Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 552 (7684), 263-267

RNA Polymerase III Limits Longevity Downstream of TORC1

Affiliations

RNA Polymerase III Limits Longevity Downstream of TORC1

Danny Filer et al. Nature.

Abstract

Three distinct RNA polymerases transcribe different classes of genes in the eukaryotic nucleus. RNA polymerase (Pol) III is the essential, evolutionarily conserved enzyme that generates short, non-coding RNAs, including tRNAs and 5S rRNA. The historical focus on transcription of protein-coding genes has left the roles of Pol III in organismal physiology relatively unexplored. Target of rapamycin kinase complex 1 (TORC1) regulates Pol III activity, and is also an important determinant of longevity. This raises the possibility that Pol III is involved in ageing. Here we show that Pol III limits lifespan downstream of TORC1. We find that a reduction in Pol III extends chronological lifespan in yeast and organismal lifespan in worms and flies. Inhibiting the activity of Pol III in the gut of adult worms or flies is sufficient to extend lifespan; in flies, longevity can be achieved by Pol III inhibition specifically in intestinal stem cells. The longevity phenotype is associated with amelioration of age-related gut pathology and functional decline, dampened protein synthesis and increased tolerance of proteostatic stress. Pol III acts on lifespan downstream of TORC1, and limiting Pol III activity in the adult gut achieves the full longevity benefit of systemic TORC1 inhibition. Hence, Pol III is a pivotal mediator of this key nutrient-signalling network for longevity; the growth-promoting anabolic activity of Pol III mediates the acceleration of ageing by TORC1. The evolutionary conservation of Pol III affirms its potential as a therapeutic target.

Conflict of interest statement

The authors declare no competing financial interests.

