Receptors determine the cellular localization of a γ-tubulin complex and thereby the site of microtubule formation
1998; Springer Nature; Volume: 17; Issue: 14 Linguagem: Inglês
10.1093/emboj/17.14.3952
ISSN1460-2075
Autores Tópico(s)Ubiquitin and proteasome pathways
ResumoArticle15 July 1998free access Receptors determine the cellular localization of a γ-tubulin complex and thereby the site of microtubule formation Michael Knop Michael Knop The Beatson Institute for Cancer Research, CRC Beatson Laboratories, Garscube Estate, Switchback Road, Bearsden, Glasgow, G61 1BD UK Search for more papers by this author Elmar Schiebel Elmar Schiebel The Beatson Institute for Cancer Research, CRC Beatson Laboratories, Garscube Estate, Switchback Road, Bearsden, Glasgow, G61 1BD UK Search for more papers by this author Michael Knop Michael Knop The Beatson Institute for Cancer Research, CRC Beatson Laboratories, Garscube Estate, Switchback Road, Bearsden, Glasgow, G61 1BD UK Search for more papers by this author Elmar Schiebel Elmar Schiebel The Beatson Institute for Cancer Research, CRC Beatson Laboratories, Garscube Estate, Switchback Road, Bearsden, Glasgow, G61 1BD UK Search for more papers by this author Author Information Michael Knop1 and Elmar Schiebel1 1The Beatson Institute for Cancer Research, CRC Beatson Laboratories, Garscube Estate, Switchback Road, Bearsden, Glasgow, G61 1BD UK The EMBO Journal (1998)17:3952-3967https://doi.org/10.1093/emboj/17.14.3952 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info The yeast microtubule organizing centre (MTOC), known as the spindle pole body (SPB), organizes the nuclear and cytoplasmic microtubules which are functionally and spatially distinct. Microtubule organization requires the yeast γ-tubulin complex (Tub4p complex) which binds to the nuclear side of the SPB at the N-terminal domain of Spc110p. Here, we describe the identification of the essential SPB component Spc72p whose N-terminal domain interacts with the Tub4p complex on the cytoplasmic side of the SPB. We further report that this Tub4p complex-binding domain of Spc72p is essential and that temperature-sensitive alleles of SPC72 or overexpression of a binding domain-deleted variant of SPC72 (ΔN-SPC72) impair cytoplasmic microtubule formation. Consequently, polynucleated and anucleated cells accumulated in these cultures. In contrast, overexpression of the entire SPC72 results in more cytoplasmic microtubules compared with wild-type. Finally, exchange of the Tub4p complex-binding domains of Spc110p and Spc72p established that the Spc110p domain, when attached to ΔN-Spc72p, was functional at the cytoplasmic site of the SPB, while the corresponding domain of Spc72p fused to ΔN-Spc110p led to a dominant-negative effect. These results suggest that different components of MTOCs act as receptors for γ-tubulin complexes and that they are essential for the function of MTOCs. Introduction Microtubules are part of the cytoskeleton of eukaryotic cells with essential functions in chromosome segregation in mitosis and meiosis, cell polarity, organelle positioning, secretion and cellular movement (reviewed in Huffaker et al., 1987; Hyman and Karsenti, 1996). Microtubules are hollow cylinders, and the wall of the cylinder consists of tubulin, a heterodimer of α- and β-tubulin (reviewed by Mandelkow and Mandelkow, 1993). Microtubules form by the self-assembly of tubulin, a process that starts in vivo at so-called microtubule organizing centres (MTOCs) (reviewed in Brinkley, 1985; Kellogg et al., 1994; Pereira and Schiebel, 1997). The generic term MTOCs groups morphologically distinct structures with a common microtubule organization activity such as centrosomes, spindle pole bodies (SPBs) and basal bodies (Pickett-Heaps, 1969). The microtubule organization capability of MTOCs frequently is cell-cycle regulated. For example, the number of microtubules at mammalian centrosomes increases 5-fold at the onset of mitosis (Kuriyama and Borisy, 1981). Another example is the SPB from Schizosaccharomyces pombe which organizes nuclear microtubules only in mitosis (Hagan and Hyams, 1988; Masuda et al., 1992). Furthermore, one cell may have multiple MTOCs which are morphologically, spatially and functionally