SEARCH

SEARCH BY CITATION

Keywords:

  • aging;
  • Caenorhabditis elegans ;
  • HSF-1;
  • insulin/IGF-1 signaling;
  • longevity;
  • mRNA translation;
  • S6 kinase;
  • target of rapamycin

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Target of rapamycin (TOR) signaling is an evolutionarily well-conserved pathway that regulates various physiologic processes, including aging and metabolism. One of the key downstream components of TOR signaling is ribosomal protein S6 kinase (S6K) whose inhibition extends the lifespan of yeast, Caenorhabditis elegans, Drosophila, and mice. Here, we demonstrate that the activation of heat shock factor 1 (HSF-1), a crucial longevity transcription factor known to act downstream of the insulin/IGF-1 signaling (IIS) pathway, mediates the prolonged lifespan conferred by mutations in C. elegans S6K (rsks-1). We found that hsf-1 is required for the longevity caused by down-regulation of components in TOR signaling pathways, including TOR and S6K. The induction of a small heat-shock protein hsp-16, a transcriptional target of HSF-1, mediates the long lifespan of rsks-1 mutants. Moreover, we show that synergistic activation of HSF-1 is required for the further enhanced longevity caused by simultaneous down-regulation of TOR and IIS pathways. Our findings suggest that HSF-1 acts as an essential longevity factor that intersects both IIS and TOR signaling pathways.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

For the last two decades, numerous genes and pathways that influence aging have been discovered using model organisms including yeast, Caenorhabditis elegans, and Drosophila melanogaster. Many of these genes and pathways have been shown to regulate lifespan in mammals as well (Fontana et al., 2010; Kenyon, 2010). Among them, insulin/IGF-1 signaling (IIS) and target of rapamycin (TOR) pathways are two of the most evolutionarily well-conserved aging-regulatory pathways (Stanfel et al., 2009; Fontana et al., 2010; Kapahi et al., 2010; Kenyon, 2010; Evans et al., 2011). Insulin/IGF-1 signaling and TOR signaling pathways link metabolic status and aging and interact with each other, although the factor(s) mediating the cross talk between these two pathways is incompletely understood.

Target of rapamycin is a protein kinase that acts as a cellular nutrient sensor to regulate various physiologic processes, including metabolism, growth, and aging (Stanfel et al., 2009; Kapahi et al., 2010; Evans et al., 2011; Laplante & Sabatini, 2012). The TOR kinase forms two distinct complexes, TORC1 and TORC2, which have complex-specific adaptors such as raptor and rictor, respectively (Dazert & Hall, 2011; Laplante & Sabatini, 2012). The activity of TORC1 is regulated by upstream signaling proteins such as Rag GTPases in response to changes in the cellular environmental conditions (Kim et al., 2008). When the nutrient supply is sufficient, TORC1 positively regulates cellular growth and proliferation by promoting protein synthesis and/or inhibiting autophagy. To exert its effects on protein synthesis, TORC1 phosphorylates its target proteins, including ribosomal protein S6 kinase (S6K) and eukaryotic translation initiation factor 4E-binding protein 1 (4E-BP1) (Laplante & Sabatini, 2012). S6K mediates the regulation of mRNA translation via phosphorylation of its substrates, including ribosomal protein subunit 6 (Erikson & Maller, 1985; Laplante & Sabatini, 2012). Recent studies have shown that genetic inhibition of TOR, S6K, or ribosomal protein subunits extends lifespan in many organisms (Fabrizio et al., 2001; Jia et al., 2004; Kapahi et al., 2004; Kaeberlein et al., 2005; Chen et al., 2007; Curran & Ruvkun, 2007; Hansen et al., 2007; Pan et al., 2007; Syntichaki et al., 2007; Selman et al., 2009) and that treatment with the TORC1 inhibitor rapamycin increases the lifespan of yeast, C. elegans, Drosophila, and mice (Powers et al., 2006; Harrison et al., 2009; Bjedov et al., 2010; Robida-Stubbs et al., 2012). These studies establish the TOR signaling pathway as an evolutionarily conserved aging-regulatory pathway and a promising target for the development of anti-aging medicines.

