Downregulation of transposable elements extends lifespan in Caenorhabditis elegans

Sturm, Ádám; Saskői, Éva; Hotzi, Bernadette; Tarnóci, Anna; Barna, János; Bodnár, Ferenc; Sharma, Himani; Kovács, Tibor; Ari, Eszter; Weinhardt, Nóra; Kerepesi, Csaba; Perczel, András; Ivics, Zoltán; Vellai, Tibor

doi:10.1038/s41467-023-40957-9

Download PDF

Article
Open access
Published: 29 August 2023

Downregulation of transposable elements extends lifespan in Caenorhabditis elegans

Nature Communications volume 14, Article number: 5278 (2023) Cite this article

11k Accesses
2 Citations
202 Altmetric
Metrics details

Subjects

Abstract

Mobility of transposable elements (TEs) frequently leads to insertional mutations in functional DNA regions. In the potentially immortal germline, TEs are effectively suppressed by the Piwi-piRNA pathway. However, in the genomes of ageing somatic cells lacking the effects of the pathway, TEs become increasingly mobile during the adult lifespan, and their activity is associated with genomic instability. Whether the progressively increasing mobilization of TEs is a cause or a consequence of ageing remains a fundamental problem in biology. Here we show that in the nematode Caenorhabditis elegans, the downregulation of active TE families extends lifespan. Ectopic activation of Piwi proteins in the soma also promotes longevity. Furthermore, DNA N⁶-adenine methylation at TE stretches gradually rises with age, and this epigenetic modification elevates their transcription as the animal ages. These results indicate that TEs represent a novel genetic determinant of ageing, and that N⁶-adenine methylation plays a pivotal role in ageing control.

Transient loss of Polycomb components induces an epigenetic cancer fate

Article Open access 24 April 2024

Evolution of tissue-specific expression of ancestral genes across vertebrates and insects

Article 15 April 2024

Genome assembly in the telomere-to-telomere era

Article 22 April 2024

Introduction

Transposable elements (TEs), also called mobile genetic elements or jumping genes, are typically 0.1–20 kilobase-long DNA stretches that can change their genomic position, causing insertional mutations in the new locations^1,2. If the acceptor chromosomal site contains functional—coding or regulatory—sequences, TE insertion often leads to a phenotypic consequence. TEs are mainly classified as relocating DNA transposons (the element is excised from its original chromosomal site and moves elsewhere in the genome) and self-replicating retrotransposons (the element is first copied by reverse transcription and then the duplicate jumps into a novel locus)². Several TEs have lost the ability to mobilize themselves through accumulating inactivating mutations during evolution. These so-called non-autonomous elements however can be mobilized by the products of intact TEs determined as autonomous elements. The functional significance of TEs is reflected by the fact that they constitute a large fraction of eukaryotic genomes, the majority of them are represented by retrotransposons, and a significant portion of them is mobile in somatic cells during lifespan. For example, at least 47% of the human genome consists of TE-related repetitive sequences, 97% of these elements are identified as retrotransposons, among which several thousand copies are estimated to be transcriptionally active^3,4,5. Although most human retrotransposons are unable to actively jump⁴, both autonomous and non-autonomous elements are capable of causing insertional mutations, which can often lead to cell death^6,7,8. This mutagenic activity of TEs is associated with genomic instability^6,9, which is a hallmark of essentially all ageing cells¹⁰.

A specific RNA silencing mechanism that protects eukaryotic genomes from the adverse mutagenic activity of TEs is the Piwi-piRNA (P element-induced wimpy testis in Drosophila-Piwi-interacting RNA) pathway¹¹, which operates in non-ageing cells including germline and cancer stem cells, as well as stem cell-like somatic cells of certain animal species, such as the freshwater hydra and flatworm planaria, but not in ageing somatic cells (the non-ageing cells share a unique feature, the capacity to proliferate indefinitely, while the ageing cells, owing to progressively accumulating cellular damage, undergo a functional decline called senescence and eventually die)¹². This raises the possibility that TEs may substantially contribute to the ageing process^13,14,15,16. Indeed, accumulating evidence indicates that TEs exhibit a gradually increasing activity during the lifespan of various eukaryotic species^{17,18,19,20,21,22,23,24}. Despite these data it is still unclear whether ageing is a cause or a consequence of TE mobility. In other words, whether TEs become progressively active in the soma because the organism ages or, conversely, ageing is a result of their growing mobilization remains unresolved (in fact, there is indication for both in the literature^11,22,24). Since active TE families consist of large copy numbers of elements in eukaryotic genomes, the simultaneous, genetic inhibition of all members within a TE family by mutations or gene silencing has not yet been achieved, leaving their potential role in the ageing process unclear. It was recently shown that in mice LINE-1 (long interspersed nucleotide element; L1) retrotransposons become transcriptionally derepressed during cellular senescence, and that inhibitors of L1 reverse transcriptase interfere with age-associated inflammation²⁵. A similar positive effect of reverse transcriptase inhibitors on longevity was previously observed in Drosophila²². These studies using pharmacological interventions to modulate TE activity suggest that mobilization of retrotransposons represents a causative factor for ageing. However, it is also possible that inhibitors used in these studies are not specific to reverse transcriptase enzymes and (an)other factor(s) may exert the detected positive effect on senescence and lifespan.

In the present study, we effectively downregulated specific TE families in the nematode Caenorhabditis elegans, the compact genome of which contains only a relatively small fraction (<12%) of TE-derived repetitive sequences represented by DNA transposons almost exclusively (Supplementary Fig. 1), and found that this genetic intervention extends lifespan. We also observed a lifespan advantage in transgenic nematodes over control when Piwi proteins were ectopically expressed from a heat-inducible promoter in somatic cells. We finally showed that adenine nucleobases at active TE stretches become progressively methylated as the organism ages, and this change is associated with an elevated transcriptional activity of TEs. Thus, TEs play an important role in driving the ageing process in nematodes and perhaps in other organisms, revealing a novel function for a large fraction of eukaryotic genomes.

Results

Downregulation of active transposable element families from the onset of adulthood extends lifespan

The C. elegans genome encodes several active DNA transposon families, in which the elements are present in relatively small copy numbers (Supplementary Fig. 1)²⁶. For example, the haploid genome of C. elegans contains 31 Tc1 (transposon C. elegans number 1) and 22 Tc3 DNA transposon stretches, which are the most mobile TEs in this organism²⁶. The limited copy numbers of active TE families in nematodes raise the possibility that individual TE families can be effectively downregulated and this genetic intervention may confer a longevity advantage. First, we used a recently developed, highly effective RNA interference (RNAi) method²⁷ to silence Tc1 elements in an otherwise wild-type genetic background. The treatment of animals with Tc1-specific double-stranded (ds) RNA was initiated from the onset of adulthood only (non-gravid adult hermaphrodites were transferred onto RNAi plates) because some so-called domesticated TE families have been reported to possess essential developmental roles²⁸. Animals depleted for Tc1 (dsRNA was targeted for the transposase coding region) lived longer at each temperature examined (15, 20 and 25 °C) than control nematodes treated with the empty RNAi vector only (Figs. 1a, c and 2a, b, and Supplementary Fig. 2a and Supplementary Table 1). A similar longevity effect was also obtained when animals were treated under conditions lacking FuDR (5-Fluoro-2′-deoxyuridine), a compound that prevents DNA replication, thereby conferring the animal sterile (Supplementary Fig. 2b and Supplementary Table 1). Thus, lifespan extension provided by Tc1 downregulation was due to a lower activity of Tc1 elements. The mutator strain Bergerac (BO) contains a much higher copy number of Tc1 (over 500 copies per haploid genome) than the wild-type N2 strain²⁹. Treating BO nematodes with dsRNA specific to Tc1 also caused a significant lifespan extension over control (Fig. 1e and Supplementary Table 1). This experiment was performed at 15 °C since BO animals become highly sterile and their physiology altered above this temperature³⁰. We next downregulated Tc3 transposase expression to ascertain whether this intervention also promotes longevity. Indeed, the lifespan of Tc3(RNAi) animals was longer than that of control under conditions with and without FuDR (Fig. 1g, and Supplementary Fig. 2c, d and Supplementary Table 1). Consistent with these data, somatic overexpression of Tc3 from a heat-inducible promoter markedly shortened the lifespan of transgenic animals under inducible conditions (Supplementary Fig. 2e, f and Supplementary Table 1). Then, we targeted Tc14/CemaT1, another active TE family, for downregulation, and observed that treated animals are long-lived relative to control (Fig. 1i, and Supplementary Fig. 2g and Supplementary Table 1). We also revealed that, except for Tc14 that may be downregulated at the protein level as it happens to many genes in this organism or, alternatively, becomes active at later adult stages only (total RNA samples were isolated at the adult stage of day 2), silencing of these TEs leads to a significant but not complete reduction in the corresponding transcript levels, thereby supporting the effectiveness of treatments (Fig. 1b, d, f, h, j and Supplementary Table 2). Cele14 represents a high copy number (~2000 per haploid genome) TE family reported to have no somatic activity³¹, and its downregulation did not modulate lifespan in the treated animals (Supplementary Fig. 2h and Supplementary Table 1). Tc2 (only 4 full copies in the haploid genome), Tc4 (with no open reading frame and only 5 copies per haploid genome) and Tc5 (only 4 copies in the haploid genome which are inactive in the wild type) were also tested but their silencing, as expected, caused no detectable impact on lifespan (data not shown). The contribution of active TE families consisting of only a few copies to the ageing process may be under detectable levels. To sum, specific downregulation of active TE families with relatively high copy numbers extends lifespan in C. elegans.

**Fig. 1: Downregulation of active transposable element (TE) families extends lifespan in *C. elegans*.**

**Fig. 2: Simultaneous downregulation of *Tc1* and *Tc3* increases lifespan additively.**

We also asked whether the positive effect of inhibiting active TE families on lifespan can act additively. Downregulation of Tc1 and Tc3 each lengthened lifespan by around 10% at 20 °C (Fig. 2a, b and Supplementary Table 1). When both of these TE families were simultaneously silenced under the same condition, the lifespan extension of treated animals relative to control was nearly the sum of those observed in the single treatments (Fig. 2c and Supplementary Table 1). In addition, the co-treatment could also reduce Tc1 transcript levels (Fig. 2d and Supplementary Table 2).

In line with literature data²⁶, the lifespan assay results above imply that several TE families should be mobile in somatic cells. This prompted us to quantify transcript levels of Tc1, Tc3 and Tc14 throughout the adulthood. We found that these TEs are transcriptionally active, and each expressed at significantly higher levels in aged adults than in younger ones (Fig. 2e, f). Parallel with this attempt, we could reconfirm that an individual copy of Tc3, R10H1.3, becomes mobile during ageing; R10H1.3 was effectively excised from its original genomic locus in aged (14-day-old) adults but not in young (1-day-old) ones (Fig. 2g, h).

