Introduction

The nucleosome, composed of DNA wrapped around histone proteins (H2A, H2B, H3, and H4), is the fundamental structural unit of chromatin. Post-translational modifications (PTMs), such as methylation, acetylation, and phosphorylation, occur on both histone tails and core regions ¹. Histone modifications function as epigenetic marks that regulate chromatin organization and stability, as well as DNA replication, repair, and transcription.

Chromatin is broadly divided into euchromatin and heterochromatin, which differ in their degrees of compaction and transcriptional activity. Histone PTMs such as acetylation, H3K4me3, and H3K36me3 are typically enriched in transcriptionally active euchromatic regions and are associated with gene activation. In contrast, repressive marks including H3K9me3 and H3K27me3 are preferentially enriched in transcriptionally silent heterochromatic regions. The distinct genomic distributions of these modifications reflect their diverse regulatory functions, and the dynamic regulation of these marks across tissues and developmental stages contributes to cell fate determination, fertility, development, and epigenetic inheritance ^2-6.

Histone PTMs are dynamically regulated by specific writers, readers, and erasers, which establish, recognize, and remove these modifications, respectively ^1,7. Writers responsible for histone methylation are histone methyltransferases (HMTs), which catalyze the methylation of lysine (K) and arginine (R) residues on histone proteins. Lysine residues can be mono-, di-, or trimethylated by lysine methyltransferases (KMTs), whereas arginine residues can be mono- or dimethylated by arginine methyltransferases (RMTs) ⁸. Most KMTs harbor a conserved SET domain that confers catalytic activity and substrate specificity ^7,9-11.

Histone H3 lysine 23 methylation (H3K23me) is widely conserved across eukaryotic species ^12-24. In organisms including Tetrahymena, C. elegans, and mammals, evidence from immunostaining, chromatin immunoprecipitation (ChIP)-based analyses, and binding studies with HP1 family proteins supports an association of H3K23 methylation with heterochromatic regions ^{14,15,17,18,20,21,24-26}. In Tetrahymena, the loss of the H3K23 methyltransferase Ezl3p leads to aberrant targeting of meiosis-induced DNA double-strand breaks to heterochromatin and compromises progeny viability ¹⁷. In C. elegans, H3K23me3 shares genomic features with classical repressive marks such as H3K9me3 and H3K27me3 and is also associated with transgenerational epigenetic silencing and accumulation at RNAi target loci ^25-30. The loss of the corresponding histone methyltransferases for H3K9me3 and H3K27me3 results in varying degrees of RNAi defects ^30-33. Two methyltransferases, SET-32 and SET-21, have been reported to contribute to H3K23 methylation ^25,26, but no obvious global reduction in H3K23 methylation has been detected in vivo upon loss of either enzyme. Nevertheless, the enzymatic basis, tissue-specific regulation, and biological functions of H3K23 methylation remain far less well characterized than those of other histone methylation marks. The C. elegans genome encodes 38 SET domain-containing KMTs ^34-45, many of which remain functionally uncharacterized. Our previous work showed that the loss of set-19 leads to a significant reduction in H3K23me3 levels in vivo ⁴⁶, implicating SET-19 as a candidate H3K23 methyltransferase.

In this work, we investigated the role of SET-19 in H3K23 methylation in C. elegans and showed that SET-19 functions as a somatic H3K23 methyltransferase required for H3K23me3 methylation in vitro and in vivo. This work provides insights into how H3K23 methylation is established and supports the idea that distinct methyltransferases may function in different tissues or cell types to promote proper development.

Results

SET-19 is required for H3K23 methylation in C. elegans

In C. elegans, the biological functions of many SET domain-containing methyltransferases remain poorly understood ^25,26,34-44. SET-19 harbors a conserved SET domain (Fig. 1a). Our previous work showed that the set-19(ok1813) mutant exhibited reduced levels of H3K23me2 and H3K23me3, as detected by western blotting ⁴⁶. However, the role of SET-19 in H3K23 methylation required further validation.

Fig. 1.

To assess the impact of set-19 loss on histone H3 methylation, we quantified global H3 methylation levels in the set-19(ok1813) mutant by mass spectrometry at the L3–L4 larval and embryonic stages. No significant changes were observed in the methylation levels of H3K4, H3K9, H3K27, or H3K36 compared with those in wild-type animals (Fig. 1b–d and Fig. S1). In contrast, H3K23 methylation was markedly reduced in the set-19(ok1813) mutant (Fig. 1b–d and Fig. S1). Specifically, H3K23me3 levels decreased from approximately 33.8% to 5.4% at the larval stage and from 10% to 8% at the embryonic stage (Fig. 1d and Fig. S1f). Mass spectrometry revealed no significant change in H3K23me2, whereas H3K23me1 levels were slightly increased at the L3–L4 larval stage (Fig. 1d), and a slight reduction of H3K9me2 was detected in embryos (Fig. S1e).

To further confirm the role of SET-19, we generated an independent set-19 allele using CRISPR/Cas9-mediated genome editing. This allele, set-19(ust644), precisely deletes the SET domain without disrupting the open reading frame (ORF) (Fig. 1a). Western blotting analysis revealed that both set-19(ok1813) and set-19(ust644) mutants exhibited a pronounced reduction in H3K23me3 and a detectable decrease in H3K23me2 at the larval stage (Fig. 1e and Fig. S2a). Two independent antibodies specific for H3K23me2 and H3K23me3, respectively, were used in the western blotting assays. Importantly, reintroduction of a GFP::SET-19 transgene (fib-1p::gfp::set-19, ustIS338) into the set-19 mutant background restored H3K23me2 and H3K23me3 levels (Fig. 1f and Fig. S2b, c). Consistent with previous reports ^25,26, the loss of set-32 did not lead to a detectable reduction in H3K23 methylation in western blotting assays (Fig. 1e and Fig. S2a).

