Mitotic apparatus-surface interaction and cell division

Summary

Results of recent investigations concerning the mechanisms of animal cell division are reviewed. The mitotic apparatus was aspirated from a blastomere of a sand dollar (Echinarachnius parma) egg before second cleavage, and the time interval between removal and the appearance of the furrow in the control companion blastomere was measured. When the mitotic apparatus is removed 4 min or less before the furrows appear in the controls, furrows also develop in the operated cells. These results show that 4 min before furrowing begins, the surface changes which lead to formation of the division mechanism have become irreversible. When the mitotic apparatus of a cylindrical cell is shifted by pushing in one of the poles when the furrow appears, a new furrow develops in association with the new position of the mitotic apparatus. The same mitotic apparatus could elicit as many as 13 furrows over a 24.5 min period following the appearance of the first furrow. The results show that, in the proper geometrical circumstances, the mitotic apparatus and the surface can interact over a longer period than they do in normal cells.

By artificially constricting sand dollar eggs with a glass loop, the normal distance relations between the astral centers and the polar and equatorial surfaces can be reversed. Constricted cells cleave normally. The blocking effect of ethyl urethane can be reversed by moving the equatorial surface closer to the spindle portion of the mitotic apparatus. Relocation of other parts of the surface closer to the mitotic apparatus was ineffective. These results help elucidate the geometrical relations that are essential for furrow formation between the mitotic apparatus and the surface.

In cylindrical sand dollar eggs, single asters and the widely separated asters of a broken mitotic apparatus can cause furrow-like constrictions in the adjacent cylindrical surface. This reaction can be blocked by treating cells with ethyl urethane, which reduces astral size. The nature of the shape change that the aster causes depends upon the surface region affected. These results aid in understanding the nature of the change in surface physical activity caused by the mitotic apparatus.     

Furrowing in altered cell surfaces.

Understanding the process which established the cell division mechanism requires analysis of the role of the responding surface as well as that of stimulatory subsurface structures. Cell surface was altered by the expansion which occurs during exovate formation. Exovates appear on the surface of fertilized Arbacia lixula, Paracentrotus lividus and Echinarachnius parma eggs in response to extreme flattening. They result from cytoplasmic outflow initiated in a very restricted portion of the egg surface. Observations of the formation process in pigmented A. lixula eggs revealed that the original surface may be expanded about 100 fold as the exovate swells. When exovates formed 15-30 minutes after fertilization contain the mitotic apparatus, they divide synchronously with flattened controls. If nucleated exovates are established after the beginning of first cleavage, furrows appear in ten minutes. Exovates established after the beginning of second cleavage develop furrows four minutes after the entrance of the the mitsotic apparatus. Cytoplasm beneath damaged exovate surfaces sometimes develops partial constrictions independently of the surface in the plane the furrow would have occupied. These results suggest that normal surface structure is unnecessary for furrow establishment and function.     

Cleavage furrow establishment by the moving mitotic apparatus

The midpoint of the mitotic apparatus is fixed in the future division plane long before the division mechanism develops, and this static relationship has been considered essential in speculations concerning division mechanism establishment. The purpose of the present investigation was to determine whether prevention of the static relationship affects the establishment process. Sand dollar eggs were reshaped into cylinders by confinement in an elastic capillary tube and, beginning about 20 min before cleavage, the mitotic apparatus was kept in reciprocal motion by alternately compressing the poles. When the movement was continuous and the excursions were 25, 50 or 75 μm, furrow activity developed near the midpoint of the region underlain by the mitotic apparatus. The acuteness of the furrow decreased as the distance the mitotic apparatus was moved increased. When the movement was made discontinuous by allowing the mitotic apparatus to pause at the end of each excursion, the results depended upon the duration of the pause. Pauses 30 s long resulted in a single furrow formed in the midpoint of the entire region underlain by the mitotic apparatus. When the pauses were 45s long, furrowing activity developed in both regions where the mitotic apparatus was allowed to pause. The results indicated that the normal static relation between the mitotic apparatus midpoint and the division plane is unnecessary for division mechanism establishment. They also demonstrate that a restricted region of contractile activity can be established in the cortex despite experimentally induced spreading and dilution of mitotic apparatus effect.     

