Gamma-tubulin is present in Drosophila melanogaster and Homo sapiens and is associated with the centrosome

The origins and evolution of eukaryotic proteins

The common ancestry of eukaryotes, archaebacteria and eubacteria is well demonstrated by amino acid sequence comparisons of numerous proteins that are common to all three groups. On the other hand, there are a few proteins, like ubiquitin, that are common to eukaryotes and archaebacteria and which have yet to be observed in eubacteria. Some proteins appear to be wholly restricted to eukaryotes; this is especially true of cytoskeletal proteins. Recently, actin has been found by crystallography to be homologous with an ATP-binding domain found in a heat shock protein and several other proteins common to all three urkingdoms. This observation is puzzling on several counts. Most cytoskeletal proteins like actin and tubulin are very slow changing and must have been so for a very long time. How is it, then, that no sequence resemblance can be discerned with their alledged prokaryotic antecedents? The question is addressed by considering two bacterial fts proteins which appear to be related to actin, on the one hand, and tubulin, on the other. One answer may be that the rate of change of these proteins changed dramatically at a key point in their history. Another possibility is that eukaryotes are much older than some of their other proteins indicate.

Three-dimensional structural characterization of centrosomes from early Drosophila embryos

An understanding of the mechanism and structure of microtubule (MT)-nucleating sites within the pericentriolar material (PCM) of the centrosome has been elusive. This is partly due to the difficulty in obtaining large quantities of centrosomes for analysis, as well as to the problem of attaining interpretable structural data with conventional EM techniques. We describe a protocol for isolating a large quantity of functional centrosomes from early Drosophila embryos. Using automated electron tomography, we have begun a three-dimensional structural characterization of these intact centrosomes with and without regrown MTs. Reconstructions of the centrosomes to approximately 6-8 nm resolution revealed no large structures at the minus ends of MTs, suggesting that if MT-nucleating material physically contacts the MTs, it must conform closely to the shape of the minus end. While many MTs originate near the centrioles, MT minus ends were found throughout the PCM, and even close to its outer boundary. The MTs criss-crossed the PCM, suggesting that nucleating sites are oriented in many different directions. Reconstructions of centrosomes without MTs suggest that there is a reorganization of the PCM upon MT regrowth; moreover, ring-like structures that have a similar diameter as MTs are apparent in the PCM of centrosomes without MTs, and may be MT-nucleating sites.

A transient association of gamma-tubulin at the midbody is required for the completion of cytokinesis during the mammalian cell division

gamma-Tubulin, a relatively new member of the tubulin gene family, is localized primarily at the centrosome throughout the mammalian cell cycle and may play a key role in nucleation of cellular microtubule assembly. A transient association of gamma-tubulin at the cytoplasmic bridge of telophase mammalian cells, the midbody, is recently documented. Using immunogold electron microscopy and serial section reconstruction analysis, we show here that the transiently associated midbody gamma-tubulin is localized at the minus ends of microtubules in the midbody structure. Using antisense RNA methods we also demonstrate that a selective depletion of transiently associated midbody gamma-tubulin causes an abortive cytokinesis due to a failure in the morphogenesis of the midbody structure.

Gamma-tubulin can both nucleate microtubule assembly and self-assemble into novel tubular structures in mammalian cells

alpha-, beta-, and gamma-tubulins are evolutionarily highly conserved members of the tubulin gene superfamily. While the abundant members, alpha- and beta-tubulins, constitute the building blocks of cellular microtubule polymers, gamma-tubulin is a low abundance protein which localized to the pericentriolar material and may play a role in microtubule assembly. To test whether gamma-tubulin mediates the nucleation of microtubule assembly in vivo, and co-assembles with alpha- and beta-tubulins into microtubules or self-assembles into macro-molecular structures, we experimentally elevated the expression of gamma-tubulin in the cell cytoplasm. In most cells, overexpression of gamma-tubulin causes a dramatic reorganization of the cellular microtubule network. Furthermore, we show that when overexpressed, gamma-tubulin causes ectopic nucleation of microtubules which are not associated with the centrosome. In a fraction of cells, gamma-tubulin self-assembles into novel tubular structures with a diameter of approximately 50 nm (named gamma-tubules). Furthermore, unlike microtubules, gamma-tubules are resistant to cold or drug induced depolymerization. These data provide evidence that gamma-tubulin can cause nucleation of microtubule assembly and can self-assemble into novel tubular structures.

Rbl2p, a yeast protein that binds to beta-tubulin and participates in microtubule function in vivo

Genetic configurations resulting in high ratios of beta-tubulin to alpha-tubulin are toxic in S. cerevisiae, causing microtubule disassembly and cell death. We identified three non-tubulin yeast genes that, when overexpressed, rescue cells from excess beta-tubulin. One, RBL2, rescues beta-tubulin lethality as efficiently as does alpha-tubulin. Rbl2p binds to beta-tubulin in vivo. Deficiencies or excesses of either Rbl2p or alpha-tubulin affect microtubule-dependent functions in a parallel fashion. Rbl2p has functional homology with murine cofactor A, a protein important for in vitro assays of beta-tubulin folding. The results suggest that Rbl2p participates in microtubule morphogenesis but not in the assembled polymer.

Polar organization of gamma-tubulin in acentriolar mitotic spindles of Drosophila melanogaster cells

The spindle pole localization of gamma-tubulin was compared in wild type and acentriolar cultured Drosophila cells using polyclonal antibodies specifically raised against the carboxy terminal amino acid sequence of Drosophila gamma-tubulin-1 (-KSEDSRSVTSAGS). During interphase, gamma-tubulin was present in the centrosome of wild type cells and accumulated around this organelle in a cell cycle dependent manner. In contrast, no such structure was observed in acentriolar cells. Wild type mitoses were homogeneously composed of biconical spindles, with two centrosome-associated gamma-tubulin spots at the poles. The mitotic apparatuses observed in the acentriolar cells were heterogeneous; multipolar mitoses, bipolar mitoses with a barrel-shaped spindle and bipolar mitoses with biconical spindles were observed. In acentriolar cells, gamma-tubulin accumulation at mitotic poles was dependent on spindle microtubule integrity. Most acentriolar spindles presented a dispersed gamma-tubulin labeling at the poles. Only well polarized and biconical acentriolar spindles showed a strong gamma-tubulin polar spot. Finally, acentriolar mitotic poles were not organized around true centrosomes. In contrast to wild type cells, in acentriolar cells the Bx63 centrosome-associated antigen was absent and the gamma-tubulin containing material dispersed readily following microtubule disassembly. These observations confirm that gamma-tubulin plays an essential role in the nucleation of microtubules even in the absence of mitotic polar organelles. In addition the data suggest that the mechanisms involved in the bipolarization of wild type and acentriolar mitoses are different, and that centrioles play a role in the spatial organization of the nucleating material containing gamma-tubulin.

