RNA polymerase III is important for C. elegans reproduction

RNA polymerase III is essential for growth and development (6, 7, 22). Indeed, when we reduced Pol III mRNA expression by feeding C. elegans bacteria expressing a dsRNA for rpc-1 from the embryonic stage, the worms exhibited growth defects and more than 90% of worms were arrested at L2-L3 larval stage, whereas all animals in control RNAi developed to adults within 3 d (Fig 1B). This confirms the requirement of Pol III for reproduction has not been examined. To determine the effect of Pol III knockdown on progeny production, we raised the worms to early adulthood on the standard laboratory food source Escherichia coli OP50-1, transferring to rpc-1 RNAi at the L4 larval stage (Fig 1C). We found that C. elegans treated with rpc-1 RNAi at this late larval stage developed into adults but displayed a dramatic reduction in their total brood size compared to control (∼40% P = 2.98 × 10^-6, Fig 1D). The reduction in brood size is likely independent of spermatogenesis as rpc-1 RNAi did not diminish the total number of sperm in self-fertilized hermaphrodites (Fig S1A and B). Nor did rpc-1 RNAi treatment lead to any change in embryonic viability, or age-specific fecundity and embryos hatched into larvae within a similar time frame as WT (Figs S1C and S2A). This suggests that Pol III inhibition from late larval development is sufficient to reduce progeny production without affecting spermatogenesis or embryo maturation ex-utero. Taken together, these data show that Pol III is required for normal organismal growth and reproductive processes.

TOR signaling interacts genetically with RNA polymerase III to control C. elegans reproduction

Reproduction and embryonic development in worms is an intense anabolic process requiring high levels of protein translation (23, 24). The TORC1 pathway acts as a master switch for governing cellular growth, whereas Pol III transcribes two of the most important non-coding RNA families (tRNAs and rRNAs) required for protein synthesis (2). To test whether Pol III could act downstream of TORC1 to control reproduction, we examined the epistatic relationship between rpc-1 knockdown and C. elegans TORC1 (CeTORC1) using progeny production as a readout. In C. elegans, the gene encoding RagA (a RAS-related GTP-binding protein) is raga-1/RagA, which acts as a positive regulator of the CeTORC1 kinase (25) (Fig 1A). Animals carrying the raga-1(ok386) mutation have constitutively low CeTORC1 signaling throughout development and adulthood (26) (Fig 1C). raga-1(ok386) mutants develop normally, but adults have a reduced brood size compared with WT (Fig 1E, P = 4.81 × 10^-6). However, treatment of raga-1(ok386) mutants with rpc-1 RNAi resulted in a further reduction in progeny numbers compared with either raga-1/RagA mutation or rpc-1 RNAi alone (Fig 1E, P = 7.91 × 10^-11 and P = 6.67 × 10^-8, respectively).

The raga-1(ok386) strain harbors a mutation predicted to remove the entire gene, and is considered null. However, TORC1 is controlled by multiple inputs, and we were concerned that the raga-1(ok386) mutation was not sufficient to completely reduce CeTORC1 signaling. To address this, we depleted the C. elegans TOR kinase itself using RNAi. C. elegans fed let-363/CeTOR RNAi from the L4 stage had a reduced brood size compared with control (Fig 1F, P = 4.9 × 10^-4). However, combining let-363/CeTOR RNAi with rpc-1 RNAi was non-additive compared with either rpc-1 or let-363/CeTOR RNAi (Fig 1F, P = 0.57 and P = 0.62, respectively). This was not because of an artifact of performing double RNAi as diluting either let-363/CeTOR or rpc-1 RNAi by 50% with EV control achieved the same fecundity-related phenotypes as undiluted RNAi (Fig S2A–C). Thus, this implies that Pol III and TOR act in the same pathway to regulate C. elegans reproduction. To explore this epistatic relationship further, we examined progeny production in raga-1(ok386) mutants treated with let-363/CeTOR RNAi. Depletion of let-363/CeTOR from early adulthood resulted in a further reduction in raga-1(ok386) brood size (Fig 1G, P = 1.31 × 10^-5), indicative that the initial raga-1/RagA mutation did not achieve complete TORC1 reduction. However, this was non-additive with rpc-1 RNAi (Fig 1G, P = 0.80). Together, our data suggest that CeTORC1 and Pol III act in the same pathway to affect organismal reproduction.