Why does a cleavage plane develop parallel to the spindle axis in conical sand dollar eggs? A key question for clarifying the mechanism of contractile ring positioning

Three types of models have been proposed about how the mitotic apparatus determines the position of the cleavage furrow in animal cells. In the first and second types, the contractile ring appears in a cortical region that least and most astral microtubules reach, respectively. The third type is that the spindle midzone positions the contractile ring. In the previous study, a new model was proposed through analyses of cytokinesis in sand dollar and sea urchin eggs. Gradients of the surface density of microtubule plus ends are assumed to drive membrane proteins whose accumulation causes the formation of contractile-ring microfilaments. In the present study, the validity of each model is examined by simulating the furrow formation in conical sand dollar eggs with the mitotic apparatus oriented perpendicular to the cone axis. The new model predicts that unilateral furrows with cleavage planes roughly parallel to the spindle axis appear between the mitotic apparatus and the vertex besides the normally positioned furrow. The predictions are consistent with the observations by Rappaport & Rappaport (1994, Dev. Biol.164, 258-266). The other three types of models do not predict the formation of the ectopic furrows. Furthermore, it is pointed out that only the new model has the ability to explain the geometrical relationship between the mitotic apparatus and the contractile ring under various experimental conditions. These results strongly suggest the real existence of the membrane proteins postulated in the model.     

Tubulin and calmodulin. Effects of microtubule and microfilament inhibitors on localization in the mitotic apparatus

Indirect immunofluorescence was used to determine the distribution of calmodulin in the mitotic apparatus of rat kangaroo PtK2 and Chinese hamster ovary (CHO) cells. The distribution of calmodulin in PtK2 cells was compared to the distribution of tubulin, also as revealed by indirect immunofluorescence. During mitosis, calmodulin was found to be a dynamic component of the mitotic apparatus. Calmodulin first appeared in association with the forming mitotic apparatus during midprophase. In metaphase and anaphase, calmodulin was found between the spindle poles and the chromosomes. While tubulin was found in the interzonal region throughout anaphase, calmodulin appeared in the interzone region only at late anaphase. The interzonal calmodulin of late anaphase condensed during telophase into two small regions, one on each side of the midbody. Calmodulin was not detected in the cleavage furrow. In view of the differences in the localization of calmodulin, tubulin, and actin in the mitotic apparatus, experiments were designed to determine the effects of various antimitotic drugs on calmodulin localization. Cytochalasin B, an inhibitor of actin microfilaments, had no apparent effect on calmodulin or tubulin localization in the mitotic apparatus of CHO cells. Microtubule inhibitors, such as colcemid and N2O, altered the appearance of tubulin- and calmodulin-specific fluorescence in mitotic CHO cells. Cold temperature (0 degrees C) altered tubulin-specific fluorescence of metaphase PtK2 cells but did not alter calmodulin-specific fluorescence. From these studies, it is concluded that calmodulin is more closely associated with the kinetichore-to-pole microtubules than other components of the mitotic apparatus.     

Slow calcium waves accompany cytokinesis in medaka fish eggs

Animal cells are cleaved by the formation and contraction of an extremely thin actomyosin band. In most cases this contractile band seems to form synchronously around the whole equator of the cleaving cell; however in giant cells it first forms near the mitotic apparatus and then slowly grows outwards over the cell. We studied the relationship of calcium to such contractile band growth using aequorin injected medaka fish eggs: we see two successive waves of faint luminescence moving along each of the first three cleavage furrows at approximately 0.5 micron/s. The first, narrower waves accompany furrow extension, while the second, broader ones, accompany the subsequent apposition or slow zipping together of the separating cells. If the first waves travel within the assembling contractile band, they would indicate local increases of free calcium to concentrations of about five to eight micromolar. This is the first report to visualize high free calcium within cleavage furrows. Moreover, this is also the first report to visualize slow (0.3-1.0 micron/s) as opposed to fast (10-100 microns/s) calcium waves. We suggest that these first waves are needed for furrow growth; that in part they further furrow growth by speeding actomyosin filament shortening, while such shortening in turn acts to mechanically release calcium and thus propagates these waves as well as furrow growth. We also suggest that the second waves act to induce the exocytosis which provides new furrow membrane.     

The role of calcium, polyamines and centrosomes in the formation and organization of cleavage furrows in amphibian eggs.