Figures

Extended Data Figure 1
Extended Data Figure 1. Inhibition of Pol III in yeast.
a, The growth of strains carrying pADH-OsTir with RPC160-AID, RPB220-AID or the control lacking any AID fusion in the presence or absence of 2.5 mM IAA (single trial). b, Chronological lifespans of the control and RPB220-AID strains treated with 0, 0.125 and 0.25 mM IAA. Top panels show a representative of two experiments, performed in parallel with RPC160-AID shown in Fig. 1b. The bottom panels show a single experiment; the improved survival of RPB220-AID was also observed at a higher IAA concentration in a second experiment. c, Yeast replicative lifespan (top panels) is not altered by 1-naphthaleneacetic acid (NAA; analogue of IAA) while cell cycle duration (bottom panels) is. Both were assessed in the pADH-OsTir RPC160-AID strain on a microfluidics dissection platform. The concentrations of NAA span the dynamic range where the degree of protein depletion can be efficiently modulated in this set-up. The same control data are shown in each panel for comparison. For each NAA concentration one experiment was performed. For replicative lifespans, 95% CIs are indicated by shading, or in brackets for median lifespan, together with log-rank p value. One-sided Mann-Whitney U test was used to test for significant differences in cell cycle duration. No adjustments were made for multiple comparisons. Dashed lines on bottom panels represent medians.
Extended Data Figure 2
Extended Data Figure 2. Inhibition of Pol III extends worm lifespan.
a, Lifespan is extended by feeding N2 worms with rpc-1 RNAi at 20°C in absence of FUDR (p<10-3 log-rank test, n= 100 control and rpc-1 RNAi treated animals). b, It is also extended at 25°C in presence of FUDR (p=9x10-3, log-rank test, n= 60 control, 77 rpc-1 RNAi animals). c, Summary of each worm lifespan experiment performed including the representative trials presented in the figures. Log-rank test p value is reported. The total number of animals in the trial = dead + censored. In general, fewer worms were censored in control vs rpc-1 RNAi conditions (average of N2 at 25°C 25% vs 38%; average of N2 at 20°C 53% vs 73%; average of VP303 at 25°C 3% vs 4% and average of VP303 at 20°C 37% vs 54%) which is likely to be due to an increased number of gut explosions in the rpc-1 RNAi treated worms. On average 84.9% of control and 85.6% of rpc-1 RNAi-treated censored events occurred before the 25th percentile of the survival curve. Overall, increasing the temperature to 25°C reduced censoring without altering our findings. d, Lifespan is extended when the RNAi against rpc-1 is restricted to the gut using VP303 strain, at 25°C in presence of FUDR (p=9x10-3, log-rank test, n= 84 control, 103 rpc-1 RNAi treated animals). In (a), (b) and (d), a representative of two trials is shown.
Extended Data Figure 3
Extended Data Figure 3. Genes corresponding to unique Pol III subunits in Drosophila.
The genes encoding the unique Pol III subunits were identified in fruit flies based on their homology to the yeast genes (BLAST, followed by reverse BLAST), or to the human orthologue.
Extended Data Figure 4
Extended Data Figure 4. Inhibition of Pol III extends fly lifespan.
a, Summary of each fly lifespan experiment performed including the representative trials presented in the figures but excluding the ones with rapamycin (see Extended Data Fig. 8a). Experiments were performed on females unless otherwise noted. The total number of animals in the trial = dead + censored. Log-rank test p value is reported. RU486 feeding does not have an effect on the lifespans of: b, UAS-dC160RNAi alone (p=0.28, log-rank test, n= 142 -RU486, 146 +RU486 animals); or c, TIGS alone controls (p=0.41, log-rank test, n= 141 -RU486, 145 +RU486 animals). d, Inducing dC53RNAi in the gut of TIGS>dC53RNAi females by RU486 extends their lifespan (p=3x10-6, log-rank test, n= 143 -RU486, 139 +RU486 animals). e, Inducing dC160RNAi predominantly in the fat body of S1106>dC160RNAi females by RU486 has no effect on their lifespan (p=0.21, log-rank test, n= 158 -RU486, 155 +RU486 animals). f, Inducing dC160RNAi in neurons of elavGS>dC160RNAi females by RU486 has a modest effect on their lifespan (p=0.03, log-rank test, n= 148 -RU486, 155 +RU486 animals). g, RU486 feeding does not have an effect on the lifespan of the GS5961 alone control (p=0.88, log-rank test, n= 89 -RU486, 91 +RU486 animals). In (b) to (g) the single trial performed is shown.
