QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

Vol. 19, Issue 1, 368-377, January 2008

This Article Abstract Full Text (PDF) Supplemental Materials All Versions of this Article: E07-08-0801v1 E07-08-0801v2 19/1/368    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Wiese, C. Search for Related Content PubMed PubMed Citation Articles by Wiese, C.

Distinct Dgrip84 Isoforms Correlate with Distinct -Tubulins in Drosophila

Department of Biochemistry, University of Wisconsin-Madison, Madison, WI 53706

Submitted August 19, 2007; Revised October 25, 2007; Accepted November 2, 2007
Monitoring Editor: Sandra Schmid

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
-Tubulin is an indispensable component of the animal centrosome and is required for proper microtubule organization. Within the cell, -tubulin exists in a multiprotein complex containing between two (some yeasts) and six or more (metazoa) additional highly conserved proteins named gamma ring proteins (Grips) or gamma complex proteins (GCPs). -Tubulin containing complexes isolated from Xenopus eggs or Drosophila embryos appear ring-shaped and have therefore been named the -tubulin ring complex (TuRC). Curiously, many organisms (including humans) have two distinct -tubulin genes. In Drosophila, where the two -tubulin isotypes have been studied most extensively, the -tubulin genes are developmentally regulated: the "maternal" -tubulin isotype (named Tub37CD according to its location on the genetic map) is expressed in the ovary and is deposited in the egg, where it is thought to orchestrate the meiotic and early embryonic cleavages. The second -tubulin isotype (Tub23C) is ubiquitously expressed and persists in most of the cells of the adult fly. In those rare cases where both -tubulins coexist in the same cell, they show distinct subcellular distributions and cell-cycle-dependent changes: Tub37CD mainly localizes to the centrosome, where its levels vary only slightly with the cell cycle. In contrast, the level of Tub23C at the centrosome increases at the beginning of mitosis, and Tub23C also associates with spindle pole microtubules. Here, we show that Tub23C forms discrete complexes that closely resemble the complexes formed by Tub37CD. Surprisingly, however, Tub23C associates with a distinct, longer splice variant of Dgrip84. This may reflect a role for Dgrip84 in regulating the activity and/or the location of the -tubulin complexes formed with Tub37CD and Tub23C.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The temporal and spatial control of microtubule nucleation and organization are essential for many fundamental cellular processes, including cell shape changes, intracellular transport, cell motility, partitioning of polarity signals during embryonic development, and, perhaps most importantly, the faithful segregation of chromosomes during cell division. In most animal cells, microtubules are nucleated and organized by the centrosome, which is composed of a pair of centrioles surrounded by the amorphous pericentriolar material (PCM) that appears to be the site of microtubule nucleation and anchoring (Gould and Borisy, 1977). Although efforts to generate an inventory of centrosomal components have made great progress in recent years (for example, Andersen et al., 2003), the molecular organization of the centrosome itself is not well understood, and mechanistic descriptions of how centrosomal proteins contribute to microtubule formation and organization are yet to be obtained.

-Tubulin (Oakley and Oakley, 1989) is an indispensable component of the animal centrosome required for microtubule nucleation and organization in all eukaryotes (reviewed in Wiese and Zheng, 2006). Human -tubulin can restore viability to Schizosaccharomyces pombe lacking endogenous -tubulin (Horio and Oakley, 1994), suggesting that -tubulins are functionally conserved in phylogenetically distant organisms. Within the cell, -tubulin exists in a multiprotein complex containing between two (in the case of Saccharomyces cerevisiae) and five or more (S. pombe, Xenopus laevis, Drosophila melanogaster, and humans) additional proteins named gamma ring proteins (Grips) or gamma complex proteins (GCPs). -Tubulin containing complexes isolated from Xenopus eggs or Drosophila embryos appear ring-shaped and have therefore been named -tubulin ring complex (TuRC; Zheng et al., 1995). The non--tubulin TuRC components are themselves highly conserved (Murphy et al., 1998, 2001; Anders et al., 2006; Tassin et al., 1998; Martin et al., 1998; Haren et al., 2006; Lüders et al., 2006). One of the major subcomplexes of the TuRC, which has been named the -tubulin small complex (TuSC; Oegema et al., 1999), appears to be a tetramer composed of two molecules of -tubulin and one molecule each of two other Grips known as Spc97p and Spc98p in S. cerevisiae, Dgrip84 and Dgrip91 in Drosophila melanogaster, or GCP2 and GCP3 in mammals (reviewed in Wiese and Zheng, 2006). The tetramer was first described in S. cerevisiae (Knop and Schiebel, 1997), where it may be the only soluble -tubulin complex. The metazoan TuRC is thought to be composed of six or seven tetramers held together and capped by several additional proteins that copurify with -tubulin from Xenopus egg extracts or Drosophila embryos (Zheng et al., 1995; Oegema et al., 1999; Gunawardane et al., 2003), but subcomplexes between -tubulin and non-TuSC Grips may also contribute to TuRC structure (Gunawardane et al., 2000).

Consistent with a role for the TuRC in microtubule nucleation, -tubulin and its binding partners are localized to the PCM at the minus ends of the microtubules (Moritz et al., 1995a,b; reviewed in Wiese and Zheng, 2006; Moritz and Agard, 2001). However, the molecular mechanism(s) for the function of the TuRC in microtubule nucleation are still unclear. In addition to facilitating nucleation of microtubules, -tubulin also seems to play a role in interphase microtubule organization (Jung et al., 2001; Horio and Oakley, 2003), spindle assembly (Paluh et al., 2000; Prigozhina et al., 2001; Vogel and Snyder, 2000a,b; Sampaio et al., 2001; Barbosa et al., 2003; Lüders et al., 2006), and spindle assembly checkpoint regulation (Vardy and Toda, 2000; Hendrickson et al., 2001; Vardy et al., 2002; Müller et al., 2006; reviewed by Blagden and Glover, 2003). Little is known to date about how -tubulin or the Grips participate in each of these functions.

