Brief Communication

43

Basal body duplication in Paramecium requires g-tubulin F. Ruiz*, J. Beisson*, J. Rossier† and P. Dupuis-Williams† First discovered in the fungus Aspergillus nidulans [1], g-tubulin is a ubiquitous component of microtubule organizing centres [2]. In centrosomes, g-tubulin has been immunolocalized at the pericentriolar material, suggesting a role in cytoplasmic microtubule nucleation [3], as well as within the centriole core itself [4]. Although its function in the nucleation of the mitotic spindle and of cytoplasmic interphasic microtubules has been demonstrated in vitro [5,6] and in vivo [7–9], the hypothesis that g-tubulin could intervene in centriole assembly has never been experimentally addressed because the mitotic arrest caused by the inactivation of g-tubulin in vivo precludes any further phenotypic analysis of putative centriole defects. The issue can be addressed in the ciliate Paramecium, which is characterized by numerous basal bodies that are similar to centrioles but the biogenesis of which is not tightly coupled to the nuclear division cycle. We demonstrate that the inactivation of the Paramecium g-tubulin genes leads to inhibition of basal body duplication. Addresses: *Centre de Génétique Moléculaire, Centre National de la Recherche Scientifique, 91198 Gif-sur-Yvette Cedex, France. †ESPCI, CNRS UMR 7637, 10 rue Vauquelin, 75231 Paris Cedex, France. Correspondence: P. Dupuis-Williams E-mail: [email protected] Received: 15 September 1998 Revised: 23 November 1998 Accepted: 23 November 1998 Published: 4 January 1999 Current Biology 1999, 9:43–46 http://biomednet.com/elecref/0960982200900043 © Elsevier Science Ltd ISSN 0960-9822

Results and discussion The duplication of centrioles and basal bodies requires both a process of microtubule nucleation and some sort of pre-pattern to assure the nine-fold symmetry [10]. The presence of γ-tubulin in basal bodies [11] and centrioles [4,12,13] has been reported, but its eventual role in their duplication has not been demonstrated. In ciliates, thousands of ciliary basal bodies duplicate at each division according to a precise spatio-temporal program [14]. The permanent association of γ-tubulin with basal bodies in Paramecium [15] suggested that it might play a role in this process. To address this issue, we have exploited a recently developed method for gene silencing in this organism [16]. Two different γ-tubulin genes, γ-PT1 and γ-PT2 (EMBL database accession numbers AJ012329 and AJ012330, respectively), were characterized; both are 1445

nucleotides long and interrupted by three short introns, typical of Paramecium [17]. They are 95% identical and encode proteins of 455 amino acids that differ at only three positions. The two genes exhibit 75–82% similarity with other γ-tubulin sequences (except for the very divergent sequences of Caenorhabditis and Saccharomyces), confirming that they are typical γ-tubulins and that their evolution obeys common functional constraints. Gene-specific inactivation can be achieved in Paramecium [16] by microinjection into the macronucleus of pure coding sequence of the gene of interest at high copy number, as previously demonstrated for several genes. Microinjection of the coding sequences of either γ-PT1 or γ-PT2 into the macronucleus of wild-type cells yielded identical results: cell division arrest after two or three divisions, then cell death. The transformed clones had a reduced cell size and an abnormal, rounded morphology. The lethality precluded analysis of the reduction in transcript and protein levels established in other cases of gene silencing [16], but the effect of γ-tubulin gene inactivation could easily be appreciated cytologically by observation of the basal body pattern and of the organization of the oral apparatus, the biogenesis of which at each division involves several rounds of basal body duplication. Figure 1 and Table 1 present data from six independent microinjection experiments using the γ-PT1 sequence. The dual syndrome — growth arrest and reduced size — was observed in 186 clones derived from a total of 197 injected cells. The transformed clones were either monitored in vivo to confirm growth arrest and eventual death or were fixed within 24 hours of growth arrest and immunostained with an anti-α-tubulin antibody to visualize basal bodies. Defects were readily spotted because the normal cortical pattern is precisely mapped [14], as is the organization of the oral apparatus. Figure 1 compares an non-injected control cell with a typical transformed cell, displaying a reduced size, a smaller and disorganized oral apparatus and an absence of post-oral fibers; more subtle defects concern the total or partial lack of doublets — cortical units with two basal bodies — normally present in well-defined territories. Table 1 recapitulates the cytological abnormalities recorded in the progeny of the transformed cells. 120 out of 124 observed cells presented reduced cell size and/or underdeveloped oral apparatus. Cell size was measured on all favorable pictures of stained cells (81 cells) and basal bodies were counted on magnified pictures for 33 transformed cells along with 11 control cells (Figure 2a). As in Figure 1, most of the transformed cells totally lacked cortical units with two basal bodies, though

