Genome folding by loop extrusion and compartmentalization

Leonid Mirny
!
leonid@mit.edu
Genome folding by loop extrusion
and compartmentalization
!

Inac%ve
Ac%ve
A/B Compartments
0.1-1 Mb
Topological Associa%on Domains (TADs)
Nora et al. Nature 2012; Dixon et al Nature 2012
2-3 fold
Multiple levels of organization:  
compartments and domains
Chr4: Mb
Rao et al. Cell 2014
active
inactive
Irina Solovei, LMU

HiGlass.io
by Nils Gehlenborg, Peter Karpedjiev, Nezar Abdennur, and others
Harvard University and MIT
HiGlass: Web-based Visual Exploration and Analysis of Genome Interaction Maps 
https://www.biorxiv.org/content/early/2017/10/30/121889

Outline
Compartments and TADs
1. Emerging evidence of loop extrusion!
2. Compartmentalization by phase separation
3. Phase separation vs extrusion
Emerging Evidence of Chromosome Folding by Loop Extrusion 

Model
36 domains=10MbC
D
Loop extrusion with boundaries => TADs
e
Formation of Chromosomal Domains by Loop Extrusion!
bioRxiv Aug 14 (2015)!
Fudenberg, Imakaev et al.!
DOI: 10.1101/024620
Article
Formation of Chromosomal Domains by Loop
(2016)

Model
36 domains=10MbC
D
Loop extrusion with boundaries => TADs
genomic distance, s (bp)
104
105
1
normalized
C
10-2
D
e
Formation of Chromosomal Domains by Loop Extrusion!
bioRxiv Aug 14 (2015)!
Fudenberg, Imakaev et al.!
DOI: 10.1101/024620
Article
Formation of Chromosomal Domains by Loop
(2016)

contact
frequency
46Mb
50Mb
48Mb
C
i
ii
iii
iv
iv
iviv
chr5, NPC (Bonev et al., 2017)
How does loop extrusion make TADs/ﬂames/peaks etc?
corner peakflame (track)insulation peak gridi ii iii iv
C
Scanning! <10% <10%

b
00.0100.0200.035
ContactProbability(simulations)
= 13
5’-GGCGGAGACCACAAGGTGGCGCCAGATCCC-3’
17.417.6
1 kb resolution
CTCF
RAD21
SMC3
Chr 1
Chr1
17.6 Mb17.4
0 0.5 1 1.5 2 2.5 3 3.5 4
Number of PeaksD
Forwardmotif
FoldChange
0
0.5
1.0
1.5
0% 20% 40% 60% 80% 100%
Percentage of peak loci bound
YY1
CTCF
RAD21
(2%)(3%)(3%)(92%)
CCACNAGGTGGCAGconsensus
x 1000
CTCF anchor
(arrowhead indicates
motif orientation)
Loop domain
Ordinary domain
290 Kb
110
Kb
190 Kb
350 Kb
270 Kb
130 Kb
450 Kb
170
Kb
F
Figure 6. Many Loops Demarcate Contact Domains; The Vast Majority of Loops Are Anchored at a Pair of Convergent CTCF/RAD21/SMC3
Binding Sites
(A) Histograms of corner scores for peak pixels versus random pixels with an identical distance distribution.
(B) Contact matrix for chr4:20.55 Mb–22.55 Mb in GM12878, showing examples of transitive and intransitive looping behavior.
(C) Percent of peak loci bound versus fold enrichment for 76 DNA-binding proteins.
(D) The pairs of CTCF motifs that anchor a loop are nearly all found in the convergent orientation.
(legend continued on next page)
1674 Cell 159, 1665–1680, December 18, 2014 ª2014 Elsevier Inc.
Border-to-border loops  
cannot reproduce Hi-C data

b
00.0100.0200.035
ContactProbability(simulations)
= 13
5’-GGCGGAGACCACAAGGTGGCGCCAGATCCC-3’
17.417.6
1 kb resolution
CTCF
RAD21
SMC3
Chr 1
Chr1
17.6 Mb17.4
0 0.5 1 1.5 2 2.5 3 3.5 4
Number of PeaksD
Forwardmotif
FoldChange
0
0.5
1.0
1.5
0% 20% 40% 60% 80% 100%
Percentage of peak loci bound
YY1
CTCF
RAD21
(2%)(3%)(3%)(92%)
CCACNAGGTGGCAGconsensus
x 1000
CTCF anchor
(arrowhead indicates
motif orientation)
Loop domain
Ordinary domain
290 Kb
110
Kb
190 Kb
350 Kb
270 Kb
130 Kb
450 Kb
170
Kb
F
Figure 6. Many Loops Demarcate Contact Domains; The Vast Majority of Loops Are Anchored at a Pair of Convergent CTCF/RAD21/SMC3
Binding Sites
(A) Histograms of corner scores for peak pixels versus random pixels with an identical distance distribution.
(B) Contact matrix for chr4:20.55 Mb–22.55 Mb in GM12878, showing examples of transitive and intransitive looping behavior.
(C) Percent of peak loci bound versus fold enrichment for 76 DNA-binding proteins.
(D) The pairs of CTCF motifs that anchor a loop are nearly all found in the convergent orientation.
(legend continued on next page)
1674 Cell 159, 1665–1680, December 18, 2014 ª2014 Elsevier Inc.
TAD ≠ border-to-border loop
Border-to-border loops  
cannot reproduce Hi-C data

Domain — systems of actively extruded loops
youtube mirnylab
http://mirnylab.mit.edu/projects/emerging-evidence-for-loop-extrusion/
https://www.youtube.com/watch?v=8FW6gOx5lPI

increased LEF
density & processivity
WT LEF depletion weakened barriers
genomic separation s, bp
WT
10
0
10
1
P(s)
10
2
10
5
10
6
10
0
10
1
10
2
10
5
10
6
10
0
10
1
10
2
10
5
10
6
10
0
10
1
10
2
10
5
10
6
0
0
0
300kb
A B C D
Testing loop extrusion
!
Predictions