Cleavage furrows of amphibian eggs exhibit characteristic morphological features: the presence of finger-like microvilli (MV) along their outer edges, the formation of furrow walls from new plasma membrane lacking MV, and the subsequent retrieval of this membrane during the infolding of the furrow. A similar structure can be induced, specifically, by certain cytoplasmic components such as centrosomes, polyamines and calcium. Their respective roles in the events associated with the furrowing process have been investigated by injecting these agents into nucleated and enucleated Pleurodeles eggs and evaluating their effects using cytochemical labelling of the egg surface with a biotin-streptavidin system. The injection of polyamines (spermine or spermidine) and in some cases, calcium into enucleated eggs provoked MV elongation and the appearance of newly formed, smooth plasma membrane. In these eggs, this membrane was not incorporated into the furrows, and as a consequence, the blastomeres did not actually separate. In contrast, the injection of centrosomes into enucleated eggs induced both the incorporation and internalization of new membrane, resulting in the formation of furrows and a true cellularization of the eggs, identical to the cleavage process observed in fertilized eggs. The present results provide further evidence that the establishment of the furrow depends on two complementary interacting systems: the contractile elements of the egg cortex which regulate the insertion of new membrane and the mitotic center which is essential for the invagination of the furrow.     

Chromosomal Proteins and Cytokinesis: Patterns of Cleavage Furrow Formation and Inner Centromere Protein Positioning in Mitotic Heterokaryons and Mid-anaphase Cells

After the separation of sister chromatids in anaphase, it is essential that the cell position a cleavage furrow so that it partitions the chromatids into two daughter cells of roughly equal size. The mechanism by which cells position this cleavage furrow remains unknown, although the best current model is that furrows always assemble midway between asters. We used micromanipulation of human cultured cells to produce mitotic heterokaryons with two spindles fused in a V conformation. The majority (15/19) of these cells cleaved along a single plane that transected the two arms of the V at the position where the metaphase plate had been, a result at odds with current views of furrow positioning. However, four cells did form an additional ectopic furrow between the spindle poles at the open end of the V, consistent with the established view. To begin to address the mechanism of furrow assembly, we have begun a detailed study of the properties of the chromosome passenger inner centromere protein (INCENP) in anaphase and telophase cells. We found that INCENP is a very early component of the cleavage furrow, accumulating at the equatorial cortex before any noticeable cortical shape change and before any local accumulation of myosin heavy chain. In mitotic heterokaryons, INCENP was detected in association with spindle midzone microtubules beneath sites of furrowing and was not detected when furrows were absent. A functional role for INCENP in cytokinesis was suggested in experiments where a nearly full-length INCENP was tethered to the centromere. Many cells expressing the chimeric INCENP failed to complete cytokinesis and entered the next cell cycle with daughter cells connected by a large intercellular bridge with a prominent midbody. Together, these results suggest that INCENP has a role in either the assembly or function of the cleavage furrow.     

Transitions regulating the timing of cytokinesis in embryonic cells

Anaphase, mitotic exit, and cytokinesis proceed in rapid succession, and while mitotic exit is a requirement for cytokinesis in yeast, it may not be a direct requirement for furrow initiation in animal cells. In this report, we physically manipulated the proximity of the mitotic apparatus (MA) to the cell cortex in combination with microinjection of effectors of the spindle checkpoint and CDK1 activity to determine how the initiation of cytokinesis is coupled to the onset of anaphase and mitotic exit. Whereas precocious contact between the MA and the cell surface advanced the onset of cytokinesis into early anaphase A, furrowing could not be advanced prior to the metaphase-anaphase transition. Additionally, while cells arrested in anaphase could be induced to initiate cleavage furrows, cells arrested in metaphase could not. Finally, activation of the mitotic checkpoint in one spindle of a binucleate cell failed to arrest cytokinesis induced by the control spindle but did inhibit the formation of furrows between the arrested MA and the control, nonarrested MA. Our experiments suggest that the competence of the mitotic apparatus to initiate cytokinesis is not dependent on cyclin degradation but does require anaphase-promoting complex (APC) activity and, thus, inactivation of the mitotic checkpoint.     

Electron microscopic observations on the maintenance of the tight junction during cell division in the epithelium of the mouse small intestine.

