Figure 6

CTCF-binding sites within the Ushg1 - Otop2 locus. The coordinates of the depicted genomic region in the mouse genome (assembly NCBI/mm9) are Chr11:115168200-115193650. (a) Position of mapped sequence reads from ChIP-seq studies using an anti-CTCF antibody and the following cells: mouse E14 ES cell line [37]; human H1-hESC ES cell line (Duke/UNC/UT-Austin/EBI ENCODE group; http://genome-test.cse.ucsc.edu); resting human CD4+T cells [38]; and five non-cancerous, karyotypically normal human cell lines (HRE, human renal epithelial cells; BJ, skin fibroblasts; HUVEC, human umbilical vein endothelial cells; SAEC, small airway epithelial cells; and NHEK, normal human epidermal keratinocytes [Duke/UNC/UT-Austin/EBI ENCODE group]). For display purposes, the coordinates of the human CTCF-binding fragments CTCF2 and CTCF3 are presented based on the coordinates in the mouse genome. Based on these data, four CTCF-binding sites (CTCF1 to CTCF4) were identified with the following species-specific occupancy: CTCF1 and CTCF4, mouse-specific; CTCF2, human-specific; and CTCF3, both mouse and human. The mouse Ushg1 and Otop2 gene structures (b) and CpG-island content (c) were derived from the RefSeq and CpG island tracks of the UCSC Genome Browser (blue and green boxes, respectively). Black boxes represent non-coding MCSs identified by both ExactPlus and PhastaCons (d) or ExactPlus only (e), respectively. The open red boxes highlight the position of the CTCF-binding motifs. (f) Sequence Logos for CTCF1 to CTCF4 graphically represent the multi-sequence alignment at the CTCF-binding sites in placental mammals; the height of each symbol reflects the relative frequency of that nucleotide at that position. (*) Indicates that hominoid sequences were not considered for the Logo generation of CTCF1 due to lack of motif conservation (i.e., hominoid-specific deletion of base 9). Consensus Logo motifs for low-, medium-, and high-occupancy CTCF-binding sites (LowOc, MedOc and HighOc, respectively) are also shown; these classes are based on the degree to which the CTCF-binding sites match the known CTCF-binding motif and the densities of sequence reads mapped at the binding sites [40, 41].