The Human EMAP-like Protein-70 (ELP70) Is a Microtubule Destabilizer That Localizes to the Mitotic Apparatus
2002; Elsevier BV; Volume: 277; Issue: 2 Linguagem: Inglês
10.1074/jbc.m106628200
ISSN1083-351X
AutoresBernd Eichenmüller, Patrick Everley, Jean Palange, Denise Lepley, Kathy A. Suprenant,
Tópico(s)14-3-3 protein interactions
ResumoIn this report, we show that theechinoderm microtubule (MT)-associated protein (EMAP) and relatedEMAP-like proteins (ELPs) share a similar domain organization with a highly conservedhydrophobic ELP (HELP) domain and a large tryptophan-aspartic acid (WD) repeat domain. To determine the function of mammalian ELPs, we generated antibodies against a 70-kDa human ELP and showed that ELP70 coassembled with MTs in HeLa cell extracts and colocalized with MTs in the mitotic apparatus. To determine whether ELP70 bound to MTs directly, human ELP70 was expressed and purified to homogeneity from baculovirus-infected Sf9 cells. Purified ELP70 bound to purified MTs with a stoichiometry of 0.40 ± 0.04 mol of ELP70/mol of tubulin dimer and with an intrinsic dissociation constant of 0.44 ± 0.13 μm. Using a nucleated assembly assay and video-enhanced differential interference contrast microscopy, we demonstrated that ELP70 reduced seeded nucleation, reduced the growth rate, and promoted MT catastrophes in a concentration-dependent manner. As a result, ELP70-containing MTs were significantly shorter than MTs assembled from tubulin alone. These data indicate that ELP70 is a novel MT destabilizer. A lateral destabilization model is presented to describe ELP70's effects on microtubules. In this report, we show that theechinoderm microtubule (MT)-associated protein (EMAP) and relatedEMAP-like proteins (ELPs) share a similar domain organization with a highly conservedhydrophobic ELP (HELP) domain and a large tryptophan-aspartic acid (WD) repeat domain. To determine the function of mammalian ELPs, we generated antibodies against a 70-kDa human ELP and showed that ELP70 coassembled with MTs in HeLa cell extracts and colocalized with MTs in the mitotic apparatus. To determine whether ELP70 bound to MTs directly, human ELP70 was expressed and purified to homogeneity from baculovirus-infected Sf9 cells. Purified ELP70 bound to purified MTs with a stoichiometry of 0.40 ± 0.04 mol of ELP70/mol of tubulin dimer and with an intrinsic dissociation constant of 0.44 ± 0.13 μm. Using a nucleated assembly assay and video-enhanced differential interference contrast microscopy, we demonstrated that ELP70 reduced seeded nucleation, reduced the growth rate, and promoted MT catastrophes in a concentration-dependent manner. As a result, ELP70-containing MTs were significantly shorter than MTs assembled from tubulin alone. These data indicate that ELP70 is a novel MT destabilizer. A lateral destabilization model is presented to describe ELP70's effects on microtubules. microtubule(s) microtubule-associated protein echinoderm MT-associated protein EMAP-like protein hydrophobic ELP 70-kDa human EMAP-like protein 1,4-piperazinediethanesulfonic acid phosphate-buffered saline restrictedly overexpressed proliferation-associated protein video-enhanced differential interference contrast microscopy tryptophan-aspartic acid Cell motility, cell morphogenesis, spindle formation, and chromosome movements require the dynamic turnover of microtubules (MTs).1 This non-equilibrium behavior is an intrinsic property of MTs driven by the irreversible hydrolysis of tubulin-liganded GTP and the structural polarity of the tubulin heterodimer (1Desai A. Mitchison T.J. Annu. Rev. Cell Dev. Biol. 1997; 13: 83-117Google Scholar, 2Margolis R.L. Wilson L. Bioessays. 