Chromatin Higher-order Structure and Dynamics€¦ · on chromatin organization has increasingly...

2010;2:a000596 originally published online April 7, 2010Cold Spring Harb Perspect Biol Christopher L. Woodcock and Rajarshi P. Ghosh Chromatin Higher-order Structure and Dynamics

References

urlshttp://cshperspectives.cshlp.org/content/2/5/a000596.full.html#related-Article cited in: http://cshperspectives.cshlp.org/content/2/5/a000596.full.html#ref-list-1

This article cites 131 articles

serviceEmail alerting

click herebox at the top right corner of the article orReceive free email alerts when new articles cite this article - sign up in the

Subject collections

(29 articles)The Nucleus � Articles on similar topics can be found in the following collections

http://cshperspectives.cshlp.org/site/misc/subscribe.xhtml go to: Cold Spring Harbor Perspectives in BiologyTo subscribe to

Copyright © 2010 Cold Spring Harbor Laboratory Press; all rights reserved

Cold Spring Harbor Laboratory Press on January 14, 2011 - Published by cshperspectives.cshlp.orgDownloaded from

http://cshperspectives.cshlp.org/content/2/5/a000596.abstract.html

http://cshperspectives.cshlp.org/content/2/5/a000596.full.html#ref-list-1

http://cshperspectives.cshlp.org/content/2/5/a000596.full.html#related-urls

http://cshperspectives.cshlp.org/cgi/alerts/ctalert?alertType=citedby&addAlert=cited_by&saveAlert=no&cited_by_criteria_resid=cshperspect;2/5/a000596&return_type=article&return_url=http://cshperspectives.cshlp.org/content/2/5/a000596.full.pdf

http://cshperspectives.cshlp.org/cgi/collection/the_nucleus

http://cshperspectives.cshlp.org/site/misc/subscribe.xhtml

http://cshperspectives.cshlp.org/

http://www.cshlpress.com

Chromatin Higher-order Structure and Dynamics

Christopher L. Woodcock1,2 and Rajarshi P. Ghosh2

1Biology Department, University of Massachusetts, Amherst, Massachusetts 010032Program in Molecular and Cellular Biology, University of Massachusetts, Amherst, Massachusetts 01003

Correspondence: [email protected]

The primary role of the nucleus as an information storage, retrieval, and replication siterequires the physical organization and compaction of meters of DNA. Although it has beenclear for many years that nucleosomes constitute the first level of chromatin compaction,this contributes a relatively small fraction of the condensation needed to fit the typicalgenome into an interphase nucleus or set of metaphase chromosomes, indicating that thereare additional “higher order” levels of chromatin condensation. Identifying these levels, theirinterrelationships, and the principles that govern their occurrence has been a challengingand much discussed problem. In this article, we focus on recent experimental advancesand the emerging evidence indicating that structural plasticity and chromatin dynamics playdominant roles in genome organization. We also discuss novel approaches likely to yieldimportant insights in the near future, and suggest research areas that merit further study.

As the cell’s primary information storage,retrieval, and duplication organelle, the

nucleus, by analogy with human-created infor-mation repositories (libraries and computerdiscs), might be expected to be highly struc-tured. However, chromatin within the livingnucleus appears to be organized in a quiteunstructured manner, and the rules that defineor encode the principles of chromatin organi-zation have been difficult to decipher. Thesefrustrations are reflected in some reviews of chro-matin organization with titles such as “Chroma-tin higher-order structure: chasing a mirage?”(van Holde and Zlatanova 1995), “Higher-orderstructures of chromatin: the elusive 30 nm fiber”(Tremethick 2007), and “Chromatin fiber struc-ture: where is the problem now?” (van Holde andZlatanova 2007). In this article, we discusswhat is known about chromatin structure, and

consider the prospects of improving our under-standing in the near future. In recent years, workon chromatin organization has increasinglyfocused on the many dynamic aspects of chro-matin, which contribute to its structural andfunctional plasticity, and it is becoming clearthat dynamics plays a crucially important func-tional role, perhaps contributing to nuclear self-organization (Misteli 2001).

CHROMATIN HIGHER-ORDER STRUCTURE

In the context of chromatin, “higher-orderstructure” may be defined as any assemblage ofnucleosomes that assumes a reproducible con-formation in 3D space. The most obvious chro-matin higher-order structure is the mitotic/meiotic chromosome in which the DNA is com-pacted some 10,000- to 20,000-fold. Metaphase

Editors: David Spector and Tom Misteli

Additional Perspectives on The Nucleus available at www.cshperspectives.org

Copyright # 2010 Cold Spring Harbor Laboratory Press; all rights reserved; doi: 10.1101/cshperspect.a000596

Cite this article as Cold Spring Harb Perspect Biol 2010;2:a000596

1




chromosomes have characteristic shapes, band-ing patterns, and locations of specific genes.Although chromosomes have a consistent struc-ture at the light microscope level in terms of lon-gitudinal positioning of bands and genes, thisconsistency is evidently modulated by an intrin-sic variability in longitudinal position of up to0.3 mm (Strukov and Belmont 2009). However,these authors reported that there was no consis-tency in the axial positioning of loci, suggestingthat chromosome architecture involves consid-erable plasticity at some level or levels of folding.Understanding how chains of nucleosomes arefolded in the creation of mitotic chromosomesand how they are arranged in interphase contin-ues to be an exciting yet technically challengingendeavor.

The concept of primary, secondary, tertiary,and quaternary structures used for proteins canalso be usefully applied to chromatin structuralhierarchies (Woodcock and Dimitrov 2001),with the beads-on-a-string organization ofnucleosomes and linker DNA constituting theprimary structure, and arrangements resultingfrom interactions between nucleosomes givingrise to secondary structures. Thus, the chroma-tin equivalent of protein secondary structuremight involve interactions analogous to thoseleading to a-helices and b-sheets. Unlike pro-teins that consist of sequences of 20 aminoacids, chromatin consists of a repeating chainof more-or-less identical nucleosomes, and thusmight be predicted to form highly “ordered”secondary structures. Note that the term “or-dered” can be used in the sense that a crystal ishighly ordered and also used to describe hier-archical levels where the concept is more akinto its meaning in “orders of magnitude.”

Although in one sense, the basic nucleo-some-linker DNA unit comprising the primarystructure of chromatin is simple, there are anumber of potentially variable parameters thatcontribute to complexity. There is structuraluncertainty even at the level of the nucleosomecore particle (NCP). Although the structure ofthe NCP is well established at the atomic level(Luger et al. 1997), portions of the four core his-tones are not seen in crystals because of theirevident mobility. For example, the unstructured

terminal regions that are extremely importantin modulating chromatin structure are notseen in X-ray data. The length of linker DNAbetween NCPs varies not only between species,but also between tissues of the same organism,and within a single nucleus (van Holde 1989).Mean linker lengths (defined as the nucleosomerepeat length minus the �145 bp of DNA pro-tected in NCPs) range from a low value of�20 bp in budding yeast to �75 bp in echino-derm sperm. For a typical vertebrate nucleusthe mean linker length is �35 bp. The rich liter-ature concerning the locations of nucleosomepositions on specific DNA sequences, revealsinstances, especially in the upstream controlregions of genes, in which nucleosomes are pref-erentially located or excluded, and this posi-tioning is often important in regulatingtranscription (Simpson 1991; Jiang and Pugh2009). For some DNA sequences, the preferencefor nucleosomes to be present at a particularlocation is retained when the nucleosomes arereconstituted on the DNA in vitro. In additionto linker DNA length, nucleosomes vary in theircomplement of core histone variants and his-tone postsynthetic modifications. All these var-iations are likely to impact chromatin secondarystructures.

The obvious experimental approaches todetermine chromatin secondary and higher-order structures using light and electron mi-croscopy yield disappointingly little informa-tion, largely because nucleosomes and linkerDNA cannot be adequately resolved in the com-pact chromatin that occurs in the nucleus. Arecent study using cryo-electron microscopy ofthin sections of chromosomes vitrified in vivo,and therefore expected to faithfully reflect thenative structure, provides an instructive exam-ple. Despite careful image processing of themicrographs, there was no evidence for anyhigher-order structure, leading the authors tosuggest that the chromatin forms a “molten”mass similar to that assumed by certain poly-mers, and that in this form, the size, shapeand trajectories of any higher-order structureswould not be resolved (Eltsov et al. 2008).This may well be the case, but nevertheless,the arrangements of arrays of nucleosomes in

C.L. Woodcock and R.P. Ghosh

2 Cite this article as Cold Spring Harb Perspect Biol 2010;2:a000596




3D space are to some extent constrained andnonrandom, and important information wouldbe obtained if it were possible to trace the pathsof chains of nucleosomes through the nucleus.The generally amorphous appearance of chro-matin in thin sections contrasts sharply withthe appearance of cytoskeletal elements suchas microtubules, which can be resolved in cellseven when in compact bundles. This strikingdifference supports the concept that tightlypacked chains of nucleosomes readily inter-digitate with each other, leading to the ratherfeatureless appearance of chromatin in EMimages (Woodcock and Horowitz 1995; Wood-cock 2006). A new technique for preparing fro-zen hydrated material that involves thinning ofsamples by milling with a focused ion beam(Marko et al. 2007, 2008) that does not involvethe compression and possible local heating ofcryosectioning is currently under developmentin several laboratories, and it will be most inter-esting to see whether this method providesimproved structural information when appliedto nuclear structures.

The 30-nm Fiber

In the absence of clear conclusions from theexamination of nuclei and chromosomes insitu, many investigators have focused on iso-lated chromatin. Through the use of nucleasesthat cut the linker DNA between nucleosomes,it is possible to isolate polynucleosomes andstudy their properties under defined conditionsin solution. Early work clearly established thatthe compaction state of polynucleosomes ishighly dependent on the ionic milieu. NakedDNA, being highly charged, is self-repulsive inlow ionic strength buffers—the self-repulsioncan be reduced or eliminated in the presenceof cations, with divalent and polyvalent cationsbeing especially effective. Chromatin behavessimilarly to DNA in this respect because theDNA negative charge is not fully neutralizedby histones. The response of chains of nucleo-somes to changes in the ionic environment canbe satisfactorily modeled (Clark and Kimura1990; Arya and Schlick 2006), suggesting thatour basic understanding of the response of

chromatin to changes in the ionic environmentis robust.

Electron microscopy of isolated polynucleo-somes clearly reveals an open beads-on-a-stringconformation in low salt and a progressive com-paction that occurs as the ionic strength israised. There is general agreement that an earlystage in compaction is the formation of a fiber�30 nm in diameter, and that, at least for poly-nucleosomes in vitro, this constitutes a bonafide chromatin secondary structure. However,the arrangement of nucleosomes and linkerDNA within isolated 30 nm fibers has been dif-ficult to study and remains controversial. Care-ful examination of isolated 30 nm fibers hasfailed to reveal any consistent clear-cut arrange-ment of nucleosomes. The realization thatnative nucleosomes are not identical, but, asdescribed earlier, vary in several parameters,suggested that these could be the source of theequivocal findings regarding 30-nm fiberorganization. To eliminate these potential prob-lems, investigators turned to fully defined artifi-cial polynucleosomes reconstituted onto DNAcontaining regularly spaced nucleosome local-ization sequences (Simpson et al. 1985). Amajor achievement using this approach wasthe successful crystallization and structuredetermination by X-ray diffraction of a tetranu-cleosome (Schlach et al. 2005). This showedunequivocally that, at least for this particularconstruct, exposed to crystallization-promotingconditions that included 90-mM Mgþþ, thecomplex assumed a zigzag arrangement, withcontacts between nucleosomes 1 and 3 andbetween 2 and 4, with the linker extendingbetween them (Fig. 1). The work clearly estab-lished the predominant internucleosome inter-actions that result in a specific secondarystructure for this particular chromatin con-struct in this ionic environment. It is not yetclear how far the zigzag structure can be ex-trapolated to long chromatin with hetero-geneity in histone content and linker length.Nevertheless, it does provide an importantcontext in which to view earlier results. Forexample, the interpretation of the first EMstudies of isolated compact 30-nm fibers interms of a simple solenoidal organization in


Cite this article as Cold Spring Harb Perspect Biol 2010;2:a000596 3




which linker DNA followed the helical trajec-tory established in the nucleosome is probablyincorrect, whereas subsequent suggestions in-volving some species of zigzag are probablycloser to the mark.