Extended Data Figure 5
Extended Data Figure 5. TIGS is active in ISCs.
Images from the posterior region of the midgut showing GFP expression driven by TIGS in the presence of RU486 and stained with: a, anti-Prospero; b, anti-HRP. GFP expression can be observed in cells with small nuclei that are Prospero-negative in (a) and those that stain with anti-HRP in (b). Examples of both types are indicated with a white “>” on the merged images. GFP-positive cells can be observed whose morphology and staining pattern correspond closely to that of the ISCs (small nucleus, small cell size, Prospero-negative, anti-HRP-positive [see ref. regarding anti-HRP]). TIGS has a complex expression pattern, showing variation between neighbouring cells of the same type and between gut regions. TIGS appears active in at least some ISCs. Images are representative of two animals.
Extended Data Figure 6
Extended Data Figure 6. Effects of dC160RNAi induction in Drosophila adult gut.
Induction of dC160RNAi in the guts of TIGS>dC160RNAi females results in: a, decreased levels of 45S pre-rRNA (p=4x10-3, MANOVA, CI= 0.95-1.1, 0.85-1.0, 0.95-1.1, 0.81-0.92 left to right; EST = 5’ external transcribed spacer, IST = internal transcribed spacer), indicating a reduction in Pol I activity as a result of Pol III – Pol I crosstalk; b, unaltered levels of mRNAs encoding ribosomal proteins (RNA-Seq data, no significant differences at 10% false discovery rate, DESeq2, n= 3 biologically independent samples), indicating no crosstalk between Pol III and Pol II; c, decreased protein synthesis (two further biological repeats and quantification related to Fig. 2e; p=4x10-3, two-sided t-test, n=3 biologically independent samples, CI= 0.65-1.4 -RU486, -0.033-0.68 +RU486). RU486 feeding of TIGS-alone control females does not result in a significant decrease in: d, levels of pre-tRNAs (p=1x10-4, MANOVA, CI= 0.96-1.0, 1.1-1.2, 0.93-1.1, 1.1-1.2, 0.91-1.1, 1.2-1.3 left to right); e, levels of 45S pre-rRNA (p=2x10-4, MANOVA, CI= 0.94-1.1, 1.1-1.3, 0.94-1.1, 1.1-1.2 left to right); f, protein synthesis (p=0.74, two-sided t-test, n=3 biologically independent samples). Induction of dC160RNAi in the guts of TIGS>dC160RNAi females does not result in significant changes to: g, total gut protein content (p=0.43, two-sided t-test); h, female fecundity (p=0.51, two-sided t-test); i, whole-fly body weight, triacylglycerol or protein content (p=0.58, 0.40, 0.16 respectively, two-sided t-test). j, RU486 feeding of TIGS-alone control females does not result in increased resistance to tunicamycin (p=0.89, log-rank test, n= 149 -RU486, 153 +RU486 animals; single trial). Bar charts show mean ± SEM, with n= number of biologically independent samples indicated and overlay showing individual data points. For gel source data see Sup. Fig. 1.
Extended Data Figure 7
Extended Data Figure 7. Regulation of Pol III activity by TORC1 in Drosophila.
a, The antibody raised against a recombinant fragment of Drosophila TOR protein and used for ChIP (Fig. 3a) recognises a single band of the expected size on western blots of S2 cell extracts. b, The same antibody can immunoprecipitate (IP) dTOR from S2 cells expressing the endogenous and FLAG-tagged dTOR. c, It can also IP endogenous dTOR and the intensity of this band is reduced upon treatment of S2 cells with dsRNA against dTOR. For (a) to (c), a single experiment was performed; the ability of the dTOR RNAi to reduce the intensity of the band was confirmed in an independent experiment. d, Relative enrichment of Pol III-transcribed genes is higher than that of Pol II-transcribed genes after ChIP using a second antibody against Drosophila TOR (raised against a peptide, p=2x10-4; LM with an a priori contrast, n= 3 biologically independent samples, CI= 1.6-2.6, 0.81-2.3, 1.1-2.7, 0.77-2.8, -0.24-2.5, -0.065-2.0, 0.13-1.7 left to right). e, No enrichment for Pol III-transcribed genes over Pol II-transcribed genes is observed after mock ChIP with no antibody (p=0.09, LM with an a priori contrast, n=3 biologically independent samples). f, Rapamycin feeding results in a decrease in total RNA content of the adult gut (p<10-4, two-sided t-test). g, Rapamycin feeding results in reduction of pre-tRNAs relative to total RNA in the fly gut (p=10-4, MANOVA). Rapamycin feeding does not result in a reduction of pre-rRNA in the fly gut relative to:h, U3 (p<10-4, MANOVA); i, total RNA (p=0.57, MANOVA). j, Feeding RU486 to TIGS>HA-Maf1 female fruit flies to induce HA-Maf1 in the gut alone extends their lifespan (p=0.006, log-rank test, n= 153 -RU486, 146 +RU486 animals; single trial). Bar charts show mean ± SEM, with n= number of biologically independent samples indicated and overlay showing individual data points. For gel source data see Sup. Fig. 1.