1998; 20: 830-836Google Scholar). Individual MTs can switch between extended periods of slow growth and rapid shrinkage. The transitions between these two states, known as catastrophe and rescue, are quite sudden and are believed to be stochastic events (3Walker R.A. O'Brien E.T. Pryer N.K. Soboeiro M.F. Voter W.A. Erickson H.P. Salmon E.D. J. Cell Biol. 1988; 107: 1437-1448Google Scholar, 4Horio T. Hotani H. Nature. 1986; 321: 605-607Google Scholar). This dynamic behavior of MTs is best described by the dynamic instability model and is generally associated with the behavior of MT plus-ends nucleated at the centrosome (5Mitchison T. Kirschner M. Nature. 1984; 312: 237-242Google Scholar, 6Mitchison T. Kirschner M. Nature. 1984; 312: 232-237Google Scholar). MT dynamics in living cells are highly regulated. For example, MTsin vivo polymerize more rapidly and undergo more frequent catastrophe and rescue events than do MTs assembled from pure tubulinin vitro (7Cassimeris L. Cell Motil. Cytoskeleton. 1993; 26: 275-281Google Scholar). These results indicate that there are cellular factors that regulate individual parameters of MT dynamics (1Desai A. Mitchison T.J. Annu. Rev. Cell Dev. Biol. 1997; 13: 83-117Google Scholar). In addition to having cell type or tissue specific factors (8Shelden E. Wadsworth P. J. Cell Biol. 1993; 120: 935-945Google Scholar), the cell must regulate the activity of each factor during cell division (9McNally F.J. Curr. Opin. Cell Biol. 1996; 8: 23-29Google Scholar, 10Andersen S.S. Trends Cell Biol. 2000; 10: 261-267Google Scholar), during cell differentiation (11Bulinski J.C. Gundersen G.G. Bioessays. 1991; 13: 285-293Google Scholar), and within different regions of the cell (12Wadsworth P. Cell Motil. Cytoskeleton. 1999; 42: 48-59Google Scholar). Recently, several proteins that destabilize MTs have been characterized. In yeast, Kar3 is a minus-end-directed kinesin that localizes to spindle poles in vivo and destabilizes MTs at their minus ends in vitro (13Endow S.A. Kang S.J. Satterwhite L.L. Rose M.D. Skeen V.P. Salmon E.D. EMBO J. 1994; 13: 2708-2713Google Scholar, 14Saunders W. Hornack D. Lengyel V. Deng C. J. Cell Biol. 1997; 137: 417-431Google Scholar). Op18/stathmin is a low molecular weight tubulin-binding protein that destabilizes MTs in tissue culture cells (15Marklund U. Larsson N. Gradin H.M. Brattsand G. Gullberg M. EMBO J. 1996; 15: 5290-5298Google Scholar) and in Xenopus egg extracts (16Belmont L.D. Mitchison T.J. Cell. 1996; 84: 623-631Google Scholar). XKCM1, a member of the Xenopus Kin I kinesin family, promotes catastrophes by favoring a conformational change at the MT ends that leads to protofilament peeling and MT shortening (17Walczak C.E. Mitchison T.J. Desai A. Cell. 1996; 84: 37-47Google Scholar, 18Desai A. Verma S. Mitchison T.J. Walczak C.E. Cell. 1999; 96: 69-78Google Scholar). Katanin uses the energy of ATP hydrolysis to sever MTs at the centrosomes (19McNally F.J. Thomas S. Mol. Biol. Cell. 1998; 9: 1847-1861Google Scholar, 20Hartman J.J. Mahr J. McNally K. Okawa K. Iwamatsu A. Thomas S. Cheesman S. Heuser J. Vale R.D. McNally F.J. Cell. 1998; 93: 277-287Google Scholar). It is clear that several proteins with different modis operandi destabilize MTs in the cell. One of the most abundant microtubule-associated proteins (MAPs) in dividing sea urchin embryos is the echinoderm MT-associated protein (EMAP) (21Keller 3rd, T.C. Rebhun L.I. J. Cell Biol. 1982; 93: 788-796Google Scholar, 22Suprenant K.A. Marsh J.C. J. Cell Sci. 1987; 87: 71-84Google Scholar, 23Vallee R.B. Bloom G.S. Proc. Natl. Acad. Sci. U. S. A. 1983; 80: 6259-6263Google Scholar). EMAP is a 75-kDa, WD repeat protein with little sequence similarity to other well-characterized MAP families (24Li Q. Suprenant K.A. J. Biol. Chem. 1994; 269: 31777-31784Google Scholar, 25Suprenant K.A. Tuxhorn J.A. Daggett M.A. Ahrens D.P. Hostetler A. Palange J.M. VanWinkle C.E. Livingston B.T. Dev. Genes Evol. 2000; 210: 2-10Google Scholar). In early embryos, EMAP localizes to the MTs of the first cleavage mitotic apparatus where it is hyperphosphorylated (26Bloom G.S. Luca F.C. Collins C.A. Vallee R.B. Cell Motil. 