SEARCH

SEARCH BY CITATION

Keywords:

  • 4-hydroxynonenal;
  • aging;
  • C. elegans;
  • electrophilic stress;
  • longevity regulation;
  • molecular biology of aging;
  • oxidative stress;
  • RNA interference;
  • RNAi;
  • stress resistance

Summary

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

Electrophilic stress caused by lipid peroxidation products such as 4-hydroxynonenal (4-HNE) and/or related compounds may contribute to aging. The major mode of 4-HNE metabolism involves glutathione conjugation catalyzed by specialized glutathione transferases. We have previously shown that glutathione transferase CeGSTP2-2, the product of the Caenorhabditis elegans gst-10 gene, has the ability to conjugate 4-HNE, and that its overexpression extends lifespan of C. elegans. We now demonstrate that the expression level of CeGSTP2-2 correlates highly with lifespan in a series of hypomorphic daf-2 mutants of C. elegans. The overexpression of CeGSTP2-2 in daf-2 is abrogated in daf-16; daf-2 mutants, indicating that expression of the gst-10 gene is modulated by insulin-like growth factor signaling. To determine whether the relationship between CeGSTP2-2 and lifespan is causal, we used RNAi to knock down CeGSTP2-2. Treatment with gst-10-specific dsRNA decreased CeGSTP2-2 protein in wild-type N2 and in daf-2 strains to an approximately equal level. The ability to conjugate 4-HNE was similarly decreased by RNAi, suggesting that the increment of that activity in daf-2 over N2 is due largely to the overexpression of CeGSTP2-2. RNAi-mediated knock-down of CeGSTP2-2 led to an increased susceptibility to 4-HNE, paraquat, and heat shock, and to a shortening of lifespan by 13% in both N2 and daf-2 strains. These results indicate that CeGSTP2-2 significantly contributes to the maintenance of the soma, and that this function is augmented in daf-2 mutants concordantly with other longevity assurance genes, probably via insulin-like growth factor signaling.


Introduction

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

Aging is a multifactorial process that is thought to reflect an imbalance between two opposing forces: damage to cellular and organismal structures, which accumulates over time and eventually leads to dysfunction and death, and mechanisms which prevent and/or repair such damage, and thus prolong life. The nature of the damaging agents and protective mechanisms most critical to longevity has been hotly debated. According to the free radical theory of aging (Harman, 1956, 2003), the damage, mostly oxidative in nature, is essentially random. On the other hand, recent findings demonstrate that aging is genetically, hormonally, and nutritionally regulated (reviewed in Kenyon, 2001, 2005), and thus far from stochastic. This dichotomy is probably only apparent. Life is an extremely complex, but basically chemical, process which is subject to largely random perturbations on a chemical (and physical) level. On the other hand, processes opposing damage accumulation are tightly regulated. Such processes act on multiple levels. Some limit the generation of radicals and oxidants (Fridell et al., 2005), but most either detoxify the reactive molecules (Gems & McElwee, 2005) or repair damaged cell components (mostly through DNA repair enzymes and heat shock proteins). Thus, while there are no ‘aging genes’, there are multiple tightly regulated longevity assurance genes.

In principle, longevity assurance genes should be able to protect and/or rejuvenate a biological system indefinitely, at least when cells can divide, as illustrated by the immortality of the germ line. While postmitotic cells in multicellular organisms may have narrower options, near-indefinite or at least very long individual lifespans should be possible. Such long lifespans are not attained, not because of a fundamental limitation, but probably because of the metabolic cost of effective somatic maintenance (Kirkwood, 2005; Partridge et al., 2005). Organisms invest resources in either reproduction or somatic maintenance. The allocation, long presumed to be made at the species level by natural selection, is also determined by individual organisms through a complex system of genetic switches which sense environmental cues and endogenous signals such as nutrient availability, crowding, temperature, metabolic status, etc. (Guarente & Kenyon, 2000). Genes may exhibit antagonistic pleiotropy, whereby the allele or induction state that favors fecundity, does so at the expense of lower somatic maintenance and thus, reduced lifespan. Although this trade-off remains hypothetical, it is supported by genetic evidence indicating antagonistic pleiotropy for some, but not all, longevity-regulating genes (Ayyadevara et al., 2001, 2003).

Experimental interventions can override the intrinsic regulatory circuits and modulate lifespan. Rapid progress has been recently made in elucidating the major signaling pathways that affect longevity, such as sirtuins, JNK signaling, and insulin/insulin-like growth factor signaling (IIS) and its components including Daf-2 (an insulin-like growth factor receptor) and Daf-16 (a FOXO transcription factor homolog) (Guarente & Picard, 2005; Kenyon, 2005; Oh et al., 2005; Wang et al., 2005). With the exception of reactive oxygen species, less attention has been given to specific chemical effector molecules that contribute to aging. In the present work, we have tentatively identified a class of such effectors, the electrophilic products of lipid peroxidation such as 4-hydroxynon-2-enal (4-HNE), and demonstrated their metabolism by a particular longevity assurance gene, a glutathione transferase (GST) which is regulated through IIS.