Together, these results indicate that SET-19 is required for H3K23 methylation, particularly H3K23me3 in C. elegans.

Recombinant SET-19 catalyzes histone H3K23 methylation in vitro

SET-19 contains a central SET domain and a nearby coiled-coil region, with intrinsically disordered regions (IDRs) at both the N- and C-termini (Fig. 2a). To test whether SET-19 is a bona fide methyltransferase, we performed in vitro histone methylation assays. Full-length SET-19 and SET-19 domain fragments were fused to GST, expressed in E. coli, and purified using glutathione beads. Although full-length SET-19 could not be successfully expressed in E. coli and was therefore excluded from subsequent analyses, we obtained purified GST-fused proteins corresponding to the SET domain alone (SET-19 SET; amino acids 37–318) and a fragment containing both the SET and coiled-coil domains (SET-19 SET+CC; amino acids 37–454) (Fig. 2a). SET-32 has previously been reported to exhibit H3K23 methyltransferase activity in vitro and was therefore included as a positive control ²⁵. Given the high evolutionary conservation of histone H3 and the conserved nature of lysine 23 across species (Fig. S3a), human histone H3 (Active Motif, #31294) was used as the methylation substrate. Purified GST-SET-19 fusion proteins or full-length GST-SET-32 were incubated with unmodified H3 in the presence of the methyl donor S-adenosylmethionine (SAM). The reaction products were analyzed by western blotting using a panel of antibodies recognizing distinct H3 methylation marks.

Fig. 2.

In the in vitro methylation assays, neither the SET-19 fragments nor full-length SET-32 induced detectable changes in H3K4me3, H3K9me3, H3K27me3, or H3K36me3 levels (Fig. 2b and Fig. S3b). In contrast, the SET-19 SET+CC fragment, but not the SET domain alone, induced a pronounced increase in H3K23me1, H3K23me2, and H3K23me3 signals (Fig. 2b and Fig. S3b), indicating that SET-19 specifically methylates H3K23 in vitro and that this activity requires sequences outside the SET domain, which likely helps the folding of the SET domain. SET-32 exhibited modest H3K23me1 methyltransferase activity, but did not appreciably catalyze H3K23me2 or H3K23me3 under the same conditions (Fig. 2b and Fig. S3b).

SET domain-containing methyltransferases differ not only in substrate specificity but also in their sensitivity to SAM concentration ^39,47-50. We then examined the enzymatic activities of GST-SET-19 SET+CC and full-length GST-SET-32 across a range of SAM concentrations. Increasing SAM levels enhanced H3K23me1/2/3 signals for both enzymes; however, SET-19 SET+CC consistently exhibited substantially higher catalytic activity than SET-32 (Fig. 2c and Fig. S3c). No methylation at other lysine residues, including H3K4, H3K9, H3K27, or H3K36, was detected under any of the tested conditions (Fig. 2c and Fig. S3c). These results indicate that although both SET-19 and SET-32 are capable of catalyzing H3K23 methylation in vitro, SET-19 shows markedly higher enzymatic activity, consistent with the western blotting analyses (Fig. 1e).

Notably, although SET-19 catalyzed mono-, di-, and trimethylation of H3K23 in vitro, both mass spectrometry and western blotting analyses revealed a predominant reduction in H3K23me3 upon loss of set-19. Together, these in vitro and in vivo data establish SET-19 as an H3K23 methyltransferase in C. elegans and indicate that SET-19 plays a prominent role in H3K23 trimethylation.

SET-19 is required for genomic H3K23me3 occupancy and gene repression

Previous studies using ChIP-seq and immunofluorescence staining have shown a strong association between H3K23 methylation (H3K23me2/3) and heterochromatin marks ^{17,18,20,21,24-26}. In C. elegans, the genome-wide distribution of H3K23me3 closely resembles those of H3K9me3 and H3K27me3 (Fig. S4a). C. elegans chromosomes are organized into broad domains, with active marks (H3K4me3 and H3K36me3) enriched in central regions and repressive marks (H3K9me1/2/3) enriched in distal arms ^34,51-53. Consistent with this chromosomal organization, H3K23me3-enriched regions were predominantly localized to chromosome arms (Fig. S4b). In addition, regions centered on H3K23me3 peaks show stronger enrichment of H3K9me3 and H3K27me3 than of the active marks H3K4me3 and H3K36me3 (Fig. 3a).

Fig. 3.

To examine whether SET-19 is required for H3K23me3 occupancy across the genome, we compared H3K23me3 ChIP-seq profiles between wild-type and set-19 mutant animals. The loss of set-19 led to a pronounced global reduction in H3K23me3 levels across the genome (Fig. 3b, c and Fig. S4c). Differential binding analysis using DiffBind ⁵⁴ revealed that 55.4% of H3K23me3 peaks were significantly reduced in the set-19 mutant relative to wild-type animals (Fig. S4d). Notably, residual H3K23me3 signals remained in the set-19 mutant (Fig. 3b, c and Fig. S4c, d), suggesting additional H3K23 methyltransferases besides SET-19 may also contribute to H3K23 methylation.

