developmental biology 194

DEVELOPMENTAL BIOLOGY 194, 47–60 (1998)ARTICLE NO. DB978815

Requirement for Microtubules in New MembraneFormation during Cytokinesis of Xenopus Embryos

M. V. Danilchik,* W. C. Funk,† E. E. Brown,* and K. Larkin‡*Department of Biological Structure and Function, Oregon Health Sciences University,Portland, Oregon 97201-3097; †Division of Biological Sciences, University of Montana,Missoula, Montana 59812; and ‡Department of Cell and Developmental Biology, OregonHealth Sciences University, Portland, Oregon 97201-3097

In cleaving Xenopus eggs, exposure to nocodazole or cold shock prevents the addition of new plasma membrane to thecleavage plane and causes furrows to recede, suggesting a specific role for microtubules in cytokinesis. Whole-mountconfocal immunocytochemistry reveals a ring of radially arranged, acetylated microtubule bundles at the base of alladvancing cleavage furrows, from the first cleavage through the midblastula stage. We hypothesize that this novel microtu-bular structure is involved in transporting maternal stores of membrane in the subcortex to a site of membrane additionnear the leading edge of the furrow. q 1998 Academic Press

Key Words: cell division; cleavage furrow; confocal microscopy; cytokinesis; egg; embryo; membrane; microtubule;midbody; midzone; Xenopus.

INTRODUCTION amphibian eggs is inserted in large amounts along the newbasolateral surfaces between dividing blastomeres (Bluemink

In animal cells, cytokinesis begins in late anaphase with and deLaat, 1973). The new plasma membrane has a compo-the formation of a thickened cortical band of microfila- sition different from that of the original egg surface (Bieliav-ments, the contractile ring (Schroeder, 1970, 1972; Szollosi, sky et al., 1992; Byers and Armstrong, 1986; Kalt, 1971;1970). Constriction by the contractile ring produces a cleav- Sanders and Singal, 1975). The source of the new membraneage furrow that advances inward to bisect the cell in a plane includes a pool of post-Golgi vesicles produced during oo-perpendicular to the mitotic spindle. This precise geometry genesis, contributing membrane lipids, glycoproteins, andhas long been thought to depend on interactions between extracellular matrix components to the growing surfacethe astral microtubules and the cortex (Rappaport and Rap- (Leaf et al., 1990; Servetnick et al., 1990). Although basolat-paport, 1974), although recent studies suggest that the posi- erally targeted membrane glycoproteins are known to ap-tion of the contractile ring is defined via localized interac- pear very quickly in newly deposited membranes (Choi ettions between telophase disk components and the cortex al., 1990; Gawantka et al., 1992; Roberts et al., 1992), the(Earnshaw and Cooke, 1991; Eckley et al., 1997; Margolis actual site or sites of vesicle fusion within the new mem-and Andreassen, 1993). It is generally thought that, once brane domain have not been identified.established, the contractile ring can function without mi- The cytoskeletal and molecular events regulating mem-crotubules, as suggested by experiments in which colchi- brane growth in the furrow are not currently understood.cine-treated sea urchin embryos complete cell division (Ha- In Xenopus, membrane addition readily proceeds in the ab-maguchi, 1975). However, microtubule-depolymerizing sence of a contractile ring following disruption with cyto-agents applied after furrowing has begun can block its com- chalasin B (Bluemink, 1971; Bluemink and deLaat, 1973) orpletion in amphibian eggs (Sawai and Yomota, 1990) and in interference with actin-regulating rho-family GTPase func-tissue culture cells (Cao and Wang, 1996; Wheatley and tion (Drechsel et al., 1997). The observation that furrowsWang, 1996). Nevertheless, no specific role for microtubules in Xenopus and Cynops eggs regress following treatmentin the late events of cytokinesis has been defined. with microtubule-disrupting drugs (Sawai, 1992; Sawai and

The rapid increase in surface area accompanying cell divi- Yomota, 1990) suggests the possibility that microtubulesare involved in membrane addition during cytokinesis.sion creates a demand for new plasma membrane, which in

47

0012-1606/98 $25.00Copyright q 1998 by Academic PressAll rights of reproduction in any form reserved.

AID DB 8815 / 6x37$$$$41 01-23-98 13:24:57 dbas

48 Danilchik et al.

were individually cold shocked as above at approximately 10-minMicrotubules have already been shown to be involved inintervals, returned to 207C for 2 min, and then scored via stereomi-a number of important events during early Xenopus devel-croscopy to determine the proportion undergoing rupture. To mini-opment. In addition to their role in meiosis and mitosis,mize variation in exposure times, transfers were performed withmicrotubules are implicated in processes that result in spa-prechilled Pasteur pipets.tial organization of the embryo, including transport to the

oocyte cortex of RNAs important for embryonic patterning Fixation and Whole-Mount Immunocytochemistry(Yisraeli et al., 1990), gathering and internalizing of germ

Procedures for examining microtubules in cleaving Xenopus eggsplasm during the first two cell cycles (Ressom and Dixon,resembled protocols developed by Gard (1993) for Xenopus oocytes,1988; Savage and Danilchik, 1993), as well as rotation ofexcept that Taxol was omitted from the fixative. Thus, embryos

the vegetal yolk mass to set up the dorsal–ventral axis (Elin- were fixed in 3.7% formaldehyde (Mallinckrodt), 0.25% glutaralde-son and Rowning, 1988). The possibility that microtubules hyde (Polysciences), and 0.2% Triton X-100 in microtubule assem-also have a role in delivery of vesicles to growing cleavage bly buffer (80 mM K-Pipes, pH 6.8, 5 mM EGTA, 1 mM MgCl2) for 2membranes would provide another example of their sig- to 4 h and then postfixed in 0207C methanol overnight. Pigmentednificance in organizing the embryonic body plan, since all embryos were bleached in a solution of 10% hydrogen peroxide

and 67% methanol for 1 to 2 h on a light table. Bleached embryosreceptor-mediated, cell–cell interactions depend on thewere then rehydrated with three 10-min rinses in phosphate-buf-composition of the membranes deposited between cells. Infered saline (PBS: 128 mM NaCl, 2 mM KCl, 8 mM NaH2PO4, 2 mMthis study, we use confocal microscopy to examine the cy-KH2PO4, pH 7.2) and incubated for 6 to 16 h at room temperature intoskeletal structures associated with cleavage furrows and100 mM NaBH4 in PBS to reduce unreacted aldehydes (Gard, 1991).find a distinctive, ring-shaped array of microtubule bundlesSpecimens were next washed with five half-hour rinses in Tris-

