Association of nucleus and centrosome: magnet or velcro?

Cell Biology International ISSN 1065-6995doi: 10.1002/cbin.10016

REVIEW

Association of nucleus and centrosome: magnet or velcro?Anton V. Burakov1 and Elena S. Nadezhdina1,2,*

1 A.N.Belozersky Institute of Physico-Chemical Biology of Lomonosov Moscow State University, Vorobjevy Gory, 1/40, Moscow 119992, Russia2 Institute of Protein Research of Russian Academy of Science, Vavilova ul., 34, Moscow 119333, Russia

Abstract

A structural link between cell's nucleus and centrosome was proposed years ago. Such a link was suggested to maintain nucleus–centrosome axis, determine polarity of interphase cells and ensure spindle assembly in mitotic cells. The idea of structural link issupported by the facts that centrosomes are usually located in close proximity to the nuclei and remain attached to the nuclei inmildly homogenated cells. However, juxtaposed location can result rather from the tendency of both organelles to occupycentral position in cell than from the existence of a specific structural link. Moreover, the nucleus was shown to be transportedtowards the centrosome along microtubules by dynein bound to nuclear envelope; inhibition of dynein results in the increase ofnucleus–centrosome distance. The interaction of both organelles is disturbed in response to actin depolymerisation, althoughthe exact role of actin filaments in this process remains unknown. The link between the nucleus and the centrosome can supportsimultaneous migration of nuclei and centrosomes in large cells and in syncytia, but its existence in interphase fibroblast-likeand epithelia-like cells was not confirmed yet. Further studies include direct visualisation of a specific link between centrosomeand nucleus and elucidation of actin role in its formation.

Keywords: centrosome; dynein; LINC; microtubule; migration; nucleus

Introduction

The concept of structural link between the centrosome andthe nucleus was facilitated by several experimental facts. First,in living cell the centrosome is usually found in closeproximity to the nucleus, and these these organelles canmovecoordinately. For example, at the end of the syncytial stage inDrosophila embryo (Archambault and Pinson, 2010) or duringeye formation in Drosophila (Mosley-Bishop et al., 1999;Patterson et al., 2004; Kracklauer et al., 2007), the centro-somes followed by nuclei migrate to the apical part of the cell.Uncoupling of centrosomal and nuclear migration occurs inmutated animals and correlates with embryo failure or witheye defects. Second, the samples of isolated nuclei oftencontain centrosomes, especially after mild homogenisationof cells (Nadezhdina et al., 1979). Third, during lateinterphase–prophase, daughter and mother centrosomesseparate from each other and migrate along the nuclearenvelope. Until nuclear envelope breakdown, the centro-somes are always closely associated with the nuclear envelopeeven they reach opposite sides of the nucleus (Gönczyet al., 1999). Nuclei, artificially displaced from prophase

asters, attract to them after cessation of treatment (Aronson,1971).

Multiple vitally important functions can be suggested fornuclear-centrosomal connection, yet almost none of themhave had experimental confirmation. Thus, close location ofthe centrosome to the nucleus can potentially facilitatecentrosome connection to the chromosomes after mitoticnuclear envelope breakdown and bipolar spindle formation.In interphase, the centrosome–nucleus axis might determineeither basal-apical or planar polarity axis of the cell (Luxtonand Gundersen, 2011). Hypothetically, the contact with thecentrosome might be a kind of reference point for non-random chromosome location in the interphase nucleus.Indeed, centrosome-derived microtubules can influencenuclear shape and chromatin arrangement (Hampoelzet al., 2011). Finally, the nuclear-centrosomal link mightcontribute to nucleus migration, assuming that the centro-some is a cytoskeletal structure connected to a number ofmicrotubules that can move independently and function as alocomotive, which carries the attached nucleus (Raff andGlover, 1989). This function may be fulfilled during flydevelopment as discussed above.

*Corresponding author: e-mail: [email protected]

Cell Biol Int 37 (2013) 95–104 � 2013 International Federation for Cell Biology 95

Several ways of association between the centrosome andthe interphase nucleus could be suggested:

(1) There is no specific physical link there, both thecentrosome and the nucleus are located in the centreof the cell or move simultaneously driven by similarforces.

(2) The centrosome is linked to the nucleus by microtubulesand microtubule motors in a magnet-like way.

(3) There are special fibrils connecting the centrosome andthe nucleus, probably, acting like velcro.

No specific mechanistic link, just co-localisation?

What if the nuclear-centrosomal link does not exist andcentral location or simultaneous movement of the twoorganelles results from the similarity of transport mecha-nisms. Indeed, there are a number of mechanisms ofcentrosome positioning in the central region of mammaliancultured cells (Burakov et al., 2003; Brodsky et al., 2007; Zhuet al., 2010; Kimura and Kimura, 2011; Maly, 2011). All ofthem consider the centrosome as an intrinsic part of thecentrosome–microtubule system. The centrosome position-ing is determined by ‘pushing’ forces, developed by both acto-myosin centripetal flow and by microtubules polymerizing atthe plus ends distal of the centrosome, as well as by strong‘pulling’ forces, applied to microtubules extending from thecentrosome and developed by dynein attached to variousorganelles and to plasmalemma (Zhu et al., 2010; Kimura andKimura, 2011). Centrosome centering in Caenorhabditiselegans blastomeres also depends on microtubules and isprobably carried out similarly to mammalian cultured cells(Hyman and White, 1987). After partial inhibition of dyneinthe centrosome is displaced from cell centre and irregularlymoves in cytoplasm driven by both dynein and kinesin(Koonce et al., 1999; Brito et al., 2005). In the absence ofmicrotubules, the centrosome centering forces are notworking (Brodsky et al., 2007).

Usually the nucleus is also positioned in the centre of cells.In most cases, the distance between the centrosome and thenucleus in cells is 1–2 mm. The forces that place the nucleusin the cell centre have been studied worse than the forcescentering the centrosome and the forces moving nuclei to cellperiphery (Dupin and Etienne-Manneville, 2011). To studycentering forces, the nucleus can be artificially displaced, forexample by centrifugation of living cells (Aronson, 1971; Faiset al., 1984). When the cells are restoring after centrifugation,nuclei return to the centre of the cell (Krokhina andNadezhdina, 1991). Acto-myosin is probably involved innucleus positioning, and a specific actin cap can exist aroundthe nucleus (Hale et al., 2008; Khatau et al., 2010). In largecells (oocytes) and in the syncytia, the nuclei can be anchoredto the cytoskeleton, mostly to actin, can move and switch

from movement to the anchoring (Starr and Han, 2003).Destruction of actin filaments by cytochalasin B slows theprocess of nucleus re-positioning in the cell centre afterartificial displacement, but the destruction of microtubuleseven accelerates re-positioning (Krokhina and Nadezhdina,1991). The movement of nuclei in migrating neuronspartially depends on actin-myosin contraction in their rear(Martini and Valdeolmillos, 2010). Nucleus migration infibroblasts is usually governed by retrograde actin-myosinflow and is regulated by Cdc42 and its effector MRCK kinase.Being crucial for centrosome movement, dynein is notinvolved in nuclear migration (Gundersen et al., 2005).