H3K23me3 has been shown to be required for transcriptional repression of nuclear RNAi-targeted genes ^25,26. To investigate the contribution of SET-19-dependent H3K23me3 to gene regulation, we performed mRNA-seq analysis in wild-type N2 and set-19 mutant animals. Genes were grouped into five categories based on their expression levels in wild-type N2 animals, and then H3K23me3 levels were quantified for each group. H3K23me3 levels showed a strong negative correlation with gene expression, with lowly expressed genes exhibiting higher H3K23me3 occupancy (Fig. 3d, e). Transcriptome analysis revealed that 520 genes were significantly upregulated and 130 genes were downregulated in the set-19 mutant compared with wild-type animals (Fig. 3f). Importantly, the upregulated genes displayed reduced H3K23me3 levels in the set-19 mutant (Fig. 3g), indicating that the loss of SET-19-dependent H3K23me3 is associated with transcriptional derepression.

Collectively, these results indicate that SET-19 is required for genomic H3K23me3 occupancy and is linked to transcriptional repression.

SET-19 is dispensable for feeding RNAi targeting both germline and somatic genes

H3K23me3 and the putative H3K23 methyltransferase SET-32 have previously been implicated in transgenerational epigenetic inheritance (TEI) ^{25,26,31,33,55}. We first tested whether SET-19 is required for feeding RNAi targeting several germline and somatic genes, including pos-1, mex-3, dpy-11, and unc-15. RNAi targeting pos-1 or mex-3 causes embryonic lethality and results in unhatched F1 embryos ^56-58. RNAi targeting dpy-11 or unc-15 induces dumpy and paralysis phenotypes, respectively ^59-61. However, neither set-19 nor set-32 mutants displayed detectable defects in these feeding RNAi responses (Fig. S5a–d). RDE-1 is an essential Argonaute protein required for feeding RNAi response ⁶². As a control, rde-1 mutants were completely resistant to feeding RNAi. To increase the sensitivity of feeding RNAi assays, we next used an enhanced RNAi background ⁶⁰. In eri-1 mutants, RNAi targeting dpy-11 resulted in a strongly enhanced dumpy phenotype in the progeny. RNAi targeting lir-1 induced larval arrest, and targeting dpy-13 induced a super dumpy phenotype ^60,63,64; both phenotypes were abolished by nrde mutation. However, the eri-1;set-19 mutant remained fully competent for feeding RNAi against dpy-11, lir-1, and dpy-13 (Fig. S5e–g).

We then asked whether SET-19 also participates in the transgenerational epigenetic inheritance of RNAi. We employed a germline-expressed pie-1_p::h2b::gfp reporter and performed feeding RNAi against gfp for one generation, after which animals were transferred to fresh plates without continued RNAi exposure. As expected, wild-type animals silenced gfp expression in the parental generation and maintained robust silencing in their F1 progeny. In hrde-1, set-25, and set-32 mutants, the parental generation remained sensitive to feeding RNAi, but silencing inheritance was abolished in the F1 progeny (Fig. 4a, b). In contrast, set-19 mutants retained effective gfp silencing in both the parental and F1 generations, comparable to that in control animals (Fig. 4a, b). Similarly, we tested whether SET-19 is required for the intergenerational inheritance of RNAi using a soma-expressed sur-5p::gfp reporter. Feeding dsRNA targeting gfp resulted in efficient silencing of GFP expression in both parental animals and F1 progeny in control animals (Fig. 4c). As expected, GFP expression was not silenced in the rde-1 mutant, consistent with the essential role of RDE-1 in feeding RNAi responses. NRDE-3 is required for the intergenerational inheritance of RNAi targeting somatic genes ⁶⁰. In nrde-3 mutants, the somatic sur-5p::gfp was silenced in the parental generation, but this silencing was abolished in the F1 progeny. In set-19 and set-32 mutants, sur-5p::gfp remained silenced in both parental and progeny generations (Fig. 4c, d).

Fig. 4.

Thus, these results suggest that SET-19 is dispensable for RNAi and its inheritance in C. elegans.

SET-19 is predominantly expressed in somatic cells and required for developmental timing

To explore the physiological roles of SET-19, we examined the development and reproduction of set-19 mutants, but did not observe significant defects in brood size or hatch rate compared with wild-type animals (Fig. 5a, b). However, set-19 mutants exhibited a modest developmental delay relative to wild-type animals (Fig. 5c).

Fig. 5.

Publicly available expression data from WormBase (version WS297) showed that set-19 mRNA is more highly expressed in embryos than in larvae (Fig. 5d). To further examine cell type-specific expression, we analyzed published single-cell RNA-seq data from L2-stage larvae ⁶⁵. set-19 expression was relatively enriched in intestinal cells and neurons (Fig. 5e).

To directly assess the expression and subcellular localization of SET-19, we generated an endogenous GFP-tagged SET-19 strain using CRISPR/Cas9-mediated genome editing. GFP::SET-19 was broadly expressed in somatic cells during both embryonic and larval stages and was predominantly localized to the nucleus (Fig. 5f–h). During embryogenesis, SET-19 expression was low in early embryos but markedly increased in late embryos (Fig. 5h). No detectable GFP::SET-19 signal was observed in germ cells (Fig. 5g). In contrast, SET-32 was highly expressed in germ cells, displayed both nuclear and cytoplasmic localization during larval stages, and was weakly expressed in embryos (Fig. S6a–c).