Extended Data Figure 8
Extended Data Figure 8. Relationship between TORC1 and Pol III.
a, Summary of fly lifespans examining the epistasis between Pol III and TORC1 inhibition (top), including the CPH analyses results (bottom). In the summary (top), log-rank test p value, relative to -RU486 -rapamycin control, is reported and the total number of animals in the trial = dead + censored. b, Induction of dC53RNAi in the adult guts by RU486 feeding of TIGS>dC53RNAi females and rapamycin feeding both extend lifespan and are not additive (for statistical analysis see [a]; n= 135 control, 135 +RU486, 120 +rapamycin and 137 +RU486 and rapamycin animals; single trial). c - f, Rapamycin but not induction of dC160RNAi in the guts of TIGS>dC160RNAi females by RU486 reduces the phosphorylation of S6K in the gut (effect of rapamycin p=3x10-4, RU486 p=0.77 and interaction p=0.55, LM, CI= 0.51-1.5, 1.0-1.2, 0.33-0.73, 0.08-0.91 left to right) and whole flies (effect of rapamycin p<10-4, RU486 p=0.10 and interaction p=0.16, LM, CI=0.77-1.2, 0.60-1.0, 0.016-0.19, 0.019-0.15 left to right). Further biological repeats related to Fig. 3g are presented in (c) for the gut and in (d) for the whole fly. These are quantified in (e) and (f) respectively. In (c) to (f), data from four biologically independent samples are shown. For gel source data see Sup. Fig. 1.
Extended Data Figure 9
Extended Data Figure 9. Inhibition of Pol III in the gut preserves organ health.
a, dC160RNAi induction in the gut of adult TIGS>dC160RNAi females by RU486 feeding supresses accumulation of pH3 positive cells in old flies (p=1x10-3, 2-tailed t-test, CI= 58-110 -RU486, 10-46 +RU486). b, dC160RNAi induction in the gut of adult TIGS>dC160RNAi females by RU486 feeding supresses loss of gut barrier function (number of “smurfs”) in old flies (p=5x10-4, χ2-test, CI=16-26 -RU486, 8.7-16 +RU486 % smurf). c, rpc-1 RNAi suppresses the severity of the age-related loss of gut barrier function in worms (effect of age p<10-4, rpc-1 RNAi p=0.51, interaction p=0.01, Ordinal Logistic Regression, CI=5.0-31, 16-50, 24-48, 25-51, 53-78, 34-66 % smurf grades 3 and 4). Age-related loss of gut barrier function has been previously described for worms. d, dC160RNAi induction in the gut of adult TIGS>dC160RNAi males by RU486 feeding results in a small but significant extension of lifespan (p=0.03, log-rank-test, n= 141 –RU486, 139 +RU486 animals; single trial). Bar charts show mean ± SEM with n= number of animals indicated and overlay showing individual data points.
Figure 1
Figure 1. Inhibition of Pol III extends lifespan.
Treatment of the RPC160-AID-myc pADH-OsTir-myc budding yeast strain with 0, 0.125 and 0.25 mM IAA: a, triggers degradation of C160-AID-myc and b, extends its chronological lifespan, measured as colony formation after normalisation for optical density and 10-fold serial dilution ([a] and [b] show a representative of two experimental trials). Feeding N2 worms with E. coli expressing the rpc-1 RNAi construct from L4 stage: c, reduces the levels of rpc-1 mRNA (p<10-4, two-tailed t-test) and d, extends their lifespan relative to vector alone at 20°C in presence of FUDR (p=0.03, log-rank test, n= 86 control, 94 rpc-1 RNAi animals; representative of three trials). Female flies heterozygous for the dC53EY allele display: e, a reduction in dC53 transcript (p=10-4, two-tailed t-test, 95% confidence intervals [CI] = 0.89-1.1 wt, 0.64-0.71 dC53EY/+) and f, extended lifespan (p=6x10-13, log-rank test, n= 152 control, 144 dC53EY/+ animals; single trial). Bar charts show mean ± Standard Error of the Mean (SEM), with n= number of biologically independent samples indicated and overlay showing individual data points. For more detailed demography and summary of worm and fly lifespan trials see Extended Data Fig. 2c and 4a. For gel source data, see Sup. Fig. 1.
Figure 2
Figure 2. Gut-specific inhibition of Pol III extends lifespan, reduces protein synthesis and increases tolerance to proteostatic stress.
