OPINION
HEAT repeats – versatile arrays of amphiphilic helices working in
crowded environments?
Shige H. Yoshimura
1,
*
and Tatsuya Hirano
2,
*
ABSTRACT
Cellular proteins do not work in isolation. Instead, they often function
as part of large macromolecular complexes, which are transported
and concentrated into specific cellular compartments and function in
a highly crowded environment. A central theme of modern cell biology
is to understand how such macromolecular complexes are
assembled efficiently and find their destinations faithfully. In this
Opinion article, we will focus on HEAT repeats, flexible arrays of
amphiphilic helices found in many eukaryotic proteins, such as
karyopherins and condensins, and discuss how these uniquely
designed helical repeats might underlie dynamic protein–protein
interactions and support cellular functions in crowded environments.
We will make bold speculations on functional similarities between the
action of HEAT repeats and intrinsically disordered regions (IDRs) in
macromolecular phase separation. Potential contributions of HEAT–
HEAT interactions, as well as cooperation between HEATs and IDRs,
to mesoscale organelle assembly will be discussed.
KEY WORDS: HEAT repeat, Karyopherin, Condensin,
Molecular crowding, IDR, Phase separation, Hydrogel
Introduction
HEAT repeats, repetitive arrays of short amphiphilic α-helices, are
found in a wide variety of eukaryotic proteins with diverse
functions. The acronym HEAT comes from four proteins that were
originally found to contain this repeat motif, that is Huntingtin,
elongation factor 3, the A subunit of protein phosphatase 2A
(PP2A) and the signaling kinase TOR1 (Andrade and Bork, 1995).
Previous structural and biophysical studies have provided evidence
that HEAT repeats undergo highly flexible and elastic
conformational changes when they interact with different
binding partners or when external forces are applied to them
(Grinthal et al., 2010). This high degree of flexibility is based on
an unusual hydrophobic core that supports intramolecular helix–
helix interactions (Kappel et al., 2010), and therefore has a
potential to respond to differential environmental factors, such as
ionic strengths and macromolecular crowding. Very little is
known, however, abo ut how these unique structural properties of
HEAT repeats might be utilized in the various functions of
macromolecules and in their specific intracellular contexts. In this
Opinion article, we will provide an overview of and discuss two
seemingly distinct cellular processes, namely, nucleo-cytoplasmic
transport and mitotic chromosome assembly, in which HEAT
repeat proteins play crucial roles. For instance, karyopherins,
which are involved in nuclear transport, flexibly change their own
conformation during nuclear translocation to move across
the amphiphilic environment inside the nuclear pore channel.
The HEAT subunits of con densin complexes appear to use their
flexibility to support the dynamic assembly of chromosome axes
in the highly crowded environment of the interior of
chromosomes. We argue here that, in both cases, the
amphiphilic nature of the HEAT repeats is at the core of these
dynamic functions. Finally, we will also draw attention to potential
similarities between HEAT-mediated protein dynamics and phase
separation, an emerging concept of macromolecular assembly that
is driven by proteins containing intrinsically disordered regions
(IDRs).
Distribution of HEAT repeats in a wide variety of eukaryotic
proteins
A single HEAT motif (∼30–40 amino acids long) is composed of a
pair of α-helices (referred to as A- and B-helices) connected by a
short linker. The motif is highly degenerate at the primary structure
level and can only be recognized by a very loose consensus
sequence (Fig. 1A) (Neuwald and Hirano, 2000). Despite the
degenerate primary structure, the secondary and tertiary structures
of the HEAT motif are highly characteristic and well conserved. The
two helices are amphiphilic (i.e. one surface is enriched with
hydrophilic residues and the other surface with hydrophobic ones),
and are arranged in an anti-parallel fashion so that their hydrophobic
surfaces are concealed (Fig. 1B). The conserved hydrophobic
residues help to define a rotational orientation of the two helices,
and proline and aspartate residues are often found in the turn region.
An additional unique property of the HEAT motif is the existence of
another proline residue within the A-helix. This proline residue
often kinks the helix and thereby affects the curvature of the
solenoid (Cingolani et al., 1999), although its functional
significance is not yet fully understood.