in the cytoplasmic bridge that may function to recruit vesi- buffered saline (TBS: 155 mM NaCl, 10 mM Tris–Cl, pH 7.4, 0.1%cles toward a site of fusion near the base of the cleavage Nonidet P-40). To facilitate staining of deeper regions, embryosfurrow. were bisected parallel or perpendicular to selected cleavage planes

with a sharp razor blade fragment prior to antibody treatment. Spec-imens were incubated overnight at 47C with mouse monoclonalantibodies directed against ab-tubulin (East Acres Biologicals) orMATERIALS AND METHODSacetylated tubulin (6-11B-1; Sigma). Controls included mousemonoclonal IgGs directed against irrelevant antigens. All antibod-Eggs and Embryosies were diluted in TBS containing 10% fetal bovine serum and 5%DMSO. After primary antibody incubations, embryos were washedAdult Xenopus laevis females were induced to ovulate by in-

jecting human chorionic gonadotropin (Sigma, 800 units per frog) with five 1-h rinses in TBS, incubated overnight at 47C with TRITC-conjugated goat anti-mouse IgG (Sigma), and finally washed withinto the dorsal lymph sac 18 h (at 167C) prior to use. Eggs were

stripped into a petri dish and fertilized by adding macerated testis five 1-h rinses in TBS. Embryos were dehydrated in methanol andcleared in Murray’s Clear (benzyl benzoate:benzyl alcohol, 2:1)and MMR/3 (MMR: 100 mM NaCl, 1.8 mM KCl, 2.0 mM CaCl2,

1.0 mM MgCl2, 5 mM Hepes, pH 7.5). Fifteen minutes after fertil- (Klymkowsky and Hanken, 1991).ization the eggs were dejellied in 2.5% cysteine in MMR/3 at pH8.0. Eggs were then washed in four rinses of MMR/3 and cultured Confocal Microscopyin the same medium at room temperature (18–217C). Live and fixed specimens were examined with a Bio-Rad MRC-

500 confocal laser-scanning microscope, mounted on a Zeiss Axiov-ert 10 inverted compound microscope equipped with a rotatingCytoskeletal Disruption with Mitotic Inhibitors stage and 10X/0.3 and 63X/1.25 Plan-Neofluar objectives. A Krohn-Hite 3202 tunable high-/low-pass filter (set to cut off at about 100Stock solutions of nocodazole, cytochalasin B, and cytochalasinkHz) was connected in-line between the scan head and the distribu-D (Sigma) in DMSO were diluted to concentrations of 10 mg/ml intion board to improve the signal to noise ratio. Fluorescent emis-deionized water or MMR/3 and applied to embryos at specifiedsions were detected with a rhodamine (GHS) filter set. Most imagestimes during early development. Paclitaxel (‘‘Taxol’’; Sigma) waswere obtained as projected stacks of Kalman averages of five slowdissolved in DMSO at 10 mg/ml, and 0.5 nl was then microinjectedscans. For stereo pairs and rotated projections, vertical stacks of 20into the animal hemisphere prior to appearance of the first or sec-to 50 optical slices were collected at approximately 0.2-mm inter-ond cleavage furrow. D2O (Sigma) was applied by incubating em-vals and projected with {107 rotation about the Y axis with NIHbryos at specified times in 50% D2O in MMR/3. Effects on cleavageImage version 1.58/ (developed at the U.S. NIH; available on thewere recorded via direct observation, time-lapse video microscopy,World Wide Web at http://rsb.info.nih.gov/nih-image).or whole-mount immunocytochemistry.

RESULTSCold Shock

Radially Arranged Microtubule Bundles Develop inCold-induced depolymerization of microtubules was performed the Cleavage Furrows at Time and Site of New

by transferring embryos to a dish of 47C MMR/3 for 4 min (Vincent Membrane Assemblyand Gerhart, 1987), after which they were transferred to either 207CWe used an antibody to tubulin in conjunction withculture medium for further culturing or fixed in 4 or 207C fixative.

In cold-induced rupture experiments, cohorts of 20–25 embryos whole-mount confocal immunofluorescence microscopy to

Copyright q 1998 by Academic Press. All rights of reproduction in any form reserved.

AID DB 8815 / 6x37$$$$41 01-23-98 13:24:57 dbas

49Microtubules in Xenopus Cleavage Furrows

FIG. 1. Microtubule bundles in cleavage furrow. Whole-mount confocal scanning microscope views of the cytoplasmic bridge duringfirst cleavage of Xenopus embryos immunostained for microtubules. (A) Embryo, fixed 20 min after appearance of first cleavage furrow,fractured parallel to the cleavage plane. Black arrow: dark line is accumulation of pigment at the boundary between original, pigmentedegg surface and new membrane domain. White arrow: furrow microtubule array (FMA). Scale bar Å 250 mm. (B) Higher magnificationdetail of embryo shown in A, showing radial arrangement of microtubule bundles surrounding cytoplasmic bridge. Bar Å 50 mm. (C)Embryo, fixed 24 min after appearance of first furrow, fractured perpendicular to cleavage plane. Bar Å 50 mm. (D) Rotated projection ofimage stack, consisting of 30 optical sections through a portion of the FMA nearest the animal pole. View is analogous to an optical planetangent to the cytoplasmic bridge. Bar Å 15 mm.

search for microtubules in the cleavage plane and discov- In amphibian embryos, cleavage is unipolar: furrowingbegins at the animal pole with the appearance of stress foldsered an unusual microtubule-containing structure near the

advancing edge of the furrow. We will herein refer to this that mark the development of a contractile actomyosinband in the cortex. As the furrow advances around the sur-structure as the furrow microtubule array (FMA). Figure 1

shows this structure in the cytoplasmic bridge in embryos face of the egg toward the equator, it begins deepening nearthe animal pole. About 9 min after furrow initiation, a do-fixed 20–24 min after initiation of the first cleavage furrow.

In a view parallel with the first cleavage plane (Figs. 1A and main of new surface membrane begins to form between thenascent blastomeres (Bluemink and deLaat, 1973). As noted1B), a circular array of short (Ç15–40 mm), thick microtu-

bule bundles is seen to trail behind the advancing edge of by Byers and Armstrong (1986) and Bieliavsky and Geuskens(1990), this new domain remains separate from the preex-the furrow. In a section perpendicular to the cleavage plane

(Fig. 1C), these bundles are seen to form a V-shaped array, isting egg surface. In the fixed embryos shown in Fig. 1A,the boundary between new and old domains can be seen aswith the apex a few microns in advance of the furrow’s

leading edge. As shown in Fig. 1D, most of the microtubule a dark, pigmented line at the base of the stress folds (blackarrow). From examination of eggs fixed at successive 4-minbundles do not extend across the apex; rather, their ends

interdigitate at the furrow’s leading edge, forming a discon- intervals during the second cell cycle, it is apparent thatthe FMA appears at about the same time that the new mem-tinuous staining pattern. Some bundles do span the furrow;

these lie generally deeper in the cytoplasm than do the dis- brane domain begins growing. As the new domain beginsexpanding, radially aligned microtubule-containing bundlescontinuous bundles.


