DNA damage signals from somatic uterine tissue arrest oogenesis through activated DAF-16 Free

The propagation of a species depends on a healthy and productive germ line. The stability of the genome is constantly under threat from extrinsic as well as cell-intrinsic genotoxic agents. Thus, all organisms invest heavily in protecting the germ line against DNA damage. Generally, in response to DNA damage, an organism deploys an array of countermeasures. Depending on the type of DNA damage, organisms employ lesion-specific DNA repair pathways that can restore damage inflicted by ultraviolet rays (UV), ionizing radiation (IR) or reactive oxygen (ROS) and nitrogen species (RNS). Apart from these highly specialized DNA repair mechanisms, organisms also depend on DNA damage response (DDR) signaling to activate damage-responsive checkpoints, leading to cell cycle arrest to repair the damage or apoptosis, when damage is beyond repair. Perturbation of the DDR can lead to unrepaired DNA damage, and genomic instability, and is the basis of many debilitating human diseases such as cancer and neurodegeneration, as well as aging (Jackson and Bartek, 2009). Unrepaired DNA lesions in the germ line can lead to infertility, reduced progeny fitness and anomalies at birth. In response to environmental cues, germ cell proliferation, as well as the critical decision of reproductive commitment, is largely influenced by somatic tissues via soma-to-germ line communication (Hubbard, 2011; Lopez et al., 2013). However, it is less well known whether or how an organism perceives intrinsic DNA damage signals in somatic tissues and regulates germ line development non-cell-autonomously to preserve progeny genome integrity.

Research in C. elegans has elucidated the role of the conserved FOXO transcription factor (TF) DAF-16 (also known as FOXO) in somatic and germ line quality assurance. The C. elegans neuroendocrine insulin/IGF-1 signaling (IIS) receptor DAF-2 signals through a conserved PI3K pathway to phosphorylate and activate downstream serine threonine kinases such as AKT-1, AKT-2 and SGK-1. These kinases, in turn, phosphorylate and sequester FOXO/DAF-16 in the cytoplasm, thereby inactivating it, ensuring normal development and reproduction. Under conditions of low signaling, as in the temperature-sensitive alleles of daf-2, FOXO/DAF-16 is dephosphorylated and moves into the nucleus to transactivate or repress gene expression, which leads to extreme longevity and resistance to multiple stresses (Lee et al., 2001; Mukhopadhyay et al., 2006; Ogg et al., 1997). Strong mutations in the IIS receptor also arrest development at dauer diapause, in a DAF-16-dependent manner (Gottlieb and Ruvkun, 1994; Larsen et al., 1995). Other than longevity and dauer diapause, the IIS-DAF-16 axis mediates arrest at the L1 larval stage when food is depleted (Baugh and Sternberg, 2006). The IIS receptor mutant animals maintain their germ line stem cell pool even at an advanced age, and therefore have delayed reproductive aging (Qin and Hubbard, 2015). The mutant worms produce better quality oocytes (Templeman et al., 2018) with low chromosomal abnormalities when compared with wild type, but the mechanism is less well understood (Luo et al., 2010). Although FOXO/DAF-16 is an important regulator of many important biological processes, it is not known whether activated FOXO/DAF-16 can modulate germ line development in response to signals of somatic DNA damage.

Here, we show that in C. elegans, a uterine tissue-specific perturbation of DDR in the IIS pathway mutant (where DAF-16 is constitutively activated) arrests germ cells at the pachytene stage of meiosis I, leading to sterility. For DDR perturbation and inducing endogenous DNA damage, we performed RNAi knockdown (KD) of the RNA polymerase II CTD kinase cdk-12, which is required for the transcription of DDR genes in worms and mammals (Bartkowiak et al., 2010; Bowman et al., 2013; Dubbury et al., 2018). In the absence of activated DAF-16, germ cells of the IIS pathway mutant fail to arrest at the pachytene stage on DDR perturbation, leading to the production of poor-quality oocytes. We show that this arrest is partly achieved by the downregulation of the ERK-MPK-1 signaling that controls pachytene exit. Thus, our study elucidates a new role of the IIS pathway and activated FOXO/DAF-16 in ensuring germ line quality in response to somatic (uterine tissue) perturbation of DDR and the associated chance of genome instability in the progeny.

The C. elegans IIS pathway regulates a plethora of phenotypes, including longevity, metabolism and development, possibly through complex interactions with as yet unknown kinases or signaling pathways. To identify previously unreported genetic interactors of the IIS pathway, a targeted RNAi screen was performed using the temperature-sensitive daf-2(e1370) (hereafter referred to as daf-2) strain. At 25°C, this allele forms 100% dauer, whereas at 22.5°C, the percentage of dauer formation is ∼40%. Focusing the RNAi screen on annotated kinases, we individually knocked down 418 C. elegans kinases in daf-2 mutants grown at 22.5°C and scored for enhancement of dauer formation. The pdk-1 RNAi served as a positive control, which increased dauer formation, whereas daf-16 RNAi suppressed dauer formation (Fig. S1A). The cdk-12 (B0285.1) RNAi significantly increased dauer formation in daf-2(e1370) worms, thereby hinting at an underlying genetic interaction. Strong cdk-12 KD using RNAi leads to developmental defects and the available cdk-12 mutant arrests in the L1 larval stage (Cassart et al., 2020). Therefore, to avoid developmental defects, we diluted the cdk-12 RNAi with control RNAi-expressing bacteria for further studies (Kamath et al., 2001).