Multiple HEAT motifs occur in a long linear array, and constitute
a HEAT repeat. The number of repeating motifs within individual
HEAT repeat proteins is variable and ranges from 15 to 50, or even
more. Owing to the loose consensus sequence, however, the exact
positions and numbers of HEAT motifs are difficult to deduce from
the primary sequences alone without any additional information
from crystal structures. Based on their overall domain organizations,
HEAT repeat proteins can be classified into three groups (Group
I–III; Fig. 1C). Proteins in Group I are composed of a long
consecutive repeat of HEAT motifs with little or no other
discernible domains. This group includes karyopherins, a large
family of nucleo-cytoplasmic transport receptors, and the A
(scaffold) subunit of protein phosphatase 2A (PP2A), one of the
founding members of HEAT repeat proteins (Xu et al., 2006; Cho
and Xu, 2007). In Group II, stretches of IDRs divide a HEAT repeat
array into several blocks; this group includes the regulatory subunits
of condensin I (CAP-D2 and CAP-G, also known as NCAPD2 and
NCAPG, respectively) and of cohesin (SA2, also known as STAG2,
1
Graduate School of Biostudies, Kyoto University, Kyoto 606-8501, Japan.
2
Chromosome Dynamics Laboratory, RIKEN, Saitama 351-0198, Japan.
*Authors for correspondence (yoshimura@lif.kyoto-u.ac.jp; hiranot@riken.jp)
T.H., 0000-0002-4219-6473
3963
© 2016. Published by The Company of Biologists Ltd
|
Journal of Cell Science (2016) 129, 3963-3970 doi:10.1242/jcs.185710
Journal of Cell Science
and PDS5), as well as the tubulin-polymerizing factor TOG (also
known as XMAP215 and CKAP5) (Fox et al., 2014). HEAT repeat
proteins of Group III possess additional well-defined structural or
functional domain(s) within single polypeptides. For instance, in
mammalian target of rapamycin (mTOR), a large N-terminal HEAT
repeat is followed by a tetratricopeptide-repeat (TPR) and a protein
kinase domain (Aylett et al., 2016). Similarly, the TATA-binding
protein-associated factor MOT1 (also known as BTAF1) contains a
Swi2/Snf2-type ATPase domain (Wollmann et al., 2011). It is also
important to note that many HEAT repeat proteins interact with
multiple proteins and often function as part of a large protein
complex (Fig. 2). From this point of view, many HEAT repeat
proteins that belong to group I function as scaffolds that
accommodate adaptable interactions with numerous different
binding partners. In contrast, HEAT repeat proteins classified into
groups II or III have a limited number of binding partners, if any.
Structural properties of HEAT repeats
A number of crystallographic studies have revealed three-
dimensional structures of HEAT repeat proteins, often together
with their binding partners (for a review, see Stewart, 2007). In
these structures, adjacent HEAT motifs are linked by short (inter-
unit) turns, and they are successively stacked with each other,
forming a two-layered helical array. Owing to twists and tilts
between adjacent motifs, the entire repeat forms a right-handed
solenoid in which A- and B-helices are aligned on the convex and
WNPCKAAGVCLMLLSTC
DEVALQGIEFWSNVCDE
PTFLVELSRVLA
PELIPQLVANVT
NEILTAIIQGMRKE
HFIMQVVCEATQ
PALFAITIEAMK
QYLVPILTQTLT
kqdendd
PHVLPFIKEHIK
IQAMPTLIELMK VVVRDTAAWTVGRICEL
CAP-D2
(condensin I)
CAP-G
(condensin I)
SA2
(cohesin)
AP-2α1
TOG/XMAP215
mTOR
Symplekin
MOT1/BTAF1
PR65A
CAND1
Karyopherins
PDS5
(cohesin)
100 aa
HEAT motif crystalized
HEAT motif predicted
Intrinsically disordered region
Appendage
AB
ATPase domain
TPR Kinase domain
Rif1
DNA binding
Group II
(HEAT with IDRs)
Group III
(HEAT with other
functional domain(s))
Group I
(HEAT only)
C
..LLP.L...
Φ
. ..D.. ..VR..A...L..L....