Ectopic expression of Piwi proteins in the soma promotes longevity

To further investigate the effect of TE expression on lifespan, we ectopically expressed certain components of the Piwi-piRNA pathway in the soma (Fig. 3, and Supplementary Figs. 3 and 4, and Supplementary Tables 1, 3). Piwi proteins by themselves are capable of degrading TE transcripts³², and several members of the pathway, including TE-derived piRNAs, are also expressed somatically (Supplementary Fig. 3)³³. Hence, it is possible that expressing individual members of the pathway in somatic cells is capable of interfering with TE activity. Indeed, we observed that lifespan is markedly (by around 30%) extended at 20 °C as compared with control when prg-1 (Piwi-related gene) encoding the primary nematode Piwi protein³⁴ is expressed from a heat-inducible promoter, which is active in the majority of somatic cells and leaky even under non-inducing conditions (Fig. 3a and Supplementary Table 1). In these transgenic animals, somatically expressed PRG-1 was fused with GFP (green fluorescent protein), forming a translational fusion reporter system, p_hsp16.2::PRG-1::GFP (Fig. 3b, c). Thus, we also generated and tested a gfp reporter-free, p_hsp16.2-driven, functional prg-1 transgene (p_hsp16.2::PRG-1). Transgenic animals expressing endogenous PRG-1 proteins in somatic cells lived significantly longer (by around 20% at 25 °C) than controls (wild-type worms and nematodes transgenic for the empty vector) at both 15 and 25 °C (Fig. 3d–f, and Supplementary Fig. 4a, b and Supplementary Table 1). Somatic expression of prg-1 from this construct strongly lowered transcript levels of Tc1 and Tc3, as revealed by a coupled RT-qPCR analysis (Fig. 3f and Supplementary Table 3), and effectively elevated prg-1 transcript levels under inducing conditions (Supplementary Fig. 5). Furthermore, we found that simultaneous downregulation of Tc1 and Tc3 in transgenic animals expressing PRG-1 somatically is incapable of further increasing lifespan (Supplementary Fig. 6 and Supplementary Table 1). Thus, somatic PRG-1 is likely to extend lifespan by inhibiting TE expression.

**Fig. 3: Somatic expression of *prg-1* (Piwi-related gene)*/Piwi* and *ppw-2* (PAZ/PIWI domain containing)*/Ago* each significantly extends lifespan.**

The somatic accumulation of PRG-2, another Piwi paralogue with unknown function in TE inhibition, also increased lifespan (Supplementary Fig. 4c, d and Supplementary Table 1) (we should emphasize that although prg-2 is predicted as a pseudogene, it is expressed in the germline and displays a functional redundancy with prg-1; see wormbase.org). We then overexpressed ppw-2 (Paz/Piwi domain-containing) in the soma which is an Argonaute-like protein functioning in the piRNA system (Supplementary Fig. 3)³⁵. This genetic intervention conferred a lifespan advantage over control at 25 °C (Fig. 3g, and Supplementary Fig. 4e and Supplementary Table 1) and increased ppw-2 transcript levels (Fig. 3h, i). Transcriptional activity of Tc1, Tc3 and Tc14 was reduced in transgenic animals that express ppw-2 somatically, as compared with control (Fig. 3j and Supplementary Table 3). Based on these results, we conclude that somatic expression of Piwi proteins and certain other components of the Piwi-piRNA pathway attenuates the rate of the ageing process, and this effect is linked to lowered TE expression.

Adenine nucleobases at active transposable elements are more methylated in aged nematodes than in young adults

It has recently been demonstrated that adenine nucleobases in DNA can be methylated not only in bacteria and plants but in divergent animal taxa ranging from nematodes to mammals^36,37,38,39, and that in Drosophila this epigenetic mark preferentially occurs at TE sequences³⁷. In addition, N⁶-methyladenine (6 mA) deposition on certain TE families was reported to affect their activity differently in Drosophila (it is associated with activation) and mammals (it is associated with silencing)^37,39, and was not investigated in nematodes. Therefore, we planned to test the effect of DNA N⁶-adenine methylation on the transcriptional activity of TEs in C. elegans.

In C. elegans, DNA N⁶-adenine methylation is catalysed by DAMT-1 enzyme (a putative DNA N⁶-adenine methyltransferase) encoded by the open reading frame C18A3.1, while demethylation of 6 mA is achieved by NMAD-1 (N⁶-methyladenine demethylase) encoded by F09F7.7 (Fig. 4a)^36,38. To examine the relationship between age and the rate of N⁶-adenine methylation in TE loci, we developed a PCR- (polymerase chain reaction) based method to detect relative 6 mA levels at any selected DNA stretch being present in a high copy number in the genome (Fig. 4b, and Supplementary Fig. 7). The method is not only relatively simple, but highly reliable (sequence specific; see Supplementary Fig. 7) unlike the most widely used techniques—including single molecule real-time sequencing (SMRTseq), liquid-chromatography-tandem mass spectrometry (LC-MS/MS) and hybridization with a 6mA-specific antibody—developed to date to identify 6 mA sites in a genome, which have been revealed to frequently generate artefacts^40,41,42,43. In addition, these previous methods are incapable of accurately identifying 6 mA levels at specific genomic sites in individual genomes isolated from a tissue sample. The PCR-based method introduced here is a further development of a previously described technique that involves a DNA methylation-sensitive/dependent enzymatic cleavage of genomic DNA and a subsequent PCR amplification of a selected target (digested) site⁴⁴. Because PCR amplification uses the intact, undigested DNA fragments as template and N⁶-adenine methylation in general is a very rare event (only a very few individual genomes within a tissue sample are N⁶-adenine methylated at a given site), the technique is not suitable for accurately determining 6 mA levels at specific DNA sequences. The improved method involves a DpnI-mediated restriction cleavage of genomic DNA samples which is sensitive to the methylated status of the adenine nucleobase found in the restriction site (5′-GA^MeTC-3′), and a subsequent ligation of an adaptor DNA fragment called linker to the cleaved site, serving as a template for the upstream part of the forward primer that is used in a subsequent PCR reaction (so, the forward primer is simultaneously specific to a downstream part of the linker and the adjacent upstream genomic sequence at the target site; Fig. 4b, and Supplementary Fig. 7). This way the quantity of the digested, methylated DNA fragments at a given site can be determined directly in a tissue sample. Since a given adenine site is randomly N⁶-methylated among the individual genomes isolated from the tissue sample examined (genomic DNA is isolated from numerous cells), our method can be used to determine precisely relative (as normalized to an internal control) 6 mA levels at the selected sites.

**Fig. 4: N⁶-methyladenine levels at active transposable elements increase with age.**

In a control experiment to evaluate our method, we found that three different TEs, Cele14, Tc1 and Tc3, display elevated 6mA levels in animals deficient in NMAD-1 function, as compared with the wild-type and DAMT-1 defective backgrounds (Fig. 4c, and Supplementary Fig. 8). Then, relative 6mA levels at these TE sequences were determined in young (1-day-old) versus aged (11-day-old) adults. According to the results, 6mA deposition was markedly increased with age on each TE family examined (Fig. 4d–g, and Supplementary Figs. 9a–c). The increase in relative 6mA levels between young and aged adults was around 5-6-fold in all cases. These data suggest that the epigenetic process DNA N⁶-adenine methylation influences the mobilization of active TEs during ageing.

NMAD-1 and DAMT-1 affect the expression of active transposable elements and influence lifespan

It was found previously that DNA N⁶-adenine methylation is positively correlated with TE expression in Drosophila³⁷ but not in mice³⁹. These fly data suggest that this epigenetic modification affects TE activity. To prove that N⁶-adenine methylation promotes transposition during the adult lifespan, thereby contributing to ageing, we measured relative transcript levels of Tc1, Tc3 (the most active TEs in C. elegans) and Tc14 elements in wild-type versus NMAD-1 defective and in control (wild-type animals fed with the empty RNAi vector only) versus DAMT-1-depleted genetic backgrounds at given adult stages [total RNA samples were isolated at day 2 from wild-type animals and nmad-1(−) mutants and at day 6 from RNAi control and damt-1(RNAi) animals]. Results we obtained indicate that a loss-of-function mutation in nmad-1, ok3133, considerably elevates, while downregulation of damt-1 markedly lowers, the expression of TEs examined relative to the corresponding control (Fig. 5a–d, and Supplementary Fig. 9d–f and Supplementary Table 4). The only exception was Tc14, transcript levels of which in the NMAD-1 defective background displayed a tendency to grow but this change was not significant compared to control (Fig. 5a, d). It is possible that Tc14 becomes active following the early adult stages (day 2) only (in nmad-1(−) mutants, total RNA samples were isolated at the adulthood of day 2). Nevertheless, one can conclude that TE expression is promoted by N⁶-adenine methylation in nematodes. This epigenetic process is likely to increase TE expression in somatic cells in an age-dependent manner. Consistent with this assumption, we found, using an NMAD-1::GFP translational fusion reporter, that nmad-1 is expressed in essentially all somatic cells during development and throughout adulthood (Fig. 5e, f).

**Fig. 5: N⁶-adenine methylation at active transposable elements promotes their expression and accelerates ageing.**

We next determined the lifespan of animals that are defective for either NMAD-1 or DAMT-1. Both nmad-1(ok3133) mutants and nmad-1(RNAi) nematodes died significantly earlier than the corresponding controls (Fig. 5g, and Supplementary Fig. 10a–c and Supplementary Table 1). In the nmad-1(ok3133) mutant background, the expression of Tc1 and Tc3 is elevated compared to wild-type (Fig. 5a–c). Thus, lifespan reduction triggered by NMAD-1 deficiency may be due to, at least in part, increased mobilization of Tc1 and Tc3. We showed that simultaneous downregulation of Tc1 and Tc3 in nmad-1(ok3133) mutant animals significantly increases lifespan as compared with RNAi control [nmad-1(ok3133) mutants treated with the empty RNAi vector only] (Supplementary Fig. 10e). In contrast, an inactivating mutation in damt-1, gk961032, and damt-1 downregulation each increased lifespan (Fig. 5h, i, and Supplementary Fig. 10d and Supplementary Table 1). Longevity effect conferred by gk961032 allele was particularly significant because the mutation, which is a large deletion, eliminates damt-1 and four additional genes from the genome (Supplementary Fig. 10f, g), and these latter genes may have a considerable contribution to viability. Because a previous study reported a slight lifespan reduction in damt-1(gk961032) mutant animals⁴⁵, prior to performing this set of lifespan assays, we outcrossed the mutant strain with the wild-type genetic background several times to avoid the effect of possible background mutations, and applied FuDR to suppress germline activity. Under these conditions, the gk961032 mutation conferred a long-lived phenotype (Fig. 5h, and Supplementary Fig. 10d and Supplementary table 1). Taken together, we propose that DNA N⁶-adenine methylation contributes to nematode ageing through activating TEs.

N ⁶-methyladenine deposition on active transposable elements progressively increases with age

Because adenine nucleobases at active TE stretches are more N⁶-methylated at advanced ages than at young adulthood (Fig. 4d–g and Supplementary Fig. 9a–c), we next asked whether the rate at which this chromatin modification process takes place is about proportional to the age of the animal. In other words, can the relative N⁶-methylation level of a given adenine site in a TE locus within a tissue sample be predictive for chronological age? To this end, we determined relative 6 mA levels at Tc1, Tc3 and Tc14 sequences throughout the adulthood, and found that they do gradually elevate as the organism ages (Fig. 6a, b, and Supplementary Table 4). 6mA levels at Tc1 stretches were also significantly higher in old adults than in young ones when animals were maintained on agar media without FuDR (Supplementary Fig. 10h). In contrast, cep-1 (C. elegans p53-like protein), which was chosen as a non-TE-like gene encoding the nematode orthologue of mammalian p53, did not exhibit such an age-dependent change in 6mA levels (Fig. 6a, b, and Supplementary Table 4). Rather, relative 6mA levels were nearly constant in the cep-1 locus during the entire adult lifespan.

