Journal: bioRxiv

Article Title: Cohesin disrupts polycomb-dependent chromosome interactions

doi: 10.1101/593970

Figure Lengend Snippet: (A) A schematic illustrating the genotype of the TIR1 and SCC1-mAID-GFP cell lines developed for Hi-C. (B) Immunofluorescence microscopy images of SCC1-mAID-GFP ESCs ± auxin (6 h). The nuclear membrane was labelled with an antibody against Lamin B1. Scale bar = 10 µm (bottom). (C) Hi-C in Control (TIR1 line + auxin) (left) and SCC1 DEG (SCC1-mAID-GFP line + auxin) (right) cells after auxin treatment visualised at 40 kb resolution. Peaks identified on the SCC1 DEG Hi-C matrix are shown as black circles. The genomic co-ordinates are illustrated below and to the right of the matrices. (D) Enrichment of histone modifications and proteins at paired interaction sites compared to the enrichments at matched random interaction sites. (E) ChIP-Seq snapshot illustrating RING1B, H2AK119ub, SUZ12 and H3K27me3 under an interaction that persists in the absence of cohesin. The Hi-C matrix is shown above at 20 kb resolution.

Article Snippet: RAD21, 1:1000 (Abcam, ab154769), RING1B (western) 1:1000 (Klose Lab), RING1B (ChIP-Seq) 1:1000 (Cell Signalling, 5694), CTCF 1:1000 (Abcam, ab70303) and TUBULIN 1:500 (Abcam, ab6046).

Techniques: Hi-C, Immunofluorescence, Microscopy, ChIP-sequencing

Journal: bioRxiv

Article Title: Cohesin disrupts polycomb-dependent chromosome interactions

doi: 10.1101/593970

Figure Lengend Snippet: (A) A representative western blot for SCC1 in the TIR1 and SCC1-mAID-GFP (=SCC1 DEG ) cell lines ± auxin. A wild type cell line is shown for comparison and tubulin is shown as a loading control. (B) Genomic distance-dependent contact probability from Hi-C in Control or SCC1 DEG cells. (C) Hi-C in Control and SCC1 DEG cells at 10 kb resolution. (D) Pearson correlation coefficient of chromosome 1 from Control and SCC1 DEG at 500 kb resolution (left). Bar plot of the genome-wide absolute Pearson correlation for Control and SCC1 DEG (right). (E) RING1B, H2AK119ub1, SUZ12, and H3K27me3 ChIP-seq signal metaplotted at RING1B peaks overlapping with interactions that persist in the absence of cohesin (blue line) or all RING1B peaks (dashed-black line). (F) A box plot showing the RING1B peak size at RING1B peaks overlapping interactions that persist in the absence of cohesin (blue right) or all RING1B peaks (grey left).

Article Snippet: RAD21, 1:1000 (Abcam, ab154769), RING1B (western) 1:1000 (Klose Lab), RING1B (ChIP-Seq) 1:1000 (Cell Signalling, 5694), CTCF 1:1000 (Abcam, ab70303) and TUBULIN 1:500 (Abcam, ab6046).

Techniques: Western Blot, Hi-C, Genome Wide, ChIP-sequencing

Journal: bioRxiv

Article Title: Cohesin disrupts polycomb-dependent chromosome interactions

doi: 10.1101/593970

Figure Lengend Snippet: (A) A schematic illustrating the genotype of the AID-RING1B cell line. (B) Immunofluorescence microscopy images of AID-RING1B ESCs ± auxin (6 h incubation) (bottom). The cells were labelled with antibodies against Lamin B1 and RING1B. Scale bar = 10 µm (bottom). (C) Genomic distance-dependent contact probability from Hi-C in Control or RING1B DEG (AID-RING1B + auxin) cells. (D) Aggregate TAD analysis of Control and RING1B DEG cells at 10 kb resolution. Effective contact probability is displayed at a published set of TAD intervals from ESC Hi-C . (E) Hi-C in Control and RING1B DEG cells at interactions that persist in the absence of cohesin (black circles) at 5 kb resolution. RING1B ChIP-seq is displayed above and to the left of the matrices. (F) Aggregate analysis of Hi-C from Control and RING1B DEG cells at interactions that persist in the absence of cohesin (n=336).

Article Snippet: RAD21, 1:1000 (Abcam, ab154769), RING1B (western) 1:1000 (Klose Lab), RING1B (ChIP-Seq) 1:1000 (Cell Signalling, 5694), CTCF 1:1000 (Abcam, ab70303) and TUBULIN 1:500 (Abcam, ab6046).