Mammalian CDK12 is an RNA Polymerase II-CTD kinase that is known to specifically regulate the transcription of DDR genes and maintain genome stability (Dubbury et al., 2018; Joshi et al., 2014). We first checked whether the cdk-12 function is conserved in C. elegans and found that DDR gene expression was significantly downregulated upon cdk-12 depletion in wild-type worms, as assessed by quantitative real-time PCR (RT-PCR) (Fig. 1A). The cdk-12 KD did not lead to global downregulation of gene expression, as determined by RNA-seq analysis and RT-PCR of non-DNA repair genes (Fig. S1B,C). To functionally validate the role of cdk-12 in genome stability, we explored the efficiency of the DDR machinery by using a chromosome fragmentation assay. In unirradiated worms, six highly condensed bivalent bodies can be seen in the oocytes; however, unrepaired DNA strand breaks in worms irradiated with IR lead to chromosome fragmentation/fusions (Clejan et al., 2006). We treated the worms with DNA damaging IR to induce double-strand break at L4 and allowed 48 h for repair of the damage. After 48 h, the oocyte nuclei were DAPI stained and imaged. We observed increased chromosome fragmentation and fusions in IR-treated wild-type worms upon KD of cdk-12 (Fig. 1B,C), which reaffirms weakened DNA damage repair functionality. Next, we exposed the L4 stage wild-type worms to different concentrations of the DNA damaging agent camptothecin (CPT) (Fig. 1D) or to varying doses of IR (Fig. 1E) and found that, in the absence of cdk-12, fewer eggs hatched, highlighting their increased sensitivity to DNA damage and reiterating a compromised DNA damage repair initiative. Assessment of RAD-51 and/or RPA-1 foci formation as a readout for endogenous DNA damage or unrepaired DNA strand breaks is widely used (Kim and Colaiácovo, 2014; Parusel et al., 2006). However, in this case, the transcript levels of the DDR genes, and consequently the proteins, were downregulated upon cdk-12 RNAi, so the GFP fusion transgenic worms could not be used. The transgenic strain hus-1::gfp was used to assess the DNA damage repair and foci formation because HUS-1 is a part of the 9-1-1 complex and is a DNA damage sensor (Hofmann et al., 2002). Upon induction of double-stranded breaks by IR (100 Gy), we observed increased HUS-1 foci formation in control worms. However, cdk-12 RNAi-fed worms had significantly reduced HUS-1 foci post-IR (Fig. S1D), implying a faulty damage checkpoint. We also observed higher apoptotic bodies per gonadal arm upon loss of cdk-12 (Fig. S1E), which may be a strategy to eliminate the defective germ cells, where endogenous damage is unrepaired (Andux and Ellis, 2008; Gartner et al., 2000; Sang et al., 2022). The increased apoptosis observed upon cdk-12 depletion, despite lower HUS-1 foci formation, suggests the induction of HUS-1 checkpoint-independent apoptosis. Previously, it has been shown that mutations in some DDR genes, e.g. atm-1 and dog-1 (Kniazeva and Ruvkun, 2019), as well as DNA damage due to free oxygen radicals, lead to elongated intestinal cells with chromosomal bridges (karyokinesis defect due to DNA damage). Interestingly, upon analyzing the intestinal nuclei of day 1 wild-type adults, we observed almost a threefold increase in intestinal chromosomal bridges upon cdk-12 KD (Fig. 1F). Furthermore, we exposed the wild-type L1 larvae to various doses of IR (0, 300 and 400 Gy) (initially, we exposed L1 worms to a range of doses from 0-400 Gy and selected doses where wild type on control RNAi showed 40-60% growth arrest) and monitored their larval development. Unirradiated worms developed beyond the L4 stage. However, upon irradiation, fewer cdk-12 RNAi worms reached the L4 stage or later by 96 h after the L1 stage, in comparison with control RNAi-fed worms, indicative of a compromised somatic DNA repair mechanism (Fig. S1F). DNA damage in worm germ line has been shown to induce stress response pathways that, in turn, confer systemic resistance and enhance somatic stress endurance (Ermolaeva et al., 2013). In agreement with the fact that the absence of cdk-12 may lead to endogenous DNA damage, we observed increased heat stress resistance (Fig. S1G) and hsp-4::gfp (endoplasmic reticulum chaperon BiP ortholog) expression (Fig. S1H). Overall, these results show a pivotal role for CDK-12 in the repair of damaged DNA, maintaining genomic integrity in both the wild-type C. elegans germ line and somatic tissues. Together, this evidence supports the functionally conserved role of CDK-12 in DDR and in maintaining genome stability from worms to mammals.

Furthermore, we evaluated the efficacy of the DDR machinery in the daf-2 mutant upon cdk-12 depletion. As in wild type, we also found that the expression of the DDR genes is downregulated by cdk-12 KD in the daf-2 mutant (Fig. S1I). We exposed the daf-2 L1 larvae to various doses of IR (0, 300 and 400 Gy) and monitored their larval development. We found that after cdk-12 KD, significantly fewer irradiated worms reached the L4 stage, indicating a compromised somatic DNA repair mechanism (Fig. 1G). We show that DNA damage indeed induces intestinal chromosomal bridges (observed after DAPI staining) in daf-2 worms treated with IR (160 Gy) at L1 and imaged at the late L4 stage (Fig. S1J). Strikingly, we found a greater than fivefold increased presence of intestinal chromatin bridges upon KD of cdk-12 in late L4 stage daf-2 worms (Fig. 1H,I). Together, these results confirm the occurrence of DNA damage upon KD of cdk-12 in daf-2 worms. Because daf-2 worms become sterile after cdk-12 RNAi, assays such as chromosome fragmentation in oocytes, egg hatching after IR, etc., could not be performed in this strain. Thus, cdk-12 is required for the DDR gene expression in wild type and during low IIS (as in the daf-2 mutant), thereby ensuring genome stability.

The daf-2 mutant is resistant to multiple external stressors, including UV-induced DNA damage, mostly in a DAF-16-dependent manner (Barsyte et al., 2001; Garsin et al., 2003; Honda and Honda, 1999; Lithgow et al., 1995; Mueller et al., 2014; Murakami and Johnson, 1996). We therefore asked how these mutants respond to DDR perturbations and the resulting DNA damage signals. Interestingly, we found that the daf-2 worms became sterile when they were grown on cdk-12 RNAi from L1 onwards at 20°C (Fig. 2A,B); these worms produced no oocytes (Fig. 2C). The sterility is DAF-16 dependent, as fertility was restored (determined by the presence of oocytes/eggs in the uterus) in daf-16;daf-2 worms, signifying that DAF-16 regulates the germ line arrest in daf-2 worms (Fig. 2A-C). Importantly, this was not due to differential RNAi efficiency in the strains (Fig. S2A,B). We also analyzed other daf-2 alleles such as e1368 and e1371, which led to sterility upon cdk-12 KD at 22.5°C, possibly owing to the stronger reduction of IIS at a higher temperature for these alleles (Fig. S2C). Additionally, KD of the canonical binding partner of CDK-12, cyclin-K (F43D2.1) (Blazek et al., 2011), produced no such sterility in daf-2 worms (Fig. S2D,E), suggesting that it may be a cyclin-independent function of CDK-12 or that a different cyclin may partner with it for this phenotype.

The C. elegans hermaphrodite gonad has two U-shaped tubular arms carrying germ line stem cell (GSC) pool near the distal end, which divide mitotically and then enter meiotic prophase as they move away from the distal tip. Germ cells first differentiate to form sperm during the larval 4 (L4) stage, and from late L4 onwards it switches from spermatogenesis to oogenesis, producing oocytes till reproductive senescence (Austin and Kimble, 1987; Kraemer et al., 1999; Schedl, 1997). Only a fraction of germ cells pass the quality control checkpoint and differentiate to form oocytes whereas the rest undergo programmed cell death at the pachytene region. The proximal gonad contains a stack of oocytes, followed by sperm residing in the spermatheca. Both arms have a common uterus, where fertilized eggs are stored until they are laid (Kimble and Crittenden, 2007) (Fig. 2D; only one gonad arm is shown).