**Fig. 6: N⁶-methyladenine levels at active transposable elements gradually increase with age during the adult lifespan.**

We also assessed the age-dependent dynamics of N⁶-adenine methylation at Tc1, Tc3 and Tc14 stretches in long-lived daf-2(−) (dauer formation defective) mutant animals, which are defective for insulin/IGF-1 (insulin-like growth factor) signalling and live two-times longer than wild-type animals under conditions that permit development into adulthood⁴⁶. Interestingly, the age-dependent rates at which relative 6mA levels increased in these TE loci were highly similar to those found in wild-type animals; 6mA deposition on Tc1, Tc3 and Tc14 sequences also gradually increased during the adult lifespan but this increment took place about half as fast as in the wild type (Fig. 6c, d, and Supplementary Table 4). These observations suggest that daf-2(−) mutant adults, which typically exhibit a lifespan roughly twice that of wild type, display a corresponding delay in reaching the relative maximum levels of 6mA mark at these genomic sites as compared with control (the wild-type background) (Fig. 6e–g).

Because daf-2(−) mutant animals live two times longer than wild-type, relative 6mA levels at the end of life (at the maximal lifespan) are about similar in both genetic backgrounds (Fig. 6e–g). Thus, relative 6mA levels in active TE loci appear to correlate to the age of the animal with a given genotype, suggesting N⁶-adenine methylation as an important process in ageing control. This is consistent with our next observation that there is a significant decrease in TE expression at a given adult stage in daf-2(−) mutant animals as compared to the wild type at the same stage (Supplementary Fig. 11 and Supplementary Table 5). In this attempt, we analysed freely available RNAseq data obtained from wild-type and daf-2(−) mutant animals at comparable (1-2-day-old) adult stages⁴⁷, and identified 46 TE families, the expression of which were significantly (P < 0.05) changed between the two genetic backgrounds. 39 out of these TEs were expressed at significantly lower levels in daf-2(−) mutants (Supplementary Fig. 11a and Supplementary Table 5). At this early adult stage, transcript levels of Tc1 and Tc3, but not Tc14, were decreased relative to control (Tc14 may be active at later—mid and advanced—adult stages only) (Supplementary Fig. 11b–d and Supplementary Table 5). We put forward that in a mutant strain living two times longer than the wild type, the cumulative expression of all active TE elements at a certain adult stage (day) is around half the level of that for the wild type at the same stage. Certain TE elements however can be overexpressed in this mutant genetic background relative to wild-type. In sum, 6mA deposition on active, specific TE sequences may serve as an epigenetic signature for determining age from DNA (i.e. a tissue sample).

The effects of DAMT-1, a putative DNA N⁶-adenine methyltransferase, and NMAD-1 N⁶-methyladenine demethylase on the methylation status of adenine nucleobases in Tc1 and Tc3 loci were also tested at the 1- and 5-day adult stages (Fig. 6h, i). We found that in the nmad-1(−) mutant background, relative 6mA levels are equally high at both adult stages. These results imply that low levels of 6mA in these loci at young adult stages is a consequence of NMAD-1 activity. In damt-1(−) mutants, however, relative 6mA levels were low but still detectable at the adult day 1 and almost completely absent at day 5 (Fig. 6h, i). Based on these data one can propose that DAMT-1 mediates N⁶-adenine methylation in these TE loci throughout adulthood, and that during development DNA N⁶-adenine methylation is also catalysed by (a) yet identified enzyme(s) that is/are distinct from DAMT-1.

We also determined the rate at which a selected individual Tc3 stretch (R10H1.3) is excised from its original genomic site during the adult lifespan in control versus nmad-1(−) and damt-1(−) mutants, as well as in daf-2(−) mutant animals (Fig. 6j). In good agreement with results above (first, relative 6mA levels at active TE sequences increase with age and N⁶-adenine methylation promotes ageing, and, second, 6 mA accumulates in TE loci at a halved rate in long-lived mutants deficient in DAF-2 activity relative to wild-type), NMAD-1 deficiency strongly increased, while defects in DAMT-1 and DAF-2 function lowered, Tc3 (R10H1.3) transposition as compared with the control genetic background (Fig. 6j, and Supplementary Fig. 12). Thus, the rate of the ageing process is strongly influenced by N⁶-adenine methylation at active TE stretches.

Single-molecule real-time sequencing verifies an age-associated increase in N ⁶-methyladenine levels at active transposable elements

To validate the above PCR-based data, according to which 6mA deposition on active TE stretches gradually increases with age, we applied SMRTseq technology serving as a direct method currently available to detect chemical modifications of cytosine and adenine nucleobases⁴⁸. Genomic DNA samples isolated from synchronized adult populations at the 1- and 5-day stages were subjected to the analysis. To ensure that the results and conclusions are not influenced by the amount and quality of genomic DNA prepared from the two age groups, that is, variations in library constructions and differences in sequencing coverage, we did not directly compare 6mA sites between 1- and 5-day-old adult animals. Rather, 6mA sites in different loci were identified within a given age group; non-TE-like versus TE sequences were separately assessed at young (1-day) and middle-age (5-day) adult stages (Fig. 7).

**Fig. 7: Single molecule real-time sequencing reveals that N⁶-adenine methylation preferably occurs at transposable elements during adulthood.**

Similar to the PCR-based 6 mA results (Fig. 6a, b), SMRTseq data revealed that 6 mA epigenetic marks increase with age at TE sequences relative to non-TE-like genes (Fig. 7a, b). Utilizing the coverage provided by SMRTseq technology, genes and TE sequences were listed and ordered as a function of their relative 6 mA quantity in 1-day-old adults. From these data we extracted how the relative quantity of 6 mA marks on the top 15 most methylated genes and TEs changes during ageing (Fig. 7c, d). We found that the top 15 most methylated genes in 1-day-old animals are an order of magnitude more methylated than the top 15 most methylated TEs (Fig. 7c). Contrary to this finding, N⁶-adenine methylation of the same set of TEs was almost one order of magnitude higher than the relative N⁶-adenine methylation level of the selected genes in 5-day-old animals (Fig. 7d). It is intriguing that the top 15 most N⁶-adenine methylated genes at day 1 became highly demethylated when adults reached the older stage. Relative N6-adenine methylation levels of the top 15 most N⁶-adenine methylated TE families at day 1 however further increased during this period.

Next, we aimed to compare the PCR-based results with those provided by the SMRTseq assay. To do this, we examined relative 6 mA levels at each Tc3 stretch in animals with various ages, and compared to that of cep-1 gene (Fig. 7e, f). We found that in 1-day-old animals, cep-1 has twice as much modified adenines than the Tc3 regions do (Fig. 7e), but in 5-day-old animals the relative N⁶-adenine methylation of the Tc3 regions was nearly four times higher than the relative N⁶-adenine methylation level of cep-1 locus (Fig. 7f). Based on both PCR- and SMRTseq-based 6 mA data (Figs. 6 and 7), we conclude that TEs are likely to exhibit a differential N⁶-adenine methylation during ageing.

Here, it is worth mentioning that SMRTseq technology has been criticized recently as it is often unable to specifically detect 6 mA sites from sample DNA due to a significant detection noise (nearby cytosine methylation), and the presence of bacterial DNA contamination and N⁶-adenine methylated RNA^42,43. Despite these limitations, potential artefacts provided by the technology should affect both samples (1 and 5 days old adult nematodes) with a similar extent, so total 6 mA quantities at distinct TE families are comparable between young and older adults (our primary goal was not to accurately identify individual 6 mA sites within a single genome, rather we compared relative 6 mA levels in specific genomic loci between two samples, using the same detection approach).

Downregulation of transposable elements increases thermotolerance in C. elegans

As shown above, N⁶-adenine methylation on TE sequences occurs at a halved rate in daf-2(−) mutants having a double lifespan relative to wild type (Fig. 6a–g). We therefore asked whether environmental stress factors can modulate 6mA accumulation at TE sites to explain the altered survival of nematodes deficient in a certain stress response pathway. To address this issue, we tested the tolerance of animals to high temperatures when TEs were inhibited (animals maintained at 20 °C were transferred at 35 °C for 5 h, then shifted back to 20 °C, and assayed for those being alive on the next day⁴⁹). Relative 6mA levels at Tc1 and Tc3 sequences were first compared in untreated control versus heat-stressed nematodes. Results obtained show that the N⁶-adenine methylation process at these specific sites is activated in response to high temperatures (Fig. 8a, b, and Supplementary Table 6). This epigenetic change was coupled with elevated Tc1 and Tc3 expression (Fig. 8c, and Supplementary Table 6). Tc1 and Tc3 elements were then downregulated simultaneously, and this intervention considerably increased the survival of treated animals relative to untreated control (Fig. 8d). To strengthen these results, we also measured heat shock tolerance in damt-1(−) and nmad-1(−) mutant animals (Fig. 8e, and Supplementary Table 6). We found that the former survive better, while the latter tends to be more sensitive to heat stress, than the wild type. Furthermore, nmad-1(−) mutation significantly suppressed the increased tolerance of long-lived mutants defective for DAF-2 function (Fig. 8e, and Supplementary Table 6). Thus, nematodes with a lowered rate at which 6mA accumulates in TE loci live longer and more resistant to heat stress than normal. In contrast, a genotype conferring a faster 6mA accumulation dynamics at TE sequences limits lifespan and renders the animal more sensitive to heat stress.

**Fig. 8: Downregulation of transposable elements promotes tolerance in nematodes to heat stress.**

Heat stress data above indicate that TEs mediate cellular stress through acting downstream of the heat stress response pathway in this organism. Indeed, neither Tc1 + Tc3 RNAi co-treatment nor inactivating mutations in damt-1 and nmad-1 could influence the expression of hsp-16.2 (heat-shock protein), which is a target gene of the nematode heat stress response pathway (Supplementary Fig. 13a, b and Supplementary Table 7)⁵⁰. We also tested the expression of sod-3 (superoxide dismutase) and gcs-1 (gamma glutamyl cysteine synthetase), targets of the C. elegans nutrient-sensing and oxidative stress response pathways, respectively^50,51, in control vs. Tc1 + Tc3 co-downregulated genetic backgrounds, but detected no obvious difference between the treated and corresponding control samples (Supplementary Fig. 13c–f and Supplementary Table 7). Similarly, downregulation of nmad-1 and damt-1 did not alter hsp-16 and gsc-1 expression (Supplementary Fig. 13g, h and Supplementary Table 7). This raises the possibility that TEs and factors that influence 6mA levels at TE sequences function downstream of stress response pathways in lifespan determination. Alternatively, TEs and factors influencing 6mA levels may act parallel to these pathways.

In the light of the above results, we next analysed available RNAseq data in wild-type versus daf-2(−) mutant animals⁵², now focusing on damt-1 and nmad-1 expression. The DAF-2/IGF-1-mediated pathway responses to nutritional stress via inhibiting the FOXO-like transcription factor DAF-16. In daf-2(−) mutant backgrounds, DAF-16 is (hyper)active and regulates target genes required for enhanced stress tolerance and longevity. The analysis revealed that DAF-16 represses damt-1 but activates nmad-1 expression (damt-1 transcript levels were lower, while nmad-1 transcript levels were higher, in mutant animals defective for DAF-2 function as compared with normal; Supplementary Fig. 14a and Supplementary Table 8). Although these gene expression changes were rather modest, but still below the significance limit (P < 0.05). In good accordance with these results, another set of available RNAseq data from wild-type, daf-2(−) single mutant and daf-2(−);daf-16(−) double mutant animals⁵³ showed that the effects of DAF-2 on damt-1 and nmad-1 expression tend to be dependent of DAF-16 activity (Supplementary Fig. 14b and Supplementary Table 8). These findings also denote that 6 mA levels increase with a lower rate at TE sequences in daf-2(−) mutants compared to wild-type.

