Transforming Acidic Coiled-coil-containing Protein 4 Interacts withCentrosomal AKAP350 and the Mitotic Spindle Apparatus*

Received for publication, February 26, 2002, and in revised form, May 2, 2002Published, JBC Papers in Press, May 15, 2002, DOI 10.1074/jbc.M201914200

Brent T. Steadman, P. Henry Schmidt, Ryan A. Shanks, Lynne A. Lapierre,and James R. Goldenring‡

From the Departments of Medicine, Surgery and Cellular Biology and Anatomy, Institute of Molecular Medicine andGenetics, Medical College of Georgia and the Augusta Veterans Affairs Medical Center, Augusta, Georgia 30912

AKAP350 is a multiply spliced family of 350–450-kDaprotein kinase A-anchoring proteins localized to thecentrosomes and the Golgi apparatus. Using AKAP350Aas bait in a yeast two-hybrid screen of a rabbit parietalcell library, we have identified a novel AKAP350-inter-acting protein, transforming acidic coiled-coil-contain-ing protein 4 (TACC4). Two-hybrid binary assays dem-onstrate interaction of both TACC3 and TACC4 withAKAP350A and AKAP350B. Antibodies raised to aTACC4-specific peptide sequence colocalize TACC4 withAKAP350 at the centrosome in interphase Jurkat cells.Mitotic cell staining reveals translocation of TACC4from the centrosome to the spindle apparatus with themajority of TACC4 at the spindle poles. TruncatedTACC4 proteins lacking the AKAP350 minimal bindingdomain found in the carboxyl coiled-coil region ofTACC4 could no longer target to the centrosome. Amino-truncated TACC4 proteins could no longer target to thespindle apparatus. Further, overexpression of TACC4fusion proteins that retained spindle localization in mi-totic cells resulted in an increased proportion of cellspresent in prometaphase. We propose that AKAP350 isresponsible for sequestration of TACC4 to the centro-some in interphase, whereas a separate TACC4 domainresults in functional localization of TACC4 to the spin-dle apparatus in mitotic cells.

Rapid transmission of membrane-initiated signals to appro-priate intracellular targets is fostered by signal compartmen-talization. Intracellular compartmentalization allows an ini-tial, general second messenger to achieve greater signalspecificity. By localizing type II cAMP-dependent protein ki-nase (protein kinase A) to a particular organelle or cytoskeletalelement, protein kinase A-anchoring proteins (AKAPs)1 effec-tively pre-position the inactive enzyme at or near its eventualsubstrate (1). Because there are a number of cAMP-dependent

signaling pathways operating simultaneously in the cell,AKAPs facilitate the stimulation of a single unique target atthe appropriate time and place (2). Most AKAPs have an am-phipathic �-helical region that binds the regulatory subunits ofthe protein kinase A heterotetramer (3). Other domains inAKAPs provide targeting motifs to enable binding of an indi-vidual AKAP to its particular intracellular compartment (4)and sequence motifs for the scaffolding of other regulatoryproteins. Indeed, many AKAPs also scaffold multiple enzymesthat are typically protein kinases, phosphatases, or other sec-ond messenger-dependent mediators (5). This complex patternof protein assembly suggests that protein kinase A-anchoringproteins have the ability to coordinate multiple intracellularsignaling events.

Several groups have shown evidence for the functional pres-ence of AKAP scaffolded intracellular signaling complexes.AKAP75, isolated from bovine brain, and its human homo-logue, AKAP79, localize protein kinase A, protein kinase C, andprotein phosphatase 2b to post-synaptic densities (6, 7). Thespecific role for each of these enzymes in this environmentremains to be identified, but AKAP79 could functionally orga-nize these multiple effectors in a strategic location for trans-mission and modulation of post-synaptic signals. Similarly,AKAP250, also known as gravin, localizes both protein kinaseA and protein kinase C to the membrane cytoskeleton andfilopodia found in human erythroleukemia cells (8). Its sub-membrane locale places it adjacent to transmembrane recep-tors, which are likely regulated in part by protein kinase A andprotein kinase C. Investigations of the Yotiao scaffolding pro-tein have revealed its essential role in tethering both proteinkinase A and the type 1 protein phosphatase to synaptic NMDAreceptors (9). Protein phosphatase 1 suppresses and proteinkinase A activates NMDA receptor up-regulation, thus definingthe role of Yotiao as a signaling system scaffold. Most recently,muscle-selective AKAP has been implicated in bidirectionalcontrol of cardiomyocyte protein kinase A activity by anchorageof both protein kinase A and cAMP-specific phosphodiesterase,PDE4D3 (10).

In recent years, our group and others have identified anprotein kinase A-anchoring protein, AKAP350, which localizesto the centrosome and the Golgi apparatus (11–15). AKAP350is the product of a multiply spliced gene from human chromo-some 7q21 that generates numerous isoforms of a large proteinscaffold with both centrosomal and non-centrosomal variants.Subsequent to our publication of AKAP350, Ono et al. (16) andOrstavik et al. (12) identified the same open reading framecDNA as CG-NAP and AKAP450, respectively. In addition to

* This work was supported by National Institutes of Health NIDDKGrants DK48370 and DK43405 and a Veterans Administration meritaward (to J. R. G.) and by a National Institutes of Health IndividualNational Research Service Award Postdoctoral Fellowship F32GM19731 (both to P. H. S.). The costs of publication of this article weredefrayed in part by the payment of page charges. This article musttherefore be hereby marked “advertisement” in accordance with 18U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in this paper has been submittedto the GenBankTM/EBI Data Bank with accession number(s) AF247727and AF372837.

‡ To whom correspondence should be addressed. Current address:James R. Goldenring, Department of Surgery, CC-2306 MCN, 1161 21stAve. S., Vanderbilt University, Nashville, TN 37232-2733. Tel.: 615-322-2096; Fax: 615-343-1355; E-mail: jim.goldenring@vanderbilt.edu.

1 The abbreviations used are: AKAP, protein kinase A-anchoringprotein; TACC, transforming acidic coiled-coil-containing protein; GFP,

green fluorescent protein; RACE, rapid amplification of cDNA ends;GST, glutathione S-transferase; PBS, phosphate-buffered saline; DAPI,4,6-diamidino-2-phenylindole.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 277, No. 33, Issue of August 16, pp. 30165–30176, 2002Printed in U.S.A.

This paper is available on line at http://www.jbc.org 30165

by guest on March 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

binding type II protein kinase A regulatory subunits, initialstudies of AKAP350/AKAP450/CG-NAP show that it interactswith the serine/threonine kinase protein kinase N, proteinphosphatase 1, and protein phosphatase 2a (13). Further stud-ies have demonstrated that CG-NAP associates with proteinkinase C� in a phosphorylation-dependent manner (16) andthat AKAP450 associates with phosphodiesterase 4D3 (17).

Two carboxyl-terminal splice variants from human gastricmucosa (AKAP350A) and human lung (AKAP350B) have beenidentified (11). AKAP450 and CG-NAP contain carboxyl ter-mini identical to AKAP350A. AKAP350C is a separate read-through termination that truncates the last 12 exons ofAKAP350A and results in a different carboxyl-terminal aminoacid sequence (GenBank accession no. AF247727). The NMDAreceptor-associated AKAP, Yotiao, is the shortest 3� splice var-iant derived from this locus (11, 18). All together, two 5� initi-ation sites, two internal splice sites, and four 3� splice variantproducts of AKAP350/AKAP450/CG-NAP have been reported.For the sake of simplicity, we will refer to this protein familythroughout as AKAP350.

Scaffolding proteins could provide a means to organize mul-tiple activities at discrete sites on the centrosome. A recentstudy of pericentrin, a known integral matrix centrosomal pro-tein (19), showed that it functions as an protein kinase A-an-choring protein, which binds cAMP-dependent protein kinase Aregulatory subunits through a novel motif (20). The functionalsignificance of the protein kinase A scaffolding at the pericen-trin centrosome is unknown, but it appears to be important inspindle formation. Spindle defects have been observed whenprotein kinase A anchoring of pericentrin is disrupted (21).Pericentrin interacts with the motor protein dynein (22) and isa substrate for protein kinase A (23). Overexpression of peri-centrin results in dynein depletion and a number of spindledefects (22). The apparent scaffolding function of pericentrinmay play an important role in the complex function of nucle-ating microtubule polymers for spindle formation in thecentrosome.

In this report, we have utilized a carboxyl-terminal fragmentof AKAP350A as bait in a yeast two-hybrid screen of a rabbitparietal cell library. Isolation and subsequent cloning haveidentified a novel AKAP350-interacting protein that is a mem-ber of the transforming acidic coiled-coil-containing (TACC)protein family, TACC4. TACC4 is the first identified AKAP350-interacting protein that is not a signal transduction enzyme.Immunolocalization with polyclonal anti-TACC4 antibodies co-localizes TACC4 with AKAP350 at the centrosome in inter-phase Jurkat cells. Once cells begin mitosis, TACC4 translo-cates to the spindle apparatus accumulating at the spindlepoles, whereas AKAP350 remains at the centrosome. Yeasttwo-hybrid binary assays and truncated GFP-TACC4 expres-sion studies identify amino acids 247–404 of TACC4 as theregion responsible for AKAP350 interaction and correspondingcentrosome localization. Further, a separate amino-terminalTACC4 region from amino acid 1 to 380 is responsible forspindle localization. These results indicate that TACC4 is se-questered to the centrosome during interphase by AKAP350interaction. This interaction is lost in mitosis, and TACC4translocates to the spindle apparatus, where it may have afunctional role in spindle dynamics.

