Age‐dependent nuclear lipid droplet deposition is a cellular hallmark of aging in Caenorhabditis elegans

Abstract Aging is the major risk factor for several life‐threatening pathologies and impairs the function of multiple cellular compartments and organelles. Age‐dependent deterioration of nuclear morphology is a common feature in evolutionarily divergent organisms. Lipid droplets have been shown to localize in most nuclear compartments, where they impinge on genome architecture and integrity. However, the significance of progressive nuclear lipid accumulation and its impact on organismal homeostasis remain obscure. Here, we implement non‐linear imaging modalities to monitor and quantify age‐dependent nuclear lipid deposition in Caenorhabditis elegans. We find that lipid droplets increasingly accumulate in the nuclear envelope, during aging. Longevity‐promoting interventions, such as low insulin signaling and caloric restriction, abolish the rate of nuclear lipid accrual and decrease the size of lipid droplets. Suppression of lipotoxic lipid accumulation in hypodermal and intestinal nuclei is dependent on the transcription factor HLH‐30/TFEB and the triglyceride lipase ATGL‐1. HLH‐30 regulates the expression of ATGL‐1 to reduce nuclear lipid droplet abundance in response to lifespan‐extending conditions. Notably, ATGL‐1 localizes to the nuclear envelope and moderates lipid content in long‐lived mutant nematodes during aging. Our findings indicate that the reduced ATGL‐1 activity leads to excessive nuclear lipid accumulation, perturbing nuclear homeostasis and undermining organismal physiology, during aging.

degradation of targeted proteins (Farese Jr. & Walther, 2009). Lipid droplets are mainly generated from ER membranes and are localized in the cytoplasm close to the outer nuclear membrane. However, accumulating evidence demonstrates the presence of LDs inside the nucleus. Thus, the association between lipid droplets and the nucleus is an emerging topic.
Nuclear lipid droplets appear to be smaller and have different lipid composition than their cytoplasmic counterparts. Moreover, when cells are cultured in a fatty acid-rich media, the number of nLDs is increased (Barbosa & Siniossoglou, 2020). Regarding their function, it is suggested that nLDs can serve as storage sites for histones, bind to transcription factors or regulate the transport of specific proteins between the nucleus and cytoplasm (Welte, 2015). A very recent study in C. elegans underlines the essential role of nLDs formation in the maintenance of intestinal and gonadal homeostasis (Mosquera et al., 2021). The enhanced number and increased size of nLDs in several human pathologies, such as atherosclerosis, obesity, fatty liver disease, and hepatic steatosis, underlines the detrimental effects of nLDs deregulation on cellular and tissue physiology (Barbosa & Siniossoglou, 2020;Sołtysik et al., 2019).
Despite the recent advances in nuclear lipid droplet research, the regulatory mechanisms dictating nuclear homeostasis collapse and nuclear lipids deposition during aging remain elusive. Here we report the use of non-linear imaging modalities to monitor and quantify nLDs and delineate the precise connection between nuclear impairment and aging. Our studies suggest that lipid content progressively accumulates with age in the nuclear envelope in several tissues, such as hypodermis and intestine of C. elegans.
Importantly, genetic interventions known to delay aging in different model organisms, such as low insulin signaling and dietary restriction, decrease the number of nuclear lipids and reduce their size. HLH-30, the homolog of the mammalian TFEB and the master regulator of lysosomal function, is required for the maintenance of nLDs homeostasis. Although autophagy is a well-known moderator of lipid content, it does not affect the age-dependent accrual of nLDs suggesting that HLH-30 regulates nLDs in an autophagyindependent manner. HLH-30 dictates the expression of several lipases, such as LIPL-4 and ATGL-1 (the homolog of the mammalian adipose triglyceride lipase (ATGL)). Interestingly, ATGL-1 is located both in cytoplasm and nucleus and regulates nuclear lipid deposition in long-lived animals. Notably, age-dependent elevation of nuclear fat deposition is linked to increased LMN-1/LMNA protein levels. Although nLDs and LMN-1 abundance are governed by HLH-30, ATGL-1 does not influence the protein levels of LMN-1 indicating that the regulation of these two phenomena is mediated by distinct HLH-30-driven mechanisms. Our findings highlight the pivotal role of HLH-30/ATGL-1 axis in restraining lipid expansion in the nucleus, thereby preserving nuclear lipid homeostasis and organismal fitness during aging.