Curiously, the genomes of several organisms, including humans and flies, encode two distinct -tubulin genes (Zheng et al., 1991; Wilson et al., 1997; Tavosanis et al., 1997; Wise et al., 2000). In Drosophila, where the two -tubulin isotypes have been studied most extensively, the -tubulin genes are developmentally regulated: the "maternal" -tubulin isotype (named Tub37CD according to its location on the genetic map) is highly expressed in the ovary and is deposited in the egg, where it is thought to orchestrate the early embryonic cleavages (Wilson et al., 1997; Llamazares et al., 1999). Consistent with this idea, mutation of Tub37CD results in female sterility (Tavosanis et al., 1997). In contrast, Tub23C is essentially ubiquitous and is required for viability, microtubule organization during somatic mitotic divisions, and male meiosis (Sunkel et al., 1995; Sampaio et al., 2001). Although both isotypes are present in early embryos, only Tub37CD is detected at centrosomes in syncytial embryos, whereas punctate structures containing Tub23C distribute throughout the cytoplasm (Wilson et al., 1997). Later during development, both Tub23C and Tub37CD localize to centrosomes in cellularized embryos (Tavosanis et al., 1997; Wilson et al., 1997; Raynaud-Messina et al., 2001). The extent to which each -tubulin isotype contributes to the assembly of female meiotic spindles remains to be resolved (Tavosanis et al., 1997; Wilson and Borisy, 1998; Llamazares et al., 1999).

Differences in the timing of their association with the centrosome have also been reported for the two -tubulin isotypes in Drosophila tissue culture cells (Raynaud-Messina et al., 2001). In Kc23 cells, the centrosomal levels of Tub37CD have been reported to vary only slightly throughout the cell cycle, whereas centrosomal Tub23C levels increased sharply during late G2. Furthermore, Tub23C was also recruited to the microtubules of the mitotic spindle, whereas Tub37CD was never found on spindle microtubules. This suggests the existence of mechanisms that allow the cell to distinguish between the two -tubulin isotypes. Here, we set out to gain a better understanding of the biochemistry of Tub23C. We report that Drosophila Tub23C isolated from tissue culture cells forms bona fide ring complexes that include most of the same Grips as the TuRCs formed by Tub37CD. However, we were surprised to find that Tub23C and Tub37CD associate with distinct splice variants of Dgrip84/GCP2 that differ by 74 amino acids in their amino termini. We speculate that Dgrip84 may be involved in regulating the subcellular localization of -tubulin.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Buffers and Reagents
The following buffers were used: HB: 50 mM K-HEPES, pH 7.6, 1 mM MgCl₂, 1 mM EGTA, 1 mM β-mercaptoethanol (β-ME) and protease inhibitor stock (1:200 final dilution; see below). H100: HB plus 100 mM NaCl; H200: HB plus 200 mM NaCl; and H500: HB plus 500 mM NaCl. Elution buffer (EB): H200 plus 100 µM GTP and 1 mg/ml Dros23C or Dros37C peptide. Homogenization buffer (for embryos): H100 plus 10%glycerol, 1 mM PMSF and protease inhibitor stock (1:100 final dilution). Cell lysis buffer (for S2 and Kc cells): H200 plus 0.5% NP-40, 1 mM GTP, and 1 mM PMSF. Protease inhibitor stock: 10 mM benzamidine-HCl, 1 mg/ml aprotinin, 1 mg/ml leupeptin, and 1 mg/ml pepstatin A in ethanol. LPC: 10 mg/ml leupeptin,10 mg/ml pepstatin A, and 10 mg/ml chymostatin dissolved in DMSO. Mounting medium: 20 mM Tris-Cl, pH 9.0, 90% glycerol, and 0.1% p-phenylenediamine. Sample buffer: 250 mM Tris, pH 6.8, 12% SDS wt/vol, 20% β-ME vol/vol, and 40% glycerol vol/vol. Tubulin was purified from the bovine brain and labeled with tetramethylrhodamine as described (http://skye.med.harvard.edu). For RNA work, aqueous solutions were treated with 0.05% diethyl pyrocarbonate (DEPC; Research Products International, Mount Prospect, IL) and autoclaved for 30 min.

Antibody Production
Dgrip84 antibodies were a kind gift from Y. Zheng (Carnegie Institution of Washington, Baltimore, MD) and are described in Oegema et al. (1999). Synthetic peptides (made by the Protein Synthesis facility of the University of Wisconsin, Madison Biotech Center) corresponding to the COOH-terminal 17 amino acids of Tub23C (Cys-PVDSKSEDSRSVTSAGS; GenBank Accession no. P23257) or Tub37CD (Cys-QIDYPQWSPAVEASKAG; GenBank Accession no. P42271) were used to raise and affinity purify rabbit polyclonal antibodies as described elsewhere (Field et al., 1998). Peptides contained an additional N-terminal cysteine residue to facilitate coupling to KLH for antibody production, or to SulfoLink coupling gel (Pierce Biotechnology, Rockford, IL) for affinity purification. Specific antibodies were purified as described (Kellogg and Alberts, 1992), except that before affinity purification of the antisera with the immunogenic peptide, each serum was passed over a column corresponding to the nonimmunogenic -tubulin isotype. This procedure depleted potential cross-reacting antibodies and thus ascertained the specificity of the affinity-purified antibodies.

Drosophila S2 Cell Tissue Culture
Schneider S2 cells (Schneider, 1972) were cultured in Schneider's Drosophila medium as described by the supplier (Invitrogen, Carlsbad, CA). Cells to be used for extract preparation were grown for 4 d at 27°C in 70 ml Drosophila medium in T-150 flasks. Cells were collected by centrifugation and washed twice with PBS, the volume of the cell pellet was estimated, and the cells were stored at –80°C until use.