These results indicate that SET-19 is predominantly expressed in somatic cells and is largely excluded from the germline.

SET-19 conducts H3K23 trimethylation predominantly in somatic cells

To further examine the role of SET-19 in H3K23me3 in vivo, we performed immunofluorescence staining and assessed H3K23me3 levels across different developmental stages and tissues. In wild-type N2 animals, H3K23me3 signals were readily detected throughout development and were broadly distributed across tissues (Fig. 6a–d). Consistent with the somatic expression pattern of SET-19, the loss of set-19 resulted in a pronounced reduction in H3K23me3 in somatic cells (Fig. 6a). However, no obvious reduction of H3K23me3 was observed in embryos (Fig. 6b). In young adult worms, the reduction in H3K23me3 was most prominent in intestinal cells (Fig. 6c), a tissue in which set-19 expression is relatively enriched (Fig. 5e). Notably, H3K23me3 levels in germ cells remained largely unchanged in set-19 mutants (Fig. 6d), consistent with the absence of detectable set-19 expression in the germline (Fig. 5g).

Fig. 6.

For comparison, we analyzed H3K23me3 levels in set-32 mutants. The loss of set-32 resulted in no, if any, detectable reduction in H3K23me3 across tissues or developmental stages (Fig. 6a–d), suggesting that SET-32 is unlikely to act as the major H3K23me3 methyltransferase in C. elegans.

Together, these results support a major role for SET-19 in establishing H3K23me3 in somatic cells.

Discussion

H3K23me3 is a histone modification presented across multiple species, yet its mechanisms of deposition and biological functions have remained poorly defined. In this study, we identified SET-19 as an H3K23me3 methyltransferase through a combination of biochemical and in vivo approaches. We show that SET-19-dependent H3K23me3 is enriched at heterochromatin-associated regions and is associated with transcriptional repression. SET-19-dependent deposition of H3K23me3 occurs predominantly in somatic tissues in C. elegans. Collectively, these findings extend our understanding of how H3K23 methylation is established and suggest that its deposition is regulated in a tissue-specific manner.

Histone lysine methylation is generally catalyzed by SET domain-containing methyltransferases. The C. elegans genome encodes at least 38 SET domain-containing lysine methyltransferases, yet only a subset of them have been functionally characterized ^35-44,66-69. To date, efforts to identify the enzymes responsible for H3K23 methylation have been relatively limited. Candidate-based analyses assessed H3K23me2 by immunostaining in primordial germ cells and the adult germline, but failed to identify a bona fide KMT for germline H3K23me2 ¹⁸. SET-32 was then shown to methylate H3K23 in vitro and to be required for nuclear RNAi-induced H3K23me3 in vivo ^25,26. Subsequently, the closely related methyltransferase SET-21 was also shown to possess H3K23 methyltransferase activity in vitro, predominantly generating H3K23me1 and H3K23me2. SET-32 and SET-21 are more strongly associated with the germline ²⁶. Yet single loss of set-32 or set-21 did not cause a noticeable reduction in overall H3K23 methylation. Here, we show that SET-19 functions as a new H3K23me3 methyltransferase that acts primarily in somatic tissues. Based on our findings together with previous reports, we therefore propose a working model in which SET-19 predominantly mediates H3K23me3 deposition in somatic cells, whereas germline H3K23 methylation may rely more heavily on SET-32 and SET-21 (Fig. 7).

Fig. 7.

Classical heterochromatin marks, such as H3K9 and H3K27 methylation, play essential roles in transcriptional repression, perinuclear anchoring of chromatin, genome stability, and RNAi-mediated gene silencing ^27-30,70-72. Previous studies linked H3K23me3, particularly SET-32-associated H3K23 methylation, to nuclear RNAi and transgenerational epigenetic inheritance ^{25,26,29,31,33,73}. Immunofluorescence staining shows that H3K23me3 forms discrete foci within the nucleus, resembling the nuclear distribution patterns reported for H3K9 and H3K27 methylation ⁷⁴. In addition, the heterochromatin protein HPL-1, a chromodomain-containing reader, has been shown to recognize both H3K9 and H3K23 methylation ¹⁸. Our data show that H3K23me3 is enriched at heterochromatin-associated regions and is linked to transcriptional repression. These observations suggest that H3K23me3 is an additional heterochromatin-associated repressive mark. Nevertheless, whether H3K23 methylation contributes to perinuclear chromatin organization remains an intriguing question. Future studies using perinuclear tethering assays could be used to assess the contribution of H3K23 methylation to nuclear positioning ⁷⁴.

H3K23 methylation appears to be functionally distinct from other heterochromatin marks. The major H3K9 and H3K27 methyltransferases in C. elegans, MET-2/SET-25 and MES-2, respectively, act broadly across tissues and developmental stages ^42,75. In contrast, SET-19 contributes to H3K23me3 predominantly in somatic cells, pointing to distinct modes of enzymatic regulation among repressive histone modifications. This tissue-specific pattern of deposition further raises the possibility that different H3K23 methyltransferases may make distinct functional contributions. Although SET-32 has been linked to nuclear RNAi and transgenerational epigenetic inheritance in the germline ^{25,26,29,31,33,73}, set-19 mutants do not exhibit noticeable defects in feeding RNAi targeting either germline or somatic genes. These results indicate that SET-19 is not an indispensable requirement for the execution of feeding RNAi and its inheritance. However, whether SET-19 contributes to RNAi-induced H3K23me3 deposition in somatic cells requires further investigation.