EXPERIMENTAL PROCEDURES

Materials—pEGFP-C2 vector, Advantage Taq, and Marathon clon-ing kits were purchased from CLONTECH. All DNA sequencing wasperformed using dye terminator chemistry automated sequencing inthe Molecular Biology Core Facility at the Medical College of Georgia.The Molecular Biology Core Facility also synthesized oligonucleotides.A rabbit polyclonal antibody (NE27) was raised against a 16-mer pep-tide sequence corresponding to amino acids 99–113 of TACC4 (NH2-

DEQGTRESPSTPTPRC-COOH) by New England Peptide. Mousemonoclonal antibodies to �-tubulin and �-tubulin were purchased fromSigma. Mouse GAL4 activation domain and binding domain monoclonalantibodies were purchased from CLONTECH. Mouse monoclonal anti-AKAP350 (14G2) was produced as previously described (11). For pro-duction of AKAP350A-specific antibodies, a peptide corresponding tothe unique region in the carboxyl terminus of AKAP350A was conju-gated to keyhole limpet hemocyanin using an Imject Immunogen EDCkit from Pierce. The AKAP350A peptide (GSTTQFHAGMRR) was syn-thesized in the Molecular Biology Core Facility. Rabbit polyclonalantibodies were raised in the Antibody Facility at the University ofGeorgia. The antibody was affinity-purified using Amino-Link systemfrom Pierce with the AKAP350A peptide. 100 mM glycine, pH 2.5, wasused to elute the AKAP350A antibody from the Amino-Link column.Species-specific Cy2-, Cy3-, and Cy5-conjugated secondary antibodieswere purchased from Jackson Immunoresearch Laboratories (WestGrove, PA). Prolong Antifade, 4�,6-diamidino-2-phenylindole, andspecies-specific Alexa 488-conjugated secondary antibodies were pur-chased from Molecular Probes, Inc. (Eugene, OR). Fail-Safe amplifica-tion kits were purchased from Epicentre. [�-32P]dCTP was purchasedfrom PerkinElmer Life Sciences. Protease inhibitor mixture and glassbeads were purchased from Sigma.

Yeast Two-hybrid Screening—A rabbit parietal cell pADGal4 two-hybrid library was screened with the final 2.7 kb of human AKAP350AcDNA in the binding domain vector pBDGal4-Cam (Stratagene). TheY190 yeast strain harboring the HIS3 and �-galactosidase reportergenes was used for screening of approximately one million clones aspreviously described (24). A single positive clone was identified andrescued into XL1-Blue bacteria with selection using ampicillin. PlasmidDNA was prepared using Qiagen miniprep kits, and the clone wassequenced using flanking vector primers and specific derived oligonu-cleotide sequences by the Medical College of Georgia Molecular BiologyCore Sequencing Facility.

5�-RACE—The 5� end of the isolated clone was highly GC-rich, andresolution of the remaining 5� sequence required use of both hightemperature cDNA production with rTth polymerase and optimizednested amplification from rabbit spleen cDNA using the Fail-Safe am-plification buffer system (Epicentre). Rabbit spleen total RNA wasprepared as previously described (25), and cDNA was prepared usingrTth polymerase (PerkinElmer Life Sciences) in the presence of man-ganese ions with oligo(dT) priming. The resulting cDNA was thenmodified by addition of linkers to prepare a linkered cDNA (Marathon,CLONTECH). Amplification from a rabbit spleen cDNA required Fail-Safe Buffer E (Epicentre) in two rounds of nested amplification. Thefirst round was performed with Adapter Primer 1 (CLONTECH) andCTCCAGGATACACAGGGGTT (5 cycles of 94 °C for 10 s, 72 °C for 3min; 5 cycles of 94 °C for 10 s, 70 °C for 3min; and 30 cycles of 94 °C for10 s, 68 °C for 3 min). This was followed by reamplification withAdapter Primer 2 and CCCGAACTGCTCCAGGTAATCGATCTC (35cycles of 94 °C for 10 s, 60 °C for 10 s, 68 °C for 2 min). The resultingproducts ranging in size from 500 to 1000 nucleotides were gel-isolatedand cloned into pTOPO-T (Invitrogen). Isolated plasmid miniprepswere sequenced as above.

Northern Blot Analysis—Northern blots containing 20 �g of totalRNA from rabbit tissues were hybridized with a random-primed TACC4cDNA probe (nucleotides 372–765). The probe had been amplified fromplasmid using polymerase chain reaction with Advantage Taq polym-erase (CLONTECH) (T400 sense, GGCCGAGACCCCAGGACCAG-GAGAC; T400 antisense, GGGGGCCCGACGCGCTCACAGG; 35 cyclesof 95 °C for 30 s, 60 °C for 30 s, and 68 °C for 60 s). Blots were washedto high stringency (0.1� SSC, 65 °C) and exposed to x-ray film (East-man Kodak Co.).

Yeast Two-hybrid Binary Assays—Carboxyl-terminal cDNAs forAKAP350B and AKAP350C of similar size to pBD-AKAP350A werecloned into pBD-Gal4. Truncations of AKAP350 including regions com-mon to AKAP350A and AKAP350B were subcloned into pBD-Gal4.Full-length TACC4 cDNA was subcloned into pAD-Gal4. TruncatedTACC4 cDNAs were amplified from full-length TACC4 cDNA and sub-cloned into pAD-Gal4. Assays were performed by dual transfection ofthe Y190 yeast strain with a binding domain pBD-Gal4 construct andan activation domain pAD-Gal4 construct followed by detection of �-ga-lactosidase production as described previously (26). Positive interac-tions were defined by identification of blue colonies within 2 h ofadministration of 5-bromo-4-chloro-3-indolyl-�-D-galactopyranoside(X-gal).

Yeast Western Blot Analysis—Yeast two-hybrid binary assays thathad negative �-galactosidase production were checked for expression ofpAD and pBD by Western blot analysis. Cotransformed yeast colonies

TACC4 Interaction with AKAP35030166

by guest on March 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

were inoculated into 50 ml of SD/�Trp/�Leu and grown overnight at30 °C. 10 ml of the overnight culture were inoculated into 50 ml of YPDand grown to an A600 � 0.5. Cultures were pelleted at 3000 � g for 10min, resuspended in 50 ml of sterile water, and pelleted again. Thepellets were quickly frozen in liquid nitrogen and kept at �70 °C.Pellets were thawed and put in lysis buffer (50 mM EDTA, 1 M NaCl, 1mM EGTA, 0.1% Triton X-100, 1 �l/ml protease inhibitor mixture).Glass beads were added to the lysis solution, and the yeast werevigorously vortexed for 10 min. Lysis solution was pulled off the top ofthe beads and spun down at 4 °C for 10 min at 20,000 � g. Supernatantwas discarded. Pellets were dissolved in 1� SDS, 10% �-mercaptoeth-anol, boiled for 15 min, and run out on a 10% SDS-PAGE gel. Gels weretransferred to Immobilon for subsequent Western blotting by anti-pAD-Gal4 and anti-pBD-Gal4. Blots were blocked with 5% nonfat dry milk in25 mM Tris-HCl, pH 7.5, 150 mM NaCl for 1 h at 25 °C. Blots were thenprobed in 0.5% nonfat dry milk in 25 mM Tris-HCl, pH 7.5, 150 mM NaClfor 2 h at 25 °C with a monoclonal antibody against pAD-Gal4 (0.1�g/ml) or pBD-Gal4 (0.1 �g/ml). After the primary incubation, the blotswere washed three times for 15 min each with 25 mM Tris-HCl, pH 7.5,150 mM NaCl and then incubated with horseradish peroxidase-conju-gated anti-mouse IgG (1:2500) for 1 h at 25 °C. The blots were washedthree times for 5 min each with 25 mM Tris-HCl, pH 7.5, 150 mM NaClfollowed by a 1-min incubation with chemiluminescence substrate(Pierce, Supersignal) and autoradiography.

Western Blot Analysis—Jurkat cell lysate was prepared by lysis of50 � 106 cells with 1� SDS, 10% �-mercaptoethanol or 1� SDS, 3 M

urea, and 10% �-mercaptoethanol. 75 �g of Jurkat cell lysate wasresolved on 10% SDS-PAGE gels and transferred to Immobilon forsubsequent Western blotting by anti-TACC4 (NE27). Blots wereblocked with 5% nonfat dry milk in 25 mM Tris-HCl, pH 7.5, 150 mM

NaCl for 1 h at 4 °C. Blots were then probed in 2.5% nonfat dry milk in25 mM Tris-HCl, pH 7.5, 150 mM NaCl for 2 h at 25 °C with a mono-clonal antibody against AKAP350 (14G2; 1:500) or with a polyclonalantibody against TACC4 (NE27; 1:3000). TACC4 peptide competitionassays were performed by adding 2.5 �M peptide (NH2-DEQGTRESP-STPTPRC-COOH) to the primary antibody incubation. After the pri-mary incubation, the blots were washed three times for 15 min eachwith 25 mM Tris-HCl, pH 7.5, 150 mM NaCl and then incubated withhorseradish peroxidase-conjugated anti-mouse IgG (1:2500) or anti-rabbit IgG (1:25,000) for 1 h at 25 °C. The blots were washed threetimes for 5 min each with 25 mM Tris-HCl, pH 7.5, 150 mM NaClfollowed by 1 min of incubation with chemiluminescence substrate(Supersignal, Pierce) and autoradiography.