PEG Precipitation
PEG (polyethylene glycol P-2139; average mol wt 5–8000; Sigma Chemical Co., St. Louis, MO) was added to a final concentration of 1–5% (from a 30% stock in H100) to clarified Drosophila embryo or S2 cell extract. The mixture was incubated on ice for 20 min, spun at 17,000 rpm for 10 min in a JA20 rotor (Beckman, Fullerton, CA), and the supernatant was removed. The pellets were resuspended in a volume corresponding the original extract volume of H200 plus 0.05% NP-40 and 100 µM GTP and were clarified at 17,000 rpm for 10 min in a JA20 rotor. Equivalent portions of supernatant and pellet fractions were then analyzed by sucrose gradient fractionation, SDS-PAGE, and Western blotting.

Immunoisolation of -Tubulin–containing Complexes from Drosophila Embryo Extract
Drosophila embryo extract was prepared by homogenizing Drosophila embryos in homogenization buffer as described (Moritz et al., 1998; Oegema et al., 1999). Embryo extract was frozen in aliquots in liquid nitrogen, stored at –80°C, and used within 2 wk of preparation. Clarified extract was prepared by centrifugation of thawed crude extract for 1 h at 50,000 rpm in a rotor (SW55; Beckman) at 4°C. -Tubulin complexes were immunoprecipitated from the supernatant essentially as described (Oegema et al., 1999), except that the PEG precipitation step was omitted. Briefly, anti--tubulin antibody (23C or 37CD; 8 µg/ml) was incubated with the supernatant at 4°C for 1 h with gentle rotation. Per 12 ml of clarified extract, 35 µl of protein A affiprep beads (Bio-Rad Laboratories, Hercules, CA) were equilibrated in wash buffer (H200 plus 0.05% NP-40, 100 µM GTP, and 1 mM PMSF) and added to the antibody-extract mix. After a further incubation for 1 h with gentle rotation, the beads were collected and washed six times with wash buffer and were incubated for 16–18 h at 4°C with an equal volume of EB containing the respective peptide. -Tubulin complexes were collected by loading an additional 40 µl of EB onto the beads and collecting the supernatant.

Immunoisolation of -Tubulin–containing Complexes from Drosophila Tissue Culture Cells
Drosophila tissue culture cell extract was prepared by resuspending thawed S2 cell pellets in 3 vol of cell lysis buffer, brief vortexing, and incubation on ice for 10 min. Cell debris was removed by a 10-min spin (at 4°C) at 2400 rpm in a Beckman table-top centrifuge. The supernatant was then incubated with antibody and processed as described above for the immunoisolation of -tubulin complexes from embryo extract.

Sucrose Gradient Sedimentation
Sucrose gradients (5–40%) were poured and run as described by Oegema et al. (1999). Briefly, gradients were poured by hand as step gradients (five steps of equal volume) in H100 plus 1 mM GTP and allowed to diffuse into continuous gradients. After centrifugation at 50,000 rpm in a Beckman TLS-55 rotor for 4.5 h at 4°C, gradients were fractionated by hand (16 fractions plus pellet). Fractions were separated on a 10% SDS-PAGE gel and analyzed by Western blot or Coomassie stain, as indicated. For some experiments, each fraction was precipitated by addition of 10% trichloroacetic acid (TCA) and centrifugation (30 min) in an Eppendorf centrifuge after a 30-min incubation on ice. Pellets were resuspended in SDS-sample buffer and analyzed by SDS-PAGE and Western blot.

Electron Microscopy
Negative staining electron microscopy of peptide-eluted complexes was described elsewhere (Zheng et al., 1995; Wiese and Zheng, 2000). Samples were viewed at 80 kV in a Philips Tecnai 12 microscope (Mahwah, NJ) equipped with a cooled CCD camera.

RT-PCR to Detect Dgrip84/GCP2 Isoforms
Total RNA was extracted from 100 µl of frozen 0–4 h Drosophila embryos or 10⁷ S2 cells by homogenizing the cells in Trizol reagent (Invitrogen) and extracting the RNA according to the manufacturer's recommendations.

The following primers were used to reverse transcribe Dgrip84 mRNA (all isoforms) and amplify specific regions of the gene to distinguish the isoforms: the first strand cDNA template was synthesized from the extracted RNA using the ImPromII-Reverse Transcription System (Promega, Madison, WI) according to the manufacturer's recommendations and using the Dgrip84/GCP2 gene-specific primer, 5'-CTCTACACACATCTTGTC-3'. Isoform-specific PCR primers used to differentiate the Dgrip84 isoforms were as follows: To distinguish the 5' end of the "A"-isoform from B/C using three PCR primers: forward primer 5'-AAAGAGCTCAATTGCC-3' (specific for the 5' untranslated region of isoform A) combined with reverse primer 5'-TCGAGCAGCATACTC-3' (amplifies all isoforms) gives an 885-base pair band if isoform A is present, no product for isoform B or C; forward primer 5'-CAGTCGATCTAGCCGTTG-3'(specific for isoforms B and C, because this primer targets a region that is spliced out in isoform A) with reverse primer 5'-TCGAGCAGCATACTC-3'(amplifies all isoforms) gives an 855-base pair band if isoform B or C is present. To test for the presence of the 3' insertion that distinguishes isoform C from isoforms A and B, we used forward primer 5'-GCTAGTCTCTATCGAG-3' and reverse primer 5'-CTGTTGATCTGCTTGAGG-3'. This primer pair amplifies a 500-base pair band if the insert is present, or a 400-base pair band if the insert is absent. PCR products were cloned and sequenced to confirm their identity.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
To analyze Tub23C and its interacting proteins we began by raising antibodies specific for Tub23C or Tub37CD (Figure 1). Sequence alignments and molecular modeling (Figure 1, A and B) revealed that the differences between Tub23C and Tub37CD (a total of 81 amino acids) did not cluster to defined regions within the -tubulin molecule but were distributed across most of the surfaces (Figures 1B and Supplementary Figure S1). The biggest difference between Tub23C and Tub37CD was a 14-amino acid insertion near the C-terminus of Tub23C. We took advantage of this difference to generate -tubulin isotype-specific antibodies by immunizing rabbits with synthetic peptides corresponding to the C-terminal 17 amino acids of each -tubulin isotype (see Materials and Methods). Dot-blot analysis confirmed that the affinity-purified antibodies were specific for their respective immunogenic peptides and did not cross-react with the nonimmunogenic peptide (Figure 1D). As expected, the antibodies recognized bands of 50 kDa (Tub37CD antibodies) and 52 kDa (Tub23C antibodies), respectively, on Western blots of embryo extracts (Figure 1E) or of purified -tubulin complexes (Figure 1F). In addition, the Tub23C antibodies recognized a band of 50 kDa in S2 (Drosophila tissue culture) cell extracts, whereas no band corresponding to -tubulin was recognized by the Tub37CD antibodies in S2 cell extracts (Figure 1E). This suggests that our S2 cell line does not express Tub37CD, or only undetectably low levels.