1985; 5: 431-446Google Scholar, 27Suprenant K.A. Dean K. McKee J. Hake S. J. Cell Sci. 1993; 104: 445-450Google Scholar, 28Brisch E. Daggett M.A. Suprenant K.A. J. Cell Sci. 1996; 109: 2885-2893Google Scholar). Although the function of EMAP in cells is unknown, its evolutionary conservation suggests that it plays an important role in MT assembly and function in a variety of cells (25Suprenant K.A. Tuxhorn J.A. Daggett M.A. Ahrens D.P. Hostetler A. Palange J.M. VanWinkle C.E. Livingston B.T. Dev. Genes Evol. 2000; 210: 2-10Google Scholar). Here, we characterize a 70-kDa human EMAP-like protein (ELP70), show that it is a novel MT destabilizer, and suggest that it is an important regulator of MT dynamics during the cell cycle. HeLa cells (American Type Culture Collection) were cultured at 37 °C in 5% CO2 with Dulbecco's modified Eagle's medium/F-12 Hams' plus 2 mm l-glutamine, 10% fetal bovine serum (Invitrogen), 10,000 units/ml (1670 units/mg) penicillin, and 10 mg/ml streptomycin. Paclitaxel (Taxol) was a gift of the Drug Synthesis and/Chemistry Branch, Developmental Therapeutics Program, Division of Cancer Treatment, NCI, National Institutes of Health. GTP (tetralithium salt) was purchased from Roche Molecular Biochemicals and was stored as a stock solution in distilled water at −80 °C. Monoclonal anti-β-tubulin (TUB 2.1) was from the Sigma Chemical Corp. Cy2- and Cy3-conjugated donkey anti-mouse and rabbit secondary antibodies were from Jackson ImmunoResearch Laboratories, Inc. ELP70 was expressed as a 6His fusion protein in baculovirus-infected Sf9 cells. The full-length ELP70 coding sequence was inserted into theKpnI and NdeI restriction sites of the baculovirus transfer vector pAcHLT-C (BD PharMingen Inc.). Sf9 insect cells were transfected with this construct and the linearized BaculoGold baculovirus DNA. Active viruses generated by homologous recombination (Ac6HisELP70) were isolated as individual plaques that were subsequently amplified and used for infection of Sf9 cells. Sf9 cell lysates were prepared at 72 h post-infection by sonication in 50 mmNa2HPO4/NaH2PO4, 0.5m NaCl, pH 7.5, 200 μm phenylmethylsulfonyl fluoride, 20 μg/ml leupeptin, 10 μg/ml pepstatin A, and 20 μg/ml aprotinin. The lysate was centrifuged (10 min at 39,000 ×g), and 6HisELP70 was purified from the supernatant fluids by cobalt-affinity chromatography (TALON metal affinity resin,CLONTECH Laboratory Inc.) and size exclusion chromatography (BIO-GEL P60, Bio-Rad Inc.). 6His ELP70 was concentrated by vacuum dialysis into 80 mm PIPES/K+, 1 mm MgSO4, 1 mm EGTA, pH 6.8 (BRB80). The final concentration of ELP70 was ∼0.35 g/liter (4.7 μm). Tubulin was purified to homogeneity from bovine brain by three cycles of MT assembly followed by phosphocellulose chromatography (29Tiwari S.C. Suprenant K.A. Anal. Biochem. 1993; 215: 96-103Google Scholar, 30Algaier J. Himes R.H. Biochim. Biophys. Acta. 1988; 954: 235-243Google Scholar). MTs were polymerized at 28 °C by the stepwise addition of paclitaxel to a final concentration of 20 μm. A constant amount of paclitaxel-stabilized MTs (10 μg) was mixed with ELP70 in BRB80 containing 1 mm GTP and 20 μm paclitaxel. After 30 min, MTs were pelleted at 100,000 × g (10 min). Bound and free ELP70 was quantitated from Coomassie Blue-stained gels as described below. Purified ELP70, for which the protein mass had been determined by a BCA assay (see below), was used as an internal standard on the same gel. Control