Results

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

CeGSTP2-2

The CeGSTP2-2 protein was originally purified from C. elegans and named CeGST 5.4 according to its isoelectric point (Engle et al., 2001). N-terminal sequencing identified the protein as the product of the gst-10 gene. On the basis of its sequence, the protein was assigned to the Pi-class of GSTs (Engle et al., 2001), and was thus renamed as CeGSTP2-2 (Ayyadevara et al., 2005) according to established rules of GST nomenclature (Chelvanayagam et al., 2001), although by C. elegans genetic convention the gene retains the 3-letter categorical designation gst-10. CeGSTP2-2 exhibits moderate but significant activity for glutathione conjugation of 4-HNE (Engle et al., 2001; Ayyadevara et al., 2005).

Correlation of lifespan with CeGSTP2-2 level

The reported lifespans of C. elegans hypomorphic daf-2 mutants vary widely (Gems et al., 1998). We determined the amount of CeGSTP2-2 in wild-type N2 and in eight daf-2 strains with reported median lifespans ranging from 25 to more than 60 days (Gems et al., 1998) by semiquantitative Western blotting on total worm homogenates. The expression level of CeGSTP2-2 correlated very well (R2 = 0.82, P = 0.006) with lifespan (Fig. 1). Note that our use of longevities determined elsewhere would only have the effect of reducing correlation. We have reproduced many of these survivals with similar rank order to Gems et al. (1998) but utilized the latter as providing the more complete data set.

image

Figure 1. Correlation between expression level of CeGSTP2-2 protein, determined by semiquantitative Western blotting in wild-type N2 and in a series of daf-2 mutant strains, and median lifespan as reported in (Gems et al., 1998). The strains used are listed in the Figure. Correlation is slightly higher (R2 = 0.82) for the fitted hyperbola shown in the figure than for a fitted straight line (R2 = 0.78).

Download figure to PowerPoint

Overexpression of CeGSTP2-2 in daf-2 mutants requires daf-16

Signaling through the Daf-2 receptor (IIS) requires the function of the Daf-16 transcription factor (reviewed in Kenyon, 2001). To determine whether the increased expression of CeGSTP2-2 is mediated through Daf-16, the level of CeGSTP2-2 was measured in a daf-16; daf-2 double mutant. Absence of daf-16 completely prevented the overexpression of CeGSTP2-2 protein (Fig. 2A,B), and reduced the elevated 4-HNE-conjugating activity of the daf-2 strain to the level of N2 controls (Fig. 2C). The smaller fractional decrease in enzymatic activity than in the level of CeGSTP2-2 protein is consistent with the presence of other 4-HNE-conjugating GSTs in C. elegans (Ayyadevara et al., 2005). Furthermore, the level of protein-4-HNE adducts was reduced in daf-2 but not in daf-16; daf-2 (Fig. 2D). The reduction, by the daf-16 mutation, of CeGSTP2-2 protein, activity, and biological outcome (ability to attenuate adduct formation) to levels observed in N2 controls establishes a causal relationship between IIS and regulation of expression of the gst-10 gene encoding CeGSTP2-2.

image

Figure 2. Overexpression of CeGSTP2-2 in daf-2 relative to wild-type N2 is prevented in a daf-16; daf-2 double-mutant worms. (A) Western blot of worm homogenates probed with anti-CeGSTP2-2 antibody. (B) Quantitation of the Western blot after normalization to actin as loading control (not shown) and to the amount of CeGSTP2-2 present in N2 animals. A representative experiment is shown. (C) 4-HNE-conjugating activity in worm lysates. Activity in daf-2 homogenate was different from that in either N2 or daf-16; daf-2 homogenates (each P < 0.005 by anova with Bonferroni correction for post hoc multiple comparisons), while activities in N2 and daf-16; daf-2 homogenates did not differ from each other (P > 0.5). (D) Level of 4-HNE-protein adducts (relative units). The adduct level in daf-2 homogenate was lower than in either N2 or daf-16; daf-2 (each P < 0.005 by General Linear Model anova with Bonferroni correction for post hoc multiple comparisons on two independent experiments after adjusting for interexperimental variability), while adduct levels in N2 and daf-16; daf-2 homogenates did not differ from each other (P > 0.5).

Download figure to PowerPoint

RNAi-mediated knock-down of CeGSTP2-2

A 242-nt fragment derived from the gst-10 coding sequence (between the initiator ATG and the unique BamHI site) was used to generate gst-10-specific dsRNA. The 44 genes annotated as encoding GSTs and 7 GST-like sequences listed in release WS135 of WormBase (http://www.wormbase.org) were aligned and inspected for sequence similarities with gst-10 using a sliding window of 22 nt. The longest uninterrupted homology block was 16 nt long (in gst-41), and very few blocks were between 9 and 15 nt in length. Other uninterrupted stretches of sequence homology were 8 nt or shorter. A BLAST search against C. elegans sequences did not reveal any substantial homology outside the GST group.