THE ROLE OF HISTONE H1

In addition to the four “core” histones presentas an octameric unit in the nucleosomes, a fifthhistone, H1 or linker histone, is present in mostnuclei. As its name implies linker histone isassociated with linker DNA, and provides par-tial nuclease protection for �20 bp of linkerDNA. H1 contains a conserved globular regionand extended amino- and carboxy-termini, thelatter being rich in lysines and able to interactstrongly with DNA. Although the exact locationof H1 in chromatin remains controversial, it isclear that it plays a crucial role in promotingchromatin higher-order structure. H1-contain-ing chromatin shows a distinct structural motifin which the entering and exiting linker DNAsegments are brought together, perhaps pro-moting an overall zigzag arrangement (Bednaret al. 1998).

Arrays of nucleosomes depleted in H1 donot readily form 30-nm fibers in vitro, and itis likely that loss of the charge neutralizationeffected by H1 is at least partly responsible.The crystals of tetranucleosomes discussedabove were created without H1—and to date,no publications of the X-ray structure of

H1-containing mono- or oligo-nucleosomeshave appeared. Modest depletion of H1 fromcells (achieved for the mouse by knocking outseveral of the six somatic H1 variants) resultsin reductions in nucleosome repeat length insome tissues whereas severe depletion is fatalfor both mice and Drosophila (Fan et al. 2003;Lu et al. 2009). H1 depletion also inhibits theproper folding of chromosomes at mitosis(Maresca and Heald 2006). Interestingly, thereis a strong linear relationship between H1 con-tent and nucleosome repeat length in vivo(Woodcock et al. 2006), a feature that will actto maintain the electrostatic balance betweenDNA and histones.

CHROMATIN FIBERS IN NUCLEI

Thin sections of typical nuclei reveal a coarsedifferentiation between strongly staining heter-ochromatin (see the following for more on thisterm) adjacent to the nuclear periphery andnucleoli, and more weakly staining euchro-matin dispersed throughout. Little internalstructural detail can be resolved in either com-ponent. There are, however, a few special casesthat are more informative. Echinoderm spermnuclei contain uniformly distributed and highlycompact chromatin, but on controlled reduc-tion in ionic strength, the chromatin swells,revealing a mass of �30-nm fibers. Althoughthe fibers appear to be quite uniform in diame-ter, tomographic 3D reconstructions reveal aninternal structure that consists of an irregulararrangement of nucleosomes and linkerDNA but with a distinctive underlying zigzagmotif (Horowitz et al. 1994). Sections of thetranscriptionally inert chicken erythrocyte nu-cleus also reveal a uniform compaction of chro-matin, which resolves into irregularly structured30-nm fibers when allowed to swell slightlyby, for example, lowering the ionic strength.In contrast, 30-nm fibers are rarely seen inmore typical nuclei from cycling cells, even aftercontrolled swelling (Woodcock and Horowitz,1995). These special cases in which the�30-nm organization is revealed on swellinghave several common features, including uni-form core histone modifications, a greater

N2

N1

N1'

N2'

Figure 1. The DNA path of a tetranucleosome asdetermined by X-ray diffraction. The structureconsists of two stacks of nucleosomes, with linkerDNA passing back and forth between them. Thus, theprimary interactions occur between alternate ratherthan adjacent nucleosomes along the DNA strand,creating a zigzag architecture. From Schlach et al. 2005.






abundance of linker histone (H1) relative tocore histones, the presence of H1 variants withmore positive charges and unusually long linkerDNA. It is not clear why these properties leadto the appearance of distinct �30-nm fibersin nucleo, but an attractive possibility is thatthese conditions favor intrafiber over interfiberinteractions.

NUCLEOSOME-NUCLEOSOMEINTERACTIONS IN ISOLATED CHROMATINFIBERS

One approach to understanding chromatin fi-ber structure is to identify nucleosome-nucleo-some interactions within the compact fiber.As noted, this cannot be achieved by directEM observation of compact fibers, but waysto examine this have been devised. One studyused the premise that the face-to-face nucleo-some contacts that occur in crystals of mono-nucleosomes (Luger et al. 1997) would likelybe seen in compact chromatin. The contactsinvolved the positively charged N-terminus ofhistone H4 and the “acidic patch” that includesa portion of histone H2A. Dorigo et al. (2004)created H4 and H2A constructs with a cysteinein the putative contact region and reconstitutedthem into nucleosomal arrays. The arrays werethen placed in conditions that promoted com-paction and formation of S-S crosslinks bet-ween close cysteines. If the basic architectureof the array were solenoidal, crosslinking wouldresult in a simple stack-of-coins conformation.In contrast, a zigzag arrangement would resultin the creation of a ladder-like structure withthe linker DNA segments forming the rungs.When cross-linking was initiated, the arraysclearly adopted the ladder-like structure,strongly supporting a zigzag architecture, albeitunder somewhat artificial conditions. An ap-proach that does not restrict internucleosomeinteractions to a specific histone-histone contactinvolves exposing nucleosomal arrays or wholecells to controlled formaldehyde crosslinkingsuch that a few of the nucleosome-nucleosomecontacts become covalently linked. Subsequ-ently, the arrays are allowed to disperse in lowsalt and imaged by electron microscopy. With

his technique, termed EMANIC (EM-assistednucleosome interaction capture) (Grigoryevet al. 2009), decondensed arrays appear in theopen beads-on-a-string conformation (Fig. 2A),whereas crosslinked sites are seen as touchingnucleosomes. Analysis of the crosslinking pat-tern of defined nucleosome arrays with H1present once again reveals a predominantly zig-zag pattern (Fig. 2 D, E). With H1 and 1 mMMgCl2 together, some crosslinking is also seenbetween adjacent nucleosomes, suggesting theseconditions favor a mixed, heteromorphic fiberarchitecture (Fig. 2F, G) in which the zigzagdominates, but is interrupted by the type ofnucleosome-nucleosome interaction expectedfor a solenoidal architecture. Interestingly, com-puter modeling predicts heteromorphic fibers,and also shows that this arrangement allowsgreater compaction as it reduces the crowdingof linker DNAs in the fiber interior (Grigoryevet al. 2009), and is more consistent with thehigher packing ratios seen in compact chroma-tin (Daban 2000; 2003).

Although there is substantial support for abasic propensity of chromatin to form zigzaghigher-order structures, results leading to alter-nate architectures have been presented. EMexamination of long reconstituted nucleosomalarrays containing linker histone reveals remark-ably uniform structures interpreted as consist-ing of solenoids in which adjacent gyresinterdigitate (Robinson et al. 2006). Thisarrangement also increases the nucleosomepacking density, and may thus better matchthe calculated in vivo packing. Additional stud-ies of long nucleosomal arrays have led to theconclusion that the architecture is stronglyinfluenced by linker length (Routh et al.2008), increasing the potential complexity ofchromatin structure in vivo, where varied linkerDNA lengths are the norm.

MECHANICAL PROPERTIES OF ISOLATEDCHROMATIN FIBERS

The development of sophisticated instrumentsusing “optical tweezers” has made it possibleto take an isolated chromatin fiber, anchorone end, and pull on the other end while






measuring the force/length relationship. Theapplied force is expected first to break theweak internucleosomal bonds holding the fibertogether, and ultimately, the much strongerintranucleosomal bonds. For a fiber with a zig-zag arrangement, the breaking of individualnucleosome-nucleosome interactions is pre-dicted to produce jumps in length with a consis-tent increment which would not be observedwith solenoid type structures. Analysis offorce/length curves for isolated chromatinfibers from chicken erythrocyte nuclei led tothe conclusion that a zigzag arrangement pre-dominated (Cui and Bustamante 2000). How-ever, force/length data from a more recentstudy using reconstituted nucleosomal arrays(Kruithof et al. 2009) were interpreted as sup-porting a solenoidal architecture. The differ-ences in starting material and ionic conditionsused in these studies makes it difficult to evalu-ate the findings. In the future, studies encom-passing a broader range of conditions and

chromatin substrates should provide moredefinitive information from this potentially val-uable technique.

INSIGHTS FROM COMPUTER MODELING

As noted earlier, computer modeling was consis-tent with the empirical finding of a heteromor-phic fiber architecture. Current “mesoscale”modeling programs are supplied with sophisti-cated structural and chemical models of thenucleosome, H1 (if desired) and linker DNA,as well as the selected ionic environment (Aryaand Schlick 2006). The changes in structureare then followed over a large number of smallincrements of time, and the final structuresanalyzed. As expected, the stochastic nature ofthe interactions involved lead to different finalstructures with each run of the program.However, consistent motifs emerge: the finalstructures are fibers, rather than spherical aggre-gates of nucleosomes, and, although zigzag

Crosslinking

A

Unfolding and EM

D’

I’

2

8

2

2

i+/–1

i+/–2

B

E F G H I

C D

Figure 2. EMANIC analysis of internucleosomal interactions. (A) Scheme of the EMANIC procedure. The twomodels for the structure of the chromatin 30-nm fiber, namely solenoid (Upper) and zigzag (Lower), lead todominant i + 1 and i + 2 internucleosome interactions, respectively. (B–D) EM of nucleosomereconstitutes crosslinked in low salt without linker histone show few crosslinks. (E-I) With H1 present, +2interactions predominate. D0 and I0 diagram the nucleosome arrays corresponding to the adjacent EMimages. From Grigoryev et al. 2009.






arrangements of nucleosomes and linker DNApredominate, they may be interspersed withthe bent linker motif characteristic of simple hel-ical arrangements of nucleosomes. Modelingresults are dependent on the accuracy of theinput data, and computational limitations stillpreclude modeling at the atomic level. Neverthe-less, this approach has the advantage of beingable to sample a wide range of conditions,informing the selection of conditions for empir-ical examination with “real” chromatin. Ascomputational speed continues to increase,modeling is likely to lead to more penetratinginsights into chromatin architecture.

LOCAL EVENTS THAT MAY MODULATECHROMATIN FIBER ARCHITECTURE

Much of recent work on chromatin secondarystructure has featured fully defined nucleoso-mal arrays in which recombinant histones areassembled in vitro onto DNA containing strongnucleosome positioning sequences. Yet, asnoted, even with highly uniform arrays, resultshave led to differing conclusions regarding theunderlying architecture. In the nucleus, localeffects of linker DNA length variability, core his-tone variants, and postsynthetic histone modi-fications are likely to contribute even moreirregularity to secondary structures. Additionalcomplexity is introduced by the presence ofnonhistone proteins known collectively asChromatin Architectural Proteins (CAPs) thatinfluence chromatin conformation (Luger andHansen 2005).

In addition to the “major” core histones,synthesized primarily in S phase and depositedat replication forks, there are numerous histonevariants encoded by separate genes, which areoften synthesized constitutively at low levelsand incorporated differently. Some of thesehave been shown to play a critical role in estab-lishing the local properties of chromatin inwhich they are embedded (Santenard andTorres-Padilla 2009). An important concept inthis regard is the distinction between the majorvariant that is synthesized during S phase anddeposited as newly replicated chromatin, andminor variants. It has been proposed that the

incorporation of different H3 variants servesas a “bar code” (Hake and Allis 2006) contribu-ting to the establishment of the local functionalpotential. Of the several variants of H2A, H2A.Zhas been particularly well studied and shown tomodulate chromatin higher-order structure.The X-ray structure of nucleosomes containingH2A.Z (Suto et al. 2000) reveals an enlargementof the ‘acidic patch’ at the nucleosome surfacethat appears to strengthen nucleosome-nucleo-some interactions, and contribute to the highlycompact chromatin surrounding centromeres(Greaves et al. 2007).