Techniques: Immunofluorescence, Microscopy, Incubation, Hi-C, ChIP-sequencing

Journal: bioRxiv

Article Title: Cohesin disrupts polycomb-dependent chromosome interactions

doi: 10.1101/593970

Figure Lengend Snippet: (A) A representative western blot for SCC1 and RING1B in the Control and AID-RING1B cell lines ± auxin. Tubulin is shown as a loading control. (B) Pearson correlation coefficient of chromosome 1 from Control and RING1B DEG at 500 kb resolution (left). Bar plot of the genome-wide absolute Pearson correlation for Control and RING1B DEG (right). (C) Hi-C in Control and RING1B DEG cells at two regions (top at 5 kb resolution and bottom at 10 kb resolution). Black circles indicate at interactions that persist in the absence of cohesin and RING1B ChIP-seq is displayed above and to the left of the matrices.

Article Snippet: RAD21, 1:1000 (Abcam, ab154769), RING1B (western) 1:1000 (Klose Lab), RING1B (ChIP-Seq) 1:1000 (Cell Signalling, 5694), CTCF 1:1000 (Abcam, ab70303) and TUBULIN 1:500 (Abcam, ab6046).

Techniques: Western Blot, Genome Wide, Hi-C, ChIP-sequencing

Journal: bioRxiv

Article Title: Cohesin disrupts polycomb-dependent chromosome interactions

doi: 10.1101/593970

Figure Lengend Snippet: (A) Hi-C illustrating an interaction that increases in strength in the SCC1 DEG cell line visualised at 20 kb resolution. RING1B ChIP-seq is displayed above and to the left of the matrices. (B) Aggregate analysis of Hi-C from Control and SCC1 DEG cells at peaks that persist in the absence of cohesin (n=336). (C) ChIP-Seq for RING1B in the Control and SCC1 DEG cells at the interacting sites shown in (A). (D) RING1B ChIP-seq signal (metaplots (left) and boxplots (right)) at RING1B peaks overlapping interactions that persist in the absence of cohesin. (E) Capture-C interaction profiles between the Nkx2-1 promoter and selected proximal and distal RING1B-occupied sites in the Control and SCC1 DEG cells. RING1B ChIP-seq peaks are shown as blue bars below. The location of the Nkx2-1 promoter is indicated with a blue arrow/bar and the interactions sites as black bars on the chromosome. Read density corresponds to normalized reads in the capture averaged across 250 DpnII restriction fragments. (F) Aggregate Capture-C signal in the Control and SCC1 DEG cells at interaction sites segregated based on distance from the capture site. Only interactions between polycomb target gene promoters and RING1B occupied sites present in SCC1 DEG are shown. Read density was normalised to Control signal at the summit and the x-axis illustrates the distance from the interaction site in DpnII fragments. (G) Average Hi-C contact strength in the Control and SCC1 DEG at interactions that persist in the absence of cohesin segregated based on distance between the interactions.

Article Snippet: RAD21, 1:1000 (Abcam, ab154769), RING1B (western) 1:1000 (Klose Lab), RING1B (ChIP-Seq) 1:1000 (Cell Signalling, 5694), CTCF 1:1000 (Abcam, ab70303) and TUBULIN 1:500 (Abcam, ab6046).

Techniques: Hi-C, ChIP-sequencing, Capture-C

Journal: bioRxiv

Article Title: Cohesin disrupts polycomb-dependent chromosome interactions

doi: 10.1101/593970

Figure Lengend Snippet: (A) Capture-C interaction profiles between the Nkx2-1 promoter and selected proximal and distal RING1B-occupied sites in the Control and RING1B DEG cell lines. RING1B ChIP-seq peaks are shown as blue bars below. The location of the Nkx2-1 promoter is indicated with a blue arrow/bar and the interactions sites as black bars on the chromosome. Read density corresponds to normalised reads in the capture averaged across 250 DpnII restriction fragments. (B) Aggregate Capture-C signal in the Control, SCC1 DEG and RING1B DEG cells at interaction sites segregated based on distance from the capture site. Only interactions between polycomb target gene promoters and RING1B occupied sites present in SCC1 DEG are shown. Read density was normalised to Control signal at the summit and the x-axis illustrates the distance from the interaction site in DpnII fragments.

Article Snippet: RAD21, 1:1000 (Abcam, ab154769), RING1B (western) 1:1000 (Klose Lab), RING1B (ChIP-Seq) 1:1000 (Cell Signalling, 5694), CTCF 1:1000 (Abcam, ab70303) and TUBULIN 1:500 (Abcam, ab6046).