Dividing epithelial cells in the mouse small intestine were examined by thin-section electron microscopy with special attention given to the mode of cytokinesis. As the columnar epithelial cells entered mitosis in the crypt, they became rounded, maintaining their junctional complexes with neighboring cells while detaching themselves from the basal lamina. In such rounded cells the mitotic apparatus was formed with its long axis parallel to the luminal surface. Replicated centrioles moved down from the apical region to locate themselves lateral to the nucleus, where they served as the poles of the mitotic spindle. During mitosis the cell retained microvilli on its luminal surface, though the terminal web became much thinner. At telophase the formation of a cleavage furrow proceeded asymmetrically from the basal side alone, and thus the contractile ring which was prominent at the base of the furrow, merged with the terminal web. Eventually, an intercellular bridge with a midbody was formed on the luminal surface. The space in the furrow was occupied by the flattened cytoplasmic processes of the neighboring cells. The tight junction was also seen on the basolateral surface of the intercellular bridge with the underlying neighboring cells. At very late telophase the intercellular bridge was disconnected from the neighboring cells and protruded into the lumen. These observations have led us to propose a mode by which the simple columnar epithelium maintain the tight junctional seal during cell division in the crypt of the small intestinal epithelium.     

Precocious cleavage furrows simultaneously move and ingress when kinetochore microtubules are depolymerized in <Emphasis Type="Italic">Mesostoma ehrenbergii</Emphasis> spermatocytes

A “precocious” cleavage furrow develops and ingresses during early prometaphase in Mesostoma ehrenbergii spermatocytes (Forer and Pickett-Heaps Eur J Cell Biol 89:607-618, 2010). In response to chromosome movements which regularly occur during prometaphase and that alter the balance of chromosomes in the two half-spindles, the precocious furrow shifts its position along the cell, moving 2–3 μm towards the half cell with fewer chromosomes (Ferraro-Gideon et al. Cell Biol Int 37:892-898, 2013). This process continues until proper segregation is achieved and the cell enters anaphase with the cleavage furrow again in the middle of the cell. At anaphase, the furrow recommences ingression. Spindle microtubules (MTs) are implicated in various furrow positioning models, and our experiments studied the responses of the precocious furrows to the absence of spindle MTs. We depolymerized spindle MTs during prometaphase using various concentrations of nocodazole (NOC) and colcemid. The expected result is that the furrow should regress and chromosomes remain in the midzone of the cell (Cassimeris et al. J Cell Sci 96:9-15, 1990). Instead, the furrows commenced ingression and all three bivalent chromosomes moved to one pole while the univalent chromosomes, that usually reside at the two poles, either remained at their poles or moved to the opposite pole along with the bivalents, as described elsewhere (Fegaras and Forer 2018). The microtubules were completely depolymerized by the drugs, as indicated by immunofluorescence staining of treated cells (Fegaras and Forer 2018), and in the absence of microtubules, the furrows often ingressed (in 33/61 cells) at a rate similar to normal anaphase ingression (~?1 μm/min), while often simultaneously moving toward one pole. Thus, these results indicate that in the absence of anaphase and of spindle microtubules, cleavage furrows resume ingression.     

Centrosomes and the Scrambled protein coordinate microtubule-independent actin reorganization

In Drosophila syncytial blastoderm embryos, centrosomes specify the position of actin-based interphase caps and mitotic furrows. Mutations in the scrambled locus prevent assembly of mitotic furrows, but do not block actin cap formation. The scrambled gene encodes a protein that localizes to the mitotic furrows and centrosomes. Sced localization, actin reorganization from caps into mitotic furrows, and centrosome-coordinated assembly of actin caps are not blocked by microtubule disruption. Our results indicate that centrosomes may coordinate assembly of cortical actin caps through a microtubule-independent mechanism, and that Scrambled mediates a second microtubule-independent process that drives mitotic furrow assembly.     