TOR and RNA polymerase III interact genetically and temporally to control adult lifespan

TOR inhibition extends lifespan in a range of model organisms from yeast to mammals (27). In C. elegans, depletion of CeTOR signaling either by raga-1/RagA loss-of-function mutation or let-363/CeTOR RNAi extends lifespan (26, 28). Given that Pol III and TORC1 act epistatically in respect to progeny production (Fig 1E–G), we wanted to test the genetic interaction between CeTOR signaling and Pol III for C. elegans adult lifespan. As above, we reduced CeTOR signaling at two different time points: from embryonic stages using raga-1/RagA mutation and from the L4 stage using let-363/CeTOR RNAi, where it consistently initiating Pol III knockdown at the L4 stage (Fig 1C). We found that although raga-1(ok386) mutants are long-lived, the addition of rpc-1 RNAi further extended this lifespan, indicating that CeTOR and Pol III are acting in parallel (Fig 2A, Table 1). To rule out the possibility that raga-1/RagA mutation is incomplete and thus masking an epistatic relationship between TOR and Pol III, we examined the simultaneous knock down of raga-1/RagA with let-363/CeTOR with and without rpc-1 RNAi (Fig 1A and C). In our hands, raga-1 mutation and let-363/CeTOR RNAi were not additive; however, the addition of rpc-1 RNAi did further increase the raga-1/RagA; let-363/CeTOR lifespan (Fig 2B, Table 1). This supports our conclusion that CeTOR signaling engages additional, Pol III independent, mediators to control lifespan if knocked out during development (Fig 2C, Table 1).

In fruit flies, RNA Pol III acts downstream of TOR to regulate lifespan. RNA Pol III is activated when TORC1 signalling is inhibited with the use of rapamycin (18). In addition, the lifespan increases incurred by rapamycin treatment and Pol III knockdown (both from adulthood) are non-additive, indicating that Pol III mediates the effects of TOR for longevity (18). To test whether a similar scenario exists in worms, we knocked down CeTOR signaling only from the L4 stage using let-363/CeTOR RNAi and examined its epistatic interaction with Pol III. Interestingly, although both knockdown of let-363/CeTOR or rpc-1 from the L4 stage extended lifespan, there was no additive effect when the RNAi treatments were given simultaneously. This indicates that Pol III mediates the effects of CeTOR signaling in adulthood to control lifespan in C. elegans (Fig 2C, Table 1). Taken together, our data suggest that the genetic interaction between CeTOR signaling and RNA Pol III is dependent on the timing of CeTOR reduction, and whereas reducing CeTOR signaling during development engages mediators in addition to Pol III to control lifespan, CeTOR reduction specifically in adulthood could act through Pol III to promote longevity.

RNA polymerase III does not interact with global protein synthesis, insulin or germline signalling to control lifespan

Reducing CeTOR signaling during development engages mediators distinct from Pol III to affect lifespan (Fig 1C). These additional mediators may work downstream of CeTOR to drive organismal development. To identify these, we tested the genetic interaction of Pol III with candidate mediators implicated in protein synthesis or developmental processes. The ribosomal S6 Kinase (S6K) is a direct target of the TOR kinase and regulates global translation (29, 30). In C. elegans, rsks-1 encodes for S6K and animals lacking this gene are long-lived (31) (Fig S3A, Table S1). We found that this longevity was additive, with that incurred by rpc-1 RNAi treatment, suggesting that Pol III and S6K do not interact to extend lifespan (Fig S3A, Table S1). In mammals, another TOR target is the translational repressor, eukaryotic initiation factor 4E-binding protein (4eBP) (29, 32). C. elegans does not encode 4eBP, but animals with mutations in the translation initiation factors, ife-2/eIF4E and ppp-1/eIF2Bgamma, exhibit disruptions in protein translation (33). However, although deletion of either ife-2 or ppp-1 increased C. elegans lifespan, this was additive with rpc-1 RNAi in both instances (Fig S3B and C, Table S1). In addition to CeTOR signalling, lifespan is also controlled by other signaling pathways implicated in development and growth (34). Mutation of the insulin receptor extends lifespan across species and has been implicated in protein synthesis and turnover in C. elegans (35, 36). In C. elegans, daf-2 encodes the insulin receptor and daf-2 loss-of-function mutants are long-lived (37) (Fig S4A, Table S2). However, knockdown of Pol III using rpc-1 RNAi from the L4 stage was additive with daf-2 lifespan, suggesting that Pol III-does not interact with insulin signalling for lifespan. Loss of germline signalling also consistently increases lifespan in model organisms (38). The C. elegans mutants glp-1(e2141ts) and glp-4 (bn2) are both sterile and live significantly longer than wild-types (39, 40). However, both of these mutations were additive with rpc-1 RNAi (Fig S4B and C, Table S2) implying that Pol III-inhibition induced longevity is independent of germline signaling. Taken together, these data show that Pol III knockdown does not seem to affect the lifespan incurred by reducing global protein synthesis, although this does not rule out the possibility that Pol III affects protein synthesis by other mechanisms not interact with global translation mechanisms.