Immunocytochemistry of NE27—Jurkat cells were pelleted onto glassslides using a Cyto-Spin (Clay-Adams). Cells were fixed in 4%paraformaldehyde for 10 min at 25 °C. Fixed cells were washed in PBS,then blocked and permeabilized with 1% milk, 0.3% Triton X-100 inPBS. Immunostaining was subsequently performed with various com-binations of anti-TACC4 (1:250), anti-AKAP350 14G2 (1:80), anti-�-tubulin (1:500), and anti-�-tubulin (1:50) for 2 h at 25 °C. The cells werethen incubated with Cy3-conjugated anti-mouse IgG and Alexa 488-conjugated anti-rabbit IgG for 60 min. Following three 5-min washes inPBS, the cells were incubated in DAPI (1 mM) for 5 min and given a final5-min wash in 50 mM sodium phosphate, pH 7.4. Slides were mountedwith Prolong Antifade (Molecular Probes, Eugene, OR). Cells wereexamined in the Imaging Core facility of the Institute of MolecularMedicine and Genetics Institute at the Medical College of Georgia on aZeiss Axiophot microscope equipped with a SPOT digitizing camera oran Amersham Biosciences confocal microscope as described previously(27). Anti-TACC4 (NE27) specificity was confirmed by competitive in-hibition of immunostaining with 500 �M TACC4 epitope peptide.

Cloning of 3� Terminal Sequences from AKAP350 Splice Variants—The 3� splice variant sequences, AKAP350A (originally reported asHGAKAP350) and AKAP350B (originally reported as HLAKAP350),were cloned previously (11). Using the Marathon RACE system, alinkered human lung cDNA (CLONTECH) was utilized to perform3�-RACE using a sense primer ATCAATACAATCTCATCTCTAAAGcorresponding to a start at nucleotide 8197 of AKAP350. The resultingRACE products were cloned into pBluescript-T, as previously described(25), and selected clones were sequenced. We identified several se-quences corresponding to the 3� regions of both AKAP350A andAKAP350B. In addition, we identified a novel sequence for a third splicevariant, AKAP350C, which is accounted for by a read-through termi-nation of the final 12 exons present in AKAP350A with a short regionof unique 3� coding sequence (GenBank accession no. AF247727).

Preparation of Recombinant Protein—TACC4 cDNA was cloned intothe pGEX5–1 vector (Amersham Biosciences) and transformed intoJM109, resulting in a vector coding for GST fused to the amino termi-

nus of full-length TACC4. A 1-liter culture of pGEX-TACC4 or emptypGEX5–1 vector was induced with 1 mM isopropyl-1-thio-�-D-galacto-pyranoside in log phase and allowed to grow for 4 h at 37 °C. Bacteriawere collected with a 5000 � g spin for 15 min, and the pellet wasresuspended in 25 ml of lysis buffer (50 mM Tris-HCl pH 8, 10 mM

EDTA, 100 mM NaCl, 0.1 mg/ml lysozyme, 5 mM benzamidine, and 0.1mM AEBSF). The lysate was sonicated on ice with four 10-s bursts.Triton X-100 was then added to a final concentration of 1%, and themixture was nutated at 4 °C for 30 min. The lysate was then centri-fuged at 20,000 � g, and the supernatant was filtered through a0.45-�m filter. This filtrate was then added to 2 ml of glutathione-Sepharose beads (Amersham Biosciences) prepared by washing twotimes in 10 volumes of PBS, one time in PBS plus 1% Triton X-100, andthree times in lysis buffer without Triton X-100. The beads and super-natant mixture were mixed on a Nutator (Clay-Adams) at 25 °C for 30min. Nonadherent protein was removed with a 5-min 500 � g centrif-ugation, and the beads were washed three times with 10 bead volumesof PBS followed by one wash in PBS with 250 mM NaCl. Beads werethen transferred to a 1.5-ml centrifuge tube, and protein was elutedwith 1 bead volume of elution buffer (10 mM reduced glutathione, 50 mM

Tris-HCl, pH 8) for 10 min at 25 °C.In Vitro Binding Assay—One hundred �l of GST-TACC4-(1–454)-

bound glutathione beads (prepared as above without the elution step),GST-bound glutathione beads, or unbound glutathione beads wereblocked in PBS, 2 mg/ml bovine serum albumin at 4 °C for 2 h. 1 ml ofrabbit stomach mucosa 100,000 � g supernatant was then added toeach aliquot of beads, and the mixtures were incubated on a Nutator at4 °C for 4 h. Nonadherent proteins were collected by centrifugation at100 � g, placed in 1% SDS stop solution, and heated at 95 °C for 5 minwith final addition of 0.1 final volume of �-mercaptoethanol. Beadswere washed three times with 1 ml of PBS and then incubated in 1.5%SDS stop solution at 95 °C for 15 min. The beads were pelleted at 500 �g for 5 min, and 0.1 volume of 2-mercaptoethanol was added to thesupernatants. Samples were resolved by SDS-PAGE (3–10% gradientgels) and transferred for 2 h at 750 mA to nitrocellulose (Sarsedt) forsubsequent Western blotting with anti-AKAP350 (14G2 1:500).

Expression of GFP-TACC4 Chimeric Proteins—TACC4 cDNA wascloned into pEGFP-C2 (CLONTECH) resulting in a vector coding forGFP fused to the amino terminus of full-length TACC4. The AKAP350binding domain of TACC4 (amino acids 247–404) was amplified by PCR(sense primer, GGGCCCATCGTGGACG; antisense primer, CTTGCT-GCGGACCTGGGCGAT) and cloned into pEGFP-C2. The TACC4 bind-ing domain of AKAP350 (amino acids 3376–3531) was amplified byPCR (sense primer, CGGATGGGGGGGCAG; antisense primer,TTATCTTCTCATGCCAGCATG) and cloned into pEGFP-C2. Jurkatcells were transfected with the GFP chimeras by electroporation. 4 �105 cells were incubated with 20 �g of GFP chimera plasmid DNA. A500-�l sample in a 0.4-cm electrode gap cuvette was pulsed at 900microfarads and 270 mV. Cells were returned to fresh media. After 30 hin suspension culture, cells were pelleted onto glass slides using aCyto-Spin (Clay-Adams). Cells were fixed, blocked, permeabilized, andstained as above with various combinations of anti-TACC4 NE27 (1:250), anti-AKAP350 14G2 (1:80), anti-�-tubulin (1:500), and anti-�-tubulin (1:50).

GFP-TACC4 Cell Quantitation—The cell cycle stage of �100 cells forthree separate transfections of GFP-TACC4, GFP-TACC4-(1–380),GFP-TACC4-(247–404), and GFP was quantitated. The number of pro-metaphase and interphase cells among transfected cells was comparedby analysis of variance with post hoc comparison of significant means byTukey’s test.

RESULTS

Yeast Two-hybrid Screen and Cloning of TACC4—Utilizingthe last 2.7 kb of human AKAP350A coding sequence as bait,yeast two-hybrid screening of a rabbit parietal cell libraryyielded a single partial clone of 1259 base pairs including a 3�poly(A) tail and an incomplete 5� end. A BLAST search of publicdata bases with this novel sequence revealed significant iden-tity with members of the TACC protein family including mu-rine TACC3, human TACC3, human TACC2, and humanTACC1. As a result, we termed the AKAP350-interacting cloneTACC4. Completion of cloning of the 5� end of TACC4 identifieda 1532-nucleotide cDNA sequence containing an upstream in-frame termination codon along with a polyadenylation signaldownstream of the termination codon (Fig. 1; GenBank acces-

TACC4 Interaction with AKAP350 30167

by guest on March 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

sion no. AF372837). The 1364-nucleotide open reading framecoded for a 454-amino acid protein with a predicted 49.2-kDamolecular mass and an acidic isoelectric point of 4.6. Structuralpredictions indicated proline rich regions in the first 255 aminoacids and a coiled-coil motif encompassing the carboxyl-termi-nal 200 amino acids, both of which are characteristics of otherTACC family members.

The recently identified TACC protein family includes humanTACC3 (28), murine TACC3 (28), human TACC2 (28)/AZU-1(29)/ECTACC (30), human TACC1 (31), Drosophila D-TACC(32), and murine AINT (33). The protein members all have aconserved coiled-coil TACC domain (34), an acidic isoelectricpoint, and proline-rich sequence outside the TACC domain.Alignment of TACC4 with the TACC members of highest ho-mology, murine TACC3 and the human TACC members, re-vealed significant identity with all members in the carboxylTACC coiled-coil domain and less identity in the remainingportion of the protein (Fig. 2). TACC4 contained the greatesttotal similarity to human TACC3 (38%) and murine TACC3(48%) with less similarity to human TACC2 (21%) and humanTACC1 (24%). The coiled-coil TACC domain contained thehighest identity for all family members ranging from 41% forhuman TACC2 to 81% for human TACC3.