distinct. In this respect, plant cells are particularly interesting, since a number of different microtubule arrays exist: these include the cortical nuclear-associated microtubules in interphase, the pre-prophase band in G2 phase, and the mitotic spindle and the phragmoplast microtubules during mitosis (reviewed in Smirnova and Bajer, 1992; Marc, 1997). There is increasing evidence that the nuclear surface of higher plants serves as a MTOC during interphase and during telophase (Stoppin et al., 1994). Other MTOCs may exist, including the phragmoplast (Cleary et al., 1992). In the fission yeast S.pombe, interphase microtubules are organized mainly by a not very well defined MTOC that is localized at the cell equator (Hagan and Hyams, 1988). However, at the onset of mitosis, microtubules are organized from the two SPBs into a typical spindle (Ding et al., 1997). The SPB of Saccharomyces cerevisiae offers an example of one MTOC that can organize functionally and spatially distinct classes of microtubules (Byers and Goetsch, 1975). The S.cerevisiae SPB is embedded in the nuclear envelope during the entire cell cycle. Substructures named the outer, central and inner plaques have been described by electron microscopy (Byers and Goetsch, 1975; Byers, 1981) (Figure 8). The outer and inner plaques organize the cytoplasmic and nuclear microtubules, respectively. The cytoplasmic microtubules have functions in nuclear positioning and nuclear movement (Palmer et al., 1992; Sullivan and Huffaker, 1992), while the nuclear microtubules are involved in spindle formation and chromosome segregation in mitosis and meiosis (Jacobs et al., 1988). Figure 1.Spc72p interacts with Spc97p and Spc98p in the two-hybrid system. (A) The potential coiled-coil regions of Spc72p (bars below the Spc72p scheme) were predicted using either the Coils software of Lupas et al. (1991) (▪) or the Paircoil program of Berger et al. (1995) (□). (B) The N-terminal domain of Spc72p interacts with Spc98p and Spc97p in the two-hybrid system. The empty plasmid was used as a control (−). The blue colony colour on X-Gal plates indicates interaction. (C) Tub4p interacts with Spc72p after co-overexpression of SPC98 and SPC97. SPC98 and/or SPC97 were expressed under the control of the Gal1 promoter from CEN-TRP1 plasmids harbouring one or both promoter fusions. The two-hybrid interactions were assayed as described (Knop and Schiebel, 1997). Download figure Download PowerPoint A universal component of MTOCs is γ-tubulin which was first discovered in the fungus Aspergillus nidulans (Weil et al., 1986; Oakley and Oakley, 1989). Since then, the function of γ-tubulin in microtubule formation has been established by antibody microinjection experiments (Joshi et al., 1992), genetic studies (Oakley et al., 1990; Horio et al., 1991; Sobel and Snyder, 1995; Marschall et al., 1996; Spang et al., 1996a) and biochemical approaches (Li and Joshi, 1995; Zheng et al., 1995). Biochemical studies using extracts of frog eggs (Zheng et al., 1995), mammalian cells (Stearns and Kirschner, 1994; Moudjou et al., 1996), S.cerevisiae (Knop and Schiebel, 1997) and A.nidulans (Akashi et al., 1997) cells revealed that γ-tubulin is part of larger complexes. Purification of such a 25S complex from Xenopus laevis eggs identified α-, β- and γ-tubulin and at least four additional proteins (Zheng et al., 1995). The S.cerevisiae γ-tubulin, Tub4p, forms a stable complex with two other proteins, Spc98p and Spc97p (Geissler et al., 1996; Knop and Schiebel, 1997), and this Tub4p complex is localized at the outer and inner plaques of the SPB (Rout and Kilmartin, 1990; Knop et al., 1997) (Figure 8). Conditional lethal mutants in SPC98 and SPC97 revealed a function of the encoded proteins in microtubule organization by the SPB (Geissler et al., 1996; Knop et al., 1997). γ-Tubulin complexes assemble in the cytoplasm of cells (Stearns and Kirschner, 1994; Moudjou et al., 1996; Pereira et al., 1998). From S.cerevisiae we know that Spc98p and Spc97p of the assembled γ-tubulin complex (Tub4p complex) bind to the N-terminal domain of the SPB component Spc110p (Knop and Schiebel, 1997). Analogously, other γ-tubulin complexes may bind to MTOCs via such γ-tubulin complex-binding proteins (GTBPs). A GTPB