The degradation of two mitotic cyclins contributes to the timing of cytokinesis

BACKGROUND: Cytokinesis occurs just as chromosomes complete segregation and reform nuclei. It has been proposed that cyclin/Cdk kinase inhibits cytokinesis until exit from mitosis; however, the timer of cytokinesis has not been experimentally defined. Whereas expression of a stable version of Drosophila cyclin B blocks cytokinesis along with numerous events of mitotic exit, stable cyclin B3 allows cytokinesis even though it blocks late events of mitotic exit. We examined the interface between mitotic cyclin destruction and the timing of cytokinesis. RESULTS: In embryonic mitosis 14, the cytokinesis furrow appeared 60 s after the metaphase/anaphase transition and closed 90 s later during telophase. In cyclin B or cyclin B3 mutant cells, the cytokinesis furrow appeared at an earlier stage of mitosis. Expression of stable cyclin B3 delayed and prolonged furrow invagination; nonetheless, cytokinesis completed during the extended mitosis. Reduced function of Pebble, a Rho GEF required for cytokinesis, also delayed and slowed furrow invagination, but incomplete furrows were aborted at the time of mitotic exit. In functional and genetic tests, cyclin B and cyclin B3 inhibited Pebble contributions to cytokinesis. CONCLUSIONS: Temporal coordination of mitotic events involves inhibition of cytokinesis by cyclin B and cyclin B3 and punctual relief of the inhibition by destruction of these cyclins. Both cyclins inhibit Pebble-dependent activation of cytokinesis, whereas cyclin B can inhibit cytokinesis by additional modes. Stable cyclin B3 also blocks the later return to interphase that otherwise appears to impose a deadline for the completion of cytokinesis.     

Tensiometric studies of cytokinesis in cleaving sand dollar eggs

Tensions exerted by cleavage furrows of Echinarachnius parma were measured by means of calibrated, flexible glass needles. The tensions exerted by the first and second furrows in isometric contraction were, respectively, 1.58 X 10(-3) dyne (S.D. 0.41 X 10(-3) dyne) and 1.43 X 10(-3) dyne (S.D. 0.44 X 10(-3) dyne). The difference between the two means is not significant. The tensions exerted by the same cleavage furrow at two different lengths, of which the shorter was about 66% the length of the longer, were not significantly different. When the progress of a second cleavage furrow was mechanically blocked, it continued to exert maximum tension up to 9 minutes after its companion blastomere, which served as a time control, completed cleavage.     

Organization of mitotic apparatus poles in etoposide-treated CHO-K1 cells

Analyzed in this study is the organization of mitotic spindle poles in CHO-K1 cells dividing after treatment with etoposide (1 h, 25 μM). At various periods after the treatment, we studied the following: (1) the distribution of γ-tubulin in mitotic cells by immunofluorescent staining, (2) the level of post-translational modification of α-tubulin in spindle microtubules by immunoelectron microscopy, and (3) the ultrastructure of mitotic apparatus poles by standard electron microscopy. 48 h after the addition of etoposide, disturbances in the ultrastructure of mitotic spindle poles were observed in etoposide-treated CHO-K1 cells with both bipolar and with multipolar mitotic apparatuses. The increased number of centrioles was unevenly distributed between the mitotic spindle poles; some centrioles did not take an obvious part in the mitotic spindle organization and differed in their number of outgrowing microtubules. Most centrioles were without fibrillar halos. Immunoelectron microscopy showed the differences in the staining of the poles of a multipolar spindle within one cell with antibodies to tyrosinated α-tubulin, whereas the staining of cells with antibodies to acetylated α-tubulin did not reveal such differences. Immunofluorescence staining for γ-tubulin also indicated differing organizations of poles in the same spindle. Our data findings provided the first evidence that the pattern of immunostaining and ultrastructure of mitotic apparatus poles can differ in cells dividing at various time periods after the action of etoposide.     

Division polarity,development and configuration of microtubule arrays in Bryophyte meiosis

Summary Immunofluorescence and TEM studies of meiosis in two mosses (Bryophyta) provide evidence that the prophasic tetrahedral system of microtubules contributes directly to the metaphase I spindle. Intense staining of tubulin, conspicuously absent around the nuclear envelope, is first seen associated with plastids. By mid-prophase, microtubules radiate from the plastids to the nuclear envelope and become organized into six bands that interconnect the four plastids, forming a tetrahedral cytoskeleton surrounding the nucleus. During transition of prophase to metaphase, the four poles of the tetrahedral microtubule system converge in pairs toward opposite cleavage furrows. Opposite furrows occupy mutually perpendicular planes and the pair of microtubule focal points straddling one furrow lies at right angles to the pair straddling the opposite furrow. Additional microtubules terminate in numerous small clusters in the concave polar regions arching over the cleavage furrows. By early anaphase, the microtubule focal points lie very close to the division axis. We conclude that microtubules recruited from the prophasic quadripolar system are incorporated into the mature metaphase I spindle and the two principal focal points at each pole are those derived from poles of the prophasic quadripolar system.     