Feeding C. elegans with E. coli which contains gst-10-targeted dsRNA caused an incomplete but nevertheless significant inhibition of CeGSTP2-2 expression in N2 animals, and an even stronger inhibition in daf-2 worms (Fig. 3A,B). The residual level of CeGSTP2-2 was similar in both C. elegans strains. Worms were fed only as adults, to avoid any possible effects on development. Since CeGSTP2-2 catalyzes the conjugation of 4-HNE with glutathione (Engle et al., 2001; Ayyadevara et al., 2005), lysates of dsRNA-exposed animals are expected to have a diminished ability to catalyze this reaction. This is indeed the case, although the inhibition of catalytic activity is less pronounced (Fig. 3C) than the decrease of the protein amount caused by RNAi (Fig. 3A,B). This suggests the presence in C. elegans of GSTs distinct from CeGSTP2-2 which possess 4-HNE-conjugating activity, in agreement with our previous findings (Ayyadevara et al., 2005).

image

Figure 3. Knock-down of CeGSTP2-2 by gst-10-directed dsRNA in N2 and daf-2 C. elegans. (A) Western blot of worm homogenates probed with anti-CeGSTP2-2 antibody. (B) Quantitation of the Western blot after normalization to the amount of CeGSTP2-2 present in N2 animals. A representative experiment is shown. (C) 4-HNE-conjugating activity in worm lysates. Brackets with asterisks indicate significant differences (P < 0.02 by anova with Bonferroni correction for post hoc multiple comparisons).

Download figure to PowerPoint

Effect of RNAi-mediated knock-down of CeGSTP2-2 on stress resistance of the organism

RNAi-mediated inhibition of CeGSTP2-2 expression led to an increased susceptibility to 4-HNE (Fig. 4A), to heat shock (Fig. 4B), and to paraquat toxicity (Fig. 4C). The survival time under the three stresses was shortened by approximately 9–13%, 13–15%, and 11–13%, respectively. Both N2 and daf-2 animals lost stress resistance upon RNAi treatment. The effect of the knock-down was statistically significant for all groups except for N2 worms exposed to paraquat (Table 1). Although in this case statistical significance was not reached for the difference between dsRNA-treated and untreated animals, the trend appeared consistent with other experimental situations, and suggested a loss of stress resistance (Fig. 4C).

image

Figure 4. Effect of RNAi-mediated knock-down of GSTs on stress resistance and lifespan in N2 (circles and triangles) and daf-2 worms (squares). Worms were fed bacteria containing an insert-free control vector (closed symbols) or a vector containing gst-10 (open symbols, A through E) or gst-8 (open symbol, F). (A) Survival in the presence of 10 mm 4-HNE. (B) Survival at 36 °C. (C) Survival in the presence of 100 mm paraquat. (D, E, F) Survival without applied stress (lifespan assays). Survival parameters and statistical analysis are shown in Table 1.

Download figure to PowerPoint

Table 1.  Effect of RNAi-mediated CeGSTP2-2 knock-down on stress resistance and lifespan
 N2daf-2
–RNAi+RNAi–RNAi+RNAi
  1. For stress resistance experiments, the listed significance is the result of bivariate Cox regression used on duplicate experiments to calculate the significance of the treatment (RNAi) after adjusting for interexperimental variability. Lifespans carried out without and with dsRNA treatment were compared by the standard chi-squared-based log rank test, as well as the corresponding randomization test using 106 Monte Carlo samples; both resulting significance levels are listed. For +RNAi experiments, the decrease in median survival time relative to the corresponding –RNAi control is listed in parentheses. Maximum likelihood estimates (MLE) of Gompertz parameters A and G (listed with their lower and upper 95% confidence limits) and of median survival times were obtained from data shown in Fig. 4 using the program WinModest (Pletcher, 1999; Promislow et al., 1999).

4-HNE treatment
 significance (P)7 × 10−63 × 10−6
 median survival (h) 4.25 3.71 (−13%) 5.88 5.33 (−9%)
 Gompertz A (× 102) 2.61 (1.68,4.05) 3.05 (1.92,4.85) 0.78 (0.45,1.35) 0.91 (0.53,1.57)
 Gompertz G (× 102)65.8 (56.9, 76.1)74.9 (64.1, 87.6)66.8 (58.6, 76.2)72.4 (63.4, 82.7)
Heat shock
 significance (P)0.00080.02
 median survival (h) 6.81 5.79 (−15%) 8.25 7.19 (−13%)
 Gompertz A (× 102) 3.41 (2.11, 5.49) 6.32 (4.20, 9.52) 2.12 (1.28, 3.51) 4.40 (2.88, 6.70)
 Gompertz G (× 102)26.8 (21.1, 34.1)21.1 (15.8, 28.3)26.7 (21.7, 32.9)19.7 (15.1, 25.7)
Paraquat treatment
 significance (P)0.100.006
 median survival (h)26.423.4 (−11%)38.333.3 (−13%)
 Gompertz A (× 102) 1.37 (0.88, 2.14) 1.99 (1.31, 3.02) 0.80 (0.51, 1.26) 1.05 (0.68, 1.62)
 Gompertz G (× 102) 4.68 (3.39, 6.46) 3.89 (2.59, 5.83) 3.81 (2.86, 5.07) 3.85 (2.84, 5.23)
No stress (lifespan)
 significance (P)3.2 × 10−8; 1.0 × 10−63.6 × 10−6; 5.0 × 10−6
 median survival (d)16.614.5 (−13%)32.428.2 (−13%)
 Gompertz A (× 102) 0.240 (0.140, 0.414) 0.177 (0.095, 0.330) 0.261 (0.152, 0.449) 0.453 (0.277, 0.741)
 Gompertz G (× 102)24.68 (21.80, 27.92)32.01 (28.17, 36.37) 9.70 (8.22, 11.45) 9.25 (7.74, 11.06)