Why does the N⁶-adenine methylation pattern at TE stretches change with age under normal, stress-free conditions? A reasonable explanation would be a life-long, progressive increase in damt-1 expression (growing N⁶-adenine methyltransferase activity) or a gradual decrease in nmad-1 expression (failing N⁶-methyladenine demethylase activity). Both alternatives would lead to the same outcome, increasingly growing 6mA levels in TE regions. Thus, we determined the expression activity of damt-1 and nmad-1 genes throughout adulthood, but found no significant changes that could underlie the gradually increasing mobilization of TEs over time (Supplementary Fig. 15). The expression of damt-1 remained nearly constant during ageing (Supplementary Fig. 15a), whereas nmad-1 exhibited even an increased tendency to produce transcripts (Supplementary Fig. 15b). The latter results were supported at the level of NMAD-1 protein accumulation (Supplementary Fig. 15c, d). An earlier study of transcriptional profile of C. elegans ageing has provided similar results regarding these genes⁵². We conclude that age-associated changes in 6 mA levels at TE sequences are driven by mechanisms other than altered expression of genes encoding enzymes involved in DNA 6 mA modification.

Discussion

Using reverse transcriptase inhibitors blocking retrotransposon mobilization, researchers could extend lifespan in Drosophila²² and delay cellular senescence in mice²⁵. However, these compounds may have multiple targets, thereby affecting the function of other proteins in addition to reverse transcriptases. Our present study provides the first effective RNA interference-mediated downregulation of active TE families in a genetic model system, C. elegans. We showed that specific silencing of Tc1 and Tc3, the most active TEs in this organism²⁶, slows down the ageing process at different temperatures (Fig. 1a, c, e, g, and Supplementary Fig. 2a–d). In addition, co-repressing these mobile DNA elements additively promoted longevity (Fig. 2a–c). Downregulation of Tc14, which is another active TE in this organism²⁶, also conferred a statistically significant lifespan advantage over control (Fig. 1i, and Supplementary Fig. 2g). In good accordance with these results, somatic expression of Piwi proteins, which normally inhibit TE activity in the non-ageing germline, lengthened lifespan in transgenic animals (Fig. 3a, d, g, and Supplementary Fig. 4). Taken together, we conclude that the rate at which somatic cells age is influenced by TE activity, and that DNA transposons contribute to lifespan determination.

C. elegans appears to be a unique model for exploring the contribution of TEs to ageing. First, only 12% of its genome derives from TE-related repetitive sequences, the majority of which are inactive elements²⁶. An active TE family in this species contains only 5-31 copies, the inhibition of which could promote longevity (Figs. 1a, c, e, g, i and 2a–c, and Supplementary Figs. 2a–d, g). Second, it was suggested earlier that TEs destabilize the genome of hyper-proliferating somatic cells, thereby protecting tissues from undergoing tumorigenesis¹⁵. Somatic expression of Piwi proteins inhibiting TE activity indeed effectively induces tumour development in mice⁵⁴, and tumorous cell lines tested so far have been proven all positive for Piwi activity¹². Thus, repression of TEs in somatic cells may lead to cancer (a pleiotropic effect), causing an early death rather than extending lifespan. However, tumours are yet to be induced in the C. elegans soma, a special feature of the organism that enables TE inhibition without causing cancer.

Although longevity advantage conferred by downregulation of single TE families was quite modest, this can be explained by the relative ineffectiveness of RNAi treatments (Fig. 1b, d, f, h, j), the high copy numbers of genes targeted for downregulation (10-31 copies per a TE family; Supplementary Fig. 1), the rather late onset of depletion during lifespan (see the Methods section), and the remaining mutagenic effect of other active TE families that were not targeted in a given trial.

Results obtained here demonstrate that active TE stretches become increasingly N⁶-methylated on adenine nucleobases during the adult lifespan (Figs. 4, 6 and 7). This epigenetic modification progressively elevates TE transcription, which, in the absence of Piwi-piRNA complexes, may cause considerable levels of mutagenic events (genomic instability) in somatic cells at advanced ages (Figs. 2e–h and 6j). DNA N⁶-adenine methylation has been recently discovered in distantly related animal species^36,37,39, but its functional significance remains largely unclear. This type of DNA modification was suggested to have a role in facilitating TE transposition in Drosophila³⁷ while inhibit TE activity in mammals³⁹. Several recent studies however debate the presence of 6mA epigenetic mark in animal genomes, and attribute it to methodological errors^40,41,42,43. Here, we developed a reliable (sequence-specific and free of artefacts caused by bacterial DNA or RNA contamination) PCR-based method to accurately determine relative 6 mA levels at selected genomic sites (Fig. 4b, and Supplementary Fig. 7). Using this technique, we showed that in C. elegans 6mA levels in TE loci gradually increase with age (Fig. 6a, b). Thus, N⁶-adenine methylation at specific genomic sites influences the rate at which somatic cells age. Since 6mA deposition on active TE stretches is proportional with the age of the organism (Fig. 6a, b), it is possible that this epigenetic modification will be a target to accurately determine age from DNA. We propose that 6mA levels at active TE stretches serve as a signature of ageing.

We also demonstrated that the excision dynamics of a Tc3 stretch correlates with the actual rate of N⁶-adenine methylation at the given genomic site (Fig. 6j). Thus, there is a causative relationship among these processes; N⁶-adenine methylation increasingly occurs at active TE sequences (Fig. 6a–g), and progressively elevates their transcriptional activity (Fig. 5a–d) and mobilization (Figs. 2e, f and 6j). This causes genomic instability (insertional mutations in functional DNA sequences) (Figs. 2g, h and 6j), leading to an accelerated ageing rate and, as a consequence, shortened lifespan (Figs. 1–3 and 5). However, the potential effects of DAMT-1 and NMAD-1 on other genes in lifespan determination cannot be excluded.

daf-2(−) mutant animals live two times longer than normal⁴⁵. In this mutant genetic background, 6mA marks accumulated with age in TE loci at a halved rate relative to wild-type (Fig. 6a–g). So, longevity correlates to a lower rate at which 6mA accumulates in TE loci. In daf-2(−) mutants, the expression of damt-1 gene was decreased, while the expression of nmad-1 gene was increased, as compared with the wild type (Supplementary Fig. 14). Thus, the transcription factor DAF-16, which operates as the effector of the DAF-2/IGF-1-mediated, nutrient-sensing stress response pathway, may modulate the activity of genes determining 6mA levels. Similar to DAF-16, other transcription factors mediating the effect of stress response pathways, such as SKN-1 and HSF-1 (these proteins participate in oxidative and heat stress responses, respectively), may also be involved in controlling 6mA levels at specific genomic sites. In a good agreement with this suppose, Tc1 and Tc3 became hypermethylated on adenine nucleobases and over activated in response to heat stress (Fig. 8a, b). The fact that TE mobilization elevates at high temperatures is well documented in other organisms as well^55,56. Moreover, oxidative, radiation and nutritional stress factors were also reported to promote TE transcription^22,57,58. Based on these data, we suggest that TEs act downstream of stress response pathways in lifespan determination (Fig. 8f).

Under normal conditions, however, other factors should modulate the rate at which 6mA mark accumulates with age in active TE loci. To examine whether this mechanism is established at the transcriptional control of damt-1 encoding a putative DNA N⁶-adenine methyltransferase and/or nmad-1 coding for a DNA N⁶-methyladenine demethylase, we assessed transcript levels of these genes at different stages of adulthood but found no significant alterations that could explain why N⁶-adenine methylation patterns of TEs increase during ageing. damt-1 transcript levels were nearly constant, whereas nmad-1 expression and NMAD-1 protein accumulation even increased during the adult lifespan (Supplementary Fig. 15). Presumably, the latter changes are merely a consequence of a genetic compensatory mechanism^59,60 to overcome a gradual loss in NMAD-1 activity. The age-related accumulation of 6 mA epigenetic mark in active TE sequences is thus not a consequence of changes in transcript/protein levels of factors mediating DNA N⁶-adenine methylation and N⁶-methyladenine demethylation. A gradual increase in 6 mA levels at specific genomic sites during adulthood may be established at different levels. These data prompted us to propose a model in which a factor unidentified to date (indicated by X in Fig. 8f) would gradually increase the activity of DAMT-1 N⁶-adenine methyltransferase or gradually lower the activity of NMAD-1 6 mA demethylase during adulthood (Fig. 8f). These two options would lead to the same result: a progressive, lifelong accumulation of 6 mA mark in active TE loci. Such a factor could be a regulatory protein or an environmental component. Indeed, oxygen was reported to act as a co-factor of NMAD-1/AlkB family of demethylases⁶¹, and hypoxic milieu is a general feature of ageing cells^62,63. Thus, a gradual reduction in intracellular oxygen levels throughout adulthood may explain why NMAD-1 activity decreases during ageing.

The important question of why TEs are differentially N⁶-adenine methylated however remains unresolved. In numerous animal species, de novo DNA methylation that occurs during early development primarily generates 5-methylcytosine (5mC) marks, most of which resides in TE loci^64,65. TEs are considered intragenomic parasites, and 5-cytosine methylation at TE sequences may represent a defence mechanism that has evolved against the invasion of mobile DNA elements to preserve genomic integrity. The mechanisms underlying 5-cytosine methylation involve a Piwi/piRNA-directed machinery, and piRNA are transcribed from TE-related sequences located in piRNA gene clusters⁶⁶. DNA N⁶-adenine methylation may also be mediated by a piRNA-driven mechanism, explaining why 6 mA preferably accumulates at TE sequences during ageing. Further studies will certainly clarify this important issue.

The regulation of ageing has been intensely studied and relatively well understood during the last decades^67,68. However, despite its medical, social and economic significance, the mechanisms underlying the ageing process, that is, genetic factors that actually cause ageing, remain largely unknown. This problem may stem from the fact that the mechanisms involve multicopy factors, the parallel inhibition of which has not yet been established. TEs constitute a large fraction of eukaryotic genomes, and their activity, as shown here, contributes to ageing. To test whether TEs represent an underlying mechanism of the ageing process, one should simultaneously block the mobilization of essentially all active TE families (both retroelements and DNA transposons) in an organism, as they are effectively repressed in the non-ageing germline by the Piwi-piRNA pathway, and detect extreme longevity.

Methods

Strains and genetics

Nematodes were maintained on standard Nematode Growth Medium (NGM) as described previously^69,70, and fed with Escherichia coli OP50 bacteria. The following C. elegans strains were used in this study.