In fibroblasts on 2D substrate, the nucleus–centrosomeaxis typically coincides with the longitudinal axis of thepolarised cell. At the same time, the movements ofcentrosome and nucleus in migrating cells are ratherindependent: in some cells the stationary nucleus can staybehind the moving centrosome, and vice versa in other cells(Gundersen et al., 2005; Luxton and Gundersen, 2011). Thus,the mechanism of centrosome centering in interphase cellsdepends mainly on microtubules and microtubule motorsand differs from the mechanism of nuclear centering basedon actomyosin activity. Perhaps central location of nuclei andcentrosomes does not indicate the existence of a specificconnection between them.

‘Magnet’ – the centrosome is attracted to nucleusby microtubules and dynein

Sometimes nuclei use microtubules for their movement incells, and in this case, if microtubules originate from thecentrosome, both organelles are connected with micro-tubules. This phenomenon is observed mostly in earlymitotic and in large (oocytes, hyphae) cells. The attachmentof the astrosphere (centrosome and surrounding micro-tubules) to the early mitotic nucleus in the blastomeres of seaurchin was for the first time described by Aronson (1971),who demonstrated that if the centrosome and the nucleus inthe blastomeres of sea urchin were separated by colcemid andcentrifugation, and the colcemid was washed out, the nucleiwere attracted to the centres of astrospheres. Migration ofnuclei directed towards the centre of microtubule aster wasobserved in case of rapid female pronuclear movement inC. elegans zygote (O'Connell et al., 2000; Schneider andBowerman, 2003). Nuclei move along the microtubules indynein- and kinesin-dependent manner in many animalzygotes and early embryos (Baker et al., 1993; Fridolfsson andStarr, 2010; Fridolfsson et al., 2010), and also in neuralprogenitor cells (Tsai et al., 2010; Kosodo et al., 2011).Nuclear movement along microtubules is evident in G2-prophase fibroblast-like cells as well (Tanenbaum et al.,2010). At the same time, separating centrosomes can movealong nuclear envelope covered with dynein (Gönczy

Association of centrosome with nucleus A. V. Burakov and E. S. Nadezhdina

96 Cell Biol Int 37 (2013) 95–104 � 2013 International Federation for Cell Biology

et al., 1999). Positioning of nuclei in G2-early prophase cellsdepends on the balance of dynein and kinesin-generatedforces; after dynein inhibition, kinesin displaces nuclei fromcell centre to the periphery. The distance between nucleusand centrosome under this condition increases to 10 mm ormore (Splinter et al., 2010). Moreover, centrosome-derivedgrowing microtubules can push the nucleus andmove it awayfrom centrosome (Zhao et al., 2012).

In interphase fibroblast-like cells, the nucleus often rotatesin a microtubule- and dynein-dependent manner (Levy andHolzbaur, 2008; Gerashchenko et al., 2009; Wu et al., 2011).Nuclear rotationwas observed in early embryos of C. elegans aswell (Fridolfsson and Starr, 2010). The nuclear rotating forceis gained as a result of the asymmetry of microtubule arraylocated near the nucleus (Wu et al., 2011). The rotating forcesare generated by nucleus-bound dynein motor. Dynein tendsto move the nucleus towards the minus-end of microtubuleslocated near the centrosome, and microtubule asymmetrygoverns nucleus rotation. During nucleus rotation thecentrosome remains in place (Levy and Holzbaur, 2008). Ifhypothetical nucleus–centrosome specific structural linksexist, in this case microtubules either ‘overcome’ them, or thelinks are highly dynamic.

Movement of the nuclei along microtubule array is usuallymediated by dynein and kinesin (Fridolfsson and Starr, 2010;Starr, 2011). In fungal hyphae, nuclear migration is regulatedby dynein accessory regulatory proteins NudC, NudF/Lis1and NudE (Xiang and Fischer, 2004). Dynein together withits cofactor dynactin is attached to the nuclear envelopemembrane by the LInkers of the Nucleoskeleton andCytoskeleton (LINC) protein complexes (Meyerzon et al.,2009; Starr, 2009; Zhang et al., 2009b; Yu et al., 2011), or bymeans of nuclear pore proteins RanBP2/Nup358 andNup133 and adaptor proteins BICD2, CENP-F and NudE/EL (Splinter et al., 2010; Tanenbaum et al., 2010; Bolhy et al.,2011) (Table 1). Kinesin-1might bind the nucleus in a similarmanner, interacting with the same protein complexes ofthe nuclear envelope (Splinter et al., 2010; Tsai et al., 2010;Yu et al., 2011, Wilson and Holzbaur, 2012). Kinesin-3 isalso involved in nuclear movement (Tsai et al., 2010). Ifkinesin is more active than dynein, the nucleus is movedtowards the periphery of microtubule aster; if dynein is moreactive than kinesin, the nucleus is moved towards thecentrosome or rotates (Splinter et al., 2010; Zhang andHan, 2010). The involvement of opposite motors helpsnuclei to overcome roadblocks (Starr, 2011). Peripheralmovement of nucleus can be also mediated by the pushingforces of growing microtubules (Zhao et al., 2012). Theinteraction of centrosome-derivedmicrotubule conewith thenucleus surface depends on nuclear geometry: the nuclearsurface should be large enough to enable lateral interactionwith microtubules. If male pronucleus in C. elegans zygoteis large enough, both prophase centrosomes retain on it,

if it is small – only one centrosome can anchor there(Meyerzon et al., 2009). However, nuclei sometimesmove along microtubules which are not connected to thecentrosome, for example in developing myoblasts (Wilsonand Holzbaur, 2012). Thus, one of the mechanisms that linkis a magnet-like dynein-driven attraction. Dynein-dependingconnection is labile because it is based on dynamicassociation of dynein with microtubules and on dynamicmicrotubules (Figure 1). Such a connection does not requirespecial properties of the centrosome: it acts as microtubule-organizing centre.