experiments were carried out with identical concentrations of ELP70 in the absence of MTs. Densitometric analysis of the gels was carried out with a Hewlett-Packard ScanJet and quantitated using IMAGE version 1.6.1 (National Institutes of Health, available at rsb.info.nih.gov/nih-image). To obtain the number of ELP70 molecules, the area under each protein peak was converted to total mass units and divided by 70,000, the relative molecular mass of ELP70. The behavior of individual MTs nucleated from Chlamydomonas flagellar axonemes was visualized using video-enhanced differential interference contrast (VE-DIC) microscopy as described previously (31Vasquez R.J. Gard D.L. Cassimeris L. J. Cell Biol. 1994; 127: 985-993Google Scholar). The microscope stage was maintained at 37 °C with an air curtain incubator, and the samples were preincubated for 30 min to reach a steady state prior to data collection. Data were collected during the following 30–45 min. A Nikon Optiphot microscope equipped with rectified DIC optics and high extinction HN-32 polarizers was used to visualize individual MTs. The light beam from a 100-watt mercury lamp passed through a fiber optic light scrambler (Technical Video, Ltd.), a 546-DF 24-nm interference filter and KG-5 heat-absorbing filter, and a 1.4-numerical aperture (NA) condenser, and images were formed with a 60× PlanApo 1.4-NA objective. Background-subtracted and contrast-enhanced images were captured by a Hamamatsu C2400 Newvicon camera and Argus-10 image processor. Super VHS format recordings were digitized with a Video Van Gogh board (Tekmatic Systems Inc.) and analyzed with the Real Time Measurement software (kindly provided by Ted Salmon, University of North Carolina, Chapel Hill, NC). MT growth and shortening rates were analyzed by a Student's t test. For protein expression, an ELP70 cDNA encoding amino acids 31–648 was cloned into theXho/BamHI site of the pET14b vector (Novagen, Inc.) and was transfected into the lysogenic Escherichia coli strain BL21(DE3) plysS (32Studier F.W. Rosenberg A.H. Dunn J.J. Dubendorff J.W. Methods Enzymol. 1990; 185: 60-89Google Scholar). Following isopropyl-1-thio-β-d-galactopyranoside induction, inclusion bodies containing the expressed protein were isolated and fractionated on a preparative SDS-polyacrylamide gel. Rabbit polyclonal antibodies were generated against the electro-eluted ELP70 (Cocalico Biologicals, Inc.). Affinity-purified antibodies were isolated by column chromatography (33Field C.M. Oegema K. Zheng Y. Mitchison T.J. Walczak C.E. Methods Enzymol. 1998; 298: 525-541Google Scholar). Briefly, purified inclusion bodies were solubilized in 0.1m sodium borate, pH 9.0, 0.5 m NaCl, and 0.5% SDS and coupled to CNBr-activated Sepharose 4B (Amersham Biosciences, Inc.) (34Vaisberg E.A. Koonce M.P. McIntosh J.R. J. Cell Biol. 1993; 123: 849-858Google Scholar). Anti-ELP70 antiserum (UK176) was diluted 1:1 in TBS (15 mm NaCl, 20 mm Tris-HCl, pH 7.4), sterile-filtered (0.2 μm), and passed over the column five times. After washing as described previously (33Field C.M. Oegema K. Zheng Y. Mitchison T.J. Walczak C.E. Methods Enzymol. 1998; 298: 525-541Google Scholar), antibodies were eluted with 0.2 m glycine-HCl, pH 2.0, or 6 m guanidine HCl in TBS. Antibody-containing eluates were dialyzed into TBS, concentrated with a Centricon 30 (Amicon Corp.), made 50% in glycerol, and stored at −20 °C. MTs were assembled and purified from 5 g of frozen HeLa cells as described previously (35Vallee R.B. Collins C.A. Methods Enzymol. 