View larger version (60K): [in this window] [in a new window]   Figure 1. Analysis of Drosophila Tub23C and Tub37CD and isotype-specific antibodies. (A) Sequence alignment between Tub37CD and Tub23C reveals that nonconserved differences between the two isotypes are scattered throughout the primary sequence. Light gray, identities; medium gray, conserved changes; red, differences that change the character of the amino acid; blue, amino acid insertions in Tub23C relative to Tub37CD; the transparent amino acids near the C-termini of the molecules are not represented in the structural model. (B) A hypothetical structural surface model of Tub23C (generated by the structure prediction program SwissModel; Peitsch, 1995; Guex and Peitsch, 1997; Schwede et al., 2003) is based on the crystal structure for human -tubulin (Aldaz et al., 2005). To highlight the residues not conserved between the isotypes, the backbone of the residues whose character is significantly different between Tub23C and Tub37CD are colored in red and the M-loop insertion in Tub23C that is not present in Tub37CD is colored in blue, as in A. Snapshot images of different faces of the molecule are shown in each panel, as indicated in the upper left-hand corner. The orientation of the snapshots are based on the observation that -tubulin can roughly be modeled as a cube, and on the orientation of and β tubulin within the microtubule lattice, as illustrated in (C). The snapshots illustrate two important points: first, the differences between the two -tubulins are not randomly distributed throughout the molecule but map mainly to the surfaces, presumably leaving the core of the structure largely unaffected (see also Supplementary Figure S1). Second, the differences map to all six faces of the cube. Corresponding ribbon models can be found in the online supplementary material, Figure S1. (D) Dot-blot analysis of affinity-purified antibodies. Aqueous solutions of each peptide were spotted onto a piece of nitrocellulose membrane, and the membrane was probed with each antibody, as indicated. The antibodies do not cross-react with each other. (E) Western blot of total extracts from embryos (emb, lanes 1 and 3) or S2 cells (S2 cells, lanes 2 and 4) probed with antibodies to Tub37CD (lanes 1 and 2) or Tub23C (lanes 3 and 4). Positions of molecular weight markers are indicated on the right. (F) Western blot of purified -tubulin complexes. Lane 1, Tub23C complexes probed with Tub37CD antibodies; lane 2, Tub37CD complexes probed with Tub23C antibodies. The 40-kDa band recognized by the Tub37C antibodies in embryo extracts probably shares an epitope with Tub37CD, as this band cross-reacts with antisera produced by several different rabbits (our unpublished observation); however, this protein apparently does not coimmunoprecipitate with the TuRC and thus did not interfere with the analysis of the purified complex.

View larger version (86K): [in this window] [in a new window]   Figure 2. Drosophila tissue culture cells express a slower migrating version of Dgrip84. (A) Western blot analysis of TuRC components in embryos, S2 cells, and Kc cells, as indicated above the lanes. Probes are indicated below the blots and are as follows: (a) Tub37CD, (b) Tub23C, (c) Dgrip98/GCP3, (d) Dgrip128/GCP5, (e) Dgrip163/GCP6, and (f) Dgrip84/GCP2. The embryonic "cognate" proteins are indicated by arrowheads on the left of each panel. Asterisks in f denote two bands (100 and 75 kDa) in tissue culture cells that are recognized by the Dgrip84 antibody. Note the absence in the tissue culture cells of a band corresponding to Tub37CD in panel a, and the absence of a band corresponding to embryonic Dgrip84 in panel f. (B) Sucrose gradient analysis of S2 cell extracts reveals that the slower migrating protein recognized by the Dgrip84 antibodies migrates similarly to Tub23C. Sucrose gradient fractions were probed for TuRC components as indicated on the right. The positions of TuSC and TuRC are indicated below the blot. Asterisks on the left denote the two proteins in tissue culture cell extracts that are recognized by the Dgrip84 antibody. Only the 100-kDa protein migrates like the embryonic Dgrip84 (compare with Figure 3).

View larger version (69K): [in this window] [in a new window]   Figure 3. Sucrose gradient analysis of -tubulins and Dgrips in embryos and tissue culture cells. (A) Embryo extracts, 0–8 h, were analyzed on 5–40% sucrose gradients. Gradient fractions were probed for Dgrips and -tubulins as indicated on the right. (B and C) Extracts made from Kc cells (B) or S2 cells (C) were analyzed on sucrose gradients as described for A. D. Individual panels from A to C were regrouped to allow a more direct comparison for each TuRC component between the migration patterns in embryos and tissue culture cells. Tub23C migrates differently in embryos, but in S2 and Kc cells resembles the behavior of Tub37CD. For Dgrip84, only the 100-kDa bands are shown in S2 and Kc cells.