44

Current Biology, Vol 9 No 1

Figure 1

(a)

Microinjection of the γ-tubulin coding sequence affects basal body pattern. Samples from the transformed clones were fixed 48 h after the injection and immunolabelled with an anti-α-tubulin antibody to visualize basal bodies. (a) Ventral surface of a control cell. (b) Ventral surface of a transformed cell. As well as the reduction in cell length, note the reduced size and abnormal morphology of the oral apparatus (oa), the absence of post-oral fibers (pof) and the total absence of double basal body units, normally present at precise locations on the cortex [25] and in particular on both sides of the anterior suture (as), the region shown in the magnifications. Bar, 5 µm.

(b)

as

oa

pof

Current Biology

some cells retained cortical units with such doublets. In transformed cells, the basal body number averaged half that of control cells. Cell length correlates well with basal body number and the 33 cells whose basal bodies were counted displayed the same distribution of cell length as the rest of the 81 photographed cells (Figure 2b).

basal body pattern, cell size, cell shape or survival. It seems reasonable, then, to conclude that the γ-tubulin genes are essential genes in Paramecium as in other species [2] and that γ-tubulin is necessary for basal body duplication, whereas it does not seem necessary for spindle pole body duplication in Saccharomyces cerevisiae [18,19].

Further microinjections were carried out for immunodetection of γ-tubulin in the transformed cells. Figure 3 shows a control cell and two transformed cells doubly labelled with an anti-α-tubulin antibody (ID5) and with an antibody raised against the carboxy-terminal part of Paramecium γ-tubulin, which decorates basal bodies, micronuclei and other microtubule organizing centres (C. Klotz et al., personal communication). The anti-γ-tubulin fluorescence present in the control cell (Figure 3b) is practically totally abolished in the transformed cells shown (Figure 3d,f). As for the effects of microinjection on cell size and basal body number (Figure 2), the level of extinction of γ-tubulin fluorescence was variable, ranging from practically total extinction (as in Figure 3) in about 50% of the stained cells, to nearly normal fluorescence in a few individuals. Altogether, these observations indicate a significant depletion of γ-tubulin in the transformed cells.

Table 1

In summary, we have shown here that microinjection of the coding sequence of either γ-PT1 or γ-PT2 results in depletion of γ-tubulin and specifically leads to a new phenotype: a block in basal body duplication and cell growth arrest. The specificity of this effect is attested by comparison with other cases of gene inactivation in Paramecium [16]: not only is a different phenotype observed in each case, that specifically reflects the absence or reduced amounts of the corresponding gene product, but the cytological analyses of these inactivations, which always included basal body staining with ID5 ([16]; F.R., unpublished observations), have never revealed any effect on

Abnormalities in basal body pattern in the progeny of microinjected cells. Experiment 1–3