AID DB 8815 / 6x37$$$$41 01-23-98 13:24:57 dbas

50 Danilchik et al.

appear at the cytoplasmic bridge (white arrow). The circular Numerous stress folds are generated in this region by thepolarized contraction of the cortex. There are relatively fewarray of microtubule bundles shrinks in diameter, accompa-

nying the closure of the cytoplasmic bridge. Thus, the mi- microtubules in the immediate vicinity of the stress folds.In contrast, thick bundles of midzone microtubules aboundcrotubule bundles are always found at the leading edge of

the rapidly expanding domain of new surface membrane, a in the cytoplasmic bridge, deep to the advancing furrow.Many of these bundles are branched or tufted at their distaltantalizing arrangement that suggests a functional role for

microtubules in membrane addition. ends (arrowheads), and most are roughly parallel with thespindle axis. With the assembly and passage of the contrac-To observe the initial assembly of microtubule bundles

in the vicinity of the furrow, we examined stereo pairs gen- tile ring, however, the bundles appear to be dragged into anew, radial alignment (arrow). These results suggest thaterated from serial confocal image stacks taken of specimens

fixed shortly after cleavage began. About 3 min after the some FMA microtubules originate as bundles in the mid-zone. However, based on the apparent lack of midzone mi-furrow appears (Fig. 2A), a distinctive microtubule-free cor-

tical zone, about 5 mm thick, lies directly beneath the fur- crotubules in the vegetal hemisphere (see Figs. 3 and 4) andalso the cold shock experiments described below, the FMArow and extends into the cortex underlying the stress folds

on either side. Below the cortex at the base of the furrow, fibers can form de novo.numerous thick, relatively short (Ç10–40 mm) microtubulebundles form a loose, tangled meshwork that extends

Microtubules Are Required for Furrowing andthrough a localized region some 70 mm below the surface.Membrane Deposition during CleavageThese bundles can be readily distinguished from the longer,

thinner, nearly parallel polar microtubule bundles deeper We used video time-lapse to record effects of a microtu-bule-disrupting agent, nocodazole, on cleavage in Xenopusin the cytoplasm of the incipient cleavage plane. Some of

the thick bundles are branched or tufted and appear to be embryos. When applied up to about 0.95 of the first cellcycle, i.e., before appearance of the cleavage furrow, nocoda-continuous with spindle fibers, but in the immediate vicin-

ity of the prospective cleavage plane, many have oblique zole completely blocked cytokinesis and no furrow devel-oped, consistent with the well-established idea that the mi-orientations (arrowhead) relative to the spindle axis. By

about 9 min after appearance of the furrow (Fig. 2B), about totic apparatus is required for defining the position of thecontractile ring. When nocodazole was applied after appear-the time that membrane addition begins, the meshwork of

microtubule bundles directly under the surface of the fur- ance of the cleavage furrow, the leading edges of the furrow,as manifested by shallow stress folds in the surface, contin-row has condensed considerably and has aligned to form

the thick, V-shaped FMA. Below the FMA, an array of Ç20- ued to progress around the egg surface for some distance.However, in the wake of the relaxing stress folds, the furrowmm-long, parallel microtubule bundles with a characteristic

central discontinuity develops in the prospective cleavage failed to deepen and quickly subsided, leaving behind distur-bances in the cortical pigmentation (Fig. 7). No new mem-plane (Fig. 2B, arrowheads). These bundles will be referred

to provisionally as midzone microtubules to reflect their brane domain developed.When microfilament-disrupting agents, cytochalasin B orresemblance to the overlapping, antiparallel interpolar bun-

dles often seen in the midzone in various somatic cells cytochalasin D, were applied to cleaving embryos, stressfolds quickly relaxed and furrows subsided. As shown pre-during late anaphase or telophase.

In the animal hemisphere, the midzone array extends viously (Bluemink and deLaat, 1973), a domain of unpig-mented membrane then began growing at the original site ofthrough a region several hundred microns in advance of the

furrow (Figs. 3A and 3B, mz). In the vegetal hemisphere, the furrow (Fig. 8). This domain of new membrane increasedsteadily in area for several cell-division cycles, eventuallya similar, but much less extensive array of microtubules

develops in the midzone in advance of the furrow base covering nearly half of the animal hemisphere surface.To test whether this localized addition of new membrane(Fig. 4).

Circular FMAs were present in every cleavage plane ex- in cytochalasin-treated embryos is microtubule-dependent,we applied nocodazole at different times following cytocha-amined up to the midblastula stage (through stage 7). In

contrast, after the midblastula stage, slender midbodies, re- lasin D treatment. As shown in Table 1, nocodazole blockedappearance of the new membrane domain, particularlysembling those of somatic cells, develop in the cytoplasmic

bridge (Fig. 5). when applied soon after furrow initiation. Time-lapse anal-ysis (not shown) indicated that this blockage of membraneThe sequence shown above in Fig. 2 suggests that the

furrow microtubule bundles form from midzone microtu- growth occurred within minutes of application of nocoda-zole. These results confirm that microtubules are involvedbules concentrated and aligned by the action of the contrac-

tile ring. This possibility was investigated by examining in depositing new membrane between blastomeres.The fact that new membrane delivery continues activelythe appearance of microtubules in a transition zone at the

leading edge of the first cleavage furrow. In the stereo pair in cytochalasin-treated embryos prompted us to examinethe region of membrane insertion for the presence of micro-shown in Fig. 6, one leading edge of the furrow is seen in its

advance from the upper right toward the lower left corner. tubules. As shown in Fig. 9, a prominent FMA persists in


AID DB 8815 / 6x37$$$$41 01-23-98 13:24:57 dbas


FIG. 2. Stereo pairs showing development of the microtubule array shortly after the initial appearance of the first cleavage furrow.Projections were generated from confocal image stacks as described under Materials and Methods. Image planes are perpendicular to thatof cleavage. (A) Embryo fixed 3 min after furrow initiation. Note that several short microtubule bundles below furrow are orientedobliquely relative to the spindle axis (e.g., at arrowhead). (B) Embryo fixed 9 min after furrow initiation. Horizontal arrowheads indicatesite of developing midzone. Bar Å 25 mm.

the region of new membrane addition during and after the treated embryos, suggesting that the FMA is directly in-volved in membrane addition.relaxation of the cleavage furrow. Importantly, microtubule

bundles retain association with the surface in cytochalasin- Cold shock, which reversibly depolymerizes microtu-


