Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 9 (8), e1003737

Deletion of microRNA-80 Activates Dietary Restriction to Extend C. Elegans Healthspan and Lifespan

Affiliations

Deletion of microRNA-80 Activates Dietary Restriction to Extend C. Elegans Healthspan and Lifespan

Mehul Vora et al. PLoS Genet.

Abstract

Caloric/dietary restriction (CR/DR) can promote longevity and protect against age-associated disease across species. The molecular mechanisms coordinating food intake with health-promoting metabolism are thus of significant medical interest. We report that conserved Caenorhabditis elegans microRNA-80 (mir-80) is a major regulator of the DR state. mir-80 deletion confers system-wide healthy aging, including maintained cardiac-like and skeletal muscle-like function at advanced age, reduced accumulation of lipofuscin, and extended lifespan, coincident with induction of physiological features of DR. mir-80 expression is generally high under ad lib feeding and low under food limitation, with most striking food-sensitive expression changes in posterior intestine. The acetyltransferase transcription co-factor cbp-1 and interacting transcription factors daf-16/FOXO and heat shock factor-1 hsf-1 are essential for mir-80(Δ) benefits. Candidate miR-80 target sequences within the cbp-1 transcript may confer food-dependent regulation. Under food limitation, lowered miR-80 levels directly or indirectly increase CBP-1 protein levels to engage metabolic loops that promote DR.