Effect of RNAi-mediated knock-down of CeGSTP2-2 on lifespan

Lifespan of N2 and daf-2 animals was decreased upon exposure to gst-10-targeted dsRNA by 13% of the original median lifespan (Fig. 4D,E, and Table 1). This effect was sequence specific as dsRNA targeted to another GST, gst-8, did not alter lifespan (Fig. 4F; P = 0.518 by log-rank test). Since the daf-2 strain has a significantly longer survival time, the approximately equal fractional reduction by gst-10 RNAi of the lifespan of daf-2 and N2 animals translates into a greater chronological median lifespan decrease for daf-2 (4.2 days) as compared with N2 (2.1 days). Maximum likelihood estimates of Gompertz parameters for all treatments are listed in Table 1.

Discussion

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

We have demonstrated that the expression level of CeGSTP2-2 correlates highly with lifespan in a series of long-lived hypomorphic daf-2 mutants of C. elegans (Fig. 1). The correlation was very strong, especially in light of the fact that we used median lifespan data reported in the literature (Gems et al., 1998) in conjunction with CeGSTP2-2 levels measured by us. Inevitable methodological differences between laboratories and other sources of variability would be expected to weaken the correlation, yet it remained remarkably strong. Nevertheless, it is only a correlation which, in itself, does not speak to the presence or absence of a causal link. Given the role of CeGSTP2-2 in the metabolism of the lipid peroxidation product 4-HNE (Esterbauer et al., 1991; Dianzani, 2003; Awasthi et al., 2004; Petersen & Doorn, 2004), which in turn is a downstream consequence of oxidative stress, it seemed attractive to hypothesize that the link between this GST and lifespan is causative rather than only correlative, and that the effect is due to 4-HNE metabolism. Consistent with this hypothesis was the finding that, in daf-2, the higher level of CeGSTP2-2 was accompanied by a lower amount of 4-HNE-protein adducts. These adducts are thought to mediate the biological effects of 4-HNE. To establish a cause–effect relationship between CeGSTP2-2 level and longevity, we knocked down CeGSTP2-2 expression using dsRNA derived from its coding sequence (gst-10). Due to relatively high sequence homology within the GST gene superfamily, which in C. elegans encompasses a staggering number of at least 44 genes, complete specificity is difficult to achieve. However, the longest uninterrupted block of sequence homology between gst-10 and any of the other C. elegans GST coding sequences was only 16 nt long, less than the approximately 22 nt used in the silencing mechanism; the vast majority of sequence homology blocks were 8 nt or less in length. Moreover, it has been estimated that 80% identity over 200 bp is needed for effective silencing (Kamath & Ahringer, 2003). The 242-bp fragment of gst-10 used to generate dsRNA was 75% identical to the corresponding fragment of gst-41, the most-similar C. elegans GST-coding sequence. Even though the similarity threshold required for cross-targeting was not reached even for the most closely related sequence, a further test of specificity was carried out. RNAi against gst-41 failed to inhibit 4-HNE-conjugating activity in N2 animals (data not shown). This result demonstrates that gst-10 was not affected by gst-41-targeted dsRNA, and indicates that RNAi procedures are able to discriminate between even closely related GST isoforms. Thus, even though some ‘off-target’ silencing cannot be ruled out (Qiu et al., 2005), the selected fragment is expected to result in dsRNA selective for gst-10.

RNAi knock-down of CeGSTP2-2 by the bacterial feeding method (Timmons et al., 2001) was effective in reducing, albeit not fully eliminating, the enzyme in wild-type N2 worms. In a daf-2 mutant strain, CeGSTP2-2 was knocked down to the same final level as in N2 animals. Changes in 4-HNE conjugation catalyzed by worm homogenates, an activity characteristic for CeGSTP2-2, largely paralleled the knock-down of protein, although the effects on activity were less pronounced. This finding suggests the presence in C. elegans of other GST(s) with the ability to conjugate 4-HNE, a conclusion consistent with our previous observations (Ayyadevara et al., 2005). The residual level of 4-HNE-conjugating activity in daf-2 worms subjected to dsRNA was slightly but significantly (P < 0.02) higher than the residual activity in N2 animals. This could indicate a limitation of the RNAi method, or could suggest an up-regulation in daf-2 of GSTs other than CeGSTP2-2. However, the difference was minor, and is unlikely to have a significant impact on the conclusions.

