way ANOVA on ranks. The data underlying this figure can be found in S1 Data. AID, auxin-inducible degron.

Fig 10. Intestinal removal of DAF-16/FoxO and DAF-12 has cell autonomous and cell nonautonomous consequences.

(A) Oil Red O staining of control and intestinal DAF-16/FoxO-depleted strains. Note a decrease in the intensity of staining upon auxin treatment of the strain with intestinal depletion of DAF-16/FoxO. (B) Quantification of lipid droplet size for strains with intestinal and panneuronal depletion of DAF-16/FoxO reveals autonomous and nonautonomous regulation of dauer lipid metabolism. (C) Survival curves for control and auxin treatment conditions for the strains of the indicated genotypes. (C1-2) Long life span of daf-2(e1370) mutants requires DAF-16. (C3) Ubiquitous depletion of DAF-16/FoxO strongly reduces life span in a daf-2(e1370) background. (C4-5) Depletion of DAF-16/FoxO in pharyngeal or body wall muscles did not significantly reduce life span in a daf-2(e1370) background. (C6): Intestinal depletion of DAF-16/FoxO strongly reduces life span in a daf-2(e1370) background. N = 40–80 worms for each group. (D) Oil Red O staining of control and intestinal and panneuronal DAF-12/VDR-depleted strains. Panneuronal DAF-12/VDR depletion has a stronger effect on dauer lipid reserves. The data underlying this figure can be found in S1 Data. AID, auxin-inducible degron.

Fig 11. DAF-16/FoxO acts cell-autonomously in muscle remodeling.

(A) Depleting DAF-16/FoxO from body wall muscle results in fewer muscle arms than in control dauers, as visualized with the trIs30[him-4p::MB::YFP] reporter [60]. (B) Quantification of the number of muscle arms per muscle segment in control and auxin-treated dauers. The data underlying this figure can be found in S1 Data. AID, auxin-inducible degron.

Fig 12. DAF-16/FoxO activation via inhibition of insulin signaling is sufficient to cell nonautonomously reduce pharyngeal pumping.

(A) Schematic showing protein domains of wild-type DAF-2 (left) and DAF-2^DN, a dominant negative form of DAF-2/InsR in which the tyrosine kinase and carboxyl-terminal domains are replaced with the blue fluorescent protein EBFP2 (right). (B) Left panels: Expression of daf-2^DN::ebfp2 in pharyngeal muscle (top), panneuronal (middle), and intestinal (bottom) tissues. Right panels: Localization of endogenously mNeonGreen-tagged DAF-16 protein in the same animals shown in the corresponding panels on the left. Black arrowheads show nuclear localization of DAF-16 in intestinal tissue. Scale bars, 20 μm. (C) Pharyngeal pumping rate in day 1 adult animals expressing daf-2^DN::ebfp2 in pharyngeal muscle (otEx7549 and otEx7550), panneuronal (otEx7541 and otEx7542), and intestinal (otEx7546 and otEx7548) tissues in the endogenously mNeonGreen-tagged daf-16(ot853) genetic background. Red horizontal lines represent median for ≥35 animals per condition. **, ****, and ns indicate p < 0.01, p < 0.0001, and p ≥ 0.05 compared to the daf-16(ot853) control strain in Dunn’s multiple comparisons test performed after 1-way ANOVA on ranks. (D) Expression of GFP-tagged DAF-16(4A) in intestinal tissue of animals. Black arrowheads show strong nuclear localization of DAF-16(4A)::GFP. Asterisk indicates non-specific autofluorescence signal. (E) Pharyngeal pumping rate in day 1 adult wild-type (N2) strain and transgenic animals expressing daf-16(4A)::gfp in intestinal tissue (otEx7585-87). Red horizontal lines represent median for ≥16 animals per condition. **** indicates p < 0.0001 compared to wild-type strain in Dunn’s multiple comparisons test performed after 1-way ANOVA on ranks. The data underlying this figure can be found in S1 Data. AID, auxin-inducible degron.

Fig 13. Dauer-specific expression of neuronal reporters depends on terminal selector-type TFs.

(A) Dauer-specific expression of a che-7 fosmid reporter (otEx7112) in BDU neurons is abolished in unc-86(n846) mutants. (B) Dauer-specific expression of a che-7 fosmid reporter (otEx7112) in AIM neurons is abolished in unc-86(n846) mutants. (C) Dauer-specific expression of a sri-9 reporter (otIs732) in NSM neurons is abolished in unc-86(n846) mutants. The data underlying this figure can be found in S1 Data. TF, transcription factor; wt, wild-type.

Fig 14. Summary and conceptual models.

(A) Tabular summary of phenotypic outcomes of tissue-specific depletion of DAF-16/FoxO and DAF-12/VDR. (B) Models of inter-tissue signaling during dauer remodeling. (C) Terminal selectors control the expression of many immutable phenotypic features of a neuron, but are also enabling hormone-responsive TFs (DAF-12 and DAF-16) to induce dauer-specific target genes (e.g., inx-6 in AIB or gcy-6 in AWC). Future experiments will determine whether these interactions occur on the level of the cis-regulatory control elements of dauer specifically induced genes. One could envision that these control elements are tuned to not be responsive to either transcriptional input alone (terminal selector, dauer-specific TF), but requiring them both together. Since DAF-12 is commonly thought to act as a repressor in its unliganded form [25], these interactions may also be indirect. Not shown here is a variation of this model that may explain how DAF-16 may act to control the down-regulation of specific, dauer-repressed genes. Those could be under control of a terminal selector in non-dauer stage animal, but DAF-16 may antagonize its ability to activate these genes(s) upon entry into the dauer stage.