HDAC1 SUMOylation promotes Argonaute-directed transcriptional silencing in C. elegans

Abstract

Eukaryotic cells use guided search to coordinately control dispersed genetic elements. Argonaute proteins and their small RNA cofactors engage nascent RNAs and chromatin-associated proteins to direct transcriptional silencing. The small ubiquitin-like modifier (SUMO) has been shown to promote the formation and maintenance of silent chromatin (called heterochromatin) in yeast, plants, and animals. Here, we show that Argonaute-directed transcriptional silencing in Caenorhabditis elegans requires SUMOylation of the type 1 histone deacetylase HDA-1. Our findings suggest how SUMOylation promotes the association of HDAC1 with chromatin remodeling factors and with a nuclear Argonaute to initiate de novo heterochromatin silencing.

Keywords: C. elegans; HDAC SUMOylation; developmental biology; germline; nuclear argonaut.

Figure 1.. SUMOylation and chromatin remodeling factors promote piRNA-mediated silencing.

(A) Schematic of the piRNA sensor screen. The piRNA sensor strain contains a gfp::csr-1 transgene that is silenced by the piRNA pathway in the presence of an active oma-1::gfp transgene (Seth et al., 2018). OMA-1::GFP localizes to the cytoplasm of oocytes. Inactivation of the piRNA pathway (by RNAi, mutation, or auxin-inducible protein depletion) desilences the transgene, resulting in GFP::CSR-1 expression in perinuclear P-granules throughout the germline, as shown in (B). (B) Differential interference contrast and epifluorescence images of dissected gonads in wild-type (wt), prg-1(tm872), and wago-9/hrde-1(ne4769) worms. PRG-1 is required to initiate silencing, while WAGO-9 is required to maintain silencing. The percentage of desilenced worms and number of worms scored are shown. (C) Analysis of SUMO and chromatin remodeling factors required for piRNA-mediated silencing. Genes identified in the RNAi-based screen of chromatin factors are listed with their human homologs and with the percentage of worms that express GFP::CSR-1 among the total number of worms analyzed (n) when function is reduced by RNAi (blue column) or by either mutation or degron-dependent protein depletion (peach column).

Figure 2.. SUMOylation of HDA-1 at K444 and K459 facilitates piRNA-mediated silencing.

(A) Domain structure of C. elegans type 1 histone deacetylase HDA-1 and C-terminal location of SUMO-acceptor sites. Sequence alignment showing poor conservation at the C-termini of C. elegans HDA-1 and Homo sapiens HDAC1. The human HDAC1 C-terminus possesses two consensus SUMO-acceptor sites, K444 and K476 (acceptor lysines in red; consensus SUMO acceptor motif in pink box). GPS-SUMO predicts two candidate non-consensus SUMOylation sites in HDA-1, both near the C-terminus, K444 and K459 (red lysines in green boxes). (B) Western blot analyses of HDA-1 and MRG-1 before (lanes 1–5) and after (lanes 6–10) affinity enrichment of SUMOylated proteins from wild-type (wt) or hda-1 SUMO acceptor-site mutants. SUMOylated proteins were enriched from worms expressing HIS10::SMO-1. Black arrowheads indicate SUMOylated HDA-1 and MRG-1; white arrowheads indicate unmodified forms of HDA-1 and MRG-1. Asterisks indicate non-specific bands. Additional higher forms (indicated by white star) were detected, suggesting Multi-monoSUMOylation or PolySUMOylation of HDA-1. (C) Analysis of piRNA-mediated silencing in SUMOylation-defective mutants and rescue by HDA-1::SMO-1 translational fusion. (Top) The color of each bar indicates the percentage of worms in which the piRNA sensor was silent (OFF, gray) or expressed (ON, green). Thirty (n = 30) worms of each genotype were examined. (Bottom) Differential interference contrast and epifluorescence images of dissected gonads from hda-1[KKRR] and hda-1[KKRR]::smo-1. (D) Western blots showing levels of HDA-1 and variants proteins expressed from the endogenous hda-1 locus. Tubulin was used as a loading control. (E) Brood size analysis of wt worms or HDA-1 SUMOylation-site mutants, and rescue by HDA-1::SMO-1 translational fusion. Worms were grown at 20°C or 25°C, as indicated. Statistical significance was determined by ordinary one-way ANOVA: *p<0.05; **p<0.01; ****p<0.0001; ns: not significant. (F) Mortal germline analyses of wt or HDA-1 SUMOylation-site mutant, and HDA-1::SMO-1 fusion worms. The wago-9/hrde-1 mutant has a severe mortal germline phenotype. Worms were passaged at 25°C for eight generations, and the average number of progeny from 10 individuals (n = 10) was determined at each generation. Error bars represent standard error of the mean (SEM).