View larger version (74K): [in this window] [in a new window]   Figure 4. Dgrip84 isoforms in Drosophila. (A) Diagram of the relationships between different Dgrip84 isoforms. The position of the 33-amino acid C-terminal insertion and the 74-amino acid N-terminal extension are indicated. Theoretical molecular weights, number of amino acids, and pI are indicated on the right for each isoform (A–D). Isoforms A–C are currently present in the database (Accession no. NP_728264, NP_728265, and NP_523409, respectively). The isoform found in S2 cells (see below) has the extension but not the insertion. We named this the "D" isoform. (B) Amino acid sequence of Dgrip84 isoform A. The N-terminal extension and C-terminal insertion are indicated by gray shading and a box, respectively. (C) Gene structure of the Dgrip84 isoforms. Numbers below the diagram indicate the number of nucleotides in each exon (top row) or intron (bottom row). (D) Western blot of staged embryos. Drosophila embryos were collected for 4 h and were then aged for the times indicated above the blots. Maximum age was 24 h. Embryos were then homogenized, loaded onto SDS-PAGE gels, and Western blotted. The blots were probed with antibodies to Dgrip163, Tub37CD, Tub23C, or Dgrip84, as indicated on the right. The positions of the slower migrating ("C") and faster-migrating ("D") Dgrip84 bands are indicated on the left. (E) Northern blot of staged embryos probed with a Dgrip84 probe common to all isoforms. Embryos were collected and aged as in D. Early embryos (0–4 h) express a shorter mRNA than later embryos or S2 cells. To determine the identity of the isoforms, we performed RT-PCR and sequenced the relevant regions of cDNAs recovered form early embryos or S2 cells (see Supplementary Figure S3). This analysis shows that early embryos express the "C" isoform, whereas S2 cells express the "D" isoform. The reduction in signal for the older embryos (starting with 12–16-h embryos) may be due to sample degradation, although our loading control (rp49; DeZazzo et al., 2000) remained unchanged.

Western blots of staged embryos probed with anti-Dgrip84 antibodies showed that Dgrip84 (marked "D" in Figure 4D) migrated more slowly in older embryos compared with younger ones. The slower-migrating form of Dgrip84 was detectable within 8–12 h of development, and its relative levels increased as the levels of the shorter isoform (marked "C" in Figure 4D) slowly declined.

To determine whether the appearance of the slower-migrating Dgrip84 protein could be explained by a switch to a longer Dgrip84 isoform, we examined the isoforms expressed in staged embryos by Northern blot analysis. Northern blots showed that the shorter isoform of Dgrip84 expressed in early embryos (0–4 h) was replaced by a longer isoform later in development that was also present in S2 cells (Figure 4E). To identify the specific isoforms present in early embryos and S2 cells, we first designed a strategy to identify the isoform by RT-PCR using specific primers (diagramed in Supplementary Figure S2). Our results were then confirmed by direct sequencing of the RT-PCR product. Consistent with results from previous work (Oegema et al., 1999; Gunawardane et al., 2000), we found that the isoform present in early embryos lacked both the N-terminal extension and the C-terminal insertion. This suggested that the embryos expressed mainly the "C"-isoform of Dgrip84. S2 cells, on the other hand, expressed mRNAs that contained the N-terminal extension. However, we were surprised to find no evidence for the C-terminal insertion. This suggested that the major isoform of Dgrip84 expressed in S2 cells is the previously unknown, novel "D" splice variant of Dgrip84 (Figure 4A). Together, these results suggested that the expression of Dgrip84 splice variants is developmentally regulated.

To confirm that the longer Dgrip84 splice variant is indeed a component of the -tubulin complex, we used our -tubulin isoform-specific or Dgrip84 antibodies to immunoprecipitate protein complexes from embryos or S2 cells and compared these by Coomassie-stained SDS-PAGE gels and Western blot (Figure 5). With the exception of the band that corresponds to Dgrip84, similar sets of proteins were precipitated from S2 tissue culture cells and from 0- to 3-h embryos using antibodies directed against Dgrip84 (Figure 5Aa), and their protein profile closely resembled the Tub37CD and Tub23C complexes isolated from embryos and S2 cells, respectively (Figure 5Ab; Oegema et al., 1999). This suggested that both Dgrip84 slice variants are components of -tubulin complexes. One important difference between S2 and embryonic -tubulin complexes was the absence of the band corresponding to the embryonic Dgrip84 and a concomitant appearance of a 100-kDa band in the S2 -tubulin complexes. Thus, consistent with the findings described above, the -tubulin complexes isolated from embryos or S2 cells differed by the associated Dgrip84 splice variant.

View larger version (40K): [in this window] [in a new window]   Figure 5. Dgrip84-D is a component of the TuRC. (A) Comparison between protein complexes immunoprecipitated with various antibodies. Panel a, Coomassie-stained gel of proteins that coimmunoprecipitate with Dgrip84 antibodies from embryo (lane 2) or S2 cell (lane 3) extracts. TuRC proteins are highlighted by arrowheads on the right, positions of molecular-weight markers are indicated on the left. Panel b, -tubulin complexes isolated from S2 cells using antibodies to Tub23C (lane 1), or from 0-to 8-h embryos using antibodies to Tub37CD (lane 2). Complexes isolated from S2 cells with antibodies to Dgrip84 or Tub23C resemble each other, as do complexes isolated from embryos by antibodies to Dgrip84 or Tub37CD. Complexes isolated from S2 cells lack a band corresponding to the embryonic Dgrip84; however, a new band appears above Dgrip91. (B) Negative-staining electron microscopy of Tub23C complexes isolated from S2 cells. Arrowheads in the left panel point to individual ring complexes. Scale bars, 25 nm in the left panel, 100 nm in the right panel. (C) Western blot of proteins that copurify with Tub37CD from early (0–8 h) embryos (lane 1), with Tub23C from late (8–16 h) embryos (lane 2), or with Tub23C from S2 cells (lane 3). Antibodies to Dgrips used to probe the Western blot are indicated on the right. The asterisks mark the two Dgrip84 isoforms that copurify with Tub37CD or Tub23C. The multiple bands in lane 3 might represent additional Dgrip84 isoforms or degradation products.