!
Experiments
control
46Mb
51Mb
46Mb
51Mb
19Mb
23Mb1 Mb 1 Mb 1 Mb
chr5 chr5 chr11
contactfrequency
0.008
0
Elphege Nora
Benoit G. Bruneau
UCSF
Francois Spitz!
Wibke Schwarzer!
!
Article
The Cohesin Release Factor WAPL
Restricts Chromatin Loop Extension
Judith H.I. Haarhuis,1,6 Robin H. van der Weide,2,6 Vincent A. Blomen,3 J. Omar Ya´ n˜ ez-Cuna,2 Mario Amendola,2,7
Marjon S. van Ruiten,1 Peter H.L. Krijger,4 Hans Teunissen,2 Rene´ H. Medema,1 Bas van Steensel,2
Thijn R. Brummelkamp,3,5 Elzo de Wit,2,* and Benjamin D. Rowland1,8,*
1Division of Cell Biology, the Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, the Netherlands
2Division of Gene Regulation, the Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, the Netherlands
3Division of Biochemistry, the Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, the Netherlands
4Hubrecht Institute, Uppsalalaan 8, 3584 CT Utrecht, the Netherlands
5Cancer Genomics Center, Plesmanlaan 121, 1066 CX Amsterdam, the Netherlands
6These authors contributed equally
7Present address: UMR_S951, Genethon, 91000 Evry, France
8Lead Contact
*Correspondence: e.d.wit@nki.nl (E.d.W.), b.rowland@nki.nl (B.D.R.)
http://dx.doi.org/10.1016/j.cell.2017.04.013
SUMMARY
The spatial organization of chromosomes influences
many nuclear processes including gene expression.
The cohesin complex shapes the 3D genome by
looping together CTCF sites along chromosomes.
We show here that chromatin loop size can be
increased and that the duration with which cohesin
embraces DNA determines the degree to which
loops are enlarged. Cohesin’s DNA release factor
the SCC2/SCC4 complex (also known as NIPBL and
respectively), and DNA release is driven by cohesin’s anta
WAPL (Ciosk et al., 2000; Gandhi et al., 2006; Kueng et al.,
The cohesin complex consists of three core subunits,
SMC3, and SCC1 (also known as RAD21 or Mcd1), that to
form a ring-shaped structure that can entrap DNA ins
lumen (Haering et al., 2008). WAPL drives cohesin’s r
from chromatin by opening up a distinct DNA exit gate
interface connecting cohesin’s SMC3 and SCC1 subunits
oue¨ t et al., 2016; Murayama and Uhlmann, 2015). In the ab
ARTICLE doi:10.1038/nature24281
Two independent modes of chromatin
organization revealed by cohesin removal
Wibke Schwarzer1
*, Nezar Abdennur2
*, Anton Goloborodko3
*, Aleksandra Pekowska4
, Geoffrey Fudenberg5
, Yann Loe-Mie6,7
,
Nuno A Fonseca8
, Wolfgang Huber4
, Christian H. Haering9
, Leonid Mirny3,5
& Francois Spitz1,4,6,7
Imaging and chromosome conformation capture studies have revealed several layers of chromosome organization,
Article
Targeted Degradation of CTCF Decouples
Local Insulation of Chromosome Domains
from Genomic Compartmentalization
Elphe` ge P. Nora,1,2,* Anton Goloborodko,3 Anne-Laure Valton,4 Johan H. Gibcus,4 Alec Uebersohn,1,2,7 Nezar Abdennur,3
Job Dekker,4 Leonid A. Mirny,3 and Benoit G. Bruneau1,2,5,6,8,*
1Gladstone Institute of Cardiovascular Disease, San Francisco, CA 94158, USA
2Roddenberry Center for Stem Cell Biology and Medicine at Gladstone, San Francisco, CA 94158, USA
3Institute for Medical Engineering and Science and Department of Physics, Massachusetts Institute of Technology, Cambridge,
MA 02139, USA
4Howard Hughes Medical Institute, Program in Systems Biology, Department of Biochemistry and Molecular Pharmacology, University of
Massachusetts Medical School, Worcester, MA 01605-0103, USA
5
regulator of chromosomal structure. Using the
auxin-inducible degron system in mouse embryonic
stem cells, we show that CTCF is absolutely and
dose-dependently required for looping between
CTCF target sites and insulation of topologically
associating domains (TADs). Restoring CTCF rein-
states proper architecture on altered chromosomes,
indicating a powerful instructive function for CTCF
in chromatin folding. CTCF remains essential for
TAD organization in non-dividing cells. Surprisingly,
active and inactive genome compartments remain
properly segregated upon CTCF depletion, revealing
that compartmentalization of mammalian chromo-
somes emerges independently of proper insulation
of TADs. Furthermore, our data support that CTCF
mediates transcriptional insulator function through
enhancer blocking but not as a direct barrier to het-
erochromatin spreading. Beyond defining the func-
tions of CTCF in chromosome folding, these results
provide new fundamental insights into the rules
governing mammalian genome organization.
INTRODUCTION
Chromosomes meet the dual challenge of packaging DNA into
the nucleus and, at the same time, enabling access to genetic in-
formation. Decades of work on chromosome organization have
tackled the link between chromosome structure and genetic
Potts et al., 20
but here we foc
Mammalian c
Euchromatin c
rich regions (G
is condensed, g
highlights the
ical, biochemic
somes. Chromo
belonging to tw
vealed by high-
(3C), with chro
loci of the same
chromosomes (
on linear genom
types forms a d
with regional c
2013; Bonev an
ment contains
B compartmen
lamina-associat
et al., 2015), wh
At a more loc
megabase segm
relatively insulat
cally associatin
et al., 2012). Th
by the binding
et al., 2012; Phi
zinc-finger nucl
930 Cell 169, 930–944, May 18, 2017 ª 2017 Elsevier Inc.

!
Experiments
control
46Mb
51Mb
46Mb
51Mb
19Mb
23Mb1 Mb 1 Mb 1 Mb
chr5 chr5 chr11
contactfrequency
0.008
0

!
Experiments
control
46Mb
51Mb
46Mb
51Mb
19Mb
23Mb1 Mb 1 Mb 1 Mb
chr5 chr5 chr11
contactfrequency
0.008
0
perturbation
100
10 1
10 2
10 3
10 4
104
105
106
107
100
10 1
10 2
10 3
10 4
104
105
106
107
Control
100
10 1
10 2
10 3
10 4
104
105
106
107
ControlControl
genomic separation s, bpgenomic separation s, bp
contactfrequency,P(s)
46Mb
51Mb
46Mb
51Mb
19Mb
23Mb