may localize to only one MTOC within a cell and thereby contribute to the characteristic microtubule organization properties of an MTOC. Spc110p is only associated with the inner plaque, while the Tub4p complex is located with the outer and inner plaques. Therefore, a protein other then Spc110p has to function as a GTBP at the outer plaque. The goal of this study was to identify the yeast GTBP at the outer plaque. Using the yeast two-hybrid system, we identified a new, essential SPB component of the outer plaque, named Spc72p, whose N-terminal domain interacts with Spc98p and Spc97p of the Tub4p complex. We further established that Spc72p fulfils similar functions in microtubule organization at the outer plaque as Spc110p at the inner plaque. Results Cloning of SPC72 and its interaction with the Tub4p complex in the two-hybrid system Previously, we have shown that the N-terminal domain of Spc110p interacts with Spc98p and Spc97p, but not with Tub4p (Knop and Schiebel, 1997) in the yeast two-hybrid system (Fields and Song, 1989). However, an interaction with Tub4p was observed after co-overexpression of SPC98 and SPC97. We used these criteria to search for further SPB components that interact with the Tub4p complex. Our screen resulted in a prey plasmid containing ∼270 codons of the 5′ end of the open reading frame (ORF) YAL047c. Since further experiments showed that ORF YAL047c encodes a SPB component, we renamed YAL047c as SPC72, for SPB component with a molecular weight of 72 kDa. SPC72 is located on chromosome I and it encodes a protein of 622 amino acids. The analysis of the amino acid sequence revealed that stretches of Spc72p have a high probability of forming coiled-coil structures (Figure 1A), a structural motif that has been found in other SPB components such as Spc42p (Donaldson and Kilmartin, 1996) and Spc110p (Kilmartin et al., 1993). Spc72p does not show significant homology to any protein in the database. Figure 2.Spc72p interacts with the Tub4p complex. (A) Co-immunoprecipitation of Spc72p-3HA with the γ-tubulin complex. Extracts of wild-type (lanes 1, YPH499) or SPC72-3HA (lanes 2, ESM479) cells were incubated with anti-HA (12CA5) antibodies covalently cross-linked to protein A–Sepharose. Spc72p, Spc98p, Spc97p and Tub4p were detected in the immunoprecipitate by immunoblotting as indicated. (B) Co-immunoprecipitation of Spc97p-3HA with Spc98p, Tub4p and Spc72p. Extracts of wild-type (lanes 1, YPH499) or SPC97-3HA (lanes 2, YMK22; Knop et al., 1997) cells were used. Spc72p, Spc98p, Spc97p and Tub4p were detected in the immunoprecipitate by immunoblotting as indicated. (C) In vitro binding of the Tub4p complex to the N-terminal domain of Spc72p. Cells of strain YMK47 (SPC97-3ProA; Knop and Schiebel, 1997) were lysed with glass beads. The cleared extract was applied to glutathione–Sepharose columns, on which GST (expressed from pGEX-5X-1), GST–Spc42p1–241 (pSM363) or GST–Spc72p1–271 (pGP68) were immobilized. After washing the columns with buffer, bound proteins were eluted with buffer containing 10 mM glutathione. Tub4p and Spc98p in the wash (W, corresponding to the last 500 μl of the wash fraction) and eluate (E) fractions were detected by immunoblotting. Spc97p-3ProA was detected due to the binding of protein A to IgGs of the primary and secondary antibodies. The antibodies specific for Tub4p were raised against a GST–Tub4p fusion protein and therefore also recognize GST and GST–Spc42p1–241 in the immunoblots. The protein of ∼70 kDa is most probably Hsp70 from E.coli which co-purifies with many fusion proteins. (D) Genetic evidence for an interaction of Spc72p with the Tub4p complex. Cells of YPH499 (rows 1–4) and ESM387–3 (rows 5–8) were transformed with p413-Gal1 (rows 1 and 5), p413-GalS–SPC72 (rows 2 and 6; pSM573), p413-Gal1–SPC72 (rows 3 and 7; pSM572) or p423-Gal1–SPC72 (rows 4 and 8; pSM574). Serial dilutions of transformants grown in raffinose medium were plated out on selective plates containing either raffinose and glucose (glucose) or raffinose and galactose (galactose) as carbon sources. Download figure Download PowerPoint Subdomains and the entire coding region of Spc72p were tested for their interactions with Tub4p, Spc98p and Spc97p using the yeast two-hybrid