Centriole disassembly in vivo and its effect on centrosome structure and function in vertebrate cells.

Glutamylation is the major posttranslational modification of neuronal and axonemal tubulin and is restricted predominantly to centrioles in nonneuronal cells (Bobinnec, Y., M. Moudjou, J.P. Fouquet, E. Desbruyères, B. Eddé, and M. Bornens. 1998. Cell Motil. Cytoskel. 39:223-232). To investigate a possible relationship between the exceptional stability of centriole microtubules and the compartmentalization of glutamylated isoforms, we loaded HeLa cells with the monoclonal antibody GT335, which specifically reacts with polyglutamylated tubulin. The total disappearance of the centriole pair was observed after 12 h, as judged both by immunofluorescence labeling with specific antibodies and electron microscopic observation of cells after complete thick serial sectioning. Strikingly, we also observed a scattering of the pericentriolar material (PCM) within the cytoplasm and a parallel disappearance of the centrosome as a defined organelle. However, centriole disappearance was transient, as centrioles and discrete centrosomes ultimately reappeared in the cell population. During the acentriolar period, a large proportion of monopolar half-spindles or of bipolar spindles with abnormal distribution of PCM and NuMA were observed. However, as judged by a quasinormal increase in cell number, these cells likely were not blocked in mitosis. Our results suggest that a posttranslational modification of tubulin is critical for long-term stability of centriolar microtubules. They further demonstrate that in animal cells, centrioles are instrumental in organizing centrosomal components into a structurally stable organelle.

Posttranslational glutamylation of alpha-tubulin.

The high degree of tubulin heterogeneity in neurons is controlled mainly at the posttranslational level. Several variants of alpha-tubulin can be posttranslationally labeled after incubation of cells with [3H]acetate or [3H]glutamate. Peptides carrying the radioactive moiety were purified by high-performance liquid chromatography. Amino acid analysis, Edman degradation sequencing, and mass spectrometric analysis of these peptides led to the characterization of a posttranslational modification consisting of the successive addition of glutamyl units on the gamma-carboxyl group of a glutamate residue (Glu445). This modification, localized within a region of alpha-tubulin that is important in the interactions of tubulin with microtubule-associated proteins and calcium, could play a role in regulating microtubule dynamics.

Reversible polyglutamylation of alpha- and beta-tubulin and microtubule dynamics in mouse brain neurons.

The relationship between microtubule dynamics and polyglutamylation of tubulin was investigated in young differentiating mouse brain neurons. Selective posttranslational labeling with [3H]glutamate and immunoblotting with a specific monoclonal antibody (GT335) enabled us to analyze polyglutamylation of both alpha and beta subunits. Nocodazole markedly inhibited incorporation of [3H]glutamate into alpha- and beta-tubulin, whereas taxol had no effect for alpha-tubulin and a stimulating effect for beta-tubulin. These results strongly suggest that microtubule polymers are the preferred substrate for polyglutamylation. Chase experiments revealed the existence of a reversal reaction that, in the case of alpha-tubulin, was not affected by microtubule drugs, suggesting that deglutamylation of this subunit can occur on both polymers and soluble tubulin. Evidence was obtained that deglutamylation of alpha-tubulin operates following two distinct rates depending on the length of the polyglutamyl chain, the distal units (4th-6th) being removed rapidly whereas the proximal ones (1st-3rd) appearing much more resistant to deglutamylation. Partition of glutamylated alpha-tubulin isoforms was also correlated with the length of the polyglutamyl chain. Forms bearing four to six units were recovered specifically in the polymeric fraction, whereas those bearing one to three units were distributed evenly between polymeric and soluble fractions. It thus appears that the slow rate component of the deglutamylation reaction offers to neurons the possibility to maintain a basal level of glutamylated alpha-tubulin in the soluble pool independently of microtubule dynamics. Finally, some differences observed in the glutamylation of alpha- and beta-tubulin suggest that distinct enzymes are involved.

Distribution of glutamylated alpha and beta-tubulin in mouse tissues using a specific monoclonal antibody, GT335.