QVARVAAGLQIKNSLTS
EHMKESTLEAIGYICQD
NNVKLAATNALLNSLEF
TRVRVAALQNLVKIMSL
A-helix
A-helix
B-helix
B-helix
Turn
Turn
33
127
171
216
258
320
365
407
67
160
204
247
289
358
396
438
NPD
DPS
NPGNS
N
PNST
S-DI
DDD
EPS
CPD
WRYRDAAVMAFGCILEG
Key
Fig. 1. Overview of HEAT-motif-containing proteins. (A) Sequence alignment of HEAT motifs in mouse importin β. A consensus sequence is shown at the
bottom. A HEAT motif is composed of a pair of α-helices (A- and B-helices) that is connected by a short linker (turn). Conserved hydrophobic residues in the
helices are marked by an orange background. Proline and positively charged (arginine or lysine) residues conserved in the A- and B-helices, respectively, are
boxed. (B) In a HEAT motif, the A- and B-helices are arranged in an antiparallel fashion through hydrophobic interactions. Hydrophobic residues are marked in
orange. (C) Domain organizations of the three groups of HEAT-motif-containing proteins in humans. HEAT motifs, IDRs (brown) and other functional domains
(pink) are shown. HEAT motifs whose crystal structures have been determined are shown in green, whereas HEAT motifs predicted solely based on their primary
sequences are shown in gray.
3964
OPINION Journal of Cell Science (2016) 129, 3963-3970 doi:10.1242/jcs.185710
Journal of Cell Science
concave surfaces, respectively (Fig. 3A). The overall conformation
of the solenoid (i.e. its diameter, curvature and pitch) varies from
protein to protein, and is also affected by the interactions formed
with their binding partners (Conti et al., 2006; Lee et al., 2000;
Forwood et al., 2010). For example, structural comparison of the
karyopherin importin β with and without the cargo has revealed
substantial differences in the curvature of the solenoid (Cingolani
et al., 1999, 2000). These results suggest that each crystal structure
represents one snapshot of a number of different conformations that
could be found in the entire energy landscape. Such structural
flexibility is thought to play an important role in the ability of the
protein to simultaneously interact with multiple binding partners. In
the case of the PP2A holoenzyme, the HEAT-containing A subunit
functions as a flexible scaffold that brings together the catalytic
subunit and a wide variety of different regulatory subunits involved
in substrate recognition (Xu et al., 2006; Cho and Xu, 2007;
Janssens et al., 2008) (Fig. 2).
The structural flexibility of HEAT repeats has been directly
characterized by spectroscopic approaches (Tsytlonok et al., 2013)
as well as by small angle X-ray scattering (Forwood et al., 2010).
Molecular dynamics simulations have also demonstrated that, when
external forces are applied at the ends of the molecule, the HEAT
repeats exhibit unique elastic properties similar to a Hookean spring,
whereby the extension is proportional to the tension applied
(Grinthal et al., 2010; Kappel et al., 2010). This means that the
HEAT repeat is highly elastic against external forces (Fig. 3B).
Remarkably, such linear extension is completely reversible, and can
be observed up to forces of ∼100 pN after which, at a certain point,
inter-helical interactions collapse (Grinthal et al., 2010). These
findings suggest that the stress imposed on the ends of the HEAT
repeats is redistributed along the entire repeat array.
What is the physiological significance of the structural
flexibility and ela sticity of HEAT repeats? One possibility is that
the HEAT array functions as a mechanosensor by sensing and
utilizing mechanical force to modify protein function ( Grinthal
et al., 2010; Viswanathan an d Auble, 2011). For example, an
external force appli ed to the HEAT subunit of PP2A could change
the mode of inter-subunit interactions, thereby modulating the
catalytic activity of the enzyme (Grinthal et al., 2010).
Alternatively, even without external forces, structural
fluctuat ions of the array could help expose binding sites for
other proteins through a ‘fly-casting’ mechanism (Tsytlonok et al.,
2013). In additio n, the convex and concave arrays of the helice s
display different degrees of elasticity (Grinthal et al., 2010),
thereby conferring highly complex elastic properties on the two-
layered helical array of HEAT repeats.
HEAT repeats in nucleo-cytoplasmic transport
Karyopherins are among the best-studied classes of HEAT repeat
proteins. They are involved in the molecular transport between the
cytoplasm and nucleoplasm through the nuclear pore complex
(NPC) that is embedded in the nuclear envelope (Peters, 2009). In
the case of importins, they bind to their cargos in the cytoplasm and
travel through the NPC, before releasing them in the nucleus
(Fig. 3C). This catch-and-release mechanism and, hence, the
directionality of the transport, is dependent on differential
Adaptor complexes
Protein phosphatase 2A
TORC MOT1/BTAF1 Rif1
Poly-adenylation
complex
CAND1
TOG/XMAP215Condensins Cohesin
Karyopherin
Roc1
Cullin
SMC2 SMC4
Condensin I
CAP-D2 CAP-G
ATP ATP
Condensin II
SMC2 SMC4
CAP-D3 CAP-G2
ATP ATP
AP-3
β 3-adaptin
δ-adaptin
Kinase
mTOR
RAPTOR
or
RICTOR
Symplekin
CPSF
Tubulin
CstF
σ1
β 1-adaptin
γ-adaptin
AP-1
GTP
Arf1
Cargo
AP-2
β 2-adaptin
α-adaptin
Cargo
HEAT(with crystal structure)
HEAT(no crystal structure)
IDR in HEAT repeat proteins
IDR in non-HEAT proteins
PPP2R1
PPP2C
PR55
etc.
