A nanobody inhibitor of Fascin-1 actin-bundling activity and filopodia formation.

Fascin-1-mediated actin-bundling activity is central to the generation of plasma membrane protrusions required for cell migration. Dysregulated formation of cellular protrusions is observed in metastatic cancers, where they are required for increased invasiveness, and is often correlated with increased Fascin-1 abundance. Therefore, there is interest in generating therapeutic Fascin-1 inhibitors. We present the identification of Nb 3E11, a nanobody inhibitor of Fascin-1 actin-bundling activity and filopodia formation. The crystal structure of the Fascin-1/Nb 3E11 complex reveals the structural mechanism of inhibition. Nb 3E11 occludes an actin-binding site on the third ß-trefoil domain of Fascin-1 that is currently not targeted by chemical inhibitors. Binding of Nb 3E11 to Fascin-1 induces a conformational change in the adjacent domains to stabilize Fascin-1 in an inhibitory state similar to that adopted in the presence of small-molecule inhibitors. Nb 3E11 could be used as a tool inhibitor molecule to aid in the development of Fascin-1 targeted therapeutics.

Phase-separated foci of EML4-ALK facilitate signalling and depend upon an active kinase conformation.

Variants of the oncogenic EML4-ALK fusion protein contain a similar region of ALK encompassing the kinase domain, but different portions of EML4. Here, we show that EML4-ALK V1 and V3 proteins form cytoplasmic foci that contain components of the MAPK, PLCγ and PI3K signalling pathways. The ALK inhibitors ceritinib and lorlatinib dissolve these foci and EML4-ALK V3 but not V1 protein re-localises to microtubules, an effect recapitulated in a catalytically inactive EML4-ALK mutant. Mutations that promote a constitutively active ALK stabilise the cytoplasmic foci even in the presence of these inhibitors. In contrast, the inhibitor alectinib increases foci formation of both wild-type and catalytically inactive EML4-ALK V3 proteins, but not a Lys-Glu salt bridge mutant. We propose that EML4-ALK foci formation occurs as a result of transient association of stable EML4-ALK trimers mediated through an active conformation of the ALK kinase domain. Our results demonstrate the formation of EML4-ALK cytoplasmic foci that orchestrate oncogenic signalling and reveal that their assembly depends upon the conformational state of the catalytic domain and can be differentially modulated by structurally divergent ALK inhibitors.

Orally bioavailable CDK9/2 inhibitor shows mechanism-based therapeutic potential in MYCN-driven neuroblastoma.

The undruggable nature of oncogenic Myc transcription factors poses a therapeutic challenge in neuroblastoma, a pediatric cancer in which MYCN amplification is strongly associated with unfavorable outcome. Here, we show that CYC065 (fadraciclib), a clinical inhibitor of CDK9 and CDK2, selectively targeted MYCN-amplified neuroblastoma via multiple mechanisms. CDK9 - a component of the transcription elongation complex P-TEFb - bound to the MYCN-amplicon superenhancer, and its inhibition resulted in selective loss of nascent MYCN transcription. MYCN loss led to growth arrest, sensitizing cells for apoptosis following CDK2 inhibition. In MYCN-amplified neuroblastoma, MYCN invaded active enhancers, driving a transcriptionally encoded adrenergic gene expression program that was selectively reversed by CYC065. MYCN overexpression in mesenchymal neuroblastoma was sufficient to induce adrenergic identity and sensitize cells to CYC065. CYC065, used together with temozolomide, a reference therapy for relapsed neuroblastoma, caused long-term suppression of neuroblastoma growth in vivo, highlighting the clinical potential of CDK9/2 inhibition in the treatment of MYCN-amplified neuroblastoma.

EML4-ALK V3 oncogenic fusion proteins promote microtubule stabilization and accelerated migration through NEK9 and NEK7.

EML4-ALK is an oncogenic fusion present in ∼5% of non-small cell lung cancers. However, alternative breakpoints in the EML4 gene lead to distinct variants of EML4-ALK with different patient outcomes. Here, we show that, in cell models, EML4-ALK variant 3 (V3), which is linked to accelerated metastatic spread, causes microtubule stabilization, formation of extended cytoplasmic protrusions and increased cell migration. EML4-ALK V3 also recruits the NEK9 and NEK7 kinases to microtubules via the N-terminal EML4 microtubule-binding region. Overexpression of wild-type EML4, as well as constitutive activation of NEK9, also perturbs cell morphology and accelerates migration in a microtubule-dependent manner that requires the downstream kinase NEK7 but does not require ALK activity. Strikingly, elevated NEK9 expression is associated with reduced progression-free survival in EML4-ALK patients. Hence, we propose that EML4-ALK V3 promotes microtubule stabilization through NEK9 and NEK7, leading to increased cell migration. This represents a novel actionable pathway that could drive metastatic disease progression in EML4-ALK lung cancer.

2-Arylamino-6-ethynylpurines are cysteine-targeting irreversible inhibitors of Nek2 kinase.

