Lifespan is governed by complex interactions between genetic and environmental factors. The perturbation of insulin/insulin-like signaling, target of rapamycin (TOR) pathway, and mitochondrial functions have been shown to extensively modulate lifespan and health (Fontana et al., 2010; Kenyon, 2010; López-Otín et al., 2013). These genetic manipulations often lead to significant changes in gene expression at both the transcriptional and translational levels. Inhibition of DAF-2, the Caenorhabditis elegans ortholog of the insulin growth factor 1 (IGF-1) receptor, doubles adult lifespan by activating the DAF-16 (FOXO) transcription factor to regulate downstream genes involved in stress resistance, detoxification, and metabolism (Kenyon et al., 1993; Kimura et al., 1997; Lin et al., 1997; Ogg et al., 1997; Murphy et al., 2003; McElwee et al., 2004).

In addition to genetic regulation, aging is also modulated by epigenetic processes. Epigenetic marks, including histone acetylation and methylation, as well as the associated chromatin states, are altered during aging. Aging-dependent loss of chromatin repression is correlated with alterations in gene expression patterns, which have been documented in species ranging from C. elegans to humans (Lee et al., 2000; Lund et al., 2002; Bennett-Baker et al., 2003; Lu et al., 2004). For example, the methylation of histone 3 lysine 4 methylation (H3K4me) is one of the posttranslational histone modifications that marks on regions of active transcription (Shilatifard, 2008). H3K4me is deposited by the MLL/COMPASS complex, which travels with elongating RNA Polymerase II during transcription (Wood et al., 2007). In C. elegans, animals with reductions in COMPASS complex subunits (wdr-5, ash-2, and set-2) live longer than wild-type individuals (Greer et al., 2010).

However, little is known about the importance of repressive histone modification for aging marks. The H3K27me3 demethylase UTX-1 regulates lifespan independently of the presence of the germline but in a manner that depends on the insulin-FOXO signaling pathway (Maures et al., 2011). MET-2, a mammalian H3K9 methyltransferase SETDB1 homolog (Loyola et al., 2006), likely monomethylates and dimethylates H3K9 in C. elegans (Towbin et al., 2012). MET-2 is necessary both for a normal lifespan (Tian et al., 2016) and for the lifespan extension of wdr-5 mutants (Lee et al., 2019). SET-6 is a putative H3K9me2/3 methyltransferase but not H3K9me1. Although the depletion of SET-6 does not significantly increase the lifespan of C. elegans, it may increase healthy aging (Yuan et al., 2020). SET-25 is a putative trimethylase for H3K9 (Towbin et al., 2012; Padeken et al., 2021). However, the deletion of SET-25 does not significantly change the worm lifespan (Woodhouse et al., 2018).

The enzymes responsible for histone lysine methylation are called histone methyltransferases (HMTs). HMTs typically contain a conserved catalytic domain called SET, which stems from Su(var)3–9, Enhancer of zeste, and Trithorax, the first HMTs known to carry this domain (Tschiersch et al., 1994). The C. elegans genome encodes 38 SET domain-containing proteins, of which 5 are essential for viability (Andersen and Horvitz, 2007; Ni et al., 2012). However, the biological roles of most SET proteins are largely unknown.

To investigate the function of repressive histone methylation in lifespan regulation, we selected a number of putative H3K9 methyltransferases as well as methyltransferases that may indirectly regulate H3K9me and tested whether the loss-of-function of these SET proteins could change the lifespan of C. elegans. Interestingly, we found that the H3K9me1/2, but not H3K9me3, mutants exhibited a synergistic lifespan extension with daf-2 mutation. These animals showed an average lifespan of approximately 60 days, which is approximately 70% longer than that of daf-2 worms and is three times as long as that of wild-type N2 animals. The double mutants exhibited a maximal lifespan of approximately 100 days. The synergistic lifespan extension of DAF-2 and lifespan-limiting HMT mutants depend on the transcription factor DAF-16 (FOXO). mRNA-seq experiments showed that DAF-16 Class I genes (genes that are activated by DAF-16) are activated in long-lived worms. Therefore, we conclude that H3K9me1/2 may limit the lifespan of daf-2 animals by repressing the expression of DAF-16 Class I genes.

Here, we identified a class of putative HMTs, as well as methyltransferases that have been suggested to indirectly regulate H3K9me, that limit the lifespan of daf-2 mutant in C. elegans. These putative HMTs are required for H3K9me1/2 modification and regulate DAF-16 Class I genes in daf-2 animals. In the absence of DAF-2 and H3K9me1/2, the DAF-16 accumulates in the nucleus and the expression of DAF-16 Class I genes increase, which promote the expression of longevity genes and anti-stress genes and extends lifespan of daf-2 animals (Figure 8—figure supplement 4D). Decreased H3K9me1/2 in daf-2 animals improved nematode tolerance to oxidative stress and heat stress. Notably, the inhibition of worm H3K9me1/2 methyltransferases by a human G9a inhibitor also extends the lifespan of daf-2 animals. Thus, targeting insulin pathway and H3K9me1/2 modification together may be a new means for the treatment of aging and aging-related diseases.

The depletion of SET proteins or the inhibition of worm H3K9me1/2 methyltransferases by G9a inhibitor significantly extended the lifespan only in daf-2 mutant worms, but has modest effects on N2 animals. Consistently, the long-lived set-6, set-19, set-20, set-21, set-32, and set-33 mutants only significantly decreased global H3K9me1/2 levels in daf-2 mutants but not in N2 animals. These results together suggested that DAF-2 limits the effects of these SET proteins.

It is unclear how and why these SET proteins only limit lifespan in daf-2 animals. There are three possibilities. First, DAF-2 may have profound effect on the chromatin accessibility. In daf-2 mutants, the chromatin may be more compact and prohibit the activity of the SET proteins to conduct H3K9me2 modification. Second, daf-2 mutation may sensitize the animals by increasing the nuclear accumulated DAF-16. Although it is unclear whether DAF-16 preferentially recognizes H3K9me2 or H3K9me3, we observed a much stronger enrichment of DAF-16 in the nucleus in daf-2;set-21 than in daf-2 animals, which suggested that H3K9me2 may prohibit the accumulation of DAF-16 in the nucleus. Third, DAF-2 may have other indirect effect on the SET proteins. Further ChIP experiments can be used to test whether daf-2 mutation could affect the association of the SET proteins with chromatin. The ATAC-seq technology may allow us to investigate whether daf-2 mutation have changed chromosome accessibility to facilitate the activity of SET proteins. In addition, identification of the substrate of SET proteins may help to elucidate their mechanism and functions.

In daf-2 animals, DAF-16 modestly accumulated in the nucleus, as shown by the newly generated single-copy GFP-tagged DAF-16 transgene. Yet in daf-2;set-21 double mutants, DAF-16 was strongly enriched in the nucleus. The depletion of DAF-16 completely blocked the synergistic lifespan extension in daf-2;set-21 animals. The mRNA-seq experiments revealed that the mRNA levels of DAF-16 Class I genes were elevated in the long-lived daf-2;set-19, daf-2;set-21, and daf-2;set-32 double mutants compared to those in control daf-2 and short-lived daf-2;set-25 mutants. Together, these data suggested that the activity to DAF-16 to Class I genes not only depends on the absence of insulin pathway, but also depends on the depletion of these SET proteins and H3K9me2. The single-copy GFP-tagged DAF-16 transgene may allow us to re-visit the mechanism of lifespan regulation pathways.

A trade-off has been proposed between fertility and lifespan, where increasing lifespan usually comes at the cost of reduced fertility (Kirkwood, 1977). In C. elegans, removing the germ cells extends lifespan, triggers the nuclear localization of the DAF-16/FOXO transcription factor, and activates target genes in the intestine (Berman and Kenyon, 2006). Some long-lived mutant worms, like daf-2 and rsks-1, show reduced brood size (Chen et al., 2013). Consistently, we found that the long-lived worms daf-2;met-2, daf-2;set-6, daf-2;set-19, daf-2;set-20, daf-2;set-21, daf-2;set-32, and daf-2;set-33 all have reduced brood size compared to that of daf-2 animals. However, additional investigations are necessary to examine whether removing the germline cells could further extend worm lifespan beyond the 100 days’ ceiling.

Lifespan is controlled by both genetic and epigenetic factors. The perturbation of insulin/insulin-like signaling, TOR pathway, and mitochondrial functions have been shown to extensively modulate the aging process and lifespan. Epigenetic marks, including histone acetylation and methylation, are altered during aging and regulate lifespan in a number of species (Lee et al., 2000; Lund et al., 2002; Bennett-Baker et al., 2003; Lu et al., 2004). In C. elegans, both H3K4 and H3K27 methylation are involved in lifespan limitation. Animals with reduced H3K4me, for example, in wdr-5, ash-2, and set-2 mutants, live longer than wild-type animals (Greer et al., 2010). Interestingly, perturbation of H3K27 methylation, via either a decrease in mes-2 mutants or an increase in utx-1 mutants, is associated with enhanced longevity, suggesting context-dependent lifespan regulation (Maures et al., 2011; Guillermo et al., 2021).

H3K9me3 is a hallmark of heterochromatin and is often enriched on repetitive sequences (Peters et al., 2003; Guenatri et al., 2004; Martens et al., 2005). In addition, H3K9me1/2/3 have all been shown enrichment on repeat sequences (Gerstein et al., 2010). H3K9 methylation and its association with the nuclear envelope help maintain silencing of repetitive sequences and may contribute to the maintenance and stability of heterochromatin-like regions in C. elegans (Liu et al., 2011), as observed in other organisms (Peng and Karpen, 2008). A hallmark of heterochromatin is the enrichment of HP1 (James and Elgin, 1986). The nematode C. elegans has two HP1 paralogous proteins: HPL-1 and HPL-2 (Couteau et al., 2002). Interestingly, HPL-2 may serve additional roles than HPL-1, as hpl-2 mutants display diverse defects while hpl-1 mutants generally lack obvious defects (Garrigues et al., 2015). The pattern of HPL-2 most closely resembles H3K9me1 and H3K9me2 but less closely resembles H3K9me3 (Garrigues et al., 2015). HPL-2-bound genes, especially those located on arms, possess significantly higher levels of H3K9me2, but show no significant difference in H3K9me3 levels when compared to genes not bound by HPL-2 (Garrigues et al., 2015). SET-25 is a putative H3K9me3 methyltransferase and is recruited to targeted genomic loci either by H3K9me2 which is deposited by MET-2 or by the somatic Argonaute protein NRDE-3 and small RNAs (Mao et al., 2015; Padeken et al., 2021). SET-25 colocalizes with both H3K9me3 and HPL-1, but does not colocalize with HPL-2 (Towbin et al., 2012).

H3K9me2 and H3K9me3 exhibit different patterns of repeat enrichment (McMurchy et al., 2017). H3K9me2 is more associated with DNA transposons and satellite repeats, similar to the heterochromatin factors HPL-2, LET-418, LIN-13, LIN-61, and MET-2, whereas H3K9me3 is particularly associated with retrotransposon families, especially LINE and SINE elements (McMurchy et al., 2017). Moreover, H3K9me2 and the five heterochromatin factors are enriched at telomeres, whereas H3K9me3 is not (McMurchy et al., 2017). Therefore, H3K9me2, rather than H3K9me3, may be more closely associated with canonical heterochromatin factors in C. elegans. It was postulated that losing repressive chromatin is detrimental to lifespan (Villeponteau, 1997; Tsurumi and Li, 2012). In humans, two premature aging diseases are caused by mutations in lamins that reduce heterochromatin and disrupt its nuclear localization (Shumaker et al., 2006). In C. elegans, Drosophila, and mammals, heterochromatin decreases as individuals grow older (Haithcock et al., 2005; Wood et al., 2010; Wood and Helfand, 2013). Additionally, mutations that increase repressive chromatin extend lifespan in eukaryotes (Kennedy et al., 1995; Jin et al., 2011; Maures et al., 2011; Larson et al., 2012). In N2 background animals, the depletion of hpl-2 was shown increased lifespan (Meister et al., 2011). However, we did not observe a further lifespan extension in daf-2;hpl-2 animals, comparing to the daf-2 mutant. Similarly, we did not observe a significant change in lifespan in daf-2;set-25 mutants. The depletion of the two heterochromatin factors HPL-1 and HPL-2 failed to further extend the lifespan in daf-2 mutant worms, suggesting that heterochromatin itself may not be sufficient to perturb lifespan in daf-2 mutant worms.

MET-2, a mammalian H3K9 methyltransferase SETDB1 homolog (Loyola et al., 2006), likely monomethylates and dimethylates H3K9 in C. elegans (Towbin et al., 2012). MET-2 is necessary both for a normal lifespan in N2 background (Tian et al., 2016) and for the lifespan extension of wdr-5 mutants (Lee et al., 2019). In met-2 mutants, the lifespan is shortened compared to that of N2 animals. However, the lifespan was significantly extended in daf-2;met-2 double mutants, suggesting context-dependent lifespan regulation by epigenetic machinery. SET-6 is a putative H3K9me2/3 methyltransferase that was postulated to prevent healthy aging rather than modulate the normal lifespan in N2 background (Yuan et al., 2020). SET-6 accelerates behavioral deterioration in C. elegans by reducing mitochondrial function and repressing the expression of nuclear-encoded mitochondrial proteins. Here, we found that the depletion of set-6 dramatically increased the lifespan of daf-2 animals, which further supports the context-dependent lifespan regulation model.

We noticed that the stress resistance is enhanced in both met-2 and daf-2;met-2 animals, yet met-2 single mutant shortens lifespan and daf-2;met-2 extends lifespan. In addition, daf-2;set-15 showed a modest increase of stress resistance, while daf-2;set-15 revealed no further lifespan extension. The reason is unknown yet. Previous research showed correlations between stress resistance and lifespan extension in many mutant worms, for example daf-2. However, since daf-2 mutation already twitched the homeostasis and increased the stress resistance of animals, it is unclear that to what extent this correlation can further increase lifespan and stress resistance. Alternatively, some of these putative lifespan-limiting HMTs may have additional functions. Further investigations are required to understand the mechanism of each putative lifespan-limiting HMT.

In this work, we examined the role of 10 putative H3K9 methyltransferases, most of which have not been directly characterized for their biochemical properties by in vitro experiments (Schwartz-Orbach et al., 2020). The presence of these many putative H3K9 methyltransferases, or methyltransferases that may indirectly regulate H3K9me, may lead to a complexity of the interaction network of histone modifications. In addition, some SET proteins may have multiple roles in mediating histone modification. For example, SET-19 is required for both H3K9me1/2 and H3K23me3 marks, further bringing up complexity of the regulation. Further investigations are necessary to reveal their enzymatic activities by in vitro experiments (Schwartz-Orbach et al., 2020). In addition, identifying the in vivo targeted genes will also be crucial to help understanding why and how these SET proteins are involved in lifespan regulation in daf-2 mutants.