While our findings support a predominantly somatic role for SET-19, the functions of H3K23 methylation in both somatic and germline cells remain to be fully defined. The unresolved questions include whether multiple H3K23 methyltransferases act redundantly or in parallel in the germline. It also remains unclear how SET-19-dependent and germline-enriched H3K23 methyltransferases are coordinated across tissues, and to what extent H3K23me3 overlaps functionally with other heterochromatin machineries. Systematic biochemical and genetic analyses of SET domain proteins in C. elegans will help define the full complement of histone methylation activities and clarify how chromatin-based regulation is directed to distinct developmental stages and cell types.

Materials and Methods

Strains

The Bristol strain N2 was used as the standard wild-type strain. All strains were grown at 20°C unless otherwise specified. The strains used in this study are listed in Table S1.

Construction of plasmids and transgenic strains

For in situ expression of GFP::SET-19, the coding sequence of 3×FLAG::GFP, fused to a linker sequence (GGAGGTGGAGGTGGAGCT), was inserted immediately downstream of the start codon of set-19 using the CRISPR/Cas9 system. The 3×FLAG::GFP fragment was PCR-amplified from YY178 genomic DNA. Homologous left and right arms (1.5 kb each) were amplified from N2 genomic DNA, and the backbone fragment was amplified from plasmid pCFJ151. All fragments were assembled into the repair plasmid using the ClonExpress MultiS One Step Cloning Kit (C113-02, Vazyme) via Gibson assembly. The injection mixture contained pDD162 (50 ng/µl), the repair plasmid (50 ng/µl), pCFJ90 (5 ng/µl), and two or three sgRNA plasmids targeting sequences near the N-terminus of the set-19 gene (20 ng/µl per sgRNA plasmid). The mixture was injected into adult animals. Three to four days after injection, F1 worms expressing pharyngeal GFP were isolated using a Leica M165 FC fluorescence stereomicroscope. Progeny from these F1 worms were subsequently screened by PCR and validated by Sanger sequencing. The sequences of primers used for construction of the in situ transgenic strain are listed in Table S2.

Construction of mutant strains via CRISPR/Cas9 technology

To generate sgRNA expression vectors, the 20 bp unc-119 sgRNA guide sequence within the pU6::unc-119 sgRNA(F+E) vector was replaced with distinct target-specific sgRNA guide sequences. A plasmid mixture containing 30 ng/µl of each of the three or four sgRNA expression vectors, 50 ng/µl of pDD162 plasmid, and 5 ng/µl of pSG259 was co-injected into wild-type N2 nematodes. Worms carrying the desired gene deletions were screened via PCR following the method described previously ⁷⁶. All sgRNA sequences used in this study are listed in Table S3.

Isolation of histones and mass spectrometry

Synchronized L3–L4 larvae and embryos (obtained by bleaching gravid adults) were washed three times with 1× M9 buffer. All samples were flash-frozen in liquid nitrogen and pulverized into a fine powder using a ball mill (60 Hz). The frozen powder was resuspended in Nuclear Extraction (NE) Buffer (100 mM HEPES, pH 7.5; 50 mM NaCl; 1% Triton X-100; 0.1% sodium deoxycholate; 1 mM EDTA; 10% glycerol; plus protease inhibitors). The resulting suspension was centrifuged at low speed to remove worm debris. To remove soluble proteins, the clarified supernatant was further centrifuged at high speed (16,000 × g for 10 min), and the resulting supernatant was discarded. The insoluble pellet was resuspended in NE buffer. Histone isolation was performed using a high-salt extraction strategy ⁷⁷. First, the insoluble pellet was further lysed in salt-free buffer to disrupt nuclear envelopes; histones were subsequently liberated from chromatin complexes via extraction with 2.5 M NaCl buffer. Histones were separated by SDS-PAGE, and the Coomassie blue-stained band containing histone H3 was cut from the gel. The gel slices underwent two rounds of propionylation with propionic anhydride to derivatize free and monomethylated lysine residues. Afterward, the samples were digested with trypsin, followed by an additional propionylation step to modify newly exposed N-terminal amines ^39,78.

The eluted peptides were analyzed on an Easy-nLC 1000 system (Thermo Fisher) coupled to a Q Exactive mass spectrometer (Thermo Fisher) by LC-MS/MS. The acquired RAW data were analyzed using EpiProfile 2.1_Celegans, an updated version of EpiProfile 2.0 ⁷⁹. For each histone modified peptide, the relative abundance (%RA) was calculated by dividing the area under the curve (AUC) of each modified peptide by the sum of the areas corresponding to all the observed forms of that peptide.