It is not clear whether dynein is associated with the nucleusthroughout interphase and prophase or binds to the nucleusonly in G2-prophase. Indeed, in G2-prophase dynein isenriched at nuclear envelope, and CENP-F- and BICD2-dependent pathways of dynein recruitment are activated(Bolhy et al., 2011; Splinter et al., 2010). However, dynein-dependent nuclear rotation is observed throughout the wholecell cycle (Levy and Holzbaur, 2008). It is uncertain whetherthe close attraction of G2-prophase nuclei to the centrosomeis a result of dynein recruitment to the nuclear envelope anddynein activation, or a result of inactivation of kinesin motor(Robinson et al., 1999; Tsai et al., 2010; Kosodo et al., 2011;Leung et al., 2011).

Physical association of centrosome and nucleuswith special links – ‘velcro’

The nucleus and the centrosome can be connected with twotypes of links: labile microtubules and dynein that functionsas a ‘magnet’, attracting nuclei and centrosomes to each other,and some specific fibres, connecting the nucleus and thecentrosome which tie these two organelles towards eachother. The phenomenon of nuclear rotation contradicts theexistence of continuous fibrils; however, velcro type ofconnection remains possible. The velcromodel was proposedfor association of nucleus with other cytoskeleton structures(Tzur et al., 2006). The physical association of the centrosomewith the interphase nucleus was detected in vitro afterdestruction of microtubules. After isolation of nuclei eitherfrom rat liver (Bornens, 1977; Nadezhdina et al., 1979) orfrom cultured cells (Gould and Borisy, 1977; Maro andBornens, 1980; Kuriyama and Borisy, 1981) the cytoplasmicfragments containing centrioles, but not microtubules,were found near the nuclei. Centrioles in this case weresurrounded by fibrillar material extended towards thenucleus. After Triton X-100 treatment, the number ofcentrioles in the nuclei specimens did not decrease, althoughboth membranes of the nuclear envelope had been destroyed.There were filaments with a diameter of about 10 nm(intermediate filaments?), entangled the centrosome andapproaching close to the nucleus (Nadezhdina et al., 1979).However, the role of intermediate filaments in nucleus–

A. V. Burakov and E. S. Nadezhdina Association of centrosome with nucleus


Table 1 Summary of proteins involved into association of centrosome and nucleus.

Name of protein

The point of centrosome-to-nucleus

association Specific cellular function (if known) Refs.

Actin Disruption of actin filaments dissociates

centrosome from nucleus

Forms ubiquitous thin filaments in

cytoplasm

Shay et al. (1974), Hubert

et al. (2011)

DISC1 Distance between nucleus and the

centrosome is extended, neuron migration is

disturbed in case of abnormal DISC1

Structural protein associated with

dynein–dynactin complex

Eastwood et al. (2010)

Asp expression

product

Mutations leads to multiple free centrosomes

in Drosophila embryos

Structural protein with unknown

functions

Gonzalez et al. (1990),

Wakefield et al. (2001)

BICD2, NudE/EL Inhibition increases distance between

centrosome and nucleus

Structural protein associated with

dynein–dynactin complex

Splinter et al. (2010),

Tanenbaum et al. (2010),

Bolhy et al. (2011)

CENP-F Inhibition increases distance between


Adaptor protein for dynein/dynactin

attachment to the nuclear envelope

membrane (also called mitosin)

Bolhy et al. (2011)

Chk2/Mnk kinase When disturbed, free centrosomes in

Drosophila embryos arises

Kinase activated by DNA damage Sibon et al. (2000), Takada

et al. (2003)

Dynein Inhibition increases distance between


Microtubule minus-end directed

motor protein

Splinter et al. (2010)

Emerin Inhibition increases distance between


Nuclear envelope lamina protein Salpingidou et al. (2007)

Greatwall/MAST-L

mitotic kinase

Inhibition increases distance between


Antagonist of Polo kinase Archambault et al. (2007)

Hook3 Inhibition increases distance between

centrosome and nucleus in migrating

neurons

Mammalian ortholog of ZYG12.

Long fibrillar protein

Walenta et al. (2001)

Katanin p60 Inhibition increases distance between


Microtubule-severing protein Toyo-Oka et al. (2005)

Klaroid Mutations detached the nuclei from the

centrosomes, nuclei located basally after

centrosomes migration to the apical surface

of the eye disc

SUN protein, Klarsicht's partner Kracklauer et al. (2007)

Klarsicht The same as Klaroid KASH protein, Klaroid's partner Mosley-Bishop et al. (1999)

Map205 Inhibition leads to centrosome's detachment

from the nucleus

Polo's partner, which sequesters the

kinase. The same as greatwall/

MAST-L

Archambault et al. (2008)

Map205 Inhibition leads to centrosome's detachment

from the nucleus

Polo's partner, which sequesters the

kinase. The same as greatwall/

MAST-L

Archambault et al. (2008)

Mars expressionproduct

Mutations leads to detachment of

centrosomes from mitotic spindles in

Drosophila embryos

Sructural protein with unknown

functions

Bennett and Alphey (2004),

Zhang et al. (2009a)

Nesprin-3 Inhibition increases distance between


Organises perinuclear cytoskeleton

in human aortic endothelial cells

Morgan et al. (2011)

PCM-1 Is associated with Hook3 long fibrillar protein of pericentriolar material Ge et al. (2010)

Polo kinase Inhibition in Drosophila leads to centro-

some's detachment from the nucleus

Mitotic kinase Sunkel and Glover (1988)



centrosome connection has not been elucidated yet. Mostprobably, vimentin filaments are not associated with thecentrosome (Maro et al., 1984) though they can concentratearound it at early stage of filament system restoration(Nadezhdina and Vaisberg, 1987; Alieva et al., 1992).

Unexpectedly, after pretreatment of cells with cytochalasinB, in vitro isolated nucleus–centriole complex could beeasily destroyed by centrifugation in sucrose (Maro andBornens, 1980). Actin filaments or another structuresassociated with actin may have been involved in nucleus–centrosome binding. Moreover, centrifugation of living cellsin cytochalasin B resulted in clear separation of nuclei andcentrosomes. Centrosomes remained in cytoplasts and nucleiwere displayed to karyoplasts (Shay et al., 1974). It seemed

that centrosomes in cytoplasts were retained by micro-tubules. Indeed, after treatment of the cells with cytochalasinB and colcemid (or nocodazole) followed by centrifugation,the centrosomes were found in karyoplasts (Karsentiet al., 1984). Dislocation of the nucleus after centrifugationof cell's monolayer entailed dislocation of the centrosome aswell. The distance of centrosome displacement was indepen-dent of microtubules and no centrosome displacement wasobserved after actin filament disruption (Fais et al., 1984).These data contradict the idea of microtubules as the onlydirect link between the centrosome and the nucleus anddenote the role of actin or actin-associated fibres in nucleus–centrosome association, probably, as a velcro-type associa-tion. Actin is involved in centrosomal separation in mitotic

Figure 1 Schemeof twomodels (magnet and velcro) of centrosome–nucleus association. Black arrow indicates the direction of the pulling forces

along microtubules, generated by dynein motors.