A monoclonal antibody (GT335) directed against polyglutamylated tubulin was obtained by immunization with a synthetic peptide which mimics the structure of the polyglutamylated site of alpha-tubulin. This peptide corresponds to the C-terminal sequence Glu441-Gly448 and was chemically modified by the addition of two glutamyl units at Glu445. The specificity of GT335 was assayed by direct and competitive enzyme-linked immunosorbent assay (ELISA) against tubulin and several synthetic peptides differing either by the structure of the added polyglutamyl chain or by their amino acid sequence. Further characterization was carried out by immunoblotting detection after one- or two-dimensional sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). The epitope appears to be formed by at least two constituents: a basic motif of monoglutamylation which is retained in the polyglutamylated forms independent of their degree of glutamylation, and some elements of the polypeptide chain close to the site of glutamylation. Given the specificity of GT335 and the delineation of its epitope, our results indicate that, in addition to alpha and beta' (class III)-tubulin, other beta-tubulin isotypes are also glutamylated. This antibody has been used to analyze the cell and tissue distributions of glutamylated tubulin. In mouse brain extracts, GT335 reacts strongly with alpha-tubulin and, to a lesser extent, with beta' (class III) and beta-tubulin. The same reactivity is also observed with cultured neurons whereas astroglial cells exhibit only low levels of glutamylated tubulin. In non-nervous mouse tissues such as spleen, lung or testis, glutamylation was shown to involve only beta-tubulin, but at far lower levels than in brain.

Differential expression of several neurospecific beta-tubulin mRNAs in the mouse brain during development.

The expression of alpha- and beta-tubulin mRNAs has been studied during mouse brain development by using in vitro translation and specific cDNA clone hybridization techniques. The results presented here indicate that about half of the isotubulins expressed in vivo by brain cells are directly encoded by mRNA, the other isotypes being of post-translational origin. Among the six brain beta-tubulin mRNAs detected after RNA size fractionation followed by in vitro translation, four (1.8a, 1.8c, 2.4 and 3.5 kb) appear to be uniquely expressed in brain, and very likely in neurons. These four beta-tubulin RNA transcripts are differently regulated during development. The 1.8a-kb mRNA, which codes for the neurospecific beta' 1-isotubulin, is expressed very early, whereas the 1.8c, 2.4 and 3.5-kb mRNAs, which all code for a beta 4-tubulin isotype, are all expressed later. We isolated a beta-tubulin cDNA clone which specifically hybridizes to the large, neural 3.5-kb mRNA. The two other ubiquitous tubulin mRNAs, 1.8b and 2.9 kb in length, both direct the synthesis of an electrophoretically distinct beta 3 isotype. These results support the idea that within the tubulin multigene family, distinct beta-tubulin isogenes could be specifically expressed at various steps of the neuronal differentiation, thus contributing to the isotubulin diversity and specificity observed in this cell type and possibly involved in the developmental changes of its structure and function.

Evolution of tubulin heterogeneity during mouse brain development.

In this report, we have characterized tubulin subunit heterogeneity and its evolution during mouse brain development, from embryonic to adult stages. A modification of the two-dimensional protein analysis was used to specify these events. The number of isotubulins increases from 6 (4 alpha and 2 beta), in the embryonic brain, to 11 (6 alpha and 5 beta), in the adult. The changes occurring in tubulin heterogeneity are developmentally controlled but it seems that alpha and beta isotubulins are independently regulated: changes in alpha tubulin occur only just before birth whereas the major evolution is concerned with the appearance and accumulation of acidic beta isotubulins throughout development.

Changes in some cytoskeletal proteins during neuroblastoma cell differentiation.

After in vitro microtubule assembly of mouse neuroblastoma crude extracts, six protein species migrate in the tubulin region of two-dimensional electrophoregrams. The evolution of these forms after morphological cell differentiation of the clone NIE115 shows two major modifications. Form 5 decreased drastically while form 6 increases during neurite formation. Peptide mapping analysis reveals that forms 5 and 6 are vimentin, a component of intermediate filaments, and beta-tubulin subunit, respectively. Sodium butyrate treatment of NIE115 cells or serum starvation of NIA103 cells, conditions blocking cell division and failing to induce morphological differentiation, prevent any modifications in the relative proportion of these proteins. It is concluded that the changes in the distribution of the tubulin isoforms and vimentin are directly related to neurite formation.