Experiment 4–6

Total

Number of clones

99

87

186

Number of cells observed

61

63

124

Basal body pattern: Normal oral apparatus, normal cortex

3

1

4

Normal oral apparatus, abnormal cortex

12

22

34

Abnormal oral apparatus, normal cortex

0

8

8

Abnormal oral apparatus, abnormal cortex

46

32

78

Microinjections were carried out either on log phase (experiments 1–3) or stationary phase (experiments 4–6) cells. Within each series, cells from different clones were pooled, fixed and immunolabelled 48 h after injection. The number of clones contributing to each pool of fixed cells is indicated, as well as the number of stained cells that were eventually analyzable. Smaller cells are more often lost during the manipulations, so the proportion of the most abnormal cells is likely to be underestimated. Consideration of the normal/abnormal organization of the oral apparatus and of the cortex led us to classify the cells into four categories. Cortex was recorded as abnormal on the basis of reduced cell size and/or of the partial or total absence of cortical units with double basal bodies, normally present in well-defined cortical territories [25]. Oral apparatus was recorded as abnormal after reduction of its size and alteration of its organization. Numbers indicate observed cells in each category for each of the two series of injections.

Brief Communication

45

Figure 2 (a) 5000

(b)

20 18

4500

16

Number of cells

4000

3500

Total Cells shown in A

14 12 10 8 6 4

3000

2 0 2500

58 65 71 74 78 80 84 87 91 94 9 107 103 126 3

Number of cortical basal bodies

Microinjection of γ-tubulin coding sequence affects basal body number and cell size. When flattened between slide and coverslip, it is possible to focus on either of the two faces of immunodecorated cells such as those of Figure 1, so that the whole complement of their cortical basal bodies can be seen and their total number counted on magnified images of the two faces. Cell length was measured on the same images. (a) The number of cortical basal bodies and cell length for 33 cells among the progeny of the injected cells. Control 1, control cells in these experiments; control 2, wild-type cell values taken from previous work [14]. For the injected cells: cells devoid of any double basal body units are indicated by s; cells retaining such units, despite other signs of inhibition of basal body duplication, are indicated by s/d. (b) The distribution of cell lengths among the sample described in (a) is compared to that of 47 other photographed cells whose basal body number was not counted.

Cell length (µm) 2000 Injected: s Injected: s/d Control 1 Control 2

1500

1000 70

80

90 100 110

120 130

Cell length (µm)

As in other systems, γ-tubulin inactivation eventually blocks cell division. The remarkable observation here is that it is possible to detect an effect on basal bodies before growth arrest. This is possible because, in Paramecium, basal body duplication and cell cycle can be uncoupled [20,21]. The result is also consistent with the apparent kinetics of γ-tubulin inactivation, which becomes effective over one cell cycle, the last one, and essentially affects basal body duplication, a late event in the division process. This function of γ-tubulin is presumably of general relevance for all centriolar structures. Our observation raises the question of how γ-tubulin intervenes in basal body and centriole duplication. It could act simply by nucleating microtubule triplets, or by contributing to the establishment of the nine-fold symmetry pattern of the structure, two functions not necessarily exclusive in view of the recently proposed model for centriole duplication [22].

Current Biology

Microinjection of PCR products Previous experiments allowed us to verify that the efficiency of gene silencing obtained after microinjection of PCR-amplified genes was equivalent to that obtained with plasmid-cloned genes [16]. PCR amplifications were carried out on the respective cloned genes, using the primers γATG (ATGCCTCGCGAAATCATCACACT) and γTGA (TCACTGTTACTCTTCGATTCTCAT), resulting in amplification of both genes precisely limited by their initiation and stop codons. After primer removal, the DNA was concentrated and microinjected into the macronucleus as previously described [16].

Immunocytochemistry Immunolabeling was performed as previously described [14] using an anti-α-tubulin antibody ID5 [23] and an anti-γ-tubulin rabbit antiserum prepared against the carboxy-terminal part of the Paramecium γ-tubulin (C. Klotz et al., personal communication). ID5 labels basal bodies, the postoral fibers, a massive microtubule bundle, and the proximal part of the microtubular rootlets of the contractile vacuole systems [24]. The affinitypurified anti-γ-tubulin antibodies used here prominently decorate the basal bodies and the micronuclei (C. Klotz et al., personal communication).

