The evolutionarily conserved longevity determinants HCF-1 and SIR-2.1/SIRT1 collaborate to regulate DAF-16/FOXO

Abstract

The conserved DAF-16/FOXO transcription factors and SIR-2.1/SIRT1 deacetylases are critical for diverse biological processes, particularly longevity and stress response; and complex regulation of DAF-16/FOXO by SIR-2.1/SIRT1 is central to appropriate biological outcomes. Caenorhabditis elegans Host Cell Factor 1 (HCF-1) is a longevity determinant previously shown to act as a co-repressor of DAF-16. We report here that HCF-1 represents an integral player in the regulatory loop linking SIR-2.1/SIRT1 and DAF-16/FOXO in both worms and mammals. Genetic analyses showed that hcf-1 acts downstream of sir-2.1 to influence lifespan and oxidative stress response in C. elegans. Gene expression profiling revealed a striking 80% overlap between the DAF-16 target genes responsive to hcf-1 mutation and sir-2.1 overexpression. Subsequent GO-term analyses of HCF-1 and SIR-2.1-coregulated DAF-16 targets suggested that HCF-1 and SIR-2.1 together regulate specific aspects of DAF-16-mediated transcription particularly important for aging and stress responses. Analogous to its role in regulating DAF-16/SIR-2.1 target genes in C. elegans, the mammalian HCF-1 also repressed the expression of several FOXO/SIRT1 target genes. Protein-protein association studies demonstrated that SIR-2.1/SIRT1 and HCF-1 form protein complexes in worms and mammalian cells, highlighting the conservation of their regulatory relationship. Our findings uncover a conserved interaction between the key longevity determinants SIR-2.1/SIRT1 and HCF-1, and they provide new insights into the complex regulation of FOXO proteins.

Figure 1. hcf-1 acts downstream of sir-2.1 to modulate lifespan and oxidative stress response.

(A–B) Lifespans of synchronized adult populations of indicated genotypes. (A) Data pooled from four independent experiments are plotted. One of four sir-2.1(ok434) hcf-1(pk924) lines is shown. (See Table S1A). (B) Pooled data from three independent experiments are displayed. One of five hcf-1(pk924);pkIs1642[sir-2.1(O/E)] lines is shown (See Table S1B). (C–F) Oxidative stress response of adult worms. (C–D) Day one adult worms were exposed to 6 mM t-BOOH on plates and their survival monitored through time. The survival curves represent pooled data from two independent experiments. (E–F) Day two adult worms were exposed to 150 mM (E) or 200 mM (F) paraquat in M9 buffer and their survival monitored through time. Survival curves are generated using pooled data from two independent experiments (E) or data from one of two representative experiments (F). See Table S1A–S1F for statistics and Figure S1C–S1F for linear mixed model analysis plots. All lifespan and stress experiments were carried out at 25°C. Quantitative data and statistical analyses are displayed in Table S1A–S1D.

Figure 2. 14-3-3 are required for lifespan extension conferred by hcf-1(pk924) mutation.

(A–B) Worms were grown on vector, daf-16, ftt-2 (Ahringer - contains multiple stretches of identical sequences to par-5), par-5 (Ahringer - contains overlapping sequences with ftt-2) , ftt-2gs (gene specific RNAi targeting 3′ UTR of ftt-2), or par-5gs (gene-specific RNAi targeting 3′ UTR of par-5) from egglay until the end of life. The lifespan experiments were carried out at 20°C. Quantitative data and statistical analyses are included in Table S2A, S2B.

Figure 3. hcf-1 inactivation and sir-2.1 overexpression similarly affect a specific subset of daf-16 downstream target genes.