We dissected the gonads of day 1 adult animals and stained them with DAPI to determine the cause of daf-16-dependent sterility upon cdk-12 KD. The DAPI images revealed that germ cells in daf-2 worms were arrested at the pachytene stage of meiosis I, but not in daf-16;daf-2 worms, upon cdk-12 KD (Fig. 2E). Because of its distinct ‘bowl of spaghetti’ morphology, pachytene nuclei stained with DAPI can be easily identified in the C. elegans germ line (Fig. S2F). The germ cells in daf-2 worms grown after cdk-12 RNAi fail to exit from the pachytene stage to enter diplotene, whereas daf-16;daf-2 germ cells bypass the arrest and produce oocytes (Fig. 2E). The pachytene stage germ cell count was drastically reduced in daf-2 worms after cdk-12 KD (Fig. 2F,G), but not so in daf-16;daf-2 worms (Fig. 2H,I). In both cases, the number of mitotic and transition zone nuclei remained relatively unchanged (Fig. 2F-I). The lower number of pachytene stage nuclei in daf-2 worms may not be simply due to increased apoptosis, as daf-16;daf-2 worms also showed similar elevated numbers (Fig. S2G). We also discovered that components of the canonical IIS signaling pathway are involved, as cdk-12 KD in age-1(hx546) (mammalian PI3K ortholog) (Morris et al., 1996; Paradis and Ruvkun, 1998) and pdk-1(sa680) (mammalian PDK ortholog) (Paradis et al., 1999) also arrested the germ line at the pachytene stage of meiosis (Fig. S2H,I), similar to daf-2 worms after cdk-12 RNAi.

Based on the observations that CDK-12 is involved in DDR in C. elegans, we hypothesize that the depletion of cdk-12 leads to genomic instability that is dependent on active DAF-16 in the daf-2 mutant, leading to the germ line arrest at the pachytene stage of meiosis, preventing defective oocyte formation. Next, we asked whether KD of other genes involved in DDR will also lead to similar arrests. We used RNAi for genes involved in DNA repair or the depletion of which could lead to genomic instability (Table S1). Interestingly, we observed DAF-16-dependent sterility in daf-2 worms only upon KD of him-10, a component of the kinetochore-associated NDC80 complex, depletion of which leads to chromosome non-disjunction (Howe et al., 2001) (Fig. S2J-K). However, at this point, it is unclear why only him-10 KD shows similar DAF-16-dependent sterility in the daf-2 worms.

Next, we sought to determine why the germ line arrests at the pachytene stage of meiosis in the daf-2 mutant grown on cdk-12 RNAi. For this, we investigated how cdk-12 KD affects oogenesis under wild-type genetic background at 20°C. Unlike circumstances where DAF-16 is activated (as in daf-2 worms, where we observed arrested oogenesis), KD of cdk-12 does not lead to sterility in wild-type worms (Fig. 3A), in which DAF-16 is predominantly cytoplasmic and transcriptionally inactive. However, the oocyte number was significantly reduced upon cdk-12 depletion in the wild type (Fig. 3B,C). On the other hand, although small but significant changes were observed in mitotic and transition zones, no significant changes were observed in the number of the pachytene germ cells in wild type (Fig. 3D) or in daf-16 (Fig. S3A). Importantly, upon depletion of cdk-12, the oocytes became mostly disorganized, mis-shapen, often lost contact with the neighboring oocytes, and developed cavities; the overall oocyte quality was severely compromised morphologically (Fig. 3E,F, Fig. S3B,C), similar to the condition found in germ lines of old worms (Luo et al., 2010). In line with these observations, we noticed increased embryonic lethality at both day 1 and day 3 of adulthood (Fig. 3G), reflective of poor oocyte quality upon cdk-12 depletion in the wild type. Notably in humans, loss of oocyte quality is associated with embryonic lethality, anomalies at birth and infertility (Cimadomo et al., 2018; Xu et al., 2015). These results emphasize the role of cdk-12 in the maintenance of the oocyte and overall germ cell quality. Additionally, in support of the above inference, the brood size was reduced (Fig. S3D), independent of sperm (Fig. S3E,F), and worms reached reproductive senescence much earlier after cdk-12 KD (Fig. S3G), both in selfed and mated worms. In addition, endomitotic oocytes, which often develop due to defective fertilization (Iwasaki et al., 1996) or poor oocyte quality and are observed in the uterus of wild-type worms after the self-sperm is depleted, were more frequent in the uterus of wild-type worms upon depletion of cdk-12 (Fig. 3H,I). Overall, these observations point towards accelerated reproductive aging due to poor oocyte or germ cell quality upon lowering of cdk-12 levels.

We have shown above that, unlike cdk-12 depletion in wild type, oogenesis of daf-2 worms arrests at the pachytene stage of meiosis; however, oogenesis is restored in daf-16;daf-2 worms. To investigate the possible reason for the arrest in daf-2, we determined the quality of the oocytes of daf-16;daf-2. Upon KD of cdk-12, daf-16;daf-2 and daf-16 worms had severe morphological defects in oocytes (Fig. 3J, Fig. S3H), reduced brood size (Fig. S3I) and increased incidences of endomitotic oocytes/uterine tumors (Fig. 3K,L), very similar to observations in wild-type worms.

Based on the above, we conclude that cdk-12 plays an important role in maintaining oocyte quality; knocking it down reduces oocyte quality and diminishes progeny health, possibly due to the accumulation of unrepaired DNA damage. Activated DAF-16, as observed in the daf-2 worms with lowered IIS, enforces a germ line quality assurance program that prevents the production of inferior quality gametes/oocytes (after cdk-12 KD), thus preserving reproductive fidelity.

To dissect the underlying mechanisms by which activated DAF-16 could halt oogenesis upon cdk-12 KD, we examined the status of the LET-60 (RAS)-MEK-2 (ERK kinase)-MPK-1 (ERK1/2) pathway that has several important roles in germ line development and maturation (Lee et al., 2007) (Fig. 4A). This pathway was the primary choice of our investigation for two reasons. First, active MAPK signaling (pMPK-1) is required for the germ cells to exit the pachytene stage and initiate oogenesis (Lee et al., 2007), a process that is stalled in daf-2 worms after cdk-12 KD. Second, the IIS pathway has been shown to couple nutrient sensing to meiotic progression via the RAS-ERK-MPK-1 pathway, such that oocyte development takes place to enable reproduction only under conditions that are favorable for offspring survival (Lopez et al., 2013). The RAS-ERK pathway works downstream of the IIS receptor daf-2. In response to nutrient availability, IIS activates MPK-1 (ERK) to promote meiotic progression. Thus, in the absence of nutrients or in low food conditions, MPK-1 inhibition results in stalling of meiosis. Upon staining with pMPK-1 antibody, we found the level of ERK activation in daf-2 worms to be significantly lower than in wild type (Fig. 4B,C). The pMPK-1 levels were rescued to wild-type levels in daf-16;daf-2 worms, showing that DAF-16 may negatively regulate pMPK-1 levels (Fig. 4B,C). This potentially explains why daf-2 worms have reduced brood size and oocyte numbers compared with daf-16;daf-2 worms (Fig. S3I, Fig. 2C). Although the level of pMPK-1 was significantly reduced upon KD of cdk-12 in wild type, the reduction was much more dramatic in daf-2 worms, possibly below a presumptive threshold level (Fig. 4B,C). This threshold is based on the earlier observations that low germ line pMPK-1 immunostaining is correlated to the inability of germ cells to exit pachytene. We observed that the basal level of pMPK-1 signal is low in daf-2 worms but increases in daf-2;daf-16 and daf-2;let-60(gf) worms (Fig. 4B,C). Thus, we hypothesize that a certain threshold of pMPK-1 is required for pachytene exit, indicated by the dashed red line in Fig. 4C, above which the worms progress to complete oogenesis. This thresholding may explain the complete arrest of the germ line at the pachytene stage upon cdk-12 KD in daf-2 worms (Fig. 2E), whereas only a reduction in oocyte number (also quality) (Fig. 3B,C,E-I) and brood size (Fig. S3D) was observed in wild type. Importantly, in the daf-16;daf-2 worms upon cdk-12 KD, the pMPK-1 levels were partially restored (Fig. 4B,C), in line with the release of pachytene arrest (Fig. 2E).

