3633Research Article

IntroductionCell division ends with the processes of nuclear division(karyogenesis) and cytosolic separation (cytokinesis). In atypical animal mitosis, coordinated actions from astral andcentral spindle microtubules to the actin cytoskeleton andmembrane systems are essential for proper cytokinesis (Eggertet al., 2006; Glotzer, 2001). Variations of the cytokinesisprocess are observed in certain biological contexts. Incompletecytokinesis very likely contributes to cancer progression, withcommon solid tumours tending to exhibit tetraploid cells(Ganem et al., 2007; Weaver and Cleveland, 2006). Indeed,these cells can appear following defects in a large number ofdifferent cytokinetic proteins (Eggert et al., 2006). In addition,mis-segregation events can lead to the presence of chromatintrapped in the cleavage furrow, inducing cytokinesis failure andtetraploidization (Mullins and Biesele, 1977). Remarkably,incomplete cytokinesis can be a programmed step in normaldevelopment, producing differentiated tetraploid progeny(Glotzer, 2001). In mammals, the differentiation process canbe associated with mitotic cycles initiated but aborted in lateanaphase with failure of both karyogenesis and cytokinesis [i.e.megakaryocytes (Ravid et al., 2002)]. In other programs, cellscomplete karyogenesis; however, during cytokinesis these cellspresent failure in constriction or in abscission. Thus,mammalian cardiomyocytes form an incorrect contractile ring,leading to the genesis of binucleated cells (Engel et al., 2006).In the male germ line, the committed spermatogenic precursorsdo not complete cytokinesis, but rather form a stable

intercellular bridge interconnecting daughter cells in asyncytium (Burgos and Fawcett, 1955).

Tetraploidy is a characteristic feature of mammalianhepatocytes (Gupta, 2000). During postnatal growth, the liverparenchyma undergoes dramatic changes characterized bygradual polyploidization during which hepatocytes of severalploidy classes emerge as a result of modified cell-divisioncycles. This process generates the successive appearance oftetraploid and octoploid cell classes with one or two nuclei.Thus, in the liver of a newborn rat, hepatocytes are exclusivelydiploid (2n). In adult rats, approximately 25% of hepatocytesare diploid, 70% are tetraploid (binucleated 2�2n ormononucleated 4n) and 5% are octoploid (binucleated 2�4nor mononucleated 8n). The degree of polyploidization variesin different mammals. In humans, the number of polyploidcells averages 30-40% in the adult liver (Kudryavtsev et al.,1993; Toyoda et al., 2005). In adults, liver polyploidization isdifferentially regulated upon loss of liver mass and liverdamage. Interestingly, partial hepatectomy induces marked cellproliferation followed by an increase in liver ploidy (Sigal etal., 1999). By contrast, during hepatocarcinoma (HCC), growthshifts to a non-polyploidizing pattern and expansion of thediploid hepatocyte population is observed in neoplastic nodules(Seglen, 1997). We previously uncovered the sequentialappearance during liver growth of binucleated 2�2n andmononucleated 4n hepatocytes from a diploid hepatocytepopulation (Guidotti et al., 2003). Furthermore, we showed invitro that binucleated 2�2n hepatocytes emerge as a result of

Cytokinesis is precisely controlled in both time and spaceto ensure equal distribution of the genetic material betweendaughter cells. Incomplete cytokinesis can be associatedwith developmental or pathological cell division programsleading to tetraploid progenies. In this study we decipher anew mechanism of incomplete cytokinesis taking place inhepatocytes during post-natal liver growth. This process isinitiated in vivo after weaning and is associated with anabsence of anaphase cell elongation. In this process,formation of a functional contractile actomyosin ring wasnever observed; indeed, actin filaments spread out alongthe cortex were not concentrated to the putative site offurrowing. Recruitment of myosin II to the cortex,

controlled by Rho-kinase, was impaired. Astralmicrotubules failed to contact the equatorial cortex and todeliver their molecular signal, preventing activation of theRhoA pathway. These findings reveal a new developmentalcell division program in the liver that prevents cleavage-plane specification.

Supplementary material available online athttp://jcs.biologists.org/cgi/content/full/120/20/3633/DC1

Key words: Tetraploidy, Hepatocytes, Cytokinesis, Weaning,Cytoskeleton

Summary

Liver tetraploidization is controlled by a new processof incomplete cytokinesisGermain Margall-Ducos1,2, Séverine Celton-Morizur1,2, Dominique Couton1,2, Olivier Brégerie3 andChantal Desdouets1,2,*1Institut Cochin, Université Paris Descartes, CNRS (UMR 8104), Paris, France2Inserm, U567, Paris, France3Inserm, U785, Université Paris sud, Centre Hépato-Biliaire, Hôpital Paul Brousse, Villejuif F-94804, France*Author for correspondence (e-mail: desdouets@cochin.inserm.fr)

Accepted 21 August 2007Journal of Cell Science 120, 3633-3639 Published by The Company of Biologists 2007doi:10.1242/jcs.016907

Jour

nal o

f Cel

l Sci

ence

3634

an incomplete cytokinesis. These tetraploid cells are capableof proliferation. In fact, binucleated hepatocytes are able toproceed through S phase and the formation of a bipolar spindleduring mitosis constituted the key step leading to the genesisof two mononucleated 4n hepatocytes (Guidotti et al., 2003).