Knock-down of CeGSTP2-2 caused a moderate but significant decrease in stress resistance (survival of exposure to elevated temperature or to 4-HNE) in both N2 and daf-2 animals, and a decrease of resistance to paraquat in daf-2 worms. Moreover, the knock-down led to a decrease in lifespan in the absence of explicit stress exposure. Interestingly, the absolute (chronological) decrease in survival time differed between N2 and daf-2, but the relative decrease (expressed as a fraction of the median lifespan) was remarkably similar for the two genotypes (Table 1).

We propose the following conceptual framework to interpret our results. We postulate that CeGSTP2-2 has a role in somatic maintenance of C. elegans. While the mechanism of the protective role of CeGSTP2-2 is not known, we speculate that it may be due to the ability of this particular GST to conjugate, and thus inactivate, the electrophilic lipid peroxidation product 4-HNE. This conjecture is based on several lines of evidence. We have recently demonstrated that either overexpression of CeGSTP2-2 or heterologous expression of the murine mGSTA4-4, an enzyme which is also capable of metabolizing 4-HNE, increase lifespan and stress resistance in C. elegans (Ayyadevara et al., 2005). Parallel experiments in a knockout mouse impaired in its ability to metabolize 4-HNE have shown a greater susceptibility to paraquat toxicity (Engle et al., 2004). Finally, we can draw on a large body of information on general physiological and pathological effects of 4-HNE (reviewed in Dianzani, 2003; Forman et al., 2003; Leonarduzzi et al., 2004; Petersen & Doorn, 2004). 4-HNE is a signaling molecule that modulates basic cell functions including proliferation, differentiation, and apoptosis. Interference with these processes is likely to have detrimental consequences. At superphysiological concentrations 4-HNE becomes patently toxic (Esterbauer et al., 1991), a situation which may be achieved when 4-HNE metabolism is experimentally impaired by CeGSTP2-2 knock-down. Moreover, mutation or overexpression of the gst-4 gene product, a GST distinct from CeGSTP2-2 which in our hands has low 4-HNE-conjugating activity (Ayyadevara et al., 2005), has no effect on lifespan of C. elegans (Leiers et al., 2003).

While an involvement of 4-HNE as an aging effector molecule appears likely, metabolism of other GST substrates – perhaps of other α,β-unsaturated carbonyl compounds, or entirely different GST functions – may also contribute to somatic maintenance. However, in the absence of contrary evidence we hypothesize that 4-HNE is the relevant mediator which contributes to aging. This hypothesis is supported by our recent finding that overexpression of either CeGSTP2-2 or murine mGSTA4-4 increases lifespan in C. elegans in proportion to the increment of 4-HNE-conjugating activity (Ayyadevara et al., 2005). A role of CeGSTP2-2-mediated metabolism of 4-HNE (or other substrates) in longevity assurance is consistent with the recently proposed ‘green theory’ of aging (McElwee et al., 2004; Gems & McElwee, 2005) which emphasizes the importance of generalized detoxification, including but not limited to antioxidant function, in somatic maintenance and thus longevity.

We found that the increased expression of CeGSTP2-2 in C. elegans carrying a hypomorphic daf-2 allele requires the presence of an intact daf-16 gene, although the basal expression of CeGSTP2-2 was not diminished in the daf-16; daf-2 double mutant. This indicates that at least the inducible expression of the gst-10 gene is modulated by IIS and, directly or indirectly, effected by the forkhead transcription factor Daf-16. Daf-16 controls a large set of longevity assurance genes (reviewed in Kenyon, 2001, 2005; Gems & McElwee, 2005), schematically depicted in Fig. 5 as genes a, b, and c downstream of Daf-16. In wild-type C. elegans, there will be a certain ratio of longevity assurance contributions of gst-10 (arrow labeled ‘1’) and of all other genes controlled by Daf-16 (arrow labeled ‘2’ in Fig. 5). Since the expression of both gst-10 and of the set of genes symbolized by ‘a, b, c’ is dependent on Daf-16, activation of the latter in daf-2 hypomorphs will, in a first approximation, equally induce gst-10 and a, b, c. This could explain the constant fractional contribution of CeGSTP2-2 to longevity in N2 and daf-2 animals, and the strong correlation we observed between CeGSTP2-2 protein level and longevity (Fig. 1) despite rather small survival effects of reducing this activity by RNAi. Under the particular experimental conditions and in the strains we used, loss of CeGSTP2-2 function(s) led to a decrease in lifespan by 13%. While this result should not be generalized to other situations, it demonstrates that CeGSTP2-2 is an important contributor to longevity assurance in C. elegans.

image

Figure 5. Schematic depiction of the role of the gst-10 gene in longevity assurance. ILP, insulin-like peptides. See text for details.

