KDM2 proteins constrain transcription from CpG island gene promoters independently of their histone demethylase activity

Abstract

CpG islands (CGIs) are associated with the majority of mammalian gene promoters and function to recruit chromatin modifying enzymes. It has therefore been proposed that CGIs regulate gene expression through chromatin-based mechanisms, however in most cases this has not been directly tested. Here, we reveal that the histone H3 lysine 36 (H3K36) demethylase activity of the CGI-binding KDM2 proteins contributes only modestly to the H3K36me2-depleted state at CGI-associated gene promoters and is dispensable for normal gene expression. Instead, we discover that KDM2 proteins play a widespread and demethylase-independent role in constraining gene expression from CGI-associated gene promoters. We further show that KDM2 proteins shape RNA Polymerase II occupancy but not chromatin accessibility at CGI-associated promoters. Together this reveals a demethylase-independent role for KDM2 proteins in transcriptional repression and uncovers a new function for CGIs in constraining gene expression.

BAP1 constrains pervasive H2AK119ub1 to control the transcriptional potential of the genome

“…For Pol II and its phosphorylated forms, cChIP-seq was done as described previously (Turberfield et al 2019). Briefly, 5 × 10 7 ESCs (untreated and OHT-treated) were cross-linked in 10 mL of 1× PBS with 1% formaldehyde (methanol-free; Thermo Scientific) for 10 min at 25°C and then quenched by addition of 125 mM glycine.…”

“…To assay chromatin accessibility, calibrated ATAC-seq was performed as described previously (Turberfield et al 2019). First, 1 ×10 7 ESCs (untreated and OHT-treated) were mixed with 4 × 10 6 Drosophila SG4 cells in 1× PBS and then lysed in 1 mL of HS lysis buffer (50 mM KCl, 10 mM MgSO 4 .7H 2 0, 5 mM HEPES, 0.05% NP40 [IGEPAL CA630], 1 mM PMSF, 3 mM DTT, 1× PIC [Roche]) for 1 min at room temperature.…”

“…Following massive parallel sequencing, reads were mapped and processed as described previously (Fursova et al 2019;Turberfield et al 2019). Briefly, for cChIP-seq, cATAC-seq, and gDNA-seq, Bowtie 2 (Langmead and Salzberg 2012) was used to align paired-end reads to the concatenated mouse and spike-in genome sequences (mm10 + dm6 for native cChIP-seq, cATAC-seq, and gDNA-seq; mm10 + hg19 for cross-linked cChIP-seq) with the "--no-mixed" and "--no-discordant" options.…”

“…To visualize the cChIP-seq, cATAC-seq, and cnRNA-seq and quantitatively compare genome enrichment profiles, chromatin accessibility, and gene expression between conditions, the data were internally calibrated using dm6 or hg19 spike-in as described previously (Fursova et al 2019;Turberfield et al 2019). Briefly, uniquely aligned mm10 reads were randomly subsampled based on the total number of spike-in (dm6 or hg19) reads in each sample.…”

The Role of Polycomb Repressive Complex in Malignant Peripheral Nerve Sheath Tumor

“…Early investigations regarding the location and function of H3K36 di-and tri-methylation marks suggested both distinct locations and opposing roles of these two marks in the regulation of the Drosophila genome [63]. This was further commented on by Turberfield et al, who used genome-wide profiling of H3K36me2 to indicate the widespread deposition of transcriptional mark throughout the genome, but with a notable absence on bodies of highly transcribed genes and CpG island-associated gene promoters [64]. Although methylation of H3K36 is commonly associated with transcriptional activation, it has been shown to participate in other cellular processes, including alternative splicing, DNA replication, as well as transcriptional repression [65].…”