The impact of chromosomes and centrosomes on spindle assembly as observed in living cells

We analyzed the role that chromosomes, kinetochores, and centrosomes play in spindle assembly in living grasshopper spermatocytes by reconstructing spindles lacking certain components. We used video-enhanced, polarization microscopy to distinguish the effect of each component on spindle microtubule dynamics and we discovered that both chromosomes and centrosomes make potent and very different contributions to the organization of the spindle. Remarkably, the position of a single chromosome can markedly affect the distribution of microtubules within a spindle or even alter the fate of spindle assembly. In an experimentally constructed spindle having only one chromosome, moving the chromosome to one of the two poles induces a dramatic assembly of microtubules at the nearer pole and a concomitant disassembly at the farther pole. So long as a spindle carries a single chromosome it will persist normally. A spindle will also persist even when all chromosomes are detached and then removed from the cell. If, however, a single chromosome remains in the cell but is detached from the spindle and kept in the cytoplasm, the spindle disassembles. One might expect the effect of chromosomes on spindle assembly to relate to a property of a specific site on each chromosome, perhaps the kinetochore. We have ruled out that possibility by showing that it is the size of chromosomes rather than the number of kinetochores that matters. Although chromosomes affect spindle assembly, they cannot organize a spindle in the absence of centrosomes. In contrast, centrosomes can organize a functional bipolar spindle in the absence of chromosomes. If both centrosomes and chromosomes are removed from the cell, the spindle quickly disappears.

Identification of intrinsic dimer and overexpressed monomeric forms of gamma-tubulin in Sf9 cells infected with baculovirus containing the Chlamydomonas gamma-tubulin sequence

A new member of the tubulin superfamily, gamma-tubulin, is localized at microtubule-organizing centers (MTOCs) in a variety of organisms. Chlamydomonas cDNA coding for the full-length sequence of gamma-tubulin was expressed in insect ovarian Sf9 cells using the baculovirus expression system. Approximately half of the induced 52 kDa gamma-tubulin was recovered in the supernatant after centrifugation of Sf9 cell lysates at 18,000 g for 15 minutes. When the cell supernatant was analyzed by FPLC on a Superdex 200 sizing column, Chlamydomonas gamma-tubulin separated into two major peaks. The lagging peak contained a monomeric form of gamma-tubulin with a sedimentation coefficient of 2.5 S, which interacted with the Superdex column in a salt-dependent manner. The leading peak, with an apparent molecular mass of 900 kDa, corresponded to a molecular chaperonin complex, and TCP1 chaperonin released folded gamma-tubulin polypeptide from the complex in the presence of MgATP. The released gamma-tubulin monomers were capable of binding to microtubules in vitro and biochemical quantities of active monomers were further purified using a combination of size-exclusion and ion-exchange column chromatography. The endogenous Sf9 cell gamma-tubulin migrated faster than Chlamydomonas gamma-tubulin with an apparent molecular mass of 49 kDa on gels. Analyses on gel filtration and sucrose density gradient centrifugation showed that, while overexpressed Chlamydomonas gamma-tubulin was present in a monomeric form, endogenous gamma-tubulin from Sf9 and HeLa cells exists as a dimer. These results may suggest the possibility that gamma-tubulin could form a heterodimer with hitherto unknown molecule(s).

Identification of an HSP70-related protein associated with the centrosome from dinoflagellates to human cells

The monoclonal antibody CTR210 raised against isolated human centrosomes strongly decorates the centrosome and more weakly a domain congruent with the Golgi apparatus in several animal cells (HeLa, 3T3, CHO, PtK2). Both decorations resist Triton extraction in conditions which totally extract the Golgi apparatus, as judged by galactosyltransferase decoration. A 67 kDa centrosomal antigen can be demonstrated in human cells with this antibody. CTR210 also decorates the centrosome or associated structures in several systems, including unicellular eukaryotes such as dinoflagellates or ciliates. A 72 kDa antigen has been identified and purified from the dinoflagellate C. cohnii and its NH2-terminal sequence partially established. It shows a close homology with HSP70 proteins. The possibility that the 72 kDa antigen belongs to this chaperone family was further supported using a mAb reacting, in most species, with HSP70. A polyclonal antibody raised against the 72 kDa antigen from C. cohnii decorates the centrosome in human cells and reacts with the CTR210 centrosomal 67 kDa antigen. These results suggest that specific chaperone proteins are associated with the centrosome in eukaryotic cells. The centrosomal chaperones could participate in the microtubule nucleation reaction or in the process of centrosome assembly.

Novel centrosomal protein reveals the presence of multiple centrosomes in turkey (Meleagris gallopavo) bnbn binucleated erythrocytes

The phenotype of the bnbn hemolytic anemia mutation in the domestic turkey is manifested as binucleation specifically in the definitive erythrocyte lineage, most likely as the consequence of anomolous centrosomal activity (Bloom et al., 1970; Searle and Bloom, 1979). Here we have identified in turkey two variants of the novel, centrosomally-associated erythroid-specific protein p23. One variant is Ca(2+)-sensitive and is highly homologous to its chick counterpart (Zhu et al., 1995, accompanying paper). The other, p21 is a truncated form resulting from a 62 amino acid deletion from the 3′ end and a 40 amino acid insertion at the 5′ end, and appears to lack Ca(2+)-sensitivity. These proteins are localized at the marginal band, centrosomes and nuclear membrane of differentiated erythrocytes. Anti-p23/p21 immunofluorescence revealed the presence of multiple centrosomes in bnbn erythrocytes. We therefore undertook a detailed genetic analysis to determine whether the p21 variant represented the bn mutation. Initial tests of normal BnBn and mutant bnbn individuals suggested that the p23/p21 proteins might be encoded by the Bn/bn genes. However, further genetic tests demonstrated independent segregation for these two genetic loci. Thus, these proteins are encoded by the heretofore undescribed genes, p23/p21, mapping to an autosomal locus in the turkey genome.