| Age-dependent alterations in nuclear structure and lipid content
Several hallmarks of aging are tightly associated with nuclear homeostasis (Hou et al., 2019;Lopez-Otin et al., 2013;Schmauck-Medina et al., 2022). Thus, it is highly appreciated that nucleus serves a paramount role in the determination of organismal healthspan and lifespan. During aging, the nucleus undergoes a functional decline that is coupled with a progressive deterioration in its morphology (Burke & Stewart, 2014;Haithcock et al., 2005;Papsdorf & Brunet, 2019;Pathak et al., 2021;Schumacher & Vijg, 2019). To examine nuclear morphology during the aging process, we utilized transgenic nematodes expressing the nuclear lamina protein LMN-1 (the ortholog of mammalian LMNA) fused to GFP under its endogenous promoter.
Several tissues, including hypodermis, body wall muscles, and intestine, showed gradual deterioration of the well-structured nuclear shape with age (Figure 1a-c; Haithcock et al., 2005;Pathak et al., 2021). Although recent studies highlight the detrimental effects of age-dependent and excessive deposition of cytosolic LDs both in adipose and non-adipose tissues on organismal physiology, the abundance of nLDs and their impact on cellular function and tissue homeostasis are still obscure (Barbosa & Siniossoglou, 2020).
We utilized third harmonic generation (THG) imaging technique to detect and quantify nLDs deposition in nematodes. Third harmonic generation imaging constitutes a powerful, label-free and noninvasive method for the precise identification of fat content in vivo at microscopic level. Indeed, THG has already been used to detect lipid deposition in various biological samples including nematodes and mammalian cells (Débarre et al., 2006;Palikaras et al., 2017;Tserevelakis et al., 2014). The visualization and quantification of lipid droplets in the nuclei of C. elegans includes the combination of THG and two-photon-excitation fluorescence (TPEF) ( Figure S1).
Simultaneous THG and TPEF 3D imaging of 1-and 10-day-old transgenic nematodes expressing fluorescent reporters of the nuclear membrane (LMN-1::GFP and EMR-1::mCherry) and the nucleoplasm (H2B::mCherry) showed the presence of nLDs in the respective nuclear compartments ( Figure S2A-C). To measure the levels of nLDs, we performed the following image analysis. Nuclei were manually selected and isolated from each slice of the respective image. A constant threshold was set for all the respective regions of interest (ROIs) for each image. The total sum of the LDs for each animal was calculated through the resulting binary images and the total mean areas (in pixels) were divided by the number of nuclei detected in each animal. Then, the representative index of the total nuclear lipid content within the examined part was depicted in a graph bar ( Figure S3A, B). Our analysis suggests that nLDs number and size gradually increase in the nematode intestinal cells during aging.
Interestingly, age-dependent nuclear lipid deposition is prominent in the nuclear envelope (LMN-1::GFP, EMR-1::mCherry), whereas in the nucleoplasm (H2B::mCherry) no significant changes are observed in either nLD abundance or size (Figure 1e). Taken together, F I G U R E 1 Nuclear structure is altered during aging. (a) Representative images of hypodermal nuclei in animals of the LW697 [p lmn-1 LMN-1::GFP] reporter strain during aging. (b) Representative images depicting alterations (normal, intermediate, and severe) in the nuclear morphology of hypodermal cells in animals of the LW697 [p lmn-1 LMN-1::GFP] reporter strain. Scale bars, 10 μm. (c) Qualitative analysis of alterations in the nuclear morphology of hypodermal cells in animals of the LW697 [p lmn-1 LMN-1::GFP] reporter strain during aging (n = 50 nuclei per condition; ns p > 0.05, **p < 0.01, ***p < 0.001, one-way ANOVA followed by Tukey HSD post hoc test). (d-i) Quantification of LD number and size in distinct nuclear compartments of intestinal nuclei in 1-day-and 10-day-old animals, using a combined THG and TPEF imaging modalities. LW697 [p lmn-1 LMN-1::GFP] and BN147 [p emr-1 EMR-1::mCherry] reporter strains were used to identify alterations in the nuclear envelope, while the YQ243 [p app-1 H2B::mCherry] reporter strain was used to identify alterations in the intra-nuclear space (n = 15 for each genetic background and time point; ns p > 0.05, **p < 0.01, ***p < 0.001, unpaired t-test). Error bars denote the standard error of the mean. ** *** *** ns these findings established non-linear phenomena as a novel, nondestructive, and label-free microscopy method to assess nLDs in vivo and underline the age-dependent accrual of nLDs in the nuclear envelope.