A multiple sequence alignment of TACC4, human TACC3,and murine TACC3 is shown in Fig. 3. The major homology isfound within the carboxyl-terminal portion of the TACC4 se-quence. As previously described, human TACC3 apparentlyhas a single exon insertion of 996 nucleotides that is not foundin murine TACC3 (28). The corresponding region appears to bemissing, for the most part, from TACC4 as well. In addition, the

coding sequence of TACC4 does not contain the amino-terminal103 amino acids shared between murine and human TACC3proteins (Fig. 2, solid box).

Northern Blot Analysis of TACC4—A Northern blot of rabbittissue total RNA was probed with a cDNA sequence specific forTACC4 (nucleotides 372–765). Fig. 4 demonstrates a single1.6-kb RNA species. This message size correlates well with the1532-nucleotide cloned TACC4 sequence. The message wasenriched in spleen, jejunum, and duodenum. It was also detect-able in distal colon, ileum, and pancreas.

Yeast Two-hybrid Mapping of the AKAP350 Binding Domainof TACC4—Binary assays were performed with several pAD-TACC4 truncations against the original pBD-AKAP350A con-struct utilized in the yeast two-hybrid screen (Fig. 5). Expres-sion of appropriately sized pAD fusion protein was confirmedby Western blot in all negative �-galactosidase production as-says. Deletion of the carboxyl-terminal 50 amino acids ofTACC4 had no effect on AKAP350A interaction. However, fur-ther deletion of 24 amino acids abolished this interaction. Al-though truncation of the 247 amino-terminal residues did notalter AKAP350A interaction, deletion of amino acids 247–284abolished interaction. Further truncations of pAD-TACC4 wereconstructed to identify the minimal AKAP350 binding domainof TACC4. pAD-TACC4-(247–380) was not able to interact withAKAP350. However, amino acids 247–404 maintained interac-tion with AKAP350, identifying the minimal AKAP350 bindingdomain of TACC4. The coiled-coil domain of TACC4 extendsfrom amino acid 255 to 454. Thus, the binary assay data sug-gest that two regions of TACC4 at the beginning and end of thecoiled-coil domain are important for interaction with AKAP350(Fig. 5, solid boxes).

Given the importance of the conserved coiled-coil TACC do-main to AKAP350A/TACC4 interaction, we evaluated whetherother TACC family members with similar coiled-coil TACCdomains could interact with AKAP350A. Truncated amino-terminal forms of each human TACC gene were cloned into thepAD-Gal4 activation domain and assayed for interaction withpBD-AKAP350A (Fig. 5). Two amino-terminal truncated frag-ments of human TACC3 both interacted with AKAP350A,whereas fragments of TACC1 and TACC2 did not interact.These results are apparently because of the high homologybetween the coiled-coil domains of human TACC3 and TACC4.

Yeast Two-hybrid Binary Assays with AKAP350B,AKAP350C, and Truncated AKAP350A Map the TACC4 Bind-ing Domain of AKAP350—AKAP350A and AKAP350B differonly in their terminal 23 amino acids. However, AKAP350Ctruncates a 700-amino acid portion of the AKAP350A carboxylterminus, eliminating one of the carboxyl leucine zippers andreplacing it with a novel 47-amino acid carboxyl-terminal se-quence. Each of the splice variant ends were amplified bypolymerase chain reaction and cloned into the pBD-Gal4 bind-ing domain vector. Binary assays were performed with pAD-TACC4 and each of the pBD-AKAP350 carboxyl-terminal splicevariants (Fig. 6). Expression of appropriately sized pBD proteinwas confirmed by Western blot in all negative �-galactosidaseproduction assays. AKAP350A and AKAP350B, but notAKAP350C, demonstrated an interaction with TACC4. Thus,the carboxyl-terminal 700 amino acid residues truncated inAKAP350C are important for TACC4 interaction, whereas the23 amino acids differing between AKAP350A and AKAP350Bresulted in no change in TACC4 interaction.

To identify the specific region of AKAP350 necessary forTACC4 interaction, amino and carboxyl pBD-AKAP350 trun-cations were constructed and assayed against pAD-TACC4. Acarboxyl truncation from amino acid 3265 to 3520 (the lastamino acid common to AKAP350A and AKAP350B) main-

FIG. 1. Nucleotide and deduced amino acid sequence ofTACC4. Cloning of TACC4 from rabbit spleen cDNA resulted in a1532-nucleotide sequence. The open reading frame codes for 454 aminoacids. The polyadenylation signal in the 3�-untranslated region is dou-ble underlined. An upstream in-frame stop codon, TGA, is underlined inthe 5�-untranslated region. A predicted cAMP-dependent protein ki-nase phosphorylation site is bold at amino acids 160–163. A predictedtyrosine phosphorylation site that is conserved in the TACC family isbold at amino acids 372–380.

TACC4 Interaction with AKAP35030168

by guest on March 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

tained interaction with TACC4 indicating that the bindingdomain of AKAP350 is in the region common to AKAP350A andAKAP350B. Further carboxyl truncation of 32 amino acids(amino acids 3265–3488) eliminated interaction with TACC4,indicating that the region of AKAP350 important for TACC4binding is within the 32 amino acids from 3488 to 3520 (Fig. 6,solid box). Finally, a minimum TACC4 binding region ofAKAP350A (amino acids 3376–3531) maintained interaction

with pAD-TACC4.In Vitro Association of AKAP350A with TACC4—In vitro

binding assays were performed to confirm the observed yeasttwo-hybrid interaction of AKAP350 and TACC4. RecombinantGST-TACC4-(1–454) fusion protein was attached to glutathi-one-Sepharose beads. A 100,000 � g gastric mucosa superna-tant was incubated with glutathione beads alone, GST-gluta-thione beads, or GST-TACC4-glutathione beads. Samples were

FIG. 2. TACC4 is a novel member of the TACC protein family. A BLAST search of available public data bases revealed TACC4 hassignificant identity to other TACC family members. An amino acid identity calculation of the highly conserved coiled-coil TACC domain (CC,hatched box) and the variable (V) region indicated rabbit TACC4 is most similar to murine TACC3 and human TACC3. Murine TACC3 and humanTACC3 share an amino-terminal region of 103 amino acids (solid) with 61% identity that is not present in rabbit TACC4 or other TACC proteins.

FIG. 3. Amino acid sequence align-ment of rabbit TACC4, humanTACC3, and murine TACC3. A se-quence alignment of the TACC familymembers with highest homology toTACC4, murine TACC3, and humanTACC3, was performed. Identical resi-dues are shaded. Rabbit TACC4, murineTACC3, and human TACC3 share highestidentity within the carboxyl 200 aminoacids in the TACC coiled-coil domain. Hu-man TACC3 and murine TACC3 alsoshare a region of high identity in the first103 amino acids that is not present inrabbit TACC4.

TACC4 Interaction with AKAP350 30169

by guest on March 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

eluted with 1% SDS, and proteins were separated on SDS-PAGE followed by Western blotting with 14G2, an anti-AKAP350 monoclonal antibody (Fig. 7A). AKAP350 immuno-reactivity was specifically recovered only from GST-TACC4beads.

Western Blot Analysis of TACC4—Anti-TACC4 polyclonalantibody (NE27) was raised against a peptide sequence specificfor TACC4 that is absent in human and murine TACC3 (Fig.8A). Using NE27, Western blot analysis of 3 M urea-treatedJurkat cell lysate detected a band of immunoreactivity (Fig.8B) with an approximate molecular mass of 52 kDa as well as

a band at �235 kDa. Notably, in samples were prepared with-out urea, the TACC4 immunoreactive band only migrated at235 kDa. Immunoreactivity in both cases was not observedwhen antibody was preincubated with 2.5 �M antigen peptide(data not shown). These results suggest that TACC4 may formSDS-insoluble aggregates.

Immunolocalization of TACC4—Jurkat cells were fixed andstained with polyclonal anti-TACC4 (NE27) and monoclonalanti-AKAP350 (14G2) (Fig. 9A) or anti-�-tubulin (Fig. 9B).Similarly, Jurkat cells were also stained with polyclonal anti-AKAP350A and monoclonal anti-�-tubulin antibodies (Fig. 9C).TACC4 colocalized with AKAP350 at the centrosome in inter-phase Jurkat cells (Fig. 9A). Anti-AKAP350A colocalized with�-tubulin, confirming the dual localization of TACC4 andAKAP350A at the centrosome (Fig. 9C). However, TACC4 lo-calized to the spindle apparatus in mitotic Jurkat cells, asrevealed by colocalization with �-tubulin (Fig. 9B). The anti-TACC4 staining decorated the spindle microtubule strands andappeared to accumulate near the spindle poles.