Next, we asked whether activating the ERK signaling could bypass the pachytene arrest in daf-2 worms after cdk-12 KD. For this, we used an activated ras allele with constitutively high pMPK-1 phosphorylation (Church et al., 1995) that also shows increased oocyte numbers compared with daf-2 worms (Fig. S4A). Remarkably, we observed that in the daf-2;let-60(gf) worms, the pMPK-1 levels were upregulated (Fig. 4B,C), and pachytene arrest was reversed, albeit partially (Fig. 4D-F, Fig. S4B). Overall, we conclude that the ERK and the IIS/PI3 kinase pathways coordinately regulate meiotic arrest on sensing DDR perturbations (by cdk-12 depletion) in daf-2 worms. Bypassing the arrest by mutating daf-16 or constitutively activating MPK-1 leads to poor-quality oocytes and loss of reproductive fidelity.

We have shown that reducing cdk-12 by RNAi leads to impaired hallmarks of DNA damage, in both soma and germ lines. We next investigated whether cdk-12 KD-mediated DNA damage in germ line would cell-autonomously induce germ line arrest in daf-2 worms or whether somatic perturbation can also non-cell-autonomously regulate germ line arrest. For this, we first used a germ line-specific RNAi system (Zou et al., 2019) to test whether tissue-restricted cdk-12 depletion was sufficient for pachytene arrest in daf-2 worms. We validated the specificity of the germ line RNAi strain by KD of the germline-specific gene glp-1 (required for germ line development) (Pepper et al., 2003), which led to sterility (Fig. S5A), showing a functional germ line RNAi machinery. Surprisingly, we found that KD of cdk-12 in only the daf-2 germ line does not lead to sterility (Fig. 5A,B), indicating that a soma-specific DDR malfunction may cause the germ line arrest non-autonomously. Similar results were obtained after KD of him-10 (Fig. S5B). Together, it appears that activated DAF-16 only promotes germ line arrest of daf-2 worms if the damage signal emanates from somatic tissues.

Next, we knocked down cdk-12 in different somatic tissues using the widely used tissue-specific RNAi strains (Calixto et al., 2010; Espelt et al., 2005; Medwig-Kinney et al., 2020; Qadota et al., 2007). Remarkably, we found that KD of cdk-12 in only the uterine tissues was sufficient to arrest the germ line of the daf-2 worms at the pachytene stage of meiosis (Fig. 5C,D). We validated the uterine-specific strain by KD of egl-43 and nhr-67 (which are important for uterine development), which resulted in protrusion of the vulva (Fig. S5C), thereby elucidating active RNAi machinery in the uterine tissue. To check the specificity, we knocked down elt-2 (expressed in the intestine) and egg-5 (expressed in the oocytes) in the wild-type or the uterine-specific RNAi strain (Fig. S5D). We found developmentally retarded worms and dead eggs after whole-body KD of elt-2 and egg-5 RNAi, respectively, in wild type (Fig. S5D). The uterine-specific RNAi strain showed no such phenotype after the elt-2 or egg-5 RNAi, highlighting that the RNAi machinery is restricted to the uterine tissue. Importantly, KD of cdk-12 in the hypodermis, muscle, intestine or neurons of daf-2 worms did not lead to sterility (Fig. S5E).

We then investigated whether defects in uterine tissue caused by cdk-12 KD were the cause of the arrest. We observed that knocking down cdk-12 led to a protruding vulva in around 30% of the worms (Fig. S6A). Protruding vulva was not the cause of daf-2(-) oogenesis arrest after cdk-12 KD because (1) the remaining ∼70% of worms did not have protruded vulva and still arrested oogenesis; and (2) when we knocked down genes known to produce a protruded vulva (egl-13, lin-12, lin-31, unc-15, lin-7, egl-26 and egl-15), none of them produced sterility in daf-2 worms even if the vulvas of these worms were defective (data not shown). We also used phalloidin staining (which marks the F-actin) but found no gross defect in the vulval muscles (Fig. S6B). Furthermore, we used the lim-7::gfp transgenic worms that express GFP in the sheath cells (Voutev et al., 2009) but also did not observe gross morphological defects (Fig. S6C). Altogether, these results imply that defects in uterine/somatic gonad morphology may not be the signal sensed for germ line pachytene arrest.

Next, to determine the tissues where DAF-16 is required to sense and mediate the germ line arrest/sterility in the daf-2 mutant upon cdk-12 KD, we used transgenic lines where the wild-type copy of daf-16 is rescued only in the neurons (using either unc-119 or unc-14 promoters), uterus, muscles or intestine of the daf-16;daf-2 mutants. cdk-12 was then knocked down in these strains using RNAi. We found maximum sterility when daf-16 is rescued in the muscle, neuron or uterine tissues of the daf-16;daf-2 mutant worms, but not in the intestine (Fig. 5E). Altogether, upon DDR perturbations in the uterine tissue, activated DAF-16 works in the somatic uterine tissues, apart from muscle and neurons to implement germ line arrest.

Furthermore, we speculated that the communication could be through hormonal signaling, which can aid in the soma-germ line crosstalk. Earlier studies have elucidated the role of the DAF-12 nuclear hormone receptor/steroid hormone signaling pathway in life span extension by germ cell perturbation (Hsin and Kenyon, 1999; Yamawaki et al., 2010). Moreover, DAF-12 bound to its ligand, dafachronic acid (DA), has also been shown to negatively affect germ cell proliferation (Mukherjee et al., 2017). We found that the germ line arrest, when cdk-12 is KD in daf-2 worms, is partially rescued upon loss of daf-12 (Fig. 5F). This suggests that inter-tissue crosstalk that mediates germ line arrest in daf-2 worms after DDR disruption by cdk-12 RNAi may require steroid hormone signaling as well as other undetermined signals.

As DDR perturbation (by cdk-12 KD) in only the uterine tissue is sufficient to mediate germ line arrest in the daf-2 mutant, we next asked whether this perturbation regulates germ cell quality (measured in terms of dead eggs and progeny development) or whether it is only considered to be an intrinsic danger signal for impending damage to the germ line that instructs the arrest. For this, we performed RNAi of cdk-12 only in the uterine tissue of wild-type animals and quantified the egg hatching. Interestingly, we found that the egg hatching was unaffected (Fig. S6D), suggesting that cdk-12 is not required in the uterine tissue to maintain germ cell quality. However, directly knocking down cdk-12 in the germ line of wild-type animals produced some dead eggs (Fig. S6E). Thus, cdk-12 regulates germ cell quality autonomously but the arrest signal generated when it is knocked down in the uterus leads to meiosis arrest in daf-2 worms. Together, these observations support a model where low IIS sensitizes uterine tissues to perturbations of DDR, leading to a non-cell-autonomous arrest of the germ line at the pachytene stage of meiosis.