Polarity and nucleation of microtubules in polarized epithelial cells

Microtubules oriented in the apicobasal axis of columnar epithelial cells are arranged with a uniform polarity with minus ends toward the apical surface, suggesting that these cytoskeletal filaments might serve as a substrate for polarized movement of membrane vesicles within the cell. It is not known whether hepatocytes, a cuboidal epithelium in which transcellular transport is a requisite step in normal apical membrane biogenesis, contain microtubules arranged with a similar polarity. In the present study, we explore the question of microtubule polarity and possible mechanisms for nucleation in the epithelial cell lines WIF-B (hepatocyte), Caco-2 (intestine), and Madin-Darby canine kidney (MDCK). Caco-2 microtubules in the apicobasal axis had uniform polarity with minus ends nearest the apical surface. After cold and nocodazole-induced depolymerization, microtubule regrowth initiated in the apical region in all three cell types. The apex of WIF-B and Caco-2 cells contained two pools of gamma-tubulin: one associated with centrosomes and the other delocalized under the apical membrane. Non-centrosomal gamma-tubulin was present in complexes that sedimented between 10S and 29S; both forms could bind microtubules. The presence of both centrosomal and noncentrosomal gamma-tubulin in apical cytoplasm suggest multiple mechanisms by which microtubule nucleation might occur in epithelial cells.

The paternal inheritance of the centrosome, the cell's microtubule-organizing center, in humans, and the implications for infertility

Successful fertilization in humans, achieved when parental chromosomes intermix at first mitosis, requires centrosome restoration and microtubule-mediated motility. Imaging of inseminated human oocytes reveals that the sperm introduces the centrosome. The centrosome then nucleates the new microtubule assembly to form the sperm aster--a step essential for successful fertilization. Oocytes from some infertile patients failed to complete fertilization because of defects in uniting the sperm and egg nuclei, indicating that failure to properly effect the cytoplasmic motions uniting the nuclei results in human infertility. These discoveries have important implications for infertility diagnosis and managing reproduction.

Construction of a transcription map surrounding the BRCA1 locus of human chromosome 17

We have used a combination of methods (exon amplification, direct selection, direct screening, evolutionary conservation, island rescue-PCR, and direct sequence analysis) to survey approximately 600 kb of genomic DNA surrounding the BRCA1 gene for transcribed sequences. We have cloned a set of fragments representing at least 26 genes. The DNA sequence of these clones reveals that 5 are previously cloned genes; the precise chromosomal location of 2 was previously unknown, and 3 have been cloned and mapped by others to this interval. Three other genes, including BRCA1 itself, have recently been mapped independently to this region. Sequences from 11 genes are similar but not identical matches to known genes; 5 of these appear to be the human homologues of genes cloned from other species. Another 7 genes have no similarity with known genes. In addition, 39 putative exons and 14 expressed sequence tags have been identified and mapped to individual cosmids. This transcript map provides a detailed description of gene organization for this region of the genome.

22 genes from chromosome 17q21: cloning, sequencing, and characterization of mutations in breast cancer families and tumors

In our effort to identify BRCA1, 22 genes were cloned from a 1-Mb region of chromosome 17q21 defined by meiotic recombinants in families with inherited breast and/or ovarian cancer. Subsequent discovery of another meiotic recombinant narrowed the region to approximately 650 kb. Genes were cloned from fibroblast and ovarian cDNA libraries by direct screening with YACs and cosmids. The more than 400 cDNA clones so identified were mapped to cosmids, YACs, and P1 clones and to a chromosome 17 somatic panel informative for the BRCA1 region. Clones that mapped back to the region were hybridized to each other and consolidated into clusters reflecting 22 genes. Ten genes were known human genes, 5 were human homologs of known genes, and 7 were novel. Each gene was sequenced, compared to genes in the databases to find homologies, and analyzed for mutations in BRCA1-linked families and tumors. Eight mutations were found in tumors or families and not in controls. In the gene encoding alpha-N-acetylglucosaminidase, approximately 100 kb proximal to the 650-kb linked region, somatic nonsense, missense, and splice junction mutations occurred in 3 breast tumors, but not in these patients' germline DNA nor in controls. In an ets-related oncogene in the linked region, a missense mutation cosegregated with breast cancer in one family and was not observed in controls. In a human homolog of a yeast pre-mRNA splicing factor, 3 different mutations cosegregated with breast cancer in 3 families and were not observed in controls. In these and the other genes in the region, 36 polymorphic variants were observed in both cases and controls.

Organization of microtubules in myelinating Schwann cells

Myelinating Schwann cells polarize their surface membrane into several ultrastructurally and biochemically distinct domains that constitute the myelin internode. Formation of these membrane domains depends on contact with appropriate axons and requires microtubule-based transport systems for site-specific targeting of membrane components. Because little is known about microtubules in myelinating Schwann cells, this study used confocal microscopy and the microtubule hook-labelling method to characterize microtubule distribution, the location of microtubule nucleation sites, and the polarity and composition of Schwann cell microtubules. In myelinating Schwann cells, microtubules were abundant within the Golgi-rich perinuclear cytoplasm; they were not attached to the centrosome. Three-fourths of the microtubules in the cytoplasmic channels located along the outer perimeter of the myelin internode had their (+) ends oriented away from the perinuclear region, whereas the remaining 25% had the opposite polarity. Depolymerization/repolymerization experiments detected microtubule nucleating sites in perinuclear cytoplasm but not along the myelin internode. Taken together, these results indicate that microtubule-mediated transport of myelin components along the internode could utilize both (+)- and (-)-end motors. Specialized microtubule tracks that target myelin proteins to specific sites were not identified on the basis of tubulin polarity or posttranslational modifications.

Recruitment of antigenic gamma-tubulin during mitosis in animal cells: presence of gamma-tubulin in the mitotic spindle

It has been claimed repeatedly that gamma-tubulin is exclusively localized at the spindle poles in mitotic animal cells, where it plays a role in microtubule nucleation. In addition to this localization, we have observed a gamma-tubulin-specific staining of the mitotic spindle in several animal cells (human, kangaroo rat, mouse, Chinese hamster, Xenopus and Drosophila) using five polyclonal antibodies raised against unique gamma-tubulin sequences and four different fixation protocols. In HeLa and PtK2 cells, gamma-tubulin was detected in the mitotic spindle from late prometaphase to telophase. In contrast, in other cell types, it was detected in metaphase only. In all cases we failed to detect gamma-tubulin in the short aster microtubules at the spindle poles. Electron microscopic observation revealed that at least part of the gamma-tubulin localized on the surface of spindle microtubules with a preferential distribution along kinetochore microtubules. In HeLa cells, the amount of antigenic gamma-tubulin was fairly constant in the spindle poles during mitosis from prometaphase to telophase. In contrast, gamma-tubulin appeared in the mitotic spindles in prometaphase. The amount of gamma-tubulin decreased in telophase, where it relocalized in the interzone. In metaphase cells about 15–25% of the total fluorescence was localized at the spindle poles, while 75–85% of the fluorescence was distributed over the rest of the spindle. These results suggest that the localization and timing of gamma-tubulin during the cell cycle is highly regulated and that is physiological role could be more complex and diverse than initially assumed.