Download figure to PowerPoint

The molecular mechanism by which CeGSTP2-2 participates in longevity assurance remains to be elucidated. One open question is the chemical effector linked to CeGSTP2-2 activity. As already discussed, 4-HNE is an attractive and likely candidate, but the conclusions of the present work do not depend on this functional assignment. The question is of interest because it focuses on a specific chemical mechanism involved in aging, in contrast to the more general statements concerning effects of genes, via unknown mediators. Another unsolved problem relates to the downstream effects of CeGSTP2-2 that are relevant to senescence. The enzyme could detoxify 4-HNE (and/or other substrates) and prevent damage of an essentially random character. Alternatively, removal of 4-HNE (which is able to modulate protein function) could feed back into regulatory pathways. Further work will be needed to elucidate these questions.

In summary, we demonstrated that CeGSTP2-2, probably via its association with 4-HNE metabolism, contributes significantly to the downstream biochemical effects of IIS. The finding highlights the role of electrophilic stress in aging.

Experimental procedures

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

Biochemical methods

Glutathione transferase activity with 4-HNE as substrate was measured according to (Alin et al., 1985). A polyclonal antibody against affinity-purified bacterially expressed CeGSTP2-2 was raised in chicken and characterized as described elsewhere (Ayyadevara et al., 2005). Western blots were probed with this antibody at a 1 : 20 000 dilution at 4 °C overnight, followed by an anti-chicken IgY secondary antibody coupled to peroxidase, and visualization by chemiluminescence detection using the ECL Plus kit (Amersham Biosciences, Piscataway, NJ, USA) on a Storm 860 phosphoimager (Molecular Dynamics, Sunnyvale, CA, USA). Equality of loading was assessed by comparing Ponceau staining intensity after electroblotting, and by reprobing the blots with an anti-actin antibody. Protein-4-HNE adducts were determined by competitive ELISA (Satoh et al., 1999) using a polyclonal antibody against 4 HNE-modified keyhole limpet hemocyanin. The antibody was a generous gift of Dr Dennis R. Petersen, University of Colorado, Denver.

C.elegans culture conditions

C. elegans strain Bristol-N2 (‘N2’) was used as the control or wild-type strain, and the daf-2(e1370) allele as the daf-2-defective mutant strain, in all experiments except those reported in Fig. 1 where a set of eight different daf-2 alleles was analyzed. Where indicated, a daf-16(m26); daf-2(e1370) double mutant was used to test epistasis of GST phenotypes. Worms were cultured in nematode growth medium (NGM: 25 mm potassium phosphate, pH 6.0, 50 mm NaCl, 0.25% (w/v) peptone, 0.5% (w/v) cholesterol, 1 mm MgCl2, 1 mm CaCl2) and fed with E. coli strain OP50 (2 × 109 cells mL−1) grown in 3XD medium (10.5 g L−1 of Na2HPO4, 4.5 g L−1 of KH2PO4, 0.6 g L−1 of NH4Cl, 15 g L−1 of casein hydrolysate, 24 g L−1 of glycerol and 3 mL L−1 of 1 m MgSO4). Animals were grown at 20 °C.

RNAi methodology

The bacterial expression vector pET30a(+)/gst-10, which contains the gst-10 coding sequence, was cut with XbaI (42 bp upstream of the initiator ATG of gst-10) and BamHI (internal to gst-10, 242 bp downstream of the translation initiation site), and the resulting fragment was subcloned into the same sites of the L4440 feeding vector (gift of A. Fire, Timmons & Fire, 1998) such that it was flanked by two identical T7 promoters. Escherichia coli HT115(DE3) (obtained from the Caenorhabditis Genetics Center, University of Minnesota, Minneapolis, MN, USA) was transformed with the recombinant plasmid or with control insert-free vector (for +RNAi and –RNAi experiments, respectively), and C. elegans were fed IPTG-induced bacteria according to Kamath and Ahringer (2003). For dsRNA-mediated silencing of gst-8, clone F11G11.1 from the Ahringer RNAi library (Kamath & Ahringer, 2003) was used.

Determination of C. elegans lifespan

Worms were synchronized by alkaline hypochlorite lysis (5 min at room temperature in 0.5 N NaOH, 1.05% hypochlorite, Emmons et al., 1979) after being grown for two generations without depletion of bacteria. Eggs were rinsed with S buffer (0.1 m NaCl, 0.05 m potassium phosphate, pH 6.0) (Brenner, 1974) and placed on fresh agar plates containing E. coli OP50. Worms were transferred onto 60-mm plates containing IPTG induced E. coli HT115(DE3) transformed with either the recombinant L4440 vector carrying the gst-10 fragment (+RNAi) or insert-free L4440 (–RNAi). Survivals were performed at 20 °C. Worms were counted and transferred daily to fresh 60-mm plates seeded with IPTG induced –RNAi or +RNAi bacteria. Worms were considered dead when they failed to move, either spontaneously or in response to touch, and showed no pharyngeal pumping. 150 worms per strain were used in each lifespan experiment.