Acknowledgements Materials and methods Molecular characterization of the Paramecium γ-tubulin genes Two pairs of primers, designed from the most conserved regions of known γ-tubulin sequences — the consensus MPREII/GGTSG and DREADGS/VLDVMR — were used for PCR amplification of genomic DNA. The resulting products of the expected sizes, corresponding respectively to the first and second third of the putative Paramecium γtubulin gene, used to probe Southern blots of genomic Paramecium DNA, suggested the existence of two different genes. The cloning of these PCR fragments, as well as those obtained by 5′ and 3′ RACE PCR, confirmed the existence of two homologous γ-tubulin sequences. Complete sequences of both were obtained from full-length genes amplified by PCR from genomic DNA and subsequent cloning.

We thank our colleagues J. Cohen, R. Karess, J.P. Mignot, L. Sperling and M. Wright for helpful suggestions and critical reading of the manuscript.

References

1. Oakley CE, Oakley BR: Identification of a g-tubulin, a new member of the tubulin superfamily encoded by mipA gene of Aspergillus nidulans. Nature 1989, 338:662-664. 2. Oakley BR: g-tubulin. In Microtubules. Edited by Hyams JS. New York: Wiley-Liss, Inc.; 1994:33-45. 3. Stearns T, Evans L, Kirschner M: g-tubulin is a highly conserved component of the centrosome. Cell 1991, 65:825-836. 4. Fuller SD, Gowen BE, Reinsch S, Sawyer A, Buendia B, Wepf R, Karsenti E: The core of the mammalian centriole contains gtubulin. Curr Biol 1995, 5:1384-1393.

46

Current Biology, Vol 9 No 1

Figure 3

(a)

(b)

mic cv

oa

cv pof

(c)

(d)

oa cv

cv

(e)

(f)

oa

Current Biology

Microinjection of γ-tubulin coding sequence affects γ-tubulin contents. Pools of ~ 30 injected cells mixed with the same number of noninjected control cells were fixed and double-labelled with the monoclonal anti-α-tubulin antibody ID5, revealed by anti-mouse rhodamine B isothiocyanate, and an affinity-purified fraction of the rabbit antiserum raised against Paramecium γ-tubulin, revealed by anti-rabbit fluorescein isothiocyanate. Cells were photographed successively under the red and green channel to detect the anti-α-tubulin (a,c,e) and the anti-γ-tubulin (b,d,f), respectively, with identical exposure times for controls and injected cells and the pictures processed under exactly the same conditions. (a) α-Tubulin and (b) γ-tubulin labelling in a control cell, viewed from its dorsal surface. In (a), the dorsal pattern of basal bodies is apparent; the strong fluorescence of the oral apparatus (oa) and post-oral fibers (pof), which are ventral organelles (well defined in Figure 1), here appears more diffuse; in contrast, the microtubules associated with the pores of the contractile vacuoles (cvs), which are dorsal structures, are visible. Basal body fluorescence is weaker than that seen in Figure 1 because ID5 was used here at a ten times lower concentration, to decrease the fluorescence passing through the red channel. In (b), the anti-γ-tubulin decoration exactly follows the basal body pattern; in addition, strong fluorescence of the two micronuclei is observed. (c–f) Two injected cells. In (c,e), the anti-α-tubulin decoration reveals the same defects in the basal body pattern and organization and size of the oral apparatus as detailed in Figure 1. In (d,f), practically no fluorescence is detected. Bar, 10 µm.