GFP-TACC4 Localizes to the Centrosome and Spindle Appa-ratus—To analyze further TACC4 targeting, full-lengthTACC4 was cloned into pEGFP-C2 to express a GFP-TACC4chimera protein containing GFP on the amino terminus (Fig.10). The expression level of GFP-TACC4 in transiently trans-

FIG. 4. Northern blot analysis of the tissue distribution of rab-bit TACC4. A random primed TACC4 cDNA probe (nucleotides 372–765) was used to probe total RNA (20 �g) from rabbit tissue (S, spleen;DC, distal colon; I, ileum; J, jejunum; D, duodenum; A, antrum; F,fundus; L, liver; B, brain; M, molecular weight markers as a nonspecificcontrol; P, pancreas). The probe recognized a single 1.6-kb message,which was enriched in spleen, jejunum, and duodenum.

FIG. 5. Y2H mapping of the AKAP350 binding domain ofTACC4. pAD-Gal4 activation domain vectors were constructed withamino and carboxyl truncated forms of TACC4. The coiled-coil region inthe TACC proteins is hatched. The resultant vectors were probedagainst binding domain pBD-AKAP350A. Carboxyl truncations toamino acid 380 (TACC4-(1–380)) revealed a region of TACC4 fromamino acid 380 to 404 (solid box), which is necessary for interactionwith AKAP350A. Amino truncations identified that a second region ofTACC4 from amino acid 247 to 284 (solid box) is also necessary forinteraction with AKAP350A. Using this data, a minimal AKAP350binding domain of TACC4 from amino acid 247 to 404 (TACC4-(247–404)) was constructed. pAD-Gal4 constructs of the highly homologouscoiled-coil TACC regions of human TACC3 (hTACC3), TACC2(hTACC2), and TACC1 (hTACC1) were constructed and tested for in-teraction with AKAP350A. Interestingly, only human TACC3 was ableto interact with AKAP350A.

FIG. 6. Y2H mapping of the TACC4 binding domain ofAKAP350. AKAP350A and AKAP350B differ only in their terminal 23amino acids (compare hatched box in AKAP350B with checkerboard boxin AKAP350A). However, AKAP350C truncates a 700-amino acid por-tion of the AKAP350A carboxyl terminus, eliminating one carboxylleucine zipper and replacing it with a novel 47-amino acid carboxyl-terminal sequence (vertically striped box). The carboxyl end ofAKAP350B was amplified from the same start site as the originalAKAP350A bait fragment and cloned into pBD-Gal4 binding domainvector as was a carboxyl portion of AKAP350C. Yeast two-hybridbinary assays were performed with full-length pAD-TACC4 and eachof the pBD-AKAP350 carboxyl-terminal splice variants. AKAP350Aand AKAP350B but not AKAP350C demonstrated interaction withTACC4. Truncation of AKAP350 to amino acid 3488 revealed a singleregion necessary for TACC4 interaction from amino acid 3488 to 3520(solid region). Using this data, a TACC4 minimal binding domainof AKAP350 was constructed from amino acid 3376 to 3531(AKAP350A-(3376–3531)).

FIG. 7. In vitro confirmation of AKAP350/TACC4 interaction.Recombinant GST-TACC4-(1–454) fusion protein was attached to glu-tathione-Sepharose beads. A 100,000 � g gastric mucosal supernatant(S) was incubated with glutathione beads alone (C), GST-glutathionebeads (G), or GST-TACC4-glutathione beads (T). Samples were elutedwith 1% SDS, subjected to SDS-PAGE, followed by Western blottingwith 14G2 anti-AKAP350 monoclonal antibody. AKAP350 immunore-activity was specifically recovered only from GST-TACC4 beads. Theseresults are representative of three separate experiments.

TACC4 Interaction with AKAP35030170

by guest on March 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

fected Jurkat cells was variable and resulted in different local-ization patterns. When expressed at low levels, GFP-TACC4(Fig. 10A) targeted predominantly to the centrosome andcaused a slight decrease in endogenous AKAP350 immuno-staining. The AKAP350 staining appeared to be dispersed toregions adjacent to the centrosome that also contained GFP-TACC4. High expression of GFP-TACC4 also targeted to thecentrosome (Fig. 10B, long arrow) and resulted in an increasednumber of cells present in prometaphase as determined byDAPI staining (Fig. 10, column 4). In the transfected promet-aphase cells, GFP-TACC4 targeted to the spindle apparatus(Fig. 10B, arrowhead) colocalizing with �-tubulin (B). Notably,high expression of GFP-TACC4 also resulted in cytosolic accu-mulations of GFP-TACC4 that were morphologically similar tothe pseudo-crystalline aggregates observed by Raff and col-leagues (34) in overexpressed TACC coiled-coil domains (34).Their electron microscopic analysis indicated that the aggre-gates result from a latticework of the coiled-coil TACC domainspresent in the TACC family members. The aggregations pres-ent are probably a result of the highly homologous TACC do-main present in TACC4.

In GFP-TACC4-transfected cells, the majority of AKAP350A,localized by anti-AKAP350A staining, remained on the centro-somes with only slight dispersal onto spindle fibers (Fig. 10B,short arrows). Thus, as with endogenous TACC4 staining (Fig.9), GFP-TACC4 disassociates from the centrosome during di-vision and relocates to the spindle apparatus, whereasAKAP350A predominantly retains localization to the centro-some. The endogenous TACC4/AKAP350 localization as well asthe GFP-TACC4 targeting led us to believe that the AKAP350/TACC4 association was dependent on the cell cycle. To assessthis idea, we constructed two more GFP fusion proteins guidedby the yeast two-hybrid binary interaction data.

GFP-TACC4-(1–380), a Non-AKAP350-interacting Protein,Localizes to the Spindle but Is Unable to Target to the Centro-some—We subcloned a carboxyl truncated TACC4-(1–380) intopEGFP-C2 and transiently transfected GFP-TACC4-(1–380)(Fig. 11) into Jurkat cells. TACC4-(1–380) did not interact withAKAP350A in yeast two-hybrid binary assays (Fig. 5). GFP-TACC4-(1–380) was dispersed throughout the cytosol and didnot target to any specific organelle in interphase cells (Fig. 11,A and B). Endogenous AKAP350A localization to the centro-some, as shown by colocalization of �-tubulin and anti-AKAP350A, was unaffected by GFP-TACC4-(1–380) expression(Fig. 11A). Importantly, there was no discernible targeting of

GFP-TACC4-(1–380) to the centrosome, demonstrating thatthe region between amino acids 380 and 454 was necessary forTACC4 localization to the centrosome in interphase cells. Thissame region of TACC4 was required for interaction withAKAP350A in yeast two-hybrid binary assays. This correlationsuggested that TACC4 interaction with AKAP350A might beresponsible for TACC4 sequestration to the centrosome duringinterphase.

The net result of GFP-TACC4-(1–380) targeting indicatedthat this TACC4 truncation differs from endogenous and full-length GFP-TACC4 by its inability to localize to the interphasecentrosome. However, GFP-(1–380) retained the same ability ofendogenous TACC4 and GFP-TACC4 to target to the mitoticspindle apparatus (Fig. 11, C and D, arrowheads). AKAP350staining, on the other hand, remained punctate on the remnantcentrosomes (Fig. 11C, long arrows). Notably, GFP-TACC4-(1–380) formed aggregates (Fig. 11C, short arrows) that weremorphologically similar to those formed by GFP-TACC4. Thisexpression pattern suggested that the region of TACC4 respon-sible for interaction with the spindle apparatus was containedwithin amino acids 1–380 and did not require interaction withAKAP350. Thus, there are distinct regions of TACC4 respon-sible for centrosome localization and spindle localization.

The AKAP350 Minimal Binding Domain of TACC4, GFP-TACC4-(247–404), Targets the Centrosome but Not the SpindleApparatus—The second construct linked the AKAP350 mini-mal binding domain of TACC4 (amino acids 247–404) to GFPforming GFP-TACC4-(247–404). Similar to GFP-TACC4, vari-able expression of GFP-TACC4-(247–404) resulted in two lo-calization patterns (Fig. 12). Low expression of GFP-TACC4-(247–404) targeted to the centrosome (Fig. 12A). Anti-TACC4staining of the same cells revealed slight dispersal of endoge-nous TACC4 from the centrosome with no observed effect inendogenous AKAP350 staining. In yeast two-hybrid binaryassays, this region of TACC4 interacted with AKAP350A.These results provide further evidence that interaction withAKAP350 correlates with TACC4 localization to thecentrosome.

High expression of GFP-TACC4-(247–404) formed anti-TACC4-reactive aggregates (Fig. 12B). This observation wasnot surprising because of the significant amount of the TACCdomain present in GFP-TACC4-(247–404). Anti-TACC4 stain-ing revealed that a significant amount of endogenous TACC4was pulled into the GFP-TACC4-(247–404) aggregates, sug-gesting that the TACC domain may elicit seeding of multimericcomplexes of endogenous TACC4. Endogenous AKAP350 wasnot pulled into these aggregates, but AKAP350 staining inten-sity at the centrosome was decreased (Fig. 12B, arrow). UnlikeGFP-TACC4 and GFP-TACC4-(1–380), GFP-TACC4-(247–404) never localized to the spindle apparatus (Fig. 12C) inmitotic cells, implying that the spindle-interacting domain ofTACC4 is present outside of amino acids 247–404.