The spindle pole body of yeast

Microtubule organizing centers play an essential cellular role in nucleating microtubule assembly and establishing the microtubule array. The microtubule organizing center of yeast, the spindle pole body (SPB), shares many functions and properties with those other organisms. In recent years considerable new information has been generated concerning components associated with the SPB, and the mechanism by which it duplicates. This article reviews our current view of the cytology and molecular composition of the SPB of the budding yeast, Saccharomyces cerevisiae, and the fission yeast, Schizosaccharomyces pombe. Genetic studies in these organisms has revealed information about how the SPB duplicates and separates, and its roles during vegetative growth, mating and meiosis.

Human gamma-tubulin functions in fission yeast

gamma-Tubulin is a phylogenetically conserved component of microtubule-organizing centers that is essential for viability and microtubule function. To examine the functional conservation of gamma-tubulin, we have tested the ability of human gamma-tubulin to function in the fission yeast Schizosaccharomyces pombe. We have found that expression of a human gamma-tubulin cDNA restores viability and a near-normal growth rate to cells of S. pombe lacking endogenous gamma-tubulin. Immunofluorescence microscopy showed that these cells contained normal mitotic spindles and interphase microtubule arrays, and that human gamma-tubulin, like S. pombe gamma-tubulin, localized to spindle pole bodies, the fungal microtubule-organizing centers. These results demonstrate that human gamma-tubulin functions in fission yeast, and they suggest that in spite of the great morphological differences between the microtubule-organizing centers of humans and fission yeasts, gamma-tubulin is likely to perform the same tasks in both. They suggest, moreover, that the proteins that interact with gamma-tubulin, including, most obviously, microtubule-organizing center proteins, must also be conserved. We have also found that a fivefold overexpression of S. pombe gamma-tubulin causes no reduction in growth rates or alteration of microtubule organization. We hypothesize that the excess gamma-tubulin is maintained in the cytoplasm in a form incapable of nucleating microtubule assembly. Finally, we have found that expression of human gamma-tubulin or overexpression of S. pombe gamma-tubulin causes no significant alteration of resistance to the antimicrotubule agents benomyl, thiabendazole and nocodazole.

Do prokaryotes contain microtubules?

In eukaryotic cells, microtubules are 24-nm-diameter tubular structures composed of a class of conserved proteins called tubulin. They are involved in numerous cell functions including ciliary motility, nerve cell elongation, pigment migration, centrosome formation, and chromosome movement. Although cytoplasmic tubules and fibers have been observed in bacteria, some with diameters similar to those of eukaryotes, no homologies to eukaryotic microtubules have been established. Certain groups of bacteria including azotobacters, cyanobacteria, enteric bacteria, and spirochetes have been frequently observed to possess microtubule-like structures, and others, including archaebacteria, have been shown to be sensitive to drugs that inhibit the polymerization of microtubules. Although little biochemical or molecular biological information is available, the differences observed among these prokaryotic structures suggest that their composition generally differs among themselves as well as from that of eukaryotes. We review the distribution of cytoplasmic tubules in prokaryotes, even though, in all cases, their functions remain unknown. At least some tend to occur in cells that are large, elongate, and motile, suggesting that they may be involved in cytoskeletal functions, intracellular motility, or transport activities comparable to those performed by eukaryotic microtubules. In Escherichia coli, the FtsZ protein is associated with the formation of a ring in the division zone between the newly forming offspring cells. Like tubulin, FtsZ is a GTPase and shares with tubulin a 7-amino-acid motif, making it a promising candidate in which to seek the origin of tubulins.

Stable expression of heterologous microtubule-associated proteins (MAPs) in Chinese hamster ovary cells: evidence for differing roles of MAPs in microtubule organization

To study the effects of microtubule-associated proteins (MAPs) on in vivo microtubule assembly, cDNAs containing the complete coding sequences of a Drosophila 205-kD heat stable MAP, human MAP 4, and human tau were stably transfected into CHO cells. Constitutive expression of the transfected genes was low in most cases and had no obvious effects on the viability of the transfected cell lines. High levels of expression, as judged by Western blots, immunofluorescence, and Northern blots, could be induced by treating cells with sodium butyrate. High levels of MAPs were maintained for at least 24-48 h after removal of the sodium butyrate. Immunofluorescence analysis indicated that all three MAPs bound to cellular microtubules, but only the transfected tau caused a rearrangement of microtubules into bundles. Despite high levels of expression of these exogenous MAPs and the bundling of microtubules in cells expressing tau, transfected cells had normal levels of assembled and unassembled tubulin. With the exception of the tau-induced bundles, microtubules in transfected cells showed the same sensitivity as control cells to microtubule depolymerization by Colcemid. Further, all three MAPs were ineffective in reversing the taxol-dependent phenotype of a CHO mutant cell line. The absence of a quantitative effect of any of these heterologous proteins on the assembly of tubulin suggests that these MAPs may have different roles in vivo from those inferred previously from in vitro experiments.

Centrosomal components immunologically related to tektins from ciliary and flagellar microtubules

Centrosomes are critical for the nucleation and organization of the microtubule cytoskeleton during both interphase and cell division. Using antibodies raised against sea urchin sperm flagellar microtubule proteins, we characterize here the presence and behavior of certain components associated with centrosomes of the surf clam Spisula solidissima and cultured mammalian cells. A Sarkosyl detergent-resistant fraction of axonemal microtubules was isolated from sea urchin sperm flagella and used to produce monoclonal antibodies, 16 of which were specific- or cross-specific for the major polypeptides associated with this microtubule fraction: tektins A, B and C, acetylated alpha-tubulin, and 77 and 83 kDa polypeptides. By 2-D isoelectric focussing/SDS polyacrylamide gel electrophoresis the tektins separate into several polypeptide spots. Identical spots were recognized by monoclonal and polyclonal antibodies against a given tektin, indicating that the different polypeptide spots are isoforms or modified versions of the same protein. Four independently derived monoclonal anti-tektins were found to stain centrosomes of S. solidissima oocytes and CHO and HeLa cells, by immunofluorescence microscopy. In particular, the centrosome staining of one monoclonal antibody specific for tektin B (tekB3) was cell-cycle-dependent for CHO cells, i.e. staining was observed only from early prometaphase until late anaphase. By immuno-electron microscopy tekB3 specifically labeled material surrounding the centrosome, whereas a polyclonal anti-tektin B recognized centrioles as well as the centrosomal material throughout the cell cycle. Finally, by immunoblot analysis tekB3 stained polypeptides of 48–50 kDa in isolated spindles and centrosomes from CHO cells.