Techniques: Capture-C, ChIP-sequencing

Journal: bioRxiv

Article Title: Cohesin disrupts polycomb-dependent chromosome interactions

doi: 10.1101/593970

Figure Lengend Snippet: (A) A schematic illustrating the genotype of the CTCF-AID-GFP cell line . (B) Live cell microscopy images of CTCF-AID-GFP cells ± auxin (48 h). Scale bar = 10 µm (bottom). (C) Aggregate analysis of Hi-C from CTCF-AID-GFP cells ± auxin at interactions that persist in the absence of cohesin (n=336). (D) Capture-C interaction profiles between the Nkx2-1 promoter and selected proximal and distal RING1B-occupied sites in the CTCF-AID-GFP cells ± auxin. RING1B ChIP-seq peaks are shown as blue bars below. The location of the Nkx2-1 promoter is indicated with a blue arrow/bar and the interactions sites as black bars on the chromosome. Read density corresponds to normalised reads in the capture averaged across 250 DpnII restriction fragments. (E) Aggregate Capture-C signal in the CTCF-AID-GFP cells ± auxin at interactions that persist in the absence of cohesin segregated based on distance from the capture site. Only interactions between polycomb target gene promoters and RING1B occupied sites present in SCC1 DEG are shown. Read density was normalised to signal at the summit in CTCF-AID-GFP cells without auxin and the x-axis illustrates the distance from the interaction site in DpnII fragments. (F) Average Hi-C contact strength in the CTCF-AID-GFP cells ± auxin at interactions that persist in the absence of cohesin segregated based on distance between the interactions.

Article Snippet: RAD21, 1:1000 (Abcam, ab154769), RING1B (western) 1:1000 (Klose Lab), RING1B (ChIP-Seq) 1:1000 (Cell Signalling, 5694), CTCF 1:1000 (Abcam, ab70303) and TUBULIN 1:500 (Abcam, ab6046).

Techniques: Microscopy, Hi-C, Capture-C, ChIP-sequencing

Journal: bioRxiv

Article Title: Cohesin disrupts polycomb-dependent chromosome interactions

doi: 10.1101/593970

Figure Lengend Snippet: (A) Capture-C interaction profiles from Hoxd10 (top) and Dlx2 (bottom) viewpoints in Control and SCC1 DEG lines. RING1B ChIP-Seq peaks are displayed as blue bars and TAD intervals as black bars. (B) Representative image of RASER-FISH showing signals classed as in contact (top pair 0.0905 µm apart) and not in contact (bottom pair 1.1629 µm apart). Probes are for Dlx2 (green) and HoxD10 (red). Scale bar = 5 µm. (C) Violin plots showing 3D distance measurements between Dlx2 and HoxD10 in the indicated cells lines. The dashed lines show the median and interquartile range of between n=376-409 cells for each cell line. (D) Absolute contact probabilities showing the percent of signals judged as colocalised from observations in (B) (see Methods).

Article Snippet: RAD21, 1:1000 (Abcam, ab154769), RING1B (western) 1:1000 (Klose Lab), RING1B (ChIP-Seq) 1:1000 (Cell Signalling, 5694), CTCF 1:1000 (Abcam, ab70303) and TUBULIN 1:500 (Abcam, ab6046).

Techniques: Capture-C, ChIP-sequencing

Journal: bioRxiv

Article Title: Cohesin disrupts polycomb-dependent chromosome interactions

doi: 10.1101/593970

Figure Lengend Snippet: (A) A representative western blot for SCC1 and RING1B in the TIR1 and AID-RING1B SCC1-mAID-GFP cell lines ± auxin. Tubulin is shown as a loading control. (B) Capture-C interaction profiles from Hoxd10 (top) and Dlx2 (bottom) viewpoints in Control and SCC1 DEG RING1B DEG cell lines. RING1B ChIP-Seq peaks are displayed as blue bars and TAD intervals are as black bars. (C) Representative Hoxd10 (red) Dlx2 (green) RASER-FISH images from the indicated cell lines. Scale bar = 5 µm. (D) Cumulative frequency distribution of 3D distance measures between Hoxd10 and Dlx2 in the indicated cells lines. Measurements as in .

Article Snippet: RAD21, 1:1000 (Abcam, ab154769), RING1B (western) 1:1000 (Klose Lab), RING1B (ChIP-Seq) 1:1000 (Cell Signalling, 5694), CTCF 1:1000 (Abcam, ab70303) and TUBULIN 1:500 (Abcam, ab6046).