AID DB 8815 / 6x37$$$$41 01-23-98 13:24:57 dbas

52 Danilchik et al.

the first cleavage, but only after furrows were present.Nearly all embryos cold-shocked between the 16- and 64-cell stages (cleavage cycles 3 to 5) ruptured. Embryos be-came progressively less cold-sensitive at later cleavagestages, when FMAs are no longer present. We hypothesizethat rupture occurred upon warming because the contractilering resumed its advance before a microtubule-dependentmembrane system had time to reassemble; the demand forincreased surface area then quickly exceeded the availablenew membrane to cover it. In an experiment to test thisidea, we applied cytochalasin B to embryos just before thecold shock/warming treatment. Rupture was prevented inall embryos that were first treated with cytochalasin B (30of 30 cases), while in the parallel control, i.e., those withoutcytochalasin B, 47% of embryos ruptured (15 of 32 cases).Thus, it is the demand for increased surface area duringfurrowing, rather than a specific effect of cold on the proper-ties of the new membrane, that caused rupture. These re-sults reveal that rapid membrane growth along the cleavage

FIG. 3. Microtubule distribution in prospective cleavage plane atbeginning of first cleavage. Embryo was fixed 9 min after appear-ance of furrow. (A) Low-magnification view, showing location ofFMA at base of first cleavage furrow in animal hemisphere. (B)Higher magnification montage of same specimen details the FMA(fma) and midzone (mz) microtubules in the prospective cleavageplane. Planar zone in the vegetal hemisphere (x) appears to be en-tirely lacking in microtubules. Bar Å 50 mm.

bules, was also used to examine the importance of microtu-bules in membrane addition. Embryos were shocked brieflywith 47C culture medium (Scharf and Gerhart, 1983) andthen allowed to recover by warming to room temperature. FIG. 4. Microtubule distribution in the vegetal portion of firstUpon warming, a large percentage of embryos immediately cleavage furrow. Note absence of antiparallel midzone microtubuleruptured, spilling their contents out along new cleavage bundles deep to the FMA. Brightly staining bodies at the vegetal

surface correspond to germ plasm islands. Bar Å 50 mm.planes. As shown in Fig. 10, rupture never occurred before


AID DB 8815 / 6x37$$$$41 01-23-98 13:24:57 dbas


peared, despite the fact that neither midzone nor spindlemicrotubules had returned to the cytoplasmic bridge (Fig.11C). This result demonstrates that the FMA does not repre-sent simply a gathering of spindle fibers, but also involvespolarized microtubule nucleation and assembly near thebase of the furrow. Although confocal analysis presented inFig. 6 suggests that the FMA could form by realigning andaccumulating preexisting spindle or midzone microtubulesalong the contractile ring, we propose that at least some ofthe FMA microtubules initiate at and grow away from thefurrow’s leading edge.

Microtubule Stabilization Affects BasolateralMembrane Growth

At high concentrations, paclitaxel stabilizes microtu-FIG. 5. Midbodies develop between dividing cells after stage 7.bules by enabling their side-to-side association and inter-Microtubule staining pattern in animal hemisphere, revealing mid-

bodies, spindles, and asters, but no circular FMAs. Dark cavity in feres with vesicle transport (Hamm-Alvarez et al., 1993;lower right corner is the blastocoel. Bar Å 50 mm. Terasaki and Reese, 1994). Injection of paclitaxel (in DMSO)

into the animal hemisphere of eggs at 0.80 to 0.90 of thefirst cell cycle effectively blocked furrowing and membraneinsertion near the site of injection in 93% of embryos (27plane must accompany closure of the contractile ring; in

the absence of microtubules, possibly those of the FMA, of 29 cases) (Fig. 12A), while injection of DMSO alone at asimilar site and time allowed normal cleavage in 83% ofthis localized growth does not occur.

The cold shock experiments also shed light on the origin embryos (15 of 18 cases) (Fig. 12B). Induction of extensivetubulin sheets in the cleavage plane by paclitaxel was con-of the FMA and its relationship to other microtubules of

the cleavage plane, i.e., midzone and spindle fibers. Embryos firmed by antitubulin staining (Fig. 12C). Although the pres-ence of tubulin sheets probably also interferes with the clo-fixed at various stages following room-temperature recovery

were examined via confocal microscopy. FMA and spindle sure of the contractile ring, the complete absence of a newmembrane domain at the site of injection is consistent withmicrotubule bundles (Fig. 11A) quickly depolymerized fol-

lowing cold shock (Fig. 11B). Within 5 min after return to vesicle transport being specifically blocked by this treat-ment.room temperature, the microtubules of the FMA reap-

FIG. 6. Stereo pair showing microtubule distribution in cortex underlying stress folds near the advancing edge of the first furrow, abouthalfway between the animal pole and equator. Projections were generated from a confocal image stack corresponding to 30 mm of totaldepth through the specimen as described under Materials and Methods. Embryo was fixed 6 min after the appearance of the first cleavagefurrow and fractured parallel to the cleavage plane. Arrowheads: examples of distal branching of the midzone microtubule bundle. Arrows:newly aligned furrow microtubule bundles.


AID DB 8815 / 6x37$$$$41 01-23-98 13:24:57 dbas

54 Danilchik et al.

FIG. 7. Regression of cleavage furrow following nocodazole treatment. Fertilized, dejellied egg was treated with 10 mg/ml nocodazole 5min after appearance of the second cleavage furrow and observed via video time-lapse. Frames correspond to successive 5-min time points,during which the second furrow (arrow) regresses. First cleavage furrow, near completion before drug application, remained unaffected.FIG. 8. Cytochalasin D-treated embryo produces a new membrane domain at the site of furrow formation. Embryo was treated with 10mg/ml cytochalasin D about 15 min after the second cleavage furrow initiated. The newly inserted membrane domain is easily distinguishedfrom the pigmented original egg surface.