The purified complexes were further analyzed by Western blotting with antibodies directed against embryonic Dgrips (Figure 5C). The Dgrips we tested (163/GCP6, 128/GCP5, and 98/GCP3) all copurified with Tub23C complexes isolated from late embryos (8–16 h) or S2 cells and migrated similarly on a 10% SDS-PAGE gel as their early embryonic counterparts (Figure 5C). Consistent with the results described above, the -tubulin complexes contained distinct but immunologically related Dgrip84 isoform(s), depending on which -tubulin isotype was present: Tub37CD correlated with a faster migrating Dgrip84, whereas the presence of Tub23C correlated with a slower migrating isoform. Interestingly, a small amount of Tub23C copurified with Tub37CD and vice versa (Figure 5C; see also Figures 1F and 6C), suggesting that some TuRCs may contain both -tubulin isotypes.

View larger version (46K): [in this window] [in a new window]   Figure 6. Tub23C migrates as a large complex in early embryos that is distinct from the embryonic TuRC. (A) Coomassie gel of proteins that precipitate from early embryos with antibodies to Tub23C (lane 2) or Tub37CD (lane 3). Positions of TuRC components are indicated on the right. Lane 1, molecular-weight markers (sizes of markers are indicated on the left). (B) Sucrose gradient analysis of extracts made from 0- to 8-h Drosophila embryos. Western blot of 0–8 h embryo extracts fractionated on a sucrose gradient probed with antibodies specific for each -tubulin isotype or Dgrips 163, 128, 98, and 84, as indicated on the right. The positions of the TuRC (fractions 13–15), the TuSC (fractions 6–8), and the peak fractions for Tub23C (fractions 11 and 12) are indicated below the blot. The gel is loaded such that the lighter sucrose fractions (starting at 5%) are on the left, and the heavier fractions (up to 40%) are on the right. Fraction numbers are indicated above the blots. Load, 1 µl of starting material was loaded as a control. (C) Tub23C and Tub37CD can be distinguished by their sensitivity to polyethyleneglycol (PEG). PEG (2.5%) was added to extracts, and precipitated proteins were separated from soluble proteins by centrifugation. Supernatant (S) and resuspended pellet (P) fractions were then fractionated on sucrose gradients, run on a SDS-PAGE gel, and blotted for Tub23C or Tub37CD as indicated on the right.

In contrast, Tub23C comigrated with the Dgrips in tissue culture cells (Figure 3; see also Figure 2B), consistent with the observation that it forms a TuRC in these cells (see above). Although it is not yet clear at which developmental stage Tub23C switches from the "embryonic" complex to become part of the TuRC, or how this process is regulated, these findings support the idea that the formation of TuRCs containing Tub23C is developmentally controlled.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Distinct patterns of recruitment to the centrosome and spindle microtubules exhibited by Tub23C and Tub37CD in tissue culture cells and syncytial embryos (Sunkel et al., 1995; Wilson et al., 1997; Raynaud-Messina et al., 2001, 2004) prompted us to examine the molecular compositions of the complexes formed by each of the two Drosophila -tubulin isotypes. We reasoned that differences in their localization might be regulated by intrinsic differences between the -tubulin isotypes or their binding partners, or both.

We analyzed -tubulin complexes in two different systems: in Drosophila embryos, where Tub37CD is the major -tubulin isotype, and in Schneider S2 tissue culture cells, where Tub23C is the major -tubulin isotype and Tub37C plays only a minor role, if any (Raynaud-Messina et al., 2004). Based on the observation that -tubulin and its interacting proteins are highly conserved among distantly related species, it was reasonable to suspect that Tub23C forms complexes in Drosophila cells that closely resemble the TuRC_37CD. Here, we provide the first experimental evidence that this is indeed the case. We show that previously identified -tubulin–interacting proteins (Dgrips) copurify with Tub23C isolated from tissue culture cells, and together these proteins form a ring-shaped complex that closely resembles the TuRC_37CD when viewed by negative staining electron microscopy. Our studies revealed two unexpected results: 1) Tub23C forms a large complex in early embryos that is distinct from the TuRC, and 2) expression of different Dgrip84 isoforms is developmentally regulated and correlates with expression of different -tubulin isoforms. These findings support the notion that the two Drosophila -tubulins interact with different binding partners and perhaps interact differently with similar binding partners.

What Is the Role of the Embryonic Tub23C Complex?
In embryos, Tub23C distributes in punctate patterns not associated with centrosomes, whereas Tub37CD localizes to centrosomes and spindle poles during early embryonic divisions (Wilson et al., 1997). It has been postulated that only Tub37CD is utilized at centrosomes during embryonic cleavages (Wilson et al., 1997). Consistent with this, female meiosis and nuclear divisions during early embryogenesis specifically require Tub37CD, and Tub37CD mutants are female sterile despite the presence of Tub23C in the embryos (Tavosanis et al., 1997; Tavosanis and Gonzales, 2003). This suggests that the -tubulin isotypes are not functionally redundant during embryogenesis. Our results provide a potential biochemical explanation for these observations. We propose that the embryonic Tub23C complex may represent a type of ‘storage form’ of Tub23C. This is suggested by the observation that Tub23C could not be immunoprecipitated from early embryos, indicating that the epitope recognized by the antibody was not accessible. We speculate that the storage particle might include molecular chaperones of the TriC protein family, which have been shown to interact with and aid in the folding of -tubulin (Melki et al., 1993). We found that the large Tub23C complexes persisted for at least the first 8 h of development, suggesting that the availability of Tub23C is very tightly regulated during early fly development. Full analysis of the embryonic Tub23C complex awaits methods to purify the complex that do not rely on antibody affinity chromatography. How the association of Tub23C with the putative storage particle is regulated and how Tub37CD escapes recruitment to the particle remains a mystery.