Another major source of chromatin varia-tion is the large number of postsyntheticmodifications of the core histones and H1(Kouzarides 2007). They fall into two major cat-egories, those that alter the charge on the his-tone, and those that establish binding sites forother proteins. Acetylation of lysines results ina reduction of the histone positive charge andweakens the DNA-histone interaction leadingto a less compact chromatin secondary struc-ture. The balance in activity between histoneacetyl transferases (HATs) and histone deacety-lases (HDACs) governs the acetylation statusof a given region of chromatin. In general,hyperacetylation is the hallmark of active chro-matin, whereas hypoacetylation is seen inrepressed chromatin. A well-studied example isthe acetylation of H4 at lysine 16. The loss of apositive charge at this site is likely to weakenthe internucleosome interaction between theamino-terminus of H4 and the acidic patch onH2A discussed above. Indeed, the effect of thismodification is to loosen the compaction ofchromatin secondary structures (Shogren-Knaak et al. 2006; Fischle et al. 2003). An exam-ple of a modification that establishes a bindingsite is the methylation of H3 at lysine 9. Themodification creates a binding site for a domainof the HP1 protein (Jacobs and Khorasanizadeh2002). As discussed below, HP1 binding isassociated with the formation and propagationof compact heterochromatin. Work on post-synthetic histone modifications has led tothe widely discussed concept that, collectively,they constitute a “histone code” that definesthe local structural and functional potential






of a region of chromatin (Jenuwein and Allis2001).

Chromatin architectural proteins, as theirname implies, do impact chromatin structure,mostly by increasing the compaction state(Luger and Hansen 2005). This appears to betheir sole common feature, because otherwisethey are highly diverse in both structure and inchromatin binding mechanism. Binding maybe restricted to specific DNA sequences (Poly-comb), methylated DNA (MeCP2), or specifichistone modifications (HP1). Only in a fewcases, such as the SIR family of DNA-bindingregulatory proteins is a detailed account of theirmolecular interactions with chromatin and witheach other emerging (Buchberger et al. 2008).Some CAPs contain more than one DNA/chro-matin binding sites, allowing them to act asinter- and/or intrafiber bridges, and thus pro-mote chromatin compaction.

Chromatin remodeling complexes may beconsidered as a distinct class of CAPs. Severalfamilies of these large multicomponent com-plexes have been identified. In vitro, they havethe ability to restructure, eject, or move nucleo-somes in an ATP-dependent manner (Clapierand Cairns 2009). This type of activity in vivowill likely influence chromatin secondary struc-ture, and, by altering the local positioning ofnucleosomes, modulate chromatin folding. Aninteresting case has been reported of the com-bined effect of a histone variant (H2A.Z), a his-tone modification (H3K9Me), and a CAP(HP1), all of which act to increase chromatincompaction (Fan et al. 2004). Thus, the pro-posed histone code based on histone modifica-tions may need to be expanded to includesynergistic effects of histone variants. A novelfunction in maintaining the pluripotential“open” state of chromatin for the remodelingfactor Chd1 (Gaspar-Maia et al. 2009) maypresage new insights into the roles of thesecomplexes.

A scenario of chromatin secondary struc-ture that, in our opinion, appears to fit all thecurrent data involves chains of nucleosomesthat are sufficiently close packed in vivo to inter-digitate, effectively merging with each other andpreventing individual fibers being seen by

microscopy. Nevertheless, the overall trajectoryof a chain of nucleosomes traces out an irregular�30-nm fiber within the nucleus. Fragmenta-tion of chromatin by limited nuclease digestionfollowed by elution from nuclei results in indi-vidual fibers becoming separated, and availablefor microscopic and biochemical characteri-zation. Irregularity dominates the architectureof isolated native chromatin fibers, reflectingthe heterogeneity in linker DNA length, andcomposition of histone variants, histone post-synthetic modification, and CAPs. The endresult is likely a hybrid (heteromorphic) orga-nization with a predominantly zigzag architec-ture interspersed with other structural motifs.Most of these sources of irregularity aredynamic, changing at different time scales.Thus, it is probably futile to expect that chroma-tin secondary structure can be defined with theprecision expected from studies of other biolog-ical fibers such as microtubules or TobaccoMosaic Virus. Thus, seeking a defined secon-dary structure for chromatin may indeed be“Chasing a Mirage” (van Holde and Zlatanova1995).

Structural Order Beyond the 30-nm Fiber—Electron Microscopy Results

The lack of definitive information on chromatinsecondary structure complicates the question ofwhether there is a bona fide set of hierarchicalstructures above the 30-nm fiber. Althoughthere is no doubt that some higher levels oforganization are present—the metaphase chro-mosome being an obvious example—preciselywhat form they take and whether they are trulyhierarchical is still a matter of intensive studyand debate. Much of this work uses the lightmicroscope and has been limited in resolutionto �250 nm by the properties of light. Theimpact of a �250-nm resolution limit is wellillustrated in a study that considers the possiblestructural arrangement of large genetic loci(�400 kb) by matching the predictions of anumber of models for large-scale chromatinstructures with actual observations (Mulleret al. 2004). Somewhat surprisingly, at the reso-lution attainable with a high quality confocal






microscope, all models fitted the data equallywell! The higher resolution of the EM, coupledwith tomography, and applied to thin sectionshas revealed short regions of fiber-like struc-tures with a variety of diameters, none particu-larly consonant with a hierarchical series basedon a 30-nm organization (Konig et al. 2007; Bel-mont et al. 1989, 1994).

More recently, however, significant improve-ments in technique have been made using an invivo immuno-gold technique that promises tolead to important insights (Kireev et al. 2008).The introduction by Belmont and colleaguesof techniques for constructing cell lines con-taining large foreign DNA segments that are“self identifying” through an ingenious use ofthe lac operator/repressor system (Belmontet al. 1999) has had a major impact within thefield and is widely used. A recent adaptationof the method involving the intracellular injec-tion of antibodies specific for the introducedchromatin followed by gold decoration of theantibodies, thin sectioning and EM observationreveals clear-cut cylindrical fibers as can be seenin the stereo pair in Figure 3 (Kireev et al. 2008).The fibers ranged in diameter from 120 nm to170 nm with different mean values in two celltypes examined. Only rarely were individual30-nm-like fibers seen, and, as the authorsnote, it remains to be seen whether the thickfibers constitute one level in a hierarchical fold-ing series. Hopefully, further exploitation of thetechnique will lead to long-awaited progress inour understanding of chromatin organizationin situ.

Structural Order Beyond the 30-nm fiber—Light Microscopy Results

The ability to identify specific segments of chro-matin using light microscopy has providedimportant insights into interphase chromatinorganization. However, in evaluating the im-pressive body of information obtained withthe light microscope, it is important to bear inmind the preparative method used in eachcase, because this may significantly modulatethe outcome of an investigation, especiallywhen close to the resolution limit. Studies

with living cells are likely to provide the mostreliable data, whereas the use of fluorescencein situ hybridization (FISH) requires a denatu-ration step which may introduce structuralperturbations.

Muller et al. (2004) reported that �400 kbsegments of nontranscribed chromatin occupyroughly spherical volumes with a diameter notmuch greater than the resolution limit of themicroscope. A �400 kb length of DNA trans-lates into �2000 nucleosomes, and, if presentas a linear �30 nm fiber with 10 nucleosomesper 10 nm length would have a length of �2mm. Thus, in this case, an untranscribed regionof chromatin must be extremely tightly packed.However, an underlying fiber-like organizationbelow the resolution of the light microscopecannot be excluded (Tumbar et al. 1999). Itwill be very interesting to see if the higherresolution offered by subdiffraction light micro-scopy will reveal any internal structure in thesecompact regions. The situation changes dra-matically on transcription, when loci typicallyexpand, often becoming linear and adopting abeaded form. In an informative study, Shoplandet al. (2006) examined the intranuclear distri-bution of a 4 Mb segment of mouse chromo-some 14 that included four gene clustersseparated by gene “deserts.” By labeling thegene clusters and deserts with different coloredfluorescent probes, it was possible to trace theconformation of this large segment of chroma-tin (Fig. 4). Four patterns occurred in roughly

Figure 3. Stereo pair of a section of nucleus in which alarge region of chromosome is decorated with goldparticles. Clear fiber-like structures of the order of100 nm are seen. Scale is 500 nm. From Kireevaet al., 2008.






equal proportions, ranging from linear “striped”arrangements to more compact clusters. Interes-tingly, the inactive segments were often clusteredand oriented closer to the nuclear peripherythan the gene-rich segments, in keeping withthe preferential location of inactive chromatinin peripheral heterochromatin.

HETEROCHROMATIN AND EUCHROMATIN

The terms heterochromatin and euchromatinrefer to states of compaction and transcriptionalpotential rather than categories of chromatinhigher-order structure per se. Nevertheless, theyare used ubiquitously in the literature and merita brief mention. Heterochromatin was originallydefined well before the discovery of DNA asregions of nuclei that stained strongly with basicdyes (Heitz, 1928) and, together with its counter-part euchromatin, has provided a useful qualita-tive indication of chromatin compaction state.In general, heterochromatin tends to be locatedat the nuclear periphery, where specific interac-tions with the envelope may occur and oftenforms blocks surrounding the nucleolus. Thesedistinctions can be clearly seen in light andelectron micrographs. Transcription is largely

confined to euchromatin, and it is interestingto note that Heitz presciently suggested a func-tional difference between the two forms.

Today, the term heterochromatin is moreloosely applied, and is often extended to includetranscriptionally silent regions of chromatinregardless of their staining properties. Animportant distinction is made between consti-tutive and facultative heterochromatin. Consti-tutive heterochromatin is always compact, andtends to be enriched in repetitive, gene-poor,and late replicating DNA sequences, whereasfacultative heterochromatin can reversiblyundergo transitions from a compact, transcrip-tionally inactive state to become more open,and transcriptionally competent. In a recentreview, Trojer and Reinberg (2007) suggestedthat facultative heterochromatin be molecularlydefined as condensed, transcriptionally silentchromatin regions that decondense and allowtranscription within temporal, spatial, orparental/heritable contexts. During embryo-genesis, for example, the amount of facultativeheterochromatin increases as unwanted sets ofgenes are progressively shut down until atmaturity, a cell expresses only the genes appro-priate for that tissue. The reverse occurs when,

Figure 4. (A) A large locus consisting of �4-Mbp region containing regions of gene “deserts” (red fluorescence)and gene clusters (green fluorescence) is seen in the nucleus in multiple configurations (C). In general, genedeserts are more closely associated with the heterochromatin at the nuclear periphery (B). Scale is 1 mm.From Shopland et al. 2006.






for example, differentiated cells are reprog-rammed to become stem cells. These eventsare typically accompanied by profound changesin histone variants, histone modifications, andthe presence of CAPS. An important feature ofheterochromatin is its propensity to “spread”to adjacent regions of chromatin (Liaw and Lus-tig 2006; Hines et al. 2009).

A recent comparison of properties of hetero-chromatin and euchromatin as diffusion bar-riers has yielded interesting and provocativeresults (Bancaud et al. 2009). Measurementsof diffusion constants of large polymers withinthese nuclear compartments confirmed thatheterochromatin constitutes a more crowdedenvironment, leading to the more efficient trap-ping of chromatin binding proteins such as his-tone H1. The crowding effect was also suggestedto assist in the maintenance of heterochroma-tin. Further, kinetic analyses indicated ananomalous component of diffusion that wasinterpreted in terms of a fractal chromatinorganization at spatial scales below 100 nm.This indicates a physical arrangement moresimilar to the “molten globule” state suggestedby the lack of structure in cryo sections thanto the classic 30-nm fiber.

The ubiquitous location of heterochroma-tin at the nuclear periphery and associationwith the nuclear lamina and nuclear envelopesuggests that this location is both structurallyand functionally important. It was thereforesurprising to read that, in the rod photoreceptorcells of nocturnal, but not diurnal, mammals,heterochromatin is concentrated in the centerof nuclei (Solovei et al. 2009). The authors pos-tulate that with this arrangement, nuclei act ascollecting lenses, providing additional sensitiv-ity in very low light environments. This un-expected finding underscores the plasticity ofchromatin organization.