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Figure 1
Figure 1. mir-80(Δ) exhibits multiple features of healthy aging.
Fig. 1A. mir-80(Δ) has low intestinal age pigment levels compared to wild type during late adult life (day 11). We grew age-synchronized WT (black), mir-80(Δ) (red), and mir-80(Δ); Ex[Pmir-80(+)] (grey) under standard conditions (20°C, on E. coli OP50-1) and scored animals for age pigment levels using a fluorimeter (n = 100 per strain/trial; day 11, as counted from the hatch; mir-80(Δ) is nDf53; mir-80(+) rescue transgene is nEx1457 [18]). Age pigment fluorescence, which increases with age, is normalized to endogenous tryptophan fluorescence, which remains relatively constant with age , (AGE/TRP ratio ∼58% decreased in mir-80(Δ) vs. wild type). Graphs represent mean data from at least 3 independent trials. Data were compared using the One-way ANOVA followed by Newman-Keuls multiple comparison test, *** - p<0.0005, * - p<0.05; WT to Ex[Pmir-80(+)] rescue p<0.12. In the rescued strain, age pigment levels might not reach WT levels due to mosaicism of the extrachromosomal transgene, the mir-80 transgene dose, or “sponge” effects of overexpression. Fig. 1B. mir-80(Δ) maintains youthful pharyngeal pumping in late adulthood. We assayed age-synchronized WT (black), mir-80(Δ) (red), and mir-80(Δ); Ex[Pmir-80(+)] (grey, nEx1457 ([18])) for pharyngeal pumping rates on Day 5 (left) and Day 11 (right) (30 s interval, n = 10/trial, 3 trials). For day 5, we included the eat-2(ad1116) mutant (blue), impaired for pharyngeal pumping to ∼30% WT rate, as a negative control. In this assay we compared healthy appearing animals (most vigorous locomotion). Graph is of cumulative data from 3 independent trials. Data were compared using the One-way ANOVA followed by Newman-Keuls multiple comparison test. * - p<0.05; ** - p<0.005, *** - p<0.0005. mir-80(Δ) pumping rate is modestly higher than WT at day 5 (p = 0.023), but note that relative pumping differences at Day 5 are small compared to differences at Day 11 (∼44% increase). Fig. 1C. mir-80(Δ) maintains youthful swimming vigor in late adulthood. We assayed age-synchronized animals, WT (black), mir-80(Δ) (red), and mir-80(Δ); Ex[Pmir-80(+)] (grey) for swimming mobility at Day 5 and Day 11 post-hatching (n≥30, 3 independent trials are combined in presented data). Data were compared using 2-tailed Student's T-test, *** - p<0.0001. Although mir-80(Δ) and WT swim similarly in young adult life, mir-80(Δ) mutants better maintain swimming prowess late in life, ∼69% increased body bend rate. Fig. 1D. mir-80(Δ) mutants have increased mean and maximum lifespans. We assayed age-synchronized WT (black), mir-80(Δ) (red), and mir-80(Δ); Ex[Pmir-80(+)] (grey) animals grown under standard conditions (20°C, OP50-1) for viability (movement away from worm pick by gentle touch) at the indicated days. We initiated trials with relatively vigorous animals on day 9 from the hatch (10 animals per plate, ≥25 per strain per trial, 3 independent trials, which are combined here). Data from individual trials are shown in Fig. S1. Statistics were calculated using the Log-rank Test. mir-80(Δ) mutants exhibit a significant extension in lifespan as compared to WT (p<0.0001) and transgenic expression of mir-80(+) reversed the longevity increase (p<.0001).
Figure 2
Figure 2. mir-80(Δ) exhibits multiple characteristics typical of DR animals.
Fig. 2A. The mir-80(Δ) mutant exhibits the DR Exmax shift. We assayed age-synchronized WT (black), mir-80(Δ) (red), and mir-80(Δ); Ex[Pmir-80(+)] (grey) 4 day old animals grown under standard conditions (20°C, OP50-1). We used a spectrofluorimeter to scan transparent animals (n = 100 per strain/trial) to generate excitation/emission profiles as in ; wavelength of excitation at maximal fluorescence is indicated. Graphs represent mean data from at least 3 independent trials. Data were compared One-way ANOVA followed by Newman-Keuls multiple comparison test, *** - p<0.0005. mir-80(Δ) exhibits a significantly down-shifted Exmax (p<0.0005) as compared to wild type under conditions of abundant food, a feature unique to DR . Fig. 2B. The mir-80(Δ) mutant exhibits low age pigment levels early in life, as occurs in C. elegans DR. We assayed age-synchronized