Additionally, we identified seven genes required for H3K9me1/2-involved lifespan extension in daf-2;set-21 animals. Among them, tts-1 is a long noncoding RNA on ribosomes of the daf-2 mutant. Depleting tts-1 in daf-2 mutants increases ribosome levels (Tepper et al., 2013; Essers et al., 2015). NHR-62 is known to be required for longevity in C. elegans (Heestand et al., 2013). INS-35 is an insulin-like peptides and may act on DAF-2 to regulate UPR^mt and mitochondrial dynamics (Chen et al., 2021). SOD-3 is a superoxide dismutase that is involved in the removal of superoxide radicals. ASM-2 is an ortholog of human acid sphingomyelinase (ASM) and is involved in ceramide biosynthetic processes and sphingomyelin catabolic processes (Lin et al., 1998; Deng et al., 2008). The C. elegans genome encodes three ASM homologs, asm-1, asm-2, and asm-3. Among them, ASM-3 is most closely related to human ASM. During development and aging, ceramide and sphingosine accumulate (Sacket et al., 2009). Interestingly, monounsaturated fatty acid (MUFA) accumulation is necessary for the lifespan extension of H3K4me3-methyltransferase-deficient worms, and dietary MUFAs are sufficient to extend lifespan (Han et al., 2017). However, when asm-1, asm-2, or asm-3 was inactivated by RNAi knockdown, a modest lifespan extension phenotype was observed (Kim and Sun, 2012). Further investigation is required to examine how and why these genes are involved in lifespan regulation in daf-2;set-21 animals.

Stichodactyla toxin (ShK, ShkT) is a 35-residue basic peptide from the sea anemone Stichodactyla helianthus that blocks a number of potassium channels with nanomolar to picomolar potency. We found that F35E8.7 and Y39G8B.7 are required for lifespan extension in daf-2;set-21 animals. Human proteins containing ShK-like domains are MMP-23 (matrix metalloprotease 23) and MFAP-2 (microfibril-associated glycoprotein 2). Whether modulating the expression of these two proteins or their downstream targets can improve longevity or healthy lifespan will stimulate much drug potential.

Sequencing data have been deposited in GSA under accession codes CRA005256. Figure 2 - Source Data 1 and Figure 8 - Source Data 2 contain the original files and figures of the blots used to generate the figures. Source Code Files 1 is used to convert BAM files to BigWig format in ChIP-seq data analysis.

1. 1. Guang S
   2. Huang M
   3. Hong M

   (2021) GSA

   ID CRA005256. H3K9me1/2 methylation limits the lifespan of C. elegans.

   https://download.cncb.ac.cn/gsa/CRA005256

1. 1. Andersen EC
   2. Horvitz HR

   (2007) Two C. elegans histone methyltransferases repress LIN-3 EGF transcription to inhibit vulval development

   Development 134:2991–2999.

   https://doi.org/10.1242/dev.009373

   • PubMed
   • Google Scholar
2. 1. Ayyadevara S
   2. Alla R
   3. Thaden JJ
   4. Shmookler Reis RJ

   (2008) Remarkable longevity and stress resistance of nematode PI3K-null mutants

   Aging Cell 7:13–22.

   https://doi.org/10.1111/j.1474-9726.2007.00348.x

   • PubMed
   • Google Scholar
3. 1. Bennett-Baker PE
   2. Wilkowski J
   3. Burke DT

   (2003) Age-Associated activation of epigenetically repressed genes in the mouse

   Genetics 165:2055–2062.

   https://doi.org/10.1093/genetics/165.4.2055

   • PubMed
   • Google Scholar
4. 1. Berman JR
   2. Kenyon C

   (2006) Germ-Cell loss extends C. elegans life span through regulation of DAF-16 by kri-1 and lipophilic-hormone signaling

   Cell 124:1055–1068.

   https://doi.org/10.1016/j.cell.2006.01.039

   • PubMed
   • Google Scholar
5. 1. Chen D
   2. Li PW-L
   3. Goldstein BA
   4. Cai W
   5. Thomas EL
   6. Chen F
   7. Hubbard AE
   8. Melov S
   9. Kapahi P

   (2013) Germline signaling mediates the synergistically prolonged longevity produced by double mutations in DAF-2 and rsks-1 in C. elegans

   Cell Reports 5:1600–1610.

   https://doi.org/10.1016/j.celrep.2013.11.018

   • PubMed
   • Google Scholar
6. 1. Chen X
   2. Xu F
   3. Zhu C
   4. Ji J
   5. Zhou X
   6. Feng X
   7. Guang S

   (2014) Dual sgrna-directed gene knockout using CRISPR/Cas9 technology in Caenorhabditis elegans

   Scientific Reports 4:7581.

   https://doi.org/10.1038/srep07581

   • PubMed
   • Google Scholar
7. 1. Chen L-T
   2. Lin C-T
   3. Lin L-Y
   4. Hsu J-M
   5. Wu Y-C
   6. Pan C-L

   (2021) Neuronal mitochondrial dynamics coordinate systemic mitochondrial morphology and stress response to confer pathogen resistance in C. elegans

   Developmental Cell 56:1770–1785.

   https://doi.org/10.1016/j.devcel.2021.04.021

   • PubMed
   • Google Scholar
8. 1. Couteau F
   2. Guerry F
   3. Muller F
   4. Palladino F

   (2002) A heterochromatin protein 1 homologue in Caenorhabditis elegans acts in germline and vulval development

   EMBO Reports 3:235–241.

   https://doi.org/10.1093/embo-reports/kvf051

   • PubMed
   • Google Scholar
9. 1. Deng X
   2. Yin X
   3. Allan R
   4. Lu DD
   5. Maurer CW
   6. Haimovitz-Friedman A
   7. Fuks Z
   8. Shaham S
   9. Kolesnick R

   (2008) Ceramide biogenesis is required for radiation-induced apoptosis in the germ line of C. elegans

   Science 322:110–115.

   https://doi.org/10.1126/science.1158111

   • PubMed
   • Google Scholar
10. 1. Dues DJ
    2. Schaar CE
    3. Johnson BK
    4. Bowman MJ
    5. Winn ME
    6. Senchuk MM
    7. Van Raamsdonk JM

    (2017) Uncoupling of oxidative stress resistance and lifespan in long-lived ISP-1 mitochondrial mutants in Caenorhabditis elegans

    Free Radical Biology & Medicine 108:362–373.

    https://doi.org/10.1016/j.freeradbiomed.2017.04.004

    • PubMed
    • Google Scholar
11. 1. Essers PB
    2. Nonnekens J
    3. Goos YJ
    4. Betist MC
    5. Viester MD
    6. Mossink B
    7. Lansu N
    8. Korswagen HC
    9. Jelier R
    10. Brenkman AB
    11. MacInnes AW

    (2015) A long noncoding RNA on the ribosome is required for lifespan extension

    Cell Reports 10:339–345.

    https://doi.org/10.1016/j.celrep.2014.12.029

    • PubMed
    • Google Scholar
12. 1. Fontana L
    2. Partridge L
    3. Longo VD

    (2010) Extending healthy life span -- from yeast to humans

    Science 328:321–326.

    https://doi.org/10.1126/science.1172539

    • PubMed
    • Google Scholar
13. 1. Frøkjaer-Jensen C
    2. Davis MW
    3. Hopkins CE
    4. Newman BJ
    5. Thummel JM
    6. Olesen S-P
    7. Grunnet M
    8. Jorgensen EM

    (2008) Single-Copy insertion of transgenes in Caenorhabditis elegans

    Nature Genetics 40:1375–1383.

    https://doi.org/10.1038/ng.248

    • PubMed
    • Google Scholar
14. 1. Garrigues JM
    2. Sidoli S
    3. Garcia BA
    4. Strome S

    (2015) Defining heterochromatin in C. elegans through genome-wide analysis of the heterochromatin protein 1 homolog HPL-2

    Genome Research 25:76–88.

    https://doi.org/10.1101/gr.180489.114

    • PubMed
    • Google Scholar
15. 1. Gerstein MB
    2. Lu ZJ
    3. Van Nostrand EL
    4. Cheng C
    5. Arshinoff BI
    6. Liu T
    7. Yip KY
    8. Robilotto R
    9. Rechtsteiner A
    10. Ikegami K
    11. Alves P
    12. Chateigner A
    13. Perry M
    14. Morris M
    15. Auerbach RK
    16. Feng X
    17. Leng J
    18. Vielle A
    19. Niu W
    20. Rhrissorrakrai K
    21. Agarwal A
    22. Alexander RP
    23. Barber G
    24. Brdlik CM
    25. Brennan J
    26. Brouillet JJ
    27. Carr A
    28. Cheung MS
    29. Clawson H
    30. Contrino S
    31. Dannenberg LO
    32. Dernburg AF
    33. Desai A
    34. Dick L
    35. Dosé AC
    36. Du J
    37. Egelhofer T
    38. Ercan S
    39. Euskirchen G
    40. Ewing B
    41. Feingold EA
    42. Gassmann R
    43. Good PJ
    44. Green P
    45. Gullier F
    46. Gutwein M
    47. Guyer MS
    48. Habegger L
    49. Han T
    50. Henikoff JG
    51. Henz SR
    52. Hinrichs A
    53. Holster H
    54. Hyman T
    55. Iniguez AL
    56. Janette J
    57. Jensen M
    58. Kato M
    59. Kent WJ
    60. Kephart E
    61. Khivansara V
    62. Khurana E
    63. Kim JK
    64. Kolasinska-Zwierz P
    65. Lai EC
    66. Latorre I
    67. Leahey A
    68. Lewis S
    69. Lloyd P
    70. Lochovsky L
    71. Lowdon RF
    72. Lubling Y
    73. Lyne R
    74. MacCoss M
    75. Mackowiak SD
    76. Mangone M
    77. McKay S
    78. Mecenas D
    79. Merrihew G
    80. Miller DM
    81. Muroyama A
    82. Murray JI
    83. Ooi SL
    84. Pham H
    85. Phippen T
    86. Preston EA
    87. Rajewsky N
    88. Rätsch G
    89. Rosenbaum H
    90. Rozowsky J
    91. Rutherford K
    92. Ruzanov P
    93. Sarov M
    94. Sasidharan R
    95. Sboner A
    96. Scheid P
    97. Segal E
    98. Shin H
    99. Shou C
    100. Slack FJ
    101. Slightam C
    102. Smith R
    103. Spencer WC
    104. Stinson EO
    105. Taing S
    106. Takasaki T
    107. Vafeados D
    108. Voronina K
    109. Wang G
    110. Washington NL
    111. Whittle CM
    112. Wu B
    113. Yan KK
    114. Zeller G
    115. Zha Z
    116. Zhong M
    117. Zhou X
    118. Ahringer J
    119. Strome S
    120. Gunsalus KC
    121. Micklem G
    122. Liu XS
    123. Reinke V
    124. Kim SK
    125. Hillier LW
    126. Henikoff S
    127. Piano F
    128. Snyder M
    129. Stein L
    130. Lieb JD
    131. Waterston RH
    132. modENCODE Consortium

    (2010) Integrative analysis of the Caenorhabditis elegans genome by the modencode project

    Science 330:1775–1787.

    https://doi.org/10.1126/science.1196914

    • PubMed
    • Google Scholar
16. 1. Goudeau J
    2. Bellemin S
    3. Toselli-Mollereau E
    4. Shamalnasab M
    5. Chen Y
    6. Aguilaniu H

    (2011) Fatty acid desaturation links germ cell loss to longevity through NHR-80/HNF4 in C. elegans

    PLOS Biology 9:e1000599.

    https://doi.org/10.1371/journal.pbio.1000599

    • PubMed
    • Google Scholar
17. 1. Greer EL
    2. Maures TJ
    3. Hauswirth AG
    4. Green EM
    5. Leeman DS
    6. Maro GS
    7. Han S
    8. Banko MR
    9. Gozani O
    10. Brunet A

    (2010) Members of the H3K4 trimethylation complex regulate lifespan in a germline-dependent manner in C. elegans

    Nature 466:383–387.

    https://doi.org/10.1038/nature09195

    • PubMed
    • Google Scholar
18. 1. Guenatri M
    2. Bailly D
    3. Maison C
    4. Almouzni G

    (2004) Mouse centric and pericentric satellite repeats form distinct functional heterochromatin

    The Journal of Cell Biology 166:493–505.

    https://doi.org/10.1083/jcb.200403109

    • PubMed
    • Google Scholar
19. 1. Guillermo ARR
    2. Chocian K
    3. Gavriilidis G
    4. Vandamme J
    5. Salcini AE
    6. Mellor J
    7. Woollard A

    (2021) H3K27 modifiers regulate lifespan in C. elegans in a context-dependent manner

    BMC Biology 19:59.

    https://doi.org/10.1186/s12915-021-00984-8

    • PubMed
    • Google Scholar
20. 1. Haithcock E
    2. Dayani Y
    3. Neufeld E
    4. Zahand AJ
    5. Feinstein N
    6. Mattout A
    7. Gruenbaum Y
    8. Liu J

    (2005) Age-Related changes of nuclear architecture in Caenorhabditis elegans

    PNAS 102:16690–16695.

    https://doi.org/10.1073/pnas.0506955102

    • PubMed
    • Google Scholar
21. 1. Han S
    2. Schroeder EA
    3. Silva-García CG
    4. Hebestreit K
    5. Mair WB
    6. Brunet A

    (2017) Mono-Unsaturated fatty acids link H3K4me3 modifiers to C. elegans lifespan

    Nature 544:185–190.

    https://doi.org/10.1038/nature21686

    • PubMed
    • Google Scholar
22. 1. Hansen M
    2. Taubert S
    3. Crawford D
    4. Libina N
    5. Lee S-J
    6. Kenyon C

    (2007) Lifespan extension by conditions that inhibit translation in Caenorhabditis elegans

    Aging Cell 6:95–110.

    https://doi.org/10.1111/j.1474-9726.2006.00267.x

    • PubMed
    • Google Scholar
23. 1. Heestand BN
    2. Shen Y
    3. Liu W
    4. Magner DB
    5. Storm N
    6. Meharg C
    7. Habermann B
    8. Antebi A

    (2013) Dietary restriction induced longevity is mediated by nuclear receptor NHR-62 in Caenorhabditis elegans

    PLOS Genetics 9:e1003651.

    https://doi.org/10.1371/journal.pgen.1003651

    • PubMed
    • Google Scholar
24. 1. Henderson ST
    2. Johnson TE

    (2001) Daf-16 integrates developmental and environmental inputs to mediate aging in the nematode Caenorhabditis elegans

    Current Biology 11:1975–1980.

    https://doi.org/10.1016/s0960-9822(01)00594-2

    • PubMed
    • Google Scholar
25. 1. James TC
    2. Elgin SC

    (1986) Identification of a nonhistone chromosomal protein associated with heterochromatin in Drosophila melanogaster and its gene

    Molecular and Cellular Biology 6:3862–3872.

    https://doi.org/10.1128/mcb.6.11.3862-3872.1986

    • PubMed
    • Google Scholar
26. 1. Jin C
    2. Li J
    3. Green CD
    4. Yu X
    5. Tang X
    6. Han D
    7. Xian B
    8. Wang D
    9. Huang X
    10. Cao X
    11. Yan Z
    12. Hou L
    13. Liu J
    14. Shukeir N
    15. Khaitovich P
    16. Chen CD
    17. Zhang H
    18. Jenuwein T
    19. Han J-DJ