CHROMATIN LOOPS

The concept of large scale chromatin loops isimportant in many scenarios of chromatinfunction. The generic term “chromatin loop”is used for a variety of phenomena that playimportant roles in nuclear organization and

function, but otherwise may have little in com-mon. Further, it is not at all clear whether thevarious loop phenomena that have beenreported constitute distinct level(s) of chroma-tin higher-order structure (Kadauke and Blobel2009). It is clear, however, that loop phenomenavary in terms of their stability. For example,enhancer-promoter loops that facilitate tran-scription are transitory, dynamic events, where-as other types of loops appear to be more stable.An early observation that has stood the test oftime is the formation of a DNA “halo” aroundisolated nuclei exposed to a mild treatmentthat releases histones (Cook and Brazell 1975).The halos were shown to consist of supercoiledDNA, suggesting that the DNA twisting thatoccurs when histones are removed from nucle-osomes is preserved by being anchored in theresidual nuclear structure. This finding is alsoconsistent with the concept of an insoluble pro-teinaceous nuclear matrix or karyoskeleton towhich chromatin loops are anchored by DNAsequences referred to as matrix attachmentregions (MARs). However, because it has notbeen possible to define its composition andstructure, the matrix remains a useful workingconcept rather than a well-accepted structurein the same sense as the cytoskeleton (Pederson2000). A somewhat different perspective in-volves the “looping out” from chromosometerritories of large segments of genome thatare related in terms of transcriptional activity.

The supposition that these different loopphenomena reflect the same underlying chro-matin organization was called into question bya surprising recent finding that halo diameterwas related to the spacing of origins of replica-tion during the previous S phase (Courbetet al. 2008). Under conditions of rapid repli-cation, origins were widely spaced, leading tolarge halos, whereas slow replication triggeredthe firing of additional replication origins, andled to smaller halos. Halos can also be generatedfrom metaphase chromosomes, and giant mei-otic lampbrush chromosomes provide a parti-cularly compelling example of large chromatinloops and their relation to transcriptionallycoupled genes. Large-scale chromatin loopsare featured in some models of chromatin and






chromosome organization, but how they areintegrated into chromatin/nuclear structureand whether they constitute a distinct hierarch-ical level of chromatin organization remains tobe established.

The interaction between enhancers and pro-moters probably represents an unrelated mani-festation of chromatin looping that is critical fortranscriptional activity. Enhancer sequences areoften many Kb distant from promoters, andmay be located upstream, downstream, or on adifferent chromosome. There seems now to bea consensus that physical interaction betweenenhancer and promoter is necessary to initiatetranscription (Visel et al. 2009), but how widelyseparated loci are brought into contact hasremained an open question. One possibility,supported by recent data (Nolis et al. 2009) isthat sequence-specific transcription factorsbind to both enhancer and promoter, and alsobind strongly with each other. In support ofthis hypothesis, “decoy” transcription factorbinding sites placed between a promoter andits enhancer blocked transcription, suggestingthat they trapped the enhancer in unproductiveloops (Nolis et al. 2009).

Finally, the recently introduced technique ofchromosome conformation capture (3C) andrelated methods, which allow mapping of phys-ical chromatin interactions in vivo, is providinggrowing evidence for physical interactionsbetween distant loci other than enhancer-promoter juxtapositions (Gondor and Ohlssen2009). Hints that there must be mechanisms forbringing specific loci together has come fromthe common occurrence of some chromosomaltranslocations, especially those leading to humandiseases. For example, Roix et al. (2003) showedthat genes associated with chromosome translo-cations leading to human lymphomas tend tobe in physical proximity and located toward thenuclear interior. The data also suggested thatthe phenomenon was dependent on the “higher-order spatial organization” of the genome ratherthan the sequences of the genes involved. Humanprostate cancer offers a system for tracking thephysical proximity of two loci. The TMPRSSand ERG genes are�3MB apart on human chro-mosome 21, but become fused in �50% of

prostate cancer cases, apparently in response tohormone levels. Using FISH to follow the loca-tions of the two genes, Mani et al. (2009) showedthat, upon hormone treatment, the two genesbecame physically close in a significant propor-tion of cells. The mechanism(s) involved in thislarge-scale motion remain unknown. Anotherrecent study of intranuclear chromatin associa-tions capitalizing on massively-parallel sequenc-ing has yielded important insights into the largescale organization of chromatin (Lieberman-Aiden et al. 2009). Although the technique inits current form is limited to megabase scale res-olution, the data clearly indicated that active andinactive chromatin occupy different domainswithin chromosome territories. Further, at thislevel of resolution, chromatin conformation wasconsistent with a knot-free, fractal globule organ-ization. In the representation of their conclusions,Lieberman-Aiden et al. show modeled nucleiwith chromatin modeled as worm-like structuresof uniform diameter, and it is important toappreciate that these structures are not intendedto correspond to a specific level of chromatinhigher-order structure. It is interesting to notethat whereas this study indicated a fractal-likeorganization at very large scales, unrelated workon the accessibility of differentially packed chro-matin suggested a fractal conformation for chro-matin below 10 nm (Bancaud et al. 2009).Hopefully, these two sets of findings obtainedthrough widely divergent experimental app-roaches, augur a future in which chromatinorganization can be studied effectively at manydifferent levels. The study of chromatin longrange organization using 3C and related techni-ques (Rusk, 2009) is clearly at an early stage indevelopment, and promises to provide importantinsights into the mechanisms that bring distantloci together and their relationship to chromo-some territories and chromatin dynamics.

The Metaphase Chromosome

The metaphase chromosome, in which theDNA is compacted some 10,000- to 20,000-fold, is the one consistent manifestation of chro-matin higher-order structure. It is thereforesobering to reflect that despite a great deal of






research, there is still no widely accepted modelfor the internal organization of this critical cel-lular component. Andrew Belmont, who hascontributed seminal work in the field, has aptlyapplied the quotation “a riddle, wrapped in amystery, inside an enigma” to illustrate thecomplexities of metaphase chromosome struc-ture. There are several features that must beaccounted for in any viable model. For a givenspecies, the diameter of chromosome arms atmetaphase is quite constant, although thereare wide variations in chromosome diameterand number from species to species. The widthconstancy appears to be independent of DNAsequence. This is seen clearly in cases in whicha chromosome bears a large insert, either as aresponse of the cell to a genetic defect that isovercome by accumulating multiple tandemrepeats of the mutated gene (e.g., Sullivan andBickmore 2000), or by insertion of large tan-demly repeated loci. Typically, these inserts,which appear at metaphase as homogeneouslystaining regions (HSRs) on account of theirlack of band/interband organization, have auniform diameter that closely matches that ofthe rest of the chromosomes. This indicatesthat something other than DNA sequence isestablishing the underlying folding patternthat governs chromosome diameter.

It is well-established that, with the exceptionof polytene systems, chromosomes are unin-eme, in that a single strand of DNA, albeit mul-tiply folded, extends from one end (telomere) tothe other (reviewed in Gall 1981). Although fora particular chromosome, metaphase bandingpatterns and gene locations are consistentfrom cell to cell, the spatial resolution of lightmicroscopy provides very limited structuralinformation in the context of a chromosomearm that may have a diameter of �500 nm.There is, however, a great deal of informationrelevant to the overall architecture of metaphasechromosomes.

An early and striking result was derivedfrom treating chromosomes to remove histones,which, like similar treatments of nuclei, revealeda “halo” of DNA loops surrounding a densercore with a size and shape comparable to thestarting chromosome (Earnshaw and Laemmli

1983). The core structures were termed “scaf-folds” and it was postulated that they formedthe structural basis of chromosome architec-ture. DNase treatment could be used to isolatethe scaffolds, which contained specific DNAsequences (scaffold-attachment regions orSARS) and were enriched for a few proteinsthought to be essential for scaffold, and hencechromosome structural integrity. Avery fruitfulapproach to chromosome structure that allowsexperimental manipulation uses amphibianegg extracts, which can be manipulated to reca-pitulate, in sperm nuclei, the essential cell cycleevents of DNA replication, chromosome forma-tion and mitosis (Lohka and Masui 1983).Importantly, the system allows the roles ofindividual components to be explored by theiraddition to or removal from the extract. Sup-port for the scaffold hypothesis was providedby the observation that, in chromosomes for-med in vitro in egg extracts, some componentsisolated from scaffolds were distributed axiallyat metaphase rather than uniformly. However,subsequent experiments established that thesituation was more complex. For example,topoisomerase II could be removed from chro-mosomes assembled in vitro without compro-mising their structure (Hirano and Mitchison1993), but was essential for proper chromosomeassembly and separation of sister chromatids atanaphase. Further, genetic screens for chromo-some segregation mutants revealed a family of“structural maintenance of chromosomes”(SMC) proteins (Hirano 2006). Also identifiedwere two complexes, cohesin and condensinthat contain SMCs and other proteins and, inadditiontootherfunctions,appeartobeinvolvedin sister chromatin adhesion and chromosomecondensation (Hagstrom and Meyer 2003).Condensin is consistently seen to have an axialdistribution in chromosomes, and in its absenceor mutation, proper condensation fails.

A completely different approach, in whichmicromechanical properties of chromosomesare measured, has introduced a dramaticallydifferent aspect of their structure. Importantly,these results have had the salutatory effect ofrequiring fresh thinking to reconcile the seem-ingly disparate sets of data. The strategy is to






isolate chromosomes either from mitotic cellsor egg extracts, hold each end with micropip-ettes, pull on one end, and measure the forceneeded for a given length increase (Houch-mandzadeh et al. 1997; Poirier and Marko2003). The environment of the chromosomecan be altered by modulating the buffer withcations etc., and the effects of nucleases or pro-teases can be determined. The results show thatchromosomes exhibit a remarkable degree ofelasticity (Fig. 5), repeatedly returning to theoriginal length after being stretched �five-fold. The general conclusion from several stud-ies is that mitotic chromosomes consists of anetwork or gel in which individual chromatinfibers are connected by crosslinking elements,presumably proteins (Fig. 6). It was also possi-ble to estimate a mean distance between cross-links of �15 Kbp by applying restrictionenzymes with different cutting frequencies tochromosomes before stretching (Poirier andMarko, 2002). The estimated distance betweencrosslinks is close to that calculated for the chro-matin loops associated with chromosome scaf-folds (Gasser et al. 1986). One surprisingfeature of the work was the large difference inflexibility between “native” chromosomesextracted from cells by micromanipulationand those assembled in egg extracts in vitro.The latter were almost two orders of magnitudemore flexible, suggesting that their non-DNAcomponents were thin, rigid, and highly exten-sible (Houchmandzadeh et al. 1997)

In its simplest form, the network modelseems to be consistent with the lack of consis-tent large-scale structures seen in cryo-sectionsof HeLa cell chromosomes (Eltsov et al. 2008) orin EM tomograms of chromosomes formed inXenopus egg extracts (Konig et al. 2007), butat odds with hierarchical models of chromo-some structure, which would predict thatstretching a chromosome would result in theunfolding of hierarchical levels with stepwisereductions in chromosome diameter. Thenetwork model also seems to contradict themany light microscopic observations of an axiallocalization of key proteins, especially conden-sins known to be involved in chromosomecompaction.

Recently, however, models that account formost of the apparently disparate observationshave been put forward. An architecture basedon observations of early chromosome conden-sation that involves hierarchical folding withan axial “glue” provided by condensin (Fig. 7)has been proposed by Kireeva et al. (2004),and a more elaborate system with specific loca-tions and roles for topoisomerase II, cohesinand condensin has been put forward by Marko(2007). In the case of the model in Figure 7, adelicate balance between the forces maintainingthe 200–250 nm and 500–750 nm fibers would

Figure 5. Time sequence showing the extension of ametaphase chromosome in which the two ends areanchored by micropipettes. The diameter remainsquite constant throughout. From Marko 2008.

Linker protein

Mitotic chromosome

30 nmchromatin fiber

Figure 6. Extensible net model of mitoticchromosome structure derived from force-extensionmeasurements. From Poirier and Marko 2002.






be needed to produce the stretching effectshown in Figure 5. As noted by Belmont(2006), many questions remain unsolved, andthere are probably additional proteins, levelsof complexity, and redundant mechanismsto be discovered before the complete storyof chromosome architecture, formation andmaintenance is uncovered. Indeed, a new activ-ity, RCA (regulator of chromosome architec-ture), that can condense chromosomes inthe absence of condensins has recently beenproposed (Vagnarelli et al. 2006). To date, nosatisfactory explanation of the constant diame-ter of chromosomes has emerged. Thus, thesimple-appearing metaphase chromosome isfar from simple, and fully understanding itscomposition, formation, maintenance andarchitecture will likely remain a challenge forsome time.

