henhaobio
Sign in

ScienceMolecular BiologyPeer-reviewed journal article

Rules of engagement for condensins and cohesins guide mitotic chromosome formation

Kumiko SamejimaiD, Johan Harmen GibcusiD, Sameer AbrahamiD, Fernanda Cisneros-SoberanisiD, Itaru SamejimaiD, Alison J. BeckettiD, Nina PucekovaiD, Maria Alba AbadiD, Christos SpanosiD, Bethan Medina‐PritchardiD, James R. PaulsoniD, Linfeng XieiD, Arockia Jeyaprakash ArulanandamiD, Ian A. PrioriD, Leonid A. MirnyiD, Job DekkeriD, Anton A. GoloborodkoiD, William C. EarnshawiD

PDFDOIPreprint
In-depth full-text digest available · Jump to digest ↓

Abstract

During mitosis, interphase chromatin is rapidly converted into rod-shaped mitotic chromosomes. Using Hi-C, imaging, proteomics and polymer modeling, we determine how the activity and interplay between loop-extruding SMC motors accomplishes this dramatic transition. Our work reveals rules of engagement for SMC complexes that are critical for allowing cells to refold interphase chromatin into mitotic chromosomes. We find that condensin disassembles interphase chromatin loop organization by evicting or displacing extrusive cohesin. In contrast, condensin bypasses cohesive cohesins, thereby maintaining sister chromatid cohesion while separating the sisters. Studies of mitotic chromosomes formed by cohesin, condensin II and condensin I alone or in combination allow us to develop new models of mitotic chromosome conformation. In these models, loops are consecutive and not overlapping, implying that condensins do not freely pass one another but stall upon encountering each other. The dynamics of Hi-C interactions and chromosome morphology reveal that during prophase loops are extruded in vivo at [~]1-3 kb/sec by condensins as they form a disordered discontinuous helical scaffold within individual chromatids.

In-depth summary

Claude Opus 5.5 · max effort · 8 Nov 2026 · from the full text of version 3 (30 Apr 2024) · CC BY 4.0

In DT40 cells, condensins displace extrusive cohesin, bypass cohesive cohesin and stall against each other while extruding ~1-3 kb/sec, yielding irregular helical chromatids of consecutive loops.

During prophase, cohesin-organized interphase chromatin is replaced by condensin loop arrays, but how condensins resolve collisions with cohesins and with each other in metazoa was underexplored, as SMC bypassing had been described for bacterial complexes and yeast condensin. Degron lines leaving specific SMC complexes active during synchronous mitotic entry let this work define those interaction rules and model chromatids built by each condensin without cohesin.

Key results

  1. ShownCondensin is needed for the rapid prophase loss of compartments and of cohesin-built TADs and CTCF-CTCF dots; simulations indicate it unloads or pushes extrusive cohesin, and condensin II appears to be the main driver.

    In WT cells, compartment, dot and TAD strength decayed within 10 min of G2 release, vanishing by nuclear envelope breakdown, but persisted to late prometaphase in SMC2-depleted cells; of 4 simulated collision rules (unloading, pushing, bypass, stalling), only unloading and pushing reproduced the fast loss; ChEP showed delayed CTCF and cohesin loss without condensin. · Condensin drives the disassembly of interphase…; Rule 2. Condensins remove extrusive cohesins

  2. SuggestedResidual cohesin on prometaphase chromosomes sits between sister chromatids, away from each chromatid's condensin axis; only polymer models in which condensin steps over cohesive cohesin reproduce this, whereas stalling or pushing yields one shared axis.

    Spike-in ChEP found ~3-6 cohesin complexes per Mb still bound at late prometaphase; SMC3-Clover/SMC2-Halo line scans (10 chromosomes in each of two replicates) put this cohesin between sisters, off the condensin axes; bypass simulations gave individualized bottlebrushes linked by distal cohesin, whereas stalling colocalized both complexes. · Substantial levels of cohesin remain on…; Cohesin localizes between sister chromatids away…; Condensin must bypass cohesive cohesins; Figure legends

  3. ShownCohesive rather than extrusive cohesin reshapes prometaphase chromosomes: removing it before mitosis shifts and sharpens the Hi-C second diagonal (a helical-turn signature) and yields shorter, wider chromatids, even when extrusive cohesin is reloaded in G2.