    (2011) Histone demethylase UTX-1 regulates C. elegans life span by targeting the insulin/IGF-1 signaling pathway

    Cell Metabolism 14:161–172.

    https://doi.org/10.1016/j.cmet.2011.07.001

    • PubMed
    • Google Scholar
27. 1. Kalinava N
    2. Ni JZ
    3. Peterman K
    4. Chen E
    5. Gu SG

    (2017) Decoupling the downstream effects of germline nuclear RNAi reveals that H3K9me3 is dispensable for heritable RNAi and the maintenance of endogenous siRNA-mediated transcriptional silencing in Caenorhabditis elegans

    Epigenetics & Chromatin 10:6.

    https://doi.org/10.1186/s13072-017-0114-8

    • PubMed
    • Google Scholar
28. 1. Kennedy BK
    2. Austriaco NR Jr
    3. Zhang J
    4. Guarente L

    (1995) Mutation in the silencing gene SIR4 can delay aging in S. cerevisiae

    Cell 80:485–496.

    https://doi.org/10.1016/0092-8674(95)90499-9

    • PubMed
    • Google Scholar
29. 1. Kenyon C
    2. Chang J
    3. Gensch E
    4. Rudner A
    5. Tabtiang R

    (1993) A C. elegans mutant that lives twice as long as wild type

    Nature 366:461–464.

    https://doi.org/10.1038/366461a0

    • PubMed
    • Google Scholar
30. 1. Kenyon CJ

    (2010) The genetics of ageing

    Nature 464:504–512.

    https://doi.org/10.1038/nature08980

    • PubMed
    • Google Scholar
31. 1. Kim Y
    2. Sun H

    (2012) ASM-3 acid sphingomyelinase functions as a positive regulator of the DAF-2/AGE-1 signaling pathway and serves as a novel anti-aging target

    PLOS ONE 7:e45890.

    https://doi.org/10.1371/journal.pone.0045890

    • PubMed
    • Google Scholar
32. 1. Kimura KD
    2. Tissenbaum HA
    3. Liu Y
    4. Ruvkun G

    (1997) Daf-2, an insulin receptor-like gene that regulates longevity and diapause in Caenorhabditis elegans

    Science 277:942–946.

    https://doi.org/10.1126/science.277.5328.942

    • PubMed
    • Google Scholar
33. 1. Kirkwood TB

    (1977) Evolution of ageing

    Nature 270:301–304.

    https://doi.org/10.1038/270301a0

    • PubMed
    • Google Scholar
34. 1. Lakowski B
    2. Hekimi S

    (1998) The genetics of caloric restriction in Caenorhabditis elegans

    PNAS 95:13091–13096.

    https://doi.org/10.1073/pnas.95.22.13091

    • PubMed
    • Google Scholar
35. 1. Langmead B
    2. Salzberg SL

    (2012) Fast gapped-read alignment with bowtie 2

    Nature Methods 9:357–359.

    https://doi.org/10.1038/nmeth.1923

    • PubMed
    • Google Scholar
36. 1. Larson K
    2. Yan SJ
    3. Tsurumi A
    4. Liu J
    5. Zhou J
    6. Gaur K
    7. Guo D
    8. Eickbush TH
    9. Li WX

    (2012) Heterochromatin formation promotes longevity and represses ribosomal RNA synthesis

    PLOS Genetics 8:e1002473.

    https://doi.org/10.1371/journal.pgen.1002473

    • PubMed
    • Google Scholar
37. 1. Lee CK
    2. Weindruch R
    3. Prolla TA

    (2000) Gene-Expression profile of the ageing brain in mice

    Nature Genetics 25:294–297.

    https://doi.org/10.1038/77046

    • PubMed
    • Google Scholar
38. 1. Lee TW-S
    2. David HS
    3. Engstrom AK
    4. Carpenter BS
    5. Katz DJ

    (2019) Repressive h3k9me2 protects lifespan against the transgenerational burden of COMPASS activity in C. elegans

    eLife 8:e48498.

    https://doi.org/10.7554/eLife.48498

    • PubMed
    • Google Scholar
39. 1. Li H
    2. Handsaker B
    3. Wysoker A
    4. Fennell T
    5. Ruan J
    6. Homer N
    7. Marth G
    8. Abecasis G
    9. Durbin R
    10. 1000 Genome Project Data Processing Subgroup

    (2009) The sequence alignment/map format and samtools

    Bioinformatics 25:2078–2079.

    https://doi.org/10.1093/bioinformatics/btp352

    • PubMed
    • Google Scholar
40. 1. Lin K
    2. Dorman JB
    3. Rodan A
    4. Kenyon C

    (1997) Daf-16: an HNF-3/forkhead family member that can function to double the life-span of Caenorhabditis elegans

    Science 278:1319–1322.

    https://doi.org/10.1126/science.278.5341.1319

    • PubMed
    • Google Scholar
41. 1. Lin X
    2. Hengartner MO
    3. Kolesnick R

    (1998) Caenorhabditis elegans contains two distinct acid sphingomyelinases

    The Journal of Biological Chemistry 273:14374–14379.

    https://doi.org/10.1074/jbc.273.23.14374

    • PubMed
    • Google Scholar
42. 1. Lin K
    2. Hsin H
    3. Libina N
    4. Kenyon C

    (2001) Regulation of the Caenorhabditis elegans longevity protein DAF-16 by insulin/IGF-1 and germline signaling

    Nature Genetics 28:139–145.

    https://doi.org/10.1038/88850

    • PubMed
    • Google Scholar
43. 1. Lithgow GJ
    2. White TM
    3. Melov S
    4. Johnson TE

    (1995) Thermotolerance and extended life-span conferred by single-gene mutations and induced by thermal stress

    PNAS 92:7540–7544.

    https://doi.org/10.1073/pnas.92.16.7540

    • PubMed
    • Google Scholar
44. 1. Liu T
    2. Rechtsteiner A
    3. Egelhofer TA
    4. Vielle A
    5. Latorre I
    6. Cheung M-S
    7. Ercan S
    8. Ikegami K
    9. Jensen M
    10. Kolasinska-Zwierz P
    11. Rosenbaum H
    12. Shin H
    13. Taing S
    14. Takasaki T
    15. Iniguez AL
    16. Desai A
    17. Dernburg AF
    18. Kimura H
    19. Lieb JD
    20. Ahringer J
    21. Strome S
    22. Liu XS

    (2011) Broad chromosomal domains of histone modification patterns in C. elegans

    Genome Research 21:227–236.

    https://doi.org/10.1101/gr.115519.110

    • PubMed
    • Google Scholar
45. 1. López-Otín C
    2. Blasco MA
    3. Partridge L
    4. Serrano M
    5. Kroemer G

    (2013) The hallmarks of aging

    Cell 153:1194–1217.

    https://doi.org/10.1016/j.cell.2013.05.039

    • PubMed
    • Google Scholar
46. 1. Loyola A
    2. Bonaldi T
    3. Roche D
    4. Imhof A
    5. Almouzni G

    (2006) Ptms on H3 variants before chromatin assembly potentiate their final epigenetic state

    Molecular Cell 24:309–316.

    https://doi.org/10.1016/j.molcel.2006.08.019

    • PubMed
    • Google Scholar
47. 1. Lu T
    2. Pan Y
    3. Kao S-Y
    4. Li C
    5. Kohane I
    6. Chan J
    7. Yankner BA

    (2004) Gene regulation and DNA damage in the ageing human brain

    Nature 429:883–891.

    https://doi.org/10.1038/nature02661

    • PubMed
    • Google Scholar
48. 1. Lund J
    2. Tedesco P
    3. Duke K
    4. Wang J
    5. Kim SK
    6. Johnson TE

    (2002) Transcriptional profile of aging in C. elegans

    Current Biology 12:1566–1573.

    https://doi.org/10.1016/s0960-9822(02)01146-6

    • PubMed
    • Google Scholar
49. 1. Mao H
    2. Zhu C
    3. Zong D
    4. Weng C
    5. Yang X
    6. Huang H
    7. Liu D
    8. Feng X
    9. Guang S

    (2015) The nrde pathway mediates small-RNA-directed histone H3 lysine 27 trimethylation in Caenorhabditis elegans

    Current Biology 25:2398–2403.

    https://doi.org/10.1016/j.cub.2015.07.051

    • PubMed
    • Google Scholar
50. 1. Martens JHA
    2. O’Sullivan RJ
    3. Braunschweig U
    4. Opravil S
    5. Radolf M
    6. Steinlein P
    7. Jenuwein T

    (2005) The profile of repeat-associated histone lysine methylation states in the mouse epigenome

    The EMBO Journal 24:800–812.

    https://doi.org/10.1038/sj.emboj.7600545

    • PubMed
    • Google Scholar
51. 1. Maures TJ
    2. Greer EL
    3. Hauswirth AG
    4. Brunet A

    (2011) The H3K27 demethylase UTX-1 regulates C. elegans lifespan in a germline-independent, insulin-dependent manner

    Aging Cell 10:980–990.

    https://doi.org/10.1111/j.1474-9726.2011.00738.x

    • PubMed
    • Google Scholar
52. 1. McElwee JJ
    2. Schuster E
    3. Blanc E
    4. Thomas JH
    5. Gems D

    (2004) Shared transcriptional signature in Caenorhabditis elegans dauer larvae and long-lived daf-2 mutants implicates detoxification system in longevity assurance

    Journal of Biological Chemistry 279:44533–44543.

    https://doi.org/10.1074/jbc.M406207200

    • PubMed
    • Google Scholar
53. 1. McMurchy AN
    2. Stempor P
    3. Gaarenstroom T
    4. Wysolmerski B
    5. Dong Y
    6. Aussianikava D
    7. Appert A
    8. Huang N
    9. Kolasinska-Zwierz P
    10. Sapetschnig A
    11. Miska EA
    12. Ahringer J

    (2017) A team of heterochromatin factors collaborates with small RNA pathways to combat repetitive elements and germline stress

    eLife 6:e21666.

    https://doi.org/10.7554/eLife.21666

    • PubMed
    • Google Scholar
54. 1. Meister P
    2. Schott S
    3. Bedet C
    4. Xiao Y
    5. Rohner S
    6. Bodennec S
    7. Hudry B
    8. Molin L
    9. Solari F
    10. Gasser SM
    11. Palladino F

    (2011) Caenorhabditis elegans heterochromatin protein 1 (HPL-2) links developmental plasticity, longevity and lipid metabolism

    Genome Biology 12:R123.

    https://doi.org/10.1186/gb-2011-12-12-r123

    • PubMed
    • Google Scholar
55. 1. Murphy CT
    2. McCarroll SA
    3. Bargmann CI
    4. Fraser A
    5. Kamath RS
    6. Ahringer J
    7. Li H
    8. Kenyon C

    (2003) Genes that act downstream of DAF-16 to influence the lifespan of Caenorhabditis elegans

    Nature 424:277–283.

    https://doi.org/10.1038/nature01789

    • PubMed
    • Google Scholar
56. 1. Ni Z
    2. Ebata A
    3. Alipanahiramandi E
    4. Lee SS

    (2012) Two set domain containing genes link epigenetic changes and aging in Caenorhabditis elegans

    Aging Cell 11:315–325.

    https://doi.org/10.1111/j.1474-9726.2011.00785.x

    • PubMed
    • Google Scholar
57. 1. Ogg S
    2. Paradis S
    3. Gottlieb S
    4. Patterson GI
    5. Lee L
    6. Tissenbaum HA
    7. Ruvkun G

    (1997) The fork head transcription factor DAF-16 transduces insulin-like metabolic and longevity signals in C. elegans

    Nature 389:994–999.

    https://doi.org/10.1038/40194

    • PubMed
    • Google Scholar
58. 1. Padeken J
    2. Methot S
    3. Zeller P
    4. Delaney CE
    5. Kalck V
    6. Gasser SM

    (2021) Argonaute NRDE-3 and MBT domain protein LIN-61 redundantly recruit an H3K9me3 HMT to prevent embryonic lethality and transposon expression

    Genes & Development 35:82–101.

    https://doi.org/10.1101/gad.344234.120

    • PubMed
    • Google Scholar
59. 1. Pan KZ
    2. Palter JE
    3. Rogers AN
    4. Olsen A
    5. Chen D
    6. Lithgow GJ
    7. Kapahi P

    (2007) Inhibition of mRNA translation extends lifespan in Caenorhabditis elegans

    Aging Cell 6:111–119.

    https://doi.org/10.1111/j.1474-9726.2006.00266.x

    • PubMed
    • Google Scholar
60. 1. Pathare PP
    2. Lin A
    3. Bornfeldt KE
    4. Taubert S
    5. Van Gilst MR

    (2012) Coordinate regulation of lipid metabolism by novel nuclear receptor partnerships

    PLOS Genetics 8:e1002645.

    https://doi.org/10.1371/journal.pgen.1002645

    • PubMed
    • Google Scholar
61. 1. Peng JC
    2. Karpen GH

    (2008) Epigenetic regulation of heterochromatic DNA stability

    Current Opinion in Genetics & Development 18:204–211.

    https://doi.org/10.1016/j.gde.2008.01.021

    • PubMed
    • Google Scholar
62. 1. Peters AHFM
    2. Kubicek S
    3. Mechtler K
    4. O’Sullivan RJ
    5. Derijck AAHA
    6. Perez-Burgos L
    7. Kohlmaier A
    8. Opravil S
    9. Tachibana M
    10. Shinkai Y
    11. Martens JHA
    12. Jenuwein T

    (2003) Partitioning and plasticity of repressive histone methylation states in mammalian chromatin

    Molecular Cell 12:1577–1589.

    https://doi.org/10.1016/s1097-2765(03)00477-5

    • PubMed
    • Google Scholar
63. 1. Raizen DM
    2. Lee RY
    3. Avery L

    (1995) Interacting genes required for pharyngeal excitation by motor neuron MC in Caenorhabditis elegans

    Genetics 141:1365–1382.

    https://doi.org/10.1093/genetics/141.4.1365

    • PubMed
    • Google Scholar
64. 1. Ramírez F
    2. Dündar F
    3. Diehl S
    4. Grüning BA
    5. Manke T

    (2014) DeepTools: a flexible platform for exploring deep-sequencing data

    Nucleic Acids Research 42:W187–W191.

    https://doi.org/10.1093/nar/gku365

    • PubMed
    • Google Scholar
65. 1. Sacket SJ
    2. Chung HY
    3. Okajima F
    4. Im DS

    (2009) Increase in sphingolipid catabolic enzyme activity during aging

    Acta Pharmacologica Sinica 30:1454–1461.

    https://doi.org/10.1038/aps.2009.136

    • PubMed
    • Google Scholar
66. 1. Samuelson AV
    2. Carr CE
    3. Ruvkun G

    (2007) Gene activities that mediate increased life span of C. elegans insulin-like signaling mutants

    Genes & Development 21:2976–2994.