LAMPBRUSH AND POLYTENECHROMOSOMES

These two special and quite rare manifestationsof chromatin provide some insight into the

surprisingly wide range of nuclear organiza-tions that “work.” Lampbrush chromosomes(LBCs) are largely confined to meiosis inoocytes, especially avian and amphibian, butare also present in the large unicellular alga Ace-tabularia. They appear to be an adaptation formaximizing transcriptional output to servethousands of cells in the developing embryoand the macroscopic cell body in Acetabularia.Light microscopy reveals an axial core fromwhich loops of varying size extend. Loops con-tain transcription units that can be seen in theelectron microscope to be loaded with RNApolymerases and continuously transcribing.Although the loops are the most conspicuousfeatures of LBCs, they represent only a smallfraction of the genome, the bulk forming thecompact chromatin of the axial cores. The factthat sperm nuclei exposed to extract fromoocytes at the lampbrush state develop intohaploid LBCs (Gall and Murphey 1998) indi-cates that the signals needed to transform achromosome into the lampbrush state evidentlyreside in the oocyte. A key question is whetherthis loop organization is unique to lampbrushchromosomes or whether it offers an especiallyclear example of a general organizing princi-ple for transcriptionally active chromatin.Although the growing evidence for the ubiqui-tous organization of chromatin into large scaleloops containing functionally related genes isconsistent with the latter interpretation, thereis also compelling evidence transcription of“normal” chromatin is organized quite differ-ently (Hu et al. 2009).

Polytene chromosomes are formed by re-peated mitoses without nuclear division andconsist of genomes precisely aligned in trans.They are especially prominent in dipteran larva,with 1024 copies of the genome being present inthe third instar of Drosophila. In the light micro-scope, a unique pattern of bands and interbandsis seen, allowing the locations of individualgenes to be mapped cytologically. It has longbeen recognized that at the sites of transcrip-tionally active genes, “puffs” appear as the chro-matin is locally decondensed. Disappointingly,polytene chromosomes have not yielded muchinsights into chromatin organization in general.

30 nmA B

100–130 nm

200–250 nm

500–750nm

Figure 7. Model of chromosome formation thatincorporates the concept of a central axis enrichedin condensins (red dots) and irregularly foldedchromatin fibers. From Kireeva et al. 2004.






In the electron microscope, their alignmentappears far from regular, and it is as difficultto recognize and follow individual chromatinfibers as in normal diploid nuclei. Polytenechromosomes may also be used to infer thestructural organization of interphase chroma-tin. For example, a component that, whendepleted or mutated, results in disorganizationof polytene chromosomes, may be involved insister chromatid adhesion. Polyteny is alsoseen in the first stages of the development ofthe nurse cells surrounding the Drosophilaovary. As nurse cells mature, the polytene align-ment breaks down, and the nuclei become pol-yploid. Surprisingly, in flies with mutations ofputative condensin 2 components, the polytenechromosomes fail to disperse at the appropriatetime, whereas the same mutations induce dis-persion of salivary gland chromosomes (Hartlet al. 2008). These findings may lead to furtherinsights into the complex events controllingpolytene chromosome integrity.

Both lampbrush and polytene chromo-somes dispel the idea that transcription is ob-ligatorily confined to “factories” that containRNA polymerases and the multiple other fac-tors necessary for RNA synthesis and processing(Cook 1999). Observation of lampbrush loopsclearly indicates that it is the polymerases thatmove along the loops rather than vice versa.Similarly, it seems very unlikely that DNA repli-cation in polytene chromosomes occurs atreplication factories through which DNA istranslated (Cook 1999). These considerationsdo not lessen the potential importance of tran-scription and replication factories in general,but again emphasize the remarkable plasticityof nuclear organization.

CHROMATIN DYNAMICS

The profound structural rearrangements thatchromatin must undergo during the cell cycleunderscores one aspect of the dynamic natureof chromatin. There is, in addition, compellingevidence of an important role for local struc-tural changes in compaction and/or positionof specific genetic loci likely to be of criticalimportance in their transcriptional regulation.

In fact, most of the functionally importantcharacteristics of chromatin show dynamic be-havior in the sense of time-dependent changes.These include the mobility of many chromatin-binding components, including chromatin ar-chitectural proteins (Phair and Misteli, 2000),and the status of posttranslational modifica-tions, especially of the core histones. Many ofthese phenomena are likely to impact localand global chromatin conformation and thusmodulate higher-order structure.

The study of chromatin dynamics in all of itsmanifestations has opened up exciting new per-spectives—indeed, a literature search revealsover 60 reviews with titles containing the terms“chromatin” and “dynamics” (e.g., Huebnerand Spector 2010). Here, we focus on aspects ofdynamics that are likely to impact higher-orderstructures. All involve changes in location and/or shape of specific genetic loci that are thoughtto have important functional implications.

As the field of chromatin dynamics hasdeveloped, it has become increasingly clearthat a number of underlying physical andmolecular phenomena are involved, and thatto understand them fully, information atmany different levels is required. For a givenlocus these may include the temporal and spa-tial scales of observed movements, their energydependence, and their location, both within thenucleus and within their specific chromosometerritory. One important goal of currentresearch is to be able to relate these changes inlocation/shape to the underlying physicalchanges in chromatin higher-order structure.For movements that are correlated with changesin transcriptional activity, it is also important todetermine whether a change in location is a pre-requisite for, or a consequence of, the alteredlevel of RNA synthesis.

Advances in light microscopy techniquesover the past �25 years have allowed chromatindynamics to be examined with steadily improv-ing temporal and spatial resolution, and hasenabled three principal types of dynamicchange to be recognized (Soutoglou and Misteli2007). These differ in the extent of motion, theirenergy dependence, and the time scales involved(Fig. 8). Early studies tracking changes in






location of large segments of chromatin andwhole chromosome territories indicated that,when corrected for nuclear rotation, the regionswere essentially immobile over distances .0.4mm (Cremer et al. 1982, Diboni and Mintz1986, Shelby et al. 1996). However, more recentwork combining higher spatial resolution within vivo labeling of smaller defined segments ofchromatin, revealed a more complex scenariowith different regions of chromatin showingvery different mobilities. For example, a seg-ment of budding yeast chromatin located neara centromere was found to show constrainedrandom walk diffusive motion with a confine-ment radius of �0.3 mm (Marshall et al.1997), that was independent of the metabolicstate of the cell. The rate of mobility of thisregion of chromatin was calculated to beapproximately three orders of magnitude lowerthan expected for free DNA of similar length(Marshall et al. 1997), leading to the suggestionthat the locus was tethered within the nucleus,perhaps to the nuclear envelope or some inter-nal structure. A detailed study of chromatin

mobility in Drosophila spermatocytes usingloci labeled with the lac repressor systemrecorded relatively large movements duringshort time intervals in early nuclei. Movementsof this magnitude would be expected to result indisplacements of over 4 mm in 1 h and cover theentire 11–12mm diameter of a nucleus in 6–7 h(Vazquez et al. 2001). However, when long timeperiods were examined, it was clear that lociwere constrained within a volume approxi-mately equivalent to a chromosome territory.Another important finding was that spermato-cytes in late G2 stage and approaching meiosisshowed greatly reduced chromatin mobility,suggesting that some form of additional tether-ing precedes chromosome condensation.

A comparison of the mobility of differentchromosomal sites in yeast also revealed strikinglocus- and cell cycle-dependent differences inchromatin dynamics (Heun et al. 2001). Somechromatin regions showed occasional large(�0.5 mm) movements over time periods asshort as 10 seconds that were inhibited in ATP-depleted cells and thus dependent on the

Short-range motion

-Locally constrainemd within ~1μm-Fast (seconds)-Random direction-ATP dependent-Frequent-Occurs by default

-Several micrometers-Slow (minutes)-Directed-ATP/motor dependent-Rare-In response to physiological cues (cell cycle)

Long-range motion

-Several micrometers-Slow (hours, days)-Global changes-Likely requires cell division-In response to physiological cues(differentiation, development)

Genome Reorganization

Figure 8. The different forms of large-scale motion that contribute to chromatin dynamics. From Soutoglou andMisteli 2007.






metabolic state of the cell. The fact that the mag-nitude of these large movements is similar fororganisms with widely different nuclear vol-umes has important structural and functionalimplications. For example, in budding yeast,which has a nuclear diameter of �2 mm, move-ment within a radius of �0.5 mm would allowaccess to a large portion of the nuclear volume.In contrast, in a typical mammalian nucleus of�10mm in diameter this motion would exploreonly a thousandth of the nuclear volume. Inyeast, the effectively large displacements of lociwithin nuclei may promote the observed highrecombination frequency, which may requiresubstantial intermingling among different re-gions of the genome (Gasser 2002).

Observations of “free” ectopic regions ofchromatin, exemplified by large (�15 kb) cir-cular plasmids in yeast underscore the contextdependence of chromatin motion. An earlystudy using a centromere-containing plasmidreported very limited movements, similar tothose observed for centromeric regions in intactchromosomes (Marshall et al. 1997). However,more recently, Gartenberg et al. (2004) usingplasmids designed to be either transcriptionallysilent or competent, found that the active plas-mids showed unconstrained movement, where-as the silent ones were strongly constrained andpreferentially located near the nuclear envelope.Also relevant to the factors controlling intranu-clear mobility is the finding that whereasdouble-stranded DNA breaks tend to staytogether through the repair process (Kruhlaket al. 2006), the absence of a critical repair factorleads to long-range movements of the brokenends (Downs et al. 2004).

A general pattern emerging from numerousstudies using different organisms is that all chro-matin loci show constrained Brownian motionwith rather similar diffusion constants, but awide range of confinement volumes. A strongrelationship between the confinement volume,and distance from the nuclear envelope hasalso been established, with loci closer to theenvelope being more constrained. The peripheryof the nucleus proximal to the nuclear envelopeis considered a transcriptionally repressive en-vironment, establishing a correlation between

transcriptional silencing and limited intranu-clear mobility (Marshall 2002). The underlyingmechanism for this correlated behavior is notclear, but there is evidence that the constrainedmotion results from physical tethering to thenuclear lamina (Heun et al. 2001; Chubb et al.2002; Gartenberg et al. 2004).

The higher spatial resolution (�20 nm) oftwo-photon microscopy, together with a tem-poral resolution of �30 msec has revealednew levels of complexity for constrained diffu-sive motion of chromatin (Levi et al. 2005). Inthese experiments, GFP-labeled repeats insertedin CHO cells showed periods of rapid con-strained diffusive motion alternating with occa-sional energy-dependent curvilinear leaps of�150 nm that lasted only 0.3–2.0 second. Theenergy-dependence of these leaps suggestedthat they may need the activity of chromatinremodeling complexes or other ATP-requiringchanges in chromatin organization that couldlead to events such as the decondensation ofcompact 30-nm fibers resulting in the rapidchanges in location. Another important out-come of this study pertinent to the extent ofphysical coupling between neighboring loci onthe same chromosome was the finding thatloci 1–2 mm apart in the nucleus moved inde-pendently (Levi et al. 2005). A similar conclu-sion was reached by Hu et al. (2009) whoshowed that for a �2 Mb region of chromatinthat appeared as a linear cluster of beads in thelight microscope, the individual beads movedindependently. These findings suggest thatneighboring loci are not structurally tiedtogether by, for example, being anchored to anunderlying nonchromatin structure. As itbecomes possible to study yet closer regions ofchromatin, there will be a point at which move-ment of the neighbors is correlated, and defin-ing this point will be very useful in being ableto infer the effective chromatin compactionlevel. Thus, although the accumulating dataclearly demonstrate that chromatin in the inter-phase nucleus undergoes constant dynamicreorganization through constrained diffusivemotion, the level(s) of chromatin higher-orderstructure that participate in this phenomenonremain to be determined.






Some hints of the mechanism whereby lociundergo directed movement have emergedfrom work on an inserted transgene that typi-cally resides in peripheral heterochromatin,but tends to relocate toward the center of thenucleus within 1–2 hour of transcriptionalup-regulation (Chuang et al. 2006). Impor-tantly, this movement was blocked by theexpression of a mutant nuclear myosin I andby a nonpolymerizable mutant of actin. Thisintriguing finding should be viewed in the con-text that several families of actin and actin-related proteins (ARPs) are found in thenucleus, many of which are components ofchromatin remodeling and histone modifica-tion complexes (Chen and Shen 2007). Also, ithas been shown that the BAF remodeling com-plex binds to the ends and branch points ofactin filaments in a PIP2-dependent manner(Rando et al. 2002). Unlike their cytoplasmiccounterparts, nuclear actins and ARPs tend tooccur as monomers rather than the filamentousstructures, and it is therefore curious that it wasan actin polymerization mutant that was defec-tive in long range movement (Chuang et al.2006). Determining the mechanism by whichactin/myosin influences the mobility and direc-tionality of loci is clearly an important futuregoal.