WT (black), mir-80(Δ) (red), and mir-80(Δ); Ex[Pmir-80(+)] (grey) 4 day old animals grown under standard conditions (20°C, OP50-1). We scanned animals (n = 100 per strain) for fluorescence over a range of wavelengths, and normalized age pigment fluorescence (AGE) to tryptophan (TRP) fluorescence as in for comparison. Graphs represent mean data from at least 3 independent trials. Data were compared using One-way ANOVA followed by Newman-Keuls multiple comparison test, *** - p<0.0005, ** - p<0.005. mir-80(Δ) exhibits low age pigment levels as compared to wild type (p<0.0005) early in life, which is true of all DR conditions previously tested (although not unique to DR). In these assays, levels were on average 66% lower in mir-80(Δ). Fig. 2C. mir-80(Δ) mutants are physically smaller than WT, typical of animals in DR. We measured age-synchronized WT (black) and mir-80(Δ) (red) 4 day old animals (examples at the right) grown under standard conditions (20°C, OP50-1) by imaging animals (WT n = 77, mir-80(Δ) n = 88) under DIC under low magnification. We measured using the segmented line tool in the ImageJ software by drawing a line across the length of the animal, and converted length in pixels to uM using a stage micrometer to assess image scale. We compared data using 2-tailed Student's T-test, *** - p<0.0005. mir-80(Δ) mutants are ∼10% shorter and look thinner than WT reared under the same conditions, typical of the scrawny appearance of animals in DR, example comparison on the right. Although size varies somewhat and is not as quantitative a measure as age pigment scores, we have used the scrawny appearance to identify likely mir-80(Δ) homozygotes in crosses. Fig. 2D. mir-80(Δ) mutants exhibit reduced fertility and an extended reproductive lifespan. We assayed egg production in age-synchronized WT (black), mir-80(Δ) (red), and DR mutant eat-2(ad1116) (blue) grown under standard conditions (20°C, OP50-1; parent n = 10, 3 independent trials). eat-2 is one trial so bars are not provided. Data were compared using 2-tailed Student's T-test. Early in adult life, mir-80(Δ) produce a reduced number of live births per day (33% decrease, p<0.05 for Day 3; 180% increase, p<0.001 Day 4–6) and exhibit a prolonged reproductive lifespan (through Day 8 for mir-80(Δ) as compared to WT Day 6, p<0.001). The constitutive DR mutant, eat-2 experiences a shift in reproductive lifespan (compared to WT, p<0.001) that is similar to mir-80(Δ). Fig. 2E. SKN-1-GFP, a molecular reporter of DR, is upregulated in mir-80(Δ) in the presence of food. SKN-1::GFP expression in the two ASI neurons is a molecular signal of some DR . We constructed strains that included an integrated rescuing skn-1-gfp fusion gene expressed from the native skn-1 promoter, Is007[skn-1-gfp] , and measured at Day 7, 20°C, growth in OP50-1 (white arrows). WT animals show low levels of ASI expression (36% with very weak expression in only one ASI), while DR-constitutive eat-2(ad1116) animals display constitutive expression of SKN-1-GFP in the ASI neurons (92% in 1 or 2 neurons, strong expression). 95% of mir-80(Δ) have 1–2 ASIs expressing at this timepoint. These data support that mir-80(Δ) mutants are in DR even when reared in the presence of ample food.
Figure 3
Figure 3. mir-80 expression is generally high in the presence of food, but low when food is lacking.
3A. Examples of expression of extrachromosomal bzEx207 [Pmir-80LmCherry] line grown in the presence of unlimited E. coli. Note that this transgenic line, typical of 4 lines that have the long mir-80 promoter region, exhibits substantial reporter expression in the first two cells of the intestine (indicated by white+sign) and in the posterior intestine (white bracket). Lower level expression is evident in several other tissues. Animals are adult day 6, but we find no bleed through of signals using red/green filter sets (Fig. S4B) so age pigments do not confound this analysis. 3B. Examples of expression of the bzEx207 [Pmir-80LmCherry] line grown in the presence of unlimited E. coli until young adulthood and then switched to no food for 48 hours. 