Reduced IIS confers lifespan extension in many species, potentially including humans (Fontana et al., 2010; Kenyon, 2010). In C. elegans, inhibition of daf-2, which encodes a homolog of insulin/IGF-1 receptor, doubles lifespan (Kenyon et al., 1993). This long lifespan requires the FOXO transcription factor DAF-16, which induces the expression of many pivotal target genes to promote longevity (Fontana et al., 2010; Kenyon, 2010). Heat shock factor 1 (HSF-1) is another transcription factor crucial for longevity downstream of IIS (Hsu et al., 2003; Morley & Morimoto, 2004). HSF-1 is activated upon reduction in IIS and is required for the long lifespan caused by inhibition of IIS, including daf-2 mutations (Hsu et al., 2003; Morley & Morimoto, 2004; Chiang et al., 2012). In addition, over-expression of hsf-1 prolongs the lifespan of C. elegans, suggesting that increase in HSF-1 activity is sufficient for longevity (Hsu et al., 2003; Morley & Morimoto, 2004). It has been shown that HSF-1 not only promotes longevity but also enhances stress resistance, innate immunity, and protein homeostasis, and that these effects are achieved mainly through increasing the levels of molecular chaperones at the transcription level (Hsu et al., 2003; Cohen et al., 2006; Singh & Aballay, 2006; Mohri-Shiomi & Garsin, 2008; Steinkraus et al., 2008). These studies demonstrated that HSF-1 acts downstream of IIS to extend lifespan; however, whether other longevity signaling pathways, including the TOR pathway, also influence HSF-1 activity is poorly understood.

In this report, we show that HSF-1 mediates the lifespan extension caused by reduction of TOR signaling in C. elegans. We found that enhanced lifespan, oxidative stress resistance, and pathogen resistance resulting from inhibition of the TOR signaling pathway are highly dependent on HSF-1. In addition, we demonstrated that mutations in rsks-1, which encodes the C. elegans ribosomal protein S6 kinase, increase HSF-1 activity, and that hsp-16, a small heat-shock protein whose induction is regulated by HSF-1, is required for the longevity of rsks-1 mutants. Furthermore, we showed that HSF-1 is synergistically activated by the inhibition of both TOR signaling and IIS pathways and is required for the further enhanced longevity of daf-2 rsks-1 double mutants. Together, our findings establish HSF-1 as a hub protein for the cross talk between TOR signaling and IIS pathways to regulate lifespan.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

While performing a genome-wide RNA interference (RNAi) screen for identification of genes that interact with rsks-1, we found that RNAi targeting hsf-1 was one of the 12 RNAi clones that specifically reduced the body size of rsks-1 mutants (Fig. S1A). We confirmed this result using hsf-1(sy441) and rsks-1(tm1714) mutations (Fig. S1B). We were intrigued by this specific genetic interaction and further investigated the relationship between rsks-1 and hsf-1. Because C. elegans hsf-1 has been established as a crucial life-extending gene (Hsu et al., 2003; Morley & Morimoto, 2004) and rsks-1 mutations prolong lifespan (Hansen et al., 2007; Pan et al., 2007), we examined the role of hsf-1 in the longevity of rsks-1 mutants. We found that hsf-1 RNAi or mutations completely suppressed the long lifespan of the rsks-1 mutants (Fig. 1A–C). The effect of hsf-1 inhibition on the longevity of rsks-1 mutants was comparable to that of mutations in daf-16/FOXO transcription factor or aak-2/AMP-activated protein kinase (Fig. 1D,E and Fig. S2; see also Fig. S2 legend for discussion) (Selman et al., 2009). Knockdown of hsf-1 using RNAi only during adulthood completely suppressed the long lifespan of rsks-1 mutants as well (Fig. 1F and Fig. S1C). This result excludes the possibility that depletion of HSF-1 during larval development negatively affected the health of the rsks-1 mutant animals and subsequently reduced the lifespan during adulthood. Together, these data indicate that down-regulation of S6 kinase requires HSF-1 to lengthen lifespan in C. elegans.

image

Figure 1. HSF-1 is required for the longevity caused by the inhibition of S6 kinase (rsks-1). (A, B) The long lifespan of rsks-1(tm1714) mutants was completely suppressed by hsf-1 RNAi (A) or hsf-1(sy441) mutations (B). Because hsf-1(sy441) mutants display larval arrest phenotype at 25 °C (Hajdu-Cronin et al., 2004), we tested whether the development of hsf-1(sy441) mutants was affected at 22.5 °C. We found that hsf-1(sy441) mutants did not display the larval arrest phenotype at 20 °C or 22.5 °C (Fig. S1D), the temperatures for the lifespan assays (A) and (B), respectively. (C) hsf-1 is required for the longevity of rsks-1 mutants without FUdR treatment. (D, E) The long lifespan of rsks-1(tm1714) mutants was suppressed by daf-16(mu86) (D) or aak-2(ok524) (E) mutations. (F) hsf-1 RNAi treatment only during adulthood (adult-only RNAi) suppressed the longevity of rsks-1(tm1714) mutants. See also Fig. S1C for the lifespan results of rsks-1(tm1714) mutants treated with hsf-1 RNAi only during adulthood without FUdR. A summary of the lifespan data presented in this and other Figures and additional repeats are included in Table S1. Lifespan experiments were carried out at 20 °C for (A) and at 22.5 °C for (B–F).