1986; 134: 116-127Google Scholar). Briefly, HeLa cells were homogenized in 2 volumes of BRB80 containing a protease inhibitor mixture that contained 158 mg/ml benzamidine, 10 mg/ml leupeptin, 2 mg/ml pepstatin, 1 mg/ml aprotinin, 1 mg/ml antipain, 1 mg/ml chymostatin, and 0.5 mm phenylmethylsulfonyl fluoride. The homogenate was centrifuged at 18,000 rpm (Beckman JA 20 rotor) for 10 min at 2 °C. The pellet was discarded, and the supernatant was centrifuged at 45,000 rpm (Beckman Ti65 rotor) for 90 min at 2 °C. Paclitaxel and GTP were added to the supernatant at a final concentration of 20 μm and 0.5 mm, respectively. The cytosolic extract was briefly warmed to 37 °C (3–5 min) and then chilled for 15 min on ice. The cytosolic extract was transferred to a chilled centrifuge tube and underlayed with ice-cold BRB80 containing 10% w/v sucrose, 20 μmpaclitaxel, and 0.5 mm GTP. MTs were pelleted through the sucrose cushion by centrifugation at 18,000 rpm (Beckman JA20 rotor) for 30 min at 2 °C. MT proteins were analyzed by SDS-PAGE and immunoblotting. HeLa cells were fixed with 3% paraformaldehyde (Electron Microscopy Sciences, Ft. Washington, PA) in PBS (15 min at room temperature). After fixation, cells were rinsed three times (∼5 min each) in PBS and permeabilized with 0.15% (w/v) saponin in PBS (5 min). Fixed and permeabilized cells were incubated in primary antibodies diluted into PBS with 0.15%(w/v) saponin (aldehyde fixation) for 1 h at 37 °C. After washing five times with PBS, the primary antibodies were detected Cy3- or Cy2-conjugated donkey anti-mouse or anti-rabbit secondary antibodies diluted in PBS containing saponin. After 1 h at 37 °C coverslips were rinsed five times in PBS and mounted on microscope slides in a solution comprised of 90% (v/v) glycerol, 10% (v/v) PBS, pH 9.0, and 5% (w/v)n-propyl gallate. For immunofluorescence microscopy, cells were viewed with a Nikon Optiphot microscope, and images were recorded on T-MAX 400 film. Negatives were digitized, and images were contrast-enhanced with Adobe Photoshop (Adobe Photosystems, Inc.). Random-priming of an ELP70 retina cDNA (H92185) generated a [32P]CTP-labeled ELP70 probe that was hybridized to a human cancer cell line Multiple Tissue Northern blots (CLONTECH Laboratories Inc.). Proteins were analyzed on SDS-denaturing polyacrylamide gels stained with Coomassie Blue (36Laemmli U.K. Beguin F. Gujer-Kellenberger G. J. Mol. Biol. 1970; 47: 69-85Google Scholar). Gels were scanned and quantitated by densitometry using the IMAGE software, version 1.61 (rsb.info.nih.gov/nih-image). For standardization, known amounts ELP70 were run on the same gel. Proteins were electrophoretically transferred from SDS-PAGE gels to polyvinylidene difluoride membranes (Millipore, Inc.) (37Towbin H. Staehelin T. Gordon J. Bio/Technology. 1992; 24: 145-149Google Scholar), probed with affinity-purified anti-ELP70 antibodies, and visualized by chemiluminescence with alkaline phosphatase-conjugated secondary antibodies (Tropix Inc.). Protein concentration was determined by the bicinchoninic acid (BCA) assay (Pierce Inc.) using bovine serum albumin as a standard. The echinodermMT-associated protein (EMAP) is a member of a family of EMAP-likeproteins (ELPs) found in many metazoans and at least one protozoan (Fig. 1). In humans, there are three EMAP-like proteins (ELPs): ELP70, ELP79/EMAPL, and ELP120/ropp120 (GenBankTM accession numbers AF103939, U97018, andAAG09279). The gene for ELP79, the first EMAP-like protein (EMAPL) identified in humans, was positionally cloned from the Usher syndrome 1A locus at 14q32 (38Eudy J.D. Ma-Edmonds M. Yao S.F. Talmadge C.B. Kelley P.M. Weston M.D. Kimberling W.J. Sumegi J. Genomics. 1997; 43: 104-106Google Scholar). The role played by ELP79 in Ushers is unknown at this time, but cytoskeletal defects are believed to be responsible for the retinitis pigmentosa, vestibular dysfunction, and deafness, characteristic of this disorder (39Barrong S.D. Chaitin M.H. Fliesler S.J. Possin D.E. Jacobson S.G. Milam A.H. Arch. Ophthalmol. 