    SMC3 depletion shifted the prometaphase second diagonal from ~6 to ~8 Mb; cohesed WT chromosomes were 1.1 µm wide (0.55 µm per sister), cohesin-free single chromatids 0.8 µm; of 4 double-synchronized cultures, the three lacking cohesive cohesin gave indistinguishable t= 30 min Hi-C, and cohesin restored after S phase rebuilt TADs and dots but not cohesion. · Cohesin depletion significantly alters mitotic…; Cohesive, but not extrusive cohesin, shapes…

  4. SuggestedCondensin II alone forms an irregular helical bottlebrush of ~400 kb consecutive loops; condensin I alone forms a non-helical, randomly folded bottlebrush of ~100 kb loops; merged parameters reproduce chromatids with both, implying additive nested loops and condensins that stall on meeting.

    Condensin II-only Hi-C had diagonals near 16, 32 and ~40-50 Mb; best fit: 400 kb loops, 80 nm gaps, 400 nm pitch holding 17 Mb. Condensin I-only P(s) fell as s^-1.5 at 2-8 Mb; fit: ~100 kb loops, 20 nm gaps. Overlapping-loop (mutual bypass) models scattered condensin, unlike the observed interior scaffold; EdU exchange gyre heights matched the SMC3-AID model pitch. · Roles of individual condensin complexes in…; Structural models for the roles of condensins in…; Rule 3. When condensins encounter one another…

  5. SuggestedTime-resolved Hi-C in synchronized cells with a single active condensin yields in vivo extrusion rates of about 1-3 kb/sec for condensin II and ~1 kb/sec for nuclear condensin I, while helical period growth outpaces extrusion, pointing to an additional condensin II spiraling process.

    Condensin II-only cells: ~200-300 kb loops at t= 2.5 min (~1.3-2 kb/sec), 400 kb at 5 min (~1.3 kb/sec), 0.5-1.3 kb/sec from the change between 2.5 and 5 min, and 2.5-3 kb/sec from rod formation by ~10 min; condensin I-only loops of ~300-400 kb by 7.5 min (~1 kb/sec). Condensin II-only spiral period rose from ~6.6 to 16.4 Mb over 15-30 min (10.9 kb/sec). · In vivo estimation of the extrusion speed; Spiraling dynamics

Methods, data and reproducibility

DT40 CDK1^as cells with auxin-inducible degrons of cohesin and/or condensin subunits were released synchronously from a 1NM-PP1 G2 block and profiled by time-resolved Hi-C, chromatin-enrichment proteomics (ChEP), imaging and serial block-face scanning EM (SBF-SEM); polymer simulations and fitted chromatid models were scored against Hi-C P(s) curves.