Same results from direct targeting of cohesinArticle
Cohesin Loss Eliminates All Loop Domains
Suhas S.P. Rao,1,2,3 Su-Chen Huang,1,2 Brian Glenn St Hilaire,1,2,4 Jesse M. Engreitz,5 Elizabeth M. Perez,5
Kyong-Rim Kieffer-Kwon,6 Adrian L. Sanborn,1,4,7 Sarah E. Johnstone,5,8 Gavin D. Bascom,9 Ivan D. Bochkov,1,2
Xingfan Huang,1,10 Muhammad S. Shamim,1,2,10,11 Jaeweon Shin,1,10 Douglass Turner,1,12 Ziyi Ye,1,10 Arina D. Omer,1,2
James T. Robinson,1,5,12 Tamar Schlick,9,13,14 Bradley E. Bernstein,5,8 Rafael Casellas,6,15 Eric S. Lander,5,16,17
and Erez Lieberman Aiden1,2,4,5,10,18,*
1The Center for Genome Architecture, Baylor College of Medicine, Houston, TX 77030, USA
2Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX 77030, USA
3Department of Structural Biology, Stanford University School of Medicine, Stanford, CA 94305, USA
4Center for Theoretical Biological Physics, Rice University, Houston, TX 77030, USA
5Broad Institute of MIT and Harvard, Cambridge, MA 02139, USA
6Lymphocyte Nuclear Biology, NIAMS, NIH, Bethesda, MD 20892, USA
7Department of Computer Science, Stanford University, Stanford, CA 94305, USA
8Department of Pathology and Center for Cancer Research, Massachusetts General Hospital and Harvard Medical School, Boston,
MA 02114, USA
9Department of Chemistry, New York University, New York, NY 10003, USA
10Departments of Computer Science and Computational and Applied Mathematics, Rice University, Houston, TX 77030, USA
11Medical Scientist Training Program, Baylor College of Medicine, Houston, TX 77030, USA
12Department of Medicine, University of California, San Diego, La Jolla, CA 92037, USA
13Courant Institute of Mathematical Sciences, New York University, New York, NY 10012, USA
14NYU-ECNU Center for Computational Chemistry, NYU Shanghai, Shanghai 200062, China
15Center of Cancer Research, NCI, NIH, Bethesda, MD 20892, USA
16Department of Biology, MIT, Cambridge, MA 02139, USA
17Department of Systems Biology, Harvard Medical School, Boston, MA 02115, USA
18Lead Contact
*Correspondence: erez@erez.com
https://doi.org/10.1016/j.cell.2017.09.026
SUMMARY
The human genome folds to create thousands of
intervals, called ‘‘contact domains,’’ that exhibit
enhanced contact frequency within themselves.
‘‘Loop domains’’ form because of tethering between
two loci—almost always bound by CTCF and cohe-
sin—lying on the same chromosome. ‘‘Compartment
domains’’ form when genomic intervals with similar
histone marks co-segregate. Here, we explore the ef-
fects of degrading cohesin. All loop domains are
eliminated, but neither compartment domains nor
histone marks are affected. Loss of loop domains
(Wendt et al., 2008) and lie at the anchors of loops (Rao et al.,
2014; Splinter et al., 2006) and the boundaries of contact do-
mains (also called ‘‘topologically constrained domains,’’ ‘‘topo-
logically associated domains,’’ or ‘‘physical domains’’) (Dixon
et al., 2012; Lieberman-Aiden et al., 2009; Nora et al., 2012;
Rao et al., 2014). This suggests that these proteins help regulate
genome folding (Merkenschlager and Nora, 2016). Consistent
with this, deletion of CTCF sites interferes with loop and contact
domain formation (Guo et al., 2015; Sanborn et al., 2015; de Wit
et al., 2015). However, initial, low-resolution experiments exam-
ining genome-wide depletion of CTCF and cohesin observed
only limited effects, reporting that compartments and contact
domains still appear to be present (Seitan et al., 2013; Sofueva
et al., 2013; Zuin et al., 2014). These results have made it difﬁcult
A
= 42
= 114
= 111
134.6133.842.140.8
42.140.8
91.995.8
91.9 95.8 Mb
134.6133.8
HCT116-RAD21-mAC
- auxin
HCT116-RAD21-mAC
+ auxin, 6hr
Chr4
Chr 4
Chr 1
Chr1Chr8
Chr 8
134.6 Mb133.8
Chr 8
= 38
= 111
42.1 Mb40.8
Chr 4
= 107
91.9 95.8 Mb
Chr 1
E
H3K27ac
H3K4me3
H3K4me1
NIPBL
Fast
+ auxin, 6hr withdraw, 20
Withdraw
68.467.6Mb68.2
= 107 = 34= 116
Chr18Chr14
0
7
0
25
0
100
0
50
Chr18
Rao et al., 2017, Cell 171, 305–320
October 5, 2017 ª 2017 Elsevier Inc.
https://doi.org/10.1016/j.cell.2017.09.026
Article
Topologically associating domains and chromatin
loops depend on cohesin and are regulated by
CTCF, WAPL, and PDS5 proteins
Gordana Wutz1,†
, Csilla Várnai2,†
, Kota Nagasaka1,†
, David A Cisneros1,†,‡
, Roman R Stocsits1
,
Wen Tang1
, Stefan Schoenfelder2
, Gregor Jessberger1
, Matthias Muhar1
, M Julius Hossain3
,
Nike Walther3
, Birgit Koch3
, Moritz Kueblbeck3
, Jan Ellenberg3
, Johannes Zuber1
,
Peter Fraser2,4
& Jan-Michael Peters1,*
Abstract
Mammalian genomes are spatially organized into compartments,
topologically associating domains (TADs), and loops to facilitate
gene regulation and other chromosomal functions. How compart-
ments, TADs, and loops are generated is unknown. It has been
proposed that cohesin forms TADs and loops by extruding chro-
matin loops until it encounters CTCF, but direct evidence for this
hypothesis is missing. Here, we show that cohesin suppresses
compartments but is required for TADs and loops, that CTCF
defines their boundaries, and that the cohesin unloading factor
WAPL and its PDS5 binding partners control the length of loops. In
the absence of WAPL and PDS5 proteins, cohesin forms extended
loops, presumably by passing CTCF sites, accumulates in axial
chromosomal positions (vermicelli), and condenses chromosomes.
Unexpectedly, PDS5 proteins are also required for boundary func-
tion. These results show that cohesin has an essential genome-
wide function in mediating long-range chromatin interactions and
support the hypothesis that cohesin creates these by loop extru-
sion, until it is delayed by CTCF in a manner dependent on PDS5
proteins, or until it is released from DNA by WAPL.
Introduction
Duplicated DNA molecules become physically connected with each
other during DNA replication. This sister chromatid cohesion is
essential for bi-orientation of chromosomes on the mitotic or
meiotic spindle and thus enables their symmetrical segregation
during cell division (Dewar et al, 2004). Cohesion is mediated by
cohesin complexes (Guacci et al, 1997; Michaelis et al, 1997;
Losada et al, 1998) which are thought to perform this function by
entrapping both sister DNA molecules inside a ring structure that is
formed by the cohesin subunits SMC1, SMC3, and SCC1 (also
known as RAD21 and Mcd1) (Haering et al, 2008).
Cohesin is present at centromeres and on chromosome arms (re-
viewed in Peters et al, 2008). At centromeres, cohesin resists the
pulling force of spindle microtubules, a function that is required
both for stabilization of microtubule–kinetochore attachments and
for chromosome bi-orientation. On chromosome arms, however, the
precise location of cohesin would not be expected to matter if cohe-
sin’s only function was to mediate cohesion. But contrary to this
expectation, cohesin is enriched at thousands of well-defined loci on
chromosome arms. In mammalian genomes, ~90% of these are
Published online: December 7, 2017
Article
A mechanism of cohesin-dependent loop extrusion
organizes zygotic genome architecture
Johanna Gassler1,†
, Hugo B Brandão2,†
, Maxim Imakaev3,4
, Ilya M Flyamer5
, Sabrina Ladstätter1
,
Wendy A Bickmore5
, Jan-Michael Peters6
, Leonid A Mirny2,3,*
& Kikuë Tachibana1,**
Abstract Introduction
Zp3-Cre
∆/∆ ∆/∆fl/flfl/fl
B
I
R
T
H
+
B
A
1.43
Scc1fl
combined
1.14
A
B
1.03
Scc1∆
combined
1.03
Average loop
-100kb
0 kb
+90kb
activeinactive
TAD
Average TAD Com
A
B
-100kb
0 kb
+90kb
TAD
activeinactivee
The EMBO Journal
G
190 Mb0 Mb
Chromosome 4
0 min auxin
190
15 min auxin 180 min auxin
J
0
10 kb 100 kb 1 Mb 10 Mb100 Mb
c
genomic distance
88 Mb 94.5 Mbchromosome 12
8
0 min auxin 15 min auxin 180 min auxin
a-tubulin
H3
0 20 40
0.0
0.2
time (min)
nor
10 30
H I
log2(observed/expected)
1.6
0.8
0
-0.8
-1.6
1 50
1
50
-450 kb 0 450 kb
chromosome 4
471 377
5 6
2
0
-2
log2enrichment
Figure 1.