RNA polymerase III knockdown acts in specific tissues to extend C. elegans lifespan

In both C. elegans and D. melanogaster the longevity phenotype caused by Pol III knockdown appears to be mediated in a tissue-specific manner (18). A detailed contribution of individual tissues has not been tested in worms, so to characterize the expression pattern of Pol III in C. elegans, we examined the expression of a rpc-1::gfp::3xflag translational reporter in wild type adult animals (41). We found RPC-1::GFP in a wide variety of tissues including nuclei of muscle, intestine, hypodermis, neurons, and germline cells (Fig 3A). Treatment of this reporter strain with rpc-1 RNAi, however, reduced expression of RPC-1::GFP in all of these tissues (Fig S5), indicating that these tissues could contribute to the Pol III-inhibition longevity phenotype.

To explore the requirement of these individual tissues for Pol III-knockdown mediated longevity, we used strains harboring a mutation in the dsRNA processing-pathway gene rde-1. The knockout mutant rde-1(ne300) is strongly resistant to dsRNA induced via the feeding RNAi method, with transgenic rescue of RDE-1 in different tissues allowing tissue-specific knockdown using RNAi (42). We knocked-down rpc-1 using RNAi from early adulthood in the muscle, intestine, hypodermis, neurons and germline, and measured their lifespan. Interestingly, rpc-1 RNAi, when knocked-down specifically in the body-wall muscles, extended the median lifespan by a modest 8%, thus indicating a role for this tissue in mediating Pol III knockdown longevity (Fig 3B, Table 2). However, we found that rpc-1 RNAi in the hypodermis, germline, neurons or intestine could not increase lifespan compared with control (Fig 3C–F, Table 2). Taken together, this suggests that Pol III acts in C. elegans muscle to inhibit lifespan under normal conditions.

Late-life knockdown of RNA polymerase III is sufficient to extend lifespan and improve age-related health

In mice, the inhibition of TOR by rapamycin has been shown to extend lifespan even when fed late in life (43). Given that the TOR-Pol III axis is important for adult lifespan (Fig 2C, Table 1), we asked whether Pol III knockdown in late adulthood could also extend longevity in C. elegans. To test this, we raised worms on E. coli OP50-1 until day 5 of adulthood and then transferred them to either control or rpc-1 RNAi plates. Excitingly, compared with control fed animals, worms on rpc-1 RNAi lived longer, suggesting that Pol III knockdown in late adulthood effectively extends longevity (Fig 4A, Table 3). Indeed, this late-life knockdown of Pol III resulted in a lifespan almost identical to that incurred by L4 RNAi treatment (Fig S6, Table 3, Filer et al, 2017 (18)). Overall, these results show that Pol III knockdown in late adulthood is sufficient to extend C. elegans lifespan, strengthening our earlier finding that the beneficial longevity effects of Pol III are post-reproduction.

Healthspan is an important indicator of youthfulness and virility and, ideally, an increase in lifespan is complemented by improvement in organismal healthspan (44). We investigated whether Pol III knockdown could also enhance the health and fitness of worms across their lives by testing their ability to move in liquid media using a thrashing assay. As expected, thrashing ability decreased with age but compared with controls, worms on rpc-1 RNAi moved better at later time points (day 7 and 11 of adulthood) than controls, suggesting improved fitness and better late-life health (Fig 4B). We then compared the ability of C. elegans with move independently on food over their lifespan by videoing their natural movement on E. coli seeded plates. Similarly to the thrashing assay, C. elegans moved around on plates less as they aged, but older worms treated with rpc-1 RNAi moved better compared with controls (Fig 4C, day 5 adults P = 0.0003, day 8 adults P = 0.002 compared with age-matched controls). Together with the thrashing data, this supports a role for Pol III in mediating life-long health and overall lifespan.