Localization of the mei-1 gene product of Caenorhaditis elegans, a meiotic-specific spindle component

Genetic evidence suggests that the product of the mei-1 gene of Caenorhabditis elegans is specifically required for meiosis in the female germline. Loss-of-function mei-1 mutations block meiotic spindle formation while a gain-of-function allele instead results in spindle defects during the early mitotic cleavages. In this report, we use immunocytochemistry to examine the localization of the mei-1 product in wild-type and mutant embryos. During metaphase of meiosis I in wild-type embryos, mei-1 protein was found throughout the spindle but was more concentrated toward the poles. At telophase I, mei-1 product colocalized with the chromatin at the spindle poles. The pattern was repeated during meiosis II but no mei-1 product was visible during the subsequent mitotic cleavages. The mei-1 gain-of-function allele resulted in ectopic mei-1 staining in the centers of the microtubule-organizing centers during interphase and in the spindles during the early cleavages. This aberrant localization is probably responsible for the poorly formed and misoriented cleavage spindles characteristic of the mutation. We also examined the localization of mei-1(+) product in the presence of mutations of genes that genetically interact with mei-1 alleles. mei-2 is apparently required to localize mei-1 product to the spindle during meiosis while mel-26 acts as a postmeiotic inhibitor. We conclude that mei-1 encodes a novel spindle component, one that is specialized for the acentriolar meiotic spindles unique to female meiosis. The genes mei-2 and mel-26 are part of a regulatory network that confines mei-1 activity to meiosis.

Nuclear mitotic apparatus protein (NuMA): spindle association, nuclear targeting and differential subcellular localization of various NuMA isoforms

We have recently shown that the nuclear mitotic apparatus protein (NuMA) is composed of at least three isoforms that differ mainly at the carboxy terminus, and are generated by alternative splicing of a common mRNA precursor from a single NuMA gene (J. Cell Sci. (1993) 104, 249–260). Transient expression of human NuMA-1 isoform (T33/p230) in Chinese hamster ovary polyoma (CHOP) cells showed that NuMA-1 was present in interphase nuclei and was concentrated at the polar regions of the spindle apparatus in mitotic cells. However, expression of two other isoforms (NuMA-m and -s) revealed a distinct subcellular localization. NuMA-m (U4/p195) and NuMA-s (U6/p194) were present in the interphase cytosol and appeared to be mainly located at the centrosomal region. When cells entered into mitosis, however, NuMA-m and -s moved to the mitotic spindle pole. Analysis of a series of linker scanning-mutants and NuMA/beta-galactosidase chimeric proteins showed that residues 1972–2007 of NuMA-1 constitute a novel nuclear localization signal (NLS) and residues 1538–2115 are necessary and sufficient for spindle association. Further analysis of the NLS by site-specific mutagenesis indicated that Lys1988 is essential for nuclear targeting, whereas Arg1984 is not. These results have allowed us tentatively to assign specific biological activities to distinct structural domains of the NuMA polypeptide.

Facilitated folding of actins and tubulins occurs via a nucleotide-dependent interaction between cytoplasmic chaperonin and distinctive folding intermediates

In the cytoplasm of eukaryotes, the folding of actins and tubulins is facilitated via interaction with a heteromeric toroidal complex (cytoplasmic chaperonin). The folding reaction consists of the formation of a binary complex between the unfolded target protein and the chaperonin, followed by the ultimate release of the native polypeptide in an ATP-dependent reaction. Here we show that the mitochondrial chaperonin (cpn60) and the cytoplasmic chaperonin both recognize a range of target proteins with different relative affinities; however, the cytoplasmic chaperonin shows the highest affinity for intermediates derived from unfolded tubulins and actins. These high-affinity actin and tubulin folding intermediates are distinct from the "molten globule" intermediates formed by noncytoskeletal target proteins in that they form relatively slowly. We show that the interaction between cytoplasmic chaperonin and unfolded target proteins depends on the chaperonin being in its ADP-bound state and that the release of the target protein occurs after a transition of the chaperonin to the ATP-bound state. Our data suggest a model in which ATP hydrolysis acts as a switch between conformational forms of the cytoplasmic chaperonin that interact either strongly or weakly with unfolded substrates.

Facilitated folding of actins and tubulins occurs via a nucleotide-dependent interaction between cytoplasmic chaperonin and distinctive folding intermediates

In the cytoplasm of eukaryotes, the folding of actins and tubulins is facilitated via interaction with a heteromeric toroidal complex (cytoplasmic chaperonin). The folding reaction consists of the formation of a binary complex between the unfolded target protein and the chaperonin, followed by the ultimate release of the native polypeptide in an ATP-dependent reaction. Here we show that the mitochondrial chaperonin (cpn60) and the cytoplasmic chaperonin both recognize a range of target proteins with different relative affinities; however, the cytoplasmic chaperonin shows the highest affinity for intermediates derived from unfolded tubulins and actins. These high-affinity actin and tubulin folding intermediates are distinct from the "molten globule" intermediates formed by noncytoskeletal target proteins in that they form relatively slowly. We show that the interaction between cytoplasmic chaperonin and unfolded target proteins depends on the chaperonin being in its ADP-bound state and that the release of the target protein occurs after a transition of the chaperonin to the ATP-bound state. Our data suggest a model in which ATP hydrolysis acts as a switch between conformational forms of the cytoplasmic chaperonin that interact either strongly or weakly with unfolded substrates.

