Simultaneous localization of myosin and tubulin in human tissue culture cells by double antibody staining

We have studied the distribution of myosin and tubulin molecules inside the same tissue culture cells by using two antibodies labeled with contrasting fluorochromes. Antimyosin raised against human platelet myosin was labeled with rhodamine. Antitubulin raised against sea urchin vinblastine-induced tubulin crystals was labeled with fluorescein. The two antibodies stained entirely different structures inside the same flat interphase cell: antimyosin bound to stress fibers and antitubulin bound to thin, wavy fibers thought to be individual microtubules. Compact interphase cells stained diffusely with both antibodies. From prophase through early anaphase both antibodies stained the mitotic spindle, although the fluorescence contrast between the spindle and the cytoplasm was much higher with antitubulin than with antimyosin. From anaphase through telophase, strong antimyosin staining occurred in the cleavage furrow, while antitubulin stained the region between the separated chromosomes. This study established the feasibility of high-resolution fluorescent antibody localization of pairs of motility proteins in the cytoplasm of single cells, an approach which will make it possible to map out the sites of the various contractile protein interactions in situ.     

Downregulation of protein 4.1R, a mature centriole protein, disrupts centrosomes, alters cell cycle progression, and perturbs mitotic spindles and anaphase

Centrosomes nucleate and organize interphase microtubules and are instrumental in mitotic bipolar spindle assembly, ensuring orderly cell cycle progression with accurate chromosome segregation. We report that the multifunctional structural protein 4.1R localizes at centrosomes to distal/subdistal regions of mature centrioles in a cell cycle-dependent pattern. Significantly, 4.1R-specific depletion mediated by RNA interference perturbs subdistal appendage proteins ninein and outer dense fiber 2/cenexin at mature centrosomes and concomitantly reduces interphase microtubule anchoring and organization. 4.1R depletion causes G(1) accumulation in p53-proficient cells, similar to depletion of many other proteins that compromise centrosome integrity. In p53-deficient cells, 4.1R depletion delays S phase, but aberrant ninein distribution is not dependent on the S-phase delay. In 4.1R-depleted mitotic cells, efficient centrosome separation is reduced, resulting in monopolar spindle formation. Multipolar spindles and bipolar spindles with misaligned chromatin are also induced by 4.1R depletion. Notably, all types of defective spindles have mislocalized NuMA (nuclear mitotic apparatus protein), a 4.1R binding partner essential for spindle pole focusing. These disruptions contribute to lagging chromosomes and aberrant microtubule bridges during anaphase/telophase. Our data provide functional evidence that 4.1R makes crucial contributions to the structural integrity of centrosomes and mitotic spindles which normally enable mitosis and anaphase to proceed with the coordinated precision required to avoid pathological events.     

Distribution of a matrix component of the midbody during the cell cycle in Chinese hamster ovary cells

Monoclonal antibodies were raised against isolated spindles of CHO (Chinese hamster ovary) cells to probe for molecular components specific to the mitotic apparatus. One of the antibodies, CHO1, recognized an antigen localized to the midbody during mitosis. Immunofluorescence staining of metaphase cells showed that although the total spindle area was labeled faintly, the antigen corresponding to CHO1 was preferentially localized in the equatorial region of the spindle. With the progression of mitosis, the antigen was further organized into discrete short lines along the spindle axis, and eventually condensed into a bright fluorescent dot at the midzone of the intercellular bridge between two daughter cells. Parallel immunostaining of tubulin showed that the CHO1-stained area corresponded to the dark region where microtubules are entrapped by the amorphous dense matrix components and possibly blocked from binding to tubulin antibody. Immunoblot analysis indicated that CHO1 recognized two polypeptides of mol wt 95,000 and 105,000. The immunoreaction was always stronger in preparations of isolated midbodies than in mitotic spindle fractions. The protein doublet was retained in the particulate matrix fraction after Sarkosyl extraction (Mullins, J. M., and J. R. McIntosh. 1982. J. Cell Biol. 94:654-661), suggesting that CHO1 antigen is indeed a component of the dense matrix. In addition to the equatorial region of spindles and midbodies, CHO1 also stained interphase centrosomes, and nuclei in a speckled pattern that was cell cycle-dependent. Thus, the midbody appears to share either common molecular component(s) or a similar epitope with interphase centrosomes and nuclei.     