In the current study, we investigated when the incompletecytokinesis process is taking place during postnatal livergrowth and how this specific division program is controlled.We deciphered, in the liver, a new physiological process ofincomplete cytokinesis triggered by weaning. Mitotichepatocytes achieved karyogenesis without establishing thecleavage plane because of the deficiencies of actin cytoskeletonand microtubule reorganization.

ResultsIn the liver, the event of incomplete cytokinesis istriggered by weaningWe set out to unveil when the incomplete cytokinesis processis initiated during postnatal liver growth. Mitotic events were

detected in vivo by a simultaneous nuclear, microtubule andplasma membrane labelling on liver tissue sections. In sucklingrats, all late-telophase hepatocytes presented a midbody and acell shape characteristic of cleavage furrow ingression (Fig.1A, left panel). In fed rats, we found that although some late-telophase hepatocytes were engaged in a normal cytokinesisprocess others presented a round shape with no midbody,indicating an absence of ingression (Fig. 1A, right panels). Weanalyzed whether this phenotype was correlated with weaning.From day 10 to day 25, we analyzed cytokinesis events; ratswere weaned at 19 days after birth. We clearly observed thatthe proportion of incomplete cytokinesis events increasedsignificantly 2 days after weaning (21-day-old rats, 10±1.3%),being maximal in 25-day-old rats (40±1.5%) (Fig. 1B). Wenext investigated why, after weaning, some hepatocytescompleted cytokinesis and others did not. Based on the locationof the blood vessels, hepatocytes of each liver lobule aredivided into two subpopulations: an upstream periportal (PP;in contact with afferent blood) and a downstream perivenous(PV; surrounding the centrolobular efferent vein) population.Hepatocytes located in the PP and PV zones of the liver lobule

Journal of Cell Science 120 (20)

Fig. 1. An incomplete mode of cytokinesis takes place in the liver,triggered by weaning. (A) Immunostaining for �-catenin (red) and �-tubulin (green) of liver sections (10- and 25-day-old rats). Images oftelophase were visualized taking into account condensed chromatinstaining (Hoechst 33342, blue). Immunostaining allowed us todistinguish between telophase that completed or did not completecytokinesis. The percentages of cells with complete cytokinesis areindicated. Bars, 3 �m. (B) Events of incomplete cytokinesis beforeand after weaning (fixed at 19 days, arrow). At each point, four ratswere independently analyzed. Percentages of complete andincomplete cytokinesis in telophase cells were calculated using �-catenin/�-tubulin/Hoechst immunostaining (see A). A total of 50telophase cells were analyzed per animal. Bars represent s.e.,P<0.0001.

Fig. 2. Events of incomplete cytokinesis occur both in periportal (PP)and periveneous (PV) hepatocytes. (A) Immunostaining forglutamine synthetase (GS; proximal PV hepatocyte staining, green,right panel), Pepck1 (proximal PP hepatocyte staining, green, leftpanel) and �-catenin (membrane labelling, red, both panels) wasperformed on rat liver sections (25-day-old rats). �-catenin/Hoechststaining allowed us to distinguish between telophase that completedcytokinesis (ingression of the membrane) and telophase that did notcomplete cytokinesis (no ingression of the membrane) (Hoechststaining in blue). Bars, 40 �m. (B) Percentage of telophase cells thatcompleted cytokinesis or did not were calculated in each region(proximal PP: GS-positive staining; proximal PV: Pepck1-positivestaining; distal PP/PV: GS/Pepck1-negative staining). A total of 50images of telophase were analyzed per animals (n=4). Bars represents.e., P<0.0001.

Jour

nal o

f Cel

l Sci

ence

3635Incomplete cytokinesis and liver growth

show remarkable differences in the levels andactivities of various enzymes and other proteins(Jungermann and Kietzmann, 1996). Bycomplementary immunolocalizations of PP andPV hepatocytes, we defined the distribution ofevents of complete/incomplete cytokinesisthroughout the entire lobule. We first confirmedthe presence of a proliferation gradient decreasingfrom the PP to PV regions (Fig. 2), as previouslydescribed (Gebhardt and Jonitza, 1991). However,we detected incomplete cytokinesis events both inthe PP and PV regions (Fig. 2), correlated with asimilar proportion of binucleated hepatocytes inthe two areas (data not shown). These resultsrepresent the first demonstration that anincomplete mode of cytokinesis takes place in theliver, initiated after weaning, with no zonationinside the hepatic lobule.