CHROMATIN DYNAMICS ANDTRANSCRIPTION

Studies of the spatial locations and transcrip-tional competence of loci with respect to theirchromosome territories has provided someimportant insights. For example, the ANT2gene located on the X chromosome is foundin the interior of the territory of the inactivehomolog, but peripherally on the active homo-log (Dietzel et al. 1999). In contrast, the relatedANT3 gene, which is located on the pseudo-autosomal region of the X chromosome wasperipherally located on the territory of bothactive and inactive homologs. This pattern ofinactive genes located inside chromosome terri-tories, and relocating to the periphery, perhapsby a “looping out” mechanism requiring a dra-matic relaxation of local chromatin structure,

has been widely observed (Chambeyron andBickmore 2005; Chambeyron et al. 2005; Volpiet al. 2000). Other studies have documentedlarge scale movements of loci undergoingchanges in transcriptional status as part of adevelopmental program. Loci were found torelocate to specific regions of the nucleus ratherthan locations within territories. An informa-tive example is provided by the nuclear reorga-nization that accompanies differentiation ofmouse lymphocytes (Brown et al. 1997, 1999).Here, lymphoid genes targeted for shutdownbecome relocated close to the heterochromaticclusters of pericentromeric chromatin charac-teristic of murine cells. Importantly, in thiscase, transcriptional repression appears to pre-cede movement, suggesting that relocation isnot a prerequisite for silencing. The relocationof genes destined for transcriptional up-regula-tion away from centromeric chromatin (Schub-eler et al. 2000) or the nuclear periphery (Kosaket al. 2002) has also been documented.

Although there is a consensus regarding thecorrelation of intranuclear and intraterritorylocation with transcription, the chromatin con-formational changes that accompany switchesin transcriptional status are still far from clear.On the one hand, it seems unlikely that thebulky transcription machinery could accessDNA and operate in compact chromatin, andin this respect, the concept of the chromatinlooping out from compact domains to facilitatetranscription is attractive. Indeed, transcrip-tional activity in LBC loops and polytene puffsis demonstrably unfolded, in some cases to thebeads-on-a-string primary structure. Althoughthere is general agreement that segments ofchromatin are more compact when transcrip-tionally active than when silenced (e.g., Mulleret al. 2001), the extent of this decondensationis unlikely to be as great as in LBC loops. Arecent study designed to examine the compac-tion state of active and inactive chromatin ininterphase nuclei has indicated quite smalltranscription-related changes in higher-orderstructure, implying that transcription can pro-ceed in quite compact chromatin (Hu et al.2009). Labeled �2 Mb chromatin domainsdesigned to support a level of transcriptional






activity typical of nuclear genes were insertedinto cells, and tracked over time. In the unin-duced state, the arrays assumed a linear, oftenbeaded conformation, with lengths indicativeof DNA packing ratios of over 1000:1. On trans-criptional induction, the arrays retained theirlinear, beaded conformation, and showed onlya modest 1.5- to threefold increase in length.No evidence was seen for large-scale loopingor unfolding to the primary beads-on-a-stringlevel of chromatin structure. Can transcriptionproceed in a compact structure that is over10-times more compact than a 30-nm fiber?In terms of access, several studies have shownthat large macromolecules can penetrate quitecompact chromatin (Verschure et al. 2003;Chen et al. 2005; Bancaud et al. 2009), perhapsas a function of the dynamic mobility of chro-matin. Also, in a simple in vitro system,conformational fluctuations allowed directaccess to the central nucleosome in a compacttrinucleosome (Poirier et al. 2009). It is alsopossible that the binding of transcription fac-tors could be followed by extensive, localunfolding, undetectable at the resolution ofthe light microscope.

The many facets of chromatin dynamicsspan orders of magnitude both in the spatialand temporal domains. Placing this complexityin the context of the level(s) of chromatinhigher-order structure that are modulated dur-ing changes in transcriptional status remains animportant challenge in understanding nuclearfunction. The difficulty is compounded by thecurrent uncertainty regarding chromatin orga-nization above the 30-nm fiber level. Neverthe-less, continuing improvements in the spatialresolution with which specific loci can beimaged should soon begin to narrow the pos-sible interpretations in terms of the level(s) ofchromatin structure that could be involved.For example, as it becomes possible to imagecloser and closer neighboring loci on the samechromatin and follow their individual dynam-ics, there should be a point at which coupledmotion will be observed. Once that point isreached, it will become possible to combineknowledge of the physical properties of arraysof nucleosomes with the observed dynamics

and decide whether, for example, the data areconsistent with a change from a compact to adecondensed 30-nm fiber. New strategies thatallow “super” resolution in light microscopywill soon be in common use and may openthe way to this type of correlative work.

CONCLUDING REMARKS

The past few years have witnessed remarkableprogress in our understanding of chromatinhigher-order structure, and this rapid trajectorycan be expected to continue. There are someareas of research that seem especially poised togenerate significant new insights in the nearfuture, and others that have been overlyneglected. Approaches that are likely to provideimportant new information in the near futureinclude the application of “super resolution”light microscopy, and further exploration ofthe role of RNAi and other small and longnuclear RNAs (for a review see Wilusz et al.2009).

A plethora of strategies for extending theresolution limitation of the light microscopewell beyond the “diffraction limit” is maturingand some have been implemented commer-cially (Huang et al. 2009). Techniques capableof generating high resolution 3D images ofthe nuclear interior using principles such asPALM and STED will be especially suitable forstudies of nuclei and chromatin. However, asexperience with electron microscopy shows,high resolution alone may not be informative,especially for compact chromatin. Progress inunderstanding the paths of putative 30-nmfibers in nuclei will require both high resolutionand a contrasting technique that highlights asingle stretch of DNA/chromatin. One applica-tion of an interference-based subdiffractiontechnique that was focused on the nuclear enve-lope showed the remarkable improvement thatcan be expected (Schermelleh et al. 2008), butlittle additional detail was seen in peripheralheterochromatin. It is also important that, asresolution improves further, preparative meth-ods that provide a level of preservation consis-tent with the spatial resolution of the imagingsystem. Such methods should be capable of






preserving membranous and cytoskeletal ele-ments, as judged by electron microscopy.

Since the discovery of RNAi and other smallnuclear RNAs, their importance in a host ofnuclear functions has been shown. Amongthese functions is a role in the formation ofthe constitutive heterochromatin that flankscentromeres (Verdel et al. 2004; Moazed et al.2006; Li et al. 2009). The process involves anRNAi-containing complex (RNA-inducedtranscriptional silencing, RITS) that binds tospecific DNA sequences via siRNA-directedbase pairing, and also to histone H3 methylatedat lysine 9, a heterochromatin-specific histonemodification. In combination, these processespromote the spreading of heterochromatin. Itseems very likely that, as the roles of smallnuclear RNAs become more fully understood,they will be found to impact other aspects ofchromatin higher-order structure. In addition,the identification and characterization of longnoncoding RNAs that are retained in thenucleus is beginning to emerge as an importantfield that will add to our understanding of geneexpression (see Wilusz et al. 2009 for a recentreview).

In the category of neglected areas that meritfurther study are the ionic composition ofnuclei, chromosomes, and chromatin, and theperhaps related issue of the impact of intra-nuclear molecular “crowding” (Hancock 2004a).Given the dramatic effects of changes in confor-mation that isolated chromatin undergoes inresponse to changes in the ionic environment,it is surprising that so little is known aboutthe intranuclear ionic environment. It is gener-ally assumed that “physiological” salt includes�100 mM monovalent ions, and a smaller con-centration (�1–2 mM) of divalent cations.Buffers for work with isolated nuclei and chro-matin in solution typically include Naþ or Kþ

for the monovalent cation, Mgþþ for the diva-lent cation, and Cl2 for the anion. It is also wellestablished that polyamines, especially spermi-dine (4þ) and spermine (8þ) are ubiquitousnuclear components with potent chromatincompaction properties. Very little is knownabout the intranuclear distributions and con-centrations of these charged species, making it

difficult to have confidence that in vitro experi-ments and modeling studies are based onmilieus that accurately reflect conditions inthe nucleus. An isolated report of the distribu-tion of cations in nuclei and isolated mitoticchromosomes using secondary ion mass spec-trometry (Strick et al. 2001) concluded thatduring the transition from interphase to mito-sis, Caþþ and Mgþþ concentrations increased�three- to four-fold, reaching �18 mM and�11 mM respectively, whereas isolated chro-mosomes contained almost twice theseamounts. Interestingly, Caþþ, but not Mgþþ

showed an axial enrichment in chromosomes.Of the monovalent ions, nuclear Naþ concen-trations were close to those measured in thecytosol, whereas Kþ was clearly enriched inboth interphase and mitosis. It is not clearwhat the effect would be of using these remark-ably high divalent ion levels in nuclear buffers,and it is unfortunate that the technique hasnot been further exploited using a wider rangeof ionic conditions.

An area of research that has received littleattention until recently, but may have a far-reaching impact on our understanding of chro-matin function and dynamics relates to thehighly crowded nuclear milieu (Hancock2004a; Richter et al. 2008; Miyoshi and Sigimoto2008; Bancaud et al. 2009). It is clear that theextremely high concentrations of macromole-cules (of the order of 100 mg/mL) that arepresent in the nucleus have important thermo-dynamic implications (Minton, 2001). Theseconcentrations cannot be effectively recapitu-lated in vitro, with the consequence that experi-ments in solution may not reflect the events thatoccur in nucleo. In one set of experiments, whenisolated nuclei were allowed to swell in low io-nic strength medium, nucleoli and PML bo-dies disassembled and nucleolar transcriptionwas dramatically reduced (Hancock 2004b).However, after adding inert macromolecules(polyethylene glycol or dextran) to 12% w/v,the compartments reassembled, and nucleolartranscription recovered. Interestingly, thisexperiment also showed that it was not the lackof ions per se that caused disassembly. A fullerunderstanding of these effects, as well as the






intranuclear levels and roles of ions is likely toexpand our understanding of chromatin andnuclear organization.

This snapshot in time of chromatin higher-order structure and dynamics reveals an excitingperiod in research, with rapid advances beingmade on many fronts. As the different inter-weaving threads begin to coalesce, we cananticipate a much more comprehensive under-standing of the structural and functional orga-nization of chromatin in the context of itsnuclear location. An overriding theme is thatchromatin and its higher-order structures fea-ture plasticity and dynamics rather thanrigidity. How this is possible in what must atsome level be a self-organizing system (Misteli2001) remains a key open question.

REFERENCES

Arya G, Schlick T. 2006. Role of histone tails in chromatinfolding revealed by a mesoscopic oligonucleosomemodel. Proc Natl Acad Sci 103: 16236–16241.

Bancaud A, Huet S, Daigle N, Mozziconacci J, Beaudouin J,Ellenberg J. 2009. Molecular crowding affects diffusionand binding of nuclear proteins in heterochromatinand reveals the fractal organization of chromatin.EMBO J doi:10.1038/emboj.2009.340.

Bednar J, Horowitz RA, Grigoryev SA, Carruthers LM,Hansen JC, Koster AJ, Woodcock CL. 1998. Nucleo-somes, linker DNA, and linker histone form a uniquestructural motif that directs the higher-order foldingand compaction of chromatin. Proc Natl Acad Sci 95:14173–14178.

Belmont AS. 2006. Mitotic chromosome structure and con-densation. Curr Opin Cell Biol 18: 632–638.

Belmont AS, Bruce K. 1994. Visualization of G1 chromo-somes: A folded, twisted, supercoiled chromonemamodel of interphase chromatid structure. J Cell Biol127: 287–302.

Belmont AS, Braunfeld MB, Sedat JW, Agard DA. 1989.Large-scale chromatin structural domains within mitoticand interphase chromosomes in vivo and in vitro. Chro-mosoma 98: 129–43.