a, Activating RNAi against rpc-1 specifically in the worm gut, using the VP303 strain, extends worm lifespan at 20°C in presence of FUDR (p=0.02, log-rank test, n= 90 control, 67 rpc-1 RNAi animals; representative of two trials). b, Feeding RU486 to TIGS>dC160RNAi female fruit flies to induce dC160RNAi in the gut alone extends their lifespan (p=6x10-16, log-rank test, n= 150 -RU486, 157 +RU486 animals; representative of three trials). c, Feeding RU486 to GS5961>dC160RNAi female fruit flies to induce dC160RNAi in the ISCs alone extends their lifespan (p=2x10-4, log-rank test, n= 139 -RU486, 142 +RU486 animals; representative of three trials). Inducing dC160RNAi in the gut with RU486 feeding of TIGS>dC160RNAi females leads to: d, reduction in pre-tRNAs (mean ± SEM, p=0.04, Multivariate Analysis of Variance [MANOVA], n= 10 biologically independent samples per -/+RU486 condition, CI= 0.91-1.1, 0.76-1.0, 0.90-1.1, 0.75-0.92, 0.88-1.1, 0.68-1.0 left to right); e, reduction in gut protein synthesis, as quantified by ex-vivo puromycin incorporation and western blotting (representative of three biologically independent repeats; see Extended Data Fig. 6c); f, improved survival in response to tunicomycin challenge (p=3x10-15, log-rank test, n=185 animals per condition; representative of two trials). For more detailed demography and summary of lifespan trials see Extended Data Fig. 2c and 4a. For gel source data, see Sup. Fig. 1.
Figure 3
Figure 3. Pol III acts downstream of TORC1 for lifespan.
a, Relative enrichment of Pol III-transcribed genes is higher than that of Pol II-transcribed genes after ChIP against TOR (p<10-4, Linear Model [LM] with an a priori contrast, n= 3 biologically independent samples). Rapamycin feeding causes a decrease in pre-tRNAs relative to U3 in: b, whole female flies (p=0.01, MANOVA, CI= 0.76-1.2, 0.58-0.79, 0.75-1.3, 0.49-0.79, 0.70-1.3, 0.48-0.78 left to right); c, their guts (p=0.01, MANOVA, CI= 0.98-1.1, 0.90-1.0, 0.87-1.1, 0.74-0.99, 0.90-1.1, 0.77-0.97 left to right). d, Induction of Maf1 in the guts of TIGS>HA-Maf1 females by RU486 feeding reduces the levels of pre-tRNAs relative to U3 (p=4x10-3, MANOVA, CI= 0.87-1.1, 0.74-1.0, 0.90-1.1, 0.82-1.0, 0.76-1.2, 0.53-1.0 left to right). e, Induction of dC160RNAi in the adult guts by RU486 feeding of TIGS>dC160RNAi females and rapamycin feeding both extend lifespan and are not additive (effect of rapamycin p=2x10-14, effect of RU486 p<2x10-16, interaction p=7x10-9, Cox Proportional Hazards [CPH], n= 135 control, 144 +RU486, 141 +rapamycin and 146 +RU486 and rapamycin animals; single trial). f, Induction of dC160RNAi in the ISCs by RU486 in GS5961>dC160RNAi females and rapamycin both extend lifespan and are not additive (effect of rapamycin p=8x10-14, effect of RU486 p=2x10-5, interaction p=3x10-7, CPH, n= 113 control, 130 +RU486, 145 +rapamycin and 144 +RU486 and rapamycin animals; single trial). Rapamycin but not dC160RNAi induction in the gut by RU468 feeding of TIGS>dC160RNAi females leads to: g, reduction in S6K phosphorylation in the gut or the whole fly (representative of four biologically independent repeats; see Extended Data Fig. 8c-f); h, reduction in egg laying (effect of rapamycin p<10-4, RU486 p=0.87 and interaction p=0.96, LM). i, Model of the relationship between TORC1, Pol III and lifespan. Bar charts show mean ± SEM, with n= number of biologically independent samples indicated and overlay showing individual data points. For more detailed demography, statistics and summary of lifespan trials see Extended Data Fig. 8a. For gel source data see Sup. Fig.1.
Figure 4
Figure 4. Stem cell-restricted Pol III inhibition improves age-related dysplasia and gut barrier function.
dC160RNAi induction in the ISCs of adult GS5961>dC160RNAi females by RU486 feeding supresses age-related accumulation of pH3 positive cells: a, images of pH3 staining in the posterior mid-gut at 70 d of age (white bar = 100 μm, white “>” marks pH3+ cells, representative of seven -RU486 and nine +RU486 animals); b, the number of pH3+ cells per gut (effect of age p<10-4, RU486 p=2x10-3, interaction p=2x10-3, LM, young: 7-9 d, old: 56-70 d, CI= 2.6-8.6, 1.8-9.4, 14-36, 6.0-14 left to right). c, dC160RNAi induction in the ISCs of adult GS5961>dC160RNAi females by RU486 feeding reduces the age-related increase in the number of flies with a leaky gut (effect of age p<10-4, RU486 p=0.09, interaction p=0.01, Ordinal Logistic Regression, young: 21 d, old: 58 d, CI= 0.19-0.28, 1.5-1.8, 18-19, 14-15 % smurf left to right). Bar charts show mean ± SEM, with n= number of biologically independent animals indicated and overlay showing individual data points.