    https://doi.org/10.1101/gad.1588907

    • PubMed
    • Google Scholar
67. 1. Schwartz-Orbach L
    2. Zhang C
    3. Sidoli S
    4. Amin R
    5. Kaur D
    6. Zhebrun A
    7. Ni J
    8. Gu SG

    (2020) Caenorhabditis elegans nuclear RNAi factor SET-32 deposits the transgenerational histone modification, h3k23me3

    eLife 9:e54309.

    https://doi.org/10.7554/eLife.54309

    • PubMed
    • Google Scholar
68. 1. Shilatifard A

    (2008) Molecular implementation and physiological roles for histone H3 lysine 4 (H3K4) methylation

    Current Opinion in Cell Biology 20:341–348.

    https://doi.org/10.1016/j.ceb.2008.03.019

    • PubMed
    • Google Scholar
69. 1. Shumaker DK
    2. Dechat T
    3. Kohlmaier A
    4. Adam SA
    5. Bozovsky MR
    6. Erdos MR
    7. Eriksson M
    8. Goldman AE
    9. Khuon S
    10. Collins FS
    11. Jenuwein T
    12. Goldman RD

    (2006) Mutant nuclear lamin a leads to progressive alterations of epigenetic control in premature aging

    PNAS 103:8703–8708.

    https://doi.org/10.1073/pnas.0602569103

    • PubMed
    • Google Scholar
70. 1. Spracklin G
    2. Fields B
    3. Wan G
    4. Becker D
    5. Wallig A
    6. Shukla A
    7. Kennedy S

    (2017) The RNAi inheritance machinery of Caenorhabditis elegans

    Genetics 206:1403–1416.

    https://doi.org/10.1534/genetics.116.198812

    • PubMed
    • Google Scholar
71. 1. Tepper RG
    2. Ashraf J
    3. Kaletsky R
    4. Kleemann G
    5. Murphy CT
    6. Bussemaker HJ

    (2013) PQM-1 complements DAF-16 as a key transcriptional regulator of DAF-2-mediated development and longevity

    Cell 154:676–690.

    https://doi.org/10.1016/j.cell.2013.07.006

    • PubMed
    • Google Scholar
72. 1. Tian Y
    2. Garcia G
    3. Bian Q
    4. Steffen KK
    5. Joe L
    6. Wolff S
    7. Meyer BJ
    8. Dillin A

    (2016) Mitochondrial stress induces chromatin reorganization to promote longevity and UPR (MT)

    Cell 165:1197–1208.

    https://doi.org/10.1016/j.cell.2016.04.011

    • PubMed
    • Google Scholar
73. 1. Towbin BD
    2. González-Aguilera C
    3. Sack R
    4. Gaidatzis D
    5. Kalck V
    6. Meister P
    7. Askjaer P
    8. Gasser SM

    (2012) Step-Wise methylation of histone H3K9 positions heterochromatin at the nuclear periphery

    Cell 150:934–947.

    https://doi.org/10.1016/j.cell.2012.06.051

    • PubMed
    • Google Scholar
74. 1. Tschiersch B
    2. Hofmann A
    3. Krauss V
    4. Dorn R
    5. Korge G
    6. Reuter G

    (1994) The protein encoded by the Drosophila position-effect variegation suppressor gene Su (var) 3-9 combines domains of antagonistic regulators of homeotic gene complexes

    The EMBO Journal 13:3822–3831.

    https://doi.org/10.1002/j.1460-2075.1994.tb06693.x

    • PubMed
    • Google Scholar
75. 1. Tsurumi A
    2. Li WX

    (2012) Global heterochromatin loss: a unifying theory of aging?

    Epigenetics 7:680–688.

    https://doi.org/10.4161/epi.20540

    • PubMed
    • Google Scholar
76. 1. Villeponteau B

    (1997) The heterochromatin loss model of aging

    Experimental Gerontology 32:383–394.

    https://doi.org/10.1016/s0531-5565(96)00155-6

    • PubMed
    • Google Scholar
77. 1. Wan Q-L
    2. Meng X
    3. Wang C
    4. Dai W
    5. Luo Z
    6. Yin Z
    7. Ju Z
    8. Fu X
    9. Yang J
    10. Ye Q
    11. Zhang Z-H
    12. Zhou Q

    (2022) Histone H3K4me3 modification is a transgenerational epigenetic signal for lipid metabolism in Caenorhabditis elegans

    Nature Communications 13:768.

    https://doi.org/10.1038/s41467-022-28469-4

    • PubMed
    • Google Scholar
78. 1. Wenzel D
    2. Palladino F
    3. Jedrusik-Bode M

    (2011) Epigenetics in C. elegans: facts and challenges

    Genesis 49:647–661.

    https://doi.org/10.1002/dvg.20762

    • PubMed
    • Google Scholar
79. 1. Wood A
    2. Shukla A
    3. Schneider J
    4. Lee JS
    5. Stanton JD
    6. Dzuiba T
    7. Swanson SK
    8. Florens L
    9. Washburn MP
    10. Wyrick J
    11. Bhaumik SR
    12. Shilatifard A

    (2007) Ctk complex-mediated regulation of histone methylation by COMPASS

    Molecular and Cellular Biology 27:709–720.

    https://doi.org/10.1128/MCB.01627-06

    • PubMed
    • Google Scholar
80. 1. Wood JG
    2. Hillenmeyer S
    3. Lawrence C
    4. Chang C
    5. Hosier S
    6. Lightfoot W
    7. Mukherjee E
    8. Jiang N
    9. Schorl C
    10. Brodsky AS
    11. Neretti N
    12. Helfand SL

    (2010) Chromatin remodeling in the aging genome of Drosophila

    Aging Cell 9:971–978.

    https://doi.org/10.1111/j.1474-9726.2010.00624.x

    • PubMed
    • Google Scholar
81. 1. Wood JG
    2. Helfand SL

    (2013) Chromatin structure and transposable elements in organismal aging

    Frontiers in Genetics 4:274.

    https://doi.org/10.3389/fgene.2013.00274

    • PubMed
    • Google Scholar
82. 1. Woodhouse RM
    2. Buchmann G
    3. Hoe M
    4. Harney DJ
    5. Low JKK
    6. Larance M
    7. Boag PR
    8. Ashe A

    (2018) Chromatin modifiers SET-25 and SET-32 are required for establishment but not long-term maintenance of transgenerational epigenetic inheritance

    Cell Reports 25:2259–2272.

    https://doi.org/10.1016/j.celrep.2018.10.085

    • PubMed
    • Google Scholar
83. 1. Yuan J
    2. Chang S-Y
    3. Yin S-G
    4. Liu Z-Y
    5. Cheng X
    6. Liu X-J
    7. Jiang Q
    8. Gao G
    9. Lin D-Y
    10. Kang X-L
    11. Ye S-W
    12. Chen Z
    13. Yin J-A
    14. Hao P
    15. Jiang L
    16. Cai S-Q

    (2020) Two conserved epigenetic regulators prevent healthy ageing

    Nature 579:118–122.

    https://doi.org/10.1038/s41586-020-2037-y

    • PubMed
    • Google Scholar

Decision letter after peer review:

Thank you for submitting your article "H3K9me1/2 methylation limits the lifespan of C. elegans" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Carlos Isales as the Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: David Katz (Reviewer #3).

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Summary of reviewer discussion:

The reviewers were all very positive about the central idea of the manuscript, that putative HMTs act synergistically with daf-2 to extend lifespan. During review and subsequent discussion, several key weaknesses of the manuscript were identified that will need to be addressed before the manuscript is suitable for publication at eLife. While many of these weaknesses can be addressed by modifying the text and including additional supplementary information likely on hand, they also include a limited number of experiments that are necessary to support the key findings of the manuscript. Of highest importance, revisions must address the lack of wild-type and mutant only controls in methylation studies and the assumption of the enzymes studied as being bona fide H3K9 methyltransferases. A revision should focus on what is being shown (changes in H3K9me1/2 that correspond with lifespan in daf-2 mutants), while either addressing or taking into account the remaining unknowns. Additionally, all the reviewer comments are included, from which the essential revisions were identified, that should be addressed before publication.

Essential revisions:

1. The 6 genes being studied are putative methyltransferases, but are frequently described in the text as though they are validated and bona fide H3K9 methyltransferases. While the data presented support that they affect H3K9 methylation status, there are both direct and indirect ways this could occur. Some evidence that these putative HMTs are actual H3K9me1/me2 MTs (not acting indirectly, as observed in Schwartz-Orbach L et al., eLife 2020) and that they and daf-2 affect H3K9 methylation (ChIP seq) would greatly strengthen the manuscript. Barring this, the circumstantial data using inhibition would need to be strengthened and the text modified to clarify the lack of evidence (e.g., the introduction currently states that MET-2 and SET-25 are responsible for the methylation of H3K9, rather than saying the genes are required for the presence of H3K9me in vivo) on the direct nature of the observed effects.

2. To properly interpret the data, wild-type and mutant controls in methylation assays must be included. Presumably, based on the lifespan results, daf-2 changes methylation status in such a way that these genes are now limiting lifespan. Without control data, it is very difficult to interpret how this may work.

3. An explanation for why 6 different H3K9me1/me2 methylases would have such similar phenotypes. This explanation could come out of #1 and/or could involve other experiments and insights, including a better discussion of the individual HMTs and how they are similar/different. For example, addressing the H3K23me2/3 data in light of the Gu lab paper referenced in #1 may help in emphasizing the complexity of the interaction network and will also highlight the biological relevance of these findings.

4. Better discussion and clarity on the relationship between brood size and fertility tradeoffs in these animals, including inclusion of daf-2 single mutant brood sizes.

5. Better clarity on replication, blinding, and inclusion of summary survival data including conditions examined per assay, N, animals censored, median lifespan (along with average lifespan), and comparison used for determination of significance. If replicates are lacking, additional replicates may be necessary (see reviewer comments).

6. A significantly more detailed transcriptomic analysis and clarity about the analysis and it's methods: e.g., unbiased analysis to identify other pathways, differentiation between DAF-16 class I/II genes, better evidence to claim class I.

7. Tempered statements on generality, or more experiments to show generality, and clarified writing to more accurately report findings.

8. Use of the data to clarify the relationship between heterochromatin, H3K9me1/2, and H3K9me3.1.

Reviewer #1 (Recommendations for the authors):

1) The C. elegans enzymes described in the intro are still only putative methyltransferases. Most of them, outside of SET-2 have not been shown by in vitro methylation assays so should be referred to as putative methyltransferases (unless the authors wish to perform in vitro methylation assays to complement the in vivo findings of others which could still be indirect). Since so much of the manuscript is focused on H3K9 methylation it would be good to show that some of these enzymes (maybe just SET-21 if it is too much to show for them all) are indeed H3K9 methyltransferases.

2) It's important to see how H3K9 methylation changes in daf-2 worms relative to WT worms, both by WB at gross level and genome wide by ChIPseq. This change in H3K9me1/me2 genome wide should then be compared to daf-16 target gene lists to see if there is an enrichment.

3) For 6b you have to present all 4 conditions for epistasis experiments. The reader needs to see what the effect of lys-7 deletion is in the daf-2 background by itself. And then 2-way anova needs to be done to test whether the effect of lys-7 in the daf-2 background is different from in the daf-2;set-21 background.

4) Some attempt should be made to explain why there are 6 redundant H3K9me1/me2 methylases (again I'm not convinced yet that there are because of point 1) that all affect lifespan in the same manner. Are these enzymes expressed in the same tissue? Do they affect H3K9me1/me2 on the same genes? At the same time? Usually it is believed that different enzymes that modify the same mark have some sort of specificity but here they are not seeing that. I don't think it's necessary to perform tissue specific experiments for this initial manuscript but some experimental justification should be provided. Do they believe that it's the amount of heterochromatin DNA in general which is regulating this? Or is it a gene specific effect?

Reviewer #2 (Recommendations for the authors):

1. As the authors mentioned in the discussion, H3K9me2 is the histone modification in C. elegans that corresponds best with canonical heterochromatin factors like HPL-2 (McMurchy et al., 2017). However, in their results, they focus on the lack of effect on lifespan seen for H3K9me3 modifiers like SET-25 to rule out any effect with heterochromatin. In fact, despite corresponding highly with H3K9me2, one of the marks implicated in further extending the lifespan of their HMT; daf-2 double mutants, hpl-1 and hpl-2 mutations had no effect in a daf-2 background (although they do seem to slightly decrease lifespan, and it would be interesting to see the statistical comparison of a log-rank test). The authors should clarify the relationship between heterochromatin, H3K9me1/2, and H3K9me3.

2. I was confused by the statements describing Figure 4C on lines 176-178. Both double and triple mutants were much longer-lived than daf-2 single mutants, rather than having similar lifespan extensions as the daf-2 single mutants. There was even a distinction between the daf-2; set-21 double mutant and the daf-2; set-21; set-2 triple mutant – the presence of a set-2 mutation represses the slight lifespan extension difference seen in the daf-2; set-21 double mutant when compared to the daf-2; set-2 double mutant. The authors should rephrase their description of their results to more accurately convey their findings.

3. My major concerns with the approach to transcriptomic analysis were conveyed in the public review, but I include more specific concerns here. Given that Class I genes are an important downstream consequence of DAF-16, it isn't clear why these are primarily unaffected in the daf-2 control condition shown in Figure 5A. Furthermore, the authors focus primarily on the top 50 Class I genes, but there are interesting effects seen in the top 50 Class II genes as well, especially given that mutations in the HMTs seem to suppress some of the upregulation observed in daf-2 control – why exclude these from their interpretations? Additionally, the brevity of the methods makes it hard to tell exactly what comparisons are being drawn – in the heatmaps shown in in Figure 5C, D, does the color refer upreglation in comparison to a wild-type control? Why not compare misregulation to a daf-2 single mutant instead? What does "Top 50" refer to – are these the fifty most-affected in these specific data sets, or are they a gene set identified by an independent metric? Why limit the analysis to only fifty genes? Finally, this analysis would be further strengthened by comparing locus-specific effects to the known locations of H3K9me1/2/3 peaks in wild-type genomes in previously published datasets (like in Li et al., 2021 or McMurchy et al., 2017) or even to H3K9me1/2/3 localization in datasets of relevant HMT mutants like set-32.

4. How were the 10 Class I DAF-16 target genes chosen? It doesn't seem like they were in the top 50 (or not that I could easily see). Furthermore, why were no Class II genes chosen? The authors should clarify how they chose to examine these specific candidates, and if necessary, extend the analysis for other targets to see whether the effects are specific to Class I targets only (as they argue in their model). Furthermore, each of these candidate genes needs to be examined in a daf-2 mutant background to fully interpret the meaning of the suppression of long-lived daf-2; set-21 mutants.

5. The authors claim that loss of H3K9me1/2 increases binding of DAF-16 at its target genes (line 282). There is no evidence to support this claim (and a lack of evidence for this claim, if we take the DAF-16 localization data in Figure S3 at complete face value). This could be part of their proposed model (outlined in Figure S7), but as written, it reads as if it were a summary of work presented in the paper. The authors should clarify their writing to ensure full accuracy for what their data can support.