Western blotting

Synchronized L3–L4 larvae and embryos obtained by bleaching gravid adults were washed three times with 1× M9 buffer. Samples were stored at -80°C until use. The samples were suspended in 2× SDS loading buffer and heated in a metal bath at 95°C for 5–10 minutes. The suspensions were then centrifuged at 16,000 × g, and the supernatants were collected. Proteins were resolved by SDS-PAGE on gradient gels (15% separation gel, 5% spacer gel) and transferred to a Hybond-ECL membrane. After washing with 1× Tris-buffered saline with Tween-20 (TBST) buffer and blocking with 5% milk-TBST, the membrane was incubated overnight at 4°C with antibodies. The membrane was washed three times for 10 minutes each with 1× TBST and then incubated with secondary antibodies at room temperature for 2 hours. The membrane was washed thrice for 10 minutes with 1× TBST and then visualized. The following antibodies were used for western blotting: anti-H3 (Abcam, ab1791), 1:8000; anti-H3K4me3 (Abcam, ab8580), 1:800; anti-H3K9me3 (Millipore, 07-523), 1:3000; anti-H3K27me3 (Millipore, 07-449), 1:8000; anti-H3K36me3 (Abcam, ab9050), 1:1000; anti-H3K23me1 (Active Motif, 39388), 1:6000; anti-H3K23me2 (Active Motif, 39653), 1:2000; anti-H3K23me2 (Beyotime, AG3990), 1:2000; anti-H3K23me3 (Active Motif, 61499), 1:3000; and anti-H3K23me3 (Beyotime, AG3889), 1:1000. Secondary antibodies: HRP-labeled goat anti-rabbit IgG (H +L) (Abcam, ab205718), 1:15000.

Recombinant protein expression and purification

The DNA segments encoding the SET domain of SET-19 (SET-19 SET; aa 37–318), the SET and coiled-coil domains of SET-19 (SET-19 SET+CC; aa 37–454) and the full-length protein of SET-32 were amplified from a cDNA library of wild-type strain N2 using PCR. Subsequently, the SET-19 SET, SET-19 SET+CC and SET-32 constructs were cloned and inserted into the pGEX-4T-1 plasmid and expressed in E. coli BL21-GOLD (DE3) cells (Novagen). The recombinant proteins were affinity-purified via GST tag binding to glutathione beads (Smart-Lifesciences, SA008100) according to the manufacturer’s instructions. The eluted protein was dialyzed against a buffer containing 20 mM Tris-HCl pH 7.5, 5% glycerol, and 200 mM NaCl and subsequently concentrated. The concentrated protein was then subjected to high-speed centrifugation; after centrifugation, the clarified protein solution was supplemented with glycerol and stored at -80°C.

In vitro histone methylation assay

GST-fused proteins (0.25 µM) were incubated with 1 µM histone H3 (Active Motif, 31294) and 80 µM S-adenosylmethionine (SAM) (NEB, B9003S) in a 20 µl reaction mixture containing methyltransferase activity buffer (20 mM Tris-HCl, pH 8.0; 50 mM NaCl; 20 mM KCl; 10 mM MgCl₂; 5% glycerol; 0.02% Triton X-100; 0.1 mg/ml BSA; and 1 mM dithiothreitol [DTT]) for 3 h at 20°C. Reactions were terminated by the addition of SDS loading buffer, followed by boiling for 15 min. Proteins were resolved on a 15% SDS-PAGE gel and analyzed by western blotting.

Imaging

For imaging larval stages, animals were immobilized in 0.1 M sodium azide and mounted on 1.5% agarose pads. For imaging embryos and germ cells, gravid adults were dissected on a coverslip in 2 μl of 0.4× M9 buffer containing 0.1 M sodium azide and then mounted on freshly prepared 1.1% agarose pads. Imaging was performed using a Leica THUNDER Imaging System equipped with a K5 sCMOS camera and HC PL FLUOTAR objectives (100×/1.40–0.70 oil, 40×/0.80, and 20×/0.80). Images were acquired using Leica Application Suite X software (v3.7.4.23463).

Immunofluorescence staining and quantification

Embryos were freeze-cracked using liquid nitrogen, then fixed in methanol for 30 seconds and post-fixed in 1% paraformaldehyde (PFA) for 2 minutes. After three washes in PBS containing 0.25% Triton X-100 (PBS-T), samples were permeabilized in PBS with 1% Triton X-100 for 20 minutes. Subsequently, the samples were blocked in PBS containing 0.5% BSA before overnight incubation with the primary antibody against H3K23me3 (Active Motif, 61499) at 4°C. Following three washes with PBS-T, samples were incubated with secondary antibodies (Alexa Fluor 488-conjugated anti-rabbit) for 2 hours at room temperature, before three final PBS-T washes and DNA staining with Hoechst 33342. For germline and intestine staining, gravid adult worms were dissected in M9 buffer, fixed in methanol for 1 minute at -20°C and in 2% PFA for 5 minutes. For L2 worm staining, worms were freeze-cracked using liquid nitrogen, then fixed in 1% paraformaldehyde (PFA) for 20 minutes. Fixed animals were permeabilized by sequential thiol-based reduction and oxidation steps. Samples were incubated in PBS containing 10% β-mercaptoethanol for 15 minutes, treated with PBS containing 10 mM DTT for 15 minutes, and then incubated in PBS containing 0.3% H₂O₂ for 15 minutes. The animals were subsequently fixed in methanol for 30 seconds and treated with the permeabilization solution (PBS containing 1% Triton X-100) for 20 minutes. Images were taken on a Leica THUNDER Imaging System with identical gain settings within each experimental set. Quantification was performed using ImageJ by measuring the mean gray value within the nuclear region and that of an adjacent background area. For each worm or tissue, fluorescence intensity was calculated as the average nuclear signal from three nuclei, subtracting the average background signal from three corresponding adjacent background regions. At least two independent experiments were performed.