Distinct intracellular localization of Lck and Fyn protein tyrosine kinases in human T lymphocytes

Two src family kinases, lck and fyn, participate in the activation of T lymphocytes. Both of these protein tyrosine kinases are thought to function via their interaction with cell surface receptors. Thus, lck is associated with CD4, CD8, and Thy-1, whereas fyn is associated with the T cell antigen receptor and Thy-1. In this study, the intracellular localization of these two protein tyrosine kinases in T cells was analyzed by immunofluorescence and confocal microscopy. Lck was present at the plasma membrane, consistent with its proposed role in transmembrane signalling, and was also associated with pericentrosomal vesicles which co-localized with the cation-independent mannose 6-phosphate receptor. Surprisingly, fyn was not detected at the plasma membrane in either Jurkat T cells or T lymphoblasts but was closely associated with the centrosome and to microtubule bundles radiating from the centrosome. In mitotic cells, fyn co-localized with the mitotic spindle and poles. The essentially non-overlapping intracellular distributions of lck and fyn suggest that these kinases may be accessible to distinct regulatory proteins and substrates and, therefore, may regulate different aspects of T cell activation. Anti-phosphotyrosine antibody staining at the plasma membrane increases dramatically after CD3 cross-linking of Jurkat T cells. The localization of lck to the plasma membrane suggests that it may participate in mediating this increase in tyrosine phosphorylation, rather than fyn. Furthermore, the distribution of fyn in mitotic cells raises the possibility that it functions at the M phase of the cell cycle.

Guanine nucleotide-dependent assembly of FtsZ into filaments

FtsZ is an essential cell division protein that is localized to the leading edge of the bacterial septum in a cytokinetic ring. It contains the tubulin signature motif and is a GTP binding protein with a GTPase activity. Further comparison of FtsZ with eukaryotic tubulins revealed some additional sequence similarities, perhaps indicating a similar GTP binding site. Examination of FtsZ incubated in vitro by electron microscopy revealed a guanine nucleotide-dependent assembly into protein filaments, supporting the hypothesis that the FtsZ ring is formed through self-assembly. FtsZ3, which is unable to bind GTP, does not polymerize, whereas FtsZ2, which binds GTP but is deficient in GTP hydrolysis, is capable of polymerization.

In vitro reconstitution of centrosome assembly and function: the central role of gamma-tubulin

The centrosome nucleates microtubule polymerization, affecting microtubule number, polarity, and structure. We use an in vitro system based on extracts of Xenopus eggs to examine the role of gamma-tubulin in centrosome assembly and function. gamma-Tubulin is present in the cytoplasm of frog eggs and vertebrate somatic cells in a large approximately 25S complex. The egg extracts assemble centrosomes around sperm centrioles. Formation of a centrosome in the extract requires both the gamma-tubulin complex and ATP and can take place in the absence of microtubules. gamma-Tubulin is not present on the sperm prior to incubation in extract, but is recruited from the cytoplasm during centrosome assembly. The gamma-tubulin complex also binds to microtubules, likely the minus end, independent of the centrosome. These results suggest that gamma-tubulin is an essential component of the link between the centrosome and the microtubule, probably playing a direct role in microtubule nucleation.

Motile kinetochores and polar ejection forces dictate chromosome position on the vertebrate mitotic spindle

We argue that hypotheses for how chromosomes achieve a metaphase alignment, that are based solely on a tug-of-war between poleward pulling forces produced along the length of opposing kinetochore fibers, are no longer tenable for vertebrates. Instead, kinetochores move themselves and their attached chromosomes, poleward and away from the pole, on the ends of relatively stationary but shortening/elongating kinetochore fiber microtubules. Kinetochores are also "smart" in that they switch between persistent constant-velocity phases of poleward and away from the pole motion, both autonomously and in response to information within the spindle. Several molecular mechanisms may contribute to this directional instability including kinetochore-associated microtubule motors and kinetochore microtubule dynamic instability. The control of kinetochore directional instability, to allow for congression and anaphase, is likely mediated by a vectorial mechanism whose magnitude and orientation depend on the density and orientation or growth of polar microtubules. Polar microtubule arrays have been shown to resist chromosome poleward motion and to push chromosomes away from the pole. These "polar ejection forces" appear to play a key role in regulating kinetochore directional instability, and hence, positions achieved by chromosomes on the spindle.

Inhibition of microtubule nucleation at the neuronal centrosome compromises axon growth

We tested the dependence of axon growth on microtubule (MT) nucleation from the neuronal centrosome. Nocodazole diminished MTs in freshly plated neurons by > 99%. Within 5 min of drug removal, MTs reassembled at the centrosome. This response was inhibited in cells microinjected with gamma-tubulin antibody. Within 2 hr of drug removal, uninjected neurons grew > 500 microns of axon. In roughly half of the antibody-injected cells, axon growth was abolished and MT levels were reduced by approximately 87% compared with uninjected cells. In the other antibody-injected cells, axon growth was compromised but not abolished, and MT levels were reduced by approximately 38%. Thus inhibition of MT nucleation at the centrosome hindered MT reassembly, and depending on the severity of this response, axon growth was either compromised or abolished.

Microtubule organizing centers and gamma-tubulin

The polar assembly of cellular microtubules is organized by microtubule organizing centers (MTOCs). Eukaryotic cells across different species, and different cell types within single species, have morphologically diverse MTOCs, which have the common function of organizing microtubule arrays by initiating microtubule assembly and anchoring microtubules by their slow-growing 'minus' ends, thus ensuring that the rapidly growing 'plus' ends extend distally. The past few years have witnessed a variety of approaches aimed at defining the molecular components of the MTOC that are responsible for regulating microtubule assembly by defining molecules common to all MTOCs.

A protein related to brain microtubule-associated protein MAP1B is a component of the mammalian centrosome

The centrosome is the main microtubule organizing center of mammalian cells. Structurally, it is composed of a pair of centrioles surrounded by a fibro-granular material (the pericentriolar material) from which microtubules are nucleated. However, the nature of centrosomal molecules involved in microtubules nucleation is still obscure. Since brain microtubule-associated proteins (MAPs) lower the critical tubulin concentration required for microtubule nucleation in tubulin solution in vitro, we have examined their possible association with centrosomes. By immunofluorescence, monoclonal and polyclonal antibodies raised against MAP1B stain the centrosome in cultured cells as well as purified centrosomes, whereas antibodies raised against MAP2 give a completely negative reaction. The MAP1B-related antigen is localized to the pericentriolar material as revealed by immunoelectron microscopy. In preparations of purified centrosomes analyzed on poly-acrylamide gels, a protein that migrates as brain MAP1B is present. After blotting on nitrocellulose, it is decorated by anti-MAP1B antibodies and the amino acid sequence of proteolytic fragments of this protein is similar to brain MAP1B. Moreover, brain MAP1B and its centrosomal counterpart share the same phosphorylation features and have similar peptide maps. These data strongly suggest that a protein homologue to MAP1B is present in centrosomes and it is a good candidate for being involved in the nucleating activity of the pericentriolar material.