system (Figure 1B). In agreement with our screening criteria, the N-terminal domain of Spc72p (Spc72p1–271; the numbers denote amino acids) showed two-hybrid interactions with Spc97p (Figure 1B, lane 4) and Spc98p (lane 5) but not with Tub4p (lane 3). We noticed that the strength of these interactions was dependent on whether Spc72p1ndash;271 was fused to Gal4p or lexA (compare lanes 4 and 8, and lanes 5 and 11). However, an indication that the two-hybrid binding of Spc72p1–271 to Spc97p is specific came from the observation that the temperature-sensitive spc97-20 (lane 16) and spc97-14 alleles (lane 14) showed no or a strongly reduced interaction with Spc72p1–271. In contrast to Spc72p1–271, no binding of the C-terminal domain (lanes 6; data not shown) or the entire Spc72p (lane 7; data not shown) to Spc98p or Spc97p was observed. Similarly to the N-terminal domain of Spc110p (Spc110p1–176) (Knop et al., 1997; Figure 1C, lane 7), a strong interaction of Spc72p1–271 with lexA–Tub4p was only observed when SPC98 and SPC97 were co-overexpressed simultaneously (Figure 1C, compare lanes 3 and 4 with 5), suggesting that the N-terminal domain of Spc72p interacted indirectly with lexA–Tub4p via binding to Spc98p and Spc97p in the Tub4p complex. It is noteworthy that the first 176 amino acids of Spc72p (Spc72p1–176) were sufficient for this interaction and behaved as Spc72p1–271 in all these experiments (lane 6; data not shown). In conclusion, the two-hybrid interactions of the subdomains of Spc72p with components of the Tub4p complex is remarkably reminiscent of Spc110p (Knop and Schiebel, 1997). In addition, the failure of Gal4–Spc72p to interact with Spc98p and Spc97p explains why the entire SPC72 was not obtained in the initial two-hybrid screen. Biochemical and genetic interactions of Spc72p with the Tub4p complex We looked for biochemical evidence for an interaction of the Tub4p complex with Spc72p. A yeast strain harbouring a functional chromosomal gene fusion of SPC72 with three copies of the haemagglutinin epitope (3HA) was constructed. SPC72-3HA cells were lysed under conditions which extracted ∼30% of SPC72p-3HA, hardly any of Spc110p and most of Spc98p, Spc97p and Tub4p from the cells (data not shown). Spc72p-3HA was then precipitated with anti-HA (12CA5) antibodies. Besides Spc72p-3HA, also Spc98p, Spc97p and Tub4p (Figure 2A), but not α- and β-tubulin (data not shown), were detected in the immunoprecipitate, indicating a physical interaction between the Tub4p complex and Spc72p. Analysis of the immunoprecipitation supernatants revealed that only a small percentage of the Tub4p complex co-precipitated with Spc72p-3HA, while >90% of the extracted Spc72p-3HA was precipitated. This suggests that only a minor fraction of the Tub4p complex is associated with Spc72p. We established that the co-precipitation of Tub4p, Spc98p and Spc97p with Spc72p-3HA was specific: no Tub4p complex was precipitated by the anti-HA antibodies from an extract of SPC72 cells (Figure 2A, lanes 1). To confirm that Tub4p, Spc98p, Spc97p and Spc72p interact, we precipitated Spc97p-3HA by anti-HA antibodies. Spc98p, Tub4p and Spc72p were detected in the precipitate and this co-precipitation was not observed from an extract containing Spc97p (Figure 2B). Figure 3.Spc72p is an essential protein that is associated with the outer plaque of the SPB. (A) The anti-Spc72p antibodies are specific. Immunoblots of crude extracts of wild-type SPC72 (YPH499; lanes 1 and 3) or SPC72-3HA cells (ESM479; lanes 2 and 4) were probed with monoclonal anti-HA (lanes 1 and 2) or affinity-purified polyclonal anti-Spc72p (lanes 3 and 4) antibodies. (B) Spc72p is an SPB component. Spc72p and tubulin of wild-type cells (YPH499) were stained by indirect immunofluorescence using affinity-purified rabbit anti-Spc72p and mouse monoclonal anti-tubulin (Wa3) antibodies. DNA was stained with DAPI. Bar: 4 μm. (C) Spc72p is associated with the outer plaque. Isolated SPBs of strain YPH499 were prepared for immunoelectron microscopy as described in Materials and methods. Spc72p was detected using the affinity-purified rabbit anti-Spc72p antibodies. Secondary antibodies were goat anti-rabbit IgGs conjugated to 15 nm gold particles. Only the outer, but not the