6 day old adults are aligned with anterior to the left, posterior gut region indicated by white bracket. Most posterior gut fluorescence is markedly diminished, although expression in the anterior two intestinal cells, the central egg laying muscles, and the very posterior gut remains high. 3C. Quantitation of fluorescence signals for a mir-80 promoter fusion reporter line in food vs. food limitation. Fluorescence of overall bzEx207[Pmir-80LmCherry] line expression after 48 hrs on no-food plates. Food limitation in these studies was by dietary deprivation , but food dilution on solid NGM media and food dilution in liquid media induced similar changes in these lines (Fig. S4). Graph represents spectrofluorimeter measurements of fluorescence levels (whole body) for at least 50 animals per DR regimen. Pairwise comparisons were made using Two-tailed Students' T-test. *** - p<0.0005. Same exposure times were used for complementary panels. 3D. Analysis of food-regulated expression of pmir-80LmCherry expression along the nematode body implicates posterior intestinal regions as a major site of regulation. We compared pmir-80LmCherry signals in transgenic ZB3042 grown either in the presence of food (blue) or switched to no food for 24 hrs (red) (measured at day 4, n = 39). We used the ImageJ program to create a 25 pixel segmented line covering the animal and measured mean fluorescence intensity along the body, dividing the length into 12 equal bins and plotting the mean fluorescence intensity at each point. Representative animals are depicted above with the approximate body positions indicated (H = head, P = pharynx, V = vulva, T = tail). Note that although food regulation is apparent in most of the body, food-regulated expression changes in the regions of the mid- and posterior intestine are most dramatic. Error bars indicate standard error for each bin measurement.
Figure 4
Figure 4. daf-16/FOXO is needed for the fluorimetric DR signature and longevity phenotypes of mir-80(Δ).
Fig. 4A. Transcription factor daf-16/FOXO is required for the Exmax shift phenotype in mir-80(Δ). We reared age-synchronized animals under standard growth conditions (20°C, OP50-1) and measured age pigment spectral properties at Day 4 (50 animals per strain) for WT (black bar), mir-80(Δ) (red), daf-16(Δ) allele mgDf50 (blue), and mir-80(Δ);daf-16(Δ) double mutant (grey). The same color coding is used for panels 4A–4D. We recorded Exmax as the highest peak detected by the Datamax software package suite (Horiba Scientific). Graphs represent mean data from at least 3 independent trials. Data were compared using 2-tailed Student's T-test. mir-80(Δ) compared to WT * - p<0.05; mir-80(Δ);daf-16(Δ) double mutant compared to WT, ns. Deletion of daf-16 reverses the Exmax shift phenotype of mir-80(Δ). Fig. 4B. daf-16/FOXO is required for low age pigment levels in mir-80(Δ). We grew age-synchronized animals under standard conditions (20°C, OP50-1) and measured total age pigment fluorescence, normalized to total tryptophan fluorescence as in (Day 4, 50 animals per trial). Graphs represent mean data from at least 3 independent trials. Error bars represent ±S.E.M. Data were compared using 2-tailed Student's T-test. *** - p<0.0005, ** - p<0.005. The low age pigment accumulation phenotype of mir-80(Δ) is reversed in the mir-80(Δ);daf-16(Δ) double mutant on day 4 (shown here) as well as on day 9 (data not shown). Fig. 4C. daf-16 is required for the lifespan extension of mir-80(Δ). We grew age-synchronized animals under standard conditions (20°C, OP50-1). At day 9, we placed 10 healthy animals per plate, ≥40 per strain per trial, and we scored viability as movement away from pick touch on the indicated days. The graphs represent data combined from 3 independent trials. Statistics are calculated using the Log-rank Test. The mir-80(Δ);daf-16(Δ) double mutant is suppressed for the longevity phenotype of mir-80(Δ) (p<0.0001). We did not, however, observe dramatic overall changes in nuclear localization of DAF-16::GFP +/− mir-80 (data not shown). Fig. 4D. mir-80(Δ) lifespan can be further extended by daf-2(RNAi). We placed age-synchronized mir-80(Δ) L1 larvae (Day 1) on empty vector control (pL4440) or daf-2 RNAi plates under standard conditions (20°C). At day 9, we placed 10 healthy animals per plate, ≥40 per strain per trial, and we scored viability as movement away from pick touch at the indicated days. The graphs represent data combined from 3 independent trials. Statistics are calculated using the Log-rank Test. daf-2(RNAi) increases the lifespan of mir-80(Δ) vector control (p<0.005), but additive effects for mir-80(Δ)+daf-2(RNAi) above the daf-2(RNAi) level are not observed (p = 0.98). Note that data from these experiments also provide a general sense of how mir-80(Δ) compares to daf-2 for lifespan extension; roughly we find mir-80(Δ) effects are slightly less than half those of daf-2(rf), see Table S4 for exact data from individual trials.