Download figure to PowerPoint

Because S6K is a key downstream kinase of TORC1 (Laplante & Sabatini, 2012), we asked whether hsf-1 plays a role in the longevity caused by down-regulation of other components in the TORC1 signaling pathway. We found that hsf-1 was required for the lifespan extension caused by RNAi knockdown of daf-15/raptor or ragc-1/Rag GTPase, or by rapamycin treatment (Fig. 2A–C). Down-regulation of TORC1 extends lifespan at least partly through decreasing mRNA translation, which is regulated by phosphorylation of ribosomal protein subunits such as RPS6 by S6K. We therefore examined the effects of hsf-1 mutations on the longevity caused by reduction in ribosomal protein subunit levels. We found that increased lifespan conferred by knockdown of rps-6 or rps-15 was suppressed by hsf-1 mutations (Fig. 2D,E). These results further support our notion that hsf-1 is crucial for lifespan extension by inhibition of TORC1 signaling.

image

Figure 2. HSF-1 is required for the long lifespan resulting from the inhibition of various components of TOR signaling. (A, B) Lifespan extension caused by inhibiting one of the components in TORC1 required hsf-1. hsf-1(sy441) mutation suppressed the prolonged lifespan caused by RNAi targeting daf-15/raptor (A) or ragc-1/Rag GTPase (B). (C) The longevity conferred by treating wild type with rapamycin (100 μm) was suppressed by RNAi knockdown of hsf-1. (D, E) Long lifespan conferred by knockdown of rps-6 (D) or rps-15 (E) was dependent on hsf-1. Note that we carried out lifespan assays by treating animals with RNAi only during adulthood, because daf-15 homozygous mutations or RNAi clones targeting ribosomal protein subunits result in larval arrest phenotypes (Jia et al., 2004; Hansen et al., 2007). All lifespan experiments in this figure were carried out at 22.5 °C.

Download figure to PowerPoint

Longevity has been shown to correlate well with increased resistance against external stresses and protein aggregation. We therefore determined the role of rsks-1 and hsf-1 in the survival rate under various stresses. Among the external stresses, rsks-1 mutations increased the resistance against oxidative stress but not against heat or osmotic stresses (Fig. 3A and Fig. S3A–C; see also Fig. S3 legend for discussion). We found that hsf-1 was required for the enhanced resistance of the rsks-1 mutants to oxidative stress (Fig. 3A). The requirement of hsf-1 for this increased oxidative stress resistance was comparable to or higher than that of daf-16/FOXO whose role in oxidative stress resistance is well known (Fig. 3B). We then examined the role of hsf-1 in enhanced protein homeostasis in rsks-1 mutants using polyglutamine-yellow fluorescent protein (Q40::YFP) transgenic animals (Morley et al., 2002), which are an established worm model of polyglutamine diseases. We found that rsks-1 mutations reduced Q40::YFP (Q40) aggregates and that hsf-1 was required for the reduction in this aggregate formation (Fig. 3C,D). These data suggest that rsks-1 mutations enhance oxidative stress resistance and protein homeostasis through HSF-1.

image

Figure 3. Enhanced resistance of rsks-1 mutants against oxidative stress and polyglutamine aggregation requires hsf-1. (A, B) Increased resistance of rsks-1(tm1714) mutants against oxidative stress caused by paraquat (20 mm) treatment was completely suppressed by hsf-1 RNAi (A) and partially suppressed by daf-16/FOXO RNAi (B). (C, D) rsks-1(tm1714) mutation decreased the level of aggregated YFP-fused polyglutamine repeats (Q40). (C) RNAi knockdown of hsf-1 reversed the reduced polyglutamine aggregation in rsks-1(tm1714) mutant animals. (D) Quantification of the number of Q40::YFP (Q40) aggregates. Error bars represent the standard error of the mean (s.e.m.) (n > 30, ***< 0.001, two-tailed Student's t-test). Statistical analysis of the data presented in this figure and additional repeats are described in Table S2.

