The WD40 Repeat Protein WDR-23 Functions with the CUL4/DDB1 Ubiquitin Ligase To Regulate Nuclear Abundance and Activity of SKN-1 in Caenorhabditis elegans

C. elegans strains.The following strains were used: wild-type N2 Bristol, GR1373 eri-1(mg366)IV, CL2166 dvIs19[pAF15(Pgst-4::GFP::NLS)], VP537 eri-1(mg366)IV;dvIs19, VP579 wdr-23(tm1817)I;dvIs19, LD001 Is007[SKN-1::GFP; rol-6], TJ356 zIs356[DAF-16::GFP; rol-6], CF1038 daf-16(mu86)I, GR1309 daf-16(mgDf47)I; daf-2(e1370)III, and KU4 sek-1(km4). Unless otherwise noted, worms were cultured at 20°C using standard methods (8).

RNAi and genome-wide RNAi screening.RNAi was performed by feeding worms a strain of Escherichia coli [HT115(DE3)] that is engineered to transcribe double-stranded RNA (dsRNA) homologous to a target gene (28). Bacteria with plasmid pPD129.36 were used as a control for nonspecific RNAi effects. This control plasmid expresses 202 bases of dsRNA that are not homologous to any predicted C. elegans gene.

Genome-wide RNAi screening was performed by supplementing the C. elegans ORFeome RNAi feeding library (Open Biosystems, Huntsville, AL) with clones from the original genomic RNAi feeding library (Geneservice, Cambridge, United Kingdom). RNAi was performed as described previously (40) with minor modifications. dsRNA-producing bacteria were grown in Luria-Bertani broth containing a selective antibiotic, washed, and then transferred to 96-well plates. Each well was filled with 50 μl of liquid nematode growth medium (NGM) lacking peptone and containing 0.2% β-lactose (40). Thirty to 50 synchronized VP537 L1 larvae were added to each well. After 3 days, the wells were screened manually for the appearance of Pgst-4::GFP (green fluorescent protein) expression with a Zeiss Stemi SV11 microscope. The RNAi worms were then exposed to 38 μM juglone and rescreened 4 h later for reduced Pgst-4::GFP induction. All genes that gave an RNAi phenotype were rescreened three additional times. Those that scored positive in all three trials were considered regulators of gst-4. The identities of gst-4 regulators were confirmed by sequencing the dsRNA plasmids.

Fluorescence assays.Fluorescence measurements were made using a COPAS Biosort (Union Biometrica, Somerville, MA) on early-adult worms, with similar lengths measured as time of flight. Pgst-4::GFP and DAF-16::GFP images were captured with a Zeiss Stemi SV11 microscope (Chester, VA) fitted with a CCD-100 DAGE-MTI camera (Michigan City, IN). Images of SKN-1::GFP were captured with a LSM510-Meta confocal microscope and a Plan-Neofluar 40×/1.3-numerical-aperture oil objective lens (Carl Zeiss MicroImaging, Inc., Thornwood, NY). Quantification of nuclear SKN-1::GFP was performed as described previously (1, 2, 23, 55) using a Plan-Neofluar 10×/0.30-numerical-aperture objective lens.

Yeast two-hybrid assays.Bait and prey plasmids were generated by gateway cloning the reading frames for full-length WDR-23 and DDB-1 into pDEST32 and pDEST22, respectively (Invitrogen, Carlsbad, CA). Interaction was tested on synthetic complete agar that lacked leucine, tryptophan, and histidine and was supplemented with 3-amino-1,2,4-triazole according to the manufacturer's protocols (ProQuest two-hybrid system; Invitrogen). Novel WDR-23 interactors were identified by sequentially transforming Saccharomyces cerevisiae with bait (pDEST32-WDR-23) and a cDNA prey library (Invitrogen). Positive clones were sequenced and tested for self-activation as described previously (9).

A GST-WDR-23 fusion construct was generated by gateway cloning the reading frame of full-length WDR-23 into pDEST 27 (Invitrogen). V5 epitope-tagged SKN-1 was generated by cloning the open reading frame of SKN-1c into pcDNA3.1/nV5-DEST (Invitrogen). Cell culture, transfection, lysis, and Western analysis of V5-SKN-1 were carried out as described previously (11).