Figure 5
Figure 5. hsf-1 is needed for the fluorimetric DR signature and longevity phenotypes of mir-80(Δ).
Fig. 5A. hsf-1(RNAi) in the mir-80(Δ) background reverses the DR Exmax shift. We grew age-synchronized animals under standard RNAi feeding conditions (20°C, HT115) and measured age pigments at Day 4 (50 animals per RNAi clone). We recorded Exmax as the highest peak detected by the Datamax software package suite (Horiba Scientific). Black bar, WT+ empty vector RNAi; red bar, mir-80(Δ)+empty vector RNAI; grey bar, mir-80(Δ)+hsf-1(RNAi). Graphs represent cumulative data from 3 independent trials. Error bars represent ±S.E.M. Data were compared using 2-tailed Student's T-test (** p<0.001). Note that hsf-1(RNAi) treatment of WT does not change Exmax (data not shown). Fig. 5B. hsf-1(RNAi) in the mir-80(Δ) background partially counters the low age pigment level phenotype of mir-80(Δ). We grew age-synchronized animals under standard conditions (20°C, HT115) and measured total age pigment fluorescence, normalized to total tryptophan fluorescence as in (Day 4 post-hatching, 50 animals per RNAi clone). Black bar, WT+ empty vector RNAi; red bar, mir-80(Δ)+empty vector RNAi; grey bar, mir-80(Δ)+hsf-1(RNAi). Graphs represent cumulative data from 3 independent trials. Error bars represent ±S.E.M. Data were compared using 2-tailed Student's T-test (*** p<0.0001, * p<0.05 compared to mir-80(Δ) empty vector). Note that hsf-1(RNAi) treatment of WT does not change age pigment scores at day 4 (data not shown). Fig. 5C. hsf-1 is required for mir-80(Δ)-induced longevity. We grew age-synchronized animals under standard conditions with low levels of FUDR to prevent progeny production (20°C, OP50-1, 50 uM FuDR). At day 9, we placed 10 healthy animals per plate, ≥40 per strain per trial, and we scored viability as movement away from pick touch at the indicated days. The graphs represent data combined from 3 independent trials. Statistics are calculated using the Log-rank Test. Error bars indicate ± S.E.M. The mir-80(Δ); hsf-1(sy441) double mutant is shorter lived than mir-80(Δ) (p<0.0001). Because RNAi knockdown is inefficient the nervous system (see [59]), the profound effects of hsf-1(RNAi) suggest that critical hsf-1 and mir-80 regulation occurs outside of the C. elegans nervous system.
Figure 6
Figure 6. CBP-1 is critical for mir-80(Δ) healthspan benefits, and is a candidate direct binding target of miR-80.
Fig. 6A. cbp-1(RNAi) in the mir-80(Δ) background reverses the DR Exmax shift. We grew age-synchronized animals under standard RNAi feeding conditions (20°C, HT115) and measured age pigments at Day 4 (50 animals per RNAi clone). We recorded Exmax as the highest peak detected by the Datamax software package suite (Horiba Scientific). Graphs represent cumulative data from 3 independent trials. Error bars represent ±S.E.M. Data were compared using 2-tailed Student's T-test (** p<0.001, * p≤0.055 compared to mir-80(Δ) empty vector). cbp-1(RNAi) Exmax is comparable to that of ad lib wild type (p = 0.729). Note that cbp-1(RNAi) treatment of WT does not change Exmax (data not shown), so this effect is specific to the DR signature of mir-80(Δ). Fig. 6B. cbp-1(RNAi) in the mir-80(Δ) background partially reverses low age pigment levels. We grew age-synchronized animals under standard conditions (20°C, HT115) and measured total age pigment fluorescence at day 4 (50 animals per RNAi clone), normalized to total tryptophan fluorescence as in ref. . Graphs represent cumulative data from 3 independent trials. Error bars represent ±S.E.M. Data were compared using 2-tailed Student's T-test (** p<0.05, * p<0.1 compared to mir-80(Δ)+empty vector RNAi). Note that cbp-1(RNAi) treatment of WT induces modest reduction