Incomplete cytokinesis is defined by anabsence of anaphase cell elongationTo further investigate the precise mechanism thatgives rise to tetraploid progenies, we culturedprimary hepatocytes from rat before and afterweaning and analyzed them by live-cell videomicroscopy. Under these conditions, hepatocytesdivided just once and a maximum mitotic index of10% is reached (Guidotti et al., 2003). Beforeweaning, time-lapse observations revealed that alldiploid hepatocytes progressed normally throughmitosis and gave rise solely to diploid progenies.As anaphase proceeded, cells elongated precedingfurrow formation and ingression (Fig. 3A andsupplementary material Movie 1). We nextmonitored hepatocyte division after weaning.Mitotic hepatocytes did not complete cytokinesisin 10±1.7% of cases (n=50) when isolated from21-day-old rats and in 28±3.6% of cases (n=50)when isolated from 25-day-old rats. Hepatocytesundergoing incomplete cytokinesis did not exhibitdynamic shape changes; there was no evidence offurrow ingression (Fig. 3A and supplementarymaterial Movie 2). To expand this finding, wemeasured cell elongation from metaphase totelophase (pole-to-pole distance) on hepatocytesisolated from 25-day-old rats. Elongation wasgreatly impaired during incomplete cytokinesis,being fourfold lower as compared with complete cytokinesis(Fig. 3B). In association with this defect, DNA-to-cortexdistance decreased (Fig. 3C); in fact, the DNA masses crushedon the cortical polar region from late anaphase to earlytelophase (supplementary material Movie 2). We confirmed thedefect in cell elongation in vivo using �-catenin staining (asshown in Fig. 1A). All hepatocytes in metaphase had the sameshape, representative of the cell rounding process (length:21.4±0.43 �m, width: 25.03±0.73 �m, n=50). In telophase,hepatocytes that completed cytokinesis had an elongated shape(length: 31±0.88 �m, width: 16.34±0.53 �m, n=50), whereashepatocytes that did not complete cytokinesis kept the sameshape as in metaphase (length: 22.51±0.75 �m, width:27.54±0.91 �m, n=50). We therefore concluded that the

physiological process of incomplete cytokinesis is associatedwith the absence of cell elongation.

Deficiency in actin cytoskeleton reorganization in cellsthat did not complete cytokinesisTo further investigate the cell elongation defect, we analyzedcytoskeleton rearrangements, which are crucial in determiningthe placement of the cleavage furrow (Murthy and Wadsworth,2005; Yumura, 2001). We examined how remodelling of F-actin was coordinated during cytokinesis. When hepatocytescompleted cytokinesis, the presence of an actin belt parallel tothe cleavage plane was observed in early telophase (Fig. 4A,left panels). By contrast, during the incomplete cytokinesisprocess, this structure was always absent (Fig. 4A, right

Fig. 3. Defect in anaphase cell elongation is observed when hepatocytes do notcomplete cytokinesis. (A) Hepatocytes were isolated from either 15- or 25-day-old rats. Mitosis of mononuclear 2n cells was monitored. Images are shown atselected time points (minutes). Complete cytokinesis: the cell divides successfully(supplementary material Movie 1, 15-day-old-rat, 100% of cytokinesis events).Incomplete cytokinesis: division leads to the genesis of a binucleated hepatocyte(supplementary material Movie 2, 25-day-old-rats, 28±3.6% of cytokinesisevents). Movies are representative of 50 cells analyzed in four independentcultures. The outlines show cell shape. (B) Cell elongation from metaphase totelophase was determined using time-lapse sequences. Time 0 correspond tochromosome alignment along the metaphase plate. Data are presented as themean±s.e. from 25 independent time-lapse sequences. (C) Chromosome-to-membrane distance was measured on the same time-lapse sequences presented inB. The histogram represents the average chromosome-to-membrane distance ofcomplete/incomplete cytokinesis. Data are presented as the mean±s.e. from 25independent time-lapse sequences, P<0.0001. Images shown are representative ofanaphase cells that will complete or not complete cytokinesis. The outlines showcell shape; arrows represent chromosome-to-membrane distance. Bars, 5 �m.

Jour

nal o

f Cel

l Sci

ence

3636

panels). We quantified pole-to-pole distribution of F-actinalong the cortex. We showed that, when hepatocytes completedcytokinesis, F-actin concentrated to the equatorial cortex,demonstrating redistribution of the protein during anaphase-to-telophase transition (Fig. 4B). By contrast, F-actin uniformlylocalized all along the cortex during the process of incomplete

cytokinesis (Fig. 4B). Because latrunculin A (LatA) preventsactin polymerization (Wakatsuki et al., 2001), we analyzed,after weaning, the effect of LatA treatment on the cellelongation defect. As described before, 30% of hepatocytesisolated from 25-day-old rats did not exhibit dynamic shapechanges; in the presence of LatA, all hepatocytes elongated(supplementary material Fig. S1). This result clearly suggeststhat, during the incomplete cytokinesis process, cell elongationis impaired because of an absence of actin cytoskeletonreorganization. In Drosophila cells, Rho kinase (ROCK) isessential for anaphase cell elongation (Dean and Spudich,2006; Hickson et al., 2006). In our system, we determinedwhether inhibition of ROCK had an effect on cell shape duringanaphase on primary cultures isolated from rats beforeweaning. Remarkably, treatment of hepatocytes with a specificROCK inhibitor resulted in a defect in cell elongation duringcytokinesis (supplementary material Movie 3). By measuringcell elongation between metaphase and telophase, we clearlyreproduced the same defect that we observed after weaning(Fig. 4C). ROCK is required for normal myosin II recruitmentto the equatorial cortex (Dean et al., 2005; Hickson et al., 2006;Straight et al., 2003). Equatorial myosin II accumulation,thereafter, derives turnover of actin filaments along the equatorto allow ingression to take place (Guha et al., 2005; Murthyand Wadsworth, 2005). We analyzed the localization of myosinII at the equatorial cortex before and after weaning. Ifhepatocytes completed cytokinesis, we always observed that,during anaphase, the phospho-regulatory light chain (phospho-RLC) of myosin II accumulated to the equatorial cortex, incontrast to hepatocytes that did not complete cytokinesis(genesis of binucleated hepatocytes) (Fig. 4D andsupplementary material Fig. S2). We conclude that theincomplete cytokinesis process is characterized by the absenceof actin cytoskeleton rearrangement.