Belmont AS, Li G, Sudlow G, Robinett C. 1999. Visualiza-tion of large-scale chromatin structure and dynamicsusing the lac operator/lac repressor reporter system.Methods Cell Biol 58: 203–222.

Brown KE, Baxter J, Graf D, Merkenschlager M, Fisher AG.1999. Dynamic repositioning of genes in the nucleus oflymphocytes preparing for cell division. Mol Cell 3:207–17.

Brown KE, Guest SS, Smale ST, Hahm K, Merkenschlager M,Fisher AG. 1997. Association of transcriptionally silentgenes with Ikaros complexes at centromeric heterochro-matin. Cell 91: 845–54.

Buchberger JR, Onishi M, Li G, Seebacher J, Rudner AD,Gygi SP, Moazed D. 2008. Sir3-nucleosome interactionsin spreading of silent chromatin in Saccharomyces cerevi-siae. Mol Cell Biol 28: 6903–6918.

Chambeyron S, Bickmore WA. 2004. Chromatin deconden-sation and nuclear reorganization of the HoxBlocus upon induction of transcription. Genes Dev 18:1119–1130.

Chambeyron S, Da Silva NR, Lawson KA, Bickmore WA.2005. Nuclear re-organisation of the Hoxb complex dur-ing mouse embryonic development. Development 132:2215–2223.

Chen M, Shen X. 2007. Nuclear actin and actin-related pro-teins in protein dynamics. Curr Opin Cell Biol 19:326–330.

Chen D, Dundr M, Wang C, Leung A, Lamond A, Misteli T,Huang S. 2005. Condensed mitotic chromatin is accessi-ble to transcription factors and chromatin structural pro-teins. J Cell Biol 168: 41–54.

Cremer T, Cremer C, Baumann H, Luedtke EK, Sperling K,Teuber V, Zorn C. 1982. Rabl’s model of the interphasechromosome arrangement tested in Chinese hamstercells by premature chromosome condensation andlaser-UV-microbeam experiments. Hum Genet 60:46–56.

Chuang CH, Carpenter AE, Fuchsova B, Johnson T, deLanerolle P, Belmont AS. 2006. Long-range directionalmovement of an interphase chromosome site. Curr Biol16: 825–31.

Chubb JR, Boyle S, Perry P, Bickmore WA. 2002. Chromatinmotion is constrained by association with nuclear com-partments in human cells. Curr Biol 12: 439–445.

Clapier CR, Cairns BR. 2009. The biology of chromatinremodeling complexes. Annu Rev Biochem 78: 273–304.

Clark DJ, Kimura T. 1990. Electrostatic mechanism of chro-matin folding. J Mol Biol 211: 883–96.

Cook PR. 1999. The organization of transcription and rep-lication. Science 284: 1790–1795.

Cook PR, Bazell IA. 1975. Supercoils in human DNA. J CellSci 19: 261–279.

Courbet S, Gay S, Arnoult N, Wronka G, Anglana M, BrisonO, Debatisse M. 2008. Replication fork movement setschromatin loop size and origin choice in mammaliancells. Nature 455: 557–560.

Cui Y, Bustamante C. 2000. Pulling a single chromatin fiberreveals the forces that maintain its higher-order structure.Proc Natl Acad Sci 97: 127–32.

Daban J-R. 2000. Physical constraints in the condensation ofeukaryotic chromosomes. Local concentration of DNAversus linear packing ratio in higher order chromatinstructures. Biochemistry 39: 3861–3866.

Daban J-R. 2003. High concentration of DNA in condensedchromatin. Biochem Cell Biol 81: 91–99

DiBoni U, Mintz AH. 1986. Curvilinear, 3-dimensionalmotion of chromatin domains and nucleoli in neuronalinterphase nuclei. Science 234: 863–866.

Dietzel S, Schiebel K, Little G, Edelmann P, Rappold GA, EilsR, Cremer C, Cremer T. 1999. The 3D positioning ofANT2 and ANT3 genes within female X chromosometerritories correlates with gene activity. Exp Cell Res252: 363–375.






Dorigo B, Schlach T, Kulangara A, Duda S, Schroeder RR,Richmond TJ. 2004. Nucleosome arrays reveal thetwo-start organization of the chromatin fiber. Science306: 1571–1573.

Downs JA, Jackson SP. 2004. A means to a DNA end: Themany roles of Ku. Nature Rev Mol Cell Biol 5: 367–378.

Earnshaw WC, Laemmli UK. 1983. Architecture of meta-phase chromosomes and chromosome scaffolds. J CellBiol 96: 84–93.

Eltsov M, Maclellan KM, Maeshima K, Frangakis AS, Dubo-chet J. 2008. Analysis of cryo-electron microscopy imagesdoes not support the existence of 30-nm chromatin fibersin mitotic chromosomes in situ. Proc Natl Acad Sci 105:19732–19737.

Fan Y, Nikitina T, Morin-Kensicki EM, Zhao J, MagnusonTR, Woodcock CL, Skoultchi AI. 2003. H1 linker histonesare essential for mouse development and affect nucleo-some spacing in vivo. Mol Cell Biol 23: 4559–72.

Fan JY, Rangasamy D, Luger K, Tremethick DJ. 2004. H2A.Zalters the nucleosome surface to promote HP1alpha-mediated chromatin fiber folding. Mol Cell 16: 655–61.

Fischle W, Wang Y, Jacobs SA, Kim Y, Allis CD, Khourasani-zadeh S. 2003. Molecular basis for the discriminationof repressive methyl-lysine marks in histone H3 byPolycomb and HP1 chromodomains. Genes Dev 17:1870–1881.

Gall JG. 1981. Chromosome structure and the C-value para-dox. J Cell Biol 91: 3s–14s.

Gall J, Murphy C. 1998. Pictures in cell biology. Lampbrushchromosomes from sperm chromatin. Trends Cell Biol 8:207.

Gartenberg MR, Neumann FR, Laroche T, Blaszczyk M,Gasser SM. 2004. Sir-mediated repression can occurindependently of chromosomal and subnuclear contexts.Cell 119: 955–67.

Gaspar-Maia A, Alajem A, Polesso F, Sridharan R, MasonMJ, Heidersbach A, Ramalho-Santos J, McManus MT,Plath K, Meshorer E, et al. 2009. Chd1 regulates openchromatin and pluripotency of embryonic stem cells.Nature 460: 863–8.

Gasser SM, 2002, Visualizing chromatin dynamics in inter-phase nuclei. Science 296: 1412–1416.

Gasser SM, Laroche T, Falquet J, Boy de la Tour E, LaemmliUK. 1986. Metaphase chromosome structure. Involve-ment of topoisomerase II. J Mol Biol 188: 613–29.

Gondor A, Ohlsson R. 2009. Chromosome cross-talk inthree dimensions. Nature 461: 199–205.

Greaves IK, Rangasamy D, Ridgway P, Tremethick DJ. 2007.H2A.Z contributes to the unique 3D structure of the cen-tromere. Proc Natl Acad Sci 104: 525–530.

Grigoryev SA, Arya G, Correll S, Woodcock CL, Schlick T.2009. Evidence for heteromorphic chromatin fibersfrom analysis of nucleosome interactions. Proc NatlAcad Sci 106: 13317–22.

Hagstrom KA, Meyer BJ. 2003. Condensin and cohesin:More than chromosome compactor and glue. Nat RevGenet 4: 520–534.

Hake SB, Allis CD. 2006. Histone H3 variants and theirpotential role in indexing mammalian genomes: The“H3 barcode hypothesis.” Proc Natl Acad Sci 103:6428–35.

Hancock R. 2004a. A role for molecular crowding effects inthe assembly and function of compartments in thenucleus. J Struct Biol 146: 281–290.

Hancock R. 2004b. Internal organisation of the nucleus:Assembly of compartments by macromolecular crowdingand the nuclear matrix model. Biol Cell 96: 595–601.

Hartl TA, Smith HF, Bosco G. 2008. Chromosome align-ment and transvection are antagonized by condensin II.Science. 322: 1384–7.

Heitz E. 1928. Das Heterochromatin der Moose, 1, JahrbWiss Bot 69: 762–818.

Heun P, Laroche T, Shimada K, Furrer P, Gasser SM. 2001.Chromosome dynamics in the yeast interphase nucleus.Science 294: 2181–2186.

Hines KA, Cryderman DE, Flannery KM, Yang H, VitaliniMW, Hazelrigg T, Mizzen CA, Wallrath L. 2009. Domainsof HP1 Required for Drosophila melanogaster hetero-chromatin spreading. Genetics. 182: 967–77.

Hirano T. 2006. At the heart of the chromosome: SMC pro-teins in action. Nat Rev Mol Cell Biol 7: 311–322.

Hirano T, Mitchison. 1993. Topoisomerase II does not play ascaffolding role in the organization of mitotic chromo-somes assembled in Xenopus egg extracts. J Cell Biol120: 601–612.

Horowitz RA, Agard DA, Sedat JM, Woodcock CL. 1994.The three dimensional archeitecture of chromatin insitu: Electron tomography reveals fibers composed of acontinuously variable zigzag nucleosomal ribbon. J CellBiol 125: 1–10.

Houchmandzadeh B, Marko JF, Chatenay D, Libchaber A.1997. Elasticity and structure of eukaryote chromosomesstudied by micromanipulation and micropipette aspira-tion. J Cell Biol 139: 1–12.

Hu Y, Kireev I, Plutz M, Ashourian N, Belmont AS. 2009.Large-scale chromatin structure of inducible genes: Tran-scription on a condensed, linear template. J Cell Biol 185:87–100.

Huang B, Bates M, Zhuang X. 2009. Super-resolution fluo-rescence microscopy. Annu Rev Biochem 78: 993–1016.

Huebner MR, Spector DL. 2010. Chromatin dynamics. InPress.

Jacobs SA, Khorasanizadeh S. 2002. Structure of the HP1chromodomain bound to a lysine 9-methylated histoneH3 tail. Science 295: 2080–2083.

Jenuwein T, Allis CD. 2001. Translating the histone code.Science 293: 1074–80.

Jiang C, Pugh BF. 2009. Nucleosome positioning and generegulation: Advances through genomics. Nat Rev Genet10: 161–72.

Kadauke S, Blobel GA. 2009. Chromatin loops in gene reg-ulation. Biochim Biophys Acta 1789: 17–25.

Kireev I, Lakonishok M, Liu W, Joshi VN, Powell R, BelmontAS. 2008. In vivo immunogold labeling confirms large-scale chromatin folding motifs. Nat Methods 5: 311–313.

Kireeva N, Lakonishok M, Kireev I, Hirano T, Belmont AS.2004. Visualization of early chromosome condensation:A hierarchical folding, axial glue model of chromosomestructure. J Cell Biol 166: 775–785.

Konig P, Braunfeld MB, Sedat JW, Agard DA. 2007. Thethree-dimensional structure of in vitro reconstituted






Xenopus laevis chromosomes by EM tomography. Chro-mosoma 116: 349–372.

Kosak ST, Skok JA, Medina KL, Riblet R, Le Beau MM,Fisher AG, Singh H. 2002. Subnuclear compartmentali-zation of immunoglobulin loci during lymphocyte devel-opment. Science 296: 158–62.

Kouzarides T. 2007. Chromatin modifications and theirfunction. Cell 128: 693–705.

Kruhlak MJ, Celeste A, Nussenzweig A. 2006. Spatio-temporal dynamics of chromatin containing DNAbreaks. Cell Cycle 5: 1910–1912.

Kruithof M, Chien FT, Routh A, Logie C, Rhodes D, vanNoort J. 2009. Single-molecule force spectroscopy revealsa highly compliant helical folding for the 30-nm chroma-tin fiber. Nat Struct Mol Biol 16: 534–40.

Li H, Motamedi MR, Yip C K, Wang Z, Waltz T, Patel DJ,Moazed D. 2009. An Alpha Motif at Tas3 C TerminusMediates RITS cis Spreading and Promotes Heterochro-matic Gene Silencing. Mol Cell 34: 155–167.

Lieberman-Aiden E, van Berkum NL, Williams L, ImakaevM, Ragoczy T, Telling A, Amit I, Lajoie BR, Sabo PJ,Dorschner MO, et al. 2009. Comprehensive mapping oflong-range interactions reveals folding principles of thehuman genome. Science 326: 289–93.