Download figure to PowerPoint

As innate immunity is influenced by various aging-regulatory genes including hsf-1, we also tested whether rsks-1 mutations affected the immunity against P. aeruginosa (PA14), a well-established model bacterial pathogen. We found that rsks-1 mutants were more resistant to PA14 than wild-type and that hsf-1 was absolutely required for this PA14 resistance (Fig. 4A). In contrast, RNAi targeting several other genes known to be crucial for pathogen resistance, such as daf-16/FOXO, nsy-1/MAP kinase kinase kinase, or pmk-1/MAP kinase (Irazoqui et al., 2010), did not fully suppress the PA14 resistance of rsks-1 mutants (Fig. 4B–D). Together, these data suggest that hsf-1 is essential for rsks-1 mutants to be resistant to oxidative stress, protein aggregation, and PA14. Given the strong correlation between longevity and stress resistance phenotypes in C. elegans, this enhanced resistance against various stresses may contribute to the long lifespan of rsks-1 mutants.

image

Figure 4. Resistance to PA14 conferred by rsks-1 mutations requires hsf-1. (A, B) hsf-1 was required for the increased survival of rsks-1(tm1714) mutants feeding on the pathogen P. aeruginosa (PA14) (A), whereas daf-16/FOXO RNAi (B) had little effects on the survival. (C, D) RNAi targeting nsy-1/MAP kinase kinase kinase (C) or pmk-1/MAP kinase (D) did not fully suppress the pathogen resistance caused by rsks-1(tm1714) mutations. A summary of the data presented in this figure and additional repeats are described in Table S2.

Download figure to PowerPoint

The requirement of hsf-1 for the longevity, and stress and pathogen resistance caused by rsks-1 mutations led us to ask whether down-regulation of TOR signaling was sufficient to increase HSF-1 activity. By performing real-time quantitative reverse transcription–PCR (qRT-PCR) analysis, we determined the mRNA expression levels of five known HSF-1 target chaperone genes, hsp-16.1/2/11 (hsp-16.1, hsp-16.2, and hsp-16.11), hsp-70, hsp-12.6, aip-1, and sip-1 (Fig. 5A–E and Fig. S4). We found that hsp-16.1/2/11 and hsp-70 were significantly induced by rsks-1 mutations in the presence and the absence of heat shock and that the induction was largely hsf-1 dependent (Fig. 5A,B and Fig. S4A,B). In contrast, hsp-12.6 and aip-1 were induced by rsks-1 mutations only upon heat-shock treatment (Fig. 5C,D and Fig. S4C,D). We also showed that GFP expression under the control of the hsp-16.2 promoter (hsp-16.2::GFP) (Link et al., 1999; Walker & Lithgow, 2003), a fluorescence reporter for HSF-1 activity, was increased by rsks-1 mutations, and this up-regulation was suppressed by hsf-1 RNAi knockdown (Fig. 5F,G). Moreover, we found that hsp-16.1/2/11 and hsp-70 were induced by RNAi knockdown of daf-15 or ragc-1, and that hsf-1 mutations suppressed the induction of these genes (Fig. 5H–K). These data indicate that at least a subset of HSF-1 targets is induced by inhibition of TOR signaling.

image

Figure 5. rsks-1 mutations extend lifespan through increasing the level of small heat-shock proteins in an HSF-1-dependent manner. (A–E) Of five HSF-1 target genes that we tested, mRNA levels of hsp-16.1/2/11 (A) and hsp-70 (C12C8.1) (B) were highly increased by the rsks-1(tm1714) mutations in an HSF-1-dependent manner in the absence of heat shock. mRNA expression of the other three tested HSF-1 target chaperon genes, hsp-12.6 (C), aip-1 (D), and sip-1 (E), was not significantly affected by rsks-1 mutations or hsf-1 RNAi. Error bars represent the standard error of the mean (s.e.m) (n = 4, **< 0.01, ***P < 0.001, two-tailed Student's t-test). (F) After incubating animals for 1 h at 32 °C, the level of GFP expression under the control of an hsp-16.2 promoter (hsp-16.2::GFP) was significantly increased by rsks-1(tm1714) mutations in an HSF-1-dependent manner. (G) Quantification of GFP pixel intensity. Experiments were performed three times independently, and error bars represent the standard error of the mean (s.e.m) (n > 30, **< 0.01, two-tailed Student's t-test). (H, I) daf-15 RNAi increased the mRNA levels of hsp-16.1/2/11 (H) and hsp-70 (I) in an hsf-1-dependent manner. (J, K) Likewise, ragc-1 RNAi induced the mRNA expression of hsp-16.1/2/11 (J) and hsp-70 (K), and hsf-1 mutations suppressed the induction of these genes. Error bars represent the standard error of the mean (s.e.m) (n = 6, *< 0.05, **< 0.01, ***P < 0.001, two-tailed Student's t-test). (L) hsp-16.1/2/11 RNAi, which targets hsp-16.1, hsp-16.2, and hsp-16.11, suppressed the longevity conferred by rsks-1(tm1714) mutations in three of four trials (two of two trials at 20 °C and one of two trials at 22.5 °C.). We speculate that the inconsistency might originate from the differences in temperatures (20 °C vs. 22.5 °C). In addition, we cannot exclude the possibility that hsp-16.1/2/11 RNAi did not work for one trial that did not display the longevity suppression or the RNAi exhibited variations among trials. Note that hsp-16.1/2/11 RNAi was designed to target hsp-16.1, but because of over 88% sequence identity among hsp-16.1, hsp-16.2, and hsp-16.11 genes, the RNAi clone is expected to target hsp-16.1, hsp-16.2, and hsp-16.11 genes. (M) RNAi knockdown of hsp-70 did not decrease the lifespan of rsks-1(tm1714) mutants or the wild-type. Note that hsp-70 RNAi increased the lifespan of wild-type at 22.5 °C, while having no effects at 20 °C (Table S1). It was previously reported that hsp-70 RNAi did not influence the lifespan of wild-type C. elegans at 25 °C (Morley & Morimoto, 2004). It is possible that hsp-70 RNAi may increase the lifespan of worms at an intermediate temperature (22.5 °C) via compensatory induction of life-extending genes, while not affecting lifespan at 20 °C or 25 °C. Currently, we do not have mechanistic explanation for this result. All lifespan experiments in this figure were carried out at 22.5 °C.