Construction of three quadruple-fluorescent MDA435 cell lines that enable monitoring of the whole chromosome segregation process in the living state

Mitotic events from prophase to telophase are defined by morphology or movement of chromatin, nuclear envelope, centrosomes and spindles. Live-cell imaging is useful for characterizing the whole chromosome segregation process in the living state. In this study, we constructed three quadruple-fluorescent MDA435 cell lines in which chromatin, kinetochores, nuclear envelope and either inner centromere, microtubules or centrosomes/spindles were differentially visualized with cyan, green, orange and red fluorescent proteins (ECFP, EGFP, mKO and DsRed). Each mitotic stage of the individual cells could be identified by capturing live-cell images without the requirement of fixing or staining steps. In addition, we obtained four-color time-lapse images of one cell line, MDA-Auro/imp/H3/AF, from prophase to metaphase and from early anaphase to telophase. These quadruple-fluorescent cell lines will be useful for precisely analyzing the mitotic events from prophase through to telophase in single cells in the future.     

Phosphorylation of centrin during the cell cycle and its role in centriole separation preceding centrosome duplication

Once during each cell cycle, mitotic spindle poles arise by separation of newly duplicated centrosomes. We report here the involvement of phosphorylation of the centrosomal protein centrin in this process. We show that centrin is phosphorylated at serine residue 170 during the G(2)/M phase of the cell cycle. Indirect immunofluorescence staining of HeLa cells using a phosphocentrin-specific antibody reveals intense labeling of mitotic spindle poles during prophase and metaphase of the cell division cycle, with diminished staining of anaphase and no staining of telophase and interphase centrosomes. Cultured cells undergo a dramatic increase in centrin phosphorylation following the experimental elevation of PKA activity, suggesting that this kinase can phosphorylate centrin in vivo. Surprisingly, elevated PKA activity also resulted intense phosphocentrin antibody labeling of interphase centrosomes and in the concurrent movement of individual centrioles apart from one another. Taken together, these results suggest that centrin phosphorylation signals the separation of centrosomes at prophase and implicates centrin phosphorylation in centriole separation that normally precedes centrosome duplication.     

T-1, a mitotic arrester, alters centrosome configurations in fertilized sea urchin eggs

T-1 induces modifications in the shape of the centrosome at division in fertilized eggs of the North American sea urchin, Lytechinus pictus. Phase contrast microscopy observations of mitotic apparatus isolated from T-1-treated (1.7-8.5 microM) eggs at first division shows that the centrosomes already begin to spread or to separate by prophase and that the mitotic spindle is barrel-shaped. When eggs are fertilized with sperm that have been preteated with T-1, the centrosomes become flattened; the spindles are of normal length. Immunofluorescence microscopy using an anti-centrosomal monoclonal antibody reveals that T-1 modifies the structure of the centrosome so that barrel-shaped spindles with broad centrosomes are observed at metaphase, rather than the expected focused poles and fusiform spindle. Higher concentrations of T-1 induce fragmentation of centrosomes, causing abnormal accumulation of microtubules in polar regions. These results indicate that T-1 directly alters centrosomal configuration from a compact structure to a flattened or a spread structure. T-1 can be classified as a new category of mitotic drugs that may prove valuable in dissecting the molecular nature of centrosomes.     

Early endosomes, late endosomes, and lysosomes display distinct partitioning strategies of inheritance with similarities to Golgi-derived membranes

The pattern of inheritance of compartments of the endocytic pathway has been rarely reported, and the precise mechanism(s) are yet to be elucidated. We used antibodies reactive to early endosomes (anti-EEA1), late endosomes (anti-LBPA and anti-LAMP-1), lysosomes (anti-LAMP-1) and trans-Golgi network (TGN) (anti-GOLGA4) to examine the inheritance of these compartments in fixed human HEp-2 cells. Prior to entering M phase, these compartments display a perinuclear bias in their cytoplasmic distribution with areas of local accumulation juxtaposed to the centrosome. The location of these compartments during mitosis was examined relative to each other, the chromosomes, centrosomes and the microtubule network. During M phase early endosomes and TGN-derived compartments share overlapping subcellular distributions. A portion of these compartments display discernible clustering around the separated and migrating centrosomes in prophase. At metaphase these compartments co-localise with the mitotic spindle, are absent at the metaphase plate and do not overlay the astral microtubules. At anaphase these compartments are concentrated between shortening kinetochore microtubules and centrosomes. In addition, they appear distributed over the elongating polar microtubules in the body of the cell. From telophase and into cytokinesis these compartments concentrate around the minus ends of the constricted remnants of polar spindle microtubules and re-establish a prominent presence juxtaposed to the centrosome. In contrast, there is little evidence of movement of late endosomes and lysosomes with migrating centrosomes in prophase, and these compartments are excluded from the mitotic spindle at metaphase. However, by the end of telophase, the subcellular distribution of a portion of late endosomes and lysosomes share overlapping distributions with that of early endosomes. We conclude a portion of endosomal compartments and Golgi-derived membranes undergo ordered partitioning based on the centrosome and mitotic spindle.     