Two specific markers for neural differentiation of embryonal carcinoma cells.

Two multipotential embryonal carcinoma (EC) cell lines, 1003 and 1009, can be induced to form preferentially neural derivatives in vitro. Synthesis of specific proteins during neural differentiation was followed by two-dimensional gel electrophoresis. The comparison of protein patterns obtained with neural and non-neural derivatives of these EC cell lines indicates that two changes are specific for the neural pathway: (i) the appearance of a new beta-tubulin isoform and (ii) the accumulation of the brain isozyme of creatine phosphokinase already present in small amounts in EC stem cells. These changes were found to take place early in the course of differentiation and to occur even when neurite outgrowth was prevented.

Control of isotubulin expression during neuronal differentiation of mouse neuroblastoma and teratocarcinoma cell lines.

Mouse neuroblastoma and teratocarcinoma constitute adequate cellular systems to study the expression of tubulin isoforms during early as well as later steps of neuronal differentiation. Tubulin heterogeneity is extensively analyzed using both isoelectric focusing and two-dimensional electrophoresis. Multipotential embryonal carcinoma cells express mainly one alpha-tubulin isoform (alpha 1) and three beta-tubulin isoforms: a major one (beta 3) and two minor ones (beta 4 and beta 5). Early events of neuronal differentiation are shown to induce the expression of an additional beta-tubulin isoform, beta'1, which is encoded by a specific mRNA. Neurite extension further increases tubulin heterogeneity and leads to the appearance of post-translationally modified isoforms: beta'2 in neuroblastoma and alpha 2 in teratocarcinoma cells. beta' 2 is shown to derive from the above mentioned beta'1 by phosphorylation, while alpha 2 is probably an acetylated form of the common alpha 1-tubulin. These results show that specific changes in tubulin heterogeneity are induced at different steps of neuronal differentiation and are controlled both at the transcriptional (or post-transcriptional) and post-translational levels.

Human serum reacting specifically with a subset of beta-tubulin isoforms.

The serum of a patient suffering from myeloma was found to decorate microtubules and mitotic spindles of cultured cells. Immunoblots performed after one- and two-dimensional electrophoresis showed a reaction with a certain subset of beta-tubulin isoforms, but not with beta'- and alpha-tubulins. The tubulin subset contained both ubiquitous (beta-3) and neurospecific (beta-4,5,6) isoforms. An IgM lambda and an IgA kappa myeloma protein were found in this serum. Immunoblots performed with specific anti-isotype second antibodies showed that the tubulin subset could be evidenced using anti-mu, alpha, lambda, and kappa-specific antisera. Moreover, the tubulin subset was also evidenced using an anti-gamma second antibody. These results, which do not exclude a participation of the myeloma proteins in the anti-tubulin reactivity, indicate, however, that the antibody response was polyclonal. The same restricted specificity of all classes of anti-tubulin antibodies of this serum favours the hypothesis that the immune response of the patient was directed against an antigen sharing epitopes with tubulin rather than with tubulin itself.

Cyclic AMP-dependent protein phosphorylation in isolated neuronal growth cones from developing rat forebrain.

We have shown recently that neuronal growth cones isolated from developing rat forebrain possess an appreciable activity of adenylate cyclase, which produces cyclic AMP and can be stimulated by various neurotransmitter receptor agonists and by forskolin. To investigate cyclic AMP-mediated biochemical mechanisms in isolated growth cones, we have centered the present study on cyclic AMP-dependent protein phosphorylation. One-dimensional gel electrophoretic analysis showed that cyclic AMP analogs increased incorporation of 32P into several phosphoproteins in molecular mass ranges of 50-58 and 76-82 kilodaltons, including those of 82, 76, and 51 kilodaltons. Two-dimensional electrophoresis, using isoelectric focusing in the first dimension, resolved phosphorylated alpha- and beta-tubulin species, actin, a very acidic protein (isoelectric point 4.0) with a molecular mass of 93 kilodaltons, and two proteins (x and x') closely neighboring beta-tubulin. Two other phosphoproteins seen in the gels had molecular masses of 56 and 51 kilodaltons (respective isoelectric points, 4.5 and 4.4) and, along with the 93-kilodalton phosphoprotein, were highly enriched in the isolated growth cones. Only the tubulin and actin species were major proteins in the isolated growth cones. Cyclic AMP analogs enhanced incorporation of 32P into phosphoproteins x and x', and, as assessed by immunoprecipitation, into beta-tubulin. Peptide digest experiments suggested that phosphoproteins x and x' are unrelated to beta-tubulin. Nonequilibrium two-dimensional electrophoresis resolved many phosphoproteins, of which a 79- and 75-kilodalton doublet, a 74-kilodalton species, and a 58-kilodalton doublet showed enhanced incorporation of 32P in the presence of cyclic AMP.