Crm1
Cargo
GTP
Ran
FG-Nups
Importin β
Cargo
GTP
Ran
or
FG-Nups
Loader
SMC1 SMC3
NIPBL
RAD21
SA2 PDS5
ATP ATP
MAU2
Appendages
Group II Group III Group I
ATPase
MOT1
TBP
DNA
binding
Rif1
αβ
CAP-H2
CAP-H
μ1
σ2
μ2
σ3
μ3
Key
Fig. 2. HEAT repeats as part of large protein complexes. Some HEAT repeat proteins function as intrinsic subunits of large protein complexes, whereas others
only temporarily interact with their partners. Many of HEAT repeat proteins that belong to Group I have numerous binding partners. In contrast, HEAT repeat
proteins classified into Groups II or III have a limited number of binding partners, or possess specific functional domains (shown in pink) within single polypeptides.
IDRs that occur within HEAT repeat proteins or HEAT-repeat-containing complexes (in-cis action) are shown in dark brown, whereas IDRs that occur within non-
HEAT protein or complexes (in-trans action) are shown in light brown.
3965
OPINION Journal of Cell Science (2016) 129, 3963-3970 doi:10.1242/jcs.185710
Journal of Cell Science
localization of the GTP- and GDP-bound forms of the small GTPase
Ran (Lee et al., 2005; Matsuura and Stewart, 2004).
A number of crystallographic studies have revealed that the
structural flexibility of importin β has important roles in its
interactions with both its cargo and RanGTP. A structural
comparison of different importin β molecules that are bound to
cargo, RanGTP or nucleoporins, has revealed conformational
differences not only in specific HEAT motifs, but also in the
entire molecule (Fukuhara et al., 2004; Lee et al., 2005). An
important implication here is that each of these distinct
conformations of importin β might not so much represent a
particular metastable structure in the entire energy diagram, but
rather corresponds to one snapshot of a wide array of possible
flexible conformations, as has been demonstrated by force-applying
molecular dynamics simulations (Grinthal et al., 2010; Kappel
et al., 2010) (Fig. 3B). Indeed, when importin β translocates through
the NPC with its cargo, it needs to inte ract with a number of different
nucleoporins (Nups) at its convex surface, while simultaneously
holding onto the cargo at its concave surface. This challenging task
requires dynamic and flexible conformational changes of the HEAT
repeat of importin β.
Recent studies have focused on the interaction between
karyopherins and intrinsically disordered Nups that contain
phenylalanine-glycine motifs (collectively referred to as FG-
Nups) (Milles et al., 2015; Bestembayeva et al., 2015; Zahn et al.,
2016). Such hydrophobic residues are believed to crosslink the
flexible polypeptide chains and to form a hydrogel, a meshwork
structure that prevents cellular macromolecules from passively
diffusing through the nuclear pore. Karyopherins interact with the
FG motifs and other hydrophobic residues of FG-Nups through a
hydrophobic pocket that is formed by adjacent A-helices of their
HEAT repeat (Bayliss et al., 2000, 2002; Liu and Stewart, 2005).
Our recent spectroscopic analysis combined with molecular
dynamics simulation of importin β has demonstrated that the
structural flexibility of HEAT repeats plays a crucial role in allowing
the migration through the crowded space of the nuclear pore channel
and is mediated through interactions with FG motifs (Yoshimura
et al., 2014). Here, a number of weak interactions between multiple
FG motifs and importin β induce temporary conformational changes
in both the HEAT repeat and the matrix of FG-hydrogels, which
enable karyopherins to migrate through the hydrogel-like
environment of the nuclear pore channel (see below for more
detailed discussion).