6. Throughout the manuscript, it's difficult to build a cohesive understanding of the functional effects of the HMT mutants. In part, this is because the paper investigates several HMTs without discussing their specific effects (in certain tissues or cell types, or at specific stages). The analyses are sometimes too broad (immunoblots of total histone modifications) and at other times too narrow (focusing only on 50 Class I DAF-16 target genes). Given that H3K9me1/2 will likely have locus specificity, it is unsatisfying to leave the analysis at the global level of immunoblots (although it is very compelling to see global effects in their mutants). Including a deeper, and unbiased, analysis of their transcriptomic data will help to bridge the gap between the global and the locus-specific, and may even point out other interesting avenues that will further bolster their claims about potential mechanism (or open new avenues of study).

[Editors’ note: further revisions were suggested prior to acceptance, as described below.]

Thank you for resubmitting your work entitled "H3K9me1/2 methylation limits the lifespan of C. elegans" for further consideration by eLife. Your revised article has been evaluated by Carlos Isales (Senior Editor) and a Reviewing Editor.

The manuscript has been improved but there are some remaining issues that need to be addressed, as outlined below:

Primarily, the three areas that remain an issue before submission are (1) a second replicate needs to be included (and ideally, performed blinded) for experiments such as Figures 1B, 4A, and 4B, (2) the appropriate epistasis controls should be included as requested by reviewer 1 to properly interpret the experiments (e.g., Figure 7), and (3) the text needs to be modified where precision remains an issue, as described by reviewer 1. There should be no technical issues with completing these experiments, and they're relatively straightforward, although I appreciate that it can take some time to complete longevity assays for these long-lived mutants. However, these experiments are necessary to experimentally validate the claims within the manuscript, along with responses to the reviewer comments below.

Reviewer #1 (Recommendations for the authors):

In the revised manuscript "H3K9me1/2 methylation limits the lifespan of C. elegans" Huang M, Hong M et al., performed additional experiments to examine the role of putative H3K9 methyltransferases in the lifespan regulation of daf-2 mutant worms. While the new experiments do bolster some portions of the manuscript I felt that the critical questions which I had asked in my initial review weren't addressed and are still important. I understand that COVID has made it difficult to perform certain experiments but with those restrictions in mind then the paper needs to be reworked to reflect what the experiments actually show rather than the model the authors wish to be true. Most importantly the role of H3K9 methylation in daf-2 mutants is still not really clear. These genes seem to have virtually no effect in WT background so the text needs to really be rewritten to reflect that. Starting with the title of the manuscript and then from the intro on it should be written from the perspective of "we wanted to examine how chromatin modifying enzymes were important in insulin receptor mutant worms".

1) As stated last review (major point #3), in epistasis experiments (such as WT and eat-2 +/- set-21 and WT and daf-2 +/- met-2, and Figure 3C +D, 7B) you need to include all 4 conditions in the experiment to make any conclusions. For example, you can't do the set-21 vs WT at a different time and assume that the set-21 is still going to have no effect in that particular experiment. There might be some unknown factor (food, temperature, etc) that makes set-21 actually have an effect in the WT worms in that experiment which would mean a very different conclusion. Then after the experiment is done you perform a 2 way ANOVA analysis to reveal whether set-21 has a significantly different effect in the WT context or the mutant context. This comment still holds true for asm-2, Y39G8B.7, etc… that I had asked for last review. You do not need to perform all possible combinations of these 3 genes but at the very least daf-2, daf-2;set-21, daf-2;asm-2 and daf-2;set-21;asm-2 if you want to make any conclusion about asm-2 relative to set-21 in the daf-2 background), examining 3 lines in an epistasis experiment rather than 4 is not really informative.

2) The text needs to be rewritten to reflect that this is all only occurring in daf-2 mutant worms and not in WT worms. This rewriting needs to start with the title ("…limits the lifespan of daf-2 mutant C. elegans".) and work its way thru the intro, results and discussion (where the authors claim H3K9me regulators should be considered for lifespan regulation). Need to include explicit writing that deletion of all these putative enzymes has only modest effects on H3K9me and lifespan in the WT context and their role is only revealed in the daf-2 mutants. Need to speculate about what that is. What makes the daf-2 mutation special to reveal a regulatory role for these genes?

Reviewer #2 (Recommendations for the authors):

Upon resubmission, this paper is greatly improved – readers now have the evidence needed to evaluate the paper's main claim that H3K9 HMTs synergistically extend lifespan in daf-2 mutants, in part via DAF-16, but also via. The additional experiments, along with a more tempered approach to discussing their findings, significantly strengthen the paper.

However, several of our concerns were not addressed by these changes, nor in the rebuttal. For example, Reviewer 2 asked that some of the key lifespan experiments be replicated at least once and also performed blinded – there are several figures where only a single replicate is shown. Key controls are still missing, like in Figure 7, where Reviewer 1 had asked for appropriate controls to interpret the epistasis experiments for example, set-21 single mutants and daf-2; lys-7 double mutants. The authors' rebuttal mentioned generating new alleles for some genes, but providing appropriate single mutant and double mutant controls should require no new strains to be built (since the strains likely already exist if the authors generated the triple mutant).

The extended analysis of the mRNA-seq is better (and opens up an interesting line of questioning for future studies!). However, the authors failed to address most of Reviewer 2's Major Point 3 about their transcriptomic analysis, nor did they include further details about how the analysis was done. The methods for the mRNA-seq will need to be more detailed. What is their unbiased approach? What is the differential expression shown in the Figure 5? For example, I had previously asked: "Does the color refer to upregulation in comparison to a wild-type control? Why not compare misregulation to a daf-2 single mutant instead?" and the answer is still not clear. Furthermore, if Class I genes are defined as being induced in daf-2 mutants, why are they downregulated in Figure 5E? And if Class II genes are defined as being not induced in daf-2 mutants, why are they upregulated in Figure 5F? I may be misunderstanding the gene classifications, but it is a demonstration of how the text is unclear. Overall, I found the new transcriptomic analyses are compelling, but more care needs to be taken to fully explain what was done.

Essential revisions:

1. The 6 genes being studied are putative methyltransferases, but are frequently described in the text as though they are validated and bona fide H3K9 methyltransferases. While the data presented support that they affect H3K9 methylation status, there are both direct and indirect ways this could occur. Some evidence that these putative HMTs are actual H3K9me1/me2 MTs (not acting indirectly, as observed in Schwartz-Orbach L et al., eLife 2020) and that they and daf-2 affect H3K9 methylation (ChIP seq) would greatly strengthen the manuscript. Barring this, the circumstantial data using inhibition would need to be strengthened and the text modified to clarify the lack of evidence (e.g., the introduction currently states that MET-2 and SET-25 are responsible for the methylation of H3K9, rather than saying the genes are required for the presence of H3K9me in vivo) on the direct nature of the observed effects.

Thanks very much for the comments.

We completely agree with the reviewers that we lack direct biochemical evidence showing that these SET proteins are bona fide H3K9 methyltransferases. We have tried very hard to conduct similar experiments as Schwartz-Orbach has done on SET-32. However, given the covid-19 pandemic and biosafety issue, we could not get access to [3H]-labeled S-adenosylmethionine (SAM). We are very sorry that we could not pursue further at this stage. Therefore, we have revised the manuscript saying that these SET proteins are putative histone methyltransferases and are required for the presence of H3K9me1/2 marks. The text of MET-2 and SET-25 has been revised similarly and saying that they are required for the presence of H3K9me in vivo.

We also added in the Discussion section that further investigations are required to reveal their enzymatic activities by in vitro experiments (lines 496-506).

2. To properly interpret the data, wild-type and mutant controls in methylation assays must be included. Presumably, based on the lifespan results, daf-2 changes methylation status in such a way that these genes are now limiting lifespan. Without control data, it is very difficult to interpret how this may work.

Thanks very much for the comments.

We have included new data comparing the global levels of H3K9me1/2/3 in N2 and daf-2 animals by Western blotting (new Figure 8—figure supplement 1A). The daf2 mutation itself did not significantly change the gross levels of H3K9me1/2/3 status.

We also compared the levels of H3K9me1/2/3, H3K4me1/2/3, H3K27me1/2/3, H3K36me1/2/3 and H3K23me2/3 in N2 and set mutants at L4 and embryonic stages (new Figure 8—figure supplement 1B-C). In N2 background, the mutation of these set genes did not significantly change gross levels of H3K9me1/2/3 either.

We speculated that the mutation of daf-2 sensitizes the animals, for example,

increases the cytoplasm-to-nuclear translocation of DAF-16. We have generated a

single-copy, in situ GFP-tagged DAF-16 transgene, crossed to daf-2, set-21, set-25

single mutants and their double mutants, and scored the nuclear localization of DAF16 at 20oC and 25oC, respectively (new Figure 5A-B, Figure 5—figure supplement 1B). The mutation of daf-2 increased the nuclear accumulation of DAF-16. The double mutants daf-2;set-21 and daf-2;set-25 further increased the nuclear accumulation of DAF-16.

Interestingly, DAF-16 was nearly 100% localized in the nucleus in both daf-2;set21 and daf-2;set-25 double mutants. Yet daf-2;set-21, but not daf-2;set-25, extended the lifespan of daf-2 animals, suggesting addition regulation required beyond the nuclear localization of DAF-16. Our work suggested that the transcription activity of DAF-16 may preferentially depend on H3K9me1/2, but not H3K9me3.

We have revised the text to include the information (lines 245-275, 357-361).

3. An explanation for why 6 different H3K9me1/me2 methylases would have such similar phenotypes. This explanation could come out of #1 and/or could involve other experiments and insights, including a better discussion of the individual HMTs and how they are similar/different. For example, addressing the H3K23me2/3 data in light of the Gu lab paper referenced in #1 may help in emphasizing the complexity of the interaction network and will also highlight the biological relevance of these findings.

Thanks very much for the comments. We are also eager to know why the 6 different H3K9me1/me2 methylases would have such similar phenotypes in this lifespan extension assay, yet have no clue at this stage.

We have generated in situ single-copy 3xFLAG-GFP tagged SET-21, SET-25 and SET-32 transgenes by CRISPR/Cas9 technology and single-copy GFP-3xFlAG-tagged SET-13 transgenic strains using the Mos1-mediated single-copy insertion (MosSCI) technology. These animals were imaged in N2 and daf-2 background at embryonic and larva stages respectively, to figure out the expression patterns and cellular localizations of SET proteins (new Figure 4—figure supplement 1). However, we failed to find a clear correlation between their expression pattern with the lifespan extension phenotype. The depletion of SET-21 and SET-32 extended the lifespan of daf-2 animals, but GFP::SET21 and GFP::SET-32 showed different expression patterns and subcellular localizations. GFP::SET-21 exclusively located in the nucleus but GFP::SET-32 located in both the nucleus and cytoplasm in embryos. At larval stage, GFP::SET-21 was expressed in soma but GFP::SET-32 was expressed in germline. Both set-13 and set-25 mutants did not show lifespan extension in N2 and daf-2 background worms. SET-13 and SET-25 were enriched in the nucleus. SET-13::GFP was expressed through 8-cell stage to lateembryonic stage, while SET-25::GFP was expressed in the whole embryonic stage and the germline in young adults.

To identify the target genes in this group of lifespan extension daf-2;set mutants, we re-analyzed the mRNA expression profile in the double mutants via an unbiased method (Figure 6, Figure 6—figure supplement 1). We have identified 49 co-upregulated genes and 11 co-downregulated genes that are specifically enriched in the long-lived double mutants daf-2;set-19, daf-2,set-21, daf-2;set-32, but not in daf-2 and daf-2;daf25 animals (Figure 6A-B and Figure 6—figure supplement 1A). Interestingly, among the 49 co-upregulated genes, 27 of them are also DAF-16 Class I genes (Figure 6—figure supplement 1B). 22 co-upregulated genes are not DAF-16 Class I genes, suggesting the existence of additional regulations.

C. elegans’ genome encodes 10 putative H3K9 methyltransferases, most of which have not been directly characterized for their biochemical properties by in vitro experiments. The presence of these many putative H3K9 methyltransferases may lead to a complexity of the interaction network of histone modifications. In addition, some SET proteins may have multiple roles in mediating histone modifications. For example, SET-19 is required for both H3K9me1/2 and H3K23me3 marks, further bringing up complexity of the regulation. Further investigations are necessary to reveal their enzymatic activities by in vitro experiments. In addition, identifying the in vivo targeted genes will also be crucial to help understanding why and how these SET proteins are involved in lifespan regulation.

We have revised the text to include the information (lines 186-201, 301-320, 496506).

4. Better discussion and clarity on the relationship between brood size and fertility tradeoffs in these animals, including inclusion of daf-2 single mutant brood sizes.

Thanks very much for the suggestion.

We have included the brood size assay of N2, daf-2, set-21 and daf-2;set-21. Knocking out set-21 reduced the brood size of daf-2(e1370) mutant worms and significantly extended lifespan (Figure 1B, Figure1—figure supplement 1C). However, we did not find significant difference between the brood size of N2 and daf-2 (new Figure 1—figure supplement 1C and Figure 3—figure supplement 1B).

A trade-off has been proposed between fertility and lifespan, where increasing lifespan usually comes at the cost of reduced fertility. In C. elegans, removing the germ cells extends lifespan by triggering the nuclear localization of the DAF-16/FOXO transcription factor and activating target genes in the intestine. Some long-lived mutant worms, like daf-2 and rsks^-1, show reduced brood size. Consistently, we found that the long-lived worms daf-2;met-2, daf-2;set-6, daf-2;set-19, daf-2;set-20, daf-2;set-21, daf-2;set-32, and daf-2;set-33 all have reduced brood size compared to that of daf-2 animals. However, additional investigations are necessary to examine whether removing the germline cells could further extend worm lifespan.

We revised the Discussion section to include this information (lines 115-117, 402411).

5. Better clarity on replication, blinding, and inclusion of summary survival data including conditions examined per assay, N, animals censored, median lifespan (along with average lifespan), and comparison used for determination of significance. If replicates are lacking, additional replicates may be necessary (see reviewer comments).

Thanks very much for the comments.

We have revised the Materials and methods section (lines 549-559) and figure legends to include the information of each life span assay. We also included a new Table S1 to summarize all lifespan experiments.

6. A significantly more detailed transcriptomic analysis and clarity about the analysis and it's methods: e.g., unbiased analysis to identify other pathways, differentiation between DAF-16 class I/II genes, better evidence to claim class I.

Thanks very much for the suggestions.

We used two approaches to analyze the mRNA-seq data.