Astral microtubules failed to contact the equatorialcortex during the incomplete cytokinesis processMitotic spindle microtubules deliver spatially restricted signalsto the cortex to promote furrowing and this occurs togetherwith the signalling pathway that regulates equatorial corticalactivity (D’Avino et al., 2005). Astral and central spindlemicrotubules are considered to be essential for cleavage-planespecification (Alsop and Zhang, 2003; Bringmann and Hyman,2005; Canman et al., 2003; Inoue et al., 2004). We thereforeexamined the behaviour of microtubules in our model. Beforeweaning, as hepatocytes proceeded through anaphase, stainingfor �-tubulin revealed the presence of anti-parallelmicrotubules (the central spindle) and microtubules towardsthe cell cortex in the furrow and polar regions (Fig. 5A). Intelophase, microtubules were compressed in the midzone as aconsequence of furrow ingression (Fig. 5A). After weaning,organization of the microtubules network was identical in allhepatocytes until early anaphase (data not shown). However,in late anaphase, we clearly noticed that 30% of hepatocytespresented a disrupted central spindle and astral equatorialmicrotubules, as well as reduced astral polar microtubules (Fig.5A). Analysis of �-tubulin fluorescence intensity at theequatorial region between anaphase and telophase revealed a1.6-fold increased labelling when hepatocytes completedcytokinesis, whereas no significant increase was observedwhen cells did not complete (supplementary material Fig. S3).

Journal of Cell Science 120 (20)

Fig. 4. Actin cytoskeleton rearrangement does not occur duringincomplete cytokinesis. (A) Hepatocytes were isolated from either15- or 25-day-old rats. Hepatocytes were stained with Alexa-Fluor-488-phalloidin (Actin) and nuclei with Hoechst (DNA). A total of200 cells from four independent experiments were analyzed. (B) TheAlexa-Fluor-488-phalloidin fluorescence profile was determinedalong the cortex in early telophase of complete (left) and incomplete(right) cytokinesis. Quantification was measured only on half of thecell, from one pole to the other (see upper scheme, red line). Thebroken vertical lines surround the equatorial area. A total of 50 earlytelophase cells were analyzed in four independent cultures. (C) Cellelongation from metaphase to telophase was determined using time-lapse sequences; hepatocytes were treated or not with HA1077 (15-day-old rats). Time 0 correspond to chromosome alignment along themetaphase plate. Data are presented as the mean±s.e. from tenindependent time-lapse sequences. (D) Hepatocytes were isolated asin A. Hepatocytes were stained with anti-phospho-Myosin lightchain 2 (Phospho-RLC) and nuclei with Hoechst (DNA). Earlytelophase cells are shown (n=200 from four independent cultures).Bars, 5 �m.

Jour

nal o

f Cel

l Sci

ence

3637Incomplete cytokinesis and liver growth

Because astral microtubules become stabilized upon contactingthe cortex (Burgess and Chang, 2005), we next analyzed theinteraction between astral microtubules and the equatorial

cortex. We demonstrated that during incomplete cytokinesis,astral microtubules failed to contact the equatorial cortex (Fig.5B). Furthermore, we analyzed localization of EB1, a proteinthat only associates with elongating microtubules and not withmicrotubules that become stabilized upon contact with thecortex (Strickland et al., 2005). We observed that the equatorialcortical EB1 staining was weak in hepatocytes that completedcytokinesis (Fig. 5C), reflecting the association of the majorityof astral equatorial microtubules to the cortex. By contrast,when cells did not complete cytokinesis, EB1 was still presenton microtubule tips, illustrating the absence of anchorage (Fig.5C). Taken together, these results demonstrate that, duringincomplete cytokinesis, the absence of astral microtubulesanchorage to the equatorial cortex induces a totaldestabilization of the microtubule network.

The equatorial zone of active RhoA is absent during theincomplete cytokinesis processWe next hypothesized that the signal delivered by microtubulesto the cortex must be impaired in this specific liver-divisionprogramme. Recent studies clearly demonstrate that proteinsthat localize to astral and/or central spindle microtubules aredelivered to the equatorial cortex in order to activate the RhoGTPase RhoA, which is indispensable to induce furrowing(Kamijo et al., 2006; Nishimura and Yonemura, 2006; Somersand Saint, 2003; Yoshizaki et al., 2004; Yuce et al., 2005). Inthis context, we analyzed the localization of MgcRacGAP(centralspindlin protein), Aurora B (chromosomal passengerprotein) and PRC1 (maintenance of central spindle) (Eggert etal., 2006). If hepatocytes completed cytokinesis, MgcRacGAPwas present during anaphase both on the central spindle andon astral equatorial microtubules and during telophase at themidbody (Fig. 6A). During incomplete cytokinesis,MgcRacGAP was localized on the remaining interdigitatingmicrotubules in anaphase B and telophase but was neverobserved on unattached astral equatorial microtubules (Fig.6A). The same profile was observed for other microtubule-associated proteins, such as aurora B, PRC1 and Plk1(supplementary material Fig. S4). These results argue the factthat, during the incomplete cytokinesis process, microtubule-dependent molecular signal is not delivered to the equatorialcortex. Finally, we investigated RhoA localization. Asdescribed before for other cell types (Yoshizaki et al., 2003;Yoshizaki et al., 2004), RhoA accumulated at the equatorial