1, perinuclear space; 2, lamins; 3, chromatin; ONM, outer nuclear membrane; INM, inner nuclear membrane. A picture of industrial velcro is from

Science photo library/barcroft media.

Table 1. (Continued)

Name of protein

The point of centrosome-to-nucleus

association Specific cellular function (if known) Refs.

RanBP2/Nup358,

Nup133

Attaching the dynein/dynactin to the nuclear

envelope

Nuclear pore proteins. Bind also

dynein and dynactin

Bolhy et al. (2011)

Spag4/Giacomo Binds to the Yuri Gagarin needed for the association of basal body of

sperm flagellum with spermatid's nucleus

Kracklauer et al. (2010)

SUN-1 SUN group protein, partner of ZYG-12 as a KASH protein Minn et al. (2009)

TBCCD1 Inhibition leads to the rapid increase of the

nucleus-centrosome distance

Protein related to tubulin cofactor

C TBCC-domain containing 1

Gonçalves et al. (2010)

Yuri Gagarin Mutation leads to aberrant nuclear migration

during Drosophila spermatogenesis

Binds to the outer nuclear envelope.

Also is involved in actin organisation

Texada et al. (2008),

Kracklauer et al. (2010)

ZYG-12 Mutations leads to the pronucleus migration

defects in C. elegans zygoteZYG12 dimer binds to to outer

nuclear membrane and to

centrosome

Malone et al. (2003)



prophase: destruction of actin filaments with latrunculininhibits splitting of centrosome into two (Uzbekovet al., 2002). Probably, the observed phenomenon does notresult from direct actin interaction with the centrosome, butis facilitated by microtubule plus-tip interaction with cellularactin cortex (Cao et al., 2010). However, actin and Arp2/3have been identified in pericentriolar material in interphasecells (Hubert et al., 2011). Obviously, the question of theassociation of actin filaments with the centrosome is poorlystudied and needs further investigations. Interestingly, YuriGagarin (see below) and Lis1 proteins (discussed below)participating in nucleus–centrosome association are alsoinvolved in actin organisation (Texada et al., 2008; Meyeret al., 2011) (Table 1).

What other structures might be involved in the centro-some–nucleus velcro? Themost appealing hypothesis impliesthe existence of a ‘bridge’ between the nucleus and thecentrosome composed of dimers or oligomers of specificproteins. Two major groups of proteins, called KASH (fromKlarsicht, ANC-1 and Syne-1) and SUN (from Sud1 andUnc-84) are involved in the interaction of nucleus withcytoskeletal structures. Proteins containing SUN-domain arelocated in the inner nuclear membrane, and recruit proteinscontaining KASH domain, protruding from the surface of thenucleus to the cytoplasm and interacting with the cytoskeletalelements (Crisp et al., 2006; McGee et al., 2006; Padmakumaret al., 2005). KASH and SUN proteins form a bridge betweenthe two membranes of the nuclear envelope – LINC-complexes (reviewed by Burke and Roux, 2009; Razafsky andHodzic, 2009). Many of KASH proteins are almost 9,000amino acids long, and can extend up to 25 mm. In cells, theseproteins are folded, and their length is much shorter thanpredicted. Nevertheless, a molecule of such a long proteinmight easily form a 2 mmbridge between the nucleus and thecentrosome.

KASH and SUN proteins are variable. Two genes encodingSUN-domains have been identified in Drosophila: spag4(giacomo), expressed exclusively in the testes (Kracklaueret al., 2010), and more widely expressed klaroid (Kracklaueret al., 2007, 2010). Both Klaroid and its KASH-partner,Klarsicht, are required to move and retain the nucleus in theapical part of the developing photoreceptor cells of eye disks(Mosley-Bishop et al., 1999; Kracklauer et al., 2007). Inklarsicht or klaroid mutants the centrosomes migrate to theapical surface of the eye disc immediately after the passing ofmorphogenic front, but the nuclei are not able to attach to thecentrosomes and remain located basally (Patterson et al.,2004; Kracklauer et al., 2007) (Table 1).

Spag4/Giacomo protein binds to the coiled-coil proteinYuri Gagarin, which lacks KASH domain, but is also locatedin the outer nuclear envelope. This connection is necessaryfor the association of basal body of sperm flagellum with aspermatid's nucleus (Kracklauer et al., 2010) (Table 1). In C.

elegans, the function of KASH protein that links centrosometo the nucleus is fulfilled by a fibrillar coiled coil protein ZYG-12. ZYG-12 is almost 750 amino acids long, indicating that itsmaximal length is only a little more than 2 mm, and onemolecule can not cover the distance from nucleus to thecentrosome. ZYG-12 binds to the outer nuclear membraneby its transmembrane domain and can form dimers, in whichprobably one molecule is located in nuclear envelope and theother is bound to the centrosome (Malone et al., 2003).Mutations in ZYG-12 result in the pronucleus migrationdefects in C. elegans zygote, although such a migration alsodepends on dynein and microtubules (Malone et al., 2003).Possibly, dynein functions as a magnet, which helps to fastenVelcro adhesion. ZYG-12 is coupled with SUN-1 protein, andboth are essential for centrosome–nucleus attachment (Minnet al., 2009) (Table 1). Disruption of nuclear envelope proteinemerin also leads to the increase of distance from nucleus tocentrosome (Salpingidou et al., 2007) (Table 1). Interactionof nuclear envelope ZYG-12 with dynein defines microtubuleorganisation and nuclear positioning in C. elegans gonad(Zhou et al., 2009).