Renewed interest in covalent inhibitors of enzymes implicated in disease states has afforded several agents targeted at protein kinases of relevance to cancers. We now report the design, synthesis and biological evaluation of 6-ethynylpurines that act as covalent inhibitors of Nek2 by capturing a cysteine residue (Cys22) close to the catalytic domain of this protein kinase. Examination of the crystal structure of the non-covalent inhibitor 3-((6-cyclohexylmethoxy-7H-purin-2-yl)amino)benzamide in complex with Nek2 indicated that replacing the alkoxy with an ethynyl group places the terminus of the alkyne close to Cys22 and in a position compatible with the stereoelectronic requirements of a Michael addition. A series of 6-ethynylpurines was prepared and a structure activity relationship (SAR) established for inhibition of Nek2. 6-Ethynyl-N-phenyl-7H-purin-2-amine [IC50 0.15 µM (Nek2)] and 4-((6-ethynyl-7H-purin-2-yl)amino)benzenesulfonamide (IC50 0.14 µM) were selected for determination of the mode of inhibition of Nek2, which was shown to be time-dependent, not reversed by addition of ATP and negated by site directed mutagenesis of Cys22 to alanine. Replacement of the ethynyl group by ethyl or cyano abrogated activity. Variation of substituents on the N-phenyl moiety for 6-ethynylpurines gave further SAR data for Nek2 inhibition. The data showed little correlation of activity with the nature of the substituent, indicating that after sufficient initial competitive binding to Nek2 subsequent covalent modification of Cys22 occurs in all cases. A typical activity profile was that for 2-(3-((6-ethynyl-9H-purin-2-yl)amino)phenyl)acetamide [IC50 0.06 µM (Nek2); GI50 (SKBR3) 2.2 µM] which exhibited >5-10-fold selectivity for Nek2 over other kinases; it also showed > 50% growth inhibition at 10 µM concentration against selected breast and leukaemia cell lines. X-ray crystallographic analysis confirmed that binding of the compound to the Nek2 ATP-binding site resulted in covalent modification of Cys22. Further studies confirmed that 2-(3-((6-ethynyl-9H-purin-2-yl)amino)phenyl)acetamide has the attributes of a drug-like compound with good aqueous solubility, no inhibition of hERG at 25 µM and a good stability profile in human liver microsomes. It is concluded that 6-ethynylpurines are promising agents for cancer treatment by virtue of their selective inhibition of Nek2.

Mitotic phosphorylation by NEK6 and NEK7 reduces the microtubule affinity of EML4 to promote chromosome congression.

EML4 is a microtubule-associated protein that promotes microtubule stability. We investigated its regulation across the cell cycle and found that EML4 was distributed as punctate foci along the microtubule lattice in interphase but exhibited reduced association with spindle microtubules in mitosis. Microtubule sedimentation and cryo-electron microscopy with 3D reconstruction revealed that the basic N-terminal domain of EML4 mediated its binding to the acidic C-terminal tails of α- and ß-tubulin on the microtubule surface. The mitotic kinases NEK6 and NEK7 phosphorylated the EML4 N-terminal domain at Ser144 and Ser146 in vitro, and depletion of these kinases in cells led to increased EML4 binding to microtubules in mitosis. An S144A-S146A double mutant not only bound inappropriately to mitotic microtubules but also increased their stability and interfered with chromosome congression. In addition, constitutive activation of NEK6 or NEK7 reduced the association of EML4 with interphase microtubules. Together, these data support a model in which NEK6- and NEK7-dependent phosphorylation promotes the dissociation of EML4 from microtubules in mitosis in a manner that is required for efficient chromosome congression.

Mitotic phosphorylation regulates Hsp72 spindle localization by uncoupling ATP binding from substrate release.

Hsp72 is a member of the 70-kDa heat shock family of molecular chaperones (Hsp70s) that comprise a nucleotide-binding domain (NBD) and a substrate-binding domain (SBD) connected by a linker that couples the exchange of adenosine diphosphate (ADP) for adenosine triphosphate (ATP) with the release of the protein substrate. Mitotic phosphorylation of Hsp72 by the kinase NEK6 at Thr66 located in the NBD promotes the localization of Hsp72 to the mitotic spindle and is required for efficient spindle assembly and chromosome congression and segregation. We determined the crystal structure of the Hsp72 NBD containing a genetically encoded phosphoserine at position 66. This revealed structural changes that stabilized interactions between subdomains within the NBD. ATP binding to the NBD of unmodified Hsp72 resulted in the release of substrate from the SBD, but phosphorylated Hsp72 retained substrate in the presence of ATP. Mutations that prevented phosphorylation-dependent subdomain interactions restored the connection between ATP binding and substrate release. Thus, phosphorylation of Thr66 is a reversible mechanism that decouples the allosteric connection between nucleotide binding and substrate release, providing further insight into the regulation of the Hsp70 family. We propose that phosphorylation of Hsp72 on Thr66 by NEK6 during mitosis promotes its localization to the spindle by stabilizing its interactions with components of the mitotic spindle.