Quantitative PCR and Western analysis.Quantitative real-time PCR was used to measure mRNA levels as described previously (9). The housekeeping ribosomal protein gene rpl-2 was used as an internal control to calculate relative expression levels by the threshold cycle method. For Western analysis, worms were sonicated in 10 volumes of buffer (50 mM Tris, pH 7.5, 50 mM NaCl, 1% sodium dodecyl sulfate, and protease inhibitor tablet [Roche, Indianapolis, IN]) and centrifuged at 16,100 × g for 5 min. Total protein was measured with a bicinchoninic acid assay (Pierce Biotechnology, Rockford, IL). Samples were diluted in NuPage LDS buffer (Invitrogen, Carlsbad, CA) and subjected to Western blot analysis using monoclonal anti-SKN-1 antibodies (7). The amount of protein transferred was determined using Ponceau S (Sigma Aldrich, St. Louis, MO) staining. Relative Western blot band and total-protein stain intensities were measured using ImageJ 1.32j software (NIH).

Stress resistance and longevity.Young adult eri-1 worms were exposed to stress-inducing compounds in liquid NGM, washed, and placed on NGM agar plates for 24 h. To assess longevity, young adult eri-1 worms were transferred to NGM agar plates containing 100 μg/ml of 5-fluorodeoxyuridine (13). Worms were considered alive if they displayed any muscle contractions in response to repeated prodding with an eyelash.

Phylogenic analysis.Basic local alignment search tool (Blast) on the National Center for Biotechnology Information website (http://www.ncbi.nlm.nih.gov/ ) was used to find the closest homologues of WDR-23 in the genomes of Homo sapiens (human), Drosophila melanogaster (fly), C. elegans (worm), Arabidopsis thaliana (plant), and Neurospora crassa (fungus). These proteins were aligned by ClustalW with Vector NTI Advance 10 (Invitrogen), and MEGA 3.1 software (36) was used to make phylogenetic trees by the neighbor-joining method with Poisson-corrected evolutionary distances (45). The branches were then tested for statistical significance by bootstrapping with 1,000 replicates.

Statistical analysis.Statistical significance was determined using a one-sample t test when normalized means were compared to control values of 1.0, a one-way analysis of variance with Dunnet or Tukey's post hoc test when three or more means were compared, Student's t test when two means were compared, and a log rank test when survival curves were compared. P values of <0.05 were taken to indicate statistical significance.

SKN-1 mediates stress-induced transcription of gst-4.Glutathione S-transferases (GSTs) are phase II detoxification enzymes that catalyze the conjugation of glutathione to electrophiles. Expression of the C. elegans gene gst-4 is induced by the redox cycling compound paraquat (54), the napthoquinone juglone (27), heavy metals (50), and acrylamide (15).

We monitored gst-4 expression using a worm strain expressing a transcriptional GFP reporter (Pgst-4::GFP) (41). Pgst-4::GFP was induced significantly by juglone, hydrogen peroxide, and paraquat. Induction by juglone was more than threefold greater than induction by either paraquat or hydrogen peroxide (Fig. 1). Quinones like juglone generate reactive oxygen species and are also electrophiles that form adducts with diverse macromolecules (56).

• Open in new tab
• Download powerpoint

FIG. 1.

gst-4 transcription is regulated by SKN-1-dependent pathways that are activated by stress. Shown are relative Pgst-4::GFP fluorescence (A) and representative fluorescence micrographs (B) of worms fed control or skn-1 dsRNA and exposed to peroxide (5 mM for 20 min), paraquat (35 mM for 1 h), or juglone (38 μM for 1 h). Fluorescence was measured 6 to 8 h after exposure. Individual data points are plotted to illustrate variability. Means are denoted by lines (n = 48 to 152 worms). All three stressors induced significant (P < 0.001) Pgst-4::GFP fluorescence in control(RNAi) worms. RNAi knockdown of skn-1 significantly (P < 0.01) suppressed Pgst-4::GFP induction by all three stressors.

To identify novel regulators of gst-4, we performed a genome-wide RNAi screen. Pgst-4::GFP worms were crossed with worms harboring a loss-of-function allele (mg366) of eri-1 that confers RNAi hypersensitivity (32). L1 larvae were fed dsRNA-producing E. coli from a bacterial feeding library covering ∼96% of the 19,427 predicted C. elegans genes (28, 51). After 3 days, the worms were screened for constitutive expression of Pgst-4::GFP in the absence of stress. The dsRNA-fed worms were then exposed to 38 μM juglone and rescreened 4 h later for reduced induction of Pgst-4::GFP.

RNAi knockdown of 14 genes, including skn-1, reduced Pgst-4::GFP in the presence of juglone. As shown in Fig. 1, skn-1 RNAi inhibited nearly all juglone-, hydrogen peroxide-, and paraquat-induced Pgst-4::GFP fluorescence, confirming and extending recent findings that SKN-1 mediates the transcription of gst-4 (15, 27).