Determination of stress resistance

Paraquat resistance was measured on worms fed –RNAi or +RNAi bacteria as described above for lifespan determination, except that 4-day-old adult worms were transferred to plates supplemented with 100 mm paraquat. For assessing heat resistance, young adult worms were incubated in 60-mm –RNAi or +RNAi plates at 36 ± 0.4 °C, and were scored every 2 hours for motility, provoked movement, and pharyngeal pumping, as described for lifespan determination. For assays of 4-HNE resistance, worms were synchronized as described for lifespan measurements. Two days after hatching, worms were transferred onto plates containing IPTG induced E. coli HT115(DE3) transformed with either the recombinant L4440 vector carrying the gst-10 fragment (+RNAi) or insert-free L4440 (–RNAi), and were maintained under these conditions (with daily transfers) for 3 days. The animals were then suspended in 50 mm KPi, pH-6.0, 100 mm NaCl, 10 mm 4-HNE containing the corresponding heat-killed transformed HT115(DE3) bacteria, and placed in 24-well plates at 50 worms in 0.2 mL buffer per well. Plates were kept at 20 °C, and the worms were scored every hour for survival. 100–150 animals per experimental group were used for all stress resistance determinations.

Statistics

Biochemical parameters were compared by anova or General Linear Model anova with Bonferroni correction for post hoc multiple comparisons. Survival and stress test curves were compared by the standard and randomization (using 106 Monte Carlo samples) log rank tests. Data from duplicate stress tests, each involving control and dsRNA-treated strains, were evaluated in aggregate by Cox regression. Maximum likelihood estimates (MLE) of Gompertz survival parameters (Arking, 1998) and of median lifespans were obtained using the program WinModest (Pletcher, 1999; Promislow et al., 1999). Statistical procedures other than MLE calculations were carried out in NCSS software (Number Cruncher Statistical Systems, Kaysville, UT, USA).