| Caloric restriction and low insulin signaling diminish nuclear lipid droplet accrual with age
Aging is a universal phenomenon that is characterized by progressive decline of cellular function and tissue homeostasis collapse leading eventually to organismal degeneration and death. Nevertheless, several environmental and genetic factors have been shown to extend lifespan through the activation of multiple molecular mechanisms, which are conserved from unicellular eukaryotic organisms to mammals (DiLoreto & Murphy, 2015;Hou et al., 2019;Kenyon, 2010;Lopez-Otin et al., 2013;Melzer et al., 2020). Among those longevity pathways dietary restriction and low insulin signaling trigger a cascade of signaling events that stimulate subsequent metabolic changes to tackle aging and age-related pathologies (DiLoreto & Murphy, 2015;Kenyon, 2010;Papsdorf & Brunet, 2019).
Previous studies have shown the protective effect of caloric restriction and low insulin signaling against age-dependent ectopic fat deposition and lipotoxicity (Huffman & Barzilai, 2009;Muzumdar et al., 2008;Palikaras et al., 2017). Consistent with the previous reports, the long-lived eat-2(ad465) and daf-2(e1370) mutant nematodes displayed low levels of nuclear lipids and a concomitant decrease in their size, compared to their respective wild-type counterparts during aging (Figure 2a-c). Moreover, the quantity of nLDs is reduced in wild-type nematodes, which are subjected to 6 h starvation during their development ( Figure S4A, B). Thus, the steady levels of nLDs throughout the lifespan of long-lived nematodes suggest that lipid metabolism is enhanced in the nuclear envelope compartments of eat-2(ad465) and daf-2(e1370) mutants, resulting in less and smaller nLDs.
Nuclear envelope is a membranous and multiprotein structure that separates the nuclear from cytosolic compartments in eukaryotic cells. A wide variety of cellular processes, including gene expression, DNA metabolism, and chromatin organization among others, are coordinated by the dynamic nature of nuclear envelope, highlighting its essential role for cellular homeostasis (Cohen-Fix & Askjaer, 2017;Hetzer, 2010). Regarding its architecture, nuclear envelope is composed of three main domains: the outer nuclear membrane, the inner nuclear membrane, and the nuclear lamina (Cohen-Fix & Askjaer, 2017;Hetzer, 2010). A recent study in Saccharomyces cerevisiae uncovered that inner nuclear membrane is a metabolic active region, where local lipid metabolism is taking place to regulate transcription and lipid homeostasis (Romanauska & Kohler, 2018). To examine the impact of aging on nuclear envelope domains, we used transgenic nematodes expressing the inner nuclear membrane protein EMR-1 fused to mCherry (the homolog of the mammalian emerin) and the nuclear lamina protein LMN-1 fused with to GFP. We found that the abundance of LMN-1 is gradually elevated with age, whereas the protein levels of EMR-1 are not altered (Figure 3a-c and Figure S5A-E). Interestingly, this differential effect of aging on LMN-1 and EMR-1 protein levels is highly correlated with nLDs accumulation, indicating an intricate association between LMN-1 and nLDs ( Figure 3d). Further supporting this notion, the protein levels and the accumulation rate of EMR-1 remain stable in the long-lived nematodes, whereas LMN-1 levels are stable in eat-2(ad465) and display a slower accumulation rate in DAF-2 deficient animals during aging (Figure 3a-c and Figure S5A-E).
Collectively the above results suggest that both caloric restriction and low insulin signaling regulate the abundance of LMN-1 protein and nLDs sustaining nuclear lipid homeostasis.
Vitellogenins are highly expressed and large proteome contributors influencing intestinal lipid localization and distribution in C. elegans during aging. Notably, vitellogenins production is particularly reduced in long-lived mutant nematodes, including eat-2(ad465) and daf-2(e1370) (Dong et al., 2007;Murphy et al., 2003). Accumulating evidence suggests that VIT-2 protein levels are increased during aging and its depletion extends lifespan in C. elegans, while the expression of vit-2 gene is suppressed in daf-2(e1370) long-lived animals (DePina et al., 2011;Goszczynski et al., 2016;Mallick & Gupta, 2020;Murphy et al., 2003). Given that vitellogenins' production in the ER continues beyond the reproductive stage, it can be speculated that nLDs may arise from mis-regulated vitellogenin production. Interestingly, knocking down of vit-2 gene does not abolish nLDs' abundance during aging ( Figure S6A, B). Though, it seems that deficiency in VIT-2 expands nLD size in both young and old ages ( Figure S6B). Moreover, VIT-2 depletion does not affect the levels of LMN-1::GFP in wild-type and eat-2(ad465) nematodes, while it increases LMN-1::GFP fluorescent signal in daf-2(e1370) mutants ( Figure S6C). Taken together, these findings indicate that different longevity interventions differentially modify cytosolic and nuclear lipid droplets accumulation, though further investigation is needed.