SMC is a motor!
Cite as: M. Ganji et al., Science
10.1126/science.aar7831 (2018).
REPORTS
The spatial organization of chromosomes is of paramount
importance to cell biology. Members of the SMC family of
protein complexes, including condensin, cohesin, and the
Smc5/6 complex, play vital roles in restructuring genomes
during the cellular life cycle (1–3). The principles by which
SMC complexes achieve these fundamental tasks are still
incompletely understood. Models based on random cross-
linking of DNA by pairwise interactions or conformational
changes in the DNA superhelicity have been proposed (4, 5).
An alternative hypothesis suggested that SMC protein com-
plexes bind to small loops in the genome to then processive-
ly enlarge them (6). More recently, the idea emerged that
condensin can start and subsequently extrude DNA loops,
which would elegantly explain how condensin mediates the
formation of mitotic chromosomes structures observed in
electron micrographs and deduced from Hi-C experiments
(7, 8). Indeed, polymer simulations showed that loop extru-
sion can, in principle, result in the efficient disentanglement
and compaction of chromatin fibers (9–11). The recent dis-
covery that condensin exhibits DNA translocase activity (12)
was consistent with, but did not provide conclusive evidence
for (13), DNA loop extrusion.
In this Report, we visualize the formation of DNA loops
by the Saccharomyces cerevisiae condensin complex in real
(ATP), we observed the accumulation of fluorescence densi-
ty at one spot along the length of the DNA (Fig. 1, D and E,
fig. S1, and movie S2). This finding shows that condensin
induces local compaction of DNA.
To visualize the compacted DNA structures in the imag-
ing plane of the microscope, we applied flow at a large angle
with respect to the double-tethered DNA. This revealed that
the bright spots were made up of extended pieces of DNA,
consistent with single large DNA loops (Fig. 1, F and G, fig.
S2, and movie S3). Importantly, we observed no DNA loop
formation by wild-type condensin in the absence of either
ATP or Mg2+
, when we replaced ATP by the non-
hydrolyzable analogs ATP S or AMPPNP, or when we used a
mutant condensin that is unable to bind ATP. Condensin
hence creates DNA loops in a strictly ATP-hydrolysis-
dependent manner, either by gradually extruding DNA or by
randomly grabbing and linking two DNA loci.
To distinguish between these two possibilities, we moni-
tored the looping process by real-time imaging of the DNA
while applying constant flow. This revealed the gradual ap-
pearance of an initially weak increase in fluorescence inten-
sity at a local spot that grew into an extended loop over time
(Fig. 2A, fig. S3, and movies S4 and S5), providing direct
visual evidence of loop extrusion and ruling out the random
Real-time imaging of DNA loop extrusion by condensin
Mahipal Ganji,1
Indra A. Shaltiel,2
* Shveta Bisht,2
* Eugene Kim,1
Ana Kalichava,1
Christian H. Haering,2
† Cees Dekker1
†
1
Department of Bionanoscience, Kavli Institute of Nanoscience Delft, Delft University of Technology, Delft, Netherlands. 2
Cell Biology and Biophysics Unit,
Structural and Computational Biology Unit, European Molecular Biology Laboratory, Heidelberg, Germany.
*These authors contributed equally to this work.
†Corresponding author. Email: christian.haering@embl.de (C.H.H.); c.dekker@tudelft.nl (C.D.)
It has been hypothesized that Structural Maintenance of Chromosomes (SMC) protein complexes such as
condensin and cohesin spatially organize chromosomes by extruding DNA into large loops. Here, we
provide unambiguous evidence for loop extrusion by directly visualizing the formation and processive
extension of DNA loops by yeast condensin in real-time. We find that a single condensin complex is able to
extrude tens of kilobase pairs of DNA at a force-dependent speed of up to 1,500 base pairs per second,
using the energy of ATP hydrolysis. Condensin-induced loop extrusion is strictly asymmetric, which
demonstrates that condensin anchors onto DNA and reels it in from only one side. Active DNA loop
extrusion by SMC complexes may provide the universal unifying principle for genome organization.
Cite as: M. Ganji et al., Science
10.1126/science.aar7831 (2018).
REPORTS
First release: 22 February 2018 www.sciencemag.org (Page numbers not final at time of first release) 1
The spatial organization of chromosomes is of paramount
importance to cell biology. Members of the SMC family of
protein complexes, including condensin, cohesin, and the
Smc5/6 complex, play vital roles in restructuring genomes
during the cellular life cycle (1–3). The principles by which
SMC complexes achieve these fundamental tasks are still
incompletely understood. Models based on random cross-
linking of DNA by pairwise interactions or conformational
changes in the DNA superhelicity have been proposed (4, 5).
An alternative hypothesis suggested that SMC protein com-
plexes bind to small loops in the genome to then processive-
ly enlarge them (6). More recently, the idea emerged that
condensin can start and subsequently extrude DNA loops,
which would elegantly explain how condensin mediates the
formation of mitotic chromosomes structures observed in
electron micrographs and deduced from Hi-C experiments
(7, 8). Indeed, polymer simulations showed that loop extru-
sion can, in principle, result in the efficient disentanglement
and compaction of chromatin fibers (9–11). The recent dis-
covery that condensin exhibits DNA translocase activity (12)
was consistent with, but did not provide conclusive evidence
for (13), DNA loop extrusion.
In this Report, we visualize the formation of DNA loops
by the Saccharomyces cerevisiae condensin complex in real
time (Fig. 1A). We tethered both ends of a double-stranded
48.5-kilobase pair (kbp) -DNA molecule to a passivated
surface (14, 15), using flow to adjust the DNA end-to-end
length to a distance much shorter than its contour length
(Fig. 1B). We then imaged DNA after staining with Sytox
Orange (SxO; Fig. 1C and movie S1). Upon flushing in 1 nM
of condensin (12) and 5 mM of adenosine triphosphate
(ATP), we observed the accumulation of fluorescence densi-
ty at one spot along the length of the DNA (Fig. 1, D and E,
fig. S1, and movie S2). This finding shows that condensin
induces local compaction of DNA.
To visualize the compacted DNA structures in the imag-
ing plane of the microscope, we applied flow at a large angle
with respect to the double-tethered DNA. This revealed that
the bright spots were made up of extended pieces of DNA,
consistent with single large DNA loops (Fig. 1, F and G, fig.
S2, and movie S3). Importantly, we observed no DNA loop
formation by wild-type condensin in the absence of either
ATP or Mg2+
, when we replaced ATP by the non-
hydrolyzable analogs ATP S or AMPPNP, or when we used a
mutant condensin that is unable to bind ATP. Condensin
hence creates DNA loops in a strictly ATP-hydrolysis-
dependent manner, either by gradually extruding DNA or by
randomly grabbing and linking two DNA loci.
To distinguish between these two possibilities, we moni-
tored the looping process by real-time imaging of the DNA
while applying constant flow. This revealed the gradual ap-