Figure 3.. SUMOylation of HDA-1 promotes its association with NuRD and other chromatin factors.

(A) Silver stained gel of proteins that co-immunoprecipitate with HDA-1::GFP. The indicated protein bands were excised from the gel and identified by mass spectrometry. (B) Scatter plot comparing the levels of proteins identified by mass spectrometry in HDA-1::GFP IPs from smo-1(RNAi) and wild-type (no RNAi) worms. The x axis shows the log value of spectral counts for each protein identified by IP-MS from wild-type worms. The y axis shows the log ratio of spectral counts for each protein in HDA-1::GFP IPs from smo-1(RNAi) vs. wild-type. (C) As in B, but comparing hda-1[KKRR]::gfp to hda-1[WT]::gfp. In (B) and (C), the spectral counts of HDA-1 (RED) were used to normalize between samples. A full list of the identified proteins is provided in Supplementary file 2. (D) Western blot analyses of proteins (indicated to left of blots) that associate with MEP-1::GTF (GFP IP) in hda-1[WT], hda-1[KKRR] and hda-1[KKRR]::smo-1 lysates. The detected proteins are indicated to the right (black arrowheads). The modified isoforms, HDA-1::SUMO and HDA-1::SUMO-UBIQUITIN are indicated with white arrow and white star, respectively. (E) Side-by-side comparison of HDA-1 isoforms detected in the HDA-1, SMO-1, and UBIQUITIN blots in (D). The black dot indicates an unknown HDA-1 isoform. The black star indicates an unknown SUMOylated protein.

Figure 4.. HDA-1 SUMOylation is required for formation of germline heterochromatin.

(A, B) Immunofluorescence micrographs of (A) wild-type (wt) and (B) hda-1[KKRR] gonads stained with anti-H3K9me2 antibody and DAPI. The dashed boxes indicated ‘a’ and ‘b’ are enlarged as shown. Two representative gonads are shown for each strain. (C) Differential interference contrast and immunofluorescence micrographs of gonads from wt, hda-1[KKRR], mep-1::gfp::degron with auxin (50 µM) worms stained with anti-H3K9Ac antibody and DAPI. (D) Genome Browser tracks (Integrated Genomics Viewer [IGV]) showing ChIP-seq peaks for NuRD complex components (HDA-1, LET-418, and MEP-1) and three histone modifications (H3K9Ac, H3K9me2, and H3K9me3) along each C. elegans chromosome (I–V and X). MEP-1(gonad) data are from worms that express MEP-1::GTF only in the germline, using the wago-1 promoter, for germline-specific CHIP.

Figure 4—figure supplement 1.. Increased levels of active H3K4me3 chromatin mark in hda-1[KKRR].

(A, B) Immunofluorescence micrographs of anti-H3K4me3 and DAPI staining in three adult gonads of (A) wild-type (wt) and (B) hda-1[KKRR].

Figure 5.. The SUMO, NuRD, and piRNA pathways regulate the same group of targets.

(A) Schematic of mRNA-seq from dissected gonads. (B) Venn diagram showing overlap between upregulated genes in degron::hda-1, mep-1::degron, ubc-9(G56R), and hda-1[KKRR] germlines. Numbers in parentheses indicate total number of upregulated genes. (C) Scatter plot of upregulated genes in hda-1[KKRR]. The x-axis represents reads in wild-type (wt), and y-axis represents reads in hda-1[KKRR]. (D) Scatter plot showing the effect of hda-1[KKRR]::smo-1 on the 430 genes (from C) upregulated in hda-1[KKRR]. The x-axis represents reads in wt, and y-axis represents reads in hda-1[KKRR]::smo-1. (E) Venn diagram showing overlap between upregulated genes in hda-1[KKRR], prg-1, rde-3, and wago-9 mutants. (F) Bar graph showing fractions of upregulated genes involved in spermatogenesis, oogenesis, neutral, or other categories. ‘Other’ indicates genes that cannot be put into one of the other categories (Ortiz et al., 2014). Genes with >1 mRNA-seq reads in wt gonad were used to generate the ‘wild-type’ dataset as a reference. The number of genes in each dataset is labeled at the top. (G–I) Scatter plots comparing mRNA-seq reads in (G) hda-1[KKRR], (H) rde-3, and (I) prg-1(ne4766) to those in wt. The blue dashed lines indicate a twofold increase or decrease in mutant compared to wt.

Figure 5—figure supplement 1.. Comparison of replicates in gonad mRNA-seq.

(A) Scatter plots comparing two independent replicates of mRNA-seq data in the indicated strains. Replicate 1 is plotted on the x-axes and replicate 2 on the y-axes. The dashed blue lines indicate a twofold change. (B) Bar graphs showing number of upregulated (green) and downregulated (orange) genes that change at least twofold in the mutants germline (p≤0.05). (C) Venn diagram showing the overlap of upregulated genes in degron::hda-1, prg-1, rde-3, and wago-9. Total number of the upregulated genes in each mutant is indicated in the parentheses.