The proteasome, CUL-4/DDB-1, and the WD40 repeat-containing protein WDR-23 function to suppress gst-4 expression.RNAi of 14 genes consistently induced Pgst-4::GFP fluorescence. As shown in Fig. 2A and B, RNAi of three 19S proteasome subunit-encoding genes, rpn-2, rpn-7, and rpn-8, induced Pgst-4::GFP expression between two- and fivefold. Knockdown of ddb-1 induced Pgst-4::GFP by over 2.7-fold. ddb-1 encodes the C. elegans orthologue of the mammalian DDB1.

• Open in new tab
• Download powerpoint

FIG. 2.

gst-4 transcription is repressed by the proteasome, DDB-1, CUL-4, and WDR-23. Shown are relative Pgst-4::GFP fluorescence (A) and representative fluorescence micrographs (B) of worms fed bacteria producing dsRNA to proteasome components (rpn-2, rpn-7, and rpn-8), ddb-1, cul-4, or wdr-23. RNAi or juglone exposure (Fig. 1) induced Pgst-4::GFP expression in multiple tissues. The strongest induction was observed in the intestine. (A) Individual data points are plotted to illustrate variability. Means are denoted by lines (n = 206 to 649 worms) and were all significantly (P < 0.001) different from a normalized value of 1.0 for control dsRNA-fed worms. (C) Relative gst-4 and gst-30 mRNA levels in worms fed bacteria producing dsRNA to ddb-1, cul-4, or wdr-23. The values are means plus standard errors (n = 4 populations of 20 to 30 worms). *, P < 0.05; **, P < 0.01; and ***, P < 0.001 compared to a normalized value of 1.0 for control dsRNA-fed worms.

DDB1 and CUL4 form a complex that mediates ubiquitinylation of diverse substrates (19, 37). In C. elegans, CUL-4/DDB-1 targets the replication-licensing factor CDT-1 for degradation (33). C. elegans RNAi screens have predicted false-negative rates of 10 to 30% (52). While CUL-4 was not detected in our screen, the identification of DDB-1 as a regulator of gst-4 transcription suggested that CUL-4 might also play a role. As shown in Fig. 2A and B, feeding worms cul-4 dsRNA-expressing bacteria induced an ∼1.3-fold increase in Pgst-4::GFP fluorescence.

The strongest repressor of Pgst-4::GFP identified in our screen was WDR-23, an uncharacterized WD40 repeat-containing protein. Knockdown of wdr-23 induced a nearly 15-fold increase in Pgst-4::GFP fluorescence (Fig. 2A and B).

Because our screen was based on transgene expression, we used real-time reverse transcription-PCR to determine if endogenous phase II detoxification gene mRNA levels were also altered by RNAi of the novel gst-4 regulators that we identified. Knockdown of ddb-1, cul-4, and wdr-23 increased gst-4 mRNA levels by 21.6-, 4.6-, and 16.6-fold, respectively (Fig. 2C). We also tested gst-30, encoding a second GST (15) that contains three SKN-1-binding motifs in its 5′ regulatory region. Knockdown of ddb-1, cul-4, and wdr-23 increased gst-30 mRNA expression by 10.2-, 1.8-, and 163-fold, respectively (Fig. 2C). Taken together, these results demonstrate that the proteasome, CUL-4/DDB-1, and the WD40 repeat-containing protein WDR-23 suppress expression of phase II detoxification genes.

CUL-4/DDB-1 and the proteasome function with WDR-23 in a common pathway.Adaptor proteins interact with CUL4/DDB1 complexes to mediate substrate specificity for ubiquitinylation and degradation (3, 18, 37). WDR23 is the mammalian orthologue of C. elegans WDR-23 (Fig. 3A and B) and was recently shown to interact with the mammalian CUL4/DDB1 complex (3, 25, 37).

• Open in new tab
• Download powerpoint

FIG. 3.

WDR-23 is evolutionarily conserved. (A) A phylogenetic tree was constructed using the neighbor-joining method with Poisson correction. BLAST searches were used to find the three closest homologues of WDR-23 in H. sapiens (human), D. melanogaster (fly), C. elegans (worm), A. thaliana (plant), and N. crassa (fungus). Human and C. elegans proteins are listed by their names; all other proteins are listed by their GenBank accession numbers. The Arabidopsis protein NP_176316 was the most divergent protein and was used to root the tree. G protein β polypeptide 3, one of the best characterized WD40-repeat proteins, is included for reference. The scale bar represents the fraction of amino acids replaced per site. (B) ClustalW alignment of C. elegans (Ce) full-length WDR-23a and WDR-23b with human (h) WDR23.1 and WDR23.2. Shading indicates conserved amino acids. Seven WD40 repeats are highlighted with solid boxes. The DWD box is highlighted with a dashed box. Asterisks indicate cysteine adjacent to a basic amino acid residue.