cortex in early telophase in hepatocytesthat complete cytokinesis; then itconcentrated at the cleavage furrow andfinally at the midbody (Fig. 6B). Duringincomplete cytokinesis, RhoA did not

Fig. 5. Organization of astral and central spindle microtubulesduring incomplete cytokinesis. (A) Hepatocytes were isolated fromeither 15- or 25-day-old rats. Hepatocytes were stained with anti-�-tubulin and nuclei with Hoechst (DNA). A total of 200 cells fromfour independent experiments were analyzed. Bars, 5 �m. (B,C)Hepatocytes (25-day-old rats) were stained with anti-�-tubulin (B,microtubules, green) or anti-EB1 (C, microtubule tips, green) andanti-�-catenin (B,C, cell cortex, red). A total of 100 cells from fourindependent experiments were analyzed. (B) Images representmagnified equatorial regions of anaphase hepatocytes. Bars, 1 �m.(C) Nuclei were stained with Hoechst (blue). Boxed areas are shownbelow: higher-magnification of the equatorial cortical region. Bars,5 �m.

Fig. 6. Incomplete cytokinesis is characterizedby an absence of active RhoA localization atthe putative cleavage plane. (A,B) Hepatocyteswere isolated from 25-day-old rats.Hepatocytes were stained with anti-MgcRacGAP (A, red) and anti-�-tubulin (A,green) or with anti-RhoA (B, green)antibodies. Nuclei were stained with Hoechst33342 (A,B, blue). A total of 200 cells fromfour independent experiments were analyzed.Bars, 5 �m.

Jour

nal o

f Cel

l Sci

ence

3638

correctly localize at the equatorial cortex, in which astralmicrotubules were almost lost; we observed a diffuserepartition near the cell centre close to central spindlemicrotubules (Fig. 6B). This result is in agreement with otherstudies that have demonstrated that RhoA zones are diffusewhen microtubules are distant from the cortex (Bement et al.,2005; Nishimura and Yonemura, 2006). In conclusion, ourresults indicate that, during incomplete cytokinesis, molecularsignals delivered by microtubules to the equatorial cortex areimpaired, preventing the activation of RhoA pathway.

DiscussionIncomplete cytokinesis is increasingly recognized as animportant source of genomic instability taking place duringcell transformation (Fujiwara et al., 2005; Shi and King, 2005;Storchova and Pellman, 2004). Paradoxically, incompletecytokinesis is also associated with developmental programmesin different species (Glotzer, 2001). In this study, weidentified, in the liver, a new developmental process ofincomplete cytokinesis triggered by weaning. Observation ofhepatocytes during the binucleation process demonstrates that,although karyokinesis is normally accomplished, the cleavageplane is never specified in these cells. Cardiomyocytes are themost related physiological model that, like hepatocytes,perform karyokinesis but only partially assemble theactomyosin ring (Engel et al., 2006). Nevertheless, in oursystem, binucleation was never associated with mitoticabnormalities that lead to nuclear bridging and micronuclei;these tetraploid progenies will be able to execute a new celldivision cycle (Guidotti et al., 2003). It is now clear that theonset of animal cell cytokinesis must be precisely controlledin both time and space in order to promote correct furrowformation (D’Avino et al., 2005; Wang, 2005). Assembly anddynamic turnover of actin and myosin II at the equatorialcortex is required to induce cell shape modification. Byfollowing the divisions of living cells after weaning, weestablished that anaphase cell elongation is clearly impairedduring incomplete cytokinesis. This phenotype has beenconfirmed in vivo; telophase hepatocytes presenting nomembrane ingression kept the same shape as in metaphase.Moreover, we demonstrate that the actin cytoskeleton is notreorganized to the cleavage plane during anaphase-telophasetransition. In Drosophila cells, this process has been shown tobe controlled by the ROCK/myosin II pathway (Dean andSpudich, 2006; Hickson et al., 2006). Myosin II is recruitedto the equatorial cortex of the cell by a Rho kinase-dependentmechanism, where it will contribute to cell elongation througha broad equatorial elongation prior to the more-restrictedcontraction of the actomyosin ring. We demonstrate that thispathway is conserved in mammalian cells. In our model,ROCK activity is clearly impaired after weaning; phospho-RLC is not recruited to the equatorial cortex leading to theabsence of actin cytoskeleton reorganization. From studies inmany laboratories, it is clear that communications betweenmicrotubules and the actin cortex somehow direct theactivation of RhoA in a precisely defined zone at the equator,promoting rapid remodelling of the cortical actomyosincytoskeleton (Kamijo et al., 2006; Nishimura and Yonemura,2006; Somers and Saint, 2003; Yoshizaki et al., 2004; Yuce etal., 2005). In particular, astral microtubules emanating fromthe spindle poles and central spindle microtubules between

