Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 May 15;34(4):556-571.
doi: 10.1101/gr.279119.124.

Suv39h-catalyzed H3K9me3 is critical for euchromatic genome organization and the maintenance of gene transcription

Christine R Keenan #  1   2 , Hannah D Coughlan #  3   2 , Nadia Iannarella  3 , Andres Tapia Del Fierro  3   2 , Andrew Keniry  3   2 , Timothy M Johanson  3   2 , Wing Fuk Chan  3   2 , Alexandra L Garnham  3   2 , Lachlan W Whitehead  3   2 , Marnie E Blewitt  3   2 , Gordon K Smyth #  3   4 , Rhys S Allan #  1   2
Affiliations

Suv39h-catalyzed H3K9me3 is critical for euchromatic genome organization and the maintenance of gene transcription

Christine R Keenan et al. Genome Res. .

Abstract

H3K9me3-dependent heterochromatin is critical for the silencing of repeat-rich pericentromeric regions and also has key roles in repressing lineage-inappropriate protein-coding genes in differentiation and development. Here, we investigate the molecular consequences of heterochromatin loss in cells deficient in both SUV39H1 and SUV39H2 (Suv39DKO), the major mammalian histone methyltransferase enzymes that catalyze heterochromatic H3K9me3 deposition. We reveal a paradoxical repression of protein-coding genes in Suv39DKO cells, with these differentially expressed genes principally in euchromatic (Tn5-accessible, H3K4me3- and H3K27ac-marked) rather than heterochromatic (H3K9me3-marked) or polycomb (H3K27me3-marked) regions. Examination of the three-dimensional (3D) nucleome reveals that transcriptomic dysregulation occurs in euchromatic regions close to the nuclear periphery in 3D space. Moreover, this transcriptomic dysregulation is highly correlated with altered 3D genome organization in Suv39DKO cells. Together, our results suggest that the nuclear lamina-tethering of Suv39-dependent H3K9me3 domains provides an essential scaffold to support euchromatic genome organization and the maintenance of gene transcription for healthy cellular function.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.
Suv39-dependent H3K9me3 loss causes predominant gene repression in euchromatic regions. (A) Heatmap showing change of expression (logRPKM) of differentially expressed (DE) genes between Suv39h1 and Suv39h2 double-knockout (DKO) and control cells. The proportions of genes down-regulated and up-regulated in Suv39DKO cells versus control are annotated. (B) Barcode enrichment plot showing ranking of aging-related genes from the GenAge database (de Magalhaes et al. 2009) among the DE genes. Genes are ranked right to left from most up-regulated to most down-regulated in DKO cells. The rank of genes associated with increased lifespan is marked by red vertical bars and that of genes associated with decreased lifespan by blue vertical bars. Red and blue worms show relative enrichment. ROAST gene set test P-values tests correlation. (C) Same as D but with directionality removed to include genes with contradictory annotated life span effects. (D,E) Gene Ontology (GO) enrichment in up-regulated (D) and down-regulated (E) DE genes. (F,G) quantification of overlap between up-regulated (F) and down-regulated (G) DE gene promoters (2 kbp upstream/500 bp downstream) and H3K4me3, H3K27ac, H3K27me3, and H3K9me3 ChIP-seq peaks from control DP thymocytes as called by HOMER. (H) Example tracks of the region surrounding the promoter of a down-regulated DE gene (Itga4) showing ATAC-seq, ChIP-seq (H3K4me3, H3K27ac, H3K27me3, H3K9me3, IgG), and RNA-seq data. (I) Plots of mean coverage from 3 kb upstream of the transcription start site (TSS) to 3 kb downstream of the transcription end site (TES) of up-regulated (DE Up), down-regulated (DE Down), and non-DE (divided into expressed and nonexpressed genes based on expression level in control cells) showing chromatin accessibility (ATAC) and H3K4me3, H3K27ac, H3K27me3, H3K9me3 H3K27ac, H3K9me3 ChIP-seq data. (J) Boxplots of ATAC, H3K4me3, H3K27ac, and H3K27me3 logRPKM across gene promoters (2 kbp upstream/500 bp downstream) that are up-regulated (DE Up), down-regulated (DE Down), and non-DE (divided into expressed and nonexpressed genes based on expression level in control cells). Box plots depict the interquartile range (IQR) ± 1.5 × IQR with median annotated. Distributions were compared by Wilcoxon test.
Figure 2.
Figure 2.