pearance of an initially weak increase in fluorescence inten-
sity at a local spot that grew into an extended loop over time
(Fig. 2A, fig. S3, and movies S4 and S5), providing direct
visual evidence of loop extrusion and ruling out the random
cross-linking model. The extruded loops were in general
stable (fig. S4), but occasionally disrupted spontaneously in
a single step (Fig. 2A and movie S6). Such a single-step dis-
ruption suggests that the DNA loop had been extruded by a
single condensin unit that spontaneously let go of the loop,
instead of a multi-step relaxation of the loop due to multiple
units.
Real-time imaging of DNA loop extrusion by condensin
Mahipal Ganji,1
Indra A. Shaltiel,2
* Shveta Bisht,2
* Eugene Kim,1
Ana Kalichava,1
Christian H. Haering,2
† Cees Dekker1
†
1
Department of Bionanoscience, Kavli Institute of Nanoscience Delft, Delft University of Technology, Delft, Netherlands. 2
Cell Biology and Biophysics Unit,
Structural and Computational Biology Unit, European Molecular Biology Laboratory, Heidelberg, Germany.
†Corresponding author. Email: christian.haering@embl.de (C.H.H.); c.dekker@tudelft.nl (C.D.)
It has been hypothesized that Structural Maintenance of Chromosomes (SMC) protein complexes such as
condensin and cohesin spatially organize chromosomes by extruding DNA into large loops. Here, we
provide unambiguous evidence for loop extrusion by directly visualizing the formation and processive
extension of DNA loops by yeast condensin in real-time. We find that a single condensin complex is able to
extrude tens of kilobase pairs of DNA at a force-dependent speed of up to 1,500 base pairs per second,
using the energy of ATP hydrolysis. Condensin-induced loop extrusion is strictly asymmetric, which
demonstrates that condensin anchors onto DNA and reels it in from only one side. Active DNA loop
extrusion by SMC complexes may provide the universal unifying principle for genome organization.
onFebruary23,2018http://science.sciencemag.org/Downloadedfrom
~8 Kb/min
for two motors 
~40Kb/min on nucleosomal fiber 
g. 2.
peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint (which was not.http://dx.doi.org/10.1101/137711doi:preprint first posted online May. 13, 2017;
Speed =4 Kb/min
Cite as: T. Terakawa et al., Science
10.1126/science.aan6516 (2017).
REPORTS
First release: 7 September 2017 www.sciencemag.org (Page numbers not final at time of first release) 1
Structural maintenance of chromosomes (SMC) complexes
are the major organizers of chromosomes in all living organ-
isms (1, 2). These protein complexes play essential roles in
sister chromatid cohesion, chromosome condensation and
segregation, DNA replication, DNA damage repair, and gene
expression. A distinguishing feature of SMC complexes is
their large ring-like architecture, the circumference of which
is made up of two SMC protein coiled-coil proteins and a sin-
gle kleisin subunit (Fig. 1A) (1–4). The ~50-nm long antipar-
allel coiled-coils are connected at one end by a stable
dimerization interface, referred to as the hinge domain, and
at the other end by globular ATP-binding cassette (ABC) fam-
ily ATPase domains (5). The ATPase domains are bound by a
protein of the kleisin family, along with additional accessory
subunits, which vary for different types of SMC complexes
(Fig. 1A). The relationship between SMC structures and their
functions in chromosome organization is not completely un-
derstood (6), but many models envision that the coiled-coil
domains allow the complexes to topologically embrace DNA
(1–4). Given the general resemblance to myosin and kinesin,
some early models postulated that SMC proteins might be
mechanochemical motors (7–10).
SMC complexes are thought to regulate genome architec-
ture by physically linking distal chromosomal loci, but how
these bridging interactions might be formed remains un-
known (1, 2, 11). An early model suggested that many three-
dimensional (3D) features of eukaryotic chromosomes might
be explained by DNA loop extrusion (Fig. 1B) (12, 13), and re-
cent polymer dynamics simulations have shown that loop ex-
trusion can recapitulate the formation of topologically
associating domains (TADs), chromatin compaction, and sis-
ter chromatid segregation (14–18). This loop extrusion model
assumes a central role for SMC complexes in actively creating
the DNA loops (11, 12). Similarly, it has been proposed that
prokaryotic SMC proteins may structure bacterial chromo-
somes through an active loop extrusion mechanism (19–21).
Yet, the loop extrusion model remains hypothetical, in large
part because the motor activity that is necessary for driving
loop extrusion could not be identified (11). Indeed, the ab-
sence of an identifiable motor activity in SMC complexes in-
stead has lent support to alternative models in which DNA
loops are not actively extruded, but instead are captured and
stabilized by stochastic pairwise SMC binding interactions to
bridge distal loci (22).
To help distinguish between possible mechanisms of SMC
protein-mediated chromosomal organization, we examined
the DNA-binding properties of condensin (23). We overex-
pressed the five subunits of the condensin complex in bud-
ding yeast and purified the complex to homogeneity (Fig. 1C
and fig. S1). Electron microscopy images confirmed that the
complexes were monodisperse (Fig. 1D). As previously de-
scribed for electron micrographs of immunopurified Xenopus
laevis or human condensin (24), we observed electron density
that presumably corresponds to the two HEAT-repeat subu-
The condensin complex is a mechanochemical motor that
translocates along DNA
Tsuyoshi Terakawa,1
* Shveta Bisht,2
* Jorine M. Eeftens,3
* Cees Dekker,3
† Christian H. Haering,2
† Eric C.
Greene1
†
1
Department of Biochemistry and Molecular Biophysics, Columbia University, New York, NY, USA. 2
Cell Biology and Biophysics Unit, Structural and Computational Unit,
European Molecular Biology Laboratory (EMBL), Heidelberg, Germany. 3
Department of Bionanoscience, Kavli Institute of Nanoscience Delft, Delft University of Technology,
Delft, Netherlands.
†Corresponding author. Email: c.dekker@tudelft.nl (C.D.); christian.haering@embl.de (C.H.H.); ecg2108@cumc.columbia.edu (E.C.G.)
Condensin plays crucial roles in chromosome organization and compaction, but the mechanistic basis for
its functions remains obscure. Here, we use single-molecule imaging to demonstrate that Saccharomyces
cerevisiae condensin is a molecular motor capable of ATP hydrolysis-dependent translocation along
double-stranded DNA. Condensin’s translocation activity is rapid and highly processive, with individual
complexes traveling an average distance of 10 kilobases at a velocity of ~60 base pairs per second. Our
results suggest that condensin may take steps comparable in length to its ~50-nanometer coiled-coil
subunits, suggestive of a translocation mechanism that is distinct from any reported DNA motor protein.
The finding that condensin is a mechanochemical motor has important implications for understanding the
mechanisms of chromosome organization and condensation.
onNovember1,2017http://science.sciencemag.org/Downloadedfrom
Speed =36 Kb/min
Theory: speed ~100-200Kb per cohesin residence time (5-20min) 
=> 10-40Kb/min
One-sided!?

