Dynamics of 3D genome organization: the role of CTCF and cohesin

Mammalian interphase genomes are functionally compartmentalized into topologically associating domains (TADs) spanning hundreds of kilobases. TADs are defined by frequent chromatin interactions within themselves and they are insulated from adjacent TADs. Most TAD or domain boundaries are strongly enriched for CTCF, an 11-zinc finger DNA-binding protein, and cohesin, a ring-shaped multi-protein complex that is thought to topologically entrap DNA. The subset of TADs which are folded into loops are referred to as loop domains and tend to be demarcated by convergent CTCF-binding sites. Targeted deletions of CTCF-binding sites as well acute depletion of either CTCF or cohesin have demonstrated that both CTCF and cohesin are causally required for the formation of loop domains. Moreover, disruption of loop domain boundaries by deletion or silencing of CTCF-binding sites allows abnormal contact between previously separated enhancers and promoters, which can induce aberrant gene activation leading to cancer or developmental defects. Thus, TADs and loop domains are critical for ensuring proper regulation of gene expression and CTCF and cohesin are the master regulators.

However, one limitation of most previous studies is that they have mainly used snap-shot techniques such as Hi-C. Thus, how they are formed and maintained remains unclear and, in particular, it was unclear whether chromatin loops are dynamic or static structures. To overcome these issues, we recently combined genome-editing to endogenously tag CTCF and cohesin with genomics and single-molecule super-resolution imaging in live cells to explore their dynamics. By ChIP-Seq, CTCF and cohesin co-occupy the same sites and by co-IP they physically interact as a biochemically stable complex – we refer to this complex as a Loop Maintenance Complex (LMC). However, using single-molecule imaging and FRAP we found that CTCF binds chromatin much more dynamically than cohesin (~1-4 min vs. 20-25 min residence time). Moreover, their search mechanisms are different: after unbinding, CTCF quickly rebinds another cognate site unlike cohesin for which the search process is long (~1 min vs. ~33 min). Finally, using super-resolution STORM imaging we confirmed that CTCF and cohesin really do co-localize at the single-molecule level inside the nucleus.

Thus, to reconcile these observations we propose a dynamic LMC model: CTCF and cohesin form a dynamic complex on chromatin with a molecular stoichiometry that changes over time since CTCF is more dynamic than cohesin. Moreover, since even cohesin does eventually dissociate, we propose that CTCF/cohesin mediate chromatin loops are dynamic structures with a life-time of minutes to tens of minutes. Our results argue against stable loops and we propose that that chromatin loops are dynamic and frequently break and reform throughout the cell cycle.

These studies were recently published in eLife: https://elifesciences.org/articles/25776

We also recently reviewed complementary evidence for dynamic TADs and loops: http://www.tandfonline.com/doi/abs/10.1080/19491034.2017.1389365

Now we are extending our studies and following up on several questions and observations:

  • Can we standardize single-molecule tracking experiments and how do we evaluate analysis methods? In close collaboration with Maxime Woringer, we validated an open-source platform (“Spot-On”) for single particle tracking analysis. Please see https://spoton.berkeley.edu/  
  • We previously observed that CTCF and cohesin form clusters inside the nucleus. In collaboration with Claudia Cattoglio, we are asking, what is the function of these clusters and how are they regulated?
  • In collaboration with Gina Dailey, we aim to develop tools to directly visualize chromatin loops inside live cells in real time?

Anders Hansen – October 2017