| HLH-30 and ATGL-1 regulates nuclear lipid deposition in long-lived animals
Emerging findings suggest that the catabolic cellular process of autophagy regulates lipid metabolism through the direct degradation of LDs in several cell types (Aman et al., 2021;Klionsky et al., 2021;Liu & Czaja, 2013;Mizushima & Levine, 2020). It is widely appreciated that autophagy efficiency declines with age leading to excessive accrual of damaged organelles, toxic protein aggregates and subsequently to cellular and tissue degeneration (Aman et al., 2021;Klionsky et al., 2021;Mizushima & Levine, 2020). To elucidate whether the age-dependent accumulation of nLDs depends on autophagy, we monitored nLDs upon depletion of the autophagyrelated factors BEC-1 and LGG-1. Notably, BEC-1 and LGG-1 deficiency (the homolog of the mammalian Beclin and GABARAP respectively) does not affect either the number or the size of nLDs in 1-and 10-day-old wild-type, daf-2(e1370), and eat-2(ad465) nematodes (Figure 4a, b). Moreover, deficiency of LGG-2, the homolog of the mammalian LC3 autophagosomal protein, does not influence the quantity and size of nLDs both in wild-type and daf-2(e1370) animals ( Figure S7A-C). Notably, efficient autophagic machinery regulates the storage and the distribution of general lipid content during nematode development (Lapierre, Silvestrini, et al., 2013). However, the developmental impact of autophagy on nLDs homeostasis cannot be discriminated by its cytoprotective function in our experimental setup. Thus, further studies should focus on this direction.  Figure S9C).
In addition to its widely cytoplasmic distribution, ATGL-1::GFP is found to be in proximity to the nuclear envelope ( Interestingly, ATGL-1::GFP levels are slightly elevated in the nuclear extracts of the long-lived daf-2(e1370) nematodes further supporting its essential role in nLDs regulation ( Figure 5d). To examine whether ATGL-1 activity is sufficient for the maintenance of nLDs, we generated transgenic nematodes overexpressing ATGL-1, without any fusion tag, under its endogenous promoter ( Figure S9D). We found that ATGL-1 overexpressing animals display diminished nLD accumulation during aging (Figure 5e, f).
To investigate further the molecular mechanism that dictates the tight association between nLDs and LMN-1 accrual with age, we monitored the LMN-1::GFP signal in ATGL-1 overexpressing nematodes. ATGL-1 overexpression does not affect either the basal protein levels or the age-dependent accumulation of LMN-1 ( Figure S9E). Moreover, the nuclear morphology deterioration is not Consistent with accumulating evidence that caloric restriction and low insulin signaling extend lifespan across species, and delay the age-dependent abnormalities of nuclear morphology, we found that eat-2(ad465) and daf-2(e1370) mutant nematodes maintain the number and the size of nLDs in their intestinal cells during aging (Charar et al., 2021;Haithcock et al., 2005;Pathak et al., 2021;Perez-Jimenez et al., 2014). Intriguingly, the overall lipid content is shown to be elevated in the long-lived daf-2(e1370) mutant nematodes (Lapierre, Silvestrini, et al., 2013;O'Rourke et al., 2009). These findings suggest that nLDs homeostasis is differentially regulated uncoupling their levels from total lipid content. Investigating the effect of aging on nuclear envelope domains, we found that the levels of the inner nuclear membrane protein EMR-1 are not changed, whereas F I G U R E 4 HLH-30 regulates nLDs deposition in wt and long-lived mutants. (a, b) Quantification of LDs volume per nucleus (a) and average size (b) in the intestinal nuclei of 1-day-and 10-day-old wild-type, daf-2(e1370) and eat-2(ad465) transgenic nematodes expressing LMN-1::GFP. The animals were treated with empty vector (control), bec-1(RNAi), lgg-1(RNAi), and hlh-30(RNAi) (n = 15 animals for each genetic background and time point, ns p > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001, one-way ANOVA followed by Tukey HSD post hoc test).