To determine whether the proteasome and the CUL-4/DDB-1 complex regulate gst-4 in a common pathway with WDR-23, we created C. elegans strain VP579, which carries the Pgst-4::GFP transgene and a wdr-23 deletion allele, tm1817, that is predicted to encode a nonfunctional protein (Fig. 4A). Worms homozygous for the tm1817 allele constitutively express high levels of Pgst-4::GFP (Fig. 4B) (18- ± 0.3-fold induction relative to the wild type; P < 0.0001) similar to that observed in animals fed wdr-23 dsRNA (Fig. 2A and B). No significant (P > 0.05) additive induction of Pgst-4::GFP expression was observed in wdr-23 deletion mutants fed bacteria producing dsRNA for rpn-2, rpn-7, rpn-8, or cul-4 (Fig. 4B), suggesting that the proteasome and CUL-4 repress gst-4 expression exclusively through pathways that require WDR-23.

• Open in new tab
• Download powerpoint

FIG. 4.

CUL-4, DDB-1, and the proteasome function with WDR-23. (A) The wdr-23 gene is predicted to encode two splice variants that differ at their amino termini (58). The tm1817 allele is a 635-bp deletion that completely removes exon 6 of the longer splice variant, WDR-23a, which is equivalent to exon 5 of the shorter splice variant, WDR-23b. This deletion is predicted to cause a frameshift and to encode a nonfunctional protein. (B) Effect of RNAi of rpn-2, rpn-7, rpn-8, ddb-1, cul-4, or F28D1.1 on Pgst-4::GFP fluorescence in wdr-23(tm1817) deletion mutants. Fluorescence is reported relative to that of deletion mutants fed control dsRNA bacteria. Individual data points are plotted to illustrate variability. Means are denoted by lines (n = 94 to 598 worms). ***, P < 0.001 greater than a normalized value of 1.0 for control dsRNA-fed wdr-23(tm1817) worms.

RNAi knockdown of ddb-1 caused a small (∼13%) but statistically significant (P < 0.001) increase in Pgst-4::GFP expression (Fig. 4B). However, the increase was ∼7% of that induced by RNAi of ddb-1 in wild-type worms (Fig. 2A). These data suggest that DDB-1 represses gst-4 expression predominantly, but perhaps not exclusively, through pathways that require WDR-23.

A possible interpretation of the data shown in Fig. 4 is that deletion of wdr-23 induces maximum gst-4 expression that cannot be increased further by independent signaling pathways. To test this possibility, we fed wdr-23 deletion mutants bacteria producing dsRNA for F28D1.1, which encodes a second WD40 repeat protein identified in our RNAi screen. Pgst-4::GFP expression in these worms increased 1.9- ± 0.02-fold (Fig. 4B) above that induced by loss of WDR-23 function alone. F28D1.1 is a homologue of the yeast protein Utp7, which functions in ribosome biogenesis, kinetochore organization, and chromosome segregation (12, 14, 26). Although the mechanism by which F28D1.1 regulates gst-4 expression is unknown, these results demonstrate that other pathways regulating gst-4 expression exist and that Pgst-4::GFP expression is not saturated in wdr-23 deletion mutants. Importantly, these results further support our hypothesis that the proteasome, DDB-1, and CUL-4 regulate gst-4 transcription in a common pathway with WDR-23.

WDR-23 interacts with DDB-1 in nuclei.The majority of the WD40 proteins that interact with CUL4/DDB1, including human WDR23, contain a conserved 16-amino-acid DWD box motif (DDB1-binding and WD40 repeat; also known as WDXR, DxR, and CDW) (3, 18, 37) (Fig. 5A). Mutations in the DWD box disrupt DDB1 binding (3, 17, 38). The DWD box is conserved in C. elegans WDR-23 (Fig. 3B and 5A), suggesting that there may be a direct interaction between DDB-1 and WDR-23. Yeast two-hybrid analysis confirmed this interaction (Fig. 5B).

• Open in new tab
• Download powerpoint

FIG. 5.

WDR-23 is expressed in cell nuclei and interacts with DDB-1. (A) Alignment of DWD boxes in human WDR23 and worm WDR-23 with the DWD box motif (h, hydrophobic; s, small; x, any residue). (B) Photograph of a histidine dropout- and 40 mM 3-amino-1,2,4-triazole-containing yeast plate streaked with yeast colonies (strain MaV203) coexpressing the GAL4 DNA binding domain (DB) fused to WDR-23 and the GAL4 activation domain (AD) alone; DB alone and AD fused to DDB-1; DB fused to WDR-23 and AD fused to DDB-1; or DB fused to a WDR-23 DWD box mutant (R389A) and AD fused to DDB-1. (C) Paired fluorescence (top) and differential interference contrast (bottom) micrographs of worms expressing the complete full-length WDR-23 fused to GFP. The WDR-23::GFP reporter is expressed in the nuclei of cells in the hypodermis, intestine, and head. Scale bars, 20 μm.

