Skip to main content
. 2017 Mar 17;6:e21064. doi: 10.7554/eLife.21064

Figure 1. Biochemical properties of PCGF6-PRC1 and its target genes in ESCs.

(A) Affinity purification of PCGF6-containing complexes in ESCs. To purify PCGF6 and associated proteins, a mouse ESC cell line stably expressing Flag-2xStrepII (FS2)-tagged PCGF6 was generated. Nuclear extract was isolated from this cell-line, PCGF6 was affinity purified, and the purified proteins were subjected to mass spectrometry. Purified PCGF6 fractions were resolved by gradient SDS-PAGE and visualized by SyproRuby staining. The purifications were performed in the absence and presence of benzonase (Benz) to exclude DNA-mediated interactions and a cell line containing only the empty vector was used as control for non-specific binding to the affinity matrix. The elutates were probed by western blot for streptavidin (Strep). (B) Identification of proteins that form stable complexes with PCGF6 in ESCs. Elutions from the PCGF6 affinity purification were directly analyzed by tryptic digestion followed by peptide identification by LC-MS/MS. The Mascot scores and peptide coverage are shown for the respective affinity purifications. (C) Confirmation of PCGF6-containing complexes by immunoprecipitation-immunoblot (IP-IB) analysis. Whole-cell extracts (WCE) from ESCs expressing FLAG-tagged PCGF6 or RINGB were subjected to IP using anti-FLAG antibody. The WCE and immunoprecipitates were separated on SDS-PAGE and analyzed by IB with the indicated antibodies. (D) Screenshot views for the distribution of PCGF6 (red) and RING1B (blue) at target genes in ESCs determined by ChIP-seq. The chromosomal positions are indicated on the x-axis. The transcription start sites (TSSs) are denoted by arrows. (E) Venn diagram representation for the overlap of PCGF6, RING1B and H3K27me3 target genes in ESCs identified by ChIP-seq. The number of genes bound by PCGF6, RING1B and H3K27me3 and included in each fraction are indicated. (F) Venn diagram representing the overlap of PCGF6, RING1B and CBX7 target genes. Published ChIP-seq data for CBX7 was obtained from NCBI GEO (accession number GSM1041373). (G) A heat map view for distribution of PCGF6, RING1B, CBX7, MAX, KDM2B and H3K27me3 in ±4 kb genomic regions around transcription start sites (TSS). Genes are classified based on their occupancy by PCGF6, RING1B and CBX7 in ESCs. The signal from a negative control (NC: FLAG-ChIP in mock transfected ESCs) was also shown.

DOI: http://dx.doi.org/10.7554/eLife.21064.002

Figure 1—source data 1. Raw data for LC-MS/MS analysis shown in Figure 1B.
DOI: http://dx.doi.org/10.7554/eLife.21064.003
elife-21064-fig1-data1.xlsx (17.3KB, xlsx)
DOI: 10.7554/eLife.21064.003

Figure 1.

Figure 1—figure supplement 1. Generation of a Pcgf6 conditional allele and properties of CpG islands at PCGF6-PRC1 target genes.

Figure 1—figure supplement 1.

(A) Schematic representation of the construct for conditional targeting of the Pcgf6 locus. The targeting construct contains an FRT (closed arrows)-flanked neomycin resistance gene (neo), and the second and the third exons (closed rectangles) of the mouse Pcgf6 gene are flanked by two loxP sites (open triangles). (B) Genomic PCR using the indicated primers demonstrating the kinetics of the excision of the loxP-flanked region in Pcgf6fl/fl; Rosa26::CreERT2tg/+ ESCs after OHT treatment. (C) Comparison of the PCGF6 ChIP-seq data in this study with those reported in a previous paper (Yang et al., 2016). (D) ChIP-qPCR data showing a strong binding of FLAG-tagged PCGF6 to representative PCGF6 targets (Ddx4, Tdrd1, Spo11 and Stag3) determined by ChIP-seq analysis shown in Figure 1 and a marginal binding to known canonical PRC1 targets (underlined). (E) Comparison of CpG island (CGI) length (left) and 5-methyl-cytosine (5mC) levels (right) of CBX7-, RING1B-, and/or PCGF6-bound genes. The number of genes included in each subset is shown at the bottom. (F) A heat map view of RING1B distribution in ±4 kb genomic regions around TSS in wild type (WT), and Kdm2b-KO ESCs. RING1B-bound genes are further subclassified by binding of PCGF6 and CBX7 (left). Average distribution and p-value for its change in Kdm2b-KO are also shown (right). (G) Gene expression changes in Kdm2b-KO ESCs. Box plot representation of gene expression changes induced by deletion of Kdm2b for genes bound by CBX7, PCGF6 and/or RING1B. The box plots represent the median (horizontal line), interquartile range (box), range (whiskers), and outliers (circles). The number of genes included in each subset is shown at the bottom. The p-values for the difference of expression changes between the indicated two groups were determined by the Student’s t-test and are indicated above each graph.
DOI: http://dx.doi.org/10.7554/eLife.21064.004