Suv39DKO reduces lamina-associated domains (LADs) without inducing gene activation. (A) Example tracks of Lamin B1 ChIP-seq from control and Suv39DKO DP thymocytes overlayed with H3K9me3, H3K27me3, H3K4me3, and H3K27ac ChIP-seq and ATAC-seq from control DP thymocytes. Disrupted LADs, not-disrupted LADS, DE genes, and expressed genes are also annotated. LADs are divided into LADs lost in DKO and LADs retained or gained. Lamin B1 ChIP-seq tracks are shown at 50-kb resolution, and all other tracks are shown at 10-kb resolution. (B) Number of LADs called by the enriched domain detector across all chromosomes in control and Suv39DKO DP thymocytes. (C) Distributions of LAD size from B in control and Suv39DKO DP thymocytes. Box plot depicts the IQR ± 1.5 × IQR and median annotated. Distributions were compared by Wilcoxon rank-sum test with continuity correction. (D) LAD strength score (seg.mean) from CBS algorithm in control and Suv39DKO DP thymocytes. Control and DKO scores were compared using a paired t-test. (E) Median coverage of H3K9me3, H3K27me3, H3K4me3, H3K27ac, and ATAC across LAD boundaries from CBS analysis showing disrupted LADs and nondisrupted LADs. (F) Distributions of H3K9me3, H3K27me3, H3K4me3, H3K27ac, and ATAC logRPKM in disrupted LADs (CBS score reduced by >0.1 in DKO vs. CON), not-disrupted LADs, and inter-LAD (iLAD) regions. Distributions are shown as box plots depicting the IQR ± 1.5 × IQR with median annotated. Distributions were compared by Wilcoxon rank-sum test with continuity correction. (G) DNA-FISH against LAD or iLAD loci in control and Suv39DKO DP thymocytes. (H) Quantification of G showing three-dimensional radial position of FISH foci between nuclear periphery (=1.0) and center of the nucleus (=0.0). (I) Proportion of DE genes between Suv39DKO and control cells that overlap disrupted LADs and nondisrupted LADs.
Figure 3.
Figure 3.
Heterochromatin loss causes a loss of chromatin interactivity in active regions and significant switching from active to repressive compartments. (A) Integrative Genomics Viewer (IGV) tracks of A/B compartments from Hi-C analysis of control and Suv39DKO DP thymocytes overlayed by H3K9me3 and H3K27ac ChIP-seq. (B) Proportion of genome that switches A/B compartments in Suv39DKO versus control cells (shown as a percentage of whole genome). (C) Overlap of down-regulated DE genes with compartment-switched regions. (D) Number of unclustered differential interactions (DIs) (FDR < 0.05) between Suv39DKO and control cells. Strengthened interactions (logFC > 0) are annotated as “up” and weakened (logFC < 0) as “down.” Overlap of the DI anchors with compartments A and B in control cells in shown. DIs in which both anchors are not contained in the same compartment or in which one or more anchor overlaps both compartments are annotated as “mixed.” (E) The proportion of DIs with anchors that overlap switched compartments. (F) Transcription factor (TF) binding motifs enriched in the anchors of DIs as determined by the HOMER pipeline. (G) CTCF ChIP-seq (Shih et al. 2012) coverage of the anchors of DIs versus the rest of the genome. Box plot depicts the IQR ± 1.5 × IQR and median annotated. Distributions were compared by Wilcoxon rank-sum test with continuity correction. (H) Hi-C contact matrices at 50-kbp resolution showing the top three DI regions between Suv39DKO and control cells. Color scale indicates the number of read bins per bin pair with visualization scaled to total library size to allow appropriate visual comparison. Unclustered DIs (FDR < 0.05) are shown as arcs (blue indicate a decrease in logFC, red an increase in logFC) in which the vertical axis is the −log10(P-value) of the DI. (I) The linear span between genomic anchors of strengthened (“up”) and weakened (“down”) DIs. Data shown as boxplot as in G. (J) Number and size of topologically associated domains (TADs) in each replicate sample of control and Suv39DKO cells. Data statistically compared by unpaired t-test with equal variance between the median of the TAD sizes and number of TADs. Boxplot for TAD size plotted as in G. (K) Number of TAD boundary changes between Suv39DKO and control cells divided into those overlapping LAD and non-LAD regions in control cells. (L) Density of ATAC, H3K27ac, H3K27me3, H3K27ac, H3K4me3, and H3K9me3 sequencing density (shown as logRPKM) of strengthened and weakened TAD boundaries in Suv39DKO and control cells. Data shown as boxplot as in G and compared by Wilcoxon rank-sum test with continuity correction. (M) Proportion of DE genes up-regulated versus down-regulated overlapping altered TAD boundaries and the rest of the genome. Data analyzed by chi-squared test.
Figure 4.
Figure 4.