…but one-sided cannot compact chromosomes!
?
S.cerevisiae
mammals

Conclusions 
1. Strong experimental support of the loop
extrusion by cohesin, 
hindered by CTCF 
!

Outline
Compartments and TADs
1. Emerging evidence of loop extrusion
2. Compartmentalization by phase separation!
3. Phase separation vs extrusion
Chromatin Organization by an Interplay of Loop Extrusion and Compartmental
Segregation 
Heterochromatin drives organization of conventional and inverted nuclei 

Kikue 
Tachibana-Konwalski 
Ilya M. Flyamer !
Johanna Gassler!
IBMA, Vienna!
!
!
!
!
!
 
Maxim Imakaev 
MIT
Hugo Brandao
Harvard Biophysics
Single-nucleus Hi-C
Flyamer, Gassler, Imakaev et al. Nature 2017!
a
TADs compartments
yesyes
yes no
Compartments and TADs are formed  
by separate mechanisms
b
active inactive
0.0 0.5
Log enrichment
-80kbinactive
0.5
Log enrichment
0.32 0.88
Effective
contact probability
-80kb loop +70kb TAD
0.0 0.8
Log enrichment
Compartmentalization Average loop Average TAD
active

TADs (but not compartments) are cohesin-dependent

Compartments and TADs are formed  
by separate mechanisms
Compartments become stronger and fragmented 
in the absence of loop extrusion
A
B
1.5
1.5
35.0 47.5 60.035.0 47.5 60.0
35.0
47.5
60.0
1.5
1.5
TAM (control)

Fine compartments match epigenetic state better than
wild-type compartments
C
D E
Eigenvector
1.5 -
-1.5 _
-0-
H3K4me3
2 -
0 _
H3K4me1
2 -
0 _
H3K36me3
2 -
0 _
H3K27me3
2 -
0 _
25.0 30.0 35.0
10 Mb
H3K27ac
2 -
0 _
Activity
chr15
20.0
30.0
40.0
20.0
30.0
40.0
20.0 30.0 40.0 55.0 70.0 72.0 88.0 103.5
B
C
TAM
chr17
1.5 1.5

Conclusions 
1. Strong experimental support of the loop extrusion by
cohesin, hindered by CTCF 
2. TAD and compartment formation — separate mechanisms
3. Innate compartments (associated with histone marks)
are partially suppressed by loop extrusion
A
1.5
1.5
35.0 47.5 60.035.0 47.5 60.0
35.0
47.5
60.0
1.5
1.5
TAM

Published online 04 August 2014 Nucleic Acids Research, 2014, Vol. 42, No. 15 9553–9561
doi: 10.1093/nar/gku698
Modeling epigenome folding: formation and dynamics
of topologically associated chromatin domains
Daniel Jost1
, Pascal Carrivain2
, Giacomo Cavalli2,*
and C´edric Vaillant1,*
1
Laboratoire de Physique, Ecole Normale Sup´erieure de Lyon, CNRS UMR 5672, Lyon 69007, France and 2
Institute
of Human Genetics, CNRS UPR 1142, Montpellier 34000, France
Received April 1, 2014; Revised July 02, 2014; Accepted July 19, 2014
ABSTRACT
Genomes of eukaryotes are partitioned into domains
of functionally distinct chromatin states. These do-
mains are stably inherited across many cell gener-
ations and can be remodeled in response to devel-
opmental and external cues, hence contributing to
the robustness and plasticity of expression patterns
and cell phenotypes. Remarkably, recent studies in-
dicate that these 1D epigenomic domains tend to fold
into 3D topologically associated domains forming
specialized nuclear chromatin compartments. How-
ever, the general mechanisms behind such compart-
mentalization including the contribution of epige-
netic regulation remain unclear. Here, we address
the question of the coupling between chromatin fold-
ing and epigenome. Using polymer physics, we ana-
lyze the properties of a block copolymer model that
accounts for local epigenomic information. Consid-
ering copolymers build from the epigenomic land-
scape of Drosophila, we observe a very good agree-
ment with the folding patterns observed in chro-
mosome conformation capture experiments. More-
over, this model provides a physical basis for the
existence of multistability in epigenome folding at
sub-chromosomal scale. We show how experiments
are fully consistent with multistable conformations
where topologically associated domains of the same
epigenomic state interact dynamically with each
other. Our approach provides a general framework
to improve our understanding of chromatin folding
during cell cycle and differentiation and its relation
to epigenetics.
INTRODUCTION
Gene expression is regulated by many sets of proteins that
associate with the genome in a cell-type and condition-
specific manner at specific regulatory elements including
proximal promoters, enhancers and repressors. The packag-
ing of eukaryotic DNA into chromatin contributes to this
regulation via the modulation of the accessibility and speci-
ficity of regulators to their nucleic sites. Locally, the chro-
matin state is characterized by various features like the nu-
cleosome positioning, the covalent modifications of DNA
and histones tails and the insertion of histone variants. This
pattern of chromatin states along the genome, the so-called
‘epigenome’, is itself regulated by the combined action of
different specialized chromatin regulators like chromatin re-
modelers, modifying enzymes and histone chaperones.
The general picture that emerges from the genome-wide
high-resolution profiling of structural and functional chro-
matin marks obtained in various organisms and cell types
(1–4), is that eukaryotic genomes are linearly organized into
distinct epigenomic domains. These domains extend over
few kilobases up to few megabases, are characterized by
a specific type of chromatin and are isolated from their
neighborhood by boundary elements such as insulators.
Euchromatin, less condensed, early replicating and con-
taining most active genes, is generally distinguished from
heterochromatin, typically highly condensed, late replicat-
ing and inhibitory to transcriptional machinery. In many
higher eukaryotes, from plants to mammals, statistical anal-
yses of hundreds of chromatin marks have identified only
a small number of main chromatin types (1,3,5,6), typi-
cally four or five, covering the well-known constitutive HP1-
like heterochromatin or the facultative (developmentally
regulated) Polycomb-like heterochromatin but also a less-
characterized ultra-repressive heterochromatin enriched in
genes that are expressed in very few tissues, the so-called
void or black chromatin (1,7).
Interestingly, within epigenomic domains, regulatory se-
quences such as enhancers may be located far from the tar-
get genes and multiple elements that are distributed over
large regions may collaborate or compete for the regula-
tion of individual genes or gene clusters. This implies the
existence of long-range mechanisms where regulatory ele-
ments could act over large genomic distances up to hun-
dreds of kilobases or more. A possible mechanism regu-
lating such long-range effects is the linear spreading of a
*To whom correspondence should be addressed. Tel: +33 4 72 72 86 34; Fax: +33 4 72 72 89 50; Email: cedric.vaillant@ens-lyon.fr
Correspondence may also be addressed to Giacomo Cavalli. Tel: +33 4 34 35 99 70; Fax: +33 4 34 35 99 01; Email: giacomo.cavalli@igh.cnrs.fr
C⃝ The Author(s) 2014. Published by Oxford University Press on behalf of Nucleic Acids Research.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which
permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.
Mechanism of compartmentalization
Johannes Nübler
Many problems associated with it are not solved yet, and the potential of its
practical use is far from being exhausted.
Recall that a block copolymer is a chain consisting of consecutively joined
blocks, each of which constitutes a long homopolymer chain. For example, the
chemical structure of a two-block copolymer is A--...--A--B--...--B. The
number of blocks in the molecule, just as the number of links in the block, can
be arbitrary.
What happens when a sufficiently concentrated solution or melt is made from
the chains of a block copolymer? From previous subsections, we know that in a
typical case, the chains of poly-A and poly-B (or in our case, the blocks) are
incompatible. A phase separation in such a system, .however, is impossible
because of the covalent bonding of the blocks into common chains.
As a result, the phase separation, which is impossible on the some of the whole
system, occurs on a certain limited length scale defined by the size of the blocks.
The arising microdomain structure is schematically shown in Figure 5.1.
If the total amount of one of the components (e.g., A) is relatively small, then
the corresponding phase enriched with A component (the A phase) occupies a
small fraction of the total volume, and it constitutes a system of spherically
shaped micelles scattered like "islets>, in a °’sea" of the phase enriched with B
component (the B phase). On increasing the fraction of A links, the spherically
shaped micelles become cylindric ones piercing the B phase like reinforcing
wires. On further increasing the A fraction, a lamellar (or layer), structure
appears, with A and B phases laid out in alternating planar layers. Finally, on
still further increasing the fraction of A links, the so-called in~zerse phases
emerge: first the cylindric phase (B phase cylinders piercing the A phase), then
the spheric one ( B "’raisins" in the A "’pudding").
To conclude, it should be noted that the microdomain (or micellar) structures
are typical not only for block copolymers but also for systems consisting of the
so-called diphilic molecules. One of their blocks has a low molecular weight, but
because of its thermodynamic properties, it cannot mix withthe other block.
Examples include phospholipid molecules consisting of a hydrophilic "head"
and a polymeric (usually not very long) "tail." The dissolution of such mole-
h ~
FIGUItE 5.1. Microdomain structure in a melt of block copolymers. (a), Spheric A phase
micelles in massive B phase. (b), Cylindric micelles. (c), Alternating planar lamellae.
cu
(a
2
.o
(
o
t
t
"
Statistical Physics of Macromolecules 1991
(micro)phase separation
extrusion segregation
counteracts
native compartmentalization
contact frequency
low
high
no loop
extrusion
(nA+nB)
-1 (nA-nB)
(nA+nB)
counts,a.u.
A
B
Chromatin Organization by an Interplay of Loop Extrusion and Compartmental
Segregation 