What Is the Importance of the Different Isoforms of Dgrip84/GCP2 in Drosophila?
To date, Drosophila Dgrip84 is the only member of the GCP2 protein family that has been reported to exist as more than one splice variant. Database analysis reveals that at least two Drosophila species possess the 74-amino acid N-terminal extension, D. melanogaster and D. pseudoobscura. It is possible that unique structural features of one or the other Dgrip84 isoform facilitate the particularly rapid centrosome assembly of the early Drosophila embryo, as has previously been postulated for Tub37CD (Wilson et al., 1997). Consistent with this hypothesis, Dgrip84 has recently been implicated in centrosome assembly and separation (Colombié et al., 2006). GCP2 (and GCP3) family members have also been implicated in TuRC recruitment to the spindle pole body in S. cerevisiae and to the centrosome in metazoa (Knop and Schiebel, 1997, 1998; Nguyen et al., 1998; Barbosa et al., 2000; Vinh et al., 2002; Takahashi et al., 2002; Zimmerman et al., 2004; Kawaguchi and Zheng, 2004; Vérollet et al., 2006). The precise role of GCP2 family members in TuRC function and recruitment remains to be elucidated.

Our analysis of Dgrip84 isoforms showed that the isoform expressed in embryos corresponds to the shortest splice variant, whereas the isoform expressed in older embryos and S2 cells corresponds to a previously unreported variant of Dgrip84 that has an N-terminal extension but lacks the reported C-terminal insertion. To date, we have been unable to find experimental evidence for the existence of Dgrip84 isoforms that possess the C-terminal insertion (isoforms A and B). It is possible that these isoforms are expressed only at very low levels or that they are expressed only in certain tissues. The roles of the extension or the insertion in Dgrip84 function are not yet clear, but we noted that the developmental switch to the longer isoform of Dgrip84 correlated with the appearance of TuRCs composed of Tub23C. Although thus far this is merely a correlation, it is tempting to speculate that the N-terminal Dgrip84 extension aids in folding or assembly of the somatic TuRC, or is required for incorporation of Tub23C into the TuRC. Experiments to address these questions are currently underway.

Chimeric TuRCs to Fine-Tune Microtubule Organization?
The metazoan TuRC contains 12–15 -tubulin molecules (Zheng et al., 1995; reviewed in Jeng and Stearns, 1999; Wiese and Zheng, 2006), most of which are thought to be arranged in tetrameric subcomplexes (TuSCs) that contain two copies of -tubulin. Having two -tubulin isotypes in the same cell prompts a number of important questions: can both -tubulin isotypes coexist in the same TuSC? Can both isotypes coexist in the same TuRC? How does the cell distinguish between the -tubulin isotypes? Do the -tubulin isotypes perform different functions? How does the cell transition from using primarily one isotype to using the other, and how does the activity of a chimeric TuRC (for which we found evidence by both sucrose gradient analysis and analysis of purified complexes) or TuSC differ from each of the homogenous versions? The results presented here are beginning to provide answers to some of these questions, but gaining a true understanding of how -tubulin complexes that differ only slightly in their composition are perhaps used to fine-tune microtubule organization will be the subject of future studies.

	ACKNOWLEDGMENTS

 
I thank M. Tremaine and P. Dobrzyn for their contributions to the experiments described here, L. O'Brien and L. Liu for critical reading of the manuscript, Y. Zheng for kind gifts of Dgrip antibodies, J. Dubnau (Cold Spring Harbor Laboratory) for the rp49 plasmid, K. Mansoorabadi (UW-Madison) for affinity purifying the Tub37CD antibodies, S. Chitteni-Pattu for the negative staining EM, and A. Steinberg and L. Vanderploeg for help with figure preparation. C.W. is a Scholar of the Sidney Kimmel Foundation for Cancer Research. Work in the Wiese lab was further supported by funds from UW-Madison WISELI, the UW-Madison Graduate School, and the Department of Biochemistry.

	Footnotes

 
This was published online ahead of print in MBC in Press (http://www.molbiolcell.org/cgi/doi/10.1091/mbc.E07-08-0801) on November 14, 2007.

Address correspondence to: Christiane Wiese (wiese{at}biochem.wisc.edu)

Abbreviations used: tub, -tubulin; TuRC, -tubulin ring complex; TuSC, -tubulin small complex; MT, microtubule.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Anders, A., Lourenco, P. C., and Sawin, K. E. (2006). Noncore components of the fission yeast -tubulin complex. Mol. Biol. Cell 17, 5075–5093.[Abstract/Free Full Text]

Andersen, J. S., Wilkinson, C. J., Mayor, T., Mortensen, P., Nigg, E. A., and Mann, M. (2003). Proteomic characterization of the human centrosome by protein correlation profiling. Nature 426, 570–574.[CrossRef][Medline]

Aldaz, H., Rice, L. M., Stearns, T., and Agard, D. A. (2005). Insights into microtubule nucleation from the crystal structure of human -tubulin. Nature 435, 523–527.[CrossRef][Medline]

Barbosa, V., Yamamoto, R. R., Henderson, D. S., and Glover, D. (2000). Mutation of a Drosophila -tubulin ring complex subunit encoded by discs degenerate-4 differentially disrupts centrosomal protein localization. Genes Dev 14, 3126–3139.[Abstract/Free Full Text]

Barbosa, V., Gatt, M., Rebollo, E., Gonzalez, C., and Glover, D. M. (2003). Drosophila dd4 mutants reveal that TuRC is required to maintain juxtaposed half spindles in spermatocytes. J. Cell Sci 116, 929–941.[Abstract/Free Full Text]

Blagden, S. P., and Glover, D. M. (2003). Polar expeditions—provisioning the centrosome for mitosis. Nat. Cell Biol 5, 505–511.[CrossRef][Medline]