Demonstration and analysis of tubulin binding sites on centrosomes.

Microtubule nucleation on centrosomes is vital to the establishment of organized microtubule arrays in cells. Despite recent advances, little is known about the sequence of molecular events which leads to microtubule assembly on centrosomes. A putative early step in the nucleation process is interaction of free tubulin dimers with centrosomes. Here, we asked if centrosomes indeed interact in a specific manner with free tubulin dimers. Using lysed cells, we show that centrosomes have a specific capacity to accumulate free tubulin molecules as compared to most other cytoplasmic cell structures. When interphasic lysed cells are incubated with rhodamine-conjugated tubulin, centrosomes emerge as conspicuous sites of tubulin accumulation while other insoluble cytoplasmic cell structures are not stained. In mitotic cells, lysed at various stages of mitosis, fluorescent tubulin stains centrosomes and other mitotic structures, such as the mitotic spindle, the midzone of the cleavage furrow, and the center part of the midbody. Fluorescent tubulin staining of centrosomes in lysed cells is not affected by addition of high concentrations of serum albumin to fluorescent tubulin solutions prior to incubation. In contrast, addition of micromolar concentrations of unlabeled tubulin, to fluorescent tubulin solutions, strongly reduces centrosomal staining. The tubulin binding capacity of centrosomes is conserved following centrosome isolation. Using quantitative methods for analysis of fluorescent tubulin binding on centrosomes, we find that centrosomes contain about 25 000 saturable tubulin binding sites. The apparent dissociation constant of tubulin-centrosome complexes is circa 5 microM. The kinetics of tubulin association with centrosomes are slow, with a half-saturation time of about 3 min and a very slow dissociation rate. Tubulin binding to centrosomes is inhibited at low temperatures, at pH above neutrality, and at NaCl concentrations above 100 mM. Our results suggest that accumulation of tubulin dimers is one intrinsic function of centrosomes. We propose that such a function is not accounted for by the presence of gamma-tubulin on centrosomes and may be an important factor in the regulation of centrosome-dependent microtubule nucleation.

Single channel currents in mouse embryonal multipotential carcinoma cells.

Electrical membrane properties of embryonal non-differentiated carcinoma cells which have been extensively used for the study of early mammalian embryogenesis were investigated by using patch clamp techniques. These multipotential cells were found to contain a restricted repertoire of a small number of ionic channels on the whole cell membrane. The most abundant type was a voltage- and calcium-activated potassium channel with characteristics similar to those described in fully differentiated cells.

Tubulin polyglutamylase: partial purification and enzymatic properties.

In this work, we report on a novel enzyme, tubulin polyglutamylase, which catalyzes the posttranslational formation of polyglutamyl side chains onto alpha- and beta-tubulin. The length of the polyglutamyl side chain regulates the interaction between tubulin and various microtubule-associated proteins. We first developed an in vitro glutamylation assay. Activity measured in brain, a tissue particularly enriched with glutamylated tubulin, decreases during postnatal development. Thus, brains from 3-day-old mice were chosen as the starting material, and the enzyme was purified approximately 1000-fold. Its Mr was estimated to be 360K and its sedimentation coefficient 10 s. The enzyme catalyzes the MgATP-dependent addition of l-glutamate onto tubulin subunits. Microtubules are much better substrates than unpolymerized tubulin, and the reaction is very specific for glutamate, other amino acids or glutamate analogues not being substrates. Moreover, glutamyl units are added sequentially onto tubulin, leading to progressive elongation of the polyglutamyl side chains. Side chains of one to six or seven glutamyl units were obtained with microtubules, whereas much longer side chains (up to 15-20 units) were formed with unpolymerized tubulin. Interestingly, such very long polyglutamyl side chains were recently detected in some situations in vivo.