Centrosome assembly in vitro: role of gamma-tubulin recruitment in Xenopus sperm aster formation

Centrioles organize microtubules in two ways: either microtubules elongate from the centriole cylinder itself, forming a flagellum or a cilium ("template elongation"), or pericentriolar material assembles and nucleates a microtubule aster ("astral nucleation"). During spermatogenesis in most species, a motile flagellum elongates from one of the sperm centrioles, whereas after fertilization a large aster of microtubules forms around the sperm centrioles in the egg cytoplasm. Using Xenopus egg extracts we have developed an in vitro system to study this change in microtubule-organizing activity. An aster of microtubules forms around the centrioles of permeabilized frog sperm in egg extracts, but not in pure tubulin. However, when the sperm heads are incubated in the egg extract in the presence of nocodazole, they are able to nucleate a microtubule aster after isolation and incubation with pure calf brain tubulin. This provides a two-step assay that distinguishes between centrosome assembly and subsequent microtubule nucleation. We have studied several centrosomal antigens during centrosome assembly. The CTR2611 antigen is present in the sperm head in the peri-centriolar region. gamma-tubulin and certain phosphorylated epitopes appear in the centrosome only after incubation in the egg extract. gamma-tubulin is recruited from the egg extract and associated with electron-dense patches dispersed in a wide area around the centrioles. Immunodepletion of gamma-tubulin and associated molecules from the egg extract before sperm head incubation prevents the change in microtubule-organizing activity of the sperm heads. This suggests that gamma-tubulin and/or associated molecules play a key role in centrosome formation and activity.

The gamma-tubulin-encoding gene from the basidiomycete fungus, Ustilago violacea, has a long 5'-untranslated region

The gene (gamma-tub) encoding gamma-tubulin (gamma-Tub) was isolated from a cosmid library constructed for Ustilago violacea by using a PCR-amplified DNA fragment as a probe. About 2.8 kb of DNA sequence was analyzed and found to encode a protein of 469 amino acids highly homologous to the gamma-Tub from other organisms. There were eight introns interrupting the coding sequence. A 'TATA'-like sequence was found 389 bp upstream from the initial Met codon. No polyadenylation signal was found in the 3' non-coding region. Southern blot analyses indicated that gamma-tub is a single-copy gene. Northern blot analyses indicated that a 1.81-kb RNA species was transcribed. Primer extension experiments determined that the transcription start point (tsp) is at 58 bp downstream from the putative TATA box, with another possible tsp at 95 bp downstream. The long 5' non-coding sequence of the RNA contained several small open reading frames; their possible roles in the regulation of gamma-tub translation are discussed.

The macronuclear gamma-tubulin-encoding gene of Euplotes octocarinatus contains two introns and an in-frame TGA

The gamma-tubulin (gamma-Tub)-encoding gene (gamma-tub) of Euplotes octocarinatus was amplified from macronuclear DNA with the help of the polymerase chain reaction (PCR) and sequenced. The polypeptide deduced from the gene consists of 462 amino acids (aa). It shares 61% aa identity with the Aspergillus nidulans gamma-Tub. The gene contains an in-frame TGA codon and two small pre-mRNA introns (36 and 26 bp). We suggest that the TGA, like TGA codons in the pheromone-encoding genes of E. octocarinatus, codes for a cysteine. This suggestion is supported by the finding that in the gamma-Tub of other organisms, a cysteine is located at this position. Sequencing the mRNA revealed that the introns are absent from the gamma-tub transcripts. The second intron constitutes the shortest one reported so far. We have also sequenced the 5'- and 3'-flanking regions of the gene up to the telomeres and report here the entire sequence of the macronuclear DNA molecule carrying gamma-tub.

Isolation, characterization and sequence of a cDNA encoding gamma-tubulin protein from the fern Anemia phyllitidis L. Sw

A complete cDNA encoding gamma-tubulin protein from Anemia phyllitidis is presented. The deduced amino acid sequence shows an average similarity of 81-86% to known gamma-tubulin genes from insects, mammals and fungi. Northern blots indicate the existence of transcripts of ca. 1.9 kb which are expressed during gametogenesis of A. phyllitidis. The gene described is the first cloned and entirely sequenced gamma-tubulin gene in any plant.

Identification of a 102 kDa protein (cytocentrin) immunologically related to keratin 19, which is a cytoplasmically derived component of the mitotic spindle pole

The mAb RK7, previously shown to recognize keratin 19, was also found to cross-react with a biologically unrelated 102 kDa protein, which becomes associated with the poles of the mitotic apparatus. This newly identified protein, called cytocentrin, is a stable cellular component, may be at least in part phosphorylated, and displays a cell cycle-dependent cellular localization. In interphase cells, it is diffusely distributed in the cytosol and shows no affinity for cytoplasmic microtubules. It becomes localized to the centrosome in early prophase, prior to nuclear envelope breakdown, separation of replicated centrosomes, and nucleation of mitotic apparatus microtubules. During metaphase, cytocentrin is located predominately at the mitotic poles, often appearing as an aggregate of small globular sub-components; it also associates with some polar microtubules. In late anaphase/early telophase cytocentrin dissociates entirely from the mitotic apparatus and becomes temporarily localized with microtubules in the midbody, from which it disappears by late telophase. In taxol-treated cells cytocentrin was associated with the center of the miniasters but also showed affinity for some cytoplasmic microtubules. Studies employing G2-synchronized cells and nocodazole demonstrated that cytocentrin can become associated with mitotic centrosomes independently of tubulin polymerization and that microtubules regrow from antigen-containing foci. We interpret these results to suggest that cytocentrin is a cytoplasmic protein that becomes specifically activated or modified at the onset of mitosis so that it can affiliate with the mitotic poles where it may provide a link between the pericentriolar material and other components of the mitotic apparatus.

Gamma-tubulin: the hub of cellular microtubule assemblies

In eukaryotic cells a specialized organelle called the microtubule organizing center (MTOC) is responsible for disposition of microtubules in a radial, polarized array in interphase cells and in the spindle in mitotic cells. Eukaryotic cells across different species, and different cell types within single species, have morphologically diverse MTOCs, but these share a common function of organizing microtubule arrays. MTOCs effect microtubule organization by initiating microtubule assembly and anchoring microtubules by their slowly growing minus ends, thus ensuring that the rapidly growing plus ends extend distally in each microtubule array. The goal is to define molecular components of the MTOC responsible for regulating microtubule assembly. One approach to defining the molecules responsible for MTOC function is to look for molecules common to all MTOCs. A newly discovered centrosomal protein, gamma-tubulin, is found in MTOCs in cells from many different organisms, and has several properties which make it a candidate for both initiation of microtubule assembly and anchorage. The hypothesis that gamma-tubulin plays a role in MTOCs in microtubule initiation and anchorage is currently being tested by a variety of experimental approaches.