Bristol (N2) as wild-type

DR1344 Bergerac BO

CB1370 daf-2(e1370)III

TJ1060 spe-9(hc88)I; rrf-3(b26)II

VC2552 nmad-1(ok3133)III

VC40319 C18A3.1(gk961032)II

TTV680 unc-119(ed3)III; eluEx390[p_hsp-16.2::Tc3 + unc-119(+)]

TTV681 unc-119(ed3)III; eluEx391[p_hsp-16.2::prg-1 + unc-119(+)]

TTV614 unc-119(ed3)III; eluEx330[p_hsp-16.2::prg-1::gfp + unc-119(+)]

TTV491 unc-119(ed3)III; eluEx364[p_hsp-16.2::prg-2 + unc-119(+)]

TTV615 unc-119(ed3)III; eluEx358[p_hsp-16.2::ppw-2 + unc-119(+)]

TTV721 unc-119(ed3)III; eluIs314[p_nmad-1::NMAD-1::GFP + unc-119(+)]

CF1553 muIs84[(pAD76) sod-3p::GFP + rol-6(su1006)]

TJ375gpIs1[hsp-16.2::GFP]

TTV804 nmad-1(ok3133)III; gpIs1

TTV805 nmad-1(ok3133)III; muIs84

TTV806 C18A3.1(gk961032)II; gpIs1

TTV807 C18A3.1(gk961032)II; muIs84

Prior to performing lifespan assays, VC2552 and VC40319 strains were isogenized by outcrossing 3 times with the wild type.

Lifespan assays

Lifespan assays were carried out at 15 °C, 20 °C and 25 °C, as described previously^69,70,71. For synchronization, 20-30 gravid, well-fed adults were transferred to a new agar plate containing NGM seeded with E. coli [OP50 or HT115(DE3)] bacteria to lay embryos for 4–5 h, and then removed. Alternatively, embryos were prepared by NaOH/hypochlorite treatment. Approximately 60-70 F1 young adults were transferred to new NGM plates supplemented with 300–400 mg/ml FUdR (5-fluoro-2′-deoxyuridine, Sigma #50-91-9) (t = 0). Sterile F1 adults were then assayed. Animals that climbed up the wall of plastic dishes or exhibited a protruded/burst vulval phenotype and thereby died prematurely were excluded from the analysis. Animals were considered dead when they stopped pharyngeal pumping and responding to touching. For measuring the lifespan of TE- (transposable element) specific double-stranded (ds) RNA-treated animals, wild-type hermaphrodites at the young, gravid adulthood were transferred onto RNAi (RNA interference) plates. This relatively late onset of treatment during lifespan allowed TEs to express potential developmental roles (during evolution certain TEs became domesticated and acquired developmental roles)²⁷. Indeed, Tc1 depletion starting from the L1 larval stage caused early lethality instead of promoting longevity. SPSS 17 software was used to calculate mean lifespan and perform statistical analysis. P values for comparing Kaplan-Meier survival curves between two groups were determined using log-rank (Mantel–Cox) tests, and P values for comparing mean lifespans were determined using Independent two-sample Student’s two-sided t tests with Bonferroni correction. E. coli HT115(DE3) RNAi feeding bacteria were grown to OD₆₀₀ = 0.5 in LB medium containing 50 µg/ml ampicillin (Sigma, #69-53-4) in final concentration. L4/young stage adults were transferred to plates containing 300–400 mg/ml FUdR, 50 µg/ml ampicillin and 0.4 mM IPTG (isopropyl β-D-1-thiogalactopyranoside, Sigma #367-93-1) in final concentration. Empty vector- (L4440 and T444T) containing bacteria were used as control.

Generation of DNA constructs and transgenic strains

Overexpressing single components of the Piwi-piRNA pathway in the soma is expected to induce the activity of the pathway because its certain components were found to accumulate somatically^12,34 and Piwi proteins, by themselves, can mediate transcript degradation³⁴. pPD118.28 served as the cloning vector for generating a translational fusion (functional) hs::PRG-1::GFP construct. pPD49.78 served as the cloning vector for somatic overexpression of prg-1 (hs::PRG-1), prg-2 (hs::PRG-2), ppw-2 (hs::PPW-2) and Tc3 (hs::Tc3) full length coding regions. These vectors contain an inducible hsp-16.2 heat shock promoter that is active in the majority of somatic cells. Note that the promoter is leaky, enabling gene expression even at normal culturing temperatures between 15-25 °C. These constructs were cloned by using the following restriction enzymes (in parenthesis), as well as forward and reverse primers. hs::PRG-1::GFP: (BamHI + AgeI) 5′-CGC GGA TCC GCG ATG GCA TCT GGA AGT GGT C-3′ and 5′-TAT ACC GGT GTC AAG AAG AAC AGC TTG TCA CG-3′; hs::PRG-1: (Gibson) 5′-CAA ACT ATA ATC ATC TCA CTG GAT CAA TGG CAT CTG GAA GTG GTC G-3′ and 5′-AAT ACC ATG GTA CCG TCG ACG CTA GTT ACA AGA AGA ACA GCT TGT CAC GAA GAC A-3′; hs::PRG-2: (BamHI + NheI) 5′-CGC GGA TCC GCG CCG ATT GAG ATG ATT CAG AAC GAC-3′ and 5′-CTA GCT AGC TAG TTA CAA GAA GAA CAG CTT GTC ACG-3′; hs::PPW-2: (XmaI + KpnI) 5′-CCC CCC GGG GGG ATG CCT GCT ACA CCG GTT-3′ and 5′-CGG GGT ACC CCG TTA AGC ATT GAC ACG GCG G-3′; hs::Tc3: (NheI + NcoI) 5′-AAA AAA GCT AGC ATG CCT CGA GGA TCT GCC C-3′ and 5′-AAA AAA CCA TGG ATA GTT AAT CGG GTT TCC TTG TGT GCG-3′.

To generate a p_nmad-1::NMAD-1::GFP translational fusion reporter construct, a 2.37 kb-long genomic fragment containing 5′ regulatory and coding regions of nmad-1 was amplified by the following forward and reverse primers: 5′-AAA ACT GCA GGG CCC TTT CCT AGT TTT TGC−3′ and 5′-AAA CCC GGG ATT TCC CCA AAT CCA CAT ATC A-3′. The resulting fragment was cloned into pPD95.75 vector, using PstI and SmaI restriction enzymes. To generate a p_damt-1::DAMT-1::GFP translational fusion construct, a 3.7 kb-long genomic fragment containing of 5′ regulatory and coding regions of damt-1 was amplified by using the following forward and reverse primers: 5′-AAA CTG CAG AGT CCA ACC AAA TCA AAA TCG−3′ and 5′-AAA GGA TCC CGA GTG AAA ACA TGT TCA GAT TG−3′. The resulting fragment was cloned into pPD95.75 vector at the PstI and BamHI restriction sites.

Transgenic animals were generated by microparticle-mediated bombardment (biolistic transformation) into the unc-119(−) (uncoordinated) mutant genetic background, using the co-transformation marker unc-119(+) gene⁷². Transgenic lines were established by picking up F1 animals with normal movement. Using a Biolistic PDS-1000/He particle delivery system (BioRad), 10-15 μg linearized plasmid DNA was bombarded onto unc-119(ed3) mutant hermaphrodites at the L4/adult stages. Fluorescent images were taken by Zeiss AxioImager Z1 epifluorescence microscope.

RNAi constructs

L4440 served as the original RNAi vector (equivalent to pPD129.36; Fire Lab 1997 Vector Kit). The 344 bp-long MCS (Multi Cloning Site) fragment from L4440 was amplified by using left and right primers that contain T7 terminator sequences at both ends, together with 5′ PciI- and 3′ NgoMIV-specific restriction sites. PCR fragment was ligated into L4440 to create a modified RNAi vector, T444T²⁷. This vector can generate specific dsRNAs only, without amplifying vector backbone sequences, thereby producing more effective phenotypes than the original L4440. Tc1, Tc3, Tc14, Cele14, nmad-1 and damt-1 genomic fragments were cloned into T444T by using the following restriction enzymes (in parenthesis), and forward and reverse primers. T444T: (PciI+ NgoMIV) 5′-TTT TTT ACA TGT CAA AAA ACC CCT CAA GAC CCG TTT AGA GGC CCC AAG GGG TTA TGC TAG TAA TAC GAC TCA CTA TAG GGA GAC CGG CAG-3′ and 5′-TTT TTT GCC GGC CAA AAA ACC CCT CAA GAC CCG TTT AGA GGC CCC AAG GGG TTA TGC TAG TAA TAC GAC TCA CTA TAG GGC GAA TTG GGT ACC G-3′; Tc1: (BglII + Acc65I) 5′-CTA GCT AGC TAG AAC AGC TTG ACA ACG ACG TG-3′ and 5′-CTA GCT AGC TAG AAC ATC CTC CGA TCA GCA AG-3′; Tc3: (BglII + Acc65I) 5′-TTT TTT AGA TCT ATG CCT CGA GGA TCT GCC CT-3′ and 5′-TTT TTT GGT ACC GGG TAA GTC TTG TTC TGA GCA TAC ACG-3′; Tc14: (BglII + Acc65I) 5′-TTT TTT AGA TCT AGA TCC GGA CGC GGA AGA CC-3′ and 5′-TTT TTG GTA CCC CAG CTC GTC CCA TGC CTT C-3′; Cele14: (Gibson) 5′-TAA TAC GAC TCA CTA TAG GGC GAA TTG GGT ACA CTG ATA AGG TAT TAC ACG TGG AGT C-3′ and 5′-TAA TAC GAC TCA CTA TAG GGA GAC CGG CAG ATC GAA TTC AAT AAT ACC ACG TGG TGT CAG G-3′; nmad-1: (Acc65I + BglII) 5′-TTT TTT AGA TCT TGT GAG ACA ACG GAA CGT GT-3′ and 5′-TTT TTT GGT ACC TCC TGA TGC ATC TCA ATG GCA-3′; damt-1: (BglII + Acc65II) 5′-TTT TTT AGA TCT CGT TGA ACG TGT CAA TCC CC-3′ and 5′-TTT TTT GGT ACC CAT TGG AAC GCT GGC AAA AAC-3′.

Transcript quantification by semi-qPCR

Total RNA was extracted from 100 synchronized animals, using TRI reagent (Sigma, #T9424), and used for first strand cDNA synthesis by RevertAid First Strand cDNA Synthesis Kit (Thermo Scientific, #K1622). Mean relative mRNA levels were determined by using cdc-42 as a reference gene. Forward and reverse primers and PCR conditions were as follows. cdc-42: 5′-CTT CGA CAA TTA CGC CGT CAC-3′ and 5′-CGA AAT TTC AGG CAC CCA TTT TTC-3′. Initial denaturation at 95 °C for 30 s, then 29 cycles of denaturation at 95 °C for 10 s, annealing at 57 °C for 30 s, extension at 68 °C for 30 s, and final extension at 68 °C for 2 min. Samples were tested at 23, 26 and 29 cycles. Tc1: 5′-TGC AAA TTC AAC GTT CTC CG-3′ and 5′-TTC TCA AAG CGA CGG ATT TC-3′. Initial denaturation at 95 °C for 30 s, then 35 cycles of denaturation at 95 °C for 10 s, annealing at 55 °C for 30 s, extension at 72 °C for 30 s, and final extension at 72 °C for 2 min. Samples were tested at cycles 29, 32, and 35. Tc3: 5′-ATT CTC GAC GCT TGG AAG TC-3′ and 5′-GTT TCC TTG TGT GCG GAT G-3′; initial denaturation at 95 °C for 30 s, then 32 cycles of denaturation at 95 °C for 10 s, annealing at 55 °C for 30 s, extension at 72 °C for 30 s, and final extension at 72 °C for 2 min. Samples were tested at 29, 32 and 35 cycles. Tc14: 5′-GTG AAC CCA TCG TTC GTT TTC-3′ and 5′-CGT GCA ATG GTG GTA GAA AC-3′; initial denaturation at 95 °C for 30 s, then 32 cycles of denaturation at 95 °C for 10 s, annealing at 55 °C for 30 s, extension at 72 °C for 30 s, and final extension at 72 °C for 2 min. Samples were tested at 29, 32 and 35 cycles. prg-1 transgene: 5′-TCT TCA CGA TGA TGC CAA TG-3′ and 5′-TGG GAC AAC TCC AGT GAA AAG-3′; initial denaturation at 95 °C for 30 s, then 29–35 cycles of denaturation at 95 °C for 10 s, annealing at 55 °C for 30 s, extension at 68 °C for 30 s, and final extension at 68 °C for 2 min. ppw-2 transgene: 5′-TTT GAA GAA AAT GAC CGA TGG-3′ and 5′-AGA TGG CGA TCT GAT GAC AG-3′; initial denaturation at 95 °C for 30 s, then 29–35 cycles of denaturation at 95 °C for 10 s, annealing at 57 °C for 30 s, extension at 68 °C for 30 s, and final extension at 68 °C for 2 min. Where products were separated by agarose gel electrophoresis we used Thermo Scientific™ GeneRuler 1 kb Plus DNA Ladder (#SM1331).

