Skip to main content
. Author manuscript; available in PMC: 2019 Sep 20.
Published in final edited form as: Cell. 2018 Aug 30;175(1):224–238.e15. doi: 10.1016/j.cell.2018.08.005

Figure 6. CTCF occupancy is lost and FMR1 is silenced when the FMR1 daSTR boundary is disrupted in FXS patients.

Figure 6.

(A-D) 5C and ChIPseq in B-lymphocytes from a male FXS patient with ~935 CGG repeats (Coriell Catalog ID GM09237), his unaffected malesibling (Coriell Catalog ID GM09236) and a male FXS patient with ~645 CGG repeats (Coriell Catalog ID GM04025). (A) Heatmap of CTCF occupancy. (B) Zoom in on two CTCF peaks with differential occupancy upstream of the FMR1 gene. Red arrow, forward CTCF motif, blue arrow, reverse CTCF motif. (C) Global 5C contact matrix in healthy B-lymphocytes showing topological context of the zoom boxes in (D), (E), and (I). The FMR1 gene is highlighted in green and the repeat with a red vertical line. (D-E) Zoom-ins on disrupted loops (boxed in green) anchored by the differential CTCF sites. A loop to a locus (D) upstream and (E) downstream of the FMR1 daSTR dissolves in for FXS patients compared to the unaffected sibling. (F) Sanger traces of CTCF motifs across samples. (G) Quantitative RT-PCR analysis of FMR1 expression. Error bars, +/− SEM (n=3 independent experiments). (H) Scatterplot of relative gene expression versus boundary strength as assessed by the directionality index at the FMR1 daSTR compared to FMR1 expression. (I) Log fold change map of 5C contacts in diseased vs WT B-lymphocytes (Coriell Catalog ID GM09237 and GM09236, respectively) with GM12878 H3K27ac tracks from ENCODE shown below. The FMR1 gene is highlighted in green and the repeat with a red vertical line. See also Figure S7.