…..add loop extrusion on top
loop
extrusion
compartmental
segregation
counteracts
no loop
extrusion
-1 1(nA-nB)
(nA+nB)
-1 1(nA-nB)
(nA+nB)
counts,a.u.
0.0
0.18
A
B
Eattr(kBT)
5 MB equiv.5 MB equiv.
50 MB equiv.
simul

…..add loop extrusion on top
5 MB equiv.
5 MB equiv.
5 MB equiv.
loop
extrusion
compartmental
segregation
counteracts
no loop
extrusion
-1 1(nA-nB)
(nA+nB)
-1 1(nA-nB)
(nA+nB)
counts,a.u.
0.0
0.18
A
B
Eattr(kBT)
50 MB equiv.
simul
compaction
1
ofile
rr.
1
50 MB equiv.
modelB
loop
extrusion
compartmental
segregation
counteracts
simulationC
example snapshot A/B number difference
-1 1(nA-nB)
(nA+nB)
counts,a.u.u.
0.0
A
B

Active mixing by loop extrusion suppresses small
compartments compartments
Nuebler et al.,
bioRxiv (2017)
Hi-C
Fig. 2
92Mb 97Mb72Mb 122Mbchr6
comp.profile
autocorr.
WTNipbl
experiments (Schwarzer 2017)

Active mixing by loop extrusion suppresses small
compartments compartments
Nuebler et al.,
bioRxiv (2017)
Hi-C
Fig. 2
small compartments
are “erased” by  
loop extrusion
92Mb 97Mb72Mb 122Mbchr6
comp.profile
autocorr.
WTNipbl
experiments (Schwarzer 2017)

Attractions: 
A-A
B-B (direct or mediated, e.g. HP1)
B-Lamina
!
- Can one disentangle these contributions?
- Which ones are more important for compartmentalization?
Irina 
Solovei,  
LMU
Yana
Fedorova,
Plovidv U
Hard to disentangle in conventional nuclei
Heterochromatin drives organization of conventional and inverted nuclei 

igures and Figure Cap ons:
inverted conventionalconventionalinverted
Irina 
Solovei,  
LMU

Modeling of compartmentalization
Comp.Stength = Sumi [XXi/toti]
Martin Falk
MIT
phase separation
in a block copolymer

CC<BC<BB<AB<AA
BC<CC<BB<AB<AA
AA<AB<BB<BC<CC
….720….

CC<BC<BB<AB<AA
BC<CC<BB<AB<AA
AA<AB<BB<BC<CC
….720….
Correct order: AA<AB<BB<BC<CC; AA≈0!
BB is the only free parameter

BB is the only free parameter
Figure 3. Polymer model reproduces micro
Hi-C features
a, Our approach is to: define a mechanistic mo
interactions; simulate an ensemble of configurati
dynamics; and compare these configurations to Hi-
BB = 0.5-0.6 kT

AA<AB<BB<BC<CC; AA≈0!
!
!
BB = 0.5-0.6 kT!
Lamina-B = 0.2 kT

Attractions: 
A-A
B-B
B-Lamina
<— for compartmentalization
<— for positioning in the nucleus

Attractions: 
A-A
B-B
B-Lamina
<— for compartmentalization
<— for positioning in the nucleus

Figures and Figure Cap ons:

inverted is the default state of the nucleus!!

Summary
Active loop extrusion!
by cohesin!
!
!
!
!
!
1. Loop extrusion can be  
universal mechanism!
• fold domains
• compacts chromosomes 
2. Compartments are formed by  
heterochromatin interactions,  
and positioned in space  
interactions with lamina
Active loop extrusion!
by cohesin!
!
!
!
!
blocked by CTCF!
1. Fundenberg J, Abnennur N, et al bioRxiv 2018
2. Gibcus, Samejima, Goloborodko A et al Science 2018
3. Falk M, et al., bioRxiv 2018
4. Nueber J et al., bioRxiv 2018
http://HiGlass.io
II

Emerging Evidence of Chromosome Folding by Loop Extrusion 
HiGlass: Web-based Visual Exploration and Analysis of Genome
Interaction Maps 
Chromatin Organization by an Interplay of Loop Extrusion and
Compartmental Segregation 
Heterochromatin drives organization of conventional and inverted
nuclei 
bioRxiv

Maxim
Imakaev 
MIT
Nezar
Abdennur  
MIT Comp/Sys
Biology
NSF, NIH: Center of Structure  
and Physics of the Genome
Hugo
Brandao
Harvard
Biophysics
Ed
Banigan
MIT
Aafke  
Van den Berg
MIT
Carolyn  
Lu 
MIT senior
Martin  
Falk
MIT
Job Dekker
UMass Medical
!
!
John Marko
Northwestern U.
Francois Spitz
Institut Pasteur
Elphege Nora
Benoit G. Bruneau
UCSF
Kick 
Tachibana- 
Konwalski
IBMA, Vienna
Bill Earnshaw
U of Edinburgh
Irina 
Solovei,  
LMU
Geoff
Fudenberg 
UCSF
Anton
Goloborodko  
MIT Physics
Johannes
Nübler
MIT

2-8 molecules (or motors) of cohesin per Mb
each consume ATP at a rate of 2 per sec per motor
< 1e5 ATP/sec!
!
Fibroblast ATP production: 1e9 ATP/sec,
!
!
hence the fraction consumed by cohesins < 0.0001  
(very modest: ~1% of the NIH budget in US GDP).