These experiments imply that KD of cdk-12 in daf-2 germ cells may lead to DNA damage, consequently resulting in poor progeny production. However, knocking cdk-12 down in the somatic uterine tissue of daf-2 worms may involve activated DAF-16-dependent quality checkpoints that lead to non-cell-autonomous germ line arrest in meiosis.

CDK-12 is a well-studied protein that is involved in DDR and genome integrity in mammalian cells (Blazek, 2012; Blazek et al., 2011). Here, we also show that in C. elegans, cdk-12 regulates the expression of DDR genes and helps maintain genome integrity. The depletion of cdk-12 makes the worms susceptible to DNA-damaging agents and induces spontaneous DNA damage in both the soma and the germ line, implying a suboptimal repair pathway. We show that cdk-12 ablation reduces gamete quality, leading to increased infertility as well as accelerated reproductive senescence, which are common outcomes of genomic instability. Thus, CDK-12 is an evolutionarily well-conserved custodian of the genome that helps maintain DNA integrity, which we have used in our study as a genetic tool to analyze the effects of tissue-restrictive DDR perturbation and DNA damage.

We have shown that uterine tissue-specific KD of cdk-12 leads to the arrest of the germ line of daf-2 worms. We believe that this is due to uterine tissue-intrinsic DDR perturbation leading to DNA damage signals that orchestrate the arrest in daf-2 worms. We acknowledge that, owing to technical challenges, there is a lack of direct evidence of DNA damage in the uterine cells per se upon cdk-12 KD. However, we have shown systemic cdk-12 depletion leads to transcriptional downregulation of DDR gene expression. We have also functionally validated the impairment of DDR in both somatic tissue and the germ line. For soma, we have measured intestinal karyokinesis defects induced by DNA damage and quantified somatic development to check the sensitivity to DNA damaging agents upon cdk-12 depletion. For examining germ line DDR, we checked chromosome fragmentation, egg hatching and apoptosis of germ cells. In both cases, we observed an increase in damage or sensitivity upon cdk-12 depletion. Together, this suggests that cdk-12 not only regulates DDR in the germ line but also in the somatic tissue. Based on these data, we could infer that uterine-specific KD of cdk-12 also causes DDR perturbation and DNA damage that led to the arrest of the germ line in daf-2 worms.

To maintain genomic integrity, organisms have evolved efficient DDR pathways that sense and repair DNA damage (Harrison and Haber, 2006). Defects in DDR are associated with reduced fitness, infertility and offspring with inherited diseases (Adriaens et al., 2009; Cleaver et al., 2009). Our study shows that activated FOXO/DAF-16 may sense DDR perturbation or DNA damage, stop reproduction by arresting the germ line, and protect the genomic integrity of the germ cell. In the absence of DAF-16, worms fail to arrest germ cell proliferation and produce oocytes of poor quality. Thus, activated FOXO/DAF-16 critically regulates reproductive decisions by sensing the intrinsic threat of genomic instability. Previous studies have shown that DAF-16 acts as a nutrient sensor and mediates somatic developmental arrest caused by starvation, as a protective mechanism (Baugh and Sternberg, 2006), by activating cell cycle inhibitors. Together, these data suggest that FOXO/DAF-16 plays a central role in sensing intrinsic as well as extrinsic stress to regulate both somatic and germ line development to ensure the best chance for survival of parents and their offspring.

We found that, upon DDR perturbation and ensuing DNA damage, FOXO/DAF-16 enforces germ line arrest partly by inactivating RAS-ERK signaling, which is essential for pachytene exit and initiation of oogenesis. In many cancers, RAS-ERK negatively regulates FOXO activity and promotes rapid proliferation (Yang and Hung, 2011). Similarly, we observed that constitutively activated RAS-ERK in the low IIS mutant (where FOXO/DAF-16 is activated) overrides the germ line arrest upon DDR perturbation, leading to the production of unhealthy progenies. Therefore, RAS-ERK and FOXO/DAF-16 regulate the activity of one another and a fine balance is important for various biological processes, including reproductive development. Apart from lowering RAS-ERK signaling to prevent oogenesis, there may be other mechanisms by which activated DAF-16 arrests germ line development, an interesting subject for future studies.

Non-cell-autonomous inter-tissue crosstalk helps an organism to perceive and respond to a changing environment. Multiple studies in C. elegans have revealed that non-cell-autonomous crosstalk is involved in the stress response and longevity (Miller et al., 2020). DAF-2 in the neuron and DAF-16 in the intestine are known to regulate longevity non-cell-autonomously (Lin et al., 2001; Wolkow et al., 2000). Muscle or intestinal FOXO/DAF-16 activity promotes a long reproductive span or better oocyte quality in the daf-2 mutant (Luo et al., 2010). Activated DAF-16 in the vulval tissue can sense damage to the eggshell vitelline layer and mount an organism-wide protective response (Sala et al., 2020). DAF-16 has also been shown to function in the uterine tissue to prevent a decline in the germ line progenitor cells with age (Qin and Hubbard, 2015). However, it is not clear how DAF-16 may regulate germ line and progeny health in response to somatic DNA damage. We show that perturbation of the DDR pathway in only the somatic uterine tissue of low IIS worm is sufficient to cause cell cycle arrest in the germ line; perturbation in the germ line itself does not lead to arrest but produces unhealthy oocytes. This suggests that the somatic tissue, not the germ line, senses the stress signal of genome instability and shunts its energy and resources toward somatic maintenance rather than reproductive commitment. This is supported by the observations of heightened stress response pathways and retarded germ line growth upon DDR perturbation. In this study, we show that the soma-germ line crosstalk may be regulated by the dafachronic acid pathway transcription factor DAF-12. However, in the future, it will be interesting to study the detailed mechanisms of the soma-germ line signaling that FOXO/DAF-16 controls.

Unless otherwise mentioned, all the C. elegans strains were maintained and propagated at 20°C on E. coli OP50 using standard procedures (Stiernagle, 2006). The strains used in this study are:

N2 var. Bristol: wild type

CB1370: daf-2(e1370) III

GR1307: daf-16(mgDf50) I

HT1890: daf-16 (mgDf50) I; daf-2 (e1370) III

CU1546: smIs34 [ced-1p::ced-1::GFP + rol-6(su1006)]

JT9609: pdk-1(sa680) X

TJ1052: age-1(hx546) II

KW2126: ckSi6[cdk-12::GFP+unc-119(+)] I; cdk-12(tm3846) III

DCL569: mkcSi13[sun-1p::rde-1::sun-1 3'UTR+unc-119(+)] II; rde-1(mkc36) V

NR350: rde-1(ne219) V; kzIs20[hlh-1p::rde-1+sur-5p::NLS::GFP]

NR222: rde-1(ne219) V; kzIs9[(pKK1260) lin-26p::NLS::GFP+(pKK1253) lin-26p::rde-1+rol6(su1006)]

VP303: rde-1(ne219) V; kbIs7[nhx-2p::rde-1+rol-6(su1006)]

NK640: rrf-3(pk1426) II; unc-119(ed4) III; rde-1(ne219) V; qyIs102[fos-1ap::rde-1(genomic)+myo-2::YFP+unc-119(+)]

TU3401: sid-1(pk3321) V; uIs69[pCFJ90 (myo-2p::mCherry)+unc-119p::sid-1] V

SD551: let-60(ga89) IV

WM27: rde-1(ne219) V

WS1433: hus-1(op241) I; unc-119(ed3) III; opIs34[hus-1p::hus-1::GFP+unc-119(+)]

SJ4005: zcIs4 [hsp-4::GFP] V.