Tubulin polyglutamylase: isozymic variants and regulation during the cell cycle in HeLa cells.

Polyglutamylation is a posttranslational modification of tubulin that is very common in neurons and ciliated or flagellated cells. It was proposed to regulate the binding of microtubule associated proteins (MAPs) and molecular motors as a function of the length of the polyglutamyl side-chain. Though much less common, this modification of tubulin also occurs in proliferating cells like HeLa cells where it is associated with centrioles and with the mitotic spindle. Recently, we partially purified tubulin polyglutamylase from mouse brain and described its enzymatic properties. In this work, we focused on tubulin polyglutamylase activity from HeLa cells. Our results support the existence of a tubulin polyglutamylase family composed of several isozymic variants specific for alpha- or beta-tubulin subunits. In the latter case, the specificity probably also concerns the different beta-tubulin isotypes. Interestingly, we found that tubulin polyglutamylase activity is regulated in a cell cycle dependent manner and peaks in G(2)-phase while the level of glutamylated tubulin peaks in mitosis. Consistent results were obtained by treating the cells with hydroxyurea, nocodazole or taxotere. In particular, in mitotic cells, tubulin polyglutamylase activity was always low while glutamylation level was high. Finally, tubulin polyglutamylase activity and the level of glutamylated tubulin appeared to be inversely related. This paradox suggests a complex regulation of both tubulin polyglutamylase and the reverse deglutamylase activity.

[Microtubules: functional polymorphisms of tubulin and associated proteins (structural and motor MAP's)]. / Les microtubules: polymorphismes fonctionnels de la tubuline et des protéines associées (MAP's structurales et motrices).

In neuronal cells, microtubules are built from a very large number of alpha- and beta-tubulin variants. This diversity is due to the expression of a multigene family and to a combination of several original posttranslational modifications. Similarly, structural and motor microtubule-associated proteins, which regulate the assembly of microtubules, the modeling of their network and the mediation of their functions, are also very heterogeneous. As a consequence, mixing of these two protein polymorphisms leads to the formation of functionally-distinct microtubules. We have shown that polyglutamylation, the major posttranslational modification of neuronal tubulin, was used as a progressive regulator in the binding of structural and motor microtubule-associated proteins, in modulating gradually the conformation of the tubulin carboxy-terminal domain, playing thus a crucial role in microtubule dynamics.

Inhibition of flagellar beat frequency by a new anti-beta-tubulin antibody.

A panel of monoclonal antibodies (mAbs) has been generated against sea urchin sperm axonemes and selected for their ability to inhibit the motility of sea urchin sperm models. The mAb C9 recognized a 50 kDa protein on blots of sea urchin sperm axonemes. Two-dimensional sodium dodecyl sulfate-polyacrylamide gel electrophoresis revealed that C9 recognized isoforms of beta-tubulin. Low concentrations of C9 (0.1-1.0 microgram/ml) blocked the motility of sea urchin sperm models by decreasing the sliding velocity and frequency of flagellar beating to less than 1 Hz and by modifying the shear angle along the axoneme, especially the distal end. Other antitubulin antibodies had little effect on motility at concentrations 100-fold higher than those effective for C9. The effects on motility were not restricted to flagella of sea urchin spermatozoa. Flagellar beating of the dinoflagellate Oxyrrhis marina was completely blocked by C9 in a manner reminiscent of that of sea urchin sperm flagella. The mAb also inhibited the motility of human spermatozoa and Chlamydomonas reinhardtii. Immunofluorescence techniques revealed that C9 stains the whole axoneme of sea urchin spermatozoa and O. marina flagella together with the cortical network of O. marina cell body. C9 is the first antitubulin antibody reported to interfere with flagellar beat frequency. The observation that this antibody arrests the motility of flagella from sea urchin sperm along with that of dinoflagellate, algae, and human sperm flagella suggests that the epitope recognized by C9 is conserved over a long period of evolution and plays an important role in sperm motility.