Incubation of embryos in D2O, another microtubule-sta- et al. (1997), microtubules in Xenopus eggs can be stabilizedwith D2O without inhibiting microtubule-based vesiclebilizing agent, showed that stabilization per se need not

inhibit delivery of new membrane. As shown by Rowning transport. When eggs were treated with D2O starting at 0.60to 0.82 of the first cell cycle, cleavage was blocked (19 of19 cases), and circular, unpigmented membrane patches ap-

TABLE 1 peared across the entire animal surface (Fig. 13A). IndirectEffect of Nocodazole on Appearance of New Membrane Domain immunofluorescence of tubulin in fixed, D2O-treated eggsin Cytochalasin B-Treated Cleaving Embryos revealed large numbers of monasters throughout the cyto-

plasm, many of which were subcortical and lay under theTime of nocodazole % embryos lacking new Total number of

new membrane patches (Fig. 13B). The appearance of newtreatment (min) membrane domain embryos examinedmembrane in D2O-treated embryos indicates that microtu-bule stabilization alone does not inhibit delivery of new0 87 45

/5 40 45 membrane, even in the absence of furrowing./10 33 30 When D2O was applied at a later time, starting at 0.85 of/15 0 32 the first cell cycle until 10 min after furrow initiation, nor-

mal furrowing occurred in 81% of embryos (44 of 54 cases),Note. Cleaving embryos were pipetted into MMR/3 containingwhile the other 19% arrested shortly after furrow formation.10 mg/ml cytochalasin B at the initiation of the second cleavageIn all of the embryos, new membrane was inserted in thefurrow (0 min). At times indicated, groups of embryos were thencurrently progressing furrow in greater amounts than nor-withdrawn and transferred to MMR/3 containing both cytochalasinmal (Fig. 13C). Confocal images of fixed eggs immuno-B and nocodazole at 10 mg/ml. Embryos were scored for the presence

of unpigmented patch at /20 min. stained for tubulin revealed a denser-than-normal network


AID DB 8815 / 6x37$$$$41 01-23-98 13:24:57 dbas


FIG. 9. Microtubule array persists in a region of new membrane delivery following cytochalasin D treatment. A–C show confocal imagesof the microtubule distributions in first-cleavage embryos fixed 5, 10, and 20 min, respectively, following exposure to 10 mg/ml cytochalasinD. In C, the surface had entirely flattened out, and the embryo resembled the live one shown in Fig. 8. Bar Å 50 mm.

of microtubules in the cytoplasm of eggs treated with D2O midbodies of telophase mouse embryos (Schatten et al.,1988). We used 6-11B-1 antibody (Piperno et al., 1987) toat first cleavage (not shown). The ability of eggs treated at

about the time of first cleavage to continue cleaving further reveal the distribution of acetylated a-tubulin in the furrow.Numerous small 6-11B-1-positive bodies were found in thedemonstrates that D2O-stabilized microtubules, which are

capable of transporting vesicles, still deliver new mem- cytoplasm until the early stages of cleavage, but acetylatedmicrotubules were not observed until the FMA appeared,brane, in this case without disrupting furrowing by the con-

tractile ring. at about the onset of new membrane growth (Figs. 14A–14C). After membrane growth began, 6-11B-1-positive mi-crotubule bundles appeared in the FMA. However, midzone

Furrow-Associated Microtubules Are Acetylated microtubules did not appear to become acetylated at anytime during cleavage. In contrast, after stage 7, when FMAsPostassembly acetylation accompanies stabilization of

microtubules in various subcellular structures (Webster and were no longer present, acetylated midbody microtubulesappeared (Fig. 14D) as previously observed by Chu andBorisy, 1989), including midzone microtubule bundles andKlymkowsky (1989). Based on their slender, linear appear-ance, these acetylated bundles represent only a subset ofthe entire midbody microtubules seen in Fig. 5.

DISCUSSION

Our results describe the development of a prominentarray of short, acetylated microtubule bundles in the cleav-age furrows of dividing Xenopus embryos, a structure wecall the FMA. The bundles of the FMA are located just deepto the progressing cleavage furrow and run radially from thecytoplasmic bridge, an orientation that suggests an activerole in the furrowing process. As each furrow extends acrossthe surface of the dividing blastomere, the FMA, whichstarts as an arc, expands to eventually form a ring at theFIG. 10. Cleavage-dependent sensitivity to cold shock-inducedleading edge of the furrow. These rings are seen only inrupture. Groups of 20 to 25 embryos were abruptly chilled by im-early cleavage stages (up to about stage 7), a period whenmersion in 47C MMR/3 for 4 min, quickly returned to room temper-massive amounts of plasma membrane deposition areature, and scored 2 min later for evidence of rupture along new

cleavage furrows as described under Materials and Methods. known to occur (Bluemink and deLaat, 1973). Thus, the


AID DB 8815 / 6x37$$$$41 01-23-98 13:24:57 dbas

56 Danilchik et al.

FIG. 11. FMA reassembles after cold shock. (A) Microtubule staining in the vegetal region of first-cleavage furrow, showing both FMAand an abundance of thick bundles in the cytoplasmic bridge. (B) Microtubule staining during cold shock. Most bundles in both the FMAand cytoplasmic bridge depolymerize. (C) Five minutes after return to room temperature, microtubule bundles have reformed in the FMA,while there are still few cytoplasmic bridge microtubules.

location, orientation, and time of appearance of the FMA FMA are generally not paired across the furrow’s leadingedge, so most of them probably do not correspond to rem-are consistent with involvement of microtubules in furrow

progression, perhaps in the growth of the basolateral surface nants of longer interpolar bundles in the cytoplasmic bridge.Second, as in other cells, midzone microtubule bundles de-membrane between blastomeres.

In animal cells undergoing cytokinesis, a midbody devel- velop in Xenopus eggs more or less directly between thespindle poles. In the earliest cleavages, the spindle is situ-ops as a collection of short microtubule bundles in the mid-

zone between the separated spindle poles. The midbody has ated in the animal hemisphere; thus, midzone microtubulesare not seen elsewhere in the prospective cleavage planeclassically been interpreted as a remnant of spindle microtu-

bules trapped in the contractile ring. Recently, however, (Fig. 3). Nevertheless, as shown in Fig. 4, FMA bundles de-velop abundantly in the vegetal portion of the furrow. TheseShu et al. (1995) found that g-tubulin appears transiently

in the cleavage plane of mammalian cells, a result sug- animal–vegetal differences are summarized in Fig. 15. Fi-nally, we have shown that FMAs reassemble quickly in thegesting that the midbody, like the FMA, may have an origin

independent of polar microtubules. Likewise, it was not absence of interpolar fibers during the recovery followingcold shock (Fig. 11). Cold treatment effectively depolymer-clear whether the FMA forms de novo or instead develops

via a more trivial process, i.e., by action of the contractile izes all microtubules in the prospective cleavage plane, in-cluding those of the FMA, midzone, and spindle. The FMA’sring to concentrate cortical or midzone microtubules. In-

deed, the data shown in Figs. 2 and 6 indicate that in the rapid reappearance upon warming, independent of other mi-crotubules, indicates a site of microtubule nucleation nearanimal hemisphere, astral microtubule bundles do realign

with the passage of the contractile ring. However, this kind the furrow base.In plant cells, microtubules have well-established rolesof process cannot account for the bulk of the FMA structure.