Unequal first cleavage in the Tubifex egg: involvement of a monastral mitotic apparatus

The first cleavage in the freshwater oligochaete Tubifex hattai is unequal and meridional, and produces a smaller cell AB and a larger cell CD. This study traces the process of furrow formation, reorganization of cortical F-actin and the assembly of a mitotic apparatus during this unequal division. Cleavage furrow formation consists of two stages: (i) when eggs are viewed from the animal pole, meridionally running furrows emerge at two points of the egg's equator that are 90° apart from each other and approach the egg axis as they deepen; and (ii) at the midpoint between the equator and the egg center, the bottoms of these furrows link to each other on the animal and vegetal surfaces of the egg and form a continuous ring of constriction in a plane parallel to the egg axis. Egg cortices, isolated during the first step and stained with rhodamine-phalloidin, show that the bottoms of recently formed furrows are underlaid by a belt of tightly packed actin bundles (i.e. a contractile arc). The transition to the second stage of furrow formation coincides with the conversion of these actin belts into a continuous ring of F-actin. Whole-mount immunocytochemistry of microtubules reveals that the first cleavage in Tubifex involves an asymmetric mitotic spindle, which initially possesses an aster at one pole but not the other. This ‘monastral’ spindle is located at the egg's center and orients itself perpendicular to the egg axis. During anaphase, astral rays elongate to reach the cell surface, so that the array of astral microtubules in the plane of the egg's equator covers a sector of 270–300°. In contrast, it is not until the transition to telophase that microtubules emanating from the anastral spindle pole approach the cell margin. If eggs are compressed along the egg axis or forced to elongate, they form monastral spindles and divide unequally. In living compressed eggs, mitotic spindles, which are recognizable as bright streaks at the egg's center, appear not to shift their position along the spindle axis during division, suggesting that without eccentric migration of spindles Tubifex eggs are able to divide unequally. These results suggest that mechanisms that translocate the mitotic spindle eccentrically do not operate in Tubifex eggs during the first cell cycle. The mechanisms that generate asymmetry in spindle organization are discussed in the light of the present results.     

中国南五味子属植物花粉形态的研究

本文用光学显微镜和扫描电子显微镜研究了国产南五味子属2个组7种花粉,记述了它们的形态,讨论了该属花粉形态的特征。该属花粉异极六沟,是属的鉴别上的一个重要特征,其中三条沟常在一极汇合成三合沟,本文报道亦有四合沟的现象,认为具合沟的一极是远极。南五味子属Kadsura Kaempf.ex Juss.集中分布于东亚至东南亚,约有24种,我国种类最为丰富,已知有13种,大都为药用经济植物。本属植物的花粉形态在国内尚未见有报道,本文试图通过对本属广布种和我国一些特有种的花粉形态的观察,探讨这个属花粉形态的特征,并与其他有关类群的花粉比较,给植物分类提供依据,为解决某些争议性问题提供资料,同时提出作者的观点。     

The organization of the mitotic apparatus poles in etoposide-treated CHO-K1 cells

In this study, we have examined the organization of the mitotic spindle poles in CHO-K1 cells dividing after treatment with the etoposide (1 h, 25 microM). We studied at various periods after the treatment: 1) the distribution of gamma-tubulin in mitotic cells by immunofluorescent staining; 2) the level of posttranslational modification of a-tubulin in the spindle microtubules by immunoelectron microscopy; 3) the ultrastructure of the mitotic apparatus poles by standard electron microscopy. In 48 h after the addition of the agent we identified considerable changes in the ultrastructure of poles in etoposide-treated CHO-K1 cells with bipolar and multipolar spindles. The number of centrioles increased. The centrioles were unevenly distributed among the poles, and some centrioles were not explicitly involved in the organization of mitotic spindle, furthermore they can differ in the number of outgrowing microtubules. Most centrioles were without fibrillar halo. In 48 h after the addition of etoposide, electron microscopy of cells after immunoperoxidase staining with antibodies to acetylated and tyrosinated alpha-tubulin has shown that different poles of a multipolar spindle within the same cell are stained differently for tyr-tubulin but not for acet-tubulin. Immunofluorescence staining for gamma-tubulin also points to different organization of poles in the same spindle. Our findings provide the first evidence that the pattern of immunostaning and the ultrastructure of mitotic apparatus poles differ in the cells dividing at various periods after etoposide treatment.