Given that Pol III knockdown in muscle is sufficient to promote longevity, we also tested whether muscle-specific rpc-1 RNAi improved age-related fitness. We found that indeed rpc-1 knockdown in the muscle was sufficient to improve overall body movement on day 5 of adulthood (Fig 4D, P = 0.001 compared with control). This supports our finding that rpc-1 is acting on muscles to control age-related health.

It is well reported that ageing contributes to disorganization of the sarcomeres in the body-wall muscle, a phenomenon called sarcopenia which is also a general hallmark of mammalian ageing (45, 46). It has also been shown that changes in muscle mitochondrial structure are strongly correlated with the decline of both sarcomeric structure and speed of movement (47). To further explore the underlying dynamics that might contribute to the improved muscle function in rpc-1 knockdown animals, we examined the mitochondrial structures using a marker of body-wall muscle mitochondria. Interestingly, knock down of rpc-1 improved the organization of mitochondrial networks in older worms compared with control (Fig 4E P = 0.033). Taken together, these results suggest that knockdown of Pol III in adulthood leads to significant increase in lifespan because of improvement in muscle-specific health parameters.

RNA polymerase III and TOR signalling interact temporally to control development and lifespan

Both TOR signalling and RNA Polymerase III are important regulators of anabolic processes, and control a wide array of cellular functions including reproduction and lifespan (16). Here we show that TOR and Pol III interact in a time-dependent manner, and that knockdown of each component individually, sequentially or in parallel influences the outcome of both reproductive and longevity processes (Fig 5). In C. elegans, we manipulated CeTOR signalling using both genetic mutants (that constitutively reduce CeTOR signalling throughout life) and let-363/CeTOR RNAi (which reduces CeTOR signalling the late L4 stage). We show that the genetic relationship between CeTOR signaling and Pol III is timing dependent. The requirement and collaboration of TOR and Pol III to support developmental processes, and the fact that their reduction in adulthood extends lifespan supports the antagonistic pleiotropy theory of ageing; whereby, certain genes are switched “on” to promote the anabolic programs leading to growth, sexual maturity and reproductive fitness, and “hyperfunction” of these genes in adulthood leads to senescence and ageing (48).

The embryonic lethality of rpc-1 and let-363/CeTOR meant that these genes must be knocked down specifically in adulthood (49, 50). Thus, our epigenetic approach meant that artifacts or reduced RNAi efficacy relating to use of combinatorial RNAi could arise (51). However, we observe the same fecundity and lifespan phenotypes for each gene whether it was knocked down individually or diluted 1:1 with control RNAi (Fig S2A–C and Table 1). In fact, these phenotypes were striking, both supporting our methodology and demonstrating the sensitivity of C. elegans to RNAi knockdown of these crucial genes.

The contrasting effects on fitness and longevity as observed by knockdown of Pol III in early development as opposed to late adulthood are similar to those observed for other signaling pathways including mTOR and insulin. Knockdown of daf-2 from early to mid-adulthood also promotes longevity whereas its knockdown from early larval stages reduces fecundity and reproductive fitness (52). Late-life interventions targeting IGF-1 receptor have also been shown to improve lifespan and healthspan in mice (53). However, whereas the lifespan of daf-2 worms cannot be further extended by let-363/CeTOR inhibition, suggesting a common mechanistic pathway (28), we did not find evidence for a genetic interaction with the C. elegans insulin receptor daf-2 (Fig S4A). In fact, daf-2 mutation and rpc-1 RNAi were additives in our hands, implying independent downstream effectors of Pol III and DAF-2. Our similar results with germline-signaling mutants (Fig S4B and C), which increase longevity by reducing germline activity, lead us to speculate that this could be in part due to the common transcriptional denominator DAF-16, which integrates signals both from the IGF-1/DAF-2 axis and germline-loss driven longevity (38, 54, 55), whereas Pol III knockdown bypasses DAF-16 requirements entirely.