The mammalian ortholog of ZYG-12 is called Hook3(Walenta et al., 2001), implicated in binding of microtubulesto Golgi and in the association of centrosome with thenucleus in migrating neurons in particular (Table 1).Inhibition of Hook3 affects neuronal migration. Hook3 isassociated with the pericentriolar satellite protein PCM-1,and in the absence of Hook3, less microtubules radiate fromthe centrosome (Ge et al., 2010) (Table 1). The observedphenomenon might cause the loss of dynein-dependentmagnet-like link between nucleus and the centrosometogether with the loss of Velcro-like Hook3-mediatedconnection. The nucleus–centrosome link in mammals ismediated also by another KASH-domain fibrillar proteinsnesprins 1–4 (Zhang et al., 2001, 2002). Expression of nesprinisoforms is tissue-specific. In human aortic endothelial cells,nesprin-3 organises perinuclear cytoskeleton. Knockdown ofnesprin-3 increases the distance between nucleus andcentrosome and inhibits centrosome re-orientation undershear stress (Morgan et al., 2011) (Table 1). Nesprin-1/2(Syn-1/2) is expressed in neural progenitor cells. Knockout ofboth Nesprins as well as both their SUN-domain partners Sun1and Sun2 lead to a remarkable increase of the distancebetween centrosome and nucleus (Table 1). Either Sun1/2 orSyn1/2 double knockout pups die immediately after birth;their brains have severe laminary defects. Syn1/2 (Nesprins 1/2) proteins are bound to dynein–dynactin complex and tokinesin and connect them to nuclear membrane (Zhanget al., 2009b). However, the reason of increased distancebetween centrosome and nucleus after Syn1/2/Nesprin1/2inhibition remains unknown: is it a direct effect or is itmediated by dynein disturbance. It has been shown thatinterkinetic nuclear migration does not depend on the



centrosome (Tsai et al., 2010) and is based on microtubuleremodelling induced by microtubule-binding Tpx2 proteinreleased from nuclei in G2-phase (Kosodo et al., 2011). Thedistance between nucleus and the centrosome is extendedand neuron migration is disturbed in case of abnormal Lis1and NDEL1 (Youn et al., 2009), as well as DISC1 (Eastwoodet al., 2010) proteins (Table 1). All these proteins eitherregulate dynein and/or modulate microtubule-nucleatingactivity of the centrosome. Their impact on neuronalmigration is rather a result of microtubule and cell motilityregulation than a result of nucleus and centrosomeconnection. Though the mechanisms of LINC-proteinsattachment to the nucleus are clear enough, the mechanismsof their binding to the centrosome are almost unknown. Itcan be assumed that this connection is labile andmediated bydynein (Robinson et al., 1999; Malone et al., 2003; Andersonet al., 2009; Wainman et al., 2009).

Other proteins involved in centrosome attachmentto the nucleus

In Drosophila embryos, detachment of centrosomes isobserved in response to DNA damage, and it is mediatedby Chk2/Mnk kinase activity (Sibon et al., 2000; Takadaet al., 2003) (Table 1). Such a detachment occurs in prophaseand is coupled to a decline of centrosome microtubule-organizing activity (Archambault and Pinson, 2010). Cen-trosome detachment from the nucleus in Drosophila embryo isalso observed with inhibition of Polo kinase (Sunkel andGlover, 1988) (Table 1). Such a phenotype can be enhancedby mutations in Scant gene encoding the mitotic kinasegreatwall/MAST-L, which acts as an antagonist of Polo(Archambault et al., 2007) (Table 1). In mutated flies, freecentrosomes together with monoastral mitotic spindlesappear. Overexpression of Polo's partner, Map205, whichsequesters the kinase, has a similar effect (Archambault et al.,2008) (Table 1). It is still unknown how Polo contributes tothe connection of the centrosome to the nucleus. Detachedcentrosomes are able to hold the astral microtubules and donot show any noticeable decrease in the content of g-tubulinor other centrosomal markers. Therefore, in this case, thedetached centrosomes are not inactivated, unlike those thatdetach due to DNA damage. Detachment of centrosomesfrom mitotic spindles occurs also in Drosophila embryosmutated in genes Mars (Bennett and Alphey, 2004; Zhanget al., 2009a) and Asp (Gonzalez et al., 1990; Wakefieldet al., 2001) (Table 1). Both genes encode structural proteinswith unknown functions. Spindle-detached centrosomesremain non-attached to nuclei, the finding that supports theidea of specific nuclear-centrosomal link.

In cultured mammalian cells, rapid increase of thenucleus–centrosome distance is sometimes unexpectedlyobserved with the inhibition of some protein expression:

microtubule-severing protein katanin p60 (Toyo-Okaet al., 2005); protein related to tubulin cofactor C TBCC-domain containing 1 (TBCCD1) (Gonçalves et al., 2010)(Table 1). The observed increase might result from thedisturbance of a specific link between the two organelles, inparticular if the association with LINC-proteins was affected,from dynein inhibition, or from the suppression ofmicrotubule-organizing activity of the centrosome. Elucida-tion of each mechanism requires further research.

Conclusion

Binding of the centrosome to the nucleus might be mediatedin animal cells by different mechanisms, which we havedesignated as magnet and velcro. Magnet involves micro-tubules and dynein, and velcro is mediated by different typesof molecular scaffolds, depending on tissue type or organismspecies. The variability of links might facilitate fine regulationof migration and positioning of nucleus and centrosome ineach case. In some cases, velcro, which connects centrosometo the nucleus, might be absent, and the connection isensured only by microtubules and dynein. Velcro betweentwo organelles might be organised by actin together withKASH-domain long proteins (Figure 1). It seems also that thecentrosomal ‘magnet’ binding to the nucleus requiresdifferential regulation of dynein's activity in various cellcompartments; in particular, the local regulation of dynein atnuclear surface. Magnet and velcro can interact: a part ofdynein is bound to nuclear envelope via LINCs. LINC-bounddynein might to compete centrosome-bound KASH proteinsand/or actin for establishing of velcro-like connection ofcentrosome. On the other hand, the direct role of dynein infastening velcro between nucleus and the centrosome cannotbe excluded.

The phenomenon of centrosome binding to the nucleuswas described about 40 years ago (Aronson, 1971;Bornens, 1977). To understand clearly the mechanisms andsignificance of this binding requires further experimentation.Magnet-like dynein-based connection is studied better thanvelcro-like binding. It remains difficult to determine the roleof actin in velcro-type connection because of multiplefunctions of actin filaments and their broad distribution incells.

KASH/SUN links need to be directly visualised in cells, andtheir dynamics in various cell types and at various stages ofcell cycle has to be studied with high-resolution lightmicroscopy with specific antibodies and truncated proteinconstructs. The mechanism of KASH protein binding tocentrosome and hypothetical KASH fibril structure remainsobscure and can be studied together with interaction of theseproteins to actin. It is also tempting to disrupt centrosome-to-nucleus connection avoiding dynein inhibition and studycell movement and entry into mitosis in such cells.



Acknowledgement and Funding

The authors are very grateful to Dr. Olga Zhapparova forcritically reading the manuscript. The work was supportedfinancially by Russian Foundation of Basic Research [11-04-01022].

References

Alieva IB, Nadezhdina ES, Vaisberg EA, Vorobjev IA (1992)Microtubule and intermediate filament patterns aroundthe centrosome in interphase cells. In: Kalnins VL, ed. TheCentrosome, San Diego: Academic Press, pp. 103–30.