RNA extraction and quantification of transcript levels by real-time qPCR

Total RNA was extracted from 50 synchronized animals per each sample, using TRI reagent (Sigma, T9424), and RNA isolation was performed according to the Direct-zol^TM RNA MiniPrep kit protocol (Zymo Research, R2050). Total RNA samples were used for first strand cDNA synthesis by RevertAid First Strand cDNA Synthesis Kit (Thermo Scientific, K1622). Real-time quantitative-PCR was performed on LightCycler 96 System (Roche, FastStart Essential DNA Green Master, 06402712001) under the following conditions: denaturation at 95 °C for 10 min, followed by 45 cycles of amplification (10 s, 95 °C; 10 s, 58 °C and 20 s, 72 °C). In this method, the fluorescence of double stranded DNA-binding SYBR-Green dye indicates the PCR products. Melting curve analysis was performed to confirm the correct PCR product size and the absence of nonspecific bands. Mean relative mRNA levels were determined by normalizing the PCR threshold cycle number of tc1, tc3, tc14, nmad-1, damt-1 and prg-1 with that of cdc-42 reference gene. For qPCR amplification, the following primers were used. cdc-42: 5′-CTT CGA CAA TTA CGC CGT CAC-3′ and 5′-CGA AAT TTC AGG CAC CCA TTT TTC-3′; tc1: 5′-AAA AAA GCT AGC ATG GAT CGC AAC ATC CTC CGA T-3′ and 5′-CC CAC GCA ACT CGA GCC-3′; tc3: 5′-GGT TGT CTT CTC C-3′ and 5′-GTT TCC TTG TGT GCG GAT G-3′; tc14: 5′-GTT TCT ACC ACC ATT GCA CG-3′ and 5′-TAC CAT TCG GCA CAG AGA AC-3′; nmad-1: forward 5′-AAC GGA ACG TGT GAA AAA GC-3′ and reverse 5′-CGA ATG TGA TCG TAA ATG AAG ACT-3′; damt-1: forward 5′-CCT CTT CGA CTT GAT CAT TGC- 3′ and reverse 5′-GAA CTT CTT CGT CCA TTT GAT ATG T-3′; prg-1: forward 5′-GTG AAC ATT CCG CTT AAA AAC AC-3′ and reverse 5′-CTT CAA GGC TTT TCG AAC AAA G-3′. Statistical analysis was performed by RStudio (version 0.99.903) software. Since data did not show a normal distribution, Mann-Whitney U-test was used to determine significance levels.

Determining N ⁶-methyl adenine (6mA) levels at TE stretches

Adenine N⁶-methylation is generally detected by SMRT (single molecule real-time) sequencing⁴⁸, which is still a rather expensive and laborious technique. Here we developed a relatively simple, PCR-based method to detect the relative N⁶-methylation level of any adenine base at any repetitive sequence. Genomic DNA from adults with age of days 1 and 11 was digested with DpnI or PvuII for 20 min at 37 °C, then a linker DNA (5′-TAG ATC TGA CCT AAC GGT AAG AGA GTT TCA TAA TAT TTT TTT TTT TTT TTT AT―TAT GAA ACT CTC TTA CCG TTA GGT CAG ATC TA-3′) was ligated to the digested genomic DNA for 24 h at 4 °C. DpnI cleaves DNA only when its restriction site (GATC) is methylated at position A, PvuII cuts any DNA sequence containing its restriction site (CAGCTG). Mean relative DNA levels were determined by using Tc3 PvuII primers as reference. Forward and reverse primers, and PCR conditions were as follows. Tc1 (DpnI): 5′-TCT TAC CGT TAG GTC AGA TCT ATC C-3′ and 5′-CAT GGG GAT AGC TTT CCA AG-3′, initial denaturation 95 °C, 30 s, then 40 cycles of denaturation at 95 °C for 10 s, annealing and extension at 50 °C for 30 sec; Tc3 (DpnI): 5′-ATG AAA CTC TCT TAC CGT TAG GTC AGA TCT ATC TC-3′ and 5′-TGA CCA AAC TTT TCA GCT GGT TGT CC-3′, initial denaturation 95 °C for 30 s, then 35 cycles of denaturation at 95 °C for 10 s, annealing and extension at 57 °C for 30 sec; Cele14 (DpnI): 5′-ACT CTC TTA CCG TTA GGT CAG ATC TAT-3′ and 5′-CCA GAA AAA TTG TGA CGT CAG CAC GCT C-3′, initial denaturation at 95 °C for 30 s, then 35 cycles of denaturation at 95 °C for 10 s, annealing and extension at 57 °C for 30 s; Tc3 (PvuII) (Reference): 5′-TGA AAC TCT CTT ACC GTT AGG TCA GAT CTA CTG-3′ and 5′-TCA ATA GTT AAT CGG GTT TCC TTG TGT GCG GAT GAT C-3′, initial denaturation at 95 °C for 30 s, then 32 cycles of denaturation at 95 °C for 10 s, annealing and extension at 57 °C for 30 s.

To determine relative 6mA levels at TE stretches and cep-1 locus throughout adulthood, genomic DNA was isolated from staged population of wild-type and daf-2(e1370) mutant adult animals grown at 20 °C, using Thermo ScientificGeneJET GenomicDNA Purification Kit (#K0721). Genomic DNA was then digested with DpnI at 37 °C for 20 min, and ligated with a linker fragment at 4 °C for overnight. 10 pg DNA was used for PCR amplification to detect relative 6mA levels at a given adenine basis.

Single-molecule real-time (SMRT) sequencing and bioinformatics

For SMRT sequencing, genomic DNA samples from adult animals staged at the 1- and 5-day were prepared as follows. Wild-type animals collected from 40 NGM plates were synchronized by bleaching (hypochlorite treatment). When the animals reached the young adult stage at 20 °C, they were transferred to 400 NGM plates containing 50 μg/ml FUdR (5-Fluoro-2′-deoxyuridine, Sigma-Aldrich, #F0503). Around 100 animals per plate were transferred. After 1 day, 200 plates were harvested for 1-day-old animals, while the remaining 200 plates were harvested for 5-day-old animals. Genomic DNA was then isolated from the samples with GeneJET Genomic DNA Purification Kit (Thermo Scientific™ GeneJET Genomic DNA Purification Kit, #K0721), and then subjected to SMART sequencing. We used command line tools of PACBIO SMRT Link (version 7.0.1) for methylation calling. First PacBio sequences (stored in unaligned BAM files) were mapped to the C. elegans reference genome (Caenorhabditis_elegans.WBcel235.dna.toplevel.fa) by pbalign (version 0.4.1). Then, ipdSummary (version 2.4) was used on the aligned BAM files to detect DNA base-modifications from kinetic signatures. The methylation levels of repeat sequences and cDNA sequences were calculated by in-house python scripts using the gff files, and the annotation files of Dfam release 3.1 (https://www.dfam.org/releases/Dfam_3.1/annotations/ce10/, https://doi.org/10.1093/nar/gkv1272). To determine the relative 6 mA content for a given DNA sequence, we divided the amount of 6 mA signals detected in the given section by the length of that segment to obtain the relative amount of 6 mA in that segment. For the TE-derived sequences, the length of the given TE sections was proportional to the size of the TE section multiplied by the copy number of the given TE. Thus, the analysis was not affected by the high copy numbers of TEs.

Detecting mobilization of a single Tc3 stretch

We developed a relatively simple, PCR-based method to detect the mobilization of a single Tc3 stretch, R10H1.3. Genomic DNA from populations of adult animals at the 1-day and 14-day stages was purified by using GeneJET Genomic DNA Purification Kit (Thermo Scientific™ GeneJET Genomic DNA Purification Kit, #K0721), and adjusted to the same concentration, using a nanodrop method. Mean relative DNA levels were determined by using Tc3 qPCR control primers as reference. Forward and reverse primers, and PCR conditions were as follows. For Tc3 mobilization: 5′-AGA CCA AAA AGA CGG TGC GTA-3′ and 5′-AAC GGT AAT TGC CGG TCC AA-3′, initial denaturation at 95 °C for 30 s, then 32 or 37 cycles of denaturation at 95 °C for 10 s, annealing and extension at 63 °C for 30 s. For Tc3 qPCR control: 5′-GTT TCC TTG TGT GCG GAT G-3′ and 5′-ATT CTC GAC GCT TGG AAG TC-3′, initial denaturation at 95 °C for 30 s, then 30 cycles of denaturation at 95°Cfor 10 s, annealing and extension at 66 °C for 30 s.

Thermotolerance assay

The assay was performed as described earlier⁴⁹. Briefly, worms were synchronized and maintained at 20 °C. Young (non-gravid) adult hermaphrodites were transferred to fresh agar plates seeded with E. coli OP50, then exposed to heat shock by transferring them at 35 °C for 5 h. Animals were then transferred back at 20 °C, and allowed to recover overnight. The survival rate was determined by counting the number of live versus dead individuals. At least 60 animals per strain were counted by light microscopy in 3 independent trials. Statistical significance was determined by independent samples t tests, using the SPSS 17 software. TE transcript levels of heat-shocked animals were determined by qPCR using the following primers: cdc-42 (control): forward 5′-CTT CGA CAA TTA CGC CGT CAC-3′ and reverse 5′-CGA AAT TTC AGG CAC CCA TTT TTC-3′; Tc1: forward 5′- TGC AAA TTC AAC GTT CTC CG-3′ and reverse 5′-TTC TCA AAG CGA CGG ATT TC-3′; Tc3: forward 5′-CGA ACG AAT TGG AGT AAA GTT GTC-3′ and reverse 5′-GTT TCC TTG TGT GCG GAT G-3′.