Download figure to PowerPoint

Next, we examined the functional significance of the induction of hsp-16 and hsp-70 for the longevity of rsks-1 mutants. We found that RNAi targeting hsp-16.1/2/11 significantly suppressed the longevity of rsks-1 mutants but hsp-70 RNAi did not (Fig. 5L,M). Thus, rsks-1 mutants may live long, at least in part, through up-regulation of hsp-16 resulting from the activation of HSF-1.

HSF-1 is a well-known longevity transcription factor downstream of the IIS pathway (Hsu et al., 2003; Morley & Morimoto, 2004; Chiang et al., 2012). We therefore explored the possibility that HSF-1 acts as a hub transcription factor for the cross talk between insulin/IGF-1 and TOR signaling pathways. Consistent with this idea, we found that the synergistic lifespan-extending effect by rsks-1 and daf-2 mutations was largely abolished by hsf-1 RNAi (Fig. 6A). In addition, daf-16 mutations fully suppressed the longevity of daf-2 rsks-1 double mutants (Fig. 6B). In contrast, enhanced dauer formation caused by rsks-1 mutations in daf-2 mutants, which requires daf-16, was not suppressed by hsf-1 RNAi (Fig. 6C). These data suggest that HSF-1 requirement for the enhanced longevity of the daf-2 rsks-1 double mutant is distinct compared with that of the dauer phenotype of the double mutant. We also found that hsp-16.1/2/11 and hsp-70 were further induced by rsks-1 mutations in daf-2/insulin/IGF-1 receptor mutants and that these highly induced mRNA levels were largely suppressed by hsf-1 RNAi (Fig. 6D,E). Together, these data indicate that HSF-1 as well as DAF-16 acts as a key transcription factor mediating the synergistic longevity conferred by simultaneous down-regulation of both IIS and TOR pathways.

image

Figure 6. HSF-1 is synergistically activated and is required for the extended longevity caused by simultaneous down-regulation of daf-2 and rsks-1. (A, B) rsks-1(tm1714) mutations further increased the long lifespan of daf-2(e1370) mutants. This very long lifespan was suppressed by hsf-1 RNAi (A) or daf-16 mutations (B). (C) rsks-1 mutations enhanced dauer formation in daf-2 mutants. This increased dauer formation was not affected by hsf-1 RNAi in either daf-2(e1370) or daf-2(e1370) rsks-1(tm1714) mutants. In contrast, daf-16 RNAi suppressed dauer formation in daf-2(e1370) and daf-2(e1370) rsks-1(tm1714) mutants. Likewise, daf-16 mutations suppressed the dauer formation of rsks-1 single mutants at 25 °C but hsf-1 RNAi did not (Fig. S5A,B). Error bars represent the standard error of the mean (s.e.m) (n = 4, *< 0.05, **< 0.01, ***< 0.001, two-tailed Student's t-test). (D, E) hsp-16.1/2/11 (D) and hsp-70 (E) mRNA levels were further increased by rsks-1(tm1714) mutations in daf-2(e1370) mutants, and hsf-1 RNAi nearly reversed the induction of these chaperone genes in daf-2(e1370) rsks-1(tm1714) double mutants. Error bars represent the standard error of the mean (s.e.m) (n = 4, *< 0.05, two-tailed Student's t-test). (F) Schematic representation of the longevity pathways by which TOR signaling and IIS regulate HSF-1. Inhibition of DAF-2/insulin/IGF-1 receptor in the IIS or positive regulators of TOR signaling, including Rag GTPase, raptor, TOR kinase, and S6 kinase, increases the activities of DAF-16 and HSF-1, which appears to lead to longevity. All lifespan experiments in this figure were carried out at 20 °C.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