AndersonMA, Jodoin JN, Lee E, Hales KG, Hays TS, Lee LA (2009)Asunder is a critical regulator of dynein–dynactin localizationduring Drosophila spermatogenesis. Mol Biol Cell 20: 2709–21.

Archambault V, Pinson X (2010) Free centrosomes. Where do theyall come from? Fly 4: 172–7.

Archambault V, Zhao X, White-Cooper H, Carpenter AT, GloverDM (2007) Mutations in Drosophila greatwall/Scant reveal itsroles in mitosis and meiosis and interdependence with polokinase. PLoS Genet 3: 200.

Archambault V, D'Avino PP, Deery MJ, Lilley KS, Glover DM(2008) Sequestration of polo kinase to microtubules byphosphopriming-independent binding to Map205 is relievedby phosphorylation at a CDK site in mitosis. Genes Dev 22:2707–20.

Aronson JF (1971) Demonstration of a colcemid-sensitiveattractive force acting between the nucleus and a center.J Cell Biol 51(21): 579–83.

Baker J, Theurkauf WE, Schubiger G (1993) Dynamic changes inmicrotubule configuration correlate with nuclear migration inthe preblastoderm Drosophila embryo. J Cell Biol 122: 113–21.

Bennett D, Alphey L (2004) Cloning and expression of mars, anovel member of the guanylate kinase associated protein familyin Drosophila. Gene Expr Patterns 4: 529–35.

Bolhy S, Bouhlel I, Dultz E, Nayak T, Zuccolo M, Gatti X, ValleeR, Ellenberg J, Doye V (2011) A Nup133-dependent NPC-anchored network tethers centrosomes to the nuclear envelopein prophase. J Cell Biol 192: 855–71.

Bornens M (1977) Is the centriole bound to the nuclearmembrane? Nature 270: 80–2.

Brito DA, Strauss J, Magidson V, Tikhonenko I, Khodjakov A,Koonce MP (2005) Pushing forces drive the comet-like motilityof microtubule arrays in Dictyostelium. Mol Biol Cell 16:3334–40.

Brodsky IB, Burakov AV, Nadezhdina ES (2007) Microtubules'interaction with cell cortex is required for their radialorganization, but not for centrosome positioning. Cell MotilCytoskeleton 64: 407–17.

Burakov A, Nadezhdina E, Slepchenko B, Rodionov V (2003)Centrosome positioning in interphase cells. J Cell Biol 162:963–9.

Burke B, Roux KJ (2009) Nuclei take a position: managing nuclearlocation. Dev Cell 17: 587–97.

Cao J, Crest J, Fasulo B, SullivanW (2010) Cortical actin dynamicsfacilitate early-stage centrosome separation. Curr Biol 20:770–6.

Crisp M, Liu Q, Roux K, Rattner JB, Shanahan C, Burke B, et al.(2006) Coupling of the nucleus and cytoplasm: role of the LINCcomplex. J. Cell Biol 172: 41–53.

Dupin I, Etienne-Manneville S (2011) Nuclear positioning:mechanisms and functions. Int J Biochem Cell Biol 43: 1698–707.

Eastwood SL, Walker M, Hyde TM, Kleinman JE, Harrison PJ(2010) The DISC1 Ser704Cys substitution affects centrosomallocalization of its binding partner PCM1 in glia in human brain.Hum Mol Genet 19: 2487–96.

Fais D, Nadezhdina ES, Chentsov YuS (1984) Evidence for thenucleus–centriole association in living cells obtained byultracentrifugation. Eur J Cell Biol 33: 190–6.

Fridolfsson HN, Starr DA (2010) Kinesin-1 and dynein at thenuclear envelope mediate the bidirectional migrations of nuclei.J Cell Biol 191: 115–28.

Fridolfsson HN, Ly N, Meyerzon M, Starr DA (2010) UNC-83coordinates kinesin-1 and dynein activities at the nuclearenvelope during nuclear migration. Dev Biol 338: 237–50.

Ge X, Frank CL, Calderon de Anda F, Tsai LH (2010) Hook3interacts with PCM1 to regulate pericentriolar materialassembly and the timing of neurogenesis. Neuron 65: 191–203.

Gerashchenko MV, Chernoivanenko IS, Moldaver MV, Minin AA(2009) Dynein is a motor for nuclear rotation while vimentinIFs is a “brake.” Cell Biol Int 33: 1057–64.

Gonçalves J, Nolasco S, Nascimento R, Lopez Fanarraga M, ZabalaJC, Soares H (2010) TBCCD1, a new centrosomal protein, isrequired for centrosome and Golgi apparatus positioning.EMBO Rep 11: 194–200.

Gönczy P, Pichler S, Kirkham M, Hyman AA (1999) Cytoplasmicdynein is required for distinct aspects of Mtoc positioning,including centrosome separation, in the one cell stageCaenorhabditis elegans embryo. J Cell Biol 147: 135–50.

Gonzalez C, Saunders RD, Casal J, Molina I, Carmena M, Ripoll P,et al. (1990) Mutations at the asp locus of Drosophila leadto multiple free centrosomes in syncytial embryos, butrestrict centrosome duplication in larval neuroblasts. J CellSci 96: 605–16.

Gould RR, Borisy GG (1977) The pericentriolar material inChinese hamster ovary cells nucleates microtubule formation.J Cell Biol 73: 601–15.

Gundersen GG, Wen Y, Eng CH, Schmoranzer J, Cabrera-Poch N,Morris EJ, et al. (2005) Regulation of microtubules by RhoGTPases in migrating cells. Novartis Found Symp 269: 106–16.

Hale CM, Shrestha AL, Khatau SB, Stewart-Hutchinson PJ,Hernandez L, Stewart CL, et al. (2008) Dysfunctional connec-tions between the nucleus and the actin and microtubulenetworks in laminopathic models. Biophys J 95: 5462–75.

Hampoelz B, Azou-Gros Y, Fabre R, Markova O, Puech PH, LecuitT (2011) Microtubule-induced nuclear envelope fluctuationscontrol chromatin dynamics in Drosophila embryos. Develop-ment 138: 3377–86.



Hubert T, Vandekerckhove J, Gettemans J (2011) Actin and Arp2/3localize at the centrosome of interphase cells. Biochem BiophysRes Commun 404: 153–8.

Hyman AA,White JG (1987) Determination of cell division axes inthe early embryogenesis of Caenorhabditis elegans. J Cell Biol 105:2123–35.

Karsenti E, Kobayashi S, Mitchison T, Kirschner M (1984) Role ofthe centrosome in organizing the interphase microtubule array:properties of cytoplasts containing or lacking centrosomes.J Cell Biol 98: 1763–76.