Figure 5—figure supplement 2.. Analysis of transposons.

(A) Bar graphs showing number of upregulated (green) or downregulated (orange) transposons that change at least twofold in the gonad mRNA-seq of the indicated mutants (p≤0.05). (B) Scatter plots displaying upregulated transposons in the indicated mutants. The x-axis shows log10 value of average read counts in repeats of control (WT), and the y-axis shows log10 value of read counts in the indicated mutants. Transposons that are increased at least twofold and adjusted p-value ≤0.05 are colored in red. Transposon families are labeled if available. WT was used as a control for all the mutants except degron::hda-1 and ubc-9(ne4446[G56R]). For degron::hda-1 with 100 µM auxin exposure, worms without auxin exposure were used as control, and for ubc-9(ne4446[G56R]);oma-1::gfp; gfp::csr-1(AE), oma-1::gfp; gfp::csr-1(AE) was used as control. The dashed lines indicate twofold changes.

Figure 5—figure supplement 3.. Loss of HDA-1 SUMOylation leads to depletion of H3k9me2 on a selected set of genes.

(A) Box plots show levels of H3K9me2 on HDA-1(KKRR) targets (upregulated in mRNA-seq) in the indicated strains. (B–E) Examples of H3K9me2 peaks (ChIP-seq reads) over protein-coding genes (B, C) and transposons (D , E). Read counts (y-axis) were normalized by total mapped reads in each sample.

Figure 5—figure supplement 4.. Upregulated spermatogenic genes in the mutants.

(A) Scatter plots comparing mRNA-seq reads in mutant gonads to those in wild-type. Axes show log10 values of average reads from replicates of control (x-axis) and the indicated mutant (y-axis). Wild-type gonads were used as control for all the mutants except degron::hda-1 and ubc-9(ne4446[G56R]). Worms without auxin exposure were used as control for degron::hda-1, and oma-1::gfp; gfp::csr-1(AE) was used as a control for ubc-9(ne4446[G56R]);oma-1::gfp; gfp::csr-1(AE). (B) Scatter plots comparing mRNA-seq reads in rde-3 vs. hda-1[KKRR], prg-1 vs. hda-1[KKRR], and prg-1 vs. rde-3. Dashed lines indicate a twofold change. A value of 0.1 was assigned to undetected genes.

Figure 5—figure supplement 5.. Analysis of small RNAs targeting upregulated genes in hda-1[KKRR].

(A) Scatter plots comparing anti-sense small RNA reads targeting each gene in hda-1[KKRR] and wild-type. The upregulated genes in hda-1[KKRR] from mRNA-seq are indicated in red. Dashed lines represent a twofold change. (B) Venn diagram showing the overlap between upregulated genes in hda-1[KKRR] from mRNA-seq and genes with increased or decreased small RNAs in hda-1[KKRR]. The total number of genes in each category is indicated in the parentheses.

Figure 6.. HDA-1 interacts with WAGO-9/HRDE-1, which requires HDA-1 SUMOylation.

(A) Western blot analysis showing that HDA-1 and HPL-2 co-immunoprecipitate with GFP::WAGO-9/HRDE-1. (B) Western blot analysis of WAGO-9 and HPL-2 in HDA-1::GFP immunoprecipitates from wild-type, HDA-1 SUMO acceptor-site mutant, or smo-1(RNAi) worms. Blotting with anti-SMO-1 antibody showed depletion of SMO-1 in the smo-1(RNAi) worms. A band the size of untagged HDA-1 (black arrowhead) in hda-1[KKRR]::gfp in input appears to be a cleavage product that removes the GFP tag. (C) Graph showing the levels of silencing induced by RNAi over three generations in wild-type, hda-1[KKRR], and hda-1[KKRR]::smo-1 worms. Worms were treated with gfp(RNAi) at P0, and F1 larvae were transferred to regular NGM plates. The percentage of worms that express OMA-1::GFP was scored (n≥22, two replicates). Error bars represent standard error of the mean (SEM).

Figure 7.. Model: HDAC1 SUMOylation promotes Argonaute-directed transcriptional gene silencing.

SUMOylation of HDA-1 enables nucleosome remodeling and deacetylase (NuRD) complex assembly in the adult germline and WAGO-9 or other Argonautes recruit the NuRD complex to piRNA targets. The Argonaute/NuRD complex, along with other histone-modifying enzymes—for example, SPR-5, MET-2—removes the active histone marks (H3K9Ac and H3K4me2/3) and establishes silencing marks (H3K9me2/3) to suppress their transcription.