Higher-order modeling of the 3D nucleome in Suv39DKO cells reveals heterochromatin loss affects the expression of genes positioned closer to the nuclear periphery in 3D space. (A) Representative Hi-C contact matrices at 200-kbp resolution showing a loss of higher-order TAD–TAD cliques in Suv39DKO cells. Color scale indicates the number of read bins per bin pair with visualization scaled to total library size to allow appropriate visual comparison. A/B compartments are also annotated as are differentially expressed genes (DEGs; blue is down-regulated and red is up-regulated). (B) Number of higher-order TAD–TAD cliques detected in Suv39DKO and control cells. Data statistically compared by two-way ANOVA. (C) Number of higher-order TAD–TAD cliques in compartment A and compartment B in Suv39DKO and control cells. Proportion of cliques in each compartment data statistically compared by chi-squared test. (D,E) Representative Chrom3D modeling of the 3D nucleome of control and Suv39DKO cells colored as individual chromosome territories (each chromosome is arbitrarily colored; D) or higher-order TAD–TAD, TAD–TAD–LAD, or LAD interactions (E). (F) Measurements of LAD positioning from Chrom3D modeling in Suv39DKO and control cells. Box plot depicts the IQR ± 1.5 × IQR with median annotated from 100 independent modeling simulations from a separate seed value, with each simulation containing 5 million iterations. Distributions were compared by a Welch's unequal variances t-test. (G,H) Measurements of DE gene positioning from Chrom3D modeling as in F. Distributions were compared by a Welch's unequal variances t-test.
Figure 5.
Figure 5.
Altered euchromatic genome interactivity near LAD domains correlates with transcriptional dysregulation in Suv39DKO cells. (A) Schematic of gene-centric interactivity analysis. (B) Heatmap of the interactivity (log2CPM) of differentially interacting genes between Suv39DKO and control cells (FDR < 0.05). (C) Gene set enrichment analysis of genes with differential interactivity relative to aging-related genes from the GenAge database. Barcode enrichment plot shows the correlation of aging-related genes relative to differential interactivity. Genes are ordered on the plot from right to left (x-axis) from most increased in interactivity to most decreased in interactivity according to the moderated F-statistic. The P-value was calculated with the fry test. (D) The logFC of gene-centric interactivity for DE genes shown relative to linear genomic distance to the nearest LAD and statistically compared by Wilcoxon rank-sum test with continuity correction. (E) Coverage of RNA expression across iLAD/LAD boundaries with disrupted LADs in the Suv39DKO plotted separately from retained LADs. Summary plot shown as the median of the coverage. (F) Barcode enrichment plot showing the correlation of DE genes relative to differential interactivity. Genes are ordered on the plot from right to left (x-axis) from most up-regulated to most down-regulated according to the moderated F-statistic. The P-value was calculated with the fry test. (G) The logFC of gene-centric interactivity relative to logFC of transcriptional change for DE genes. (H) Schematic of putative mechanism by which the loss of heterochromatin causes transcriptional repression in euchromatic regions.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Allan RS, Zueva E, Cammas F, Schreiber HA, Masson V, Belz GT, Roche D, Maison C, Quivy JP, Almouzni G, et al. 2012. An epigenetic silencing pathway controlling T helper 2 cell lineage commitment. Nature 487: 249–253. 10.1038/nature11173 - DOI - PubMed
    1. Bahar R, Hartmann CH, Rodriguez KA, Denny AD, Busuttil RA, Dollé ME, Calder RB, Chisholm GB, Pollock BH, Klein CA, et al. 2006. Increased cell-to-cell variation in gene expression in ageing mouse heart. Nature 441: 1011–1014. 10.1038/nature04844 - DOI - PubMed
    1. Bannister AJ, Zegerman P, Partridge JF, Miska EA, Thomas JO, Allshire RC, Kouzarides T. 2001. Selective recognition of methylated lysine 9 on histone H3 by the HP1 chromo domain. Nature 410: 120–124. 10.1038/35065138 - DOI - PubMed
    1. Becker JS, Nicetto D, Zaret KS. 2016. H3K9me3-dependent heterochromatin: barrier to cell fate changes. Trends Genet 32: 29–41. 10.1016/j.tig.2015.11.001 - DOI - PMC - PubMed
    1. Becker JS, McCarthy RL, Sidoli S, Donahue G, Kaeding KE, He Z, Lin S, Garcia BA, Zaret KS. 2017. Genomic and proteomic resolution of heterochromatin and its restriction of alternate fate genes. Mol Cell 68: 1023–1037.e15. 10.1016/j.molcel.2017.11.030 - DOI - PMC - PubMed

Publication types

LinkOut - more resources

-