Acknowledgments

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

We thank Eric R. Siegel, M.S., Division of Biostatistics, University of Arkansas for Medical Sciences, for statistical advice. This work was supported in part by National Institutes of Health grants R01 AG18845 and R01 ES07804 (to P.Z.), and Program Project grant P01 AG20641 (to R.J.S.R.). P.Z. and R.J.S.R. are recipients of VA Research Career Scientist Awards. Some nematode strains used in this work were provided by the Caenorhabditis Genetics Center, which is funded by the NIH National Center for Research Resources (NCRR). We thank Dr Dennis R. Petersen, University of Colorado, Denver, for the generous gift of an antibody against protein-4-HNE adducts.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  • Alin P, Danielson UH, Mannervik B (1985) 4-Hydroxyalk-2-enals are substrates for glutathione transferase. FEBS Lett 179, 267270.
  • Arking R (1998) Measuring age-related changes in populations. In Biology of Aging: Observations and Principles (ArkingR, ed.). Sunderland, MA: Sinauer Associates, pp. 2759.
  • Awasthi YC, Yang YS, Tiwari NK, Patrick B, Sharma A, Li J, Awasthi S (2004) Regulation of 4-hydroxynonenal-mediated signaling by glutathione S-transferases. Free Radic. Biol. Med. 37, 607619.
  • Ayyadevara S, Ayyadevara R, Hou S, Thaden JJ, Shmookler Reis RJ (2001) Genetic mapping of quantitative trait loci governing longevity of Caenorhabditis elegans in recombinant-inbred progeny of a Bergerac-BO × RC301 interstrain cross. Genetics 157, 655666.
  • Ayyadevara S, Ayyadevara R, Vertino A, Galecki A, Thaden JJ, Shmookler Reis RJ (2003) Genetic loci modulating fitness and life span in Caenorhabditis elegans: categorical trait interval mapping in CL2a × Bergerac-BO recombinant-inbred worms. Genetics 163, 557570.
  • Ayyadevara S, Engle MR, Singh SP, Dandapat A, Lichti CF, Benes H, Shmookler Reis RJ, Liebau E, Zimniak P (2005) Life span and stress resistance of Caenorhabditis elegans are increased by expression of glutathione transferases capable of metabolizing the lipid peroxidation product 4-hydroxynonenal. Aging Cell 4, 257271.
  • Brenner S (1974) The genetics of Caenorhabditis elegans. Genetics 77, 7194.
  • Chelvanayagam G, Parker MW, Board PG (2001) Fly fishing for GSTs: a unified nomenclature for mammalian and insect glutathione transferases. Chem. Biol. Interact. 133, 256260.
  • Dianzani MU (2003) 4-Hydroxynonenal from pathology to physiology. Mol. Aspects Med. 24, 263272.
  • Emmons SW, Klass MR, Hirsh D (1979) Analysis of the constancy of DNA sequences during development and evolution of the nematode Caenorhabditis elegans. Proc. Natl Acad. Sci. USA 76, 13331337.
  • Engle MR, Singh SP, Czernik PJ, Gaddy D, Montague DC, Ceci JD, Yang Y, Awasthi S, Awasthi YC, Zimniak P (2004) Physiological role of mGSTA4–4, a glutathione S-transferase metabolizing 4 hydroxynonenal: generation and analysis of mGsta4 null mouse. Toxicol. Appl. Pharmacol. 194, 296308.
  • Engle MR, Singh SP, Nanduri B, Ji X, Zimniak P (2001) Invertebrate glutathione transferases conjugating 4-hydroxynonenal: CeGST 5.4 from Caenorhabditis elegans. Chem. Biol. Interact. 133, 244248.
  • Esterbauer H, Schaur RJ, Zollner H (1991) Chemistry and biochemistry of 4-hydroxynonenal, malonaldehyde and related aldehydes. Free Radic. Biol. Med. 11, 81128.
  • Forman HJ, Dickinson DA, Iles KE (2003) HNE – signaling pathways leading to its elimination. Mol. Aspects Med. 24, 189194.
  • Fridell Y-WC, Sanchez-Blanco A, Silvia BA, Helfand SL (2005) Targeted expression of the human uncoupling protein 2 (hUCP2) to adult neurons extends life span in the fly. Cell Metab. 1, 145152.
  • Gems D, McElwee JJ (2005) Broad spectrum detoxification: the major longevity assurance process regulated by insulin/IGF-1 signaling? Mech. Ageing Dev. 126, 381387.
  • Gems D, Sutton AJ, Sundermeyer ML, Albert PS, King KV, Edgley ML, Larsen PL, Riddle DL (1998) Two pleiotropic classes of daf-2 mutation affect larval arrest, adult behavior, reproduction and longevity in Caenorhabditis elegans. Genetics 150, 129155.
  • Guarente L, Kenyon C (2000) Genetic pathways that regulate ageing in model organisms. Nature 408, 255262.
  • Guarente L, Picard F (2005) Calorie restriction – the SIR2 connection. Cell 120, 473482.
  • Harman D (1956) Aging: a theory based on free radical and radiation chemistry. J. Gerontol. 11, 298300.
  • Harman D (2003) The free radical theory of aging. Antioxid. Redox Signal. 5, 557561.
  • Kamath RS, Ahringer J (2003) Genome-wide RNAi screening in Caenorhabditis elegans. Methods 30, 313321.
  • Kenyon C (2001) A conserved regulatory system for aging. Cell 105, 165168.
  • Kenyon C (2005) The plasticity of aging: insights from long-lived mutants. Cell 120, 449460.
  • Kirkwood TB (2005) Understanding the odd science of aging. Cell 120, 437447.
  • Leiers BR, Kampkotter A, Grevelding CG, Link CD, Johnson TE, Henkle-Duhrsen K (2003) A stress-responsive glutathione S-transferase confers resistance to oxidative stress in Caenorhabditis elegans. Free Radic. Biol. Med. 34, 14051415.
  • Leonarduzzi G, Robbesyn F, Poli G (2004) Signaling kinases modulated by 4-hydroxynonenal. Free Radic. Biol. Med. 37, 16941702.
  • McElwee JJ, Schuster E, Blanc E, Thomas JH, Gems D (2004) Shared transcriptional signature in Caenorhabditis elegans dauer larvae and long-lived daf-2 mutants implicates detoxification system in longevity assurance. J. Biol. Chem. 279, 4453344543.
  • Oh SW, Mukhopadhyay A, Svrzikapa N, Jiang F, Davis RJ, Tissenbaum HA (2005) JNK regulates lifespan in Caenorhabditis elegans by modulating nuclear translocation of forkhead transcription factor/DAF-16. Proc. Natl Acad. Sci. USA 102, 44944499.
  • Partridge L, Gems D, Withers DJ (2005) Sex and death: what is the connection? Cell 120, 461472.
  • Petersen DR, Doorn JA (2004) Reactions of 4-hydroxynonenal with proteins and cellular targets. Free Radic. Biol. Med. 37, 937945.
  • Pletcher SD (1999) Model fitting and hypothesis testing for age-specific mortality data. J. Evol. Biol. 12, 430439.
  • Promislow DEL, Tatar M, Pletcher SD, Carey JR (1999) Below-threshold mortality: implications for studies in evolution, ecology and demography. J. Evol. Biol. 12, 314328.
  • Qiu S, Adema CM, Lane T (2005) A computational study of off-target effects of RNA interference. Nucl. Acids Res. 33, 18341847.
  • Satoh K, Yamada S, Koike Y, Igarashi Y, Toyokuni S, Kumano T, Takahata T, Hayakari M, Tsuchida S, Uchida K (1999) A 1-hour enzyme-linked immunosorbent assay for quantitation of acrolein- and hydroxynonenal-modified proteins by epitope-bound casein matrix method. Anal. Biochem. 270, 323328.
  • Timmons L, Fire A (1998) Specific interference by ingested dsRNA. Nature 395, 854.
  • Timmons L, Court DL, Fire A (2001) Ingestion of bacterially expressed dsRNAs can produce specific and potent genetic interference in Caenorhabditis elegans. Gene 263, 103112.
  • Wang MC, Bohmann D, Jasper H (2005) JNK extends life span and limits growth by antagonizing cellular and organism-wide responses to insulin signaling. Cell 121, 115125.