separating chromosomes are essential for the activation ofequatorial cortical regions (D’Avino et al., 2005). Indeednumerous signalling proteins – including the centralspindlincomplex MKLP1 and MgcRacGAP, the passenger proteinAurora B, and the Rho GTPase exchange factor ECT2 – areinvolved in relaying information from the spindle to the cortexand ultimately anchor the crucial cytokinesis activator RhoA(Eggert et al., 2006). We show that, in hepatocytes that do notcomplete cytokinesis, microtubule network behaviour isnormal until early anaphase; thereafter, elongatingmicrotubules fail to interact with the equatorial cortex. As aconsequence, the spindle collapses in telophase. Consistentwith these results, we demonstrate that the microtubulemolecular signal essential for furrow induction is not deliveredto the equatorial cortex. Consequently, active RhoA does notconcentrate at the putative site of furrow formation, leading toan absence of activation of its downstream signals. Finally, itis quiet captivating that a specific cell division programme canbe controlled by animals weaning. In the liver, weaning isthe trigger of complex physiological changes, such asnutriment/ hormonal balance and modification of thecircadian cycle (Gupta, 2000). Further studies will be neededto show whether one or several of these factors could controlastral equatorial cortical capture by modifying either theintegrity of plus-tips proteins, which are essential to mediatecortical interactions, and/or cell membrane composition,preventing anchorage of astral equatorial microtubules to thecortex.

Materials and MethodsAnimalsMale Wistar rats (IFFACREDO, France) were treated in accordance with EuropeanUnion regulations on animal care. The rats were housed under standard light/darkconditions and received pelleted food and water ad libitum. All rats were weanedat 19 days after birth.

Cell cultures and inhibitorHepatocytes were isolated from rat livers by a two-step perfusion as previouslydescribed (Guidotti et al., 2003). Latrunculin A (Calbiochem) was used at 2 �M;HA1077 (Sigma-Aldrich) was used at 10 �M.

AntibodiesCommercial primary antibodies used were as follows: mouse anti-�-catenin (BDTransduction Laboratories, 1:200), mouse anti-�-tubulin (Tub 2.1, Sigma, 1:400),mouse anti-RhoA (Santa Cruz, 1:100), mouse anti-EB1 (BD TransductionLaboratories, 1:100), mouse anti-GS (BD Transduction Laboratories, clone 6,1:200), mouse anti-phospho-Myosin light chain 2 (nonmuscle) (3675, CellSignaling, 1:50), mouse anti-AIM-1 (BD Transduction Laboratories, 1:100), mouseanti-Plk1 (Euromedex-upstate, clone 35-206, 1:200), rabbit anti-Pepck1 (fromLamers, AMC Liver Center, Amsterdam, The Netherlands, 1:1000), rabbit anti-MgcRacGAP (from T. Kitamura, Institute of Medical Research Science, Tokyo,Japan, 1:100), rabbit anti-PRC1 (from W. Jiang, Burnham Institute for MedicalResearch, La Jolla, CA, 1:100) and 165 nM Alexa-Fluor-488-phalloidin (Molecularprobes). Secondary antibodies to rabbit and mouse IgG were conjugated either withAlexa-Fluor-488 or Alexa-Fluor-594 (Molecular probes, 1:500).

Live imagingHepatocytes were grown on 35�10 mm coverslips coated with collagen solution(Sigma) and mounted on the microscope after 60 hours of culture. During imaging,hepatocytes were on a stage heated at 37°C under a 5% CO2 atmosphere. Cells werefilmed every 90 seconds with a Leica DMIRBE using a 63� lens (numericalaperture, 1.4), a condenser (working distance, 23 mm; numerical aperture, 0.53) anda Pentamax cooled CCD camera (Popper Scientific) coupled to an electronic shutter.Metamorph 7.1 was used for computer-based image acquisition and analysis of livecell data. The single images shown were prepared using Adobe Photoshop CS.

ImmunohistochemistryLiver tissues fixed in 10% phosphate buffered formalin were embedded in paraffin.Tissue sections (3 �m) were obtained using a conventional microtome. Sections

Journal of Cell Science 120 (20)

Jour

nal o

f Cel

l Sci

ence

3639Incomplete cytokinesis and liver growth

were deparaffinized in xylene and placed in 100% ethanol. Sections were rehydratedin a descending gradient of ethanol-water and then boiled for 2�5 minutes in Tris0.1 M Tween 0.1%. All subsequent antibody incubations were carried out asdescribed below.

ImmunofluorescenceCoverslips containing hepatocytes were collected and washed in PBS. The cellswere fixed either in–20°C MetOH for 3 minutes (to stain �-tubulin, �-catenin, EB1,MgcRacGAP, Aurora B, Prc1 and Plk1), or in 4°C 10% TCA for 15 minutes (tostain RhoA), or in 4°C 4% PFA for 15 minutes (to stain actin and phospho-RLC).After blocking in PBS with 10% goat serum (30 minutes), cells were incubated for1 hour with primary antibodies, washed in PBS containing 0.1% Tween 20 andincubated with secondary antibodies for 30 minutes, all at room temperature.Hoechst 33342 (0.2 �g/ml, Sigma) was included in the final wash to counterstainnuclei. Samples were mounted on slides in Fluorescent mounting medium.