Fluorescent microscopy

Animals were synchronized, and approximately 60-70 L4 larvae were transferred to a new NGM plate supplemented with 300 mg/ml FUdR (5-fluoro-2′-deoxyuridine, Sigma). Adult animals were then assayed at days 1, 5 and 10 at 20 °C. In the case of assessing hsp-16.2::gfp expression levels, nematodes were heat shocked at 35 °C for 30 min and microscopy was performed after a recovery at 20 °C for 3 h. Transgenic worms were immobilized by adding 0.1 M sodium-aside in M9 buffer. Images for quantitative analysis were taken by a Zeiss AxioImager Zeiss M2 epifluorescence microscope (objective EC Plan-NeoFluar 10 × 0.3 NA). Expression data were obtained from whole animals. Statistical significance was determined by independent samples t tests (SPSS 17 software). The software ImageJ was used for quantitative analysis.

Statistics and reproducibility

Each statistical test implemented in this study is indicated either within the figure legends or the supplementary materials tables. Experiments which display results from representative experiments, including gel images and microscopic images, were conducted with a minimum of three biological replicates. This number was chosen based on accepted guidelines within the scientific community and is designed to balance the need for robust, reproducible data with considerations of experimental efficiency.

Analysis of published data sets

All external data sets used in Supplementary Fig. 14 were downloaded from GEO GSE93724, and published by Senchuk et al.⁵² and Chen et al.⁵³. Expression levels were determined by aligning the raw sequencing data to the Repbase⁷³, a C. elegans-specific TE database, using custom Python scripts.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

All data supporting the findings of this study are provided within the paper and its Supplementary Information. Source data are provided with this paper. Any data are available from the authors upon request. Raw SMRT-sequencing reads of the 1 day old and 5 days old animals are available in the NCBI Sequence Read Archive (SRA) database with the BioProject accession code PRJNA682481. All data, code, and materials used in the analysis are available upon reasonable request for collaborative studies regulated by materials/data transfer agreements (MTA/DTAs) to the corresponding author (vellai.tibor@ttk.elte.hu). Uncropped versions of every picture used in the study can be downloaded from the link below: https://osf.io/c9wxj/?view_only=b67f2ae5eb9a4f9abbe0c5baa563dedb. Source data are provided with this paper.

References

McClintock, B. The origin and behaviour of mutable loci in maize. Proc. Natl Acad. Sci. USA 36, 344–355 (1950).
ADS CAS PubMed Central PubMed Google Scholar
Levin, H. L. & Moran, J. V. Dynamic interactions between transposable elements and their hosts. Nat. Rev. Genet. 12, 615–627 (2011).
CAS PubMed Central PubMed Google Scholar
Lander, E. S. et al. Initial sequencing and analysis of the human genome. Nature 409, 860–921 (2001).
ADS CAS PubMed Google Scholar
Brouha, B. et al. L1s accounts for the bulk retrotransposition in the human population. Proc. Natl Acad. Sci. USA 100, 5280–5285 (2003).
ADS CAS PubMed Central PubMed Google Scholar
Hormozdiari, F. et al. Alu repeat discovery and characterization within human genomes. Genome Res. 21, 840–849 (2011).
CAS PubMed Central PubMed Google Scholar
Gasior, S. L., Wakeman, T. P., Xu, B. & Deininger, P. L. The human LINE-1 retrotransposon creates DNA double-strand breaks. J. Mol. Biol. 357, 1383–1393 (2006).
CAS PubMed Central PubMed Google Scholar
Daskalos, A. et al. Hypomethylation of retrotransposable elements correlates with genomic instability in non-small cell lung cancer. Int. J. Cancer 124, 81–87 (2009).
CAS PubMed Google Scholar
Belgnaoui, S. M., Gosden, R. G., Semmes, O. J. & Haoudi, A. Human LINE-1 retrotransposon induces DNA damage and apoptosis in cancer cells. Cancer Cell Int. 6, 13 (2006).
PubMed Central PubMed Google Scholar
Hedges, D. J. & Deininger, P. L. Inviting instability: transposable elements, double-strand breaks, and the maintenance of genome integrity. Mutat. Res. 616, 46–59 (2007).
CAS PubMed Google Scholar
Vijg, J. & Dong, X. Pathogenic mechanisms of somatic mutation and genome mosaicism in aging. Cell 182, 12–23 (2020).
CAS PubMed Central PubMed Google Scholar
Vagin, V. V. et al. A distinct small RNA pathway silences selfish genetic elements in the germline. Science 313, 320–324 (2006).
ADS CAS PubMed Google Scholar
Ross, R. J., Weiner, M. M. & Lin, H. PIWI proteins and PIWI-interacting RNAs in the soma. Nature 505, 353–359 (2014).
ADS CAS PubMed Central PubMed Google Scholar
Murray, V. Are transposons a cause of aging? Mutat. Res. 237, 59–63 (1990).
CAS PubMed Google Scholar
Sturm, Á., Ivics, Z. & Vellai, T. The mechanism of aging: primary role of transposable elements in genome disintegration. Cell. Mol. Life. Sci. 72, 1839–1847 (2015).
CAS PubMed Google Scholar
Sturm, Á., Perczel, A., Ivics, Z. & Vellai, T. The Piwi-piRNA pathway: road to immortality. Aging Cell 16, 906–911 (2017).
CAS PubMed Central PubMed Google Scholar
Gorbunova, V. et al. The role of retrotransposable elements in ageing and age-associated diseases. Nature 596, 43–53 (2021).
ADS CAS PubMed Central PubMed Google Scholar
Egilmez, N. K. & Shmookler Reis, R. J. Age-dependent somatic excision of transposable element Tc1 in Caenorhabditis elegans. Mutat. Res. 316, 17–24 (1994).
CAS PubMed Google Scholar
Li, W. et al. Activation of transposable elements during aging and neuronal decline in Drosophila. Nat. Neurosci. 16, 529–531 (2013).
CAS PubMed Central PubMed Google Scholar
De Cecco, M. et al. Transposable elements become active and mobile in the genomes of aging mammalian somatic tissues. Aging (Albany NY) 5, 867–883 (2013).
PubMed Google Scholar
Patterson, M. N. et al. Preferential retrotransposition in aging yeast mother cells is correlated with increased genome instability. DNA Repair (Amst.). 34, 18–27 (2015).
CAS PubMed Central PubMed Google Scholar
Chen, H., Zheng, X., Xiao, D. & Zheng, Y. Age-associated derepression of retrotransposons in the Drosophila fat body, its potential cause and consequence. Aging Cell 15, 542–552 (2016).
CAS PubMed Central PubMed Google Scholar
Wood, J. G. et al. Chromatin-modifying genetic interventions suppress age-associated transposable element activation and extend life span in Drosophila. Proc. Natl Acad. Sci. USA 113, 11277–11282 (2017).
ADS Google Scholar
O’Donnell, K. A. & Burns, K. H. Mobilizing diversity: transposable element insertions in genetic variation and disease. Mob. DNA 1, 21 (2010).
PubMed Central PubMed Google Scholar
Van Meter, M. et al. SIRT6 represses LINE1 retrotransposons by ribosylating KAP1 but this repression fails with stress and age. Nat. Commun. 5, 5011 (2014).
ADS PubMed Google Scholar
De Cecco, M. et al. L1 drives IFN in senescent cells and promotes age-associated inflammation. Nature 566, 73–78 (2019).
ADS PubMed Central PubMed Google Scholar
Bessereau, J.-L. Transposons in C. elegans, WormBook (The C. elegans Research Community, WormBook, 2016).
Sturm, Á., Saskői, É., Kovács, T., Weinhardt, N. & Vellai, T. Highly efficient RNAi and Cas9-based auto-cloning systems for C. elegans research. Nucleic Acids Res. 46, e105 (2018).
PubMed Central PubMed Google Scholar
Bouallègue, M., Rouault, J. D., Hua-Van, A., Makni, M. & Capy, P. Molecular evolution of piggyBac superfamily: from selfishness to domestication. Genome Biol. Evol. 9, 323–339 (2017).
PubMed Central PubMed Google Scholar
Plasterk, R. H. Differences between Tc1 elements from the C. elegans strain Bergerac. Nucleic Acids Res. 15, 10050 (1987).
CAS PubMed Central PubMed Google Scholar
Johnson, T. E. & Wood, W. B. Genetic analysis of life-span in Caenorhabditis elegans. Proc. Natl Acad. Sci. USA 79, 6603–6607 (1982).
ADS CAS PubMed Central PubMed Google Scholar
Surzycki, S. A. & Belknap, W. R. Repetitive DNA elements are similarly distributed on Caenorhabditis autosomes. Proc. Natl Acad. Sci. USA 97, 245–249 (2000).
ADS CAS PubMed Central PubMed Google Scholar
Tóth, K. F., Pezic, D., Stuwe, E. & Webster, A. The piRNA pathway guards the germline genome against transposable elements. Adv. Exp. Med. Biol. 886, 51–77 (2016).
PubMed Central PubMed Google Scholar
Ninova, M., Griffiths-Jones, S. & Ronshaugen, M. Abundant expression of somatic transposon-derived piRNAs throughout Tribolium castaneum embryogenesis. Genome Biol. 18, 184 (2017).
PubMed Central PubMed Google Scholar
Cox, D. N. et al. A novel class of evolutionarily conserved genes defined by piwi are essential for stem cell self-renewal. Genes Dev. 12, 3715–3727 (1998).
CAS PubMed Central PubMed Google Scholar
Vastenhouw, N. L. et al. A genome-wide screen identifies 27 genes involved in transposon silencing in C. elegans. Curr. Biol. 13, 1311–1316 (2003).
CAS PubMed Google Scholar
Greer, E. L. et al. DNA Methylation on N6-Adenine in C. elegans. Cell 161, 868–878 (2015).
CAS PubMed Central PubMed Google Scholar
Zhang, G. et al. N6-methyladenine DNA modification in Drosophila. Cell 161, 893–906 (2015).
CAS PubMed Google Scholar
Wang, S. Y. et al. The demethylase NMAD-1 regulates DNA replication and repair in the Caenorhabditis elegans germline. PLoS Genet. 15, e1008252 (2019).
CAS PubMed Central PubMed Google Scholar
Wu, T. P. et al. DNA methylation on N6-adenine in mammalian embryonic stem cells. Nature 532, 329–333 (2016).
ADS CAS PubMed Central PubMed Google Scholar
O’Brown, Z. K. et al. Sources of artifact in measurements of 6mA and 4mC abundance in eukaryotic genomic DNA. BMC Genomics 20, 445 (2019).
PubMed Central PubMed Google Scholar
Schiffers, S. et al. Quantitative LC-MS provides no evidence for m6dA or m4dC in the genome of mouse embryonic stem cells and tissues. Angew. Chem. Int. Ed. Engl. 56, 11268–11271 (2017).
CAS PubMed Google Scholar
Douvlataniotis, K., Bensberg, M., Lentini, A., Gylemo, B. & Nestor, C. E. No evidence for DNA N 6-methyladenine in mammals. Sci. Adv. 6, eaay3335 (2020).
ADS CAS PubMed Central PubMed Google Scholar
Kong, Y. et al. Critical assessment of DNA adenine methylation in eukaryotes using quantitative deconvolution. Science 375, 515–522 (2022).
ADS CAS PubMed Central PubMed Google Scholar
Yao, B. et al. DNA N6-methyladenine is dynamically regulated in the mouse brain following environmental stress. Nat. Commun. 8, 1122 (2017).
ADS PubMed Central PubMed Google Scholar
Lieberman Greer, E., Becker, B., Latza, C., Antebi, A. & Shi, Y. Mutation of C. elegans demethylase spr-5 extends transgenerational longevity. Cell Res. 26, 229–238 (2016).
Google Scholar
Kenyon, C., Chang, J., Gensch, E., Rudner, A. & Tabtiang, R. A. C. elegans mutant that lives twice as long as wild type. Nature 366, 461–464 (1993).
ADS CAS PubMed Google Scholar
Senchuk, M. M. et al. Activation of DAF-16/FOXO by reactive oxygen species contributes to longevity in long-lived mitochondrial mutants in Caenorhabditis elegans. PLoS Genet. 14, e1007268 (2018).
PubMed Central PubMed Google Scholar
Flusberg, B. A. et al. Direct detection of DNA methylation during single-molecule, real-time sequencing. Nat. Methods 7, 461–465 (2010).
CAS PubMed Central PubMed Google Scholar
Zevian, S. C. & Yanowitz, J. L. Methodological considerations for heat shock of the nematode Caenorhabditis elegans. Methods 68, 450–457 (2014).
CAS PubMed Central PubMed Google Scholar
Hsu, A.-L., Murphy, C. T. & Kenyon, C. Regulation of aging and age-related disease by DAF-16 and heat-shock factor. Science 300, 1142–1145 (2003).
ADS CAS PubMed Google Scholar
Papp, D., Csermely, P. & Soti, C. A role for SKN-1/Nrf in pathogen resistance and immunosenescence in Caenorhabditis elegans. PLoS Pathog. 8, e1002673 (2012).
CAS PubMed Central PubMed Google Scholar
Lund, J. et al. Transcriptional profile of aging in C. elegans. Curr. Biol. 12, 1566–1573 (2002).
CAS PubMed Google Scholar
Chen, A. T.-Y. et al. Longevity genes revealed by integrative analysis of isoform-specific daf-16/Foxo mutants of Caenorhabditis elegans. Genetics 201, 613–629 (2015).
CAS PubMed Central PubMed Google Scholar
Siddiqi, S., Terry, M. & Matushansky, I. Hiwi mediated tumorigenesis is associated with DNA hypermethylation. PLoS ONE 7, e33711 (2012).
ADS CAS PubMed Central PubMed Google Scholar
Jardim, S. S., Schuch, A. P., Pereira, C. M. & Loreto, E. L. S. Effects of heat and UV radiation on the mobilization of transposon mariner-Mos1. Cell Stress Chaperones 20, 843–851 (2015).
CAS PubMed Central PubMed Google Scholar
Sun, L. et al. Heat stress-induced transposon activation correlates with 3D chromatin organization rearrangement in Arabidopsis. Nat. Commun. 11, 1886 (2020).
ADS CAS PubMed Central PubMed Google Scholar
Horvath, V., Merenciano, M. & González, J. Revisiting the relationship between transposable elements and the eukaryotic stress response. Trends Genet. 33, 832–841 (2017).
CAS PubMed Google Scholar
de Oliveira, D. S., Rosa, M. T., Vieira, C. & Loreto, E. L. S. Oxidative and radiation stress induces transposable element transcription in Drosophila melanogaster. J. Evol. Biol. 34, 628–638 (2021).
PubMed Google Scholar
El-Brolosy, M. A. et al. Genetic compensation triggered by mutant mRNA degradation. Nature 568, 193–197 (2019).
ADS CAS PubMed Central PubMed Google Scholar
Billes, V. et al. Developmentally regulated autophagy is required for eye formation in Drosophila. Autophagy 14, 1499–1519 (2018).
CAS PubMed Central PubMed Google Scholar
Boulias, K. & Greer, E. L. Means, mechanisms and consequences of adenine methylation in DNA. Nat. Rev. Genet. 23, 411–428 (2022).
CAS PubMed Central PubMed Google Scholar
Valli, A., Harris, A. L. & Kessler, B. M. Hypoxia metabolism in ageing. Aging (Albany NY) 7, 465–466 (2015).
CAS PubMed Google Scholar
Snyder, B., Wu, H.-K., Tillman, B. & Floyd, T. F. Aged mouse hippocampus exhibits signs of chronic hypoxia and an impaired HIF-controlled response to acute hypoxic exposures. Cells 11, 423 (2022).
CAS PubMed Central PubMed Google Scholar
Yoder, J. A., Walsh, C. P. & Bestor, T. H. Cytosine methylation and the ecology of intragenomic parasites. Trends Genet. 13, 335–340 (1997).
CAS PubMed Google Scholar
Sturm, A. & Vellai, T. How does maternal age affect genomic stability in the offspring? Aging Cell 21, e13612 (2022).
CAS PubMed Central PubMed Google Scholar
Deniz, Ö., Frost, J. M. & Branco, M. R. Regulation of transposable elements by DNA modifications. Nat. Rev. Genet. 20, 417–431 (2019).
CAS PubMed Google Scholar
Kenyon, C. The genetics of aging. Nature 464, 504–512 (2010).
ADS CAS PubMed Google Scholar
Vellai, T. How the amino acid leucine activates the key cell-growth regulator mTOR. Nature 596, 192–194 (2021).
ADS CAS PubMed Google Scholar
Hotzi, B. et al. Sex-specific regulation of aging in Caenorhabditis elegans. Aging Cell 17, e12724 (2018).
PubMed Central PubMed Google Scholar
Kutnyánszky, V. et al. Sex-specific regulation of neuronal functions in Caenorhabditis elegans: the sex-determining TRA-1 represses goa-1/Gα(_i/o). Mol. Genet. Genomics 295, 357–371 (2020).
PubMed Google Scholar
Vellai, T. et al. Genetics: Influence of TOR Kinase on Lifespan in C. elegans. Nature 426, 620 (2003).
ADS CAS PubMed Google Scholar
Berezikov, E., Bargmann, C. I. & Plasterk, R. H. Homologous gene targeting in Caenorhabditis elegans by biolistic transformation. Nucleic Acids Res. 32, e40 (2004).
PubMed Central PubMed Google Scholar
Bao, W., Kojima, K. K. & Kohany, O. Repbase Update, a database of repetitive elements in eukaryotic genomes. Mob. DNA 6, 11 (2015).
PubMed Central PubMed Google Scholar