First, as shown in Fig. 1D, the microtubule bundles in the in cytokinesis, including that of directing Golgi-derived ves-


AID DB 8815 / 6x37$$$$41 01-23-98 13:24:57 dbas


FIG. 12. Injection of paclitaxel blocks furrowing and new membrane formation. Injection of 0.5 nl paclitaxel (10 mg/ml in DMSO) atthe animal pole just prior to cleavage prevented the furrow from progressing and new membrane from being deposited (A), while a similarinjection of DMSO did not alter cleavage (B). (C) Large numbers of paclitaxel-induced tubulin sheets occupy the cytoplasmic bridge, belowthe furrow base.FIG. 13. D2O fails to block furrowing or new membrane formation. (A) Egg treated with 50% D2O at 0.82 of the first cell cycle did notform a furrow but inserted new membrane in random spots in the animal hemisphere at the time that untreated eggs began cleavage. (B)Monasters are abundant in the cytoplasm of eggs treated with 50% D2O before first cleavage. Bar Å 30 mm. (C) Egg treated late in thesecond cleavage cycle adds membrane along the active third cleavage furrow.

icles toward the growing cell plate (reviewed in Gunning, preted as remnants of the mitotic apparatus, with the impli-cation that they play no further role in furrowing (Selman1982). In contrast, during animal cell cytokinesis, potential

functions of microtubules have been generally overlooked, and Perry, 1970). A few studies have proposed a mechanicalor supportive role for midbody microtubule bundles, e.g.,perhaps owing to the robustness of the contractile ring

model (reviewed in Rappaport, 1996). In classic models of in the cytoplasmic bridges of ciliates (Tucker, 1971) and inearly cleavages of Ilyanassa (Conrad et al., 1992).animal cell cytokinesis, which are based largely on work

with echinoderm eggs, the sole function of microtubules The response of amphibian embryos to microtubule as-sembly inhibitors may differ from that of cleaving echinoidfollowing chromosome separation in anaphase is to provide

a stimulus to the cortex to define the location of the acto- embryos. Sea urchin embryos exposed to colchicine cancomplete cleavages initiated before drug treatment (Hama-myosin contractile ring (Devore et al., 1989; Oegema and

Mitchison, 1997; Rappaport and Rappaport, 1974). In early guchi, 1975; see also Rappaport, 1996). Thus, it is generallyaccepted that, by the time a furrow appears, its progressionultrastructural studies (e.g., Gulyas, 1973; Kalt, 1971), mi-

crotubules in the cleavage plane were generally identified has become independent of microtubules. In contrast, inamphibians, the furrowing mechanism remains microtu-as components of the midzone or midbody and often inter-


AID DB 8815 / 6x37$$$$41 01-23-98 13:24:57 dbas

58 Danilchik et al.

FIG. 14. Whole-mount preparations stained with anti-acetylated tubulin antibody 6-11B-1. (A) 5, (B) 20, and (C) 25 min after appearanceof first cleavage furrow; (D) stage 8 midbodies. Bar Å 50 mm.

bule-dependent for much longer, perhaps until cleavage is specific structure in amphibian embryos for localized tar-geting of membrane insertion. We are currently examiningcompleted. Sawai and Yomota (1990) and Sawai (1992) ob-

served furrow regression in amphibian eggs following expo- other eggs and cells with emphasis on the distribution ofmicrotubules in the vicinity of the furrow.sure to colchicine and proposed that microtubules are re-

quired for propagation of the furrow around the egg surface. The evidence provided here strongly supports the hypoth-esis that microtubules function in the growth of newRecent evidence indicates that in mammalian tissue culture

cells, midzone microtubules are required for closure of the plasma membrane along the cleavage plane. In apparentcontrast to this, Roberts et al. (1992) showed that vesiclescytoplasmic bridge (Wheatley and Wang, 1996). Likewise,

Katsumoto (1993) proposed that the midzone microtubules containing a basolaterally targeted viral glycoprotein canincorporate into established basolateral membranes in no-interact directly with cortical microfilaments to initiate fur-

row formation. It is possible that mammalian tissue culture codazole-arrested eight-cell embryos, indicating that vesiclefusion per se does not require intact microtubules. How-cells also contain both FMA and midzone bundles but that

the two structures cannot be resolved in the cytoplasmic ever, this addition of vesicles to preexisting basolateral do-mains probably represents a process independent of thebridge of such small cells. Midbodies could be involved in

anaphase functions of the mitotic apparatus, with the FMA rapid growth of the new membrane and does not precludethe role of microtubules in the recruitment of membranecarrying out more furrow-related functions in cytokinesis.

On the other hand, since amphibian eggs, apparently unlike stores during cleavage.How furrow-associated microtubules function in estab-cultured mammalian cells or sea urchin eggs, add relatively

large amounts of new plasma membrane to the cleavage lishing the new basolateral domain during cytokinesis re-mains enigmatic. The FMA might play a direct role in mem-plane itself during cytokinesis, the FMA may represent a

FIG. 15. Summary of animal–vegetal differences in the development of early FMAs. In the animal hemisphere, a meshwork of fibersdevelops directly beneath the early furrow. Discontinuous midzone bundles are found generally deeper in the cytoplasm, between thespindle poles. As the furrow deepens and new membrane is inserted along the cleavage plane, a prominent FMA develops at the furrowbase, possibly by condensation of the earlier meshwork. Discontinuous midzone bundles precede the furrow; whether they are ultimatelyincorporated into the FMA or are passed by during the furrow’s advance is uncertain. In the vegetal hemisphere, the FMA develops largelyin the absence of midzone bundles.


AID DB 8815 / 6x37$$$$41 01-23-98 13:24:57 dbas


Cao, L. G., and Wang, Y. L. (1996). Signals from the spindle midzonebrane growth: for example, microtubule bundles could pro-are required for the stimulation of cytokinesis in cultured epithe-vide tracks for transporting vesicles toward targets in thelial cells. Mol. Biol. Cell 7, 225–232.growing membrane. In such a scenario, one would expect

Choi, Y. S., Sehgal, R., McCrea, P., and Gumbiner, B. (1990). Afusion events to be concentrated at, or very near, the basecadherin-like protein in eggs and cleaving embryos of Xenopusof the advancing furrow. The surface radiolabeling experi-laevis is expressed in oocytes in response to progesterone. J. Cell

ments of Byers and Armstrong (1986) indicate that the base Biol. 110, 1575–1582.of the furrow includes membrane components from the Chu, D. T., and Klymkowsky, M. W. (1989). The appearance of ace-original egg surface, so a site of membrane delivery at the tylated alpha-tubulin during early development and cellular dif-very base of the furrow appears unlikely. On the other hand, ferentiation in Xenopus. Dev. Biol. 136, 104–117.analysis of the movement of carbon particles in the cleavage Colombo, R. (1983). Actin in Xenopus yolk platelets: A peculiar

and debated presence. J. Cell Sci. 63, 263–270.plane of newt embryos reveals that membrane insertion isConrad, A. H., Paulsen, A. Q., and Conrad, G. W. (1992). The rolelocalized to a site within a few microns of the furrow base

of microtubules in contractile ring function. J. Exp. Zool. 262,(Sawai, 1987; cited in Rappaport, 1996; and similar unpub-154–165.lished observations by M. V. Danilchik in Xenopus). Alter-