Polyglutamylation of nucleosome assembly proteins.

Polyglutamylation is an original posttranslational modification, discovered on tubulin, consisting in side chains composed of several glutamyl units and leading to a very unusual protein structure. A monoclonal antibody directed against glutamylated tubulin (GT335) was found to react with other proteins present in HeLa cells. After immunopurification on a GT335 affinity column, two prominent proteins of approximately 50 kDa were observed. They were identified by microsequencing and mass spectrometry as NAP-1 and NAP-2, two members of the nucleosome assembly protein family that are implicated in the deposition of core histone complexes onto chromatin. Strikingly, NAP-1 and NAP-2 were found to be substrates of an ATP-dependent glutamylation enzyme co-purifying on the same column. We took advantage of this property to specifically label and purify the polyglutamylated peptides. NAP-1 and NAP-2 are modified in their C-terminal domain by the addition of up to 9 and 10 glutamyl units, respectively. Two putative glutamylation sites were localized for NAP-1 at Glu-356 and Glu-357 and, for NAP-2, at Glu-347 and Glu-348. These results demonstrate for the first time that proteins other than tubulin are polyglutamylated and open new perspectives for studying NAP function.

Developmental regulation of polyglutamylated alpha- and beta-tubulin in mouse brain neurons.

Polyglutamylation is an important posttranslational modification of tubulin that is very active in nerve cells, where it accounts for the main factor responsible for tubulin heterogeneity. In the present work, we have analyzed quantitative and qualitative changes in glutamylated alpha- and beta-tubulin occurring during neuronal differentiation in culture. Glutamylated alpha- and beta-tubulin both markedly accumulate during this process with a time course remarkably similar to that observed in vivo during brain development. However, the characteristics of the glutamylation of the two subunits are not exactly the same. Glutamylated alpha-tubulin is already abundant in very young neurons and displays, at this stage, a wide range of its degree of glutamylation (1 to 6 glutamyl units present in the lateral polyglutamyl chain), which remains unchanged during the entire period of the culture. Glutamylated beta-tubulin is present at very low levels in young neurons and its accumulation during differentiation is accompanied by a progressive increase in its degree of glutamylation from 2 to 6 glutamyl units. Posttranslational incorporation of [3H]glutamate into alpha- and beta-tubulin decreases during differentiation, as well as the rate of the reverse deglutamylation reaction, suggesting that accumulation of glutamylated tubulin is accompanied by a decrease in the turnover of glutamyl units onto tubulin. Neuronal differentiation is also accompanied by an increase of other posttranslationally modified forms of tubulin, including acetylated and non-tyrosinatable alpha-tubulin, which can occur in combination with polyglutamylation and contributes to increase the complexity of tubulin in mature neurons.

Differential distribution of glutamylated tubulin during spermatogenesis in mammalian testis.

The distribution of glutamylated tubulin has been analyzed in mammalian testis using the specific mAb GT335 by immunoelectron microscopy and immunoblotting. In spermatozoa of various species, immunogold labeling showed the presence of glutamylated tubulin in all of the microtubules of axoneme and centrioles, whereas the microtubule network of the spermatid manchette was unlabeled. In earlier germ cells, centriole was the only microtubule structure to be labeled. A similar distribution was observed using the anti-acetylated tubulin antibody (6-11B-1), confirming previous results of Hermo et al. [Anat. Rec. 229:31-50, 1991]. However, among testicular somatic cells, microtubules of some Sertoli cell branches were not acetylated but glutamylated. 2-D PAGE of mouse and hamster sperm extracts showed a high level of alpha and beta-tubulin heterogeneity, comparable to that found in brain. Immunoblotting with GT335 revealed a large amount of glutamylated tubulin resolved into numerous alpha as well as beta-tubulin isoforms. This suggests that the major testis-specific tubulin isotypes (m alpha 3/7 and m beta 3) are also glutamylatable. These results show a subcellular sorting of posttranslationally modified tubulin isoforms in spermatids, glutamylation being associated with the most stable microtubule structures.