F I G U R E 5
The triglyceride lipase ATGL-1 sustains nuclear lipid content in long-lived nematodes. (a, b) Quantification of LDs volume per nucleus (a) and average size (b) in the intestinal nuclei of wild-type, eat-2(ad465) and daf-2(e1370) nematodes expressing LMN-1::GFP.
Despite their role as transient storage depots, LDs could also supply histones for rapid chromatin remodeling (Li et al., , 2014. Additionally, nLDs have the potential to influence chromatin organization since they are formed at the envelope and, thereby, penetrating nuclear lamina to enter the nucleoplasm (Mosquera et al., 2021;Ohsaki et al., 2016;Romanauska & Kohler, 2018). A very recent study conducted in C. elegans reported that nLDs are coated by LMN-1 and/or heterochromatin, suggesting that nLDs accumulation could mediate the disposal of peripheral heterochromatin (Mosquera et al., 2021). Interestingly, loss of heterochromatin and derepression of several silenced genes are highly appreciated as major contributors to premature aging (Tsurumi & Li, 2012). Moreover, excessive accrual of nLDs could disrupt chromosome territories leading eventually to nuclear damage and cellular dysfunction. Indeed, the loss of the intestinal nuclei is a well-characterized feature of C. elegans aging that could be driven by the runaway accumulation of nLDs (Gems & Riddle, 2000;Mosquera et al., 2021;Poteryaev et al., 2005).
Genetic studies in C. elegans demonstrated that functional autophagy during early developmental stages is required for lipid droplet homeostasis (Lapierre, Silvestrini, et al., 2013). Indeed, BEC-1, LGG-1, UNC-51, and VSP-34 depleted nematodes display reduced general lipid content both in wild-type and long-lived daf-2(e1370) nematodes, highlighting the essential developmental role of autophagy in lipid storage and distribution (Lapierre, Silvestrini, et al., 2013). Although we knocked down several autophagy-related factors from early development, we did not observe any alteration in nuclear morphology, LMN-1::GFP abundance and nLDs number and size, suggesting that nLDs storage can be differentially regulated from cytosolic LDs. However, further investigation is needed to discriminate between the developmental and cytoprotective functions of autophagy, focusing on its impact in nLDs regulation.
Accumulating evidence suggests that HLH-30/TFEB is a crucial regulator of autophagy, lysosomal and lysosomal lipases gene expression. Our findings clearly demonstrate that HLH-30 is required for the reduced accrual of nLDs in long-lived nematodes. The lysosomal lipase LIPL-4 is highly appreciated as a longevity modulator (Folick et al., 2015;Ramachandran et al., 2019;Savini et al., 2022;Wang et al., 2008). LIPL-4 is highly expressed in long-lived mutant animals and its depletion abolishes the lifespan extension