The structure of the mitotic spindle and nucleolus during mitosis in the amebo-flagellate Naegleria

Mitosis in the amebo-flagellate Naegleria pringsheimi is acentrosomal and closed (the nuclear membrane does not break down). The large central nucleolus, which occupies about 20% of the nuclear volume, persists throughout the cell cycle. At mitosis, the nucleolus divides and moves to the poles in association with the chromosomes. The structure of the mitotic spindle and its relationship to the nucleolus are unknown. To identify the origin and structure of the mitotic spindle, its relationship to the nucleolus and to further understand the influence of persistent nucleoli on cellular division in acentriolar organisms like Naegleria, three-dimensional reconstructions of the mitotic spindle and nucleolus were carried out using confocal microscopy. Monoclonal antibodies against three different nucleolar regions and α-tubulin were used to image the nucleolus and mitotic spindle. Microtubules were restricted to the nucleolus beginning with the earliest prophase spindle microtubules. Early spindle microtubules were seen as short rods on the surface of the nucleolus. Elongation of the spindle microtubules resulted in a rough cage of microtubules surrounding the nucleolus. At metaphase, the mitotic spindle formed a broad band completely embedded within the nucleolus. The nucleolus separated into two discreet masses connected by a dense band of microtubules as the spindle elongated. At telophase, the distal ends of the mitotic spindle were still completely embedded within the daughter nucleoli. Pixel by pixel comparison of tubulin and nucleolar protein fluorescence showed 70% or more of tubulin co-localized with nucleolar proteins by early prophase. These observations suggest a model in which specific nucleolar binding sites for microtubules allow mitotic spindle formation and attachment. The fact that a significant mass of nucleolar material precedes the chromosomes as the mitotic spindle elongates suggests that spindle elongation drives nucleolar division.     

Clathrin in mitotic spindles

Subconfluent cultures ofMadin-Darby canine kidney (MDCK) and CV-1 cells were immunostained withtwo monoclonal antibodies (MAbs), MAb X-22 and MAb 23, against clathrinheavy chain and with polyclonal antiserum against a conserved region ofall mammalian clathrin light chains. In interphase MDCK and CV-1 cells,staining by all three antibodies resulted in the characteristicintracellular punctate vesicular and perinuclear staining pattern. Inmitotic cells, all three anti-clathrin antibodies strongly stained the mitotic spindle. Staining of clathrin in the mitotic spindle was colocalized with anti-tubulin staining of microtubular arrays in thespindle. Staining of the mitotic spindle was evident in mitotic cellsfrom prometaphase to telophase and in spindles in mitotic cellsreleased from a thymidine-nocodazole block. In CV-1 cells, staining ofclathrin in the mitotic spindle was not affected by brefeldin A. OnWestern blots, clathrin was detected, but not enriched, in isolatedspindles. The immunodetection of clathrin in the mitotic spindle maysuggest a novel role for clathrin in mitosis. Alternatively, therecruitment of clathrin to the spindle may suggest a novel regulatorymechanism for localization of clathrin in mitotic cells.

     

TRANSIENT LOCALIZATION OF γ-TUBULIN AROUND THE CENTRIOLES IN THE NUCLEAR DIVISION OF BOERGESENIA FORBESII (SIPHONOCLADALES, CHLOROPHYTA)