Mitotic spindle association of TACC3 requires Aurora-A-dependent stabilization of a cryptic α-helix.

Aurora-A regulates the recruitment of TACC3 to the mitotic spindle through a phospho-dependent interaction with clathrin heavy chain (CHC). Here, we describe the structural basis of these interactions, mediated by three motifs in a disordered region of TACC3. A hydrophobic docking motif binds to a previously uncharacterized pocket on Aurora-A that is blocked in most kinases. Abrogation of the docking motif causes a delay in late mitosis, consistent with the cellular distribution of Aurora-A complexes. Phosphorylation of Ser558 engages a conformational switch in a second motif from a disordered state, needed to bind the kinase active site, into a helical conformation. The helix extends into a third, adjacent motif that is recognized by a helical-repeat region of CHC, not a recognized phospho-reader domain. This potentially widespread mechanism of phospho-recognition provides greater flexibility to tune the molecular details of the interaction than canonical recognition motifs that are dominated by phosphate binding.

Characterization of Three Druggable Hot-Spots in the Aurora-A/TPX2 Interaction Using Biochemical, Biophysical, and Fragment-Based Approaches.

The mitotic kinase Aurora-A and its partner protein TPX2 (Targeting Protein for Xenopus kinesin-like protein 2) are overexpressed in cancers, and it has been proposed that they work together as an oncogenic holoenzyme. TPX2 is responsible for activating Aurora-A during mitosis, ensuring proper cell division. Disruption of the interface with TPX2 is therefore a potential target for novel anticancer drugs that exploit the increased sensitivity of cancer cells to mitotic stress. Here, we investigate the interface using coprecipitation assays and isothermal titration calorimetry to quantify the energetic contribution of individual residues of TPX2. Residues Tyr8, Tyr10, Phe16, and Trp34 of TPX2 are shown to be crucial for robust complex formation, suggesting that the interaction could be abrogated through blocking any of the three pockets on Aurora-A that complement these residues. Phosphorylation of Aurora-A on Thr288 is also necessary for high-affinity binding, and here we identify arginine residues that communicate the phosphorylation of Thr288 to the TPX2 binding site. With these findings in mind, we conducted a high-throughput X-ray crystallography-based screen of 1255 fragments against Aurora-A and identified 59 hits. Over three-quarters of these hits bound to the pockets described above, both validating our identification of hotspots and demonstrating the druggability of this protein-protein interaction. Our study exemplifies the potential of high-throughput crystallography facilities such as XChem to aid drug discovery. These results will accelerate the development of chemical inhibitors of the Aurora-A/TPX2 interaction.

A moving target: structure and disorder in pursuit of Myc inhibitors.

The Myc proteins comprise a family of ubiquitous regulators of gene expression implicated in over half of all human cancers. They interact with a large number of other proteins, such as transcription factors, chromatin-modifying enzymes and kinases. Remarkably, few of these interactions have been characterized structurally. This is at least in part due to the intrinsically disordered nature of Myc proteins, which adopt a defined conformation only in the presence of binding partners. Owing to this behaviour, crystallographic studies on Myc proteins have been limited to short fragments in complex with other proteins. Most recently, we determined the crystal structure of Aurora-A kinase domain bound to a 28-amino acid fragment of the N-Myc transactivation domain. The structure reveals an α-helical segment within N-Myc capped by two tryptophan residues that recognize the surface of Aurora-A. The kinase domain acts as a molecular scaffold, independently of its catalytic activity, upon which this region of N-Myc becomes ordered. The binding site for N-Myc on Aurora-A is disrupted by certain ATP-competitive inhibitors, such as MLN8237 (alisertib) and CD532, and explains how these kinase inhibitors are able to disrupt the protein-protein interaction to affect Myc destabilization. Structural studies on this and other Myc complexes will lead to the design of protein-protein interaction inhibitors as chemical tools to dissect the complex pathways of Myc regulation and function, which may be developed into Myc inhibitors for the treatment of cancer.

Structural basis of N-Myc binding by Aurora-A and its destabilization by kinase inhibitors.

Myc family proteins promote cancer by inducing widespread changes in gene expression. Their rapid turnover by the ubiquitin-proteasome pathway is regulated through phosphorylation of Myc Box I and ubiquitination by the E3 ubiquitin ligase SCFFbxW7 However, N-Myc protein (the product of the MYCN oncogene) is stabilized in neuroblastoma by the protein kinase Aurora-A in a manner that is sensitive to certain Aurora-A-selective inhibitors. Here we identify a direct interaction between the catalytic domain of Aurora-A and a site flanking Myc Box I that also binds SCFFbxW7 We determined the crystal structure of the complex between Aurora-A and this region of N-Myc to 1.72-Å resolution. The structure indicates that the conformation of Aurora-A induced by compounds such as alisertib and CD532 is not compatible with the binding of N-Myc, explaining the activity of these compounds in neuroblastoma cells and providing a rational basis for the design of cancer therapeutics optimized for destabilization of the complex. We also propose a model for the stabilization mechanism in which