Danilchik, M. V., and Denegre, J. M. (1991). Deep cytoplasmic rear-natively, microtubules might play a more indirect role inrangements during early development in Xenopus laevis. Devel-basolateral membrane growth, for example, by regulatingopment 111, 845–856.

localized vesicle fusion. In Xenopus, furrow progression re- Devore, J. J., Conrad, G. W., and Rappaport, R. (1989). A model forquires localized calcium release (Miller et al., 1993; Muto astral stimulation of cytokinesis in animal cells. J. Cell Biol. 109,et al., 1996). Microtubules of the FMA could locally mediate 2225–2232.this calcium release by mechanically perturbing the endo- Drechsel, D. N., Hyman, A. A., Hall, A., and Glotzer, M. (1997). Aplasmic reticulum (Jaffe, 1993; Mehlmann et al., 1995), requirement for Rho and Cdc42 during cytokinesis in Xenopus

embryos. Curr. Biol. 7, 12–23.thereby facilitating the localized fusion of vesicles with theEarnshaw, W. C., and Cooke, C. A. (1991). Analysis of the distribu-surface. A third possible function of the FMA in cytokinesis

tion of the INCENPs throughout mitosis reveals the existencemight be to facilitate the extensive ingression of subcorticalof a pathway of structural changes in the chromosomes duringcytoplasm along the early cleavage planes. Ingression com-metaphase and early events in cleavage furrow formation. J. Cellpletely reorganizes the maternally established egg organiza-Sci. 98, 443–461.

tion (Danilchik and Denegre, 1991). By mechanically inter-Eckley, D. M., Ainsztein, A. M., Mackay, A. M., Goldberg, I. G.,

acting with the extensive subcortical networks of cytokera- and Earnshaw, W. C. (1997). Chromosomal proteins and cytoki-tin (Gard et al., 1997; Klymkowsky et al., 1987) or actin nesis: Patterns of cleavage furrow formation and inner centro-(Colombo, 1983), the FMA could play an important role in mere protein positioning in mitotic heterokaryons and mid-ana-the localization of cytoplasmic determinants. phase cells. J. Cell Biol. 136, 1169–1183.

Elinson, R. P., and Rowning, B. (1988). A transient array of parallelmicrotubules in frog eggs: Potential tracks for a cytoplasmic rota-tion that specifies the dorso–ventral axis. Dev. Biol. 128, 185–ACKNOWLEDGMENT197.

Gard, D. L. (1991). Organization, nucleation, and acetylation of mi-This work was supported by NSF IBN-9396063 and 9631765. crotubules in Xenopus laevis oocytes: A study by confocal immu-

nofluorescence microscopy. Dev. Biol. 143, 346–362.Gard, D. L. (1993). Confocal immunofluorescence microscopy of

REFERENCES microtubules in amphibian oocytes and eggs. Methods Cell Biol.38, 241–264.

Gard, D. L., Cha, B. J., and King, E. (1997). The organization andBieliavsky, N., and Geuskens, M. (1990). Interblastomeric plasmaanimal–vegetal asymmetry of cytokeratin filaments in stage VImembrane formation during cleavage of Xenopus laevis embryos.Xenopus oocytes is dependent upon F-actin and microtubules.J. Submicrosc. Cytol. Pathol. 22, 445–457.Dev. Biol. 184, 95–114.Bieliavsky, N., Geuskens, M., Goldfinger, M., and Tencer, R. (1992).

Gawantka, V., Ellinger-Ziegelbauer, H., and Hausen, P. (1992). BetaIsolation of plasma membranes, Golgi bodies and mitochondria1-integrin is a maternal protein that is inserted into all newlyof Xenopus laevis morulae. Identification of plasma membraneformed plasma membranes during early Xenopus embryogenesis.proteins. J. Submicrosc. Cytol. Pathol. 24, 335–349.Development 115, 595–605.Bluemink, J. G. (1971). Cytokinesis and cytochalasin-induced fur-

Gulyas, B. J. (1973). Cytokinesis in the rabbit zygote: Fine-struc-row regression in the first-cleavage zygote of Xenopus laevis. Z.tural study of the contractile ring and the mid-body. Anat. Rec.Zellforsch. Mikroskop. Anat. 121, 102–126.177, 195–207.Bluemink, J. G., and deLaat, S. W. (1973). New membrane forma-

Gunning, B. E. S. (1982). The cytokinetic apparatus: Its develop-tion during cytokinesis in normal and cytochalasin B-treated eggsment and spatial regulation. In ‘‘The Cytoskeleton in Plantof Xenopus laevis. I. Electron microscope observations. J. CellGrowth and Development’’ (C. W. Lloyd, Ed.), pp. 229–292. Aca-Biol. 59, 89–108.demic Press, London.Byers, T. J., and Armstrong, P. B. (1986). Membrane protein redistri-

Hamaguchi, Y. (1975). Microinjection of colchicine into sea urchinbution during Xenopus first cleavage. J. Cell Biol. 102, 2176–2184. eggs. Dev. Growth Differ. 17, 111–117.


AID DB 8815 / 6x37$$$$41 01-23-98 13:24:57 dbas

60 Danilchik et al.

Hamm-Alvarez, S. F., Kim, P. Y., and Sheetz, M. P. (1993). Regula- Savage, R. M., and Danilchik, M. V. (1993). Dynamics of germplasm localization and its inhibition by ultraviolet irradiation intion of vesicle transport in CV-1 cells and extracts. J. Cell Sci.

106, 955–966. early cleavage Xenopus embryos. Dev. Biol. 157, 371–382.Sawai, T. (1987). Surface movement in the region of the cleavageJaffe, L. F. (1993). Classes and mechanisms of calcium waves. Cell

furrow of amphibian eggs. Zool. Sci. 4, 825–832.Calcium 14, 736–745.Sawai, T. (1992). Effect of microtubular poisons on cleavage furrowKalt, M. R. (1971). The relationship between cleavage and blasto-

formation and induction of furrow-like dent in amphibian eggs.coel formation in Xenopus laevis. II. Electron microscopic obser-Dev. Growth Differ. 34, 669–675.vations. J. Embryol. Exp. Morphol. 26, 51–66.