Mitosis in Boergesenia forbesii (Harvey) Feldman was studied by immunofluorescence microscopy using anti-β–tubulin, anti-γ–tubulin, and anti-centrin antibodies. In the interphase nucleus, one, two, or rarely three anti-centrin staining spots were located around the nucleus, indicating the existence of centrioles. Microtubules (MTs) elongated randomly from the circumference of the nuclear envelope, but distinct microtubule organizing centers could not be observed. In prophase, MTs located around the interphase nuclei became fragmented and eventually disappeared. Instead, numerous MTs elongated along the nuclear envelope from the discrete anti-centrin staining spots. Anti-centrin staining spots duplicated and migrated to the two mitotic poles. γ–Tubulin was not detected at the centrioles during interphase but began to localize there from prophase onward. The mitotic spindle in B. forbesii was a typical closed type, the nuclear envelope remaining intact during nuclear division. From late prophase, accompanying the chromosome condensation, spindle MTs could be observed within the nuclear envelope. A bipolar mitotic spindle was formed at metaphase, when the most intense staining of γ-tubulin around the centrioles could also be seen. Both spindle MT poles were formed inside the nuclear envelope, independent of the position of the centrioles outside. In early anaphase, MTs between separating daughter chromosomes were not detected. Afterward, characteristic interzonal spindle MTs developed and separated both sets of the daughter chromosomes. From late anaphase to telophase, γ-tubulin could not be detected around the centrioles and MT radiation from the centrioles became diminished at both poles. γ-Tubulin was not detected at the ends of the interzonal spindle fibers. When MTs were depolymerized with amiprophos methyl during mitosis, γ-tubulin localization around the centrioles was clearly confirmed. Moreover, an influx of tubulin molecules into the nucleus for the mitotic spindle occurred at chromosome condensation in mitosis.     

Cell division of binuclear cells induced by caffeine: Spindle organization and determination of division plane

Synchronously dividing binuclear cells were induced in root tips ofTriticum turgidum by caffeine treatment. Spindle and other microtubular configurations of such cells were studied using tubulin immunofluorescence and electron microscopy. The binuclear cells developed one, two or three preprophase microtubule bands longitudinally, transversely or rarely in a cross configuration. During the mitotic entry binuclear cells formed prophase spindles separately around each nucleus. When the nuclei were located fairly apart, their spindle structures developed independently throughout all mitotic phases. But when the nuclei were located closely together their metaphase and anaphase spindles shared a common polar region. However, the two spindles in such cells retained their functional autonomy. They display structurally independent minipoles in the common polar region. After anaphase the neighbouring nonsister chromosome groups of nuclei divided by a common polar region come to lie close together and in telophase, become enclosed by a common nuclear envelope. During cytokinesis of binuclear cells cell plates were formed only between sister nuclei. These cell plates may develop normally or may curve or branch giving rise to aberrant daughter cell walls. The peculiar mode of spindle and spindle polar region organization of binuclear cells and determination of the division plane in them are discussed.     

Spindles and centrosomes during male meiosis in Drosophila melanogaster

We have studied the spatial distribution of chromosomes, spindle fibers and centrosomes throughout the first meiotic division in males of Drosophila melanogaster. There seem to be two different types of spindle fibers: those which connect the poles to the chromosomes, and others arranged as cup-shaped hemispheres that reach from the poles to an unstained area on the equator of the cell. These pole-equator fibers could be responsible for positioning the nucleus and distributing cytoplasmic organelles around the nucleus during prophase, so that after meiosis, the daughter cells are provided with equal amounts of preorganized cytoplasmic organelles. These fibers remain until after the daughter nuclei have formed during telophase. An antigen associated with the centrosomes of mitotic spindles appears during meiosis as dispersed particles surrounding the nucleus; these particles might provide the developing spermatids with microtubule-organizing centers.     

ULTRASTRUCTURE AND TIME COURSE OF MITOSIS IN THE FUNGUS FUSARIUM OXYSPORUM

Mitosis in Fusarium oxysporum Schlect. was studied by light and electron microscopy. The average times required for the stages of mitosis, as determined from measurements made on living nuclei, were as follows: prophase, 70 sec; metaphase, 120 sec; anaphase, 13 sec; and telophase, 125 sec, for a total of 5.5 min. New postfixation procedures were developed specifically to preserve the fine-structure of the mitotic apparatus. Electron microscopy of mitotic nuclei revealed a fibrillo-granular, extranuclear Spindle Pole Body (SPB) at each pole of the intranuclear, microtubular spindles. Metaphase chromosomes were attached to spindle microtubules via kinetochores, which were found near the spindle poles at telophase. The still-intact, original nuclear envelope constricted around the incipient daughter nuclei during telophase.     