Image acquisition and analysisImages were taken using a Nikon Statif Eclipse E600 microscope with 60�magnification, 1.4-0.7 NA PL-APO objectives, a DXM1200 cooled CCD camera(Nikon) and ACT-1 (Universal Imaging). To measure the fluorescence intensities oftubulin, actin and myosin, z-axis stacks were collected using a piezoelectric devicemounted at the base of a 63� magnification, a 1.4 NA PL-APO objective on a ZeissDMRA2 microscope and a Coolsnap HQ camera controlled by Metamorph software(Universal Imaging). A total of 20-30 planes (0.2 �m slice) were captured for eachcell and compiled as single two-dimensional projections using ImageJ software. Allimages were imported into Adobe Photoshop CS for contrast manipulation andfigure assembly. To measure actin and myosin fluorescence intensity, Linescanfunction of Metamorph software was used.

The authors would like to thank A. Echard for his critical evaluationof this manuscript and all members of the lab for fruitful discussion.We also thank the imaging facility at the Jacques Monod Institute(Paris, France). G.M.-D. was supported by a doctoral fellowship fromthe Research Minister. This study was supported by grants fromInstitut National de la Santé et de la Recherche Médicale (INSERM),by ARC3259, by la Ligue Comité de Paris (RS06/75-57) and byANR-05-JCJC-0168-01.

ReferencesAlsop, G. B. and Zhang, D. (2003). Microtubules are the only structural constituent of

the spindle apparatus required for induction of cell cleavage. J. Cell Biol. 162, 383-390.Bement, W. M., Benink, H. A. and von Dassow, G. (2005). A microtubule-dependent

zone of active RhoA during cleavage plane specification. J. Cell Biol. 170, 91-101.Bringmann, H. and Hyman, A. A. (2005). A cytokinesis furrow is positioned by two

consecutive signals. Nature 436, 731-734.Burgess, D. R. and Chang, F. (2005). Site selection for the cleavage furrow at

cytokinesis. Trends Cell Biol. 15, 156-162.Burgos, M. H. and Fawcett, D. W. (1955). Studies on the fine structure of the mammalian

testis. I. Differentiation of the spermatids in the cat (Felis domestica). J. Biophys.Biochem. Cytol. 1, 287-300.

Canman, J. C., Cameron, L. A., Maddox, P. S., Straight, A., Tirnauer, J. S.,Mitchison, T. J., Fang, G., Kapoor, T. M. and Salmon, E. D. (2003). Determiningthe position of the cell division plane. Nature 424, 1074-1078.

D’Avino, P. P., Savoian, M. S. and Glover, D. M. (2005). Cleavage furrow formationand ingression during animal cytokinesis: a microtubule legacy. J. Cell Sci. 118, 1549-1558.

Dean, S. O. and Spudich, J. A. (2006). Rho kinase’s role in myosin recruitment to theequatorial cortex of mitotic Drosophila S2 cells is for myosin regulatory light chainphosphorylation. PLoS ONE 1, e131.

Dean, S. O., Rogers, S. L., Stuurman, N., Vale, R. D. and Spudich, J. A. (2005).Distinct pathways control recruitment and maintenance of myosin II at the cleavagefurrow during cytokinesis. Proc. Natl. Acad. Sci. USA 102, 13473-13478.

Eggert, U. S., Mitchison, T. J. and Field, C. M. (2006). Animal cytokinesis: from partslist to mechanisms. Annu. Rev. Biochem. 75, 543-566.

Engel, F. B., Schebesta, M. and Keating, M. T. (2006). Anillin localization defect incardiomyocyte binucleation. J. Mol. Cell. Cardiol. 41, 601-612.

Fujiwara, T., Bandi, M., Nitta, M., Ivanova, E. V., Bronson, R. T. and Pellman, D.(2005). Cytokinesis failure generating tetraploids promotes tumorigenesis in p53-nullcells. Nature 437, 1043-1047.

Ganem, N. J., Storchova, Z. and Pellman, D. (2007). Tetraploidy, aneuploidy andcancer. Curr. Opin. Genet. Dev. 17, 157-162.

Gebhardt, R. and Jonitza, D. (1991). Different proliferative responses of periportaland perivenous hepatocytes to EGF. Biochem. Biophys. Res. Commun. 181, 1201-1207.

Glotzer, M. (2001). Animal cell cytokinesis. Annu. Rev. Cell Dev. Biol. 17, 351-386.Guha, M., Zhou, M. and Wang, Y. L. (2005). Cortical actin turnover during cytokinesis

requires myosin II. Curr. Biol. 15, 732-736.Guidotti, J. E., Bregerie, O., Robert, A., Debey, P., Brechot, C. and Desdouets, C.

(2003). Liver cell polyploidization: a pivotal role for binuclear hepatocytes. J. Biol.Chem. 278, 19095-19101.

Gupta, S. (2000). Hepatic polyploidy and liver growth control. Semin. Cancer Biol. 10,161-171.

Hickson, G. R., Echard, A. and O’Farrell, P. H. (2006). Rho-kinase controls cell shapechanges during cytokinesis. Curr. Biol. 16, 359-370.