1992; 110: 706-710Google Scholar, 40Weil D. Blanchard S. Kaplan J. Guilford P. Gibson F. Walsh J. Mburu P. Varela A. Levilliers J. Weston M.D. Nature. 1995; 374: 60-61Google Scholar). ELP70 was identified and cloned by homology to sea urchin EMAP and is the most similar to sea urchin EMAP (41Lepley D.M. Palange J.M. Suprenant K.A. Gene. 1999; 237: 343-349Google Scholar), sharing 57% sequence identity and 77% sequence similarity over the length of the polypeptide. At 981 amino acids in length, ELP120/ropp120 is the largest member of the ELP family in humans. ELP120/ropp120 is less conserved over the full-length of the polypeptide, but only because there are long NH2- and COOH-terminal sequences not found in either ELP79 or ELP70. Interestingly, ELP120/ropp120 was originally identified as therestrictedly overexpressedproliferation-associated protein (ropp), which is dramatically overexpressed during mitosis (42Heidebrecht H.J. Buck F. Pollmann M. Siebert R. Parwaresch R. Genomics. 2000; 68: 348-350Google Scholar). In addition to the human ELPs, complete ELP-coding sequences are predicted for the ciliate Euplotes octocarinatus (AJ251505), the sea urchins, Strongylocentrotus purpuratus (U15551) andLytechinus variegatus (AF136234), the fruit fly,Drosophila melanogaster (CT32830), and the nematode,Caenorhabditis elegans (Z92833). A comparison of eight ELP sequences revealed a highly conserved 43-amino acid domain that begins with an invariant proline residue (Fig. 1, a and b). This hydrophobicELP (HELP) domain is 97% conserved between human and sea urchin ELPs and has not been detected in other known proteins. This high degree of conservation among ELP family members indicates that the HELP domain is likely to be critical to ELP function. For example, a preliminary study has shown that the NH2 terminus of sea urchin EMAP, which contains the HELP domain, is sufficient for MT binding in vitro (43Eichenmüller B. Ahrens D.P. Li Q. Suprenant K.A. Cell Motil. Cytoskeleton. 2001; 50: 161-172Google Scholar). The NH2-terminal region of ELP70 is also very rich in the amino acids proline, serine, and arginine. Although there is no striking sequence homology between this region and the known MT binding domains of other MAPs, serine- and proline-rich domains have been identified in MAP-4 (44Olson K.R. McIntosh J.R. Olmsted J.B. J. Cell Biol. 1995; 130: 639-650Google Scholar). In addition, all members of the ELP family share a conserved SGGG motif (amino acids 310–314, ELP70) perhaps analogous to the PGGG repeat domain of MAP-2, MAP-4, and tau (45Lewis S.A. Wang D.H. Cowan N.J. Science. 1988; 242: 936-939Google Scholar, 46West R.R. Tenbarge K.M. Olmsted J.B. J. Biol. Chem. 1991; 266: 21886-21896Google Scholar). Unlike the latter MAPs, the SGGG motif is only found once in the ELP sequence. Over 75% of the ELP70 polypeptide is comprised of a series of 11 domains that satisfy the regular expression pattern of a WD repeat domain (Fig. 1c). WD repeats are found in a variety of proteins involved with RNA processing, transcriptional regulation, cytoskeleton assembly, and vesicular trafficking (bmerc-www.bu.edu/wdrepeat). The crystal structure of the Gβ subunit of the heterotrimeric G protein, transducin, revealed that a WD repeat adopts a β-propeller fold comprised of a four-stranded anti-parallel β-sheet (47Smith T.F. Gaitatzes C. Saxena K. Neer E.J. Trends Biochem. Sci. 1999; 24: 181-185Google Scholar). By analogy to β-transducin, ELP70 may fold into an 11-bladed propeller structure. ELP70 is expressed in a variety of normal tissues from a single gene located on chromosome 19 (41). To determine if ELP70 was expressed in human cancer cell lines, we probed a Northern blot with a 550-bp