Brood Size

L4-stage hermaphrodites were singled onto plates and transferred daily as adults until embryo production ceased, and the progeny numbers were scored.

Hatch ratio

Synchronized adult hermaphrodites were transferred to NGM plates to lay eggs, which were subsequently removed. The numbers of embryos and hatched larvae were scored, and the hatch ratio was calculated as the number of hatched larvae divided by the total number of embryos.

Development rate

Synchronized adult hermaphrodites were transferred to NGM plates to lay eggs, which were subsequently removed. The growth of the progeny was monitored every 24 hours at 20°C.

RNAi

RNAi experiments were performed at 20°C by placing synchronized embryos on RNAi feeding plates as previously described ⁸⁰. HT115 bacteria expressing the empty vector L4440 (a gift from A. Fire) were used as controls. Bacterial clones expressing target-specific double-stranded RNAs (dsRNAs) were obtained from the Ahringer RNAi library ⁸¹ and their identities were verified by Sanger sequencing.

Total RNA isolation

Synchronized L4-stage worms were washed three times with 1× M9 to remove bacterial residues. For RNA extraction, 400 µl of TRIzol Reagent (Ambion, 15596026) was added to a 100 µl worm aliquot, followed by 7–8 cycles of freezing in liquid nitrogen and thawing in a 42°C water bath. Afterward, 100 µl of DNA/RNA Extraction Reagent (Solarbio LIFE SCIENCES, P1014) was added to the samples, which were then centrifuged at 16000 × g for 15 minutes at 4°C. The collected supernatant was further treated with 400 µl isopropanol and 400 µl pre-chilled 75% ethanol, and subjected to DNase I digestion (Thermo Fisher Scientific, EN0521). Finally, RNA was eluted into 20 µl of nuclease-free water and used for mRNA library preparation.

mRNA-seq analysis

The Illumina-generated raw reads were first filtered to remove adapters, low-quality reads, and contaminants, yielding high-quality clean reads. The clean reads were aligned to the reference genome of WBcel235 via HISAT2 (v2.1.0)⁸². For quantification, featureCounts (v1.6.0) ⁸³ was used to calculate gene-level read counts. Library size factors were calculated using DESeq2 (v1.46.0) ⁸⁴ with its standard median-of-ratios method based on genome-wide gene expression counts. Genes with an adjusted P value (padj) < 0.01 and an absolute log2FoldChange > 1 were considered significantly differentially expressed. Three biological replicates were analyzed per condition. All statistical analyses and plots were generated using custom R scripts.

Chromatin immunoprecipitation (ChIP)

ChIP experiments were performed as previously described ³⁰. For L3–L4 worms, frozen animals were first ground to a fine powder in liquid nitrogen ⁸⁵. Then, the samples were crosslinked in 2% formaldehyde for 10 minutes at room temperature with gentle rotation. Crosslinking was quenched by adding 0.125 M glycine (final concentration) for 5 minutes at room temperature. Samples were sonicated for 13 cycles (30 s on and 30 s off per cycle) at high output with a Bioruptor Plus (Diagenode), with the sample tubes kept in an ice-water bath throughout the process. The lysates were precleared and immunoprecipitated with a rabbit anti-H3K23me3 antibody (Active Motif, 61499) overnight at 4°C. Chromatin/antibody complexes were recovered with Dynabeads^TM Protein A (Invitrogen, 10002D) followed by extensive sequential washes with 150 mM, 500 mM, and 1 M NaCl, respectively. Crosslinks were reversed overnight at 65°C. The recovered DNA was treated with RNase (Roche) for 30 minutes at 65°C, and all DNA samples were purified using a DNA purification kit (TianGen, #DP204).

ChIP-Seq

The DNA samples from the ChIP experiments were subjected to quality control (QC) prior to sequencing and then sequenced at high depth at Novogene Bioinformatics Technology Co., Ltd. (Beijing, China). Briefly, the ChIP DNA was processed for library construction following the standard protocol: the DNA fragments were combined with an End Repair/A-tailing mix and incubated in a single reaction to complete end repair and 3’-end adenylation. The resulting A-tailed DNA fragments were then incubated with sequencing adapters in ligation mix to achieve adapter ligation. The adapter-ligated DNA was subjected to size selection and several rounds of PCR amplification to obtain the final DNA library. The size distribution of the library fragments was analyzed on the Agilent 5400 system (Agilent, USA; with matching Agilent assay reagents). The library was quantified by qPCR to a final concentration of 1.5 nM. The qualified libraries were pooled according to their effective concentration and required data volume and then further amplified on cBot to generate clusters on the flow cell and sequenced with a paired-end 150 bp (PE150) method on an Illumina platform.

Published ChIP-seq datasets of histone modifications in C. elegans were downloaded from the NCBI GEO or ENCODE databases. The datasets used in this study are listed in Table S4.