The following strains were generated in-house using standard cross-over techniques:

daf-2(e1370) III; daf-12(rh61rh411) X

daf-2(e1370) III: rde-1(ne219) V

daf-2(e1370) III; let-60(ga89) IV

daf-2(e1370) III; mkcSi13[sun-1p::rde-1::sun-1 3'UTR+unc-119(+)] II]; rde-1(mkc36) V

daf-2(e1370)III; rde-1(ne219) V; kzIs20[hlh-1p::rde-1+sur-5p::NLS::GFP]

daf-2(e1370) III; rde-1(ne219) V; kzIs9[(pKK1260) lin-26p::NLS::GFP+(pKK1253) lin-26p::rde-1+rol-6(su1006)]

daf-2(e1370) III; rde-1(ne219) V; kbIs7[nhx-2p::rde-1+rol-6(su1006)]

daf-2(e1370) III; sid-1(pk3321) V; uIs69[pCFJ90 (myo-2p::mCherry)+unc-119p::sid-1] V

daf-2(e1370) III; unc-119(ed4) III; rde-1(ne219) V; qyIs102[fos-1ap::rde-1(genomic)+myo-2::YFP+unc-119(+)]

daf-2(e1370) III; smIs34 [ced-1p::ced-1::GFP + rol-6(su1006)]

daf-16(mgDf50) I; daf-2(e1370) III; smIs34 [ced-1p::ced-1::GFP + rol-6(su1006)].

RNAi plates were poured using autoclaved nematode growth medium supplemented with 100 µg/ml ampicillin and 2 mM IPTG. Plates were dried at room temperature for 2-3 days. Bacterial culture harboring an RNAi construct was grown in Luria Bertani (LB) media supplemented with 100 µg/ml ampicillin and 12.5 µg/ml tetracycline, overnight at 37°C in a shaker incubator. Saturated cultures were re-inoculated the next day in fresh LB media containing 100 µg/ml ampicillin by using 1/100th volume of the primary inoculum and grown in a 37°C shaker until OD₆₀₀ reached 0.5-0.6. The bacterial cells were pelleted down by centrifuging the culture at 3214 g for 10 min at 4°C and resuspended in 1/10th volume of M9 buffer containing 100 µg/ml ampicillin and 1 mM IPTG.

Strong cdk-12 KD leads to developmental defects and the cdk-12 mutants are non-viable. We therefore diluted the cdk-12 RNAi with control RNAi-expressing bacteria, according to the experimental requirements. Different dilutions of RNAi were made by mixing with the control RNAi feed. For cdk-12 RNAi plates, we have used a 1:3 dilution of cdk-12:control RNAi. Around 300 µl of this suspension was seeded onto RNAi plates and left at room temperature for 2-3 days for drying, followed by storage at 4°C until further use.

Gravid adult worms, initially grown on E. coli OP50 bacteria, were collected using M9 buffer in a 15 ml Falcon tube. Worms were washed thrice by first centrifuging at 652 g for 60 s followed by resuspension of the worm pellet in 1× M9 buffer. After the third wash, the worm pellet was resuspended in 3.5 ml of 1× M9 buffer and 0.5 ml 5 N NaOH and 1 ml of 4% sodium hypochlorite solution was added. The mixture was vortexed for 5-7 min until the entire worm body dissolved, leaving behind the eggs. The eggs were washed five or six times, by first centrifuging at 1258 g, decanting the 1× M9 and resuspending in fresh 1× M9 buffer to remove traces of bleach and alkali. After the final wash, eggs were kept in 15 ml Falcon tubes with ∼10 ml of 1× M9 buffer and rotated at ∼15 rpm for 17 h to obtain L1 synchronized worms for all strains. The L1 worms were obtained by centrifugation at 805 g followed by resuspension in ∼100-200 µl of M9 and added to different RNAi plates.

Worms grown on control or cdk-12 RNAi were collected using 1× M9 buffer and washed thrice to remove bacteria. Trizol reagent (200 μl; Takara Bio, Kusatsu, Shiga, Japan) was added to the 50 μl worm pellet and the sample was subjected to three freeze-thaw cycles in liquid nitrogen with intermittent vortexing to break open worm bodies. The samples were then frozen in liquid nitrogen and stored at −80°C until further use. For experiments, 200 μl of Trizol was again added to the worm pellet and the sample was vortexed vigorously. Chloroform (200 μl) was then added and the tube was gently inverted several times followed by 3 min of incubation at room temperature. The sample was then centrifuged at 12,000 g for 15 min at 4°C. The RNA containing the upper aqueous phase was gently removed into a fresh tube without disturbing the bottom layer and interphase. To this aqueous solution, an equal volume of isopropanol was added and the reaction was allowed to sit for 10 min at room temperature followed by centrifugation at 12,000 g for 10 min at 4°C. The supernatant was carefully discarded without disturbing the RNA-containing pellet. The pellet was washed using 1 ml 70% ethanol solution followed by centrifugation at 12,000 g for 5 min at 4°C. The RNA pellet was further dried at room temperature and later dissolved in nuclease-free water (Qiagen) followed by incubation at 65°C for 10 min with intermittent tapping. The concentration of RNA was determined by measuring absorbance at 260 nm using a NanoDrop UV spectrophotometer (Thermo Fisher Scientific) and quality checked using denaturing formaldehyde-agarose gel.

First-strand cDNA synthesis was carried out using the Iscript cDNA synthesis kit (Bio-Rad) following the manufacturer's guidelines. The prepared cDNA was stored at −20°C. Gene expression levels were determined using the Brilliant III Ultra-Fast SYBR Green QPCR master mix (Agilent) and Agilent AriaMx Real-Time PCR system, according to manufacturer's guidelines. The relative expression of each gene was determined by normalizing the data to actin expression levels. The list of primers is summarized in Table S2.

Worms were grown under control RNAi conditions from L1 stage onwards, and ∼50 L4 worms for each strain were transferred to control or cdk-12 RNAi-seeded NGM plates in triplicate and transferred to an incubator maintained at 20°C. About 48 h post-transfer, the RNAi plates were transferred to an incubator maintained at 35°C. After this, survival was scored every 2 h until all worms were dead.

HSP-4::GFP transgenic worms were grown under control or cdk-12 RNAi conditions. Day 1 adults were mounted on 2% agarose pad-coated microscope slides in 20 mM sodium azide and imaged using an Axioimager M2 fluorescence microscope (Zeiss).