inner plaque of the SPB was labelled. The inner plaque is still associated with the nuclear microtubules, while the cytoplasmic microtubules of the outer plaque were lost during the SPB purification (Rout and Kilmartin, 1990). Abbreviations: c, central plaque; i, inner plaque; m, microtubule; o, outer plaque. Bar: 100 nm. Download figure Download PowerPoint We then tested whether the Tub4p complex binds to GST–Spc72p1–271 purified from Escherichia coli, as is the case for GST–Spc110p1–204 (Knop and Schiebel, 1997). Yeast extract containing Tub4p complex was incubated with recombinant GST, GST–Spc42p1–214 or GST–Spc72p1–271 bound to a glutathione resin. Binding of Tub4p, Spc98p and Spc97p to the resins was analysed by immunoblotting, because these species are minor components in total yeast cell lysate. A fraction of the Tub4p complex present in the crude lysate bound to GST–Spc72p1–271, but not to GST or GST–Spc42p1–214, indicating specific binding (Figure 2C). Genetic evidence for an interaction of Spc72p with the Tub4p complex came from overexpression studies of SPC72. We found that moderate overexpression of SPC72 from the centromere-based GalS promoter caused a slight growth defect (Figure 2D, compare lanes 1 and 2, galactose). This defect was increased further by expressing SPC72 from the stronger Gal1 promoter (lane 3, galactose) and was finally lethal when the Gal1–SPC72 promoter fusion was on a 2 μm multicopy plasmid (lane 4, galactose). In contrast, all cells grew equally on the repressing glucose plates. Interestingly, the toxic effects of SPC72 overexpression were much weaker in strain ESM387-3 which carries the chromosomal Gal1–TUB4, Gal1–SPC98 and Gal1–SPC97 derivatives (Figure 2D, lanes 5–8). This result is explained most easily by a binding of Spc72p to the assembled Tub4p complex. Taken together, our biochemical and genetic analyses confirmed that the N-terminal domain of Spc72p interacts with the Tub4p complex. Spc72p is an essential component of the outer plaque of the SPB Spc72p may be a cytoplasmic Tub4p complex-binding protein, it could represent an additional subunit of the Tub4p complex, or it may function as a GTBP at the outer plaque. Only in the latter case we would expect to find Spc72p exclusively at the outer plaque of the SPB. To address these possibilities, we investigated the localization of Spc72p by indirect immunofluorescence and immunoelectron microscopy. We used affinity-purified anti-Spc72p1–271 antibodies for these experiments which were specific for Spc72p. This is indicated by the fact that predominantly one protein band with ∼85 kDa was detected in a cell lysate of SPC72 cells (Figure 3A, lane 3) and this band was shifted towards a higher molecular weight in an SPC72-3HA cell extract (compare lanes 3 and 4). A comparison with the anti-HA antibody (lanes 1 and 2) showed that the anti-Spc72p antibodies (lanes 3 and 4) were more specific. On immunoblots, some Spc72p was shifted to higher molecular weights, and sometimes even multipe bands were resolved. One reason for this behaviour could be phosphorylation of Spc72p. By indirect immunofluorescence, Spc72p was detected as one or two dots at the nuclear periphery of all cells of an unsynchronized culture by the anti-Spc72p antibodies (Figure 3B). Double labelling experiments with anti-tubulin antibodies established that Spc72p is associated with the spindle poles, exactly where the SPBs are situated. An identical cellular distribution was observed with a functional Spc72p–green fluorescent protein (Spc72p–GFP) fusion (data not shown). Figure 4.spc72-7 is defective in nuclear migration and nuclear division. (A) Strategy for the construction of temperature-sensitive alleles of SPC72 which carry mutations specifically in the Tub4p complex-binding domain of Spc72p. Codons 1–176 of SPC72 were mutagenized by PCR. The PCR product was combined with the non-mutagenized 3′ region of SPC72 by homologous recombination as described (Muhlrad et al., 1992). The regions where recombination occurs reside within the primer of the PCR product and, therefore, were not mutagenized. (B) SPC72 (YMK179-9) and spc72-7 (YMK179–3) cells were arrested by α-factor in the cell cycle in G1 with a 1N content (t = 0 h). Cells were