Overexpression of GFP-TACC4 and GFP-TACC4-(1–380) Re-sults in Prometaphase Arrest—When immunostaining of thechimeric GFP-TACC4 proteins was performed, we observed ageneral increase in the number of transfected mitotic cells. Toquantitate the observed increase, we assessed the cell cyclestage in DAPI staining of �100 transfected cells from threeseparate transfections. Cell cycle status was evaluated for cellstransfected with GFP, GFP-TACC4, GFP-TACC4-(1–380), andGFP-TACC4-(247–404). The most common mitotic morphologyobserved, condensed chromatin and a dissolved nuclear enve-lope, corresponded to prometaphase. Therefore, we statisticallycompared the number of cells present in interphase versusprometaphase (Fig. 13). The number of transfected cells pres-ent in prometaphase was significantly increased for GFP-

FIG. 8. Polyclonal anti-TACC4 antibody Western blot of Jurkatcell lysate. Anti-TACC4 polyclonal antibody (NE27) was raised againsta peptide sequence (A) specific for TACC4 and absent in human andmurine TACC3. Using NE27, Western blot analysis of 3 M urea treatedJurkat cell lysate detected two bands of immunoreactivity (B) withapproximate molecular mass values of 52 and 235 kDa. Notably, theTACC4 immunoreactive band only migrated at 235 kDa when sampleswere prepared without urea. Immunoreactivity in both cases was notobserved when antibody was preincubated with antigen peptide.

TACC4 Interaction with AKAP350 30171

by guest on March 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

TACC4 and GFP-TACC4-(1–380) compared with GFP andGFP-TACC4-(247–404). These results suggested a link be-tween localization of GFP-TACC4 proteins to the spindle ap-paratus and prometaphase arrest. Thus, increased number ofprometaphase cells required overexpression of GFP-TACC4proteins that contained the region of TACC4 implicated inspindle apparatus targeting.

DISCUSSION

The AKAP350 family of proteins interacts with numeroussignaling proteins from multiple signal transduction pathways.The shortest 3� splice variant, Yotiao protein, associates with

the neuronal NMDA receptor and contains a single binding sitefor the RII subunit of type II cAMP-dependent protein kinase(18, 35). The other AKAP350 splice variants contain two bind-ing sites for the RII subunit of type II cAMP-dependent proteinkinase (11–13). In addition, these proteins also contain bindingregions for phosphatase 1 and protein kinase N in their Yotiaohomology regions, as well as binding regions for protein phos-phatase 2a and protein kinase C� in their carboxyl-terminalregion (13, 16). Most recently, Tasken et al. (17) have shownthat a central region of AKAP350 interacts with cAMP-depend-ent phosphodiesterase 4D3 (PDE4D3). Notably, all of the pub-lished AKAP350 interaction partners are enzymatic proteins

FIG. 9. TACC4 targets the centrosome and mitotic spindle apparatus. Jurkat cells were fixed and stained with the labeled antibodies andDAPI (column 3) for cell cycle identification. Triple-labeled pseudocolor images were constructed using blue for DAPI, red for column 1 antibodystaining, and green for column 2 antibody staining. Colocalization of the red and green images is seen as yellow in the triple label (column 4). Dualimmunostaining of interphase Jurkat cells with anti-TACC4 and anti-AKAP350 (A) identified TACC4 colocalization with AKAP350 at thecentrosome. Anti-TACC4 colocalized with anti-�-tubulin on the spindle apparatus in mitotic Jurkat cells (B). Anti-AKAP350A staining colocalizedwith �-tubulin at the centrosome in interphase Jurkat cells (C).

FIG. 10. Overexpression of GFP-TACC4 targets the centrosome and spindle apparatus. Full-length TACC4 was cloned into pEGFP-C2and transiently transfected into Jurkat cells. The cells were fixed and stained with the labeled antibodies and DAPI (column 4) for cell cycleidentification. Triple-labeled pseudocolor images were constructed using blue for column 3 antibody staining, red for column 1 antibody staining,and green for column 2 GFP-TACC4 expression. Colocalization of the red and green images is seen as yellow in the triple label (column 5).Colocalization of the green and yellow images is seen as aqua blue. Areas of colocalization between all three images are seen as white in the triplelabel. Low expression of GFP-TACC4 (A) targeted to the centrosome with endogenous anti-TACC4 staining and anti-AKAP350 staining. AKAP350staining appeared dispersed or decreased in some cells. High expression of GFP-TACC4 (B) resulted in aggregate formation and localization to thecentrosome (long arrows) in interphase cells and the spindle apparatus (arrowheads) in dividing cells. The majority of anti-AKAP350A (B, shortarrows) staining remained at the centrosome in interphase and mitotic Jurkat cells.

TACC4 Interaction with AKAP35030172

by guest on March 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

that regulate signaling pathways by phosphorylation statechanges or changes in second messenger levels. Utilizing yeasttwo-hybrid screening with AKAP350A as bait, we have nowidentified the first evidence for two AKAP350 interaction part-ners, TACC3 and TACC4, which have putative non-enzymaticeffector functions.

The original rabbit clone identified from the AKAP350Ascreen contained high homology to the TACC gene family.Thus, we named the putative interacting clone TACC4. Subse-quent cloning of the full-length gene for TACC4 identified thatTACC4 has significant homology with human and murineTACC3. A polyclonal antibody, NE27, was raised to a 15-merpeptide sequence, which is present only in TACC4 and not inhuman TACC3. Anti-TACC4 stained the centrosome through-out interphase and the spindle apparatus during mitosis inJurkat cells. This staining pattern is significantly differentfrom previously published anti-TACC3 localization in whichTACC3 only stained a diffuse area around the centrosomeduring mitosis in HeLa cells (34). Anti-TACC4 also did notrecognize a human GFP-TACC3 chimera by immunocytochem-istry (data not shown). This distinct staining pattern suggeststhat TACC4 is a novel gene product. Nevertheless, we cannotrule out the possibility that TACC4 is a splice variant of TACC3

with an alternative 5�-start site. Current public data baseinformation does not permit us to identify definitively all of thegenomic sequence flanking the human TACC3 gene.

We have confirmed the yeast two-hybrid TACC4 interactionwith AKAP350 by both biochemical and immunocytochemicalapproaches. First, using a biochemical GST-TACC4 pull-downassay, we demonstrated that only GST-TACC4 beads were ableto pull down an anti-AKAP350-reactive band. Second, anti-AKAP350 and anti-TACC4 staining colocalized to the centro-some in interphase Jurkat cells. These results support theyeast two-hybrid interaction findings that TACC4 interactswith AKAP350. Further, yeast two-hybrid binary assays re-vealed that the only mammalian TACC family members thatcan interact with AKAP350 are TACC3 and TACC4. Binaryassays also demonstrated that TACC4 interacts withAKAP350A and AKAP350B, but not AKAP350C. Truncationsof AKAP350 localize the TACC4 binding domain of AKAP350 to31 amino acids from 3488 to 3520 present in the carboxyl endsof both AKAP350A and AKAP350B.

Our data suggest that AKAP350 interaction with TACC4 isnecessary for localization of TACC4 to the centrosome in inter-phase Jurkat cells. Immunocytochemistry with anti-AKAP350and anti-TACC4 antibodies demonstrated a cell cycle-depend-

FIG. 11. GFP-TACC4-(1–380) overexpression shows cytoplasmic dispersal and spindle targeting. TACC4 (amino acids 1–380) wascloned into pEGFP-C2 and transiently transfected into Jurkat cells. The cells were fixed and stained with the labeled antibodies and DAPI (column4) for cell cycle identification. Triple-labeled pseudocolor images were constructed using blue for column 3 antibody staining, red for column 1antibody staining, and green for column 2 GFP-TACC4-(1–380) expression. Immunostaining with anti-AKAP350A and anti-�-tubulin (A) identifiedcytoplasmic dispersal of GFP-TACC4-(1–380). Notably, GFP-TACC4-(1–380) expression did not appear to affect endogenous AKAP350A localiza-tion to the centrosome. High expression levels of cytoplasmic GFP-TACC4-(1–380) appeared to perturb microtubules (anti-�-tubulin), whereasthere was no effect on anti-AKAP350A centrosomal staining (B). GFP-TACC4-(1–380) expression in mitotic cells (C) targeted to the spindleapparatus (arrowheads) and formed aggregates (short arrows) in a manner similar to GFP-TACC4. Anti-TACC4 staining was also seen on thespindle apparatus and the aggregates, whereas anti-AKAP350 staining remained (long arrows) at the centrosome structures. Whether or notGFP-TACC4-(1–380) pulled in endogenous TACC4 could not be addressed because GFP-TACC4-(1–380) contains the antigen peptide epitope usedto raise anti-TACC4. GFP-TACC4-(1–380) localization to the spindle apparatus was confirmed by colocalization with �-tubulin (D).

TACC4 Interaction with AKAP350 30173

by guest on March 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

ent variation in TACC4 colocalization with AKAP350.Throughout the cell cycle, AKAP350 remains on the centro-some with only a slight dispersal onto spindle microtubulefibers during mitosis. In interphase cells, anti-AKAP350 stain-ing is located on the centrosome. Anti-TACC4 staining duringinterphase is also present at the centrosome, suggesting thatAKAP350 may sequester TACC4 to the centrosome in non-mitotic cells. Once cells enter mitosis, TACC4 loses its associ-ation with the centrosome and translocates to the spindle ap-paratus where anti-TACC4 staining decorated spindle strandsand accumulated at the spindle poles. Thus, endogenous stain-ing patterns suggest that TACC4 only interacts with AKAP350in interphase.

