الانتقال إلى مراجعة PREreview

مراجعة PREreview لChromatin boundary permeability is controlled by CTCF conformational ensembles

نُشرت
DOI
10.5281/zenodo.24355003
الترخيص
CC BY 4.0

Overall assessment

In this manuscript, the authors report a mechanism of chromatin context dynamically shaping CCCTC-binding factor (CTCF) conformations in controlling the boundary permeability in cohesin-mediated loop extrusion beyond physical DNA-protein affinity. The authors reveal and characterize the diverse and sequential conformation of CTCF on DNA. In details, the peripheral unwrapped CTCF-DNA state, captured by nanopore sequencing GpC footprinting assay, shows higher cohesin occupancy. Molecular dynamics (MD) simulation confirms the high mobility of peripheral ZFs. Fitting the Hidden Markov Model (HMM) to single-molecule optical tweezer assays revealed an ensemble of CTCF conformations with DNA and matches with the unwrapped footprinting in cells with the probability model of CTCF stalling in cells. Changing the chromatin context reshapes the CTCF conformations, where PDS5A can dock on CTCF and prime CTCF for more dominant unwrapped conformation. Depletion of PDS5A weakens the contacts at PDGFRA insulator at CTCF.

Overall this is an outstanding and comprehensive study that addresses a long-standing question of CTCF boundaries permeability. This manuscript has innovative assays and high-quality data and proposes an evidence-backed and compelling model. While there are minor areas where data presentation and interpretation could be enhanced, these should be easy to address.

Major comments:

  1. The connection between CTCF conformation, PDS5 docking and stabilized loop anchors: In figure 3G, unzipping assays and HMM state modeling demonstrate that PDS5A engages with DNA-bound CTCF and shifts conformation towards unwrapped conformation II. In figure 4, rupture measurements demonstrate that PDS5 stabilized the loop anchors. However, there is a distinction to be made between PDS5-induced CTCF conformational shifts capturing cohesin and PDS5 acting on cohesin at CTCF, which can be independent of CTCF conformational shift. Meanwhile, a direct measure of cohesin kinetics in different conditions can provide a stronger link between CTCF conformation and cohesin capturing. By performing unzipping assays with sequential addition of CTCF and Cohesin/PDS5+Cohesin, the distinction between priming and stabling can be made.

  2. To test the author’s main claim that PDS5 engagement with CTCF enhances cohesin capture, it would be good to show that cohesin dwell time at CTCF increases by the priming of PDS5, e.g., as the authors do for PDS5 in Figure 3E. Can you perform the experiment as shown in Figure 3D and E, and then add cohesin with or without PDS5 priming first to determine whether PDS5 enhances cohesin-CTCF colocalization. This experiment can provide strong evidence of PDS5 priming CTCF as a stronger impediment before the engagement of cohesin, rather than CTCF better captures PDS5-Cohesin.

  3. In supplementary figure S6, authors determine the optimal number of hidden states for CTCF conformations across all conditions. The choice of optimal numbers looks most affected by one condition, CTCF WT + cohesin + PDS5. A more rigorous analysis of the number of possible states at each condition can strengthen the proposed model of how CTCF conformation affects boundary permeabilities. The number of states included can alter the results and the interpretation of CTCF conformations across conditions.

Minor comments:

  1. For figure 1C on page 5, consider adding text to support and explain the estimation calculations for “At CTCF-bound reads, protected molecules are estimated to make up about 60%...”, as the proportion is calculated across all motifs and the heatmap displays only one example motif. Currently, the link to the methods section does not explain the calculation.

  2. For figure 1E, to show that high cohesin-occupied CTCF has a distinct pattern from low cohesin-occupied CTCF, consider performing a paired t-test across all zinc fingers/clusters, instead of the position value per cluster. Also, additional text in the legend or methods to explain the reasoning behind grouping would improve interpretation.

  3. For figure 4G, the heatmap should be quantified to show that the loop anchor is destabilized, as opposed to the overall interaction along the loop extrusion trajectory (i.e. stripe vs dot, anchor vs continuous extrusion) is weakened or both need to be discussed. Adding a genome-wide quantification of the change in boundary strength before and after PDS5A depletion would support the claim that PDS5 primes CTCF for cohesin capture as a more universal mechanism.

  4. To test their model that flanking Zn fingers are dissociated from DNA to facilitate association with cohesin, the authors could consider making mutants in those Zn fingers (or in the DNA sequence that binds these fingers) that reduce DNA binding affinity. The prediction would be that this would favor conformation II, which in turn should increase the affinity for PDS5 and make a stronger impediment.

I am one of 2 participants from the same research lab that reviewed this preprint as part of the Transparent and Accountable Peer Review training program. This report was written with the support of my research mentor.

Competing interests

The authors declare that they have no competing interests.

Use of Artificial Intelligence (AI)

The authors declare that they did not use generative AI to come up with new ideas for their review.