In plants and mammals, the interaction between WD40 repeat proteins and DDB1 is disrupted by mutation of an arginine residue to alanine at position 16 in the DWD box (3, 17, 38). Mutation of this arginine in C. elegans WDR-23 prevents it from interacting with DDB-1 (Fig. 5B).

The DDB1/CUL4 ubiquitin ligase complex has well-defined roles in regulating DNA damage repair, DNA replication, and chromatin modifications in the nuclei of eukaryotes (19, 37). In C. elegans, ddb-1 and cul-4 mRNAs are expressed ubiquitously and DDB-1 is localized to cell nuclei (33, 59). The data shown in Fig. 4 and 5 indicate that WDR-23 functionally and physically interacts with CUL-4/DDB-1 ubiquitin ligase complexes to regulate gst-4 transcription. If this is correct, then WDR-23 should be expressed with DDB-1 in cell nuclei. To determine where WDR-23 is expressed, we generated a transgenic worm strain expressing full-length WDR-23 fused to GFP that was under the control of the wdr-23 promoter. As shown in Fig. 5C, WDR-23::GFP is expressed in nuclei of the hypodermis, intestine, and head neurons.

WDR-23 interacts with and regulates nuclear accumulation of SKN-1.To further define the mechanism by which WDR-23 regulates gst-4 transcription, we performed a yeast two-hybrid screen for novel interacting proteins. The full-length WDR-23 protein was used as bait to screen 6.3 × 10⁵ cDNA clones. We isolated 17 strongly positive clones, and all encoded SKN-1 (Fig. 6A). SKN-1 is expressed as three splice variants that differ in the lengths of their amino termini (WormBase; California Institute of Technology, Cold Spring Harbor Laboratory, Washington University—St. Louis, and The Wellcome Trust Sanger Institute). All of the WDR-23-interacting clones included the entire open reading frames of SKN-1c and various lengths of full-length SKN-1a. GST affinity assays confirmed the interaction between WDR-23 and SKN-1c (Fig. 6A).

• Open in new tab
• Download powerpoint

FIG. 6.

WDR-23 interacts with and regulates nuclear accumulation of SKN-1. (A) (Top) Photograph of a histidine dropout- and 40 mM 3-amino-1,2,4-triazole-containing plate streaked with yeast colonies (strain MaV203) coexpressing the complete reading frame of SKN-1c fused to the GAL4 activation domain (AD) and either the GAL4 DNA binding domain alone (DB) or the GAL4 DNA binding domain fused to full-length WDR-23 (DB::WDR-23). SKN-1- and WDR-23-coexpressing yeast cells also tested positive for uracil dropout, LacZ, and 5-fluoorotic acid phenotypes (not shown). (Bottom) Anti-V5 Western blot of lysates from CHO cells coexpressing V5-tagged SKN-1c and either GST alone or GST fused to full-length WDR-23. (B) Relative Pgst-4::GFP fluorescence (left) and representative fluorescence micrographs (right) of wdr-23(tm1817) deletion worms fed skn-1 or GFP dsRNA-producing bacteria. Individual data points are plotted to illustrate variability. Means are denoted by lines (n = 37 to 248 worms) and were significantly (P < 0.001) different from a normalized value of 1.0 for control dsRNA-fed worms. (C) Representative fluorescence micrographs of SKN-1::GFP-expressing worms fed control, ddb-1, cul4, or wdr-23 dsRNA-producing bacteria. The arrowheads mark the locations of nuclear SKN-1::GFP. Scale bar, 20 μm. (D) Quantification of nuclear SKN-1::GFP. Low, no visible nuclear GFP; medium, nuclear GFP visible only in anterior and/or posterior intestinal cells; high, nuclear GFP visible throughout the intestine.

As described previously (15, 27) and shown in Fig. 1, SKN-1 is required for gst-4 transcription. To determine if SKN-1 mediates gst-4 transcription in wdr-23 deletion mutants, we measured Pgst-4::GFP expression in VP579 worms that were fed bacteria producing control, skn-1, or GFP dsRNA. As shown in Fig. 6B (left), Pgst-4::GFP expression was reduced nearly 80% by skn-1 RNAi, demonstrating that the transcription factor mediates gst-4 transcription downstream from WDR-23. The remaining Pgst-4::GFP expression was largely limited to the pharynx. Similar residual Pgst-4::GFP expression also remained in worms fed GFP dsRNA and may represent RNAi resistance in this tissue (Fig. 6B, right).