There are two distinct targeting regions of TACC4. Oneregion is responsible for AKAP350 binding and sequestrationto the centrosome during interphase. The other region is re-sponsible for TACC4 localization to the spindle apparatus dur-ing mitosis. These two distinct TACC4 regions are deducedfrom our yeast two-hybrid binaries and GFP-TACC4 expres-sion studies. The TACC4 region responsible for interactionwith AKAP350 requires a domain of TACC4 from amino acid247 to 404. A requirement of the subregion from amino acid 380to 404 is shown by the inability of pAD-TACC4-(1–380) tointeract with pBD-AKAP350A and the loss of AKAP350 colo-calization and centrosome targeting by GFP-TACC4-(1–380).Amino-terminal truncation of TACC4 defined a second subre-gion at the beginning of the coiled-coil TACC domain betweenamino acids 247 and 284 that is necessary for AKAP350Ainteraction. A TACC4 construct from amino acid 247 to 404maintained interaction with AKAP350A. Expression of thesame region, GFP-TACC4-(247–404), targeted GFP to the cen-trosome but did not localize to the spindle apparatus duringmitosis. Thus, amino acids 247–404 of TACC4 are responsiblefor centrosome localization of TACC4 and contain theAKAP350 minimal binding domain. However, this domain isnot the region responsible for TACC4 spindle localization.Thus, localization of TACC4 to the centrosome, but not the

spindle apparatus, requires interaction with AKAP350A.TACC4 localization to the spindle apparatus during mitosis

requires a distinct region in the amino terminus of the protein.GFP-TACC4 overexpression recapitulated the observed anti-TACC4 staining by localizing to the spindle apparatus in mi-totic cells. However, although carboxyl truncation of aminoacids 380–454 of TACC4 in GFP-TACC4-(1–380) resulted in aloss of interphase centrosomal localization, GFP-TACC4-(1–380) still targeted to the spindle apparatus in mitotic cells.These results, along with the observation that GFP-TACC4-(247–404) did not localize to the spindle apparatus, indicatethat the region of TACC4 necessary for TACC4 spindle local-ization is contained in the amino end of the protein betweenamino acids 1 and 247. Notably, overexpression of GFP-TACC4constructs with the ability to localize to the spindle apparatusalso resulted in an increase in the number of cells present inprometaphase. By contrast, overexpression of GFP-TACC4-(247–404) did not result in increased numbers of prometaphasecells. Thus, the AKAP350 minimal binding region of TACC4retains neither the region necessary for spindle localization northe ability to elicit prometaphase arrest.

The explanation for why overexpression of the spindle inter-action region of TACC4 causes an increase in the number ofprometaphase cells is not immediately apparent. The mostlikely possibility is that TACC4 overexpression perturbs spin-dle structure, resulting in an activation of the spindle check-point and arrest of the dividing cell at prometaphase (reviewedin Refs. 36–38). Alternatively, TACC4 may be a protein mem-ber of the spindle checkpoint machinery. Overexpression ofTACC4 would then result in perturbation of the checkpointmachinery dynamics, causing an increase in prometaphasecells. In either case, the observed prometaphase arrest sug-gests that interaction of TACC4 with the spindle apparatus isof purposeful importance to normal spindle function.

The dynamics of interaction of TACC4 with the spindle ap-paratus remain to be elucidated, but recent studies of theDrosophila TACC homologue, D-TACC, suggest it may interact

FIG. 12. Overexpression of the minimal AKAP350 binding domain of TACC4, GFP-TACC4-(247–404), targets the centrosome butnot the spindle apparatus. The minimal AKAP350 binding domain of TACC4 (amino acids 247–404) was cloned into EGFP-C2 and transientlytransfected into Jurkat cells. The cells were fixed and stained with the labeled antibodies and DAPI (column 4) for cell cycle identification.Triple-labeled pseudocolor images were constructed using red for column 1 antibody staining, green for column 2 GFP-TACC4-(247–404)expression, and blue for column 3 antibody staining. Low expression of GFP-TACC4-(247–404) targeted to the centrosome and colocalized withendogenous anti-TACC4 and anti-AKAP350 staining (A). High expression of GFP-TACC4-(247–404) exhibited aggregate formation that displayedanti-TACC4 staining (B) but did not target to the centrosome. Anti-TACC4 and anti-AKAP350 colocalization to the centrosome was unperturbed.Notably, expression of GFP-TACC4-(247–404) targeted to the spindle apparatus (C).

TACC4 Interaction with AKAP35030174

by guest on March 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

with a conserved microtubule-associated protein family impor-tant in microtubule dynamics (39, 40). D-TACC was initiallyidentified by copurification with microtubules in a microtubulespin-down experiment of Drosophila embryo extracts (32). D-TACC contains a carboxyl coiled-coil TACC domain homolo-gous to mammalian TACC family members. Overexpression ofthis TACC domain formed aggregates in HeLa cells, suggestinga conserved function of the TACC domain in TACC proteins(34). Dtacc mutants displayed abnormally short spindle micro-tubules and eventually died as a result of chromosomal segre-gation defects.

Recent work by the same group (39) and Cullen and col-leagues (40) have implicated interaction of D-TACC with Msps,a Drosophila member of the mammalian XMAP215/ch-TOGmicrotubule-associated protein family. Vertebrate members ofthis microtubule-associated protein family stabilize microtu-bules and antagonize the kinesin XKCM1 in controlling plus-end microtubule dynamics in vitro (41). Defective tripolar mei-otic acentrosomal spindles were formed in both dtacc and mspsmutants (40). D-TACC and Ncd, a minus-end-directed micro-tubule motor, were involved in efficient localization of Msps toacentrosomal meiotic spindle poles. Similarly, Msps colocalizedwith D-TACC at mitotic spindle poles in Drosophila embryosand localized to overexpressed D-TACC TACC domain aggre-gates (39).

Notably, the interaction of these two Drosophila proteinsappears to be conserved in human homologues TACC3 andch-TOG; overexpressed TACC3 aggregates concentrated ch-TOG (39). The ch-TOG gene product, TOGp, localizes to perinu-clear cytoplasm in interphase and to the spindle apparatuswith concentration at the spindle poles during mitosis (42). Itseems plausible that TOGp or a protein complex that contains

TOGp may be responsible for TACC4 localization to the spindleapparatus in a similar manner to that seen in Drosophila withD-TACC and Msps.

A remaining question of whether unknown or knownAKAP350 interaction partners play a functional role in TACC4binding to AKAP350 remains to be determined. Notably,Tasken and colleagues (17) have recently reported thatAKAP350 has the ability to form functional regulatory com-plexes with known binding proteins. Targeting of the RII sub-unit to AKAP350 spatially localizes protein kinase A to thecentrosome region whereas temporal control is exerted in thesame anchored region by phosphodiesterase 4D3 degradationof cAMP. Therefore, it is reasonable to hypothesize that asimilar complex may exist between TACC4 and otherAKAP350 regulatory interaction proteins. The existence ofsuch an AKAP350 signaling complex would provide a plausibleexplanation for the observed cell cycle-dependent translocationof TACC4. Similar functional translocation is seen in manyproteins, in particular protein kinases (reviewed in Refs. 1 and43). Cell cycle phosphorylation-dependent translocation is seenwith cyclin B-p34cdc2 kinase (CDK1) activation. During G2

phase of the cell cycle, CDK1 is held in an inactive state byphosphorylation at residues Thr-14 and Tyr-15 by WEE1 andMYT1 (44). In late G2 phase or mitotic prophase, phosphataseCDC25C dephosphorylates and activates CDK1, resulting inthe translocation of CDK1 to the nucleus, an event that isrequired for cell cycle progression (45, 46). Notably, recentwork by Carlson and colleagues (47) has shown a cell cyclephosphorylation-dependent interaction between RII� andAKAP350. CDK1 is localized to the centrosome at the begin-ning of mitosis and phosphorylates RII� on Thr-54 (48). Thisphosphorylation by CDK1 results in dissociation of RII� from

FIG. 13. Overexpression of GFP-TACC4 and GFP-TACC4-(1–380) butnot GFP-TACC4-(247–404) resulted inan increased percentage of cells pres-ent at prometaphase. When visualiza-tion of the GFP-TACC4 proteins was per-formed, we observed a general trend ofincreased cells present at prometaphase.We quantitated the cell cycle stage for�100 cells for three separate transfec-tions of GFP-TACC4, GFP-TACC4-(1–380), GFP-TACC4-(247–404), and GFP.GFP-TACC4 and GFP-TACC4-(1–380)transfections resulted in a significant in-crease in the number of cells in promet-aphase compared with GFP-TACC4-(247–404) and GFP (*, p � 0.001compared with GFP-transfected cells).

TACC4 Interaction with AKAP350 30175

by guest on March 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

AKAP350. A similar event in which TACC4 is phosphorylatedor dephosphorylated could explain redistribution of TACC4from AKAP350 centrosome sequestration to the spindle appa-ratus. A responsible kinase or phosphatase remains to bedetermined.