RNA polymerase III inhibition as a modulator of late-life health and lifespan

Studies in a wide variety of model organisms show that mTOR inhibition increases longevity and improves organismal health, spurring interest in mTOR inhibitors for slowing down human ageing (56). However, this can occur at the cost of development, e.g., treating female flies with rapamycin increased lifespan, but significantly reduced brood size (57, 58). In addition, clinical studies have shown that chronic exposure to mTOR inhibitors can result in serious side effects including immunosuppression and metabolic dysfunction, making them unattractive anti-ageing therapies (27). Consequently, transient knockdown of mTOR genetically, or by inhibitors, is a fast emerging paradigm for targeting mTOR complexes. Studies targeting mTORC1 by rapamycin even late in life have shown to extend lifespan in mice models, without major side effects (43, 59). Here we show that whereas Pol III knockdown during early larval stages leads to developmental abnormalities and larval arrest (Fig 1B), Pol III knockdown post-reproduction on the fifth day of adulthood, extends lifespan, suggesting that, similar to TOR, even late perturbations in Pol III activity are sufficient to promote longevity (Fig 4A). Importantly, we also show that Pol III inhibition is sufficient to improve age-related decline in movement (Fig 4C and D), which has consistently been correlated with fitness and improved lifespan in worms (60, 61, 62). Overall, our results make RNA Pol III a promising target molecule for ameliorating post-reproductive age-related decline.

RNA polymerase III acts in muscle to drive ageing

It is known that aging progresses heterogeneously and various tissue-specific responses have been identified to perturb organismal aging. Earlier studies have shown that knockdown of Pol III in the fly gut improves gut barrier function and is sufficient to ameliorate age-related impairments in gut health which may ultimately promote longevity (18). In C. elegans, multiple studies have shown tissue-specific requirements for longevity genes (55, 63) and Pol III RNAi mediated longevity can also be mediated via the intestine (18). The broad expression of rpc-1::GFP (Figs 3A and S5), led us to explore its longevity and health functions in other tissues and found that muscle-specific knockdown of rpc-1 was sufficient to extend lifespan (Fig 3B).

Sarcopenia, which leads to debilitating conditions among the elderly, is characterised by low skeletal muscle mass or strength and is believed to be the most frequent cause of disability and a major risk factor of health-related conditions (64, 65). In C. elegans, even reducing translation in body-wall muscle during development shortens lifespan and increases reproduction (66). Conversely, multiple studies, in both worms and mammals, have shown that improving body-wall muscle function can improve overall health and longevity (67, 68, 69), with muscle mitochondrial organisation being a critical contributing factor. For example, loss of RAGA-1/RagA in the nervous system increases lifespan via maintaining mitochondrial networks in muscle (70). Here we show that reducing Pol III/rpc-1 preserves both mitochondrial organisation (Fig 4E) and suppresses the age-related decline observed in older worms (Figs 4D and S7). This strongly supports a role for Pol III in maintaining muscle function with age, with the potential to be used as a novel target molecule for interventions aimed at reducing sarcopenia in human ageing.

Previously, we showed that rpc-1 RNAi in the intestine-specific RNAi strain VP303 significantly extends lifespan as reported in Filer et al, 2017 (18) (Fig S8). This observation is in contrast to our published data wherein rpc-1 RNAi extends the lifespan of a different intestine-specific RNAi strain rde-1(ne219). However, this difference is likely attributed to the allele-specific differences between rde-1(ne219) and rde-1(ne300); ne219 retains significant RNAi processing capacity and is not completely null; whereas, ne300 deletion allele is completely resistant to RNAi (Watts et al (42)). Thus, the rde-1(ne300) strain we initially tested is likely to be “leaky” and RNAi knockdown might not be limited to the gut of the worms.

Lowering translation or mTORC1 activity in adulthood consistently increases lifespan (20), reduces brood size and improves motility (31, 71). Surprisingly though, suppressing CeTORC1 specifically in body-wall muscle increased reproduction and slowed motility, indicating a role for the muscle in CeTOR-mediated activity (72). We could not test if muscle-specific reduction in Pol III during development altered lifespan owing to the strong larval arrest phenotype of rpc-1 RNAi but suspect that such an intervention may perturb global protein translation and remodel inter-tissue communication via one or many of the small-RNAs transcribed by Pol III.

Exploring additional mediators of the Pol III response

It has been well documented that inhibition of translation delays ageing and evidence suggests that reducing global translation can maintain protein homeostasis, dysregulation of which leads to aging and senescence (71, 73, 74). It is likely that Pol III reduction leads to a decrease in two major RNA types, tRNAs and 5s rRNA, subsequently down-regulating global protein translation. Hence, we were surprised to find that rpc-1 RNAi induced longevity was additive with C. elegans mutants of other key translation mediators (Fig S3). This implies a more nuanced role of Pol III in maintaining a healthy proteome, possibly via improving protein folding or reducing insoluble protein loads, which would be important for lifespan and age-related disease. In the future, addressing the impact of Pol III inhibition on protein levels and quality should prove interesting.