released from their cell-cycle block by removing α-factor and they were simultaneously shifted to 37°C. Samples were taken every hour and analysed for their DNA content by flow cytometry. (C) spc72-7 cells fail to organize cytoplasmic microtubules and have a nuclear migration defect. Synchronized spc72-7 cells of (B) were fixed with formaldehyde after 3 h at 37°C. The fixed cells were analysed by indirect immunofluorescence using anti-tubulin antibodies. DNA was stained with DAPI. Cells were also inspected by phase contrast microscopy. Bar: 6 μm. (D) Phenotypes of spc72-7 cells. The spindle and nuclear migration defects of >200 synchronized SPC72, spc72-7 and spc72-14 cells incubated for 3 h at 37°C were determined. The distribution is given in percentages. Download figure Download PowerPoint The substructural localization of Spc72p at the SPB was investigated by immunoelectron microscopy. Isolated SPBs were incubated with the anti-Spc72p antibodies, followed by secondary antibodies coupled to colloid gold (15 nm). One to six gold particles were associated with the outer plaque region of the SPBs (n = 30; Figure 3C), suggesting that Spc72p is a component of the outer plaque. It is noteworthy that the inner plaque which is recognizable by the attached microtubules was never stained by the anti-Spc72p antibodies. To determine whether SPC72 is an essential gene, the entire coding region of SPC72 was disrupted in the diploid yeast strain YPH501 using the kanMX4 gene as disruption marker (ESM418). Analysis of tetrads from ESM418 suggested that SPC72 is indeed essential (data not shown). The essential function of SPC72 was confirmed by a plasmid shuffle experiment (Figure 7B). In summary, SPC72 encodes an essential SPB component that is associated with the outer plaque of the SPB. Figure 5.Overexpression of ΔN-SPC72 is dominant lethal and results in multi- and anucleated cells, while overexpression of SPC72 causes an increase in cytoplasmic microtubule staining. (A) Overexpression of ΔN-SPC72 is lethal. Cells of YPH499 containing the control plasmid p415-GalS, p415-GalS–ΔN-SPC72 (pMK252) or p415-GalS–SPC72 (pMK253) were grown on selective plates containing either glucose and raffinose (Glc) or galactose and raffinose (Gal) as carbon sources. Note that the GalS promoter is repressed by glucose and induced by galactose. (B) Expression levels of GalS–ΔN-SPC72 and GalS–SPC72. YPH499 cells containing plasmids p415-GalS, p415-GalS–ΔN-SPC72 or p415-GalS–SPC72NotI were grown in raffinose/galactose medium for 3 h at 30°C. Expression of the Spc72p derivatives was determined by immunoblotting using anti-Spc72p antibodies. Since YPH499 (SPC72) carrying p415-GalS–ΔN-SPC72 also expressed SPC72, Spc72p and ΔN-Spc72p were detected in the immunoblot. However, the signals are not comparable, because our anti-Spc72p antibodies are directed against the N-terminal 271 amino acids of Spc72p, of which the first 176 are missing in ΔN-Spc72p. (C) Overproduced ΔN-Spc72p and Spc72p replace Spc72p–GFP but not Spc42p–GFP from the SPB. Cells of SPC42–GFP (ESM440) or SPC72–GFP (ESM504) were transformed with the p415-GalS–ΔN-SPC72 and p415-GalS–SPC72 plasmids. Transformants were grown in selective medium containing either raffinose/glucose or raffinose/galactose as carbon sources for 9 h at 30°C (corresponding to two doubling times of the GalS cells). The cells were fixed briefly and the DNA was stained with DAPI. The number of GFP signals associated with ∼100 nuclei was determined by fluorescence microscopy. (D) Overexpression of ΔN-SPC72 results in multi- and anucleated cells, while overexpression of SPC72 gives rise to more cytoplasmic microtubules. Wild-type cells (YPH499) carrying plasmids p415-GalS, p415-GalS–ΔN-SPC72 or p415-GalS–SPC72 were grown in glactose medium as described in (C). Microtubules of fixed cells were analysed by indirect immunofluorescence using anti-tubulin antibodies. DNA was stained with DAPI. Cells were also inspected by phase contrast microscopy. Bar: 5 μm. Download figure Download PowerPoint Temperature-sensitive SPC72 mutants are defective in spindle elongation, cytoplasmic microtubule organization and nuclear migration The phenotype of mutants provides information about the function of the wild-type