Comment in

Similar articles

See all similar articles

Cited by 18 PubMed Central articles

See all "Cited by" articles

References

    1. Roeder RG, Rutter WJ. Multiple forms of DNA-dependent RNA polymerase in eukaryotic organisms. Nature. 1969;224:234–237. - PubMed
    1. Arimbasseri AG, Maraia RJ. RNA Polymerase III Advances: Structural and tRNA Functional Views. Trends in biochemical sciences. 2016;41:546–559. doi: 10.1016/j.tibs.2016.03.003. - DOI - PMC - PubMed
    1. Kennedy BK, Lamming DW. The Mechanistic Target of Rapamycin: The Grand ConducTOR of Metabolism and Aging. Cell Metab. 2016;23:990–1003. doi: 10.1016/j.cmet.2016.05.009. - DOI - PMC - PubMed
    1. Vannini A, Cramer P. Conservation between the RNA polymerase I, II, and III transcription initiation machineries. Molecular cell. 2012;45:439–446. doi: 10.1016/j.molcel.2012.01.023. S1097-2765(12)00089-5 [pii] - DOI - PubMed
    1. Moir RD, Willis IM. Regulation of pol III transcription by nutrient and stress signaling pathways. Bba-Gene Regul Mech. 2013;1829:361–375. doi: 10.1016/j.bbagrm.2012.11.001. - DOI - PMC - PubMed

Additional references for Methods

    1. Verduyn C, Postma E, Scheffers WA, Van Dijken JP. Effect of benzoic acid on metabolic fluxes in yeasts: a continuous-culture study on the regulation of respiration and alcoholic fermentation. Yeast. 1992;8:501–517. doi: 10.1002/yea.320080703. - DOI - PubMed
    1. Lee SS, Avalos Vizcarra I, Huberts DH, Lee LP, Heinemann M. Whole lifespan microscopic observation of budding yeast aging through a microfluidic dissection platform. Proc Natl Acad Sci U S A. 2012;109:4916–4920. doi: 10.1073/pnas.1113505109. - DOI - PMC - PubMed
    1. Huberts DH, et al. Construction and use of a microfluidic dissection platform for long-term imaging of cellular processes in budding yeast. Nature protocols. 2013;8:1019–1027. doi: 10.1038/nprot.2013.060. - DOI - PubMed
    1. Papagiannakis A, Jonge J, Zhang Z, Heinemann M. Quantitative characterization of the auxin-inducible degron: a guide for dynamic protein depletion in single yeast cells. Scientific Reports. 2017 in press. - PMC - PubMed
    1. Gelino S, et al. Intestinal Autophagy Improves Healthspan and Longevity in C. elegans during Dietary Restriction. PLoS genetics. 2016;12:e1006135. doi: 10.1371/journal.pgen.1006135. - DOI - PMC - PubMed

Publication types

MeSH terms

Substances

Feedback