HEAT repeats in mitotic chromosome dynamics
Condensins are large protein complexes that play a fundamental role
in chromosome organization and segregation (Hirano, 2016). Most
eukaryotes have two different types of condensin complexes
(condensins I and II), each of which is composed of five subunits
(Fig. 2). The two complexes share the same pair of structural
maintenance of chromosomes (SMC) ATPase subunits, but have
distinct sets of non-SMC regulatory subunits. Among these,
condensins I and II have different pairs of HEAT subunits, CAP-
D2 and CAP-G, and CAP-D3 (NCAPD3) and CAP-G2 (NCAPG2),
respectively. Although condensin-like complexes are also found
among most bacterial and archaeal species, the HEAT-containing
subunits are unique to eukaryotic condensins, implying that the
HEAT subunits might be involved in eukaryote-specific aspects of
large-scale chromosome organization.
However, exactly how this type of elaborate protein machine
works to organize mitotic chromosomes is not fully understood. A
recent study using Xenopus cell-free egg extracts has provided
evidence that the HEAT subunits of condensin I have crucial roles in
the dynamic assembly of chromosome axes (Kinoshita et al., 2015).
Interestingly, the two HEAT subunits appear to have distinct roles in
this process and are possibly involved in both construction and
deconstruction of chromosomes that occur upon mitotic entry and
exit, respectively. These findings raise the possibility that regulated
HEAT– HEAT interactions between different condensin complexes
underlie the organization of chromosome axes. At present, however,
there is no direct evidence that supports this idea. To provide
evidence for such a mechanism, several issues need to be taken into
account. Firstly, if the predicted HEAT–HEAT interactions take
place, then they would not involve stereospecific, stable
interactions. Rather they would consist of an ensemble of
multivalent, weak interactions that reflect the flexible and elastic
nature of HEAT repeats (Kappel et al., 2010). Secondly, such
interactions would be highly dynamic; condensins turn over rapidly
under the control of their SMC ATPase activity, as has been implied
from experiments using mutant complexes in cell-free extracts
(Kinoshita et al., 2015) or from fluorescence recovery after
RanGTP
Cargo
Importin
β
Nuclear pore
complex
Cytoplasm
Nucleoplasm
B-helix
HEAT
motif
A-helix
HEAT
motif
HEAT
motif
B-helix layer
(concave)
A-helix layer
(convex)
A
B
C
Top
view
Side
view
Top
view
Side
view
Fig. 3. Structural properties of HEAT repeats and the action of
karyopherins. (A) Structure of a HEAT repeat array of yeast importin β (PDB
code: 3ND2). Multiple HEAT motifs, each being composed of a pair of
α-helices (A- and B-helices), are stacked with each other, forming a
two-layered array. (B) Structural flexibility of HEAT repeats. Because the two-
layered array of amphiphilic helices are organized by weak hydrophobic
interactions, HEAT repeats are highly flexible and elastic; they have the
potential to undergo large conformational changes by either interacting with
other proteins, or responding to external forces or environmental changes.
A-helices present in the convex surface are shown in light green, whereas
B-helices present in the concave surface are shown in dark green.
(C) Transport model of importin β through the nuclear pore complex. Importin β
binds to its cargo in the cytoplasm and travels through the NPC, which is
composed of flexible FG-Nups. Conformational changes occurring in the
HEAT repeat facilitate the translocation of the importin–cargo complex through
the crowded environment of the diffusion barrier. In the nucleoplasm, RanGTP
binds to importin β and releases the cargo from importin β.
3966
OPINION Journal of Cell Science (2016) 129, 3963-3970 doi:10.1242/jcs.185710
Journal of Cell Science
photobleaching (FRAP) experiments in vivo (Gerlich et al., 2006).
Thirdly, these postulated interactions would occur only on (or
inside) chromosomes, and would not take place when the condensin
complexes are not bound to chromosomes. In fact, no physical
interaction between purified condensin complexes has been
detected thus far. Furthermore, the molecular environment
surrounding and constituting mitotic chromosomes might also be
crucial as discussed below.