Khatau SB, Kim DH, Hale CM, Bloom RJ, Wirtz D (2010) Theperinuclear actin cap in health and disease. Nucleus 1: 337–42.

Kimura K, Kimura A (2011) A novel mechanism of microtubulelength-dependent force to pull centrosomes toward the cellcenter. Bioarchitecture 1: 74–79.

Koonce MP, Köhler J, Neujahr R, Schwartz JM, Tikhonenko I,Gerisch G (1999) Dynein motor regulation stabilizes interphasemicrotubule arrays and determines centrosome position.EMBO J 18: 6786–92.

Kosodo Y, Suetsugu T, Suda M, Mimori-Kiyosue Y, Toida K, BabaSA, et al. (2011) Regulation of interkinetic nuclear migrationby cell cycle-coupled active and passive mechanisms in thedeveloping brain. EMBO J 30: 1690–704.

Kracklauer MP, Banks SM, Xie X, Wu Y, Fischer JA (2007)Drosophila klaroid encodes a SUN domain protein required forKlarsicht localization to the nuclear envelope and nuclearmigration in the eye. Fly (Austin) 1: 75–85.

Kracklauer MP, Wiora HM, Deery WJ, Chen X, Bolival B Jr,Romanowicz D, et al. (2010) The Drosophila SUN proteinSpag4 cooperates with the coiled-coil protein Yuri Gagarin tomaintain association of the basal body and spermatid nucleus.J Cell Sci 123: 2763–72.

Krokhina TB, Nadezhdina ES (1991) The role of the cytoskeletalelements during the return to the cell center of their nucleidisplaced by centrifugation. Tsitologiia 33: 28–35 (Russian).

Kuriyama R, Borisy GG (1981) Centriole cycle in Chinese hamsterovary cells as determined by whole-mount electron microscopy.J Cell Biol 91: 814–21.

Leung L, Klopper AV, Grill SW, Harris WA, Norden C (2011)Apical migration of nuclei during G2 is a prerequisite for allnuclear motion in zebrafish neuroepithelia. Development 138:5003–13.

Levy JR, Holzbaur EL (2008) Dynein drives nuclear rotation duringforward progression ofmotile fibroblasts. J Cell Sci 121: 3187–95.

Luxton GW, Gundersen GG (2011) Orientation and function ofthe nuclear-centrosomal axis during cell migration. Curr OpinCell Biol 23: 579–88.

Malone CJ,Misner L, Le Bot N, TsaiMC, Campbell JM, Ahringer J,et al. (2003) The C. elegans hook protein, ZYG-12, mediates theessential attachment between the centrosome and nucleus. Cell115: 825–36.

Maly IV (2011) Systems biomechanics of centrosome positioning:a conserved complexity. Commun Integr Biol 4(2): 230–5.

Maro B, Bornens M (1980) The centriole–nucleus association:effects of cytochalasin B and nocodazole. Biol Cell 39: 287–90.

Maro B, Paintrand M, Sauron ME, Paulin D, Bornens M (1984)Vimentin filaments and centrosomes. Are they associated? ExpCell Res 150: 452–8.

Martini FJ, Valdeolmillos M (2010) Actomyosin contraction at thecell rear drives nuclear translocation in migrating corticalinterneurons. J Neurosci 30: 8660–70.

McGee MD, Rillo R, Anderson AS, Starr DA (2006) UNC-83 is aKASH protein required for nuclear migration and is recruited tothe outer nuclear membrane by a physical interaction with theSUN protein UNC-84. Mol Biol Cell 17: 1790–801.

Meyer I, Kuhnert O, Gräf R (2011) Functional analyses oflissencephaly-related proteins in Dictyostelium. Semin Cell DevBiol 22: 89–96.

Meyerzon M, Gao Z, Liu J, Wu JC, Malone CJ, Starr DA (2009)Centrosome attachment to the C. elegans male pronucleus isdependent on the surface area of the nuclear envelope. Dev Biol327: 433–46.

Minn IL, Rolls MM,Hanna-RoseW,Malone CJ (2009) SUN-1 andZYG-12, mediators of centrosome-nucleus attachment, are afunctional SUN/KASH pair in Caenorhabditis elegans. Mol BiolCell 20: 4586–95.

Morgan JT, Pfeiffer ER, Thirkill TL, Kumar P, Peng G, FridolfssonHN, et al. (2011) Nesprin-3 regulates endothelial cell morpho-logy, perinuclear cytoskeletal architecture, and flow-inducedpolarization. Mol Biol Cell 22: 4324–34.

Mosley-Bishop KL, Li Q, Patterson L, Fischer JA (1999) Molecularanalysis of the klarsicht gene and its role in nuclear migrationwithin differentiating cells of the Drosophila eye. Curr Biol 9:1211–20.

Nadezhdina ES, Vaisberg EA (1987) Reorganization of the systemof intermediate filaments and detection of the organizationcenters after centrifugation of cells attached to a substrate.Tsitologiia 29: 543–8.

Nadezhdina ES, Fais D, Chentsov YS (1979) On the associationof centrioles with the interphase nucleus. Eur J Cell Biol 19:109–15.

O'Connell KF, Maxwell KN, White JG (2000) The spd-2 geneis required for polarization of the anteroposterior axis andformation of the sperm asters in the Caenorhabditis elegans zygote.Dev Biol 222: 55–70.

Padmakumar VC, Libotte T, Lu W, Zaim H, Abraham S, NoegelAA, Gotzmann J, Foisner R, Karakesisoglou I (2005) The innernuclear membrane protein Sun1 mediates the anchorage ofNesprin-2 to the nuclear envelope. J Cell Sci 118: 3419–30.

Patterson K, Molofsky AB, Robinson C, Acosta S, Cater C, FischerJA (2004) The functions of klarsicht and nuclear lamin indevelopmentally regulated nuclear migrations of photoreceptorcells in the Drosophila eye. Mol Biol Cell 15: 600–10.

Raff JW, Glover DM (1989) Centrosomes, and not nuclei, initiatepole cell formation in Drosophila embryos. Cell 57: 611–9.

Razafsky D, Hodzic D (2009) Bringing KASH under the SUN: themany faces of nucleo-cytoskeletal connections. J Cell Biol 186:461–72.

Robinson JT, Wojcik EJ, Sanders MA, McGrail M, Hays TS (1999)Cytoplasmic dynein is required for the nuclear attachment and



migration of centrosomes during mitosis in Drosophila. J CellBiol 146: 597–608.

Salpingidou G, Smertenko A, Hausmanowa-Petrucewicz I, HusseyPJ, Hutchison CJ (2007) A novel role for the nuclear membraneprotein emerin in association of the centrosome to the outernuclear membrane. J Cell Biol 178: 897–904.