C. elegans maintenance and strains

C. elegans were routinely grown and maintained on Nematode Growth Media (NGM) seeded with E. coli OP50-1. The wild-type strain was Bristol N2. The following strains were used:VC222 raga-1(ok386),RB1206 rsks-1(ok1255), RB579 ife-2 (ok306), ppp-1(syb7781), DR1567 daf-2, CB4037 glp-1(e2141), SJ4103 myo-3p::mitGFP, SS104 glp-4(bn2), IG1839 frSi17 II; frIs7 IV; rde-1(ne300) V (intestine-specific RNAi), DCL569 mkcSi13 II; rde-1(mkc36) V (germline-specific RNAi). WM118 rde-1(ne300) V; neIs9 X (muscle-specific RNAi). TU3401 sid-1(pk3321) V; uIs69 V (neuronal-specific RNAi), NR222 rde-1(ne219) V; kzIs9 (hypodermis-specific RNAi), VP303 rde-1(ne219) V; kbIs7 (intestine-specific RNAi). NB: The raga-1(ok386) allele harbors a 1,242 bp deletion at the raga-1/RagA locus that removes almost the entire coding region of the gene.

Lifespan assay

Longevity assays were performed as previously described (Filer et al, 2017 (18)). Briefly, synchronous L1 animals were placed on NGM plates seeded with E. coli OP50-1 until they reached L4. From L1 to L4 larval stage, animals were maintained at 20°C. At L4 stage, worms were shifted to plates seeded with E. coli HT1115 carrying appropriate RNAi clones. These RNAi clones were obtained from the Ahringer RNAi library. The plates were supplemented with 50 μM FuDR solution and lifespans were carried out at 25°C. For combinatorial RNAi knockdown, the two respective overnight RNAi cultures were mixed 1:1 before seeding the plates. In these experiments, controls were also diluted 1:1 with HT115 expressing empty pL4440. Live, dead and censored worms were counted every 2–3 d in the worm populations by scoring movement with gentle prodding when necessary. Data were analyzed and statistics performed using the online freeware OASIS (http://sbi.postech.ac.kr/oasis/surv/).

Brood size measurements

Animals were grown on E. coli OP50-1 plates at 20°C till L4 larval stage. At L4, ∼15 animals were placed on individual NGM plates seeded with the appropriate RNAi. For combinatorial RNAi, see protocol in lifespan section. Progeny production was assessed at 25°C and each parent animal moved to a fresh plate daily until egg-laying ceased. The total number of eggs and larvae were counted after 24 h. Plates in which the animals went missing or showed intestinal bursting were not included in the final analysis. NB: At 25°C (the temperature that we did these experiments) very few animals suffered internal hatching and those that did were taken out of the analysis.

Thrashing assay

Animals were raised to L4 for lifespan assays. At L4, the animals were moved to control RNAi or rpc-1 RNAi seeded plates. On days 4, 7, and 11 of adulthood, worms were put in the isotonic M9 buffer and allowed to acclimatize for 1 min before measuring the number of side-to-side mid-body movements per minute. Fifteen animals were analyzed per condition.

Microscopy

C. elegans carrying the rpc-1::gfp::3Xflag reporter were grown in fed conditions to the L4 stage. myo-3::mitGFP imaging, animals were imaged on day fifth of adulthood with rpc-1 RNAi treatment initiated on day 1. When imaging, animals were immobilized on 2% agarose pads using 0.06% Levamisole and imaged immediately using a Zeiss Confocal Ultra at 40X zoom.

Speed measurement

C. elegans were raised and maintained as for the lifespan assays. At each timepoints, 20–30 worms were moved to a thinly seeded 60 mm plate and were allowed to acclimatize for 10 min. Their individual movements were then recorded using Wormtracker (MBF Bioscience) for 2 min with 60 fpm settings. Average speeds for individual animals were calculated using MS Excel.

Spermatocyte measurement

Hermaphrodite C. elegans were raised and maintained for brood size assays. At the early L4 stage, 30–40 worms were moved to RNAi plates. After 24 h (day 1 of adulthood), worms were washed with 0.01% Tween PBS to remove excess bacteria, fixed in methanol for 5 min. Worms were washed once again with 0.1% Tween PBS. DAPI dye was added immediately and kept for 10 min. Worms were washed once again with 0.1% Tween PBS and mounted on 2.5% agarose pads for visualization. DIC and DAPI images were acquired using a Olympus IX 81 microscope. Individually, distinct spermatocytes for each worm were counted.