The number of engulfed cell corpses was analyzed using CED-1::GFP-expressing transgenic worms, where CED-1 is a transmembrane protein expressed on phagocytic cells that engulf cell corpses. Transgenic worms, expressing CED-1 fused to GFP under ced-1 promoter [ced-1p::ced-1::GFP(smIs34)], were bleached and their eggs were allowed to hatch in 1× M9 buffer for 17 h to obtain L1 synchronized worms. Approximately 200 L1 worms were placed onto control or cdk-12 RNAi in triplicate. At day 1 adult stage, worms were visualized under a LSM-980 confocal microscope (Zeiss). The number of cell corpses per gonad was counted.

Worms were grown under control or cdk-12 RNAi conditions from L1 stage onwards. Day 1 adults were collected in 1× M9 buffer in a 1.5 ml Eppendorf tube and worms were allowed to settle down. Using a glass Pasteur pipette, the 1× M9 was discarded, leaving behind a ∼100 µl worm suspension. 1 ml chilled 100% methanol was added to the worm pellet and incubated for 30 min at −20°C. The methanol-fixed worm pellet was placed on a glass slide and after the methanol had dried, Fluoroshield with DAPI (Invitrogen) was added. For staining dissected gonads, worms were placed in 1× M9 onto a glass slide and the gonads were obtained by cutting the pharynx or tail end of the worm using a sharp 25 G needle. After cutting gonads for 10 min, 500 µl chilled 100% methanol was added onto the slide and allowed to dry. Fluoroshield with DAPI (Invitrogen) was added. The slides were imaged using an LSM 980 confocal microscope (Zeiss).

Worms were grown under control or cdk-12 RNAi conditions from L1 onwards and upon reaching the young adult stage, five worms were picked onto fresh RNAi plates, in triplicate, and allowed to lay eggs for 24 h. The worms were then transferred to fresh plates every day until worms ceased to lay eggs, and the eggs laid on the previous day's plate were counted. These plates were again counted after 48 h to document the number of hatched worms and the unhatched eggs were considered dead eggs. The total number of hatched progenies per worm is defined as brood size.

The number of days from the L4 larval stage to the last day of self-progeny production is defined as the reproductive span. The number of progenies per worm was plotted for each day to observe the reproductive span. For the reproductive span of mated worms, the hermaphrodites were mated on 35 mm cross plates with fresh males (1:3 ratio of hermaphrodite:male) each day for 4 h and shifted back to 60 mm RNAi plates. The rest was performed according to the reproductive span of selfed worms. Egg quality was determined by calculating the percentage of hatched progeny in different conditions.

pMPK-1 immunostaining was performed as described previously (Gervaise and Arur, 2016). Briefly, on day 1, adult worms were dissected in 1× M9 buffer to obtain gonads. The dissections were performed on a glass slide and within a 5 min window to prevent the loss of the pMPK-1 signal. Following this, the dissected gonads and remaining worms were transferred to a 10 ml round bottomed glass tube using a glass pipette. To this, 2 ml of 3% paraformaldehyde (PFA) was added and incubated at room temperature for 10 min. Next, 3 ml of 1×PBST was added for washing to remove PFA. The tube was allowed to stand until the dissected gonads and residual intact worms settled to the bottom. After removing the supernatant, the washing was repeated twice more. After the final PBST wash, 2 ml of 100% methanol was added and the tubes were incubated at −20°C for 1 h. Three 1×PBST washes were then given, as described previously, and after the final wash, the worms were transferred to a 1.5 ml glass tube. Excess PBST was removed carefully using a glass Pasteur pipette. Blocking was performed at room temperature for 1 h using 100 µl of 30% normal goat serum (NGS) per tube. After blocking, 100 µl of pMPK-1 antibody (M8159, Sigma) diluted 1:400 in 30% NGS was added to each tube, and tubes were capped, sealed with parafilm to prevent loss by evaporation and stored at 4°C overnight. After three 1×PBST washes, 100 µl of secondary antibody in 30% NGS was added to each tube and incubated at room temperature for 2 h. Again, three 1×PBST washes were given and excess PBST was removed. Glass Pasteur pipettes were used to pick the stained gonads in 1×PBST onto the glass slide. Quickly, before the slides were completely dried, Fluoroshield with DAPI (Invitrogen) was added and a coverslip was slowly positioned using a needle, to avoid air gaps. The coverslip was gently pressed, and the edges were sealed with transparent nail paint. The slides were imaged using an LSM 510 confocal microscope (Zeiss). The pMPK-1 signal was quantified using ImageJ software.

Worms were grown from L1 onwards under control or cdk-12 RNAi conditions. Differential interference contrast (DIC) images of the oocytes were captured on day 1 and day 3 of adulthood. Oocyte images were categorized into three groups based on their morphology (cavities, shape, size and organization). Based on the severity of the phenotype, oocytes were categorized as normal, mild or severe. No cavities, no small oocytes and no disorganized oocytes were categorized as normal. The presence of two small oocytes, two cavities or two disorganized oocytes was scored as a mild phenotype. More than two cavities, small oocytes or mis-shapen oocytes were categorized as severe.

Worms were bleached and their eggs were allowed to hatch in 1× M9 buffer for 17 h to obtain L1 synchronized worms. Approximately 100 L1 worms were placed onto different RNAi plates, in triplicate. At the day 1 adult stage, bright-field images were captured using an Axioimager M2 brightfield microscope (Zeiss). Worms with more than five eggs in the uterus were considered fertile.

Approximately 100 L1 worms were placed onto control or cdk-12 RNAi plates in triplicate. Images of DAPI-stained day 1 adult germline were captured using an LSM 980 confocal microscope (Zeiss). The best plane (maximum germ cells in that plane) was selected during imaging. Based on morphology, we quantified the germ cells in each zone manually: mitotic zone (i.e. distal tip to transition zone), transition zone (start to end of crescent-shaped nuclei) and pachytene zone (after the transition zone up to the turn region).

Approximately 100 L1 worms were placed onto control or cdk-12 RNAi plates in triplicate. At day 1 adult stage, the worms were DAPI stained and imaged using an LSM 980 confocal microscope (Zeiss). The oocytes were counted by quantifying diakinesis nuclei in the DAPI-stained adult gonads.

Early young adults (four or five eggs in a gonad) were DAPI stained and imaged using an LSM 980 confocal microscope (Zeiss). The best plane (maximum germ cells in that plane) was selected during imaging. Numbers of sperm per gonadal arm were quantified manually.

Approximately 100 L1 worms were placed onto control or cdk-12 RNAi plates in triplicate. The worms were irradiated with ionizing radiation (IR) of different doses at the L4 stage. After 48 h, the worms were stained with DAPI, and the oocyte chromosomes were imaged in z-stack using a LSM980 confocal microscope (Zeiss). For scoring chromosome fragmentation, images were converted into maximum intensity projections (MIPs) and scored. We initially used a range of IR doses for standardization and finally used an appropriate dose with which the difference between control and cdk-12 KD could be resolved.