probe corresponding to the COOH terminus of ELP70 (Fig. 2). The ELP70 probe recognized a major 2.5-kb band in HeLa (S3), chronic myelogenous leukemia (K-562), lymphoblastic leukemia (MOLT-4), colorectal adenocarcinoma (SW480), lung carcinoma (A549), and melanoma (G361) cells. The ELP70 probe also hybridized to a 3.2-kb major band in promyelocytic leukemia (HL-60) and Burkitt's lymphoma (Raji), indicating that there may be tissue-specific alternative splicing of ELP70 transcripts. The 3.2-kb band was also observed in HeLa, colorectal adenocarcinoma, lung carcinoma, and melanoma cells but at much lower levels (data not shown). These results are interesting in light of the fact that, in normal human tissues, the 3.2-kb band is only detected in brain and spinal cord (41Lepley D.M. Palange J.M. Suprenant K.A. Gene. 1999; 237: 343-349Google Scholar). This raises the possibility that ELP70 RNA processing is altered in tumor cells. Because ELP70 transcripts were abundant in HeLa cells, we used antibodies generated against a bacterially expressed ELP70 fusion protein to determine if ELP70 proteins were associated with HeLa MTs. Affinity-purified ELP70 antibodies recognized a polypeptide with a molecular mass consistent with the conceptually translated ELP70 cDNA (70 kDa) and not the related ELP79 (79 kDa) or ELP120 (120 kDa) cDNAs (Fig. 3). To determine whether HeLa ELP70 was an MT-binding protein, endogenous HeLa MTs were analyzed by SDS-PAGE. The immunoblot shown in Fig. 3shows that the 70-kDa ELP70 polypeptide was greatly enriched in the MT pellet and depleted in the MT-depleted supernatant. These results are consistent with the ELP70 polypeptide being an MT-binding protein in HeLa cells. Because ELP70 is associated with MTs in vitro, the affinity-purified antibodies were used to examine the intracellular distribution of ELP70 in dividing HeLa cells. The images presented in Fig. 4 show that ELP70 was concentrated in the region of the mitotic spindle apparatus. To determine whether ELP70 bound directly to MTs, a 6His-ELP70 fusion protein was purified from baculovirus-infected Sf9 insect cells and incubated with paclitaxel-stabilized MTs (Fig. 5). Purified ELP70 bound directly to purified bovine brain MTs. In similar assays no ELP70 sedimented in the presence of actin filaments, intermediate filaments, or in the absence of MTs. The results of several binding assays with a constant amount of paclitaxel-stabilized MTs indicated that ELP70 bound to MTs in a concentration-dependent manner. From these data, an equilibrium-binding isotherm was obtained (Fig.6). ELP70 binding to bovine brain MTs was saturated at a stoichiometry of 0.40 ± 0.04 mol of ELP70/mol of tubulin dimer and the intrinsic dissociation constant (Kd) for ELP70 binding to MTs was determined to be 0.44 ± 0.13 μm. This level of saturation indicated that ELP70 binds along the length of the MT polymer. In support of this conclusion, we found no evidence that ELP70 bound preferentially to MT ends. The same quantity of ELP70 copelleted with longer MTs as with an equal mass of shorter MTs produced by shearing with a syringe (data not shown). To examine the effect of ELP70 on MT dynamics in vitro, individual MTs nucleated from axonemes were visualized by VE-DIC microscopy. As shown in Fig. 7, seeded nucleation was dependent upon the ELP70 concentration. The number of MTs emanating from the plus-ends of individual axonemes was reduced from an average value of 4.67 ± 1.44 MTs per axoneme (no ELP70) (n = 15, where n = number of axonemes) to 3.43 ± 1.17 MTs per axoneme (0. 