First, the depletion of DAF-2 reduces the insulin signaling pathway, promotes DAF-16 nuclear translocation, and leads to both upregulation and downregulation of large sets of genes, referred to as Class I and II genes, respectively. Class I genes are induced in daf-2 mutants but are repressed in daf-2;daf-16 double mutants. Class II genes are not induced in daf-2 mutants but are induced in daf-2; daf-16 double mutants. 1663 genes are classified as positive (class I) DAF-16 targets and 1733 genes are classified as negative (class II) DAF-16 targets of DAF-16. Class I genes are enriched for the Gene Ontology categories including oxidation, reduction, and energy metabolism, whereas class II genes are enriched for genes involved in biosynthesis, growth, reproduction, and development. We performed mRNA-seq and analyzed Class I and II DAF-16 genes to identify the mis-regulated genes in the daf-2;set mutants. Interestingly, the mRNA levels of DAF-16 Class I, but not Class II, genes are consistently activated in long-lived daf-2;set-19, daf-2;set-21 and daf-2;set-32 worms, than in the control daf-2 and daf-2;set-25 animals (Figure 5E-F and new Figure 5figure supplement 2).

Second, to identify the target genes in the group of lifespan extension daf-2;set mutants, we re-analyzed the mRNA expression profile in the double mutants via an unbiased method (new Figure 6, Figure 6—figure supplement 1). We have identified 49 co-upregulated genes and 11 co-downregulated genes that are specifically enriched in the long-lived double mutants daf-2;set-19, daf-2,set-21, daf-2;set-32, but not in daf-2 and daf-2;daf-25 animals (Figure 6A-B and Figure 6—figure supplement 1A). Interestingly, among the 49 co-upregulated genes, 27 of them are also DAF-16 Class I genes (new Figure 6—figure supplement 1B). 22 co-upregulated genes are not DAF-16 Class I genes, suggesting the existence of additional regulations.

We then analyzed a number of known transcription factors for their binding to the co-regulated targeted genes (new Figure 6C-D, Figure 6—figure supplement 1C-E). Among them, we found that DAF-16 and NHR-80 were specifically enriched at the transcription start sites (TSS) of the 49 co-upregulated genes. NHR-80 is a homolog of mammalian hepatocyte nuclear factor 4 and is an important nuclear hormone receptor involved in the control of fat consumption and fatty acid composition in C. elegans. Among the genes targeted by DAF-16 and NHR-80, twelve of them are co-regulated by both factors, which is consistent with previous report that daf-16 and nhr-80 function in parallel pathway for lipid metabolism.

We revised the text to include this information (lines 284-320).

7. Tempered statements on generality, or more experiments to show generality, and clarified writing to more accurately report findings.

Thanks very much for the suggestion. We have revised the manuscript to tone down these SET proteins as putative H3K9me1/2 methyltransferases. And we also carefully revised the text to more accurately report our findings.

8. Use of the data to clarify the relationship between heterochromatin, H3K9me1/2, and H3K9me3.1.

Thanks very much for the suggestion. Actually, the relationship between heterochromatin, H3K9me1/2, and H3K9me3 is quite intriguing.

H3K9me3 is a hallmark of heterochromatin and is often enriched on repetitive sequences. In addition, H3K9me1/2/3 have all been shown enrichment on repeat sequences (GERSTEIN et al., 2010). H3K9 methylation and its association with the nuclear envelope help maintain silencing of repetitive sequences and may contribute to the maintenance and stability of heterochromatin-like regions in C. elegans, as observed in other organisms. A hallmark of heterochromatin is the enrichment of HP1. The nematode C. elegans has two HP1 paralogous proteins: HP1 Like 1 and 2 (HPL-1 and HPL-2). Interestingly, HPL-2 may serve additional roles than HPL-1, as hpl-2 mutants display diverse defects while hpl-1 mutants generally lack obvious defects. The pattern of HPL-2 most closely resembles H3K9me1 and H3K9me2 but less closely resembles H3K9me3. HPL-2-bound genes, especially those located on arms, possess significantly higher levels of H3K9me2, but show no significant difference in H3K9me3 levels when compared to genes not bound by HPL-2. SET-25 is a putative H3K9me3 methyltransferase and is recruited to targeted genomic loci either by H3K9me2 which is deposited by MET-2 or by the somatic Argonaute protein NRDE-3 and small RNAs. SET-25 colocalizes with both H3K9me3 and HPL-1, but does not colocalize with HPL-

2.

H3K9me2 and H3K9me3 exhibit different patterns of repeat enrichment. H3K9me2 is more associated with DNA transposons and satellite repeats, similar to the heterochromatin factors HPL-2, LET-418, LIN-13, LIN-61, and MET-2, whereas H3K9me3 is particularly associated with retrotransposon families, especially LINE and SINE elements. Moreover, H3K9me2 and the five heterochromatin factors are enriched at telomeres, whereas H3K9me3 is not. Therefore, H3K9me2, rather than H3K9me3, may be more closely associated with canonical heterochromatin factors in C. elegans. It was postulated that losing repressive chromatin is detrimental to lifespan. In humans, two premature aging diseases are caused by mutations in lamins that reduce heterochromatin and disrupt its nuclear localization. In C. elegans, Drosophila, and mammals, heterochromatin decreases as individuals grow older. Additionally, mutations that increase repressive chromatin extend lifespan in eukaryotes. In N2 background animals, the depletion of hpl-2 was shown increased lifespan. However, we did not observe a further lifespan extension in daf-2;hpl-2 animals, comparing to the daf-2 mutant. Similarly, we did not observe a significant change in lifespan in daf2;set-25 mutants. The depletion of the two heterochromatin factors HPL-1 and HPL-2 failed to further extend the lifespan, suggesting that heterochromatin itself may not be sufficient to perturb lifespan in daf-2 mutant worms.

We have revised the Discussion section to include the above information (lines 426-469).

Reviewer #1 (Recommendations for the authors):

1) The C. elegans enzymes described in the intro are still only putative methyltransferases. Most of them, outside of SET-2 have not been shown by in vitro methylation assays so should be referred to as putative methyltransferases (unless the authors wish to perform in vitro methylation assays to complement the in vivo findings of others which could still be indirect). Since so much of the manuscript is focused on H3K9 methylation it would be good to show that some of these enzymes (maybe just SET-21 if it is too much to show for them all) are indeed H3K9 methyltransferases.

Thanks very much for the suggestion.

We completely agree with the reviewer that we lack direct evidence showing that these SET proteins are bona fide H3K9 methyltransferases. We have tried very hard to conduct similar experiments as Schwartz-Orbach et al., has done on SET-32. However, given the covid-19 pandemic and biosafety issue, we could not get access to [3H]labeled S-adenosylmethionine (SAM). We are very sorry that we could not pursue further at this stage. Therefore, we have revised the manuscript saying that these SET proteins are putative histone methyltransferases and are required for the presence of H3K9me1/2.

C. elegans’ genome encodes 10 putative H3K9 methyltransferases, most of which have not been directly characterized for their biochemical properties by in vitro experiments (SCHWARTZ-ORBACH et al., 2020). The presence of these many putative H3K9 methyltransferases may lead to a complexity of the interaction network of histone modifications. In addition, some SET proteins may have multiple roles in mediating histone modification. For example, SET-19 is required for both H3K9me1/2 and H3K23me3 marks, further bringing up complexity of the regulation. Further investigations are necessary to reveal their enzymatic activities by in vitro experiments

(SCHWARTZ-ORBACH et al., 2020). In addition, identifying the in vivo targeted genes will also be crucial to help understanding why and how these SET proteins are involved in lifespan regulation.

We have revised the Discussion section to include the above information (lines 496-506).

2) It's important to see how H3K9 methylation changes in daf-2 worms relative to WT worms, both by WB at gross level and genome wide by ChIPseq. This change in H3K9me1/me2 genome wide should then be compared to daf-16 target gene lists to see if there is an enrichment.

Thanks very much for the comments.

We have included new data comparing the global levels of H3K9me1/2/3 in N2 and daf-2 animals by Western blotting (new Figure 8—figure supplement 1A). The daf2 mutation itself did not significantly change the gross levels of H3K9me1/2/3 status.

We also compared the levels of H3K9me1/2/3, H3K4me1/2/3, H3K27me1/2/3, H3K36me1/2/3 and H3K23me2/3 in N2 and set mutants at L4 and embryonic stages (new Figure 8—figure supplement 1B-C). In N2 background, the mutation of these set genes did not significantly change gross levels of H3K9me1/2/3 either.

However, for technical reasons, we could not successfully conduct ChIP-seq experiments on daf-2 and daf-2;set larva animals.

We speculated that the mutation of daf-2 sensitizes the animals, for example,

increases the cytoplasm-to-nuclear translocation of DAF-16. We have generated a

single-copy, in situ GFP-tagged DAF-16 transgene, crossed to daf-2, set-21, set-25

single mutants and their double mutants, and scored the nuclear localization of DAF16 at 20^oC and 25^oC, respectively (new Figure 5A-B, Figure 5—figure supplement 1B). The mutation of daf-2 increased the nuclear accumulation of DAF-16. The double mutants daf-2;set-21 and daf-2;set-25 further increased the nuclear accumulation of DAF-16.

Interestingly, DAF-16 was nearly 100% localized in the nucleus in both daf-2;set21 and daf-2;set-25 double mutants. Yet daf-2;set-21, but not daf-2;set-25, extended the lifespan of daf-2 animals, suggesting addition regulation required beyond the nuclear localization of DAF-16. Our work suggested that the transcription activity of DAF-16 may preferentially depend on H3K9me1/2, but not H3K9me3.

We have revised the text to include the information (lines 245-275, 357-361).

3) For 6b you have to present all 4 conditions for epistasis experiments. The reader needs to see what the effect of lys-7 deletion is in the daf-2 background by itself. And then 2-way anova needs to be done to test whether the effect of lys-7 in the daf-2 background is different from in the daf-2;set-21 background.

Thanks very much for the suggestions. These are really valuable experiments to understand the mechanism of each putative target gene. However, for the time issue, we are still at the stage to generate multiple alleles and transgenes of nhr-62, sod-3, asm-2, F35E8.7, and Y39G8B.7. Hopefully, we can carefully investigate how and why each gene is involved in lifespan regulation in daf-2;set double mutant animals in the near future.

4) Some attempt should be made to explain why there are 6 redundant H3K9me1/me2 methylases (again I'm not convinced yet that there are because of point 1) that all affect lifespan in the same manner. Are these enzymes expressed in the same tissue? Do they affect H3K9me1/me2 on the same genes? At the same time? Usually it is believed that different enzymes that modify the same mark have some sort of specificity but here they are not seeing that. I don't think it's necessary to perform tissue specific experiments for this initial manuscript but some experimental justification should be provided. Do they believe that it's the amount of heterochromatin DNA in general which is regulating this? Or is it a gene specific effect?

Thanks very much for the comments. We are also eager to know why the 6 different H3K9me1/me2 methylases would have such similar phenotypes in this lifespan extension assay, yet have no clue at this stage.

We have generated in situ single-copy 3xFLAG-GFP tagged SET-21, SET-25 and SET-32 transgenes by CRISPR/Cas9 technology and single-copy GFP-3xFlag-tagged SET-13 transgenic strains using the Mos1-mediated single-copy insertion (MosSCI) technology. These animals were imaged in N2 and daf-2 background at embryonic and larva stages respectively, to figure out the expression patterns and cellular localizations of SET proteins (new Figure 4—figure supplement 1). However, we failed to find a clear correlation between their expression pattern with the lifespan extension phenotype. The depletion of SET-21 and SET-32 extended the lifespan of daf-2 animals, but GFP::SET21 and GFP::SET-32 showed different expression pattern and subcellular localization. GFP::SET-21 exclusively located in the nucleus but GFP::SET-32 located in both the nucleus and cytoplasm in embryos. At larval stage, GFP::SET-21 was expressed in soma but GFP::SET-32 was expressed in germline. Both set-13 and set-25 mutants did not show lifespan extension in N2 and daf-2 background worms. SET-13 and SET-25 were enriched in the nucleus. SET-13::GFP was expressed through 8-cell stage to lateembryonic stage, while SET-25::GFP was expressed in the whole embryonic stage and the germline in young adults.

To identify the target genes in the group of lifespan extension daf-2;set mutants, we re-analyzed the mRNA expression profile in the double mutants via an unbiased method (new Figure 6, Figure 6—figure supplement 1). We have identified 49 coupregulated genes and 11 co-downregulated genes that are specifically enriched in the long-lived double mutants daf-2;set-19, daf-2,set-21, daf-2;set-32, but not in daf-2 and daf-2;daf-25 animals (new Figure 6A-B and Figure 6—figure supplement 1A). Interestingly, among the 49 co-upregulated genes, 27 of them are also DAF-16 Class I genes (new Figure 6—figure supplement 1B). 22 co-upregulated genes are not DAF-16 Class I genes, suggesting the existence of additional regulations.

C. elegans’ genome encodes 10 putative H3K9 methyltransferases, most of which have not been directly characterized for their biochemical properties by in vitro experiments. The presence of these many putative H3K9 methyltransferases may lead to a complexity of the interaction network of histone modifications. In addition, some SET proteins may have multiple roles in mediating histone modifications. For example, SET-19 is required for both H3K9me1/2 and H3K23me3 marks, further bringing up complexity of the regulation. Further investigations are necessary to reveal their enzymatic activities by in vitro experiments. In addition, identifying the in vivo targeted genes will also be crucial to help understanding why and how these SET proteins are involved in lifespan regulation.

We have revised the text to include the information (lines 186-201, 301-320, 496-506).

Reviewer #2 (Recommendations for the authors):

1. As the authors mentioned in the discussion, H3K9me2 is the histone modification in C. elegans that corresponds best with canonical heterochromatin factors like HPL-2 (McMurchy et al., 2017). However, in their results, they focus on the lack of effect on lifespan seen for H3K9me3 modifiers like SET-25 to rule out any effect with heterochromatin. In fact, despite corresponding highly with H3K9me2, one of the marks implicated in further extending the lifespan of their HMT; daf-2 double mutants, hpl-1 and hpl-2 mutations had no effect in a daf-2 background (although they do seem to slightly decrease lifespan, and it would be interesting to see the statistical comparison of a log-rank test). The authors should clarify the relationship between heterochromatin, H3K9me1/2, and H3K9me3.

Thanks very much for the suggestion. Actually, the relationship between heterochromatin, H3K9me1/2, and H3K9me3 is quite intriguing.

H3K9me3 is a hallmark of heterochromatin and is often enriched on repetitive sequences. In addition, H3K9me1/2/3 have all been shown enrichment on repeat sequences. H3K9 methylation and its association with the nuclear envelope help maintain silencing of repetitive sequences and may contribute to the maintenance and stability of heterochromatin-like regions in C. elegans, as observed in other organisms. A hallmark of heterochromatin is the enrichment of HP1. The nematode C. elegans has two HP1 paralogous proteins: HP1 Like 1 and 2 (HPL-1 and HPL-2). Interestingly, HPL-2 may serve additional roles than HPL-1, as hpl-2 mutants display diverse defects while hpl-1 mutants generally lack obvious defects. The pattern of HPL-2 most closely resembles H3K9me1 and H3K9me2 but less closely resembles H3K9me3. HPL-2bound genes, especially those located on arms, possess significantly higher levels of H3K9me2, but show no significant difference in H3K9me3 levels when compared to genes not bound by HPL-2. SET-25 is a putative H3K9me3 methyltransferase and is recruited to targeted genomic loci either by H3K9me2 which is deposited by MET-2 or by the somatic Argonaute protein NRDE-3 and small RNAs. SET-25 colocalizes with both H3K9me3 and HPL-1, but does not colocalize with HPL-2.