Sawai, T., and Yomota, A. (1990). Cleavage plane determination inKatsumoto, T., Naguro, T., Yan, Y. L., Iino, A., and Onodera, K.amphibian eggs. Ann. N.Y. Acad. Sci. 582, 40–49.(1993). The spatial distribution of spindle microtubules in ana-

Scharf, S. R., and Gerhart, J. C. (1983). Axis determination in eggsphase 3Y1 cells. Biol. Cell 77, 247–253.of Xenopus laevis: A critical period before first cleavage, identi-Klymkowsky, M. W., and Hanken, J. (1991). Whole-mount stainingfied by the common effects of cold, pressure and ultraviolet irra-of Xenopus and other vertebrates. Methods Cell Biol. 36, 419–diation. Dev. Biol. 99, 75–87.441.

Schatten, G., Simerly, C., Asai, D. J., Szoke, E., Cooke, P., andKlymkowsky, M. W., Maynell, L. A., and Polson, A. G. (1987). PolarSchatten, H. (1988). Acetylated alpha-tubulin in microtubulesasymmetry in the organization of the cortical cytokeratin systemduring mouse fertilization and early development. Dev. Biol. 130,of Xenopus laevis oocytes and embryos. Development 100, 543–74–86.557.

Schroeder, T. E. (1970). The contractile ring. I. Fine structure ofLeaf, D. S., Roberts, S. J., Gerhart, J. C., and Moore, H. P. (1990).dividing mammalian (HeLa) cells and the effects of cytochalasinThe secretory pathway is blocked between the trans-Golgi andB. Z. Zellforsch. Mikroskop. Anat. 109, 431–449.the plasma membrane during meiotic maturation in Xenopus

Schroeder, T. E. (1972). The contractile ring. II. Determining itsoocytes. Dev. Biol. 141, 1–12.brief existence, volumetric changes, and vital role in cleavingMargolis, R. L., and Andreassen, P. R. (1993). The telophase disc:Arbacia eggs. J. Cell Biol. 53, 419–434.Its possible role in mammalian cell cleavage. Bioessays 15, 201–

Selman, G. G., and Perry, M. M. (1970). Ultrastructural changes in207.the surface layers of the newt’s egg in relation to the mechanism

Mehlmann, L. M., Terasaki, M., Jaffe, L. A., and Kline, D. (1995).of its cleavage. J. Cell Sci. 6, 207–227.

Reorganization of the endoplasmic reticulum during meioticServetnick, M., Schulte-Merker, S., and Hausen, P. (1990). Cell sur-

maturation of the mouse oocyte. Dev. Biol. 170, 607–615. face proteins during early Xenopus development: Analysis of cellMiller, A. L., Fluck, R. A., McLaughlin, J. A., and Jaffe, L. F. (1993). surface proteins and total glycoproteins provides evidence of a

Calcium buffer injections inhibit cytokinesis in Xenopus eggs. J. maternal glycoprotein pool. Roux’s Arch. Dev. Biol. 198, 433–Cell Sci. 106, 523–534. 442.

Muto, A., Kume, S., Inoue, T., Okano, H., and Mikoshiba, K. (1996). Shu, H. B., Li, Z., Palacios, M. J., Li, Q., and Joshi, H. C. (1995). ACalcium waves along the cleavage furrows in cleavage-stage Xen- transient association of gamma-tubulin at the midbody is re-opus embryos and its inhibition by heparin. J. Cell Biol. 135, quired for the completion of cytokinesis during the mammalian181–190. cell division. J. Cell Sci. 108, 2955–2962.

Oegema, K., and Mitchison, T. J. (1997). Rappaport Rules—Cleav- Szollosi, D. (1970). Cortical cytoplasmic filaments of cleaving eggs:age furrow induction in animal cells. Proc. Natl. Acad. Sci. UA A structural element corresponding to the contractile ring. J. Cell94, 4817–4820. Biol. 44, 192–209.

Piperno, G., LeDizet, M., and Chang, X. J. (1987). Microtubules Terasaki, M., and Reese, T. S. (1994). Interactions among endoplas-containing acetylated alpha-tubulin in mammalian cells in cul- mic reticulum, microtubules, and retrograde movements of theture. J. Cell Biol. 104, 289–302. cell surface. Cell Motil. Cytoskel. 29, 291–300.

Rappaport, R. (1996). ‘‘Cytokinesis in Animal Cells.’’ Cambridge Tucker, J. B. (1971). Microtubules and a contractile ring of micro-Univ. Press, Cambridge, UK. filaments associated with a cleavage furrow. J. Cell Sci. 8, 557–

Rappaport, R., and Rappaport, B. N. (1974). Establishment of cleav- 571.Vincent, J. P., and Gerhart, J. C. (1987). Subcortical rotation in Xen-age furrows by the mitotic spindle. J. Exp. Zool. 189, 189–196.

opus eggs: An early step in embryonic axis specification. Dev.Ressom, R. E., and Dixon, K. E. (1988). Relocation and reorganiza-Biol. 123, 526–539.tion of germ plasm in Xenopus embryos after fertilization. Devel-

Webster, D. R., and Borisy, G. G. (1989). Microtubules are ace-opment 103, 507–518.tylated in domains that turn over slowly. J. Cell Sci. 92, 57–65.Roberts, S. J., Leaf, D. S., Moore, H. P., and Gerhart, J. C. (1992). The

Wheatley, S. P., and Wang, Y. (1996). Midzone microtubule bundlesestablishment of polarized membrane traffic in Xenopus laevisare continuously required for cytokinesis in cultured epithelialembryos. J. Cell Biol. 118, 1359–1369.cells [published erratum appears in J. Cell Biol. 135(6, Pt. 1), 1679Rowning, B. A., Wells, J., Wu, M., Gerhart, J. C., Moon, R. T., and(1996)]. J. Cell Biol. 135, 981–989.Larabell, C. A. (1997). Microtubule-mediated transport of organ-

Yisraeli, J. K., Sokol, S., and Melton, D. A. (1990). A two-step modelelles and localization of beta-catenin to the future dorsal side offor the localization of maternal mRNA in Xenopus oocytes:Xenopus eggs. Proc. Natl. Acad. Sci. US A 94, 1224–1229.Involvement of microtubules and microfilaments in the translo-Sanders, E. J., and Singal, P. K. (1975). Furrow formation in Xenopuscation and anchoring of Vg1 mRNA. Development 108, 289–298.embryos. Involvement of the golgi body as revealed by ultrastruc-

tural localization of thiamine pyrophosphatase activity. Exp. Cell Received for publication September 29, 1997Accepted November 21, 1997Res. 93, 219–224.


AID DB 8815 / 6x37$$$$41 01-23-98 13:24:57 dbas

developmental biology 194

Documents