In this study, we show that HSF-1 is a crucial longevity transcription factor acting downstream of the TOR signaling pathway. We found that genetic inhibition of hsf-1 completely abrogated the long lifespan caused by inhibition of TOR signaling components. We also demonstrated that inhibition of C. elegans S6 kinase (rsks-1), a key downstream kinase of TOR, increased the level of a subset of HSF-1 targets. In addition, hsf-1 was required for the resistance of rsks-1 mutants to various internal and external stresses, which has been shown to strongly correlate with longevity. Furthermore, we showed that HSF-1 was required for the longevity caused by simultaneous inhibition of both IIS and TOR signaling. Together, our findings suggest that HSF-1 may act as a converging transcription factor downstream of these two most evolutionarily well-conserved longevity pathways (Fig. 6F).

We noted the specific genetic interaction between rsks-1 and hsf-1 initially through an unbiased modifier RNAi screen using the slow-growth phenotype of rsks-1 mutants. We then found the requirement of HSF-1 in other physiological phenotypes of rsks-1 mutants, including lifespan extension, stress resistance, and pathogen resistance. Because hsf-1 RNAi influenced the growth of rsks-1 mutants, HSF-1 might be considered a limiting factor for rsks-1 mutants to normally develop to adulthood and to live long. However, we prefer our model proposing that HSF-1 acts as an instructive factor for the longevity and the stress resistance of rsks-1 mutants for the following reasons. First, we showed that hsf-1 RNAi only during adulthood was sufficient for suppressing the longevity of rsks-1 mutants, excluding the possibility that hsf-1 RNAi results in general sickness during growth. Second, we found that rsks-1 mutations increased a subset of HSF-1 target gene expression and that one HSF-1 target gene, hsp-16, was at least partly required for the long lifespan. Third, we showed that TOR signaling and IIS synergistically increased the level of HSF-1 target genes. These data are consistent with our model that the reduction in TOR signaling activates HSF-1, which in turn extends lifespan.

How does TOR signaling influence the activity of HSF-1? Although a recent study using cultured mammalian cells showed that mammalian TOR (mTOR) phosphorylates and activates HSF-1 (Chou et al., 2012), this may not be directly translated into our study. First, the serine residue in HSF-1 that is phosphorylated by mTOR (Chou et al., 2012) is not conserved across organisms, including yeast, C. elegans, Drosophila, and zebrafish. Thus, this regulatory mechanism of HSF-1 by TOR may have evolved relatively recently. Moreover, in contrast to Chou et al.'s paper suggesting the activation of HSF-1 by TOR, we found that reduction in TOR signaling activates HSF-1. Thus, the role of phosphorylation of HSF-1 by TOR in lifespan regulation remains to be determined in future studies.

What will be the functional significance of up-regulation of HSF-1 activity by reduced TOR signaling? A previous report showed that dietary restriction increases lifespan and enhances resistance to polyglutamine aggregation in an hsf-1-dependent manner (Steinkraus et al., 2008). As dietary restriction reduces TOR signaling (Kaeberlein & Kennedy, 2011), it seems likely that dietary restriction promotes longevity and stress resistance via inhibiting TOR signaling and subsequently increasing HSF-1 activity. Inhibition of TOR signaling is also well known to decrease mRNA translation, which leads to reduction in the overall protein level. Several studies have indicated that reduced translation by TOR signaling increases lifespan and stress resistance (Stanfel et al., 2009; Kapahi et al., 2010; Evans et al., 2011; Laplante & Sabatini, 2012). Under these conditions, organisms may need to maintain biological functions using available proteins that were already synthesized. Therefore, the thrifty usage of the available proteins would be very critical for survival. Our data imply that reduced TOR signaling results in the activation of HSF-1 and subsequent up-regulation of molecular chaperones, which helps to improve overall protein quality control. This enhanced protein quality control in general may in turn increase lifespan and stress resistance. Alternatively, activated HSF-1 may induce specific chaperones, which facilitate maintaining or activating a subset of client proteins that are important for longevity and stress resistance. Consistent with this idea, only a subset of known HSF-1 target genes is induced by rsks-1 mutations. Moreover, a small heat-shock protein that has been shown to be necessary and sufficient for the longevity caused by reduced IIS (Hsu et al., 2003; Walker & Lithgow, 2003; Morley & Morimoto, 2004) was required for the long lifespan of rsks-1 mutants. This indicates that the client proteins of HSP-16 are pivotal longevity proteins. It will be interesting to investigate these two alternative but not mutually exclusive possibilities to further understand the molecular basis of lifespan regulation by TOR signaling. Future studies using biochemical approaches to determine how the activity of HSF-1 is increased by reduced TOR signaling in animals will shed light on this issue.