H3K9me2 and H3K9me3 exhibit different patterns of repeat enrichment. H3K9me2 is more associated with DNA transposons and satellite repeats, similar to the heterochromatin factors HPL-2, LET-418, LIN-13, LIN-61, and MET-2, whereas H3K9me3 is particularly associated with retrotransposon families, especially LINE and SINE elements. Moreover, H3K9me2 and the five heterochromatin factors are enriched at telomeres, whereas H3K9me3 is not. Therefore, H3K9me2, rather than H3K9me3, may be more closely associated with canonical heterochromatin factors in C. elegans. It was postulated that losing repressive chromatin is detrimental to lifespan. In humans, two premature aging diseases are caused by mutations in lamins that reduce heterochromatin and disrupt its nuclear localization. In C. elegans, Drosophila, and mammals, heterochromatin decreases as individuals grow older. Additionally, mutations that increase repressive chromatin extend lifespan in eukaryotes. In N2 background animals, the depletion of hpl-2 was shown increased lifespan. However, we did not observe a further lifespan extension in daf-2;hpl-2 animals, comparing to the daf-2 mutant. Similarly, we did not observe a significant change in lifespan in daf2;set-25 mutants. The depletion of the two heterochromatin factors HPL-1 and HPL-2 failed to further extend the lifespan, suggesting that heterochromatin itself may not be sufficient to perturb lifespan in daf-2 mutant worms.

We have revised the Discussion section to include the above information (lines 426469).

We have revised the Materials and methods section (lines 549-559) and figure legends to include the information of each lifespan assay. We also included a new Table S1 to summarize all lifespan experiments.

2. I was confused by the statements describing Figure 4C on lines 176-178. Both double and triple mutants were much longer-lived than daf-2 single mutants, rather than having similar lifespan extensions as the daf-2 single mutants. There was even a distinction between the daf-2; set-21 double mutant and the daf-2; set-21; set-2 triple mutant – the presence of a set-2 mutation represses the slight lifespan extension difference seen in the daf-2; set-21 double mutant when compared to the daf-2; set-2 double mutant. The authors should rephrase their description of their results to more accurately convey their findings.

Sorry for the confusion.

We have revised the text as “the double mutants daf-2;set-2 and daf-2;set-21 and

the triple mutants daf-2;set-2;set-21 exhibited similar lifespan extensions relative to that of daf-2 animals (Figure 4C)” (lines 221-223).

3. My major concerns with the approach to transcriptomic analysis were conveyed in the public review, but I include more specific concerns here. Given that Class I genes are an important downstream consequence of DAF-16, it isn't clear why these are primarily unaffected in the daf-2 control condition shown in Figure 5A. Furthermore, the authors focus primarily on the top 50 Class I genes, but there are interesting effects seen in the top 50 Class II genes as well, especially given that mutations in the HMTs seem to suppress some of the upregulation observed in daf-2 control – why exclude these from their interpretations? Additionally, the brevity of the methods makes it hard to tell exactly what comparisons are being drawn – in the heatmaps shown in in Figure 5C, D, does the color refer upreglation in comparison to a wild-type control? Why not compare misregulation to a daf-2 single mutant instead? What does "Top 50" refer to – are these the fifty most-affected in these specific data sets, or are they a gene set identified by an independent metric? Why limit the analysis to only fifty genes? Finally, this analysis would be further strengthened by comparing locus-specific effects to the known locations of H3K9me1/2/3 peaks in wild-type genomes in previously published datasets (like in Li et al., 2021 or McMurchy et al., 2017) or even to H3K9me1/2/3 localization in datasets of relevant HMT mutants like set-32.

Thanks very much for the suggestions.

We used two approaches to analyze the mRNA-seq data.

First, the depletion of DAF-2 reduces the insulin signaling pathway, promotes DAF-16 nuclear translocation, and leads to both upregulation and downregulation of large sets of genes, referred to as Class I and II genes, respectively. Class I genes are induced in daf-2 mutants but are repressed in daf-2;daf-16 double mutants. Class II genes are not induced in daf-2 mutants but are induced in daf-2; daf-16 double mutants. 1663 genes are classified as positive (class I) DAF-16 targets and 1733 genes are classified as negative (class II) DAF-16 targets of DAF-16. Class I genes are enriched for the Gene Ontology categories including oxidation, reduction, and energy metabolism, whereas class II genes are enriched for genes involved in biosynthesis, growth, reproduction, and development. We performed mRNA-seq and analyzed Class I and II DAF-16 genes to identify the mis-regulated genes in the daf-2;set mutants. Interestingly, the mRNA levels of DAF-16 Class I, but not Class II, genes are consistently activated in long-lived daf-2;set-19, daf-2;set-21 and daf-2;set-32 worms, than in the control daf-2 and daf-2;set-25 animals (Figure 5E-F and new Figure 5—figure supplement 2).

Second, to identify the target genes in the group of lifespan extension daf-2;set mutants, we re-analyzed the mRNA expression profile in the double mutants via an unbiased method (new Figure 6, Figure 6—figure supplement 1). We have identified 49 co-upregulated genes and 11 co-downregulated genes that are specifically enriched in the long-lived double mutants daf-2;set-19, daf-2,set-21, daf-2;set-32, but not in daf-2 and daf-2;daf-25 animals (Figure 6A-B and Figure 6—figure supplement 1A). Interestingly, among the 49 co-upregulated genes, 27 of them are also DAF-16 Class I genes (new Figure 6—figure supplement 1B). 22 co-upregulated genes are not DAF-16 Class I genes, suggesting the existence of additional regulations.

We then analyzed a number of known transcription factors for their binding to the co-regulated targeted genes (new Figure 6C-D, Figure 6—figure supplement 1C-E). Among them, we found that DAF-16 and NHR-80 were specifically enriched at the transcription start sites (TSS) of the 49 co-upregulated genes. NHR-80 is a homolog of mammalian hepatocyte nuclear factor 4 and is an important nuclear hormone receptor involved in the control of fat consumption and fatty acid composition in C. elegans. Among the genes targeted by DAF-16 and NHR-80, twelve of them are co-regulated by both factors, which is consistent with previous report that daf-16 and nhr-80 function in parallel pathway for lipid metabolism.

We revised the text to include this information (lines 284-320).

4. How were the 10 Class I DAF-16 target genes chosen? It doesn't seem like they were in the top 50 (or not that I could easily see). Furthermore, why were no Class II genes chosen? The authors should clarify how they chose to examine these specific candidates, and if necessary, extend the analysis for other targets to see whether the effects are specific to Class I targets only (as they argue in their model). Furthermore, each of these candidate genes needs to be examined in a daf-2 mutant background to fully interpret the meaning of the suppression of long-lived daf-2; set-21 mutants.

Thanks very much for the suggestion.

To identify the target genes in these group of lifespan extension daf-2;set mutants, we analyzed the mRNA expression profile in the double mutants via an unbiased method (new Figure 6, Figure 6—figure supplement 1). We have identified 49 coupregulated genes and 11 co-downregulated genes that are specifically enriched in the long-lived double mutants daf-2;set-19, daf-2,set-21, daf-2;set-32, but not in daf-2 and daf-2;daf-25 animals (Figure 6A-B and Figure 6—figure supplement 1A). Interestingly, among the 49 co-upregulated genes, 27 of them are also DAF-16 Class I genes (new Figure 6—figure supplement 1B). 22 co-upregulated genes are not DAF-16 Class I genes, suggesting the existence of additional regulations.

To confirm the change in target gene expression, we chose 10 genes from the 49 co-upregulated genes. Among them, nhr-62, dao-3 and tts^-1 are direct binding targets of both DAF-16 and NHR-80. sod-3 is a DAF-16 direct binding target. lys-7 and asm2 are NHR-80 direct binding targets. All of the 10 genes have been shown to be upregulated in daf-2 compared to daf-2;daf-16 animals. We quantified the mRNA levels by quantitative real-time PCR (qRT–PCR) in daf-2, daf-2;set-21 and daf-2;set-25 animals (Figure 7A). Consistently, all the 10 genes were up-regulated in daf-2;set-21 double mutants than in daf-2 and daf-2;set-25 animals (Figure 7A).

We are very interested in how and why each gene is involved in lifespan regulation in daf-2;set double mutant animals, we are trying to generate multiple alleles and transgenes for nhr-62, sod-3, asm-2, F35E8.7, and Y39G8B.7. However, for the time issue, we are still at the very early stage to generate more lines. Hopefully, we can answer this question in the near future.

5. The authors claim that loss of H3K9me1/2 increases binding of DAF-16 at its target genes (line 282). There is no evidence to support this claim (and a lack of evidence for this claim, if we take the DAF-16 localization data in Figure S3 at complete face value). This could be part of their proposed model (outlined in Figure S7), but as written, it reads as if it were a summary of work presented in the paper. The authors should clarify their writing to ensure full accuracy for what their data can support.

Thanks very much for the comments. We completely agree with the reviewer that we lack direct data showing that these set mutations increased the binding activity of DAF-16 to target genes by ChIP analysis.

We speculated that the mutation of daf-2 sensitizes the animals, for example, increases the cytoplasm-to-nuclear translocation of DAF-16. We have generated a single-copy, in situ GFP-tagged DAF-16 transgene, crossed to daf-2, set-21, set-25 single mutants and their double mutants, and scored the nuclear localization of DAF16 at 20^oC and 25^oC, respectively (new Figure 5A-B, Figure 5—figure supplement 1B). The mutation of daf-2 increased the nuclear accumulation of DAF-16. The double mutants daf-2;set-21 and daf-2;set-25 further increased the nuclear accumulation of DAF-16.

Interestingly, DAF-16 was nearly 100% localized in the nucleus in both daf-2;set21 and daf-2;set-25 double mutants. Yet daf-2;set-21, but not daf-2;set-25, extended the lifespan of daf-2 animals, suggesting addition regulation required beyond the nuclear localization of DAF-16. Our work suggested that the transcription activity of DAF-16 may preferentially depend on H3K9me1/2, but not H3K9me3.

This result is unlike a previous widely used multi-copied DAF-16::GFP transgene. In the multi-copied DAF-16::GFP transgene animals, DAF-16::GFP localized in the cytoplasm and a daf-2 mutation completely re-localizes DAF-16::GFP to the nucleus (Figure 5—figure supplement 1A). This result suggests further investigations are required to understand the mechanism of DAF-2/DAF-16 axis in lifespan regulation.

We have revised the text to include the information (lines 357-361).

We also revised the working model in the Discussion section (lines 391-396) as:

“These putative histone methyltransferases are required for H3K9me1/2 modification and regulate DAF-16 Class I genes. In the absence of H3K9me1/2, the DAF-16 accumulates in the nucleus and the expression of DAF-16 Class I genes increase, which promote the expression of longevity genes and anti-stress genes and extends lifespan of daf-2 animals (Figure 8—figure supplement 4D).”

6. Throughout the manuscript, it's difficult to build a cohesive understanding of the functional effects of the HMT mutants. In part, this is because the paper investigates several HMTs without discussing their specific effects (in certain tissues or cell types, or at specific stages). The analyses are sometimes too broad (immunoblots of total histone modifications) and at other times too narrow (focusing only on 50 Class I DAF-16 target genes). Given that H3K9me1/2 will likely have locus specificity, it is unsatisfying to leave the analysis at the global level of immunoblots (although it is very compelling to see global effects in their mutants). Including a deeper, and unbiased, analysis of their transcriptomic data will help to bridge the gap between the global and the locus-specific, and may even point out other interesting avenues that will further bolster their claims about potential mechanism (or open new avenues of study).

Thanks very much for the suggestions.

We have generated in situ single-copy 3xFLAG-GFP tagged SET-21, SET-25 and SET-32 transgenes by CRISPR/Cas9 technology and single-copy GFP-3xFLAG-tagged SET-13 transgenic strains using the Mos1-mediated single-copy insertion (MosSCI) technology. These animals were imaged in N2 and daf-2 background at embryonic and larva stages respectively, to figure out the expression patterns and cellular localizations of SET proteins (new Figure 4—figure supplement 1). However, we failed to find a clear correlation between their expression pattern with the lifespan extension phenotype. The depletion of SET-21 and SET-32 extended the lifespan of daf-2 animals, but GFP::SET21 and GFP::SET-32 showed different expression patterns and subcellular localizations. GFP::SET-21 exclusively located in the nucleus but GFP::SET-32 located in both the nucleus and cytoplasm in embryos. At larval stage, GFP::SET-21 was expressed in soma but GFP::SET-32 was expressed in germline. Both set-13 and set-25 mutants did not show lifespan extension in N2 and daf-2 background worms. SET-13 and SET-25 were enriched in the nucleus. SET-13::GFP was expressed through 8-cell stage to lateembryonic stage, while SET-25::GFP was expressed in the whole embryonic stage and the germline in young adults.

To identify the target genes in these group of lifespan extension daf-2;set mutants, we have analyzed the mRNA expression profile in the double mutants via an unbiased method (new Figure 6, Figure 6—figure supplement 1). We have identified 49 coupregulated genes and 11 co-downregulated genes that are specifically enriched in the long-lived double mutants daf-2;set-19, daf-2,set-21, daf-2;set-32, but not in daf-2 and daf-2;daf-25 animals (new Figure 6A-B and Figure 6—figure supplement 1A). Interestingly, among the 49 co-upregulated genes, 27 of them are also DAF-16 Class I genes (new Figure 6—figure supplement 1B). 22 co-upregulated genes are not DAF-16

Class I genes, suggesting the existence of additional regulations.

C. elegans’ genome encodes 10 putative H3K9 methyltransferases, most of which have not been directly characterized for their biochemical properties by in vitro experiments. The presence of these many putative H3K9 methyltransferases may lead to a complexity of the interaction network of histone modifications. In addition, some SET proteins may have multiple roles in mediating histone modification. For example, SET-19 is required for both H3K9me1/2 and H3K23me3 marks, further bringing up complexity of the regulation. Further investigations are necessary to reveal their enzymatic activities by in vitro experiments. In addition, identifying the in vivo targeted genes will also be crucial to help understanding why and how these SET proteins are involved in lifespan regulation.

We have revised the text to include the information (lines 186-201, 301-320, 496506).

[Editors’ note: what follows is the authors’ response to the second round of review.]

The manuscript has been improved but there are some remaining issues that need to be addressed, as outlined below:

Primarily, the three areas that remain an issue before submission are (1) a second replicate needs to be included (and ideally, performed blinded) for experiments such as Figures 1B, 4A, and 4B,

Thanks very much for the suggestions.

We have included a blinded replicate to compare the lifespan of daf-2 vs daf-2;set-21 mutants (new Figure 1A). We have also conducted new experiments to compare N2 vs all set-21 alleles (new Figure 1A). We did not found significant lifespan difference between N2 and all set-21 alleles, but daf-2;set-21 double mutants strongly extended the lifespan of daf-2 worms.

We have included blinded replicates for experiments in Figures 4A and 4B. The conclusions were unchanged.