5. Stearns T, Kirschner M: In vitro reconstitution of centrosome assembly and function: the central role of gamma-tubulin. Cell 1994, 76:623-637. 6. Zheng Y, Wong ML, Alberts B, Mitchison T: Nucleation of microtubule assembly by a gamma-tubulin-containing ring complex. Nature 1995, 378:578-583. 7. Horio T, Usawa S, Jung MK, Oakley BR, Tanaka K, Yanagida M: The fission yeast g-tubulin is essential for mitosis and is localized at microtubule organizing centers. J Cell Sci 1991, 99:693-700. 8. Joshi HC, Palacios MJ, McNamara L, Cleveland DW: Gamma-tubulin is a centrosomal protein required for cell cycle-dependant microtubule nucleation. Nature 1992, 356:80-83. 9. Sunkel CE, Gomes R, Sampaio P, Perdigao J, Gonzales C: g-tubulin is required for the structure and function of the microtubule organizing center in Drosophila neuroblasts. EMBO J 1995, 14:28-36. 10. Dippel RV: The development of basal bodies in Paramecium. Proc Natl Acad Sci USA 1968, 61:461-468. 11. Muresan V, Joshi HC, Besharse JC: Gamma-tubulin in differentiated cell types: localization in the vicinity of basal bodies in retinal photoreceptors and ciliated epithelia. J Cell Sci 1993, 104:1229-1237. 12. Komarova Yu A, Ryabov EV, Alieva IB, Uzbekov RE, Uzbekova SV, Vorobjev LA: Polyclonal antibodies against human gamma-tubulin stain centrioles in mammalian cells from different tissues. Membr Cell Biol 1997, 10:503-513. 13. Moudjou M, Bordes N, Paintrand M, Bornens M: Gamma-tubulin in mammalian cells: the centrosomal and the cytosolic forms. J Cell Sci 1996, 109:875-887. 14. Iftode F, Cohen J, Ruiz F, Torres Rueda A, Chen-Shan L, Adoutte A, et al.: Development of surface pattern during division in Paramecium: I. mapping of duplication and reorganization of cortical cytoskeletal structures in the wild type. Development 1989, 105:191-211. 15. Liang A, Ruiz F, Heckmann K, Klotz C, Tollon Y, Beisson J, et al.: Gamma-tubulin is permanently associated with basal bodies in ciliates. Eur J Cell Biol 1996, 70:331-338. 16. Ruiz F, Vayssie L, Klotz C, Sperling L, Madeddu L: Homologydependent gene silencing in Paramecium. Mol Biol Cell 1998, 9:931-943. 17. Dupuis P: The b-tubulin genes of Paramecium are interrupted by two 27 bp introns. EMBO J 1992, 11:3713-3719. 18. Marschall LG, Jeng RL, Mulholland J, Stearns T: Analysis of Tub4p, a yeast gamma-tubulin-like protein: implications for microtubuleorganizing center function. J Cell Biol 1996, 134:443-454. 19. Spang A, Geissler S, Grein K, Schiebel E: Gamma-tubulin-like Tub4p of Saccharomyces cerevisiae is associated with the spindle pole body substructures that organize microtubules and is required for mitotic spindle formation. J Cell Biol 1996, 134:429-441. 20. Jerka-Dziadosz M, Ruiz F, Beisson F: Uncoupling of basal body duplication and cell division in crochu, a mutant of Paramecium hypersensitive to nocodazole. Development 1998, 125:1305-1314. 21. Ruiz F, Garreau de Loubresse N, Beisson J: A mutation affecting basal body duplication and cell shape in Paramecium. J Cell Biol 1987, 104:417-430. 22. Mignot JP: New hypothesis on the replication of centrioles and basal bodies. C R Acad Sci III 1996, 319:1093-1099. 23. Wehland J, Weber K: Turnover of the carboxy-terminal tyrosine of a-tubulin and means of reaching elevated levels of detyrosination in living cells. J Cell Sci 1987, 88:185-203. 24. Dupuis-Williams P, Klotz C, Mazarguil H, Beisson J: The tubulin gene family of Paramecium: characterization and expression of the alphaPT1 and alphaPT2 genes which code for alpha-tubulins with unusual C-terminal amino acids, GLY and ALA. Biol Cell 1996, 87:83-93. 25. Sonneborn TM: Positional information and nearest neighbour interactions in relation to spatial patterns in ciliates. Ann Biol 1975, XIV:565-584.