indicating its cytoprotective function (Wang et al., 2008). Even though LIPL-4 deficiency does not influence the lifespan of wild-type nematodes, its overexpression is sufficient to promote longevity (Folick et al., 2015). Although LIPL-4 serves as a lipolysis mediator, its depletion abolishes nLDs accumulation with age in wild-type animals. ATGL-1 overexpression is sufficient to sustain nLDs abundance during aging and promote lifespan extension in wild-type nematodes (Zaarur et al., 2019). In addition to nematode data, ATGL (the human homolog of ATGL-1) is detected both in LDs and the nucleoplasm of multiple human cell lines (www.prote inatl as.org), highlighting its possible function as nLDs regulator across species (Thul et al., 2017;Uhlén et al., 2015). Notably, ATGL was shown to interact directly with the autophagosomal protein LC3, suggesting an interplay between lipolysis and lipophagy (Martinez-Lopez et al., 2016). However, LGG-1, LGG-2, and BEC-1 deficiency do not display any effect in nLDs number and size both in aged wild-type and long-lived mutants, indicating that nuclear fat deposition is not regulated by autophagy. Accumulating evidence showed that nuclear export inhibition improves organismal proteostasis and longevity by promoting HLH-30/TFEB nuclearization and autophagy induction in nematodes, flies, and mammalian cells (Silvestrini et al., 2018). Intriguingly, the autophagy protein LGG-1 modulates nucleolar size and ribosomal function through the degradation of key components in nucleolar dynamic and protein translation, such as FIB-1 and RPL-11 to promote longevity upon nuclear export inhibition (Kumar et al., 2022;Tiku et al., 2018). These findings further support the notion that autophagy, apart from being a simple degradation pathway, it has a broader role in lipid remodeling and nuclear homeostasis (Lapierre, Silvestrini, et al., 2013).
The cumulative results of the current study suggest that the reduced levels and the impaired activity of HLH-30 results in elevated LMN-1 abundance and diminished ATLG-1 expression, which subsequently trigger excessive nuclear lipids accumulation ( Figure 6). Furthermore, our findings underscore the dynamic interplay between autophagy, lipolytic enzymes, and lipases, which is pivotal for the maintenance of nuclear envelope integrity, nucleolus morphology and nLDs content to promote nuclear homeostasis and prevent age-dependent cellular impairment and physiology decline. C. elegans is a versatile genetic model that can be used as a screening platform to unravel novel tissue-specific modulators of nLDs distribution. Investigating further the molecular pathways that regulate nLDs formation and accrual will enlighten new avenues for therapeutic intervention strategies to tackle metabolic and ageassociated diseases.