2) the appropriate epistasis controls should be included as requested by reviewer 1 to properly interpret the experiments (e.g., Figure 7),

Thanks very much for the comments. We have included new epistasis control experiments and data analysis for Figures 1A, 1B, 3-S3, and 7B. The conclusions were unchanged.

and (3) the text needs to be modified where precision remains an issue, as described by reviewer 1.

Thanks very much for the comments. We have revised the manuscript following reviewer #1’ s suggestions to state that the lifespan extension effect is likely more specific to daf-2(-) background animals.

Reviewer #1 (Recommendations for the authors):

1) As stated last review (major point #3), in epistasis experiments (such as WT and eat-2 +/- set-21 and WT and daf-2 +/- met-2, and Figure 3C +D, 7B) you need to include all 4 conditions in the experiment to make any conclusions. For example, you can't do the set-21 vs WT at a different time and assume that the set-21 is still going to have no effect in that particular experiment. There might be some unknown factor (food, temperature, etc) that makes set-21 actually have an effect in the WT worms in that experiment which would mean a very different conclusion. Then after the experiment is done you perform a 2 way ANOVA analysis to reveal whether set-21 has a significantly different effect in the WT context or the mutant context. This comment still holds true for asm-2, Y39G8B.7, etc… that I had asked for last review. You do not need to perform all possible combinations of these 3 genes but at the very least daf-2, daf-2;set-21, daf-2;asm-2 and daf-2;set-21;asm-2 if you want to make any conclusion about asm-2 relative to set-21 in the daf-2 background), examining 3 lines in an epistasis experiment rather than 4 is not really informative.

Thanks very much for the suggestions.

­

We have conducted new experiments to compare N2 vs all set-21 alleles (new Figure 1B). We did not found significant lifespan difference between N2 and all set-21 alleles, but daf-2;set-21 double mutants strongly extended the lifespan of daf-2 worms.

In new Figure 1C, we also included N2 vs set-21(ust68). We did not found significant lifespan difference between N2 and set-21(ust68) mutants, but eat-2;set-21 double mutants strongly extended the lifespan of eat-2 worms.

We conducted new experiments to compare the stress resistance of N2 vs set mutants (new Figure 3-S3A-B). We did not find strong difference of stress resistance between N2 vs set mutants, yet there are significant difference of stress resistance between daf-2 vs daf-2;set mutants (new Figure 3D-E).

We included new replicates for Figure 4A-B. The conclusions are not changed.

We conducted new experiments to compare the lifespan of daf-2, daf-2;y39g8b.7, daf-2;set-21 double mutants vs daf-2;set-21;y39g8b.7 triple mutant and etc… (new Figure 7B). Although we did not find significant lifespan difference between daf-2 vs daf-2;y39g8b.7, daf-2;asm-2, daf-2;nhr-62 double mutants, the mutations of y39g8b.7, asm-2 and nhr-62 could partially reverted the lifespan extension of daf-2;set-21 in triple mutants. Actually, none of the ten selected targeted genes could shorten the lifespan of daf-2 animals but seven of them specifically shorten the lifespan in triple mutants.

2) The text needs to be rewritten to reflect that this is all only occurring in daf-2 mutant worms and not in WT worms. This rewriting needs to start with the title ("…limits the lifespan of daf-2 mutant C. elegans". ) and work its way thru the intro, results and discussion (where the authors claim H3K9me regulators should be considered for lifespan regulation). Need to include explicit writing that deletion of all these putative enzymes has only modest effects on H3K9me and lifespan in the WT context and their role is only revealed in the daf-2 mutants. Need to speculate about what that is. What makes the daf-2 mutation special to reveal a regulatory role for these genes?

Thanks very much for the suggestions.

We have revised the manuscript as suggested, stating that these SET proteins and H3K9me2 may preferentially limit the lifespan of daf-2 mutant worms.

The depletion of SET proteins or the inhibition of worm H3K9me1/2 methyltransferases by G9a inhibitor significantly extended the lifespan only in daf-2 mutant worms, but has modest effects on N2 animals. Consistently, the long-lived set-6, set-19, set-20, set-21, set-32 and set-33 mutants only significantly decreased global H3K9me1/2 levels in daf-2 mutants but not in N2 animals. These results together suggested that DAF-2 limits the effects of these SET proteins.

It is unclear how and why these SET proteins only limit lifespan in daf-2 animals. There are three possibilities. First, DAF-2 may have profound effect on the chromatin accessibility. In daf-2 mutants, the chromatin may be more compact and prohibit the activity of the SET proteins to conduct H3K9me2 modification. Second, daf-2 mutation may sensitize the animals by increasing the nuclear accumulated DAF-16. Although it is unclear whether DAF-16 preferentially recognizes H3K9me2 or H3K9me3, we observed a much stronger enrichment of DAF-16 in the nucleus in daf-2;set-21 than in daf-2 animals, which suggested that H3K9me2 may prohibit the accumulation of DAF-16 in the nucleus. Third, DAF-2 may have other indirect effect on the SET proteins. Further chromatin-immunoprecipitation experiments can be used to test whether daf-2 mutation could affect the association of the SET proteins with chromatin. The ATAC-seq technology may allow us to investigate whether daf-2 mutation have changed chromosome accessibility to facilitate the activity of SET proteins. In addition, identification of the substrate of SET proteins may help to elucidate their mechanism and functions.

In daf-2 animals, DAF-16 modestly accumulated in the nucleus, as shown by the newly generated single-copy GFP-tagged DAF-16 transgene. Yet in daf-2;set-21 double mutants, DAF-16 was strongly enriched in the nucleus. The depletion of DAF-16 completely blocked the synergistic lifespan extension in daf-2;set-21 animals. The mRNA-seq experiments revealed that the mRNA levels of DAF-16 class I genes were elevated in the long-lived daf-2;set-19, daf-2;set-21, and daf-2;set-32 double mutants compared to those in control daf-2 and short-lived daf-2;set-25 mutants. Together, these data suggested that the activity to DAF-16 to class I genes not only depends on the absence of insulin pathway, but also depends on the depletion of these SET proteins and H3K9me2. The single-copy GFP-tagged DAF-16 transgene may allow us to re-visit the mechanism of lifespan regulation pathways.

We have revised the Discussion section to include the above information (lines 435-468).

Reviewer #2 (Recommendations for the authors):

Upon resubmission, this paper is greatly improved – readers now have the evidence needed to evaluate the paper's main claim that H3K9 HMTs synergistically extend lifespan in daf-2 mutants, in part via DAF-16, but also via. The additional experiments, along with a more tempered approach to discussing their findings, significantly strengthen the paper.

However, several of our concerns were not addressed by these changes, nor in the rebuttal. For example, Reviewer 2 asked that some of the key lifespan experiments be replicated at least once and also performed blinded – there are several figures where only a single replicate is shown. Key controls are still missing, like in Figure 7, where Reviewer 1 had asked for appropriate controls to interpret the epistasis experiments for example, set-21 single mutants and daf-2; lys-7 double mutants. The authors' rebuttal mentioned generating new alleles for some genes, but providing appropriate single mutant and double mutant controls should require no new strains to be built (since the strains likely already exist if the authors generated the triple mutant).

Thanks very much for the comments.

We conducted new experiments to compare the lifespan of daf-2, daf-2;y39g8b.7, daf-2;set-21 double mutants vs daf-2;set-21;y39g8b.7 triple mutant and etc… (new Figure 7B). Although we did not find significant lifespan difference between daf-2 vs daf-2;y39g8b.7, daf-2;asm-2, daf-2;nhr-62, the mutations of y39g8b.7, asm-2 and nhr-62 could partially reverted the lifespan extension of daf-2;set-21 in triple mutants. Actually, none of the ten selected targeted genes could shorten the lifespan of daf-2 animals but seven of them specifically shorten the lifespan in triple mutants.

We also included new blinded replicates for Figures 1A, 4A-B and 7B. The conclusions are not changed.

The extended analysis of the mRNA-seq is better (and opens up an interesting line of questioning for future studies!). However, the authors failed to address most of Reviewer 2's Major Point 3 about their transcriptomic analysis, nor did they include further details about how the analysis was done. The methods for the mRNA-seq will need to be more detailed. What is their unbiased approach? What is the differential expression shown in the Figure 5? For example, I had previously asked: "Does the color refer to upregulation in comparison to a wild-type control? Why not compare misregulation to a daf-2 single mutant instead?" and the answer is still not clear. Furthermore, if Class I genes are defined as being induced in daf-2 mutants, why are they downregulated in Figure 5E? And if Class II genes are defined as being not induced in daf-2 mutants, why are they upregulated in Figure 5F? I may be misunderstanding the gene classifications, but it is a demonstration of how the text is unclear. Overall, I found the new transcriptomic analyses are compelling, but more care needs to be taken to fully explain what was done.

Thanks very much for the comments.

We have analyzed the mRNA-seq data in two ways. The first one is comparing the expression levels of reported DAF-16 target genes in daf-2 and daf-2;daf-25 animals and in the long-lived daf-2;set-19, daf-2,set-21, daf-2;set-32 double mutants (Figure 5E and Figure 5—figure supplement 2A). The second method is directly comparing the gene expression in short-lived daf-2;daf-25 animals and the long-lived daf-2;set-19, daf-2,set-21, daf-2;set-32 double mutants to the gene expression in daf-2 animals (Figure 6A and Figure 6—figure supplement 1A). This method identified 49 co-upregulated genes and 11 co-downregulated genes that are specifically enriched in the long-lived double mutants but not in control daf-2 and daf-2;daf-25 animals. We then analyzed a number of known transcription factors for their binding to the co-regulated targeted genes.

Figure 5E showed heatmap of the standardized FPKM of reported top 50 DAF-16 Class I and Class II genes by mRNA-seq in the indicated animals. Figure 5—figure supplement 2A showed heatmap of the standardized FPKM for reported 1576 DAF-16 Class I and 1653 Class II genes by mRNA-seq in the indicated animals. Statistical analysis was performed to obtain the expression levels of each gene in each strain and the average of the expression levels in the five strains. The gene expression levels in a single strain were compared with the average of the five strains, and the ratio obtained was processed by log2. The resulting value > 0 means the expression level of the gene in the indicated strain is higher than the average expression levels of this gene in the five strains, as shown from yellow to red. The resulting value < 0 means the expression level of the gene in the indicated strain is lower than the average expression levels of this gene in the five strains, as shown in blue. The expression levels are indicated by the color bar. Most of the DAF-16 Class I genes are yellow and red in the long-lived daf-2;set-19, daf-2,set-21, daf-2;set-32 double mutants, but blue in daf-2 and daf-2;daf-25. The mRNA levels of DAF-16 Class I genes are consistently elevated in long-lived daf-2;set-19, daf-2;set-21 and daf-2;set-32 worms, but not in the control daf-2 and daf-2;set-25 animals.

We have revised the figure legends to describe our data analysis in more detail (lines 1118-1127, 1225-1234).

Author details

1. Meng Huang

   The USTC RNA Institute, Ministry of Education Key Laboratory for Membraneless Organelles & Cellular Dynamics, Department of Obstetrics and Gynecology, The First Affiliated Hospital of USTC, School of Life Sciences, Division of Life Sciences and Medicine, Biomedical Sciences and Health Laboratory of Anhui Province, University of Science and Technology of China, Hefei, China

   Contribution

   Conceptualization, Formal analysis, Investigation, Visualization, Methodology, Writing - original draft, Project administration

   Contributed equally with

   Minjie Hong

   Competing interests

   No competing interests declared

2. Minjie Hong

   The USTC RNA Institute, Ministry of Education Key Laboratory for Membraneless Organelles & Cellular Dynamics, Department of Obstetrics and Gynecology, The First Affiliated Hospital of USTC, School of Life Sciences, Division of Life Sciences and Medicine, Biomedical Sciences and Health Laboratory of Anhui Province, University of Science and Technology of China, Hefei, China

   Contribution

   Formal analysis, Validation, Investigation, Visualization, Methodology

   Contributed equally with

   Meng Huang

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-1510-6396

3. Xinhao Hou

   The USTC RNA Institute, Ministry of Education Key Laboratory for Membraneless Organelles & Cellular Dynamics, Department of Obstetrics and Gynecology, The First Affiliated Hospital of USTC, School of Life Sciences, Division of Life Sciences and Medicine, Biomedical Sciences and Health Laboratory of Anhui Province, University of Science and Technology of China, Hefei, China

   Contribution

   Software, Formal analysis, Visualization

   Competing interests

   No competing interests declared

4. Chengming Zhu

   The USTC RNA Institute, Ministry of Education Key Laboratory for Membraneless Organelles & Cellular Dynamics, Department of Obstetrics and Gynecology, The First Affiliated Hospital of USTC, School of Life Sciences, Division of Life Sciences and Medicine, Biomedical Sciences and Health Laboratory of Anhui Province, University of Science and Technology of China, Hefei, China

   Contribution

   Data curation, Supervision

   Competing interests

   No competing interests declared

5. Di Chen

   State Key Laboratory of Pharmaceutical Biotechnology and MOE Key Laboratory of Model Animals for Disease Study, Model Animal Research Center, Institute for Brain Sciences, Nanjing University, Nanjing, China

   Contribution

   Resources, Methodology

   For correspondence

   chendi@nju.edu.cn

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-0514-7947

6. Xiangyang Chen

   The USTC RNA Institute, Ministry of Education Key Laboratory for Membraneless Organelles & Cellular Dynamics, Department of Obstetrics and Gynecology, The First Affiliated Hospital of USTC, School of Life Sciences, Division of Life Sciences and Medicine, Biomedical Sciences and Health Laboratory of Anhui Province, University of Science and Technology of China, Hefei, China

   Contribution

   Data curation, Supervision, Writing - review and editing

   For correspondence

   xychen91@ustc.edu.cn

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-6530-7191

7. Shouhong Guang

   1. The USTC RNA Institute, Ministry of Education Key Laboratory for Membraneless Organelles & Cellular Dynamics, Department of Obstetrics and Gynecology, The First Affiliated Hospital of USTC, School of Life Sciences, Division of Life Sciences and Medicine, Biomedical Sciences and Health Laboratory of Anhui Province, University of Science and Technology of China, Hefei, China
   2. CAS Center for Excellence in Molecular Cell Science, Chinese Academy of Sciences, Hefei, China

   Contribution

   Conceptualization, Resources, Supervision, Funding acquisition, Methodology, Writing - original draft, Project administration, Writing - review and editing

   For correspondence

   sguang@ustc.edu.cn

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-7700-9634

8. Xuezhu Feng

   The USTC RNA Institute, Ministry of Education Key Laboratory for Membraneless Organelles & Cellular Dynamics, Department of Obstetrics and Gynecology, The First Affiliated Hospital of USTC, School of Life Sciences, Division of Life Sciences and Medicine, Biomedical Sciences and Health Laboratory of Anhui Province, University of Science and Technology of China, Hefei, China

   Contribution

   Supervision, Project administration, Writing - review and editing

   For correspondence

   fengxz@ustc.edu.cn

   Competing interests

   No competing interests declared

• 1,529

  Page views

• 434

  Downloads

• 4

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.