Inoue, Y. H., Savoian, M. S., Suzuki, T., Mathe, E., Yamamoto, M. T. and Glover, D.M. (2004). Mutations in orbit/mast reveal that the central spindle is comprised of twomicrotubule populations, those that initiate cleavage and those that propagate furrowingression. J. Cell Biol. 166, 49-60.

Jungermann, K. and Kietzmann, T. (1996). Zonation of parenchymal andnonparenchymal metabolism in liver. Annu. Rev. Nutr. 16, 179-203.

Kamijo, K., Ohara, N., Abe, M., Uchimura, T., Hosoya, H., Lee, J. S. and Miki, T.(2006). Dissecting the role of Rho-mediated signaling in contractile ring formation.Mol. Biol. Cell 17, 43-55.

Kudryavtsev, B. N., Kudryavtseva, M. V., Sakuta, G. A. and Stein, G. I. (1993).Human hepatocyte polyploidization kinetics in the course of life cycle. Virchows Arch.B Cell Pathol. 64, 387-393.

Mullins, J. M. and Biesele, J. J. (1977). Terminal phase of cytokinesis in D-98s cells.J. Cell Biol. 73, 672-684.

Murthy, K. and Wadsworth, P. (2005). Myosin-II-dependent localization and dynamicsof F-actin during cytokinesis. Curr. Biol. 15, 724-731.

Nishimura, Y. and Yonemura, S. (2006). Centralspindlin regulates ECT2 and RhoAaccumulation at the equatorial cortex during cytokinesis. J. Cell Sci. 119, 104-114.

Ravid, K., Lu, J., Zimmet, J. M. and Jones, M. R. (2002). Roads to polyploidy: themegakaryocyte example. J. Cell. Physiol. 190, 7-20.

Seglen, P. O. (1997). DNA ploidy and autophagic protein degradation as determinants ofhepatocellular growth and survival. Cell Biol. Toxicol. 13, 301-315.

Shi, Q. and King, R. W. (2005). Chromosome nondisjunction yields tetraploid ratherthan aneuploid cells in human cell lines. Nature 437, 1038-1042.

Sigal, S. H., Rajvanshi, P., Gorla, G. R., Sokhi, R. P., Saxena, R., Gebhard, D. R., Jr,Reid, L. M. and Gupta, S. (1999). Partial hepatectomy-induced polyploidy attenuateshepatocyte replication and activates cell aging events. Am. J. Physiol. 276, G1260-G1272.

Somers, W. G. and Saint, R. (2003). A RhoGEF and Rho family GTPase-activatingprotein complex links the contractile ring to cortical microtubules at the onset ofcytokinesis. Dev. Cell 4, 29-39.

Storchova, Z. and Pellman, D. (2004). From polyploidy to aneuploidy, genomeinstability and cancer. Nat. Rev. Mol. Cell Biol. 5, 45-54.

Straight, A. F., Cheung, A., Limouze, J., Chen, I., Westwood, N. J., Sellers, J. R. andMitchison, T. J. (2003). Dissecting temporal and spatial control of cytokinesis with amyosin II Inhibitor. Science 299, 1743-1747.

Strickland, L. I., Wen, Y., Gundersen, G. G. and Burgess, D. R. (2005). Interactionbetween EB1 and p150glued is required for anaphase astral microtubule elongationand stimulation of cytokinesis. Curr. Biol. 15, 2249-2255.

Toyoda, H., Bregerie, O., Vallet, A., Nalpas, B., Pivert, G., Brechot, C. and Desdouets,C. (2005). Changes to hepatocyte ploidy and binuclearity profiles during humanchronic viral hepatitis. Gut 54, 297-302.

Wakatsuki, T., Schwab, B., Thompson, N. C. and Elson, E. L. (2001). Effects ofcytochalasin D and latrunculin B on mechanical properties of cells. J. Cell Sci. 114,1025-1036.

Wang, Y. L. (2005). The mechanism of cortical ingression during early cytokinesis:thinking beyond the contractile ring hypothesis. Trends Cell Biol. 15, 581-588.

Weaver, B. A. and Cleveland, D. W. (2006). Does aneuploidy cause cancer? Curr. Opin.Cell Biol. 18, 658-667.

Yoshizaki, H., Ohba, Y., Kurokawa, K., Itoh, R. E., Nakamura, T., Mochizuki, N.,Nagashima, K. and Matsuda, M. (2003). Activity of Rho-family GTPases during celldivision as visualized with FRET-based probes. J. Cell Biol. 162, 223-232.

Yoshizaki, H., Ohba, Y., Parrini, M. C., Dulyaninova, N. G., Bresnick, A. R.,Mochizuki, N. and Matsuda, M. (2004). Cell type-specific regulation of RhoAactivity during cytokinesis. J. Biol. Chem. 279, 44756-44762.

Yuce, O., Piekny, A. and Glotzer, M. (2005). An ECT2-centralspindlin complexregulates the localization and function of RhoA. J. Cell Biol. 170, 571-582.

Yumura, S. (2001). Myosin II dynamics and cortical flow during contractile ringformation in Dictyostelium cells. J. Cell Biol. 154, 137-146.

Jour

nal o

f Cel

l Sci

ence