| Non-linear microscopy setup
The experimental setup was similar to the one described in our previous studies (Palikaras et al., 2017;Tserevelakis et al., 2014).
Two different femtosecond (

| Imaging and statistical analysis
All the samples have been imaged under constant irradiation con-

| Caenorhabditis elegans strains and culture methods
We followed standard procedures for C. elegans strain maintenance (Stiernagle, 2006

| Molecular cloning
For generating the ATGL-1 O/E vector, we amplified the complete atgl-1 genomic sequence, including atgl-1 promoter region (approximately 1.5 kb upstream from the start codon) and its endogenous 3'-UTR (approximately 400 bp downstream from the stop codon). The resulting fragment was ligated to pCRII-TOPO and was used directly for the generation of transgenic animals.
For engineering, the atgl-1, lgg-2, vit-2, and lipl-4 RNAi constructs, gene-specific fragments of interest were obtained by PCR amplification directly from C. elegans genomic DNA. The PCR-generated fragments were subcloned into the pL4440 plasmid vector. The resulting constructs were transformed into HT115(DE3) Escherichia coli bacteria deficient for RNase III. Bacteria carrying an empty vector were used in control experiments. The primers used for each construct are provided in Table S1. In our study we have also used RNAi vectors against bec-1, lgg-1, hlh-30, and vps-34, which had been previously generated in our lab (Samara et al., 2008). The entire list of the primers used for these genes are also included in Table S1.

| RNA isolation and qRT-PCR analysis
Total RNA from synchronized day-1 adult animals was extracted by using the TRIzol reagent (Invitrogen). cDNA synthesis was performed by using the iScript™ cDNA Synthesis Kit (Bio-Rad). Quantitative Real-Time PCR (qRT-PCR) was performed in a Bio-Rad CFX96 Real-

| Nuclear fractionation
For nuclear fractionation, we followed a previously described were centrifuged for 5 min at 500 g at 4°C for the precipitation of worm debris. The supernatant was transferred to a new tube and centrifuged again for 5 min at 500 g, 4°C. Similarly, the resulting supernatant was transferred to a new tube and 30 μl was kept as the input fraction. The nuclei were subsequently pelletized by centrifuging the supernatant at 4000 g for 5 min at 4°C. The supernatants were subjected to another round of centrifugation at 17000 g for 30 min at 4°C, to produce the cytoplasmic fraction.
The pellets were washed twice with hypotonic buffer and centrifuged at 4000 g for 5 min at 4°C. Nuclear pellets were finally resuspended in 50 μl hypertonic buffer (15 mM HEPES KOH pH 7.6, 400 mM KCl, 5 mM MgCl 2 , 0.1 mM EDTA, 0.1% Tween-20, 10% Glycerol, 1 mM DTT, 2× protease inhibitor cocktails) and transferred to a new tube. After snap freezing, all samples were stored at −80°C for further analysis by Western blotting.

| BODIPY staining
L4 transgenic larvae expressing p app-1 H2B::mCherry transgene were placed on nematode growth media Escherichia coli (OP50) plates seeded with 100 μl of 5 μM C1BODIPY-C12 (Invitrogen, D3823) diluted in M9 buffer. 1-day-old and 5-day-old stained nemaotdes were washed twice with M9 buffer. Then, the animals were immobilized with 10 mM levamisole/M9 buffer before mounting on 2% agarose pads for microscopic examination with a KEYENCE BZ-X800 epifluorescence microscope. The number of nLDs were calculated for each intestinal nuclei in these images using the ImageJ software (http://rsb.info. nih.gov/ij/). In each experiment, at least 20 animals and intestinal 50 nuclei were examined for each strain/condition.
Each assay was repeated at least three times. We used the Prism software package (GraphPad Software) for statistical analyses.
Finally, the membrane was developed by chemiluminescence (Supersignal chemiluminescent substrate pico and femto, Thermo Fisher Scientific).

AUTH O R CO NTR I B UTI O N S
K.P. and M.M. conceived and conceptualized the project. K.P., M.M., and C.P. wrote the paper. K.P., M.M., C.P., and A.P. performed and analyzed the experiments. All authors edited the paper. K.P., G.F., and N.T. supervised the project.

ACK N OWLED G M ENTS
Some nematode strains used in this work were provided by the Caenorhabditis Genetics Center, which is funded by the National Center for Research Resources of the National Institutes of Health, and S. Mitani (National Bioresource Project) in Japan. We thank A.
Fire for plasmid vectors.

CO N FLI C T O F I NTE R E S T
The authors declare that there is no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are included in the main and supplementary figures that accompany the manuscript.
The data are also available from the corresponding author upon request.