Schneider SQ, Bowerman B (2003) Cell polarity and thecytoskeleton in the Caenorhabditis elegans zygote. Annu RevGenet 37: 221–49.

Shay JW, Porter KR, Prescott DM (1974) The surface morphologyand fine structure of CHO (Chinese hamster ovary) cellsfollowing enucleation. Proc Natl Acad Sci USA 71: 3059–63.

Sibon OC, Kelkar A, Lemstra W, Theurkauf WE (2000) DNA-replication/DNA-damage-dependent centrosome inactivationin Drosophila embryos. Nat Cell Biol 2: 90–5.

Splinter D, Tanenbaum ME, Lindqvist A, Jaarsma D, Flotho A, YuKL, et al. (2010) Bicaudal D2, dynein, and kinesin-1 associatewith nuclear pore complexes and regulate centrosome andnuclear positioning during mitotic entry. PLoS Biol 8: e1000350.

Starr DA (2009) A nuclear-envelope bridge positions nuclei andmoves chromosomes. J Cell Sci 122: 577–86.

Starr DA (2011) Watching nuclei move: insights into how kinesin-1 and dynein function together. Bioarchitecture 1: 9–13.

Starr DA, Han M (2003) ANChors away: an actin basedmechanism of nuclear positioning. J Cell Sci 116: 211–6.

Sunkel CE, Glover DM (1988) Polo, a mitotic mutant of Drosophiladisplaying abnormal spindle poles. J Cell Sci 89: 25–38.

Takada S, Kelkar A, Theurkauf WE (2003) Drosophila checkpointkinase 2 couples centrosome function and spindle assembly togenomic integrity. Cell 113: 87–99.

TanenbaumME, Akhmanova A,Medema RH (2010) Dynein at thenuclear envelope. EMBO Rep 11: 649.

Texada MJ, Simonette RA, Johnson CB, Deery WJ, BeckinghamKM (2008) Yuri Gagarin is required for actin, tubulin and basalbody functions in Drosophila spermatogenesis. J Cell Sci 121:1926–36.

Toyo-Oka K, Sasaki S, Yano Y, Mori D, Kobayashi T, Toyoshima YY(2005) Recruitment of katanin p60 by phosphorylated NDEL1,an LIS1 interacting protein, is essential for mitotic cell divisionand neuronal migration. Hum Mol Genet 14: 3113–28.

Tsai JW, Lian WN, Kemal S, Kriegstein AR, Vallee RB (2010)Kinesin 3 and cytoplasmic dynein mediate interkinetic nuclearmigration in neural stem cells. Nat Neurosci 13: 1463–71.

Tzur YB,Wilson KL, Gruenbaum Y (2006) SUN-domain proteins:‘Velcro’ that links the nucleoskeleton to the cytoskeleton. NatRev Mol Cell Biol 7: 782–8.

Uzbekov R, Kireyev I, Prigent C (2002) Centrosome separation:respective role of microtubules and actin filaments. Biol Cell 94:275–88.

Wainman A, Creque J, Williams B, Williams EV, Bonaccorsi S,Gatti M, et al. (2009) Roles of the Drosophila NudE protein inkinetochore function and centrosome migration. J Cell Sci 122:1747–58.

Wakefield JG, Bonaccorsi S, Gatti M (2001) The drosophila proteinasp is involved in microtubule organization during spindleformation and cytokinesis. J Cell Biol 153: 637–48.

Walenta JH, Didier AJ, Liu X, Krämer H (2001) The golgi-associated hook3 protein is a member of a novel family ofmicrotubule-binding proteins. J Cell Biol 152: 923–34.

Wilson MH, Holzbaur EL (2012) Opposing microtubule motorsdrive robust nuclear dynamics in developing muscle cells. J CellSci 125: 4158–69.

Wu J, Lee KC, Dickinson RB, Lele TP (2011) How dynein andmicrotubules rotate the nucleus. J Cell Physiol 226: 2666–74.

Xiang X, Fischer R (2004) Fungal nuclear migration andpositioning in filamentous fungi. Genet Biol 41: 411–9.

Youn YH, Pramparo T, Hirotsune S, Wynshaw-Boris A (2009)Distinct dose-dependent cortical neuronal migration andneurite extension defects in Lis1 and Ndel1 mutant mice.J Neurosci 29: 15520–30.

Yu J, Lei K, Zhou M, Craft CM, Xu G, Xu T, et al. (2011) KASHprotein Syne-2/Nesprin-2 and SUN proteins SUN1/2 mediatenuclear migration during mammalian retinal development.Hum Mol Genet 20: 1061–73.

Zhang X, Han M (2010) Nuclear migration: rock and rollfacilitated by dynein and kinesin. Curr Biol 20: R1027–9.

ZhangQ, Skepper JN, Yang F, Davies JD, Hegyi L, Roberts RG, et al.(2001) Nesprins: a novel family of spectrin repeat-containingproteins that localize to the nuclear membrane in multipletissues. J Cell Sci 114: 4485–98.

Zhang Q, Ragnauth C, Greener MJ, Shanahan CM, Roberts RG(2002) The nesprins are giant actin-binding proteins, orthol-ogous to Drosophila melanogaster muscle protein MSP-300.Genomics 80: 473–81.

Zhang G, Breuer M, Forster A, Egger-Adam D, Wodarz A (2009)Mars, a Drosophila protein related to vertebrate HURP, isrequired for the attachment of centrosomes to the mitoticspindle during syncytial nuclear divisions. J Cell Sci 122:535–45.

Zhang X, Lei K, Yuan X,WuX, Zhuang Y, Xu T, et al. (2009) SUN1/2 and Syne/Nesprin-1/2 complexes connect centrosome to thenucleus during neurogenesis and neuronal migration in mice.Neuron 64: 173–87.

Zhao T, Graham OS, Raposo A, St Johnston D (2012) Growingmicrotubules push the oocyte nucleus to polarize the Drosophiladorsal-ventral axis. Science 336: 999–1003.

Zhou K, Rolls MM, Hall DH, Malone CJ, Hanna-RoseW (2009) AZYG-12-dynein interaction at the nuclear envelope definescytoskeletal architecture in the C. elegans gonad. J Cell Biol 186:229–41.

Zhu J, Burakov A, Rodionov V, Mogilner A (2010) Finding the cellcenter by a balance of dynein and myosin pulling andmicrotubule pushing: a computational study. Mol Biol Cell21: 4418–27.

Received 2 October 2012; accepted 12 November 2012.Final version published online 2 January 2013.



Documents

Association of nucleus and centrosome: magnet or velcro?