gene. We were interested especially in the phenotype of spc72 mutants with a defect in the N-terminal Tub4p complex-binding domain. Therefore, we constructed an N-terminal variant of SPC72 (ΔN-SPC72), that carried a deletion of amino acids 2–176. However, ΔN-SPC72 was unable to keep cells alive in the absence of the wild-type SPC72 (Figure 7B). To overcome this problem, we constructed and analysed temperature-sensitive spc72(ts) mutants with defects in the Tub4p complex-binding region of Spc72p (Figure 4A). Synchronized cultures of SPC72 and spc72-7 were shifted from 23 to 37°C. The DNA content, microtubule phenotypes and the distribution of the DNA were determined over time. SPC72 and spc72-7 cells replicated their DNA with similar kinetics (Figure 4B). However, while SPC72 cells continued in the cell cycle, as indicated by the appearance of cells with 1N DNA content after 3 h, a spc72-7 culture also showed cells with a DNA content >2N after 3 h. Figure 6.Thin section electron microscopic analysis of GalS, GalS–ΔN-SPC72 and GalS–SPC72 cells. YPH499 cells containing p415-GalS, p415-GalS–ΔN-SPC72 (pMK252) and p415-GalS–SPC72 (pMK253) were grown in galactose medium as described in Figure 5D. Thin serial sections of embedded cells were inspected by electron microscopy. Note that the contrast of the specimen is relatively weak due to growth in synthetic complete medium without leucine. (A) A wild-type spindle of GalS cells. (B) A section through a GalS–SPC72 SPB. (C) A spindle of a GalS–ΔN-SPC72 cell. (D) A GalS–ΔN-SPC72 cell with two separated nuclei. (E) An SPB of a GalS–ΔN-SPC72 cell. Arrowheads in (A) and (C) point towards the SPBs. Bars in (A), (C) and (D) are 320 nm and in (B) and (E) 200 nm. Abbreviations: c; cytoplasm; n, nucleoplasm. Download figure Download PowerPoint When spc72-7 cells were analysed using immunofluorescence microscopy, it was clear that most cells had strongly reduced cytoplasmic microtubule arrays (Figure 4C; for comparison, see Figures 3B and 5D, first column). We also noticed cells with an anaphase spindle in one cell body that still exhibited cytoplasmic microtubule remnants associated with the SPBs (Figure 4C, arrow). Figure 4D shows the distribution of the various microtubule phenotypes observed in wild-type, spc72-7 and spc72-14 cells 3 h after release fom the α-factor arrest at 37°C. As expected from this cytoplasmic microtubule deficiency, 29% of spc72-7 cells were anucleated and lacked any microtubules, and about the same percentage of spc72-7 cells had two separate 4,6′-diamidino-2-phenylindole (DAPI)-staining regions in the schmoo-containing mother cell (Figure 4C and D). These two DAPI regions were still connected by long misaligned nuclear microtubules, indicating that nuclear division was not complete. The residual cells (∼46%) had one DAPI-staining region with a short spindle of random orientation. The nucleus was either in the mother cell or in a cell without a bud. Nearly identical results were obtained with spc72-14 cells which differ from spc72-7 cells in their temperature sensitivity (data not shown and see Figure 4D). Taken together, our spc72(ts) cells show a clear defect in cytoplasmic microtubule functions, as indicated by the nuclear migration and spindle orientation defects. However, they show additional defects in spindle elongation or nuclear division that are not easily explainable. Figure 7.The Tub4p-binding domain of Spc110p is functional at the outer plaque. (A) Construction of N-SPC72–SPC110 and N-SPC110–SPC72 hybrids. NotI restriction sites were introduced by recombinant PCR after codon 1 and codon 176 of SPC72 and SPC110 [now named (pMK231) and SPC72NotI (pMK234)]. Using these NotI sites, we constructed ΔN-SPC72 (pMK232), ΔN-SPC110 (pMK230), N-SPC72–SPC110 (pMK233) and N-SPC110–SPC72 (pMK235). (B) The Tub4p complex-binding domain of Spc110p is essential for its function and N-SPC110–SPC72 rescues a SPC72 null mutant. Strain ESM335 (Δspc110::HIS3 pRS316-SPC110 'SPC110 shuffle strain') was transformed with plasmid pRS414 (sector 'control plasmid'), and pRS414 derivatives carrying SPC110 (pSM187), ΔN-SPC110 (pMK230) or (pMK231). Transformants were incubated on 5-FOA-containing plates at 30°C. Strain ESM448 (Δspc72::kanMX4, p
Referência(s)