  • Time-resolved Hi-C of WT and degron DT40 CDK1^as lines from G2 arrest through prometaphase (2.5 to 30 min after release) · DT40 vertebrate cell line (chromosome Z identified) · GEO GSE262525; BioProject PRJNA1091327
  • ChEP proteomics (SILAC ratios, iBAQ counts normalized to histone H4, Halo-H4 spike-in absolute counts) of chromatin-bound SMC complexes during mitotic entry · DT40 vertebrate cell line · n = n= 6 (SILAC/iBAQ); n= 4 (spike-in)
  • SBF-SEM whole-cell 3D reconstructions of prometaphase chromosomes in WT, SMC3-AID, SMC3-AID/CAP-H-AID and SMC3-AID/CAP-H2-AID cells · DT40 vertebrate cell line · EMPIAR-11919
  • Live-cell imaging of Halo-lamin B1 and 3xGFP-NES reporter cells timing chromatin condensation and nuclear envelope breakdown · DT40 vertebrate cell line
  • Fixed-cell line scans of SMC3-Clover and SMC2-Halo across prometaphase chromosomes · DT40 vertebrate cell line · n = 10 chromosomes in each of two replicate experiments
  • Nocodazole/hypotonic chromosome spreads used to measure chromosome 1 length across WT and degron lines at t= 30 min · DT40 vertebrate cell line · n = ≥16 chromosomes or chromatids from 3 independent experiments
  • EdU-labeled harlequin spreads with partial sister chromatid exchanges used to measure gyre height · DT40 vertebrate cell line · n = SMC3-AID: n=4, 92 measurements; WT: n=3, 81 measurements
Methods & tools
Hi-C with P(s) contact-frequency decay analysis, Auxin-inducible degron (AID) system with OsTIR1, CDK1^as with 1NM-PP1 G2 block and release, Palbociclib (CDK4/6 inhibitor) G1/S synchronization, ChEP proteomics with SILAC and iBAQ quantification, Halo-histone H4 spike-in absolute protein quantification, HaloTag ligands JFX646 and JFX549, SBF-SEM with DNA contrast-enhancing labeling, Loop-extrusion polymer simulations of cohesin-condensin collision rules, Bottlebrush chromatid polymer models fitted to Hi-C P(s), EdU sister chromatid exchange labeling, FACS sorting
Reagents & environment
DT40 CDK1^as lines expressing OsTIR1 with AID alleles of SMC3, SMC2, CAP-H and/or CAP-H2 (Table S1); tagged SMC3-Clover, SMC2-Halo, Halo-CAP-H2, Halo-lamin B1 and 3xGFP-NES reporters; JFX646 and JFX549 Halo ligands; 1NM-PP1, palbociclib, auxin.
Experimental conditions
SBF-SEM whole-cell reconstructions; FACS sorting by GFP; live imaging of seven z-sections (1 µm interval) every 1.5 min; palbociclib block of 11 h plus 3 h; 0.5 µg/ml nocodazole for 30 min and 75 mM KCl for 10 min before spreading.
Reproducibility gaps
GEO Hi-C series GSE262525 to be released only on publication; no repository or URL is given for polymer simulation or model-fitting code; Methods and supplementary tables, figures and movies are referenced but not part of the supplied text.
Limitations (6)
  • Spike-in ChEP counts of chromatin-bound cohesin cannot separate arm from centromeric cohesin, or cohesive from extrusive complexes, so the residual ~3-6 per Mb is not assigned by location or function. · stated by the authors
  • Extrusion speeds are approximations: when condensin II switches on after 1NM-PP1 reversal is unknown, and the loop-size change between 2.5 and 5 min likely underestimates speed because loops may have peaked within that interval. · stated by the authors
  • Modeling leaves open whether condensin pushes extrusive cohesin aside or facilitates its unloading; neither process has been seen in single-molecule assays, and the authors favor unloading because condensins are weak motors and ChEP shows faster cohesin loss when condensin is present. · stated by the authors
  • Light-microscopy chromosome dimensions vary strongly with spreading, nocodazole or hypotonic treatment and fixation, and are resolution-limited, so chromosome volumes for modeling had to come from SBF-SEM. · stated by the authors
  • All perturbations use one DT40 CDK1^as background with chemical G2 arrest and release; whether the same engagement rules and extrusion rates apply in other vertebrate cells, such as mammalian lines, is not tested here. · noted in this summary
  • Bypass, stalling and nested-loop conclusions rest on agreement of multi-parameter polymer models with population-averaged Hi-C and imaging; SMC collisions are not observed directly, so other mechanisms giving similar contact profiles are not excluded. · noted in this summary

Read this if you study how cohesin and condensin interact to build mitotic chromosomes, or need in vivo condensin extrusion rates and Hi-C-constrained polymer models of chromatid loop and helix architecture.

Citations
80
FWCI
21.52
Field percentile
top 1%

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISYunusova '25Schalbetter '17Abramo '19Zhao '25Golfier '19Dey '22
81% semantic matchGenomics
Anastasia Yunusova, Egor Tsoi, A. V. Smirnov+9 more·bioRxiv·Aug 8, 2025· 1 citations