Several previous studies have revealed that TOR signaling and IIS pathways employ common transcription factors, including SKN-1/NRF-2 and DAF-16/FOXO to regulate the lifespan of C. elegans (Sheaffer et al., 2008; Tullet et al., 2008; Wang et al., 2010; Robida-Stubbs et al., 2012). Here, we show that HSF-1 functions as an intersecting point or downstream of the intersecting point of TOR and IIS pathways. Future studies to elucidate the mechanisms by which these transcription factors interact with one another upon reduction in TOR signaling and IIS pathways will be important for providing insights into solving the secrets of human aging, particularly because TOR signaling and IIS pathways are two of the most crucial longevity pathways.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Strains

The following strains were analyzed in this study.

N2 wild-type, VB633 rsks-1(sv31) III, FX01714 rsks-1(tm1714) III, IJ109 rsks-1(tm1714) III outcrossed 4 times to Lee-lab N2, which is the same as Kenyon-lab N2, CF2725 aak-2(ok524) X outcrossed 3 times to Kenyon-lab N2, CF2495 hsf-1(sy441) I outcrossed three times to Kenyon-lab N2, CF1042 daf-16(mu86) I outcrossed 11 times to Kenyon-lab N2, IJ184 rmIs133[unc-54p::Q40::YFP] outcrossed 4 times to Kenyon-lab N2, CL2070 dvIs70[hsp-16.2p::GFP; rol-6D], IJ120 rsks-1(tm1714) III; aak-2(ok524) X, IJ121 hsf-1(sy441) I; rsks-1(tm1714) III, IJ123 daf-16(mu86) I; rsks-1(tm1714) III, IJ185 rsks-1(tm1714) III; rmIs133[unc-54p::Q40::YFP], IJ186 rsks-1(tm1714) III; dvIs70[hsp-16.2p::GFP; rol-6D], CF1041 daf-2(e1370) III outcrossed 6 times to Kenyon-lab N2, CF1085 daf-16(mu86) I; daf-2(e1370) III, IJ124 daf-2(e1370) rsks-1(tm1714) III, IJ413 daf-16(mu86) I; daf-2(e1370) rsks-1(tm1714) III.

Lifespan assays

Lifespan assays were performed as described previously with some modifications (Lee et al., 2010). Worms were synchronized by allowing gravid adults to lay eggs for 12 h, and the progeny were allowed to develop. When worms were at young adult stage, approximately 30 worms were transferred onto each new NGM plate containing 5 μm 2′fluoro-5′deoxyuridine (FUdR, Sigma). For the experiment shown in Fig. 1C and Fig. S1C, approximately 30 worms were synchronized at young adult stage and worms were transferred to fresh plates without FUdR every 2–3 days. We carried out lifespan assays at 20 °C or 22.5 °C for the following reasons. When we measured the lifespan of rsks-1 mutants at 20 °C, the animals tend to display a vulval extrusion phenotype, which required censoring of many animals. To circumvent this problem, we performed the lifespan assays mostly at 22.5 °C after maintaining the worms at 22.5 °C for at least two generations. We carried out the lifespan experiments for daf-2(e1370) or daf-2(e1370) rsks-1(tm1714) mutants at 20 °C, because many of these animals arrested as dauer larvae at 22.5 °C. We specified the temperature at which each lifespan was measured in the Figure Legends and the Supporting Tables. Animals that ruptured, exhibited internal hatching of progeny, burrowed, or crawled off the plates were censored but included in the lifespan analysis as censored animals, as described previously (Lee et al., 2010). OASIS (online application of survival analysis, http://sbi.postech.ac.kr/oasis) was used for statistical analysis (Yang et al., 2011). P values were calculated using the log-rank (Mantel–Cox method) test.

Rapamycin treatment

Rapamycin (LC Laboratories, Woburn, MA, USA) was dissolved in DMSO to obtain a 50 mg mL−1 stock solution and added onto plates to achieve the final concentration of 100 μm. The control plates were prepared to contain the same amount of DMSO without rapamycin.