(A–D) Heat maps representing the expression patterns of differentially expressed genes identified by Significance Analysis of Microarrays (SAM) and Venn diagrams showing the overlap between different datasets. For heat maps, each column represents a biological replicate and each row is a gene. Pink = upregulated, Yellow = downregulated, Black = not changed. (A) Heat maps comparing hcf-1(-) and sir-2.1(O/E) arrays. Gene clusters are categorized as: (a) = Genes similarly changed in hcf-1(-) and sir-2.1(O/E) (866), (b) = genes oppositely changed in hcf-1(-) and sir-2.1(O/E) (98), (c) = genes uniquely changed in hcf-1(-) (66), (d) = genes uniquely changed in sir-2.1(O/E) (73). (B) Venn diagram summarizing overlap in (A). (C) Heat maps comparing hcf-1(-), sir-2.1(O/E), and daf-2(-) arrays. Genes are clustered as (a) = similarly expressed in all 3 profiles (693), (b) = similar in only hcf-1(-) and sir-2.1(O/E) (173), (c) = uniquely changed in sir-2.1(O/E) (130), (d) = similar in only sir-2.1(O/E) and daf-2(-) (26), (e) = uniquely changed in hcf-1(-) (140), (f) = similar in only hcf-1(-) and daf-2(-) (46), (g) = uniquely changed in daf-2(-) (1750) (See also Table S3). (D) Venn diagram summarizing overlaps in (C). (E) Most highly enriched GO terms (See also Table S4A, S4B) are summarized based on general biological process.

Figure 4. C. elegans HCF-1 physically interacts with SIR-2.1 and 14-3-3 proteins.

(A) Lysates from sir-2.1(-) (sir-2.1(ok434)), sir-2.1(O/E) (geIn3[sir-2.1(O/E)]), and sir-2.1(O/E);hcf-1(-) (hcf-1(ok559);geIn3[sir-2.1(O/E)]) worms were either immunoprecipitated using anti-HCF-1 antibody (left panel) or anti-SIR-2.1 antibody (right panel). The immunoprecipitated protein complexes were subsequently immunoblotted using anti-HCF-1, or anti-SIR-2.1 antibodies. (B) Lysates from ftt-2::mCherry or hcf-1::GFP;ftt-2::mCherry were immunoprecipitated with an anti-GFP antibody and blotted with anti-mCherry or anti-GFP antibodies. (C) Wild-type or HCF-1::GFP expressing worm lysates were immunoprecipitated using anti-GFP antibodies and blotted with anti-PAR-5 or anti-GFP antibody.

Figure 5. Mammalian HCF-1 and HCF-2 regulate the expression of FOXO target genes.

(A) INS-1 cells treated with HCF-1 or control siRNA. (B) INS-1 cells treated with HCF-2 or control siRNA. mRNA levels of Bim, Gadd45a, p27,and IGFBP1 were quantified using RT-qPCR and normalized to the level of β-actin. The mean normalized RNA level for each gene in sicontrol treated cells was set to 1. The data represented are pooled from three independent experiments and are represented as mean +/− SEM. * denotes a p-value<0.05 relative to sicontrol.

Figure 6. Mammalian HCF-1 physically associates with FOXO3 and SIRT1.

(A, B) HEK293T cells were transfected with plasmids encoding Flag-FOXO3 (A) or Flag-SIRT1 (B). Cell lysates were collected 48 hours later and incubated with anti-Flag-conjugated agarose beads. Immunoprecipitated protein complexes were analyzed by western blot using anti-HCF-1, anti-FOXO3 or anti-SIRT1 antibodies. HCF-1 is known to be proteolytically processed and is detected as multiple bands on SDS-PAGE . * denotes a non-specific band.

Figure 7. Conserved regulation of DAF-16/FOXO by HCF-1 and SIR-2.1/SIRT1.

We propose that C. elegans HCF-1 and SIR-2.1 coordinate to fine-tune the transcriptional activity of DAF-16 on a distinct subset of potential target genes. DAF-16 target genes responsive to the hcf-1/sir-2.1 pathway largely overlap with a small subset of IIS-regulated genes, and are specialized in longevity determination, cellular defense, and lipid/fatty acid/amino acid homeostasis. HCF-1 likely represses DAF-16 by forming a complex with SIR-2.1 and 14-3-3, and antagonizing their abilities to stimulate DAF-16. This functional relationship is highly conserved as mammalian HCF proteins also repress the expression of multiple FOXO/SIRT1 target genes and reside in protein complexes with SIRT1 and FOXO3. Our results highlight HCF proteins to be key components of the regulatory network linking SIR-2.1/SIRT1 and DAF-16/FOXO in diverse organisms.