The nuclear localization of DDB-1 (33) and WDR-23 (Fig. 5C) and the strong SKN-1/WDR-23 interaction suggest that WDR-23 and the ubiquitin ligase function to regulate SKN-1 accumulation in the nucleus. To test this possibility, we used a transgenic worm strain expressing a region of SKN-1 that includes SKN-1b and SKN-1c fused to GFP (1). Under nonstressed conditions, SKN-1::GFP is undetectable except for the two ASI sensory neurons in the head (1, 5). Exposure to stressors or inhibition of the proteasome causes SKN-1::GFP to accumulate in the nuclei of intestinal cells (1, 27). As shown in Fig. 6C and D, RNAi of wdr-23, ddb-1, and cul-4 caused accumulation of SKN-1::GFP in intestinal nuclei.

ASI sensory neurons mediate dietary-restriction-induced longevity in C. elegans. Increased life span requires SKN-1b, which is constitutively localized to the nuclei of these neurons (5). SKN-1b is the shortest SKN-1 variant and includes an exon not present in SKN-1a or SKN-1c. Interestingly, we did not detect SKN-1b in our WDR-23 yeast two-hybrid screen. This suggests that SKN-1b may not interact with WDR-23, which in turn may account for its constitutive localization to ASI nuclei.

Nuclear accumulation of SKN-1::GFP could simply reflect increases in skn-1 mRNA levels. To test this possibility, we performed real-time PCR on wild-type worms that were fed bacteria producing dsRNA for wdr-23, ddb-1, or cul-4. As shown in Fig. 7A, skn-1 mRNA levels were not significantly (P > 0.05) increased by knockdown of these genes, demonstrating that WDR-23 and the CUL-4/DDB-1 complex regulate nuclear accumulation of SKN-1 protein independently of skn-1 mRNA levels.

• Open in new tab
• Download powerpoint

FIG. 7.

WDR-23 regulates SKN-1 protein levels. (A) skn-1 mRNA levels in worms fed control, ddb-1, cul4, or wdr-23 dsRNA-producing bacteria. The values are means plus standard errors (n = 4 populations of 20 to 30 worms). (B) Ponceau S total-protein stain (left) and anti-SKN-1 immunoblot (right) of lysates from worms fed bacteria producing control or skn-1 dsRNA. (C) Ponceau S total-protein stain (left) and anti-SKN-1 immunoblot of lysates from N2 and wdr-23(tm1817) worms. Quantification of total protein and SKN-1 levels is shown on the far right. The values are means plus standard errors (n = 4 populations of ∼1,000 worms). **, P < 0.01 compared to N2.

To determine if nuclear accumulation of SKN-1 reflects increased protein abundance, we performed Western analysis of N2 and wdr-23(tm1817) worms using an anti-SKN-1 monoclonal antibody (7). This antibody detects a 67.5-kDa protein that is sensitive to skn-1 RNAi (Fig. 7B). As shown in Fig. 7C, wdr-23 loss of function increased SKN-1 protein levels by 2.8-fold, demonstrating that WDR-23 regulates SKN-1 protein abundance.

WDR-23 functions independently of DAF-16.Like SKN-1, the FOXO transcription factor DAF-16 accumulates in the cell nucleus in response to stressors (29, 35), including juglone (Fig. 8A). In addition, recent studies have demonstrated that gst-4 expression is regulated in part by DAF-16 (55). These results suggest that WDR-23 may play a role in regulating DAF-16. To test this possibility, we fed DAF-16::GFP-expressing worms wdr-23 dsRNA bacteria. Figure 8A demonstrates that loss of wdr-23 function does not promote nuclear accumulation of DAF-16. In addition, Pgst-4::GFP expression in wdr-23 deletion mutants was not inhibited by DAF-16 RNAi (Fig. 8B), which is sufficient to eliminate DAF-16::GFP expression (Fig. 8C). Finally, RNAi of wdr-23 increased gst-4 and gst-30 mRNA levels in a daf-16 deletion mutant (mu86) (Fig. 8D), verifying that DAF-16 is not required for target gene activation. Taken together, the data in Fig. 8 demonstrate that DAF-16 is not regulated by WDR-23 and indicate that loss of WDR-23 function induces gst-4 transcription specifically through SKN-1.

• Open in new tab
• Download powerpoint

FIG. 8.