Quantitative RT–PCR assays

Synchronized worms at young (day 1) adult stage were prepared for quantitative RT–PCR assays. RNA isoplus (Takara, Shiga, Japan) reagent was used for the extraction of total RNA from C. elegans. Total cDNA was synthesized using the reverse transcription system (Promega, Madison, WI, USA) and used as templates for qPCR with the SYBR green dye (Takara, Shiga, Japan). The PCR reaction was carried out in a StepOne Real-Time PCR system (Applied Biosystems, USA). mRNA of hsp-16.1/2/11, hsp-70, aip-1, sip-1, or hsp-12.6 was amplified with gene-specific primers and normalized to ama-1 mRNA level. The sequences of the gene-specific primers are described in the Supporting Information.

Oxidative stress resistance assays

Oxidative stress resistance assays were performed using paraquat as described previously with some modifications (Lee et al., 2010). Gravid worms were transferred onto NGM plates seeded with Escherichia coli to lay eggs for 12 h. Progeny were grown on the plates to reach adulthood, and approximately 80 young (day 1) adults were transferred onto bacteria-containing plates treated with paraquat (methyl viologen; Sigma, St. Louis, MO, USA) at a final concentration of 20 mm and with 5 μm FUdR. Worms were incubated at 20 °C and scored as dead when they did not respond to prodding with a platinum pick. 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.

PA14 slow-killing assays

PA14 slow-killing assays were carried out as described previously with minor modifications (Tan et al., 1999). P. aeruginosa (PA14) was cultured in LB media at 37 °C overnight, and 5 μL of the culture was seeded onto high-peptone (0.35%) agar plates with 5 μm FUdR. The bacteria were then incubated at 37 °C for 24 h followed by incubation at 25 °C for 8–16 h for slow-killing assay. Approximately 100 young adult worms were transferred onto two PA14-seeded plates for each experiment. Worms were scored alive if they responded to prodding. OASIS was used for statistical analysis, and P values were calculated using the log-rank (Mantel–Cox method) test.

Fluorescence microscopy analysis

The number of polyglutamine (Q40::YFP) aggregates defined as discrete structures with edges recognizable from surrounding fluorescence was counted as described previously using Axio Scope A1 compound microscope (Zeiss, Oberkochen, Germany) (Hsu et al., 2003). For measuring the level of hsp-16.2::GFP expression, young (day 1) adult worms were subjected to heat shock for 1 h at 32 °C, then recovered at 20 °C for 6 h, and monitored for GFP level using the Axio Scope A1 compound microscope. The fluorescence pixel intensity was quantified using Imagej software (Rasband, W.S., ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA, http://rsb.info.nih.gov/ij/).

Dauer assays

Dauer assay was performed as described previously with some modifications (Gaglia et al., 2012). At least 150 worms were synchronized by allowing 15–30 adult worms to lay eggs at 22.5 °C for 6 h. Because the rsks-1(tm1714) mutations caused developmental delay in daf-2 mutants, dauers were scored after 2 days for daf-2(e1370) mutants and 3 days for daf-2(e1370) rsks-1(tm1714) mutants. The assay was independently repeated four times.

Acknowledgments

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (NO. NRF-2012R1A4A1028200), and by a grant of the Korean Health Technology R&D Project, Ministry of Health & Welfare, Republic of Korea (A111656) to S-J.L., and by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (NO. NRF-2010-0019706) to S.K.J. Some C. elegans strains used in this study were provided by Drs. Cynthia Kenyon and Shohei Mitani, and by the Caenorhabditis Genetics Center, which is funded by the NIH National Center for Research Resources. We thank Dr. Dennis Kim for providing PA14, and all Lee and Jang laboratory members for help and discussion.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
acel12140-sup-0001-FigureS1-S5.pdfapplication/PDF5520K

Fig. S1 Genetic interaction between hsf-1 and rsks-1 for growth during development and adult lifespan phenotypes.

Fig. S2 Decrease in the lifespan of rsks-1 mutants by daf-16 RNAi treatment.

Fig. S3 The effects of rsks-1 mutations on the survival responses to heat and osmotic stresses.

Fig. S4 rsks-1 mutations induce a subset of HSF-1 target genes in an hsf-1-dependent manner upon treatment with heat shock.

Fig. S5 Increased dauer formation by rsks-1 mutation requires daf-16 but not hsf-1.

acel12140-sup-0002-TableS1-S2.docxWord document111K

Table S1 Lifespan analysis.

Table S2 Survival analysis other than lifespan analysis.

Data S1 Experimental procedures.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.