Download references

Acknowledgements

This work was supported by the grants OTKA (Hungarian Scientific Research Fund; K132439), VEKOP (VEKOP-2.3.2-16-2017-00014), HunProtExc (2018-1.2.1-NKP-2018, provided by the National Research, Development and Innovation Fund of Hungary), Szint+ (ELTE Thematic Excellence Programme supported by the Hungarian Ministry for Innovation and Technology) to T.V., and NKFIH (National Research, Development and Innovation Office) PD (131839) to E.A. C.K. was supported by the European Union project RRF-2.3.1-21-2022-00004 within the framework of the Artificial Intelligence National Laboratory, Hungary. This work was completed in the ELTE Institutional Excellence Program (1783-3/2018/FEKUTSRAT) supported by the Hungarian Ministry of Human Capacities. Á.S., A.T., J.B. and T.V. were supported by the ELKH/MTA-ELTE Genetics Research Group (01062). Some C. elegans strains were provided by the Caenorhabditis Genetics Center (CGC) supported by the National Institutes of Health – Office of Research Infrastructure Programs (P40 OD010440). The authors express their gratitude to András Fodor for establishing genetic research on the model organism Caenorhabditis elegans in Hungary.

Funding

Open access funding provided by Eötvös Loránd University.

Author information

Authors and Affiliations

Department of Genetics, Eötvös Loránd University (ELTE), 1117, Budapest, Hungary
Ádám Sturm, Éva Saskői, Bernadette Hotzi, Ferenc Bodnár, Himani Sharma, Tibor Kovács, Eszter Ari, Nóra Weinhardt & Tibor Vellai
Eötvös Loránd Research Network (ELKH)-ELTE Genetics Research Group, 1117, Budapest, Hungary
Ádám Sturm, Anna Tarnóci, János Barna & Tibor Vellai
HCEMM-BRC Metabolic Systems Biology Research Group, 6726, Szeged, Hungary
Eszter Ari
Synthetic and Systems Biology Unit, Institute of Biochemistry, Biological Research Centre, Eötvös Loránd Research Network (ELKH), Temesvári krt. 62, 6726, Szeged, Hungary
Eszter Ari
Institute for Computer Science and Control (SZTAKI), 1111, Budapest, Hungary
Csaba Kerepesi
Brigham and Women’s Hospital & Harvard Medical School, Boston, MA, 02115, USA
Csaba Kerepesi
Laboratory of Structural Chemistry and Biology & Hungarian Academy of Sciences (MTA)-ELTE Protein Modelling Research Group, Institute of Chemistry, Eötvös Loránd University, 1117, Budapest, Hungary
András Perczel
Division of Medical Biotechnology, Paul Ehrlich Institute, 63225, Langen, Germany
Zoltán Ivics
Vellab Biotech Ltd., 6722, Szeged, Hungary
Tibor Vellai

Authors

Ádám Sturm
View author publications
You can also search for this author in PubMed Google Scholar
Éva Saskői
View author publications
You can also search for this author in PubMed Google Scholar
Bernadette Hotzi
View author publications
You can also search for this author in PubMed Google Scholar
Anna Tarnóci
View author publications
You can also search for this author in PubMed Google Scholar
János Barna
View author publications
You can also search for this author in PubMed Google Scholar
Ferenc Bodnár
View author publications
You can also search for this author in PubMed Google Scholar
Himani Sharma
View author publications
You can also search for this author in PubMed Google Scholar
Tibor Kovács
View author publications
You can also search for this author in PubMed Google Scholar
Eszter Ari
View author publications
You can also search for this author in PubMed Google Scholar
Nóra Weinhardt
View author publications
You can also search for this author in PubMed Google Scholar
Csaba Kerepesi
View author publications
You can also search for this author in PubMed Google Scholar
András Perczel
View author publications
You can also search for this author in PubMed Google Scholar
Zoltán Ivics
View author publications
You can also search for this author in PubMed Google Scholar
Tibor Vellai
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

Á.S. designed and performed experiments (lifespan assays, construct generation, measuring transcript and 6 mA levels, bioinformatical analysis) and wrote the manuscript; É.S. performed lifespan and semi-qPCR measurements; B.H. generated transgenic strains and performed lifespan assays; A.T. performed lifespan assays; J.B. performed lifespan assays; F.B. determined 6 mA levels; T.K. performed RT-qPCR analysis and evaluated data; H.S. determined relative 6 mA levels during ageing; E.A. identified E02H1.1 as a potential DNA N6-adenine methyltransferase; N.W. performed bioinformatical and statistical analyses; C.K. performed bioinformatical analysis of SMRT and RNAseq data; A.P. designed experiments and wrote the manuscript; Z.I. designed experiments and wrote the manuscript; T.V. designed and performed experiments, and wrote the manuscript.

Corresponding author

Correspondence to Tibor Vellai.

Ethics declarations

Competing interests

The authors declare the following competing interests: T.V. and A.S. share 6mA detection method-related patent applications (accurate determination of age from DNA) with submission numbers of P2100409, W2200015, P2200206 and W2200017. The remaining authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Reporting Summary

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Sturm, Á., Saskői, É., Hotzi, B. et al. Downregulation of transposable elements extends lifespan in Caenorhabditis elegans. Nat Commun 14, 5278 (2023). https://doi.org/10.1038/s41467-023-40957-9

Download citation

Received: 25 July 2018
Accepted: 17 August 2023
Published: 29 August 2023
DOI: https://doi.org/10.1038/s41467-023-40957-9

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.