How can SMCs extrude loops?
2. From translocation to extrusion

How can SMCs extrude loops?
2. From translocation to extrusion
Figure 4.
a. Walking as a possible mechanism of SMC translocation, with SMC arms in yellow and orang
kleisin in blue, creating a shackled walker.
possible implementation of 
two-sided

Mechanism of loop extrusion
long-range interactions are formed by 1D process
number of sites L ) 1: At time t = 0, M motile element
pairs are dispersed randomly, each pair initially occupying
adjacent sites of this lattice. The DNA-binding motile
elements (referred to below as ‘motors’) then move
along the DNA with rates independent of position; steps
that move a motor away from its partner (‘forward’ steps
that extrude a DNA loop) occur at a rate r+ and steps that
move a motor back toward its partner (‘reverse steps’ that
retract the loop) occur at a rate rÀ (Figure 1). We suppose
the motion to be directed by energy gained from ATP
hydrolysis, with r+ > rÀ (when r+ ¼ rÀ there is 1D diffu-
sion of each motor; when rÀ > r+, the motors are driven
together which is not of interest here). The motor pairs are
assumed to have left/right symmetry, i.e. the left and right
motors move with the same rates.
increment Át to the event is distributed over the
0 Át 1 exponentially, with probability distr
PðÁtÞ ¼ ReÀRt
: The actual realization of Át is
from this continuous distribution. Which of the K
tions actually occurs is determined from their prob
distribution pi ¼ ri=R: This second, discrete, distrib
be used to select which of the K candidates actually
Once Át and i are determined, the state of the sy
changed, and time is increased to t+Át: The algor
then repeated to propagate the system forward from
to event, for as many transition steps as one r
(or for as long a total time as is required). The r
a series of transition events, distributed in time acc
to the rates that define the model. There is no tim
cretization; events can occur separated by arbitraril
Figure 1. Schematic drawing of machine positions on the lattice as time progresses; lattice model equivalent is sketched below each pan
dumbbell shapes (and arrows in the lattice sketch) depict enzymes and green lines show DNA. Panel (a) depicts the starting point and the pro
of infinitely processive machines, while Panel (b) shows machines with lower processivity (disassociation rate is still relatively small,
Panel (c) depicts a single step, with ATP binding, hydrolysis and release associated with extrusion of a small amount of DNA.
Nucleic Acids Research, 2
Self-organization of domain structures by
DNA-loop-extruding enzymes
Elnaz Alipour1,
* and John F. Marko2,
*
1
Center for Cell Analysis and Modeling, University of Connecticut Health Sciences Center, Farmington,
CT 06030 and 2
Departments of Physics and Astronomy and Molecular Biosciences, Northwestern University,
Evanston, IL 60208, USA
Received June 1, 2012; Revised August 17, 2012; Accepted September 13, 2012
ABSTRACT
The long chromosomal DNAs of cells are organized
into loop domains much larger in size than individual
DNA-binding enzymes, presenting the question of
how formation of such structures is controlled. We
present a model for generation of defined chromo-
somal loops, based on molecular machines consist-
ing of two coupled and oppositely directed motile
elements which extrude loops from the double helix
along which they translocate, while excluding one
another sterically. If these machines do not dissoci-
ate from DNA (infinite processivity), a disordered,
exponential steady-state distribution of small loops
is obtained. However, if dissociation and rebinding
of the machines occurs at a finite rate (finite
processivity), the steady state qualitatively changes
to a highly ordered ‘stacked’ configuration with sup-
pressed fluctuations, organizing a single large,
stable loop domain anchored by several machines.
The size of the resulting domain can be simply
regulated by boundary elements, which halt the
progress of the extrusion machines. Possible real-
izations of these types of molecular machines
are discussed, with a major focus on structural main-
tenance of chromosome complexes and also with
discussion of type I restriction enzymes. This mech-
anism could explain the geometrically uniform
folding of eukaryote mitotic chromosomes, through
nucleoids. It has been proposed that chromosomes
might simply occupy maximum-entropy conformations,
in the manner of confined random-coil polymers (1,2).
However, sequence position analyses reveal DNA to be
spatially ordered. Chromosomes of Escherichia coli (3–5)
and Caulobacter crescentus (6) have loci precisely pos-
itioned inside the cell, with fluctuations too small to be
consistent with random-polymer statistics (7). In eukary-
ote cells, interphase chromosomes in differentiated cells
occupy distinct territories (8). Furthermore, analyses of
DNA juxtapositions inside eukaryote nuclei reveal that
loci up to tens of megabases apart along chromosomes
are positioned near one another in the nucleus (9,10),
with statistical properties inconsistent with random-
polymer organization (10).
Detailed characterizations of specific cases of in cis
gene regulation also indicate that chromosomes have a
well-defined ‘loop domain’ organization, with specific
but distant sequences along the same chromosome pos-
itioned to be near one another (11). It is thought that
‘chromatin-bridging’ proteins (12) somehow stabilize
these loop structures, but the processes by which
sequence-defined chromatin loops are established and
maintained are unknown.
Strong correlations of juxtaposed DNA sequences are
especially clear during eukaryote mitosis, when chromo-
somes are compactly folded, following their replication.
Chromosomes are ‘condensed’ by folding along their
length into linear paired-chromatid noodle-like structures,
with a well-defined thickness and strikingly uniform struc-
tural and mechanical properties (13). As mitotic chromo-
Nucleic Acids Research, 2012, 1–11
doi:10.1093/nar/gks925
Nucleic Acids Research Advance Access published October 15, 2012
atMITLibrariesonNovember13,2012http://nar.oxfordjournals.org/Downloadedfrom
Marko 2013
Nasmyth 2001
on July 13, 2010rstb.royalsocietypublishing.orgDownloaded from
doi: 10.1098/rstb.1990.0012
, 285-2973261990Phil. Trans. R. Soc. Lond. B
A. D. Riggs
Folding and Enhancer Function
Memory, and Type 1 DNA Reeling Could Aid Chromosome
DNA Methylation and Late Replication Probably Aid Cell
References http://rstb.royalsocietypublishing.org/content/326/1235/285#related-urls
Article cited in:
Rapid response
http://rstb.royalsocietypublishing.org/letters/submit/royptb;326/1235/285
Respond to this article
Email alerting service hereright-hand corner of the article or click
Receive free email alerts when new articles cite this article - sign up in the box at the top
http://rstb.royalsocietypublishing.org/subscriptionsgo to:Phil. Trans. R. Soc. Lond. BTo subscribe to
on July 13, 2010rstb.royalsocietypublishing.orgDownloaded from
Riggs 1990

Genome folding by loop extrusion and compartmentalization

Recommended

Recommended

More Related Content

Similar to Genome folding by loop extrusion and compartmentalization

Similar to Genome folding by loop extrusion and compartmentalization (20)

Recently uploaded

Recently uploaded (20)

Genome folding by loop extrusion and compartmentalization