Colombie, N., Verollet, C., Sampaio, P., Moisand, A., Sunkel, C. E., Bourbon, H.-M., Wright, M., and Raynaud-Messina, B. (2006). The Drosophila -tubulin small complex subunit Dgrip84 is required for structural and functional integrity of the spindle apparatus. Mol. Biol. Cell 17, 272–282.[Abstract/Free Full Text]

DeZazzo, J., Sandstrom, D., de Belle, S, Velinzon, K., Smith, P., Grady, L., DelVecchio, M., Ramaswami, M., and Tully, T. (2000). nalyot, a mutation of the Drosophila myb-related Adf1 transcription factor, disrupts synapse formation and olfactory memory. Neuron 27, 145–158.[CrossRef][Medline]

Field, C. M., Oegema, K., Zheng, Y., Mitchison, T. J., and Walczak, C. E. (1998). Purification of cytoskeletal proteins using peptide antibodies. Meth. Enzym 298, 525–541.[Medline]

Gould, R. R., and Borisy, G. G. (1977). The pericentriolar material in Chinese hamster ovary cells nucleates microtubule formation. J. Cell Biol 73, 601–615.[Abstract/Free Full Text]

Gunawardane, R. N., Martin, O. C., Cao, K., Zhang, L., Dej, K., Iwamatu, A., and Zheng, Y. (2000). Characterization and reconstitution of Drosophila -tubulin ring complex subunits. J. Cell Biol 151, 1513–1523.[Abstract/Free Full Text]

Gunawardane, R. N., Martin, O. C., and Zheng, Y. (2003). Characterization of a new TuRC subunit with WD repeats. Mol. Biol. Cell 14, 1017–1026.[Abstract/Free Full Text]

Guex, N., and Peitsch, M. C. (1997). SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling. Electrophoresis 18, 2714–2723.[CrossRef][Medline]

Haren, L., Remy, M. H., Bazin, I., Callebaut, I., Wright, M., and Merdes, A. (2006). NEDD1-dependent recruitment of the -tubulin ring complex to the centrosome is necessary for centriole duplication and spindle assembly. J. Cell Biol 172, 505–515.[Abstract/Free Full Text]

Hendrickson, T. W., Yao, J., Bhadury, S., Corbett, A. H., and Joshi, H. C. (2001). Conditional mutations in -tubulin reveal its involvement in chromosome segregation and cytokinesis. Mol. Biol. Cell 12, 2469–2481.[Abstract/Free Full Text]

Horio, T., and Oakley, B. R. (1994). Human -tubulin functions in fission yeast. J. Cell Biol 126, 1465–1473.[Abstract/Free Full Text]

Horio, T., and Oakley, B. R. (2003). Expression of Arabidopsis -tubulin in fission yeast reveals conserved and novel functions of -tubulin. Plant Physiol 33, 1926–1934.

Jeng, R., and Stearns, T. (1999). -Tubulin complexes: size does matter. Trends Cell Biol 9, 339–342.[CrossRef][Medline]

Jung, M. K., Prigozhina, N., Oakley, C. E., Nogales, E., and Oakley, B. R. (2001). Alanine-scanning mutagenesis of Aspergillus -tubulin yields diverse and novel phenotypes. Mol. Biol. Cell 12, 2119–2136.[Abstract/Free Full Text]

Kawaguchi, S., and Zheng, Y. (2004). Characterization of a Drosophila centrosome protein CP309 that shares homology with Kendrin and CG-NAP. Mol. Biol. Cell 15, 37–45.[Abstract/Free Full Text]

Kellogg, D. R., and Alberts, B. M. (1992). Purification of a multiprotein complex containing centrosomal proteins from the Drosophila embryo by chromatography with low-affinity polyclonal antibodies. Mol. Biol. Cell 3, 1–11.[Abstract]

Knop, M., and Schiebel, E. (1997). Spc98p and Spc97p of the yeast gamma-tubulin complex mediate binding to the spindle pole body via their interaction with Spc110p. EMBO J 16, 6985–6995.[CrossRef][Medline]

Llamazares, S., Tavosanis, G., and Gonzalez, C. (1999). Cytological characterization of the mutant phenotypes produced during early embryogenesis by null and loss-of-function alleles of the -Tub37C gene in Drosophila. J. Cell Sci 112, 659–667.[Abstract]

Lüders, J., Patel, U. K., and Stearns, T. (2006). GCP-WD is a -tubulin targeting factor required for centrosomal and chromatin-mediated microtubule nucleation. Nat. Cell Biol 8, 137–147.[CrossRef][Medline]

Martin, O. C., Gunawardane, R. N., Iwamatsu, A., and Zheng, Y. (1998). Xgrip109: a -tubulin-associated protein with an essential role in -tubulin ring complex (TuRC) assembly and centrosome function J. Cell Biol 141, 675–687.[CrossRef]

Melki, R., Vainberg, I. E., Chow, R. L., and Cowan, N. J. (1993). Chaperonin-mediated folding of vertebrate actin-related protein and -tubulin. J. Cell Biol 122, 1301–1310.[Abstract/Free Full Text]

Moritz, M., Braunfeld, M. B., Sedat, J. W., Alberts, B. M., and Agard, D. A. (1995a). Microtubule nucleation by -tubulin-containing rings in the centrosome. Nature 378, 638–640.[CrossRef][Medline]

Moritz, M., Braunnfeld, M. B., Fung, J. C., Sedat, J. W., Alberts, B. M., and Agard, D. A. (1995b). Three-dimensional structural characterization of centrosomes from early Drosophila embryos. J. Cell Biol 130, 1149–1159.[Abstract/Free Full Text]

Moritz, M., Zheng, Y., Alberts, B. M., and Oegema, K. (1998). Recruitment of the -tubulin ring complex to Drosophila salt-stripped centrosome scaffolds. J. Cell Biol 142, 775–786.[Abstract/Free Full Text]

Moritz, M., and Agard, D. A. (2001). -Tubulin complexes and microtubule nucleation. Curr. Opin. Struct.