Discovering a binary CTCF code with a little help from BORIS

Abstract

CCCTC-binding factor (CTCF) is a conserved, essential regulator of chromatin architecture containing a unique array of 11 zinc fingers (ZFs). Gene duplication and sequence divergence during early amniote evolution generated the CTCF paralog Brother Of the Regulator of Imprinted Sites (BORIS), which has a DNA binding specificity identical to that of CTCF but divergent N- and C-termini. While healthy somatic tissues express only CTCF, CTCF and BORIS are normally co-expressed in meiotic and post-meiotic germ cells, and aberrant activation of BORIS occurs in tumors and some cancer cell lines. This has led to a model in which CTCF and BORIS compete for binding to some but not all genomic target sites; however, regulation of CTCF and BORIS genomic co-occupancy is not well understood. We recently addressed this issue, finding evidence for two major classes of CTCF target sequences, some of which contain single CTCF target sites (1xCTSes) and others containing two adjacent CTCF motifs (2xCTSes). The functional and chromatin structural features of 2xCTSes are distinct from those of 1xCTS-containing regions bound by a CTCF monomer. We suggest that these previously overlooked classes of CTCF binding regions may have different roles in regulating diverse chromatin-based phenomena, and may impact our understanding of heritable epigenetic regulation in cancer cells and normal germ cells.

Keywords: 1xCTS; 2xCTS; BORIS; CTCF; CTCFL; ChIP-seq; chromatin.

Figure 1.

The DNA binding domains of CTCF and BORIS are highly conserved and bind to shared and distinct sites in the human genome. (A) Schematic representations of the genomic architecture of the CTCF and BORIS genes. C- and N-terminal exons are represented by thin blue (CTCF) or yellow (BORIS) lines. Conserved ZF-encoding exons are represented by thick colored boxes, with exons corresponding to each ZF colored differently and the exon(s) numbers according to which ZF they encode. ZFs 4, 7, 9, and 11 are split between two exons; in these cases, the ZF number is written above the intervening intron. Portions of exons corresponding to ZF linkers are represented by thin grey boxes. Introns are represented by black lines. (B) Alignment of the 11-ZF DNA binding domains of CTCF and BORIS. Zinc fingers are in bold and numbered. Zinc-coordinating residues are highlighted in blue and DNA-contacting residues are highlighted in gold. The asterisk (*) indicates amino acid conservation, the colon (:) represents a strongly conservative amino acid substitution, and the period (.) represents a weakly conservative amino acid substitution. The alignment was generated with Clustal Omega. (C) Human K562 cell ChIP-seq and ChIP-reChIP-seq data demonstrating the presence of two major classes of CTCF and BORIS binding regions. A robust CTCF&BORIS-bound 2xCTS element at the PRSS50 (TSP50) gene promoter has been characterized in detail in an earlier study and verified by analysis of DNase I footprinting data as well as ChIP-reChIP-seq (BORIS ChIP with antibodies verified for lack of CTCF crossreactivity followed by CTCF reChIP-seq) is highlighted in blue. Two CTCF-only peaks are highlighted in yellow, and a BORIS-only peak is highlighted in green.

Figure 2.

2xCTSes are resolved by nuclease footprinting. (A) UCSC Genome Browser comparison of ChIP-seq and DNase I footprinting data at a selected 1xCTS and 2xCTS regions. K562 CTCF and BORIS ChIP-seq and NHDF BORIS ChIP-seq data are from Pugacheva et al. NHDF CTCF ChIP-seq and DNase I overlap signal data are from the ENCODE consortium. DNase I footprints are boxed in red. (B) Heatmaps showing digital genomic footprinting (DGF) data at 1xCTSes in BORIS-positive K562 cells and 2xCTSes in BORIS-negative NHDF cells and K562 cells. A model of the organization of CTSes and bound molecules within each class of region is shown. (C) Schematic of a possible explanation for the failure of standard ChIP-seq to resolve individual binding sites within 2xCTSes.