In summary, we have identified a novel TACC family mem-ber, TACC4, which interacts with AKAP350 at the centrosomein interphase cells and the spindle apparatus in mitotic cells.We propose that AKAP350 is responsible for sequesteringTACC4 to the centrosome in non-dividing cells. AKAP350 an-choring of TACC4 provides for a spatial localization of TACC4for its functional role in regulating the interaction with thespindle apparatus during mitosis. The numerous interactionpartners of AKAP350 suggest that the centrosomal AKAP350complex may play a role in activation and subsequent translo-cation of TACC4 to the spindle.

Acknowledgments—We thank Makio Iwashima and Robert Seymourfor their assistance in performing the Jurkat cell work. We also thankIvan Still for the expression vectors of TACC1, TACC2, and TACC3.

REFERENCES

1. Lester, L. B., and Scott, J. D. (1997) Recent Prog. Horm. Res. 52, 409–4302. Edwards, A. S., and Scott, J. D. (2000) Curr. Opin. Cell Biol. 12, 217–2213. Carr, D. W., Stofko-Hahn, R. E., Fraser, I. D., Bishop, S. M., Acott, T. S.,

Brennan, R. G., and Scott, J. D. (1991) J. Biol. Chem. 266, 14188–141924. Colledge, M., and Scott, J. D. (1999) Trends Cell Biol. 9, 216–2215. Faux, M. C., and Scott, J. D. (1996) Cell 85, 9–126. Coghlan, V. M., Perrino, B. A., Howard, M., Langeberg, L. K., Hicks, J. B.,

Gallatin, W. M., and Scott, J. D. (1995) Science 267, 108–1117. Nadal-Ginard, B., Smith, C. W., Patton, J. G., and Breitbart, R. E. (1991) Adv.

Enzyme Regul. 31, 261–2868. Nauert, J. B., Klauck, T. M., Langeberg, L. K., and Scott, J. D. (1997) Curr.

Biol. 7, 52–629. Westphal, R. S., Tavalin, S. J., Lin, J. W., Alto, N. M., Fraser, I. D., Langeberg,

L. K., Sheng, M., and Scott, J. D. (1999) Science 285, 93–9610. Dodge, K. L., Khouangsathiene, S., Kapiloff, M. S., Mouton, R., Hill, E. V.,

Houslay, M. D., Langeberg, L. K., and Scott, J. D. (2001) EMBO J. 20,1921–1930

11. Schmidt, P. H., Dransfield, D. T., Claudio, J. O., Hawley, R. G., Trotter, K. W.,Milgram, S. L., and Goldenring, J. R. (1999) J. Biol. Chem. 274, 3055–3066

12. Witczak, O., Skalhegg, B. S., Keryer, G., Bornens, M., Tasken, K., Jahnsen, T.,and Orstavik, S. (1999) EMBO J. 18, 1858–1868

13. Takahashi, M., Shibata, H., Shimakawa, M., Miyamoto, M., Mukai, H., andOno, Y. (1999) J. Biol. Chem. 274, 17267–17274

14. Keryer, G., Rios, R. M., Landmark, B. F., Skalhegg, B., Lohmann, S. M., andBornens, M. (1993) Exp. Cell Res. 204, 230–240

15. Dransfield, D. T., Yeh, J. L., Bradford, A. J., and Goldenring, J. R. (1997)Biochem. J. 322, 801–808

16. Takahashi, M., Mukai, H., Oishi, K., Isagawa, T., and Ono, Y. (2000) J. Biol.

Chem. 275, 34592–3459617. Tasken, K. A., Collas, P., Kemmner, W. A., Witczak, O., Conti, M., and Tasken,

K. (2001) J. Biol. Chem. 276, 21999–2200218. Lin, J. W., Wyszynski, M., Madhavan, R., Sealock, R., Kim, J. U., and Sheng,

M. (1998) J. Neurosci. 18, 2017–202719. Dictenberg, J. B., Zimmerman, W., Sparks, C. A., Young, A., Vidair, C., Zheng,

Y., Carrington, W., Fay, F. S., and Doxsey, S. J. (1998) J. Cell Biol. 141,163–174

20. Diviani, D., Langeberg, L. K., Doxsey, S. J., and Scott, J. D. (2000) Curr. Biol.10, 417–420

21. Diviani, D., and Scott, J. D. (2001) J. Cell Sci. 114, 1431–143722. Purohit, A., Tynan, S. H., Vallee, R., and Doxsey, S. J. (1999) J. Cell Biol. 147,

481–49223. Inaba, K., Morisawa, S., and Morisawa, M. (1998) J. Cell Sci. 111, 1105–111524. Lapierre, L. A., Tuma, P. L., Navarre, J., Goldenring, J. R., and Anderson,

J. M. (1999) J. Cell Sci. 112, 3723–373225. Goldenring, J. R., Shen, K. R., Vaughan, H. D., and Modlin, I. M. (1993) J. Biol.

Chem. 268, 18419–1842226. Lapierre, L. A., Kumar, R., Hales, C. M., Navarre, J., Bhartur, S. G., Burnette,

J. O., Provance, D. W., Jr., Mercer, J. A., Bahler, M., and Goldenring, J. R.(2001) Mol. Biol. Cell 12, 1843–1857

27. Wang, X., Kumar, R., Navarre, J., Casanova, J. E., and Goldenring, J. R. (2000)J. Biol. Chem. 275, 29138–29146

28. Still, I. H., Vince, P., and Cowell, J. K. (1999) Genomics 58, 165–17029. Chen, H. M., Schmeichel, K. L., Mian, I. S., Lelievre, S., Petersen, O. W., and

Bissell, M. J. (2000) Mol. Biol. Cell 11, 1357–136730. Pu, J. J., Li, C., Rodriguez, M., and Banerjee, D. (2001) Cytokine 13, 129–13731. Still, I. H., Hamilton, M., Vince, P., Wolfman, A., and Cowell, J. K. (1999)

Oncogene 18, 4032–403832. Gergely, F., Kidd, D., Jeffers, K., Wakefield, J. G., and Raff, J. W. (2000)

EMBO J. 19, 241–25233. Sadek, C. M., Jalaguier, S., Feeney, E. P., Aitola, M., Damdimopoulos, A. E.,

Pelto-Huikko, M., and Gustafsson, J. A. (2000) Mech. Dev. 97, 13–2634. Gergely, F., Karlsson, C., Still, I., Cowell, J., Kilmartin, J., and Raff, J. W.

(2000) Proc. Natl. Acad. Sci. U. S. A. 97, 14352–1435735. Feliciello, A., Cardone, L., Garbi, C., Ginsberg, M. D., Varrone, S., Rubin, C. S.,

Avvedimento, E. V., and Gottesman, M. E. (1999) FEBS Lett. 464, 174–17836. Sudakin, V., Chan, G. K., and Yen, T. J. (2001) J. Cell Biol. 154, 925–93637. Amon, A. (1999) Curr. Opin. Genet. Dev. 9, 69–7538. Hardwick, K. G. (1998) Trends Genet. 14, 1–439. Lee, M. J., Gergely, F., Jeffers, K., Peak-Chew, S. Y., and Raff, J. W. (2001) Nat

Cell Biol. 3, 643–64940. Cullen, C. F., and Ohkura, H. (2001) Nat. Cell Biol. 3, 637–64241. Andersen, S. S. (2000) Trends Cell Biol. 10, 261–26742. Charrasse, S., Schroeder, M., Gauthier-Rouviere, C., Ango, F., Cassimeris, L.,

Gard, D. L., and Larroque, C. (1998) J. Cell Sci. 111, 1371–138343. Ferrell, J. E. Jr (1998) Trends Biochem. Sci 23, 461–46544. Heald, R., McLoughlin, M., and McKeon, F. (1993) Cell 74, 463–47445. Ookata, K., Hisanaga, S., Okano, T., Tachibana, K., and Kishimoto, T. (1992)

EMBO J. 11, 1763–177246. Pines, J., and Hunter, T. (1991) J. Cell Biol. 115, 1–1747. Carlson, C. R., Witczak, O., Vossebein, L., Labbe, J. C., Skalhegg, B. S.,

Keryer, G., Herberg, F. W., Collas, P., and Tasken, K. (2001) J. Cell Sci.114, 3243–3254

48. Keryer, G., Yassenko, M., Labbe, J. C., Castro, A., Lohmann, S. M., Evain-Brion, D., and Tasken, K. (1998) J. Biol. Chem. 273, 34594–34602

TACC4 Interaction with AKAP35030176

by guest on March 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

GoldenringBrent T. Steadman, P. Henry Schmidt, Ryan A. Shanks, Lynne A. Lapierre and James R.

AKAP350 and the Mitotic Spindle ApparatusTransforming Acidic Coiled-coil-containing Protein 4 Interacts with Centrosomal

doi: 10.1074/jbc.M201914200 originally published online May 15, 20022002, 277:30165-30176.J. Biol. Chem. 

  10.1074/jbc.M201914200Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/277/33/30165.full.html#ref-list-1

This article cites 48 references, 26 of which can be accessed free at

by guest on March 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from