Chaperonin-mediated folding of vertebrate actin-related protein and gamma-tubulin

The folding of actin and tubulin is mediated via interaction with a heteromeric toroidal complex (cytoplasmic chaperonin) that hydrolyzes ATP as part of the reaction whereby native proteins are ultimately released. Vertebrate actin-related protein (actin-RPV) (also termed centractin) and gamma-tubulin are two proteins that are distantly related to actin and tubulin, respectively: gamma-tubulin is exclusively located at the centrosome, while actin-RPV is conspicuously abundant at the same site. Here we show that actin-RPV and gamma-tubulin are both folded via interaction with the same chaperonin that mediates the folding of beta-actin and alpha- and beta-tubulin. In each case, the unfolded polypeptide forms a binary complex with cytoplasmic chaperonin and is released as a soluble, monomeric protein in the presence of Mg-ATP and the presence or absence of Mg-GTP. In contrast to alpha- and beta-tubulin, the folding of gamma-tubulin does not require the presence of cofactors in addition to chaperonin itself. Monomeric actin-RPV produced in in vitro folding reactions cocycles efficiently with native brain actin, while in vitro folded gamma-tubulin binds to polymerized microtubules in a manner consistent with interaction with microtubule ends. Both monomeric actin-RPV and gamma-tubulin bind to columns of immobilized nucleotide: monomeric actin-RPV has no marked preference for ATP or GTP, while gamma-tubulin shows some preference for GTP binding. We show that actin-RPV and gamma-tubulin compete with one another, and with beta-actin or alpha-tubulin, for binary complex formation with cytoplasmic chaperonin.

Pathobiochemical aspects of cytoskeleton components

This review summarizes pathobiochemical aspects of diseases, in which cytoskeletal components play a crucial role in pathogenesis. An attempt to classify the disorders on the basis of phenotypic changes that occur in microfilaments, intermediate filaments and microtubuli was unsuccessful. Three groups of disorders are presented: 1. cytoplasmic inclusions in specific diseases (merely descriptive); 2. diseases with genetic defects in cytoskeletal proteins (a chain of causality from defect to phenotype, in some cases with large gaps); 3. diseases with suspected involvement of cytoskeleton (hypothetical causal chain). Microfilaments are involved in certain pathogenetic processes on account of defects in their associated proteins; in Duchenne muscular dystrophy, dystrophin is defective, while the defective protein in Rett syndrome is synapsin. Defects in spectrin and membrane anchor proteins lead to disorders of the red cell membrane skeleton (congenital haemolytic anaemias). Intermediate filaments accumulate in some types of cytoplasmic inclusions, together with ubiquitin (Mallory bodies, desmin accumulation in some myopathies and others). A pathogenetic interpretation of this phenomenon is lacking. A genetic defect in certain types of keratin is the cause of epidermolysis bullosa. Interesting preliminary results are reviewed that reveal the crucial role of cytoskeletal components in a further group of diseases (intrahepatic cholestasis, Alzheimer disease, pemphigus). These disorders are currently under investigation, or are of theoretical interest with respect to the cytoskeleton. Specific reactions of cytoskeletal components in serum, which might be used diagnostically, have not been found.

The gamma-tubulin gene of the malaria parasite Plasmodium falciparum

By screening of cDNA and genomic libraries of Plasmodium falciparum with a DNA probe derived from the cognate beta-tubulin gene, gene Pf gamma tub has been identified that codes for gamma-tubulin, a newly discovered member of the tubulin superfamily that is indispensible for nuclear division and microtubule assembly [12]. Gene Pf gamma tub is not interrupted by introns and only present as a single-copy in the parasite genome. Its encoded amino acid sequence (452 amino acids; M(r) 50,560) has a 63% similarity to the gamma-tubulins encoded by Aspergillus nidulans, Schizosaccharomyces pombe, Drosophila melanogaster, Xenopus laevis and Homo sapiens. This figure is significantly (approx. 8%) lower than the average identity between the gamma-tubulins of the latter five species suggesting that during evolution the genes have been exposed to different selection pressures. The identity of gamma-tubulin to the Plasmodium falciparum encoded alpha- and beta-tubulins is 30 and 33%, respectively.

A nucleotide substitution in one of the beta-tubulin genes of Trichoderma viride confers resistance to the antimitotic drug methyl benzimidazole-2-yl-carbamate

We characterized a Trichoderma viride strain that is resistant to the antimitotic drug methyl benzimidazole-2-yl-carbamate (MBC). This species has two beta-tubulin genes (tub1 and tub2) and by reverse genetics we showed that a mutation in the tub2 gene confers MBC resistance in this strain. Comparison of the tub2 sequence of the mutant strain with that of the wild type revealed that a single amino acid substitution of tyrosine for histidine at a position 6 is responsible for the MBC tolerance. Furthermore, we showed that this gene can be used as a homologous dominant selectable marker in T. viride transformation. Both tubulin genes were completely sequenced. They differ by 48 residues and the degree of identity between their deduced amino acid sequences is 86.3%.

Drosophila gamma-tubulin is part of a complex containing two previously identified centrosomal MAPs

gamma-tubulin is a minor tubulin that is localized to the microtubule organizing center of many fungi and higher eucaryotic cells (Oakley, B. R., C. E. Oakley, Y. Yoon, and M. C. Jung. 1990. Cell. 61: 1289-1301; Stearns, T., L. Evans, and M. Kirschner. 1991. Cell. 65:825-836; Zheng, Y., M. K. Jung, and B. R. Oakley. 1991. Cell. 65:817-823). Here we show that gamma-tubulin is a component of a previously isolated complex of Drosophila proteins that contains at least two centrosomal microtubule-associated proteins called DMAP190 and DMAP60. Like DMAP190 and DMAP60, the gamma-tubulin in extracts of early Drosophila embryos binds to microtubules, although this binding may be indirect. Unlike DMAP190 and DMAP60, however, only 10-50% of the gamma-tubulin in the extract is able to bind to microtubules. We show that gamma-tubulin binds to a microtubule column as part of a complex, and a substantial fraction of this gamma-tubulin is tightly associated with DMAP60. As neither alpha- nor beta-tubulin bind to microtubule columns, the gamma-tubulin cannot be binding to such columns in the form of an alpha:gamma or beta:gamma heterodimer. These observations suggest that gamma-tubulin, DMAP60, and DMAP190 are components of a centrosomal complex that can interact with microtubules.