Both condensins I and II are enriched at the axial core of
metaphase chromosomes, and their cooperative actions have crucial
roles in determining the shape and physical properties of eukaryotic
chromosomes (see Box 1). Condensin II associates with
chromosomes in prophase earlier than condensin I, and is found
more internally than condensin I in metaphase chromosomes
(Fig. 4A). Why and how the different condensins are enriched at
these chromosomal regions is unknown. Although the interior of
mitotic chromosomes is highly crowded (Hancock, 2012;
Wachsmuth et al., 2008), at the same time, it is also a network of
well-solvated chromatin that is held together by noncovalent
crosslinking proteins (Poirier and Marko, 2002) and is readily
accessible by macromolecules (Hihara et al., 2012). Thus, the
interior of chromosomes could share some of the physico-chemical
properties of a hydrogel. In fact, micromechanical experiments
using micropipettes have shown that mitotic chromosomes are
highly elastic objects that return to their native lengths even after
five-fold extensions (Marko, 2008). Compared with the interior of
chromosomes, their periphery is expected to be less dense and to
behave like a liquid, as has been predicted for interphase chromatin
(Maeshima et al., 2016). Along these lines, an intriguing
observation from an early study is that the condensin subunit
SMC2 first appears at the surface of condensing chromosomes in
middle prophase, and then suddenly translocates into the interior of
chromosomes where axial structures are formed by late prophase
(Kireeva et al., 2004) (Fig. 4A). We speculate that this relocalization
is accompanied by conformational changes in condensin subunits,
in particular, the HEAT-containing subunits. HEAT-mediated
condensin–condensin interactions could then occur in a
B
A
Condensin II
loading
Phase
transition?
Condensin I
loading
Middle
prophase
Late
prophase
Metaphase
Liquid phase?
Hydrogel
phase?
Cross section of
a chromosome
Condensin II
Fig. 4. Dynamic behaviors of condensins during mitosis. (A) Architecture
and subunit composition of the eukaryotic condensin complexes are shown in
Fig. 2. Condensins I and II have different pairs of HEAT subunits, CAP-D2–
CAP-G and CAP-D3–CAP-G2, respectively. Condensin II (dark green) first
appears on the surface of chromatin in middle prophase, and then translocates
into the interior of chromatids (represented by the dashed cylinder) to form their
central axes by late prophase. Upon nuclear envelope breakdown in
prometaphase, condensin I (light green) gains access to chromatids and
accumulates around the condensin-II-positive chromosome axes by
metaphase. (B) Hypothetical actions of condensin II. Condensin II binds to
chromosomes at their periphery, possibly through an ATP-dependent
entrapment mechanism. The ATPase cycle of the SMC subunits could further
modulate any conformational changes of the HEAT subunits and also trigger
HEAT-mediated condensin–condensin interactions in the interior of
chromosomes (Kinoshita et al., 2015). The translocation of condensin II from
the exterior to the interior of chromosomes and the resulting assembly of
chromosome axes could steer a phase transition of chromatin from a liquid-like
structure to a hydrogel.
Box 1. Chromosome size and shape – relevance of two
condensin com plexes
400 nm
100 nm
200 nm
Somatic
chromatid
Embryonic
chromatid
Fission yeast
chromatid
Condensin I Condensin II
1
:
0.2 1
:
1
I
:
II ratio
1
:
0
Most eukaryotic species have both condensins I and II, but some lack
condensin II (Hirano, 2016). Shown here are cross-sections of three
different chromatids with different diameters. For instance, small
chromatids, such as those in fission yeast, contain condensin I only
(shown on the left). In vertebrates, embryonic (middle) and somatic
(right) chromatids have different ratios of condensins I and II, and display
different shapes; the embryonic chromatids are thin and long, whereas
the somatic ones are thick and short. Experiments using Xenopus cell-
free egg extracts have provided evidence that the relative ratio between
condensin I and II determines chromosome shape (Shintomi and Hirano,
2011). Moreover, condensin II is located more internally than condensin
I, which is found along the axial core of chromatids (Ono et al., 2003). On
the basis of these and other data, condensin II has been proposed to
contribute to lengthwise shortening, especially of large and thick
chromatids (Shintomi and Hirano, 2011), and to confer their physical
rigidity (Houlard et al., 2015). Thus, if condensins indeed act as
crosslinkers for chromatin networks as has been predicted (Marko,
2008), condensin II would be a more robust crosslinker than condensin
I. This raises the question of whether the different pairs of HEAT repeat
subunits present in the two condensin complexes confer different
crosslinking properties. To that end, it will be of interest to carefully
compare their biochemical properties in order to understand how they
might differentially contribute to assembly and maintenan ce of
chromosomes with characteristic physico-chemical properties. Critical
comparisons with non-biological, amphiphilic materials that self-
assemble to form a rod-shaped structure (Qiu et al., 2015) will be
another exciting direction of future research.
3967
OPINION Journal of Cell Science (2016) 129, 3963-3970 doi:10.1242/jcs.185710
Journal of Cell Science