Tubulin composition and microtubule nucleation of a griseofulvin-resistant Chinese hamster ovary cell mutant with abnormal spindles

A griseofulvin-resistant Chinese hamster ovary (CHO) mutant (Grs-2) which has an altered beta-tubulin subunit as well as wild-type beta-tubulin is temperature-sensitive (ts) for growth at 40.5 degrees C. This growth defect appears to result from the formation of abnormal mitotic spindles at the non-permissive temperature (Abraham, I et al., J cell biol 97 (1983) 1055) [19]. Light microscopy of spindles isolated from mutant cells cultured at the permissive temperature showed a typical bipolar morphology, whereas spindles isolated at the non-permissive temperature were multipolar. In order to study the role of tubulin in spindle formation, we analyzed the tubulin composition of the multipolar spindles. Two-dimensional gels and immunoblotting analysis of one-dimensional electrophoretic gels stained with monoclonal anti-Chinese hamster brain beta-tubulin antibody revealed that both mutant and wild-type beta-tubulins were present in similar proportions in both bipolar spindles at 37 degrees C and multipolar spindles at 40.5 degrees C. The ratio between wild-type and mutant tubulin in spindles was also found to be the same as in the cytoplasmic microtubule network in interphase cells, providing evidence that the mutant beta-tubulin appeared to be incorporated in a similar manner into both interphase and mitotic microtubule structures. In vitro microtubule polymerization onto centrosomes prepared from mutant Grs-2 demonstrated that 80% of the sites for microtubule nucleation were without centrioles, suggesting fragmentation of pericentriolar material away from centrioles. This may be one of the causes of multipolar spindle formation in the mutant cells. These results, therefore, suggest that abnormal formation of spindles in mutant cells is due not to the presence of the mutant tubulin per se, but to the abnormal behavior of this mutant tubulin in the cellular environment during mitosis or abnormal interaction with other components in the spindle at 40.5 degrees C.     

AlphaB-crystallin phosphorylated at Ser-59 is localized in centrosomes and midbodies during mitosis

We have reported that the three serine residues in alphaB-crystallin are phosphorylated under various stress conditions. We prepared affinity-purified antibodies recognizing each of the phosphorylated serine residues (Ser-19, Ser-45, and Ser-59, respectively) in alphaB-crystallin with peptides (p19S, p45S, or p59S) that contained the corresponding phosphorylated serine residue. Immunocytochemically anti-p45S antibodies stained the cytoplasm of mitotic cells (J. Biol. Chem. 273, 28,346-28,354). We have now found that the anti-p59S antibodies recognize centrosomes and midbodies of dividing cells. alphaB-Crystallin was the only protein recognized by the anti-p59S antibodies in Western blot analyses of isolated centrosome fractions. alphaB-Crystallin phosphorylated at Ser-59 was localized at the microtubule organizing centers by means of double staining with anti-beta-tubulin antibody in aster formation analysis and was co-localized with gamma-tubulin in centrosomes. Gamma-Tubulin was co-immunoprecipitated with alphaB-crystallin in U373 glioma cell extracts. On the other hand, the location of the phosphorylated alphaB-crystallin deviated from that of alpha-tubulin or gamma-tubulin in the midbody region. Taken together with the evidences that several chaperones are distributed to centrosomes, these results suggest that alphaB-crystallin as a chaperone might be also involved in the quality control of proteins.     

Cytoplasmic dynein plays a role in mammalian mitotic spindle formation

The formation and functioning of a mitotic spindle depends not only on the assembly/disassembly of microtubules but also on the action of motor enzymes. Cytoplasmic dynein has been localized to spindles, but whether or how it functions in mitotic processes is not yet known. We have cloned and expressed DNA fragments that encode the putative ATP- hydrolytic sites of the cytoplasmic dynein heavy chain from HeLa cells and from Dictyostelium. Monospecific antibodies have been raised to the resulting polypeptides, and these inhibit dynein motor activity in vitro. Their injection into mitotic mammalian cells blocks the formation of spindles in prophase or during recovery from nocodazole treatment at later stages of mitosis. Cells become arrested with unseparated centrosomes and form monopolar spindles. The injected antibodies have no detectable effect on chromosome attachment to a bipolar spindle or on motions during anaphase. These data suggest that cytoplasmic dynein plays a unique and important role in the initial events of bipolar spindle formation, while any later roles that it may play are redundant. Possible mechanisms of dynein's involvement in mitosis are discussed.     