Levi V, Ruan Q, Plutz M, Belmont AS, Gratton E. 2005,Chromatin dynamics in interphase cells revealed bytracking in a two-photon excitation microscope. BiophysJ 89: 4275–85.

Liaw H, Lustig AJ. 2006. Sir3 C-terminal domain involve-ment in the initiation and spreading of heterochromatin.Mol Cell Biol 26: 7616–7631.

Lohka MJ, Masui Y. 1983. Formation in vitro of sperm pro-nuclei and mitotic chromosomes induced by amphibianooplasmic components. Science 220: 719–721.

Lu X, Wontakal SN, Emelyanov AV, Morcillo P, Konev AY,Fyodorov DV, Skoultchi AI. 2009. Linker histone H1 isessential for Drosophila development, the establishmentof pericentric heterochromatin, and a normal polytenechromosome structure. Genes Dev 23: 452–65.

Luger K, Hansen JC. 2005. Nucleosome and chromatin fiberdynamics. Curr Opin Struct Biol 15: 188–196.

Luger K, Mader AW, Richmond RK, Sargent DF, RichmondTJ. 1997. Crystal structure of the nucleosome core par-ticle at 2.8 A resolution. Nature 389: 251–60.

Mani RS, Tomlins SA, Callahan K, Ghosh A, Nyati MK, Var-ambally S, Palanisamy N, Chinnaiyan AM. 2009. Inducedchromosomal proximity and gene fusions in prostatecancer. Science 326: 1230.

Marko JF. 2008. Micromechanical studies of mitotic chro-mosomes. Chromosome Res 16: 469–497.

Marko M, Hsieh C, Salmon N, Rodriguez M, Frank J, Man-ella C. 2008. FIB milling of frozen hydrated cells and tis-sues for TEM cryo-tomography. Microsc Microanal 13:968CD.

Marko M, Hsieh C, Schalek R, Frank J, Mannella C. 2007.Focused-ion-beam thinning of frozen-hydrated biologi-cal specimens for cryo-electron microscopy. Nat Methods4: 215–7.

Maresca TJ, Heald R. 2006. The long and the short of it:Linker histone H1 is required for metaphase chromo-some compaction. Cell Cycle 5: 589–91.

Marshall WF. 2002. Order and disorder in the cell nucleus.Curr Biol 12: R185–R192.

Marshall WF, Straight A, Marko JF, Swedlow J, Dernberg A,Belmont A, Murray AW, Agard DA, Sedat J. 1997. Inter-phase chromosomes undergo constrained diffusionalmotion in living cells. Curr Biol 7: 930–939.

Minton AP. 2001. The influence of macromolecular crowd-ing and macromolecular confinement on biologicalreactions in physiological media. J Biol Chem 276:10577–10580.

Misteli T. 2001. The concept of self-organization in cellulararchitecture. J Cell Biol 155: 181–185.

Miyoshi D, Sigimoto N. 2008. Molecular crowding effectson structure and stability of DNA. Biochimie 90:1040–1051.

Moazed D, Buhler M, Buker SM, Colmenares SU, Gerace EL,Gerber SA, Hong EJ, Motamedi MR, Verdel A, Villen J,Gygi SP. 2006. Studies on the mechanism of RNAi-dependent heterochromatin assembly. Cold Spring HarbSymp Quant Biol 71: 461–471.

Muller WG, Rieder D, Kreth G, Cremer C, Trajanoski Z,McNally JG. 2004. Generic features of tertiary chromatinstructure as detected in natural chromosomes. Mol CellBiol 24: 9359–9370.

Muller WG, Walker D, Hager G, McNally J. 2001. Large-scalechromatin decondensation and recondensation regulatedby transcription from a natural promoter. J Cell Biol 154:33–48.

Nolis IK, McKay DJ, Mantouvalou E, Lomvaradas S, MerikaM, Thanos D. 2009. Transcription factors mediate long-range enhancer-promoter interactions. Proc Natl Sci106: 20222–7.

Pederson T. 2000. Half a century of “the nuclear matrix.” MolBiol Cell 11: 799–805.

Phair RD, Misteli T. 2000. High mobility of proteins in themammalian nucleus. Nature 404: 604–609.

Poirier MG, Marko JF. 2002. Mitotic chromosomes are chro-matin networks without a mechanically contiguous pro-tein scaffold. Proc Natl Acad Sci 99: 15393–15397.

Poirier MG, Marko JF. 2003. Micromechanical studies ofmitotic chromosomes. Curr Top Dev Biol 55: 75–141.

Poirier MG, Oh E, Tims HS, Widom J. 2009. Dynamics andfunction of compact nucleosome arrays. Nat Struct MolBiol 16: 938–44.

Rando OJ, Zhao K, Janmey P, Crabtree GR. 2002.Phosphatidylinositol-dependent actin filament bindingby the SWI/SNF-like BAF chromatin remodeling com-plex. Proc Natl Acad Sci 99: 2824–2829.

Richter K, Nessling M, Lichter P. 2008. Macromolecularcrowding and its potential impact on nuclear function.Biochim Biophys Acta 1783: 2100–2107.

Robinson PJ, Fairall L, Huynh VA, Rhodes D. 2006. EMmeasurements define the dimensions of the “30-nm”chromatin fiber: Evidence for a compact, interdigitatedstructure. Proc Natl Acad Sci 103: 6506–6511.

Roix JJ, McQueen PG, Munson PJ, Parada LA, Misteli T.2003. Spatial proximity of translocation-prone gene lociin human lymphomas. Nat Genet 34: 287–291.

Routh A, Sandin S, Rhodes D. 2008. Nucleosome repeatlength and linker histone stoichiometry determine chro-matin fiber structure. Proc Natl Acad Sci 105: 8872–8877.






Rusk N. 2009. When ChIA PETs meet Hi-C. Nature Methods6: 863.

Santenard A, Torres-Padilla ME. 2009. Epigenetic reprog-ramming in mammalian reproduction: Contributionfrom histone variants. Epigenetics 4: 80–4.

Schermelleh L, Carlton PM, Haase S, Shao L, Winoto L,Kner P, Burke B, Cardoso MC, Agard DA, GustafssonMG, et al. 2008. Subdiffraction multicolor imaging ofthe nuclear periphery with 3D structured illuminationmicroscopy. Science 320: 1332–1336.

Schlach T, Duda S, Sargent DF, Richmond TJ. 2005. X-raystructure of a tetranucleosome and its implications forthe chromatin fibre. Nature 436: 138–141.

Schubeler D, Francastel C, Cimbora DM, Reik A, Martin DI,Groudine M. 2000. Nuclear localization and histone ace-tylation: A pathway for chromatin opening and transcrip-tional activation of the human beta-globin locus. GenesDev 14: 940–50.

Shelby RD, Hahn KM, Sullivan KF. 1996. Dynamic elasticbehavior of alpha satellite domains visualized in livingcells. J Cell Biol 135: 545–557.

Simpson RT. 1991. Nucleosome positioning: Occurrence,mechanisms, and functional consequences. Prog NucleicAcid Res Mol Biol 40: 143–84.

Shogren-Knaak M, Ishii H, Sun JM, Pazin MJ, Davie JR,Peterson CL. 2006. Histone H4-K16 acetylation controlschromatin structure and protein interactions. Science311: 844–847.

Shopland LS, Lynch CR, Peterson KA, Thornton K, KepperN, Hase J, Stein S, Vincent S, Molloy KR, Kreth G, et al.2006. Folding and organization of a contiguous chromo-some region according to the gene distribution pattern inprimary genomic sequence. J Cell Biol 174: 27–38.

Simpson RT, Thoma F, Brubaker J. 1985. Chromatin recon-stituted from tandemly repeated cloned DNA fragmentsand core histones: A model system for study of higherorder structure. Cell 42: 799–808.

Solovei I, Kreysing M, Lanctot C, Kosem S, Peichl L, CremerT, Guck J, Joffe B. 2009. Nuclear architecture of rod pho-toreceptor cells adapts to vision in mammalian evolution.Cell. 137: 356–68.

Soutoglou E, Misteli T. 2007. Mobility and immobility ofchromatin in transcription and genome stability. CurrOpin Genet Dev 17: 435–4.

Strick R, Strissel PL, Gavrilov K, Levi-Setti R. 2001.Cation-chromatin binding as shown by ion microscopyis essential for the structural integrity of chromosomes.J Cell Biol 155: 899–910.

Strukov YG, Belmont AS. 2009. Mitotic chromosome struc-ture: Reproducibility of folding and symmetry betweensister chromatids. Biophys J 96: 1617–28.

Sullivan BA, Bickmore WA. 2000. Unusual chromosomearchitecture and behaviour at an HSR. Chromosoma109: 181–9.

Suto RK, Clarkson MJ, Tremethick DJ, Luger K. 2000. Crys-tal structure of a nucleosome core particle containing thehistone variant H2A.Z. Nature Struct Biol 7: 1121–1124.

Tremethick DJ. 2007. Higher-order structures of chromatin:The elusive 30 nm fiber. Cell 128: 651–654.

Trojer P, Reinberg D. 2007. Facultative heterochromatin: Isthere a molecular signature. Mol Cell 28: 1–12.

Tumbar T, Sudlow G, Belmont AS. 1999. Large-scale chro-matin unfolding and remodeling induced by VP16 acidicactivation domain. J Cell Biol 145: 1341–54.

Vagnarelli P, Hudson DF, Ribeiro SA, Trinkle-Mulcahy L,Spence JM, Lai F, Farr CJ, Lamond AI, Earnshaw WC.2006. Condensin and Repo-Man-PP1 co-operate in theregulation of chromosome architecture during mitosis.Nat Cell Biol 8: 1133–1142.

van Holde K. 1989. Chromatin. Springer Verlag, New York.

van Holde K, Zlatanova J. 1995. Chromatin structure: Chas-ing a mirage? J Biol Chem 270: 8373–8376.

van Holde K, Zlatanova J. 2007. Chromatin fiber structure:Where is the problem now? Sem Cell Dev Biol 18: 651–658.

Vazquez J, Bemont AS, Sedat JW. 2001. Multiple regimes ofconstrained chromosome motion are regulated in theinterphase Drosophila nucleus. Curr Biol 11: 1227–1239.

Verdel A, Jia S, Gerber S, Sugiyama T, Gygi S, Grewal SI,Moazed D. 2004. RNAi-mediated targeting of hetero-chromatin by the RITS complex. Science 303: 672–676.

Verschure PJ, van der Kraan I, Manders EM, Hoogstraten D,Houtsmuller AB, van Driel R. 2003. Condensed chroma-tin domains in the mammalian nucleus are accessible tolarge macromolecules. EMBO Rep 4: 861–866.

Visel A, Rubin EM, Pennacchio LA. 2009. Genomic views ofdistant-acting enhancers. Nature 461: 199–205.

Volpi EV, Chevret E, Jones T, Vatcheva R, Williamson J, BeckS, Campbell RD, Goldsworthy M, Powis SH, Ragoussis J,et al. 2000. Large-scale chromatin organization of themajor histocompatibility complex and other regions ofhuman chromosome 6 and its response to interferon ininterphase nuclei. J Cell Sci 113: 1565–2576.

Wilusz JE, Sunwoo H, Spector DL. 2009. Long non-codingRNAs: Functional surprises from the RNA world. GenesDev 23: 1494–1504.

Woodcock CL. 2006. Chromatin architecture. Curr OpinStruct Biol 16: 213–220.

Woodcock CL, Dimitrov S. 2001. Higher-order structure ofchromatin and chromosomes. Curr Opin Genet Dev 11:130–135.

Woodcock CL, Horowitz RA. 1995. Chromatin organizationre-viewed. Trends Cell Biol 5: 272–277.

Woodcock CL, Skoultchi AI, Fan Y. 2006. Role of linkerhistone in chromatin structure and function: H1 stoichi-ometry and nucleosome repeat length. Chromosome Res14: 17–25.






Chromatin Higher-order Structure and Dynamics€¦ · on chromatin organization has increasingly...

Documents

Transcript of Chromatin Higher-order Structure and Dynamics€¦ · on chromatin organization has increasingly...