ChIP-seq data analysis

ChIP-seq data analysis was performed as previously described ⁸⁶. ChIP-seq reads were aligned to the WBcel235 assembly of the C. elegans genome using Bowtie2 (v2.3.5.1) ⁸⁷ by Ben Langmead with the default settings. The SAMtools (v0.1.19) ⁸⁸ “view” utility was used to convert the alignments to BAM format, and the “sort” utility was used to sort the alignment files. ChIP-seq peaks were called using MACS2 (v2.1.1) ⁸⁹ with subcommand. For the H3K4me3 and H3K36me3 datasets, which display narrow enrichment patterns, peaks were identified using the parameters “-g ce -B -f BAM -q 0.01”. For the H3K27me3, H3K9me3, and H3K23me3 datasets, which show broad histone modification domains, peaks were identified using the parameters “-g ce -B -f BAMPE --broad --broad-cutoff 0.01”. Deeptools subcommand bamCoverage (v3.5.0) was used to produce bigWig tracks for data visualization with defined parameters (--binSize 20 --normalizeUsing BPM --smoothLength 60 --extendReads 150) from bam files. The Integrative Genomics Viewer genome browser ⁹⁰ was applied to visualize signals and peaks. The genomic regions were defined as chromosome arms and centers based on previous works ⁵¹. For visualization of peak locations, we used the R package ggplot2 (v4.0.0).

For heatmap analysis, the called peaks of H3K23me3 rep1 were used. Deeptools subcommand computeMatrix (v3.4.3) was used to calculate the score matrix for the heatmap with defined parameters (reference-point --referencePoint center -b 3000 -a 3000 --skipZeros). The heatmap was plotted with Deeptools subcommand plotHeatmap (v3.5.0).

ChIPseqSpikeInFree normalization for H3K23me3

H3K23me3 ChIP-seq experiments were performed with two replicates each in both wild-type (wt) and set-19 mutant backgrounds. To compare the H3K23me3 signals between the two backgrounds in the absence of spike-in, ChIPseqSpikeInFree (v1.2.4) ⁹¹ was used to calculate the scaling factor (SF) for every sample (SF_i). The effective ChIP-seq library size for sample i was calculated as N_i ∗ SF_i, where N_i is the original library size. The effective library size was then used to normalize the read count from sample i during the downstream differential analysis and heatmap visualization. For the two replicates of wt and set-19 mutants, pairwise comparisons were performed, and the resulting SF values were averaged, yielding SF_wt = 1 and SF_set-19 = 2.115. The “-scale” parameter for bedtools genomecov was calculated from the effective library size using the formula: Scale parameter = 15000000/(Ni × SFi), where 15000000 is the preset reference read count. Normalized BEDGraph files were generated by setting the “-scale” parameter to the value from the above formula. The genome was then partitioned into fixed 20 bp windows, and the normalized signal was mapped into these windows using a bedtools map with weighting by the actual overlap length. The log₂(IP/Input) enrichment value for each window was subsequently calculated after adding a pseudocount of 1 to both IP and Input signals to avoid log₂(0) errors. Finally, the sorted BEDGraph file was converted into BigWig format for visualization. For the differential analysis of H3K23me3 peaks, we used the R package DiffBind (v3.16.0) ⁵⁴ and integrated the sample-specific SF into the total read count normalization via the underlying DESeq2 (v1.46.0) ⁸⁴ framework.

Statistics

Bar graphs with error bars are presented with mean and standard deviation (SD). All experiments were conducted with independent C. elegans animals for the indicated N times. Statistical analysis was performed with a two-tailed Student’s t test.

Figure supplements

Fig. S1.

Fig. S2.

Fig. S3.

Fig. S4.

Fig. S5.

Fig. S6.

Data availability

ChIP-seq and RNA-seq data have been deposited in the Genome Sequence Archive (GSA) at the National Genomics Data Center (NGDC) under accession number CRA037051. The mass spectrometry proteomics data have been deposited in the iProX repository with the accession codes IPX0015174000 and PXD073043. All data generated or analyzed during this study are included in the manuscript and supporting files. Source data files have been provided for all figures.

Acknowledgements

We are grateful to the members of the Guang laboratory for their comments and suggestions. We are grateful to the International C. elegans Gene Knockout Consortium and the National Bioresource Project for providing the strains. Some strains were provided by the CGC, which is funded by the NIH Office of Research Infrastructure Programs (P40 OD010440).

Additional information

Funding

This work was supported by grants from the National Key R&D Program of China (2022YFA1302700), the National Natural Science Foundation of China (32230016, 32270583, 32300438, 32400435, and 32470633), the Research Funds of Center for Advanced Interdisciplinary Science and Biomedicine of IHM (QYPY20230021), and the Fundamental Research Funds for the Central Universities (WK9100250107).

Author contributions

Y.S., M. Huang, X.F. and S.G. conceptualized the research; M.X., X.F. and S.G. designed the research; M.X., Z.F., M. Huang, and C.Y. performed the research; X.C., X. Huang, C.Z., M. Hong and X. Hou contributed new reagents; J.C., X. Hou, S.L. and M.L. provided technical guidance on bioinformatics analysis and protein purification; M.X., X.F. and S.G. wrote the paper.

Funding

MOST | National Key Research and Development Program of China (NKPs) (2022YFA1302700)

• Shouhong Guang

MOST | National Natural Science Foundation of China (NSFC) (32230016)

• Shouhong Guang

MOST | National Natural Science Foundation of China (NSFC) (32270583)

• Chengming Zhu

MOST | National Natural Science Foundation of China (NSFC) (32300438)

• Xinya Huang

MOST | National Natural Science Foundation of China (NSFC) (32400435)

• Xiangyang Chen

MOST | National Natural Science Foundation of China (NSFC) (32470633)

• Xuezhu Feng

MOE | Fundamental Research Funds for the Central Universities (Fundamental Research Fund for the Central Universities) (WK9100250107)

• Xiangyang Chen

Additional files

Tables S1-S4.