Techniques: Western Blot, Capture-C, ChIP-sequencing

Journal: bioRxiv

Article Title: Cohesin disrupts polycomb-dependent chromosome interactions

doi: 10.1101/593970

Figure Lengend Snippet: (A) An MA plot of gene expression alterations in Scc1-mAID-GFP cells ± auxin (6 h). The number of genes with increased or decreased expression (p-adj < 0.05 and > 1.5-fold) is shown in red. The density of Log2 Fold changes is shown on the right. (B) RING1B binding (+/-1kb from the TSS, blue bars) at gene promoters that show reductions in gene expression following cohesin removal (left) compared to all genes (right). Empirical p-value for RING1B-bound genes enrichment within the downregulated genes: p=0 (n=10000 random tests). (C) The magnitude of gene expression change at expressed RING1B bound genes that do (right) or do not (left) interact with another RING1B bound site in Hi-C. (D) Hi-C (left at 40 kb and right at 10 kb resolution), cRNA-Seq and RING1B ChIP-Seq for two examples of genes with interactions in Hi-C that are strengthened after cohesin removal and whose gene expression decreases.

Article Snippet: RAD21, 1:1000 (Abcam, ab154769), RING1B (western) 1:1000 (Klose Lab), RING1B (ChIP-Seq) 1:1000 (Cell Signalling, 5694), CTCF 1:1000 (Abcam, ab70303) and TUBULIN 1:500 (Abcam, ab6046).

Techniques: Expressing, Binding Assay, Hi-C, ChIP-sequencing

Journal: bioRxiv

Article Title: Cohesin disrupts polycomb-dependent chromosome interactions

doi: 10.1101/593970

Figure Lengend Snippet: (A) Density distribution of Log2 transformed RPKM of transcripts in untreated Scc1-mAID cells. Note the bimodal distribution. Red line marks the threshold (RPKM=0.712) to distinguish expressed from unexpressed genes. (B) Boxplots show transcriptional levels in untreated Scc1-mAID cells. Transcription is shown for expressed genes with RING1B-ocupied promoters that either have no detectable Hi-C interactions in SCC1-AID cells treated with auxin (light blue), or interact with other sites that are RING1B occupied (dark blue). Numbers of genes in each group are indicated below the x-axis.

Article Snippet: RAD21, 1:1000 (Abcam, ab154769), RING1B (western) 1:1000 (Klose Lab), RING1B (ChIP-Seq) 1:1000 (Cell Signalling, 5694), CTCF 1:1000 (Abcam, ab70303) and TUBULIN 1:500 (Abcam, ab6046).

Techniques: Transformation Assay, Hi-C

Journal: BMC Genomics

Article Title: Recent acquisition of imprinting at the rodent Sfmbt2 locus correlates with insertion of a large block of miRNAs

doi: 10.1186/1471-2164-12-204

Figure Lengend Snippet: Methylation of Sfmbt2 CpG Island . A. Five amplicons representing 140 CpG sites across the transcriptional start sites for both the sense and antisense transcripts at the Sfmbt2 locus were queried for methylation in sperm and oocyte genomic DNA by bisulfite mutagenesis. Two amplicons, at the 5' and 3' extremes of the CpG island, Me3 and CG3, were also examined in C57BL6 X Castaneus F1 placenta genomic DNA to determine parental allele methylation. Black circles represent CpGs that survived bisulfite conversion; open circles represent CpGs that were converted to TG; red circles represent asymmetric cytosines in oocyte genomic DNA that survived conversion in most clones; grey circles represent cytosines in a CpG context that were methylated on only the bottom strand in oocyte genomic DNA. Each line represents a unique clone; duplicate clones were not scored. Bisulfite conversion was approximately 95%. The apparent two CpG sites in CG1 are only separated by a single base pair, and are therefore assumed to be a single site. B. Relative location of bisulfite amplicons within CpG island. Hatched boxes represent amplicons for which only one strand was queried; solid box (CG3) was analysed on both the top and bottom strands. Hash marks indicate the relative location of the methylated cytosines in oocyte genomic DNA; black = CpG, red = asymmetric. Of 25 methylated cytosines, 18 were confirmed to be asymmetric, including the four CpG sites in CG3 illustrated in panel A.

Article Snippet: The Broad Institute ChIP-seq database indicates that there is very strong association of the CpG island with Ring1b, EZH2, SUZ12, H3K4 Me2/3 and H3K27Me3, but not H3K9 Me2 or Me3, nor H4K20Me3 in ES cells.

Techniques: Methylation, Mutagenesis, Clone Assay