5 μmELP70) (n = 21) and 2.23 ± 0.93 MTs per axoneme (0.8 μm ELP70) (n = 13). The reduction in seeded nucleation was statistically significant (p < 0.05) at both concentrations of ELP70. In addition to reducing seeded nucleation, plus-end MT lengths were dramatically shorter (Fig. 8, TableI). In the presence of 0.8 μm ELP70, more than 55% of the MTs were shorter than 3 μm (compared with none without ELP70) and MTs rarely exceeded a length of 6 μm. At these concentrations of ELP70 and tubulin, there were very few minus-end MTs that could be measured. At higher ratios of ELP70 to tubulin, the MTs were too few and too short to gather meaningful data regarding further increases in the frequency of catastrophe or reduction of growth rate.Table IELP70 reduces microtubule elongation velocity and increases catastrophes at the plus-ends of microtubulesPlus-ends13 μm tubulin13 μm tubulin + 0.5 μm ELP7016 μm tubulin16 μm tubulin + 0.5 μm ELP7016 μm tubulin + 0.8 μm ELP70Ve (μm/min)0.78 ± 0.230.47 ± 0.150.94 ± 0.240.71 ± 0.170.64 ± 0.21 n6637587833Vs (μm/min)32.6 ± 8.033.9 ± 12.035.8 ± 10.329.7 ± 9.335.4 ± 11.4 n4627385323fcat (min−1)0.23 ± 0.030.45 ± 0.070.20 ± 0.030.31 ± 0.040.43 ± 0.07 n6640507540fres (min−1)0.31 ± 0.16n.o.0.16 ± 0.110.35 ± 0.180.22 n 4 2 4 1Lavg (μm)3.3 ± 1.11.0 ± 0.44.7 ± 1.22.3 ± 0.51.5 ± 0.5 Open table in a new tab Individual parameters of dynamic instability such as the elongation and shortening rates as well as catastrophe and rescue frequencies were analyzed at video frame rates. As shown in Tables I and II, ELP70 had major effects on the dynamic behavior of MTs. The elongation rate was reduced significantly in the presence of sub-stoichiometric levels of ELP70. For example, the addition of 0.5 μm ELP70 to 16 μm tubulin significantly reduced the elongation rate from 0.94 ± 0.24 μm/min to 0.71 ± 0.17 μm/min (p < 0.001). Increasing the concentration of ELP70 from 0.5 to 0.8 μm resulted in an additional significant decrease in the elongation rate from 0.71 ± 0.17 to 0.64 ± 0.21 μm/min (p < 0.05). The ELP70-dependent reduction in the elongation rate was concentration-dependent and was more apparent at a tubulin concentration of 13 μm, where the addition of 0.5 μm ELP70 decreased the elongation rate from 0.78 ± 0.23 to 0.47 ± 0.15 μm/min (p < 0.001). In contrast to the effects on elongation, ELP70 appeared to have little effect on the velocity of shortening.Table IIParameters for microtubule dynamics at the minus-ends of microtubulesMinus-ends16 μm tubulin16 μm tubulin + 0.5 μm ELP70Ve (μm/min)0.19 ± 0.070.21 ± 0.10 n2727Vs (μm/min)44.3 ± 19.745.3 ± 19.8 n1914fcat (min−1)0.17 ± 0.040.26 ± 0.05 n2224fres (min−1)10.2 ± 3.14.9 ± 1.9 n11 7Lavg (μm)1.1 ± 0.40.8 ± 0.4 Open table in a new tab These ELP70-dependent effects on the elongation rate were specific to the plus-ends of the MTs. Minus-end MTs grew at the same rate in the presence and absence of ELP70 (TableII). ELP70 also increased the transition from a growing MT to shortening MT (catastrophe frequency), in a concentration-dependent manner. For example, in the presence of 0.8 μm ELP70 and 16 μm tubulin, plus-end catastrophes were observed every 140 s (40 catastrophes in 5581 s of elongation) compared with one catastrophe every 300 s for 16 μm tubulin alone (50 catastrophes in 15,000 s of elongation). This is a 2.2-fold increase in the catastrophe frequency. An increase in the catastrophe frequency was observed for the minus-end MTs as well. Although the frequency of catastrophe increased at both the plus- and minus-ends of MTs in the presence of ELP70, there was no apparent effect on the shortening rates at either the plus- or minus-ends. In addition, there was no apparent effect on the transition from a shortening MT to a growing MT (rescue frequency) at the plus-end. Generally, plus-end MTs, polymerized at these tubulin concentrations, shortened all the way back to the nucleation site, and rescue events were observed very infrequen
Referência(s)