WDR-23 does not regulate DAF-16. (A) Numbers of nuclei with visible nuclear DAF-16::GFP expression (left) and representative fluorescence micrographs (right) of DAF-16::GFP-expressing worms treated with juglone (38 μM for 1 h) or fed bacteria producing wdr-23 dsRNA. Nuclear DAF-16::GFP was assessed in cells located between the pharynx and vulva. The values are means plus standard errors (n = 5 worms). **, P < 0.01 compared to control worms. (B) Relative Pgst-4::GFP fluorescence (left) and representative fluorescence micrographs (right) of wdr-23(tm1817) deletion mutants fed daf-16 dsRNA-producing bacteria. Individual data points are plotted to illustrate variability. Means are denoted by lines (n = 80 to 106 worms). (C) Representative fluorescence micrographs of DAF-16::GFP-expressing worms treated with control or daf-16 RNAi. Knockdown of DAF-16::GFP verified the efficiency of daf-16 RNAi. (D) Relative gst-4 and gst-30 mRNA levels in daf-16(mu86) worms fed bacteria producing wdr-23 dsRNA. The values are means plus standard errors (n = 4 populations of 20 to 30 worms). ***, P < 0.001.

WDR-23 may function downstream from GSK-3, DAF-2, and SEK-1.Previous studies have identified protein kinase pathways that phosphorylate SKN-1 and alter its nuclear accumulation (2, 23, 55). However, it is unclear how SKN-1 phosphorylation contributes to nuclear accumulation and activation during stress. The data in Fig. 2 and 4 to 7 suggest that WDR-23 represses SKN-1 nuclear accumulation by recruiting it to the CUL-4/DDB-1 ubiquitin ligase complex and demonstrate that disruption of this complex is by itself sufficient to promote nuclear accumulation and activation. To determine if stressors activate gst-4 transcription via WDR-23 regulation, we exposed VP579 wdr-23 loss-of-function worms to peroxide, paraquat, and juglone and measured Pgst-4::GFP fluorescence. As shown in Fig. 9A and B, none of these stressors, including the robust gst-4 inducer juglone (Fig. 1), increased gst-4 transcription any more than wdr-23 loss of function alone. These results suggest that stressors activate gst-4 transcription through a mechanism that disrupts WDR-23 repression of SKN-1. In contrast, the effect of juglone on gst-4 expression was additive with RNAi of F28D1.1 (2.4- ± 0.05-fold induction; P < 0.001), further supporting our conclusion that this WD40 protein functions in a pathway distinct from that of WDR-23.

• Open in new tab
• Download powerpoint

FIG. 9.

Stress, GSK-3, DAF-2, and SEK-1 may function upstream from WDR-23. Shown are relative Pgst-4::GFP fluorescence (A) and representative fluorescence micrographs (B) of wdr-23(tm1817) deletion mutants exposed to peroxide (5 mM for 20 min), paraquat (35 mM for 1 h), or juglone (38 μM for 1 h). Fluorescence was measured 6 to 8 h after exposure. Individual data points are plotted to illustrate variability. Means are denoted by lines (n = 137 to 269 worms). (C) Relative gst-4 mRNA levels in N2 and wdr-23(tm1817) worms fed bacteria producing control or gsk-3 dsRNA. (D) Relative gst-4 mRNA levels in N2 and daf-16(mgDf47);daf-2(e1370) worms fed bacteria producing control or wdr-23 dsRNA. (E) Relative gst-4 and gst-30 mRNA levels in N2 and sek-1(km4) worms fed bacteria producing control or wdr-23 dsRNA. The values in panels C, D, and E are means plus standard errors (n = 3 or 4 populations of 20 to 30 worms).

Phosphorylation by GSK-3 and kinases downstream from the DAF-2 insulin-like pathway (AKT-1, AKT-2, and SGK-1) inhibit SKN-1 nuclear accumulation and activation (2, 55). To determine if these kinases repress SKN-1 in a pathway with WDR-23, we performed epistasis analysis using real-time PCR quantification of gst-4 mRNA. gsk-3 RNAi increased gst-4 mRNA by 3.3-fold in N2 worms but had no additive effect in wdr-23 loss-of-function animals (Fig. 9C). As shown by Tullet et al. (55), daf-2 loss of function increases gst-4 expression independently of daf-16 (Fig. 9D). However, daf-2 loss of function had no additive effect on gst-4 induction in wdr-23(RNAi) worms (Fig. 9D). These data suggest that GSK-3 and insulin-like signaling may function together with WDR-23 to repress SKN-1 activity.