of age pigment levels (p = 0.01, data not shown). Fig. 6C. mir-80(Δ) longevity is dependent on cbp-1. We placed age-synchronized L1 larvae on empty vector control (pL4440) plates under standard conditions (20°C) until Day 4 (day 1 of adult life) at which time animals were moved to either empty vector control (L4440) or cbp-1(RNAi) plates. At day 9, we placed 10 healthy animals per plate (≥40 per strain per trial), and we scored viability as movement away from pick touch on the indicated days. The graphs represent data combined from 3 independent trials. Statistics are calculated using the Log-rank Test. cbp-1(RNAi) decreases the lifespan of mir-80(Δ) (p<0.0001 compared to vector control. Because RNAi knockdown is inefficient the nervous system (see [59]), the profound effects of cbp-1(RNAi) suggest that critical cbp-1/mir-80 regulation occurs outside of the C. elegans nervous system. Fig. 6D. The cbp-1 transcript includes two predicted binding sites for miR-80. Exon structure of cbp-1 is indicated by thick blue bars, introns in thin black lines (see WormBase for details). The rna22 algorithm , which searches for target sites outside the 3′UTR, predicts that miR-80 binds cbp-1 within the 5′ UTR and within exon 8. The potential alignments of miR-80 (red) to C. elegans cbp-1 (blue) sequences are indicated. Note that the seed match to the exon 8 region is a perfect 10 bp match for C. elegans, and that the target sequence is conserved in mouse and human CBP1 (see Fig. S7). Fig. 6E. Endogenous CBP-1 protein levels are increased in 7 day old mir-80(Δ) mutants. We grew age-synchronized animals under standard conditions (20°C, OP50-1) and extracted total protein at Day 7 (100 animals per strain) for Western blot analysis (top). Graphs represent CBP-1 levels for each strain normalized to own TUB-1 levels. Error bars represent ±S.E.M. Data were compared using 2-tailed Student's T-test (** p<0.005). The graphs represent data combined from 3 independent trials. We noted that during young adulthood, native levels of CBP-1 seemed comparable to WT in mir-80(Δ), suggesting that additional regulatory controls are exerted on CBP-1 expression levels in development or early adulthood.
Figure 7
Figure 7. A model for miR-80 regulation of DR metabolism.
In adults, when food is abundant, mir-80 is expressed at a high level, and miR-80 binds to metabolic and signaling targets to down-regulate their expression. The cbp-1 transcript, which includes two potential binding sites for miR-80, one in the 5′ UTR and one in exon 8 (exons thick dark blue lines, promoter lighter blue), and is essential for mir-80(Δ) benefits, is one candidate target (light blue represents relatively low CBP-1 concentration in food). When food is limiting, miR-80 levels drop, and translational repression of cbp-1 could be relieved (dark blue circle represents higher concentration CBP-1). The CBP-1 protein associates with DAF-16 and HSF-1 to promote expression of genes required for DR metabolism and longevity. Note that although cbp-1 is essential for mir-80(Δ) DR benefits, direct targeting remains to be proved and it is likely that additional targets help modulate the DR state. Since we cannot rule out that daf-16, hsf-1, and cbp-1 disruptions make animals too generally sick to gain mir-80(Δ) benefits, alternative models are possible.

Similar articles

See all similar articles

Cited by 30 PubMed Central articles

See all "Cited by" articles

References

    1. Kenyon CJ (2010) The genetics of ageing. Nature 464: 504–512. - PubMed
    1. Tissenbaum HA (2012) Genetics, life span, health span, and the aging process in Caenorhabditis elegans. J Gerontol A Biol Sci Med Sci 67: 503–510. - PMC - PubMed
    1. Fontana L, Partridge L, Longo VD (2010) Extending healthy life span–from yeast to humans. Science 328: 321–326. - PMC - PubMed
    1. Greer EL, Brunet A (2009) Different dietary restriction regimens extend lifespan by both independent and overlapping genetic pathways in C. elegans. Aging Cell 8: 113–127. - PMC - PubMed
    1. Lee GD, Wilson MA, Zhu M, Wolkow CA, de Cabo R, et al. (2006) Dietary deprivation extends lifespan in Caenorhabditis elegans. Aging Cell 5: 515–524. - PMC - PubMed

Publication types

MeSH terms

LinkOut - more resources

Feedback