Immunocytochemical localization of glucocorticoid receptor in human gingival fibroblasts and evidence for a colocalization of glucocorticoid receptor with cytoplasmic microtubules

The cellular distribution of the glucocorticoid receptor (GR) in relation to various intracellular and plasma membrane structures in human fibroblasts was studied using indirect immunofluorescence techniques with monoclonal and polyclonal antibodies. During interphase, GR was located predominantly in the cytoplasm, showing a similar pattern as tubulin. In mitotic cells, GR and tubulin were localized in mitotic spindles and in telophase midbodies. Colchicine and vinblastine induced a similar redistribution of GR and tubulin to the cell periphery. This redistribution was reversible for colchicine but not for vinblastine. Vinblastine also induced paracrystals containing GR and tubulin. These results support the hypothesis that GR interacts in vivo with cytoplasmic microtubules.     

p53 localization at centrosomes during mitosis and postmitotic checkpoint are ATM-dependent and require serine 15 phosphorylation

We recently demonstrated that the p53 oncosuppressor associates to centrosomes in mitosis and this association is disrupted by treatments with microtubule-depolymerizing agents. Here, we show that ATM, an upstream activator of p53 after DNA damage, is essential for p53 centrosomal localization and is required for the activation of the postmitotic checkpoint after spindle disruption. In mitosis, p53 failed to associate with centrosomes in two ATM-deficient, ataxiatelangiectasia-derived cell lines. Wild-type ATM gene transfer reestablished the centrosomal localization of p53 in these cells. Furthermore, wild-type p53 protein, but not the p53-S15A mutant, not phosphorylatable by ATM, localized at centrosomes when expressed in p53-null K562 cells. Finally, Ser15 phosphorylation of endogenous p53 was detected at centrosomes upon treatment with phosphatase inhibitors, suggesting that a p53 dephosphorylation step at centrosome contributes to sustain the cell cycle program in cells with normal mitotic spindles. When dissociated from centrosomes by treatments with spindle inhibitors, p53 remained phosphorylated at Ser15. AT cells, which are unable to phosphorylate p53, did not undergo postmitotic proliferation arrest after nocodazole block and release. These data demonstrate that ATM is required for p53 localization at centrosome and support the existence of a surveillance mechanism for inhibiting DNA reduplication downstream of the spindle assembly checkpoint     

ZYG-9, A Caenorhabditis elegans Protein Required for Microtubule Organization and Function, Is a Component of Meiotic and Mitotic Spindle Poles

We describe the molecular characterization of zyg-9, a maternally acting gene essential for microtubule organization and function in early Caenorhabditis elegans embryos. Defects in zyg-9 mutants suggest that the zyg-9 product functions in the organization of the meiotic spindle and the formation of long microtubules. One-cell zyg-9 embryos exhibit both meiotic and mitotic spindle defects. Meiotic spindles are disorganized, pronuclear migration fails, and the mitotic apparatus forms at the posterior, orients incorrectly, and contains unusually short microtubules. We find that zyg-9 encodes a component of the meiotic and mitotic spindle poles. In addition to the strong staining of spindle poles, we consistently detect staining in the region of the kinetochore microtubules at metaphase and early anaphase in mitotic spindles. The ZYG-9 signal at the mitotic centrosomes is not reduced by nocodazole treatment, indicating that ZYG-9 localization to the mitotic centrosomes is not dependent upon long astral microtubules. Interestingly, in embryos lacking an organized meiotic spindle, produced either by nocodazole treatment or mutations in the mei-1 gene, ZYG-9 forms a halo around the meiotic chromosomes. The protein sequence shows partial similarity to a small set of proteins that also localize to spindle poles, suggesting a common activity of the proteins.     

Indirect immunofluorescence of microtubules in Dictyostelium discoideum. A study with polyclonal and monoclonal antibodies to tubulins

Amebae of D. discoideum on coverslips were fixed in situ with glutaraldehyde and permeabilized with Triton X-100. Of six antibodies tested, only a monoclonal antibody to yeast tubulin consistently gave bright fluorescence. Counterstaining with DAPI facilitated the identification of interphase and mitotic stages. Most microtubules (MTs) in interphase amebae emanated from a nucleus-associated centre that had a non-fluorescent core. Amebae in early stages of mitosis lacked cytoplasmic MTs almost entirely. The nascent spindle in prophase appeared as a brightly fluorescent dot, whereas the prometaphase spindle was a short rod. Spindles in metaphase and anaphase nuclei were more elongate, some consisting of several fluorescent lines. Astral MTs were prominent on spindles in anaphase and telophase. Asters are obviously converted to the interphase complex of MTs in post-mitotic cells, while the shaft-like remnant of the central spindle disappears. The cyclical changes in the MT system related to cell division resemble those observed in higher eukaryotes and probably reflect changes in the locomotory behavior of the amebae rather than changes in cell shape.