Nuclear accumulation and activation of SKN-1 require phosphorylation by the conserved p38 MAPK pathway (23). Oxidative stress activates the p38 MAPK PMK-1, which phosphorylates SKN-1 at residues required for its nuclear accumulation and activation (23). The MAPK kinase SEK-1 functions immediately upstream from PMK-1 and is required for oxidative-stress-induced activation of PMK-1 and SKN-1 (23). To determine if this pathway functions upstream from WDR-23, we quantified the effects of sek-1 loss of function on gst-4 and gst-30 induction in wdr-23(RNAi) worms. As shown in Fig. 9E, sek-1 loss of function did not suppress wdr-23(RNAi) induction of gst-4 or gst-30, suggesting that WDR-23 regulates SKN-1 downstream from the p38 MAPK pathway.

WDR-23 regulates stress resistance, longevity, and growth.Loss of skn-1 function reduces stress tolerance and life span (1, 2, 55), while overexpression of SKN-1 enhances stress tolerance and increases longevity (2, 55). Given that WDR-23 strongly represses SKN-1 activity (Fig. 2, 6, and 7), we predicted that loss of wdr-23 function would enhance stress tolerance and extend life span. As shown in Fig. 10A, wdr-23 RNAi increased the survival of worms exposed to peroxide and juglone by 1.2- and 2.6-fold (P < 0.05), respectively. Knockdown of wdr-23 by RNAi also significantly extended the median and mean life spans by 4.0 and 2.4 days, respectively (Fig. 10B). A similar extension of life span by wdr-23 knockdown has been observed in a recent RNAi screen for longevity genes (10).

• Open in new tab
• Download powerpoint

FIG. 10.

WDR-23 regulates stress resistance and longevity. (A) Percent survival of control and wdr-23 dsRNA-fed worms exposed to peroxide (5 mM for 30 min) or juglone (100 μM for 1 h). Survival was measured 24 h after exposure. The values are means plus standard errors (n = 8 populations of 22 to 143 worms). *, P < 0.05, and ***, P < 0.001 compared to control worms. (B) Longevity curves for control and wdr-23(RNAi) worms. Median longevities were 20 and 24 days for control and wdr-23 dsRNA-fed worms, respectively, and were significantly (P < 0.0001) different (n = 251 to 261 total worms from two experiments). A total of 26 control and 11 wdr-23(RNAi) worms were censored because of intestinal protrusion through the vulva. Day zero corresponds to worms at the L1 stage. Experiments were performed in RNAi-hypersensitive eri-1 mutant worms. (C) Loss of WDR-23 slows growth in a SKN-1-dependent manner. Relative worm length (time of flight) was measured with a COPAS Biosort beginning with eggs at day 0 and synchronized L1 larvae at day 1. The values are means plus standard errors (n = 99 to 803 worms). *, P < 0.001.

Given that WDR-23 functions normally to repress stress resistance and longevity, we examined the effect of loss of wdr-23 function on growth and development under nonstress conditions. As shown in Fig. 10C, wdr-23(tm1817) worms have impaired growth relative to N2 animals. This effect was partially suppressed by skn-1 RNAi (Fig. 10C), suggesting that in nonstressed animals, repression of SKN-1 activity by WDR-23 is required for normal growth and development.

Does WDR23 regulate Nrf2?Transcription of phase II detoxification genes in mammals is predominantly activated by the NF-E2-related factor (Nrf) basic-leucine zipper protein Nrf2 (34). Nrf2 protects mammals from oxidative stress and age-related diseases, such as cancer, neurodegeneration, and chronic inflammation (34, 39). In addition, aberrant Nrf2 activity has recently been implicated in the development of multidrug resistance in cancer cells (16, 53, 57). These observations have made pharmacological modulation of Nrf2 an important approach for treating several human diseases (58). Detailed molecular characterization of all Nrf2 regulatory pathways is thus essential for gaining a comprehensive understanding of its roles in disease and for developing new therapeutic strategies and targets.

It is interesting to speculate that human WDR23 may perform a role similar to that of C. elegans WDR-23. Nuclear accumulation and activity of Nrf2 in Drosophila and vertebrate cells are controlled by Keap1, which functions as an actin binding protein and as a substrate recognition subunit for the Cullin ubiquitin ligase CUL3 (34, 49). Keap1 functions with CUL3 ubiquitin ligases in the cytoplasm (34, 49) and may also transiently shuttle into the nucleus (46).

Studies with Nrf2 mutants that do not bind to Keap1 demonstrate that it is targeted for degradation by at least one Keap1-independent mechanism (43, 47). C. elegans SKN-1 and mammalian Nrf2 share regions of high homology, including one that is postulated to mediate Keap1-independent ubiquitinylation in the nucleus (43). Humans have one WDR23 orthologue (Fig. 3), and it interacts with CUL4/DDB1 ubiquitin ligases (3). It will be critical to determine if WDR23 or other nuclear CUL4/DDB1 substrate adaptors regulate Nrf2 in mammals, which would allow temporal and spatial specialization of Nrf2 regulation.