Approximately 50 L1 worms were placed onto control or cdk-12 RNAi plates in triplicate. Day 1 adult worms were stained with DAPI and imaged in z-stack using a LSM980 confocal microscope (Zeiss). Enlarged nuclei densely stained with DAPI (Iwasaki et al., 1996) were identified as endomitotic oocytes (EMOs) and, for scoring, the images were converted into maximum intensity projections (MIPs). The percentage of worms with one or more EMOs in the uterus or gonad arms were determined.

Approximately 50-80 L1 worms were placed onto control or cdk-12 RNAi plates in triplicate. Day 1 adult worms were stained with DAPI and the intestinal nucleus was imaged using a LSM980 confocal microscope (Zeiss).

For the sterility assay, ∼100 L1 worms were placed onto control RNAi plates and treated with different doses of IR. Day 1 adult worms were imaged under an Axioimager M2 bright-field microscope (Zeiss). Worms with more than five eggs in the uterus were considered to be fertile. Initially a range of IR doses was used for standardization. As L1s are known to be resistant to IR, higher doses were selected.

Similarly, for the developmental assay, ∼100-130 L1 worms were treated with different doses of IR while growing on control RNAi. IR-treated L1 were then transferred to different RNAi plates and scored for worms that reached L4 or above after 100 h.

Worms were grown under control or cdk-12 RNAi conditions. At the young adult stage, worms were exposed to IR doses ranging between 0 and 40 Gy. The IR-treated worms were allowed to recover for 3-4 h, after which five worms were transferred to respective RNAi plates, in duplicate, and then incubated for 18-20 h at 20°C. The adults were sacrificed, and the number of eggs laid on the plates was counted. About 48 h later, the number of hatched progenies was also counted.

The working stock of CPT (1 or 2 µM) was made in 10× concentrated bacterial feed suspended in 1×M9 buffer. Worms were added to wells containing CPT in liquid bacterial feed. The plates were wrapped with foil and incubated at 20°C for 18-20 h, with gentle shaking. The worms were then transferred to Eppendorf tubes and washed twice with 10% Triton X-100 (in 1×M9 buffer), followed by two washes with 1×M9 buffer. The worms were then placed on RNAi plates to recover for 3-4 h, followed by tight egg-laying for 3-4 h on fresh, respective RNAi plates. The adults were sacrificed and the number of eggs laid was counted. About 48 h later, the number of hatched progenies was also counted.

HUS-1::GFP transgenic worms were grown under control or cdk-12 RNAi conditions. At the young adult stage, worms were treated with 0 or 100 Gy IR dose and incubated for 28 h at 20°C. The adults were then mounted on 2% agarose pad-coated microscope slides in 20 mM sodium azide and z-stacked images were captured using a 488 nm laser channel in a confocal microscope (LSM980 Zeiss). Under normal conditions, HUS-1::GFP is diffused in the cytoplasm and only weakly localized in the chromatin of germ cells. However, upon exposure to DNA damage (e.g. gamma radiation), HUS-1::GFP concentrates at nuclear foci in all germ cells and is seen as bright foci. The HUS-1::GFP bright foci were quantified in 40 proliferating germ cells (in ten gonads per condition).

The different IR doses used are due to differences in the sensitivity of the assays. Although 60 Gy of IR dose is sufficient for chromosome breakage (Fig. 1B,C), at this dose the HUS-1::GFP foci were found to be less different between the untreated and irradiated control RNAi-fed worms. Upon increasing the IR dose to more than 60 Gy, extensive chromosome breakage was found in control RNAi-fed worms itself, obscuring the role of cdk-12. Therefore, a 60 Gy dose was selected for chromosome fragmentation assay, while a higher dose (100 Gy) was used for the HUS-1::GFP foci detection assay as the fluorescence of foci had to be significant enough for detection.

Synchronized late-L4 worms grown under control or cdk-12 RNAi conditions were collected using 1×M9 buffer, after washing it thrice to remove bacteria. Total RNA was isolated from these worm pellets using the Trizol method. The concentration of RNA was determined by measuring absorbance at 260 nm using a NanoDrop UV spectrophotometer (Thermo Fisher Scientific) and RNA quality was checked using RNA 6000 NanoAssay chip on a Bioanalyzer 2100 machine (Agilent Technologies). RNA with an integrity number of eight or more was included in the study. In one batch the sequencing Libraries were constructed using a NEBNext Poly(A) mRNA Magnetic Isolation Module (E7490L, New England Biolabs) and a NEBNext UltraTM II Directional RNA Library Prep kit (E7765L, New England Biolabs), according to the manufacturer's instructions. For sequencing, equimolar quantities of all libraries were pooled and sequenced on Illumina Hiseq 2500 sequencer (Illumina) as per the manufacturer's instructions using Hiseq Rapid v2 single end 50 cycles kit (1×50 cycles). In another independent batch, libraries were constructed using Truseq stranded mRNA library (for human/animal/plant) and sequencing was performed using a NovaSeq 6000 platform, 100 bp paired-end (PE) with 30 million reads. The RNA sequencing data have been deposited in ArrayExpress under accession number E-MTAB-11189.

Sequencing reads were subjected to quality control using the FASTQC kit. Alignment of the reads to the WBcel235 genome was carried out with Tophat2 (Kim et al., 2013) version 2.1.0 with an average 95% alignment rate. No novel junctions and novel insertions-deletions were considered with the parameters ‘-no-novel-junc’ and ‘no-novel-indel’, respectively. Gene counts were obtained with feature counts (Liao et al., 2014) version 1.6.3 and WBcel235 Ensembl annotation v95. Gene expression analysis was performed using the DeSeq2 (Love et al., 2014) package. Differentially expressed genes were defined as those with P-values below 5%. Genes with a cut-off of fold change >2 and fold change <−2 were considered upregulated and downregulated, respectively. For downstream analysis, the function variance stabilizing transformations (VST) (Anders and Huber, 2010) in the DeSeq2 package were implemented. Enrichment analysis was performed using the online tool DAVID 6.8 with a cutoff of FDR<10%. The dot plot was plotted with ggplot2 in R. The heatmap was plotted with the help of the heatmap function in R.

All statistical tests were performed with GraphPad Prism 9.5 using inbuilt functions. The tests used for each comparison are indicated in the figure legends. When we compared the two conditions, we used the Student's t-test with Welch's correction, with no assumptions for consistent standard deviations. When more conditions were compared, we used two-way ANOVA with Tukey's multiple comparison test. The statistical tests used in each figure are summarized in Table S3.

We thank all members of Molecular Aging for their support, and Dr K. Subramaniam and A. Ghosh-Roy for their comments. Some strains were provided by the CGC, which is funded by the NIH Office of Research Infrastructure Programs (P40 OD010440) and the National Bioresource Project (NBRP), Japan. The GC1285 strain [daf-16(m26);daf-2(e1370);naEx239[pGC629(Pfos-1a::gfp::daf-16)+pRF4] was kindly provided by Dr E. Jane Albert Hubbard. This work was supported by core funding from the National Institute of Immunology. The schematic representations were created with BioRender.com.

The peer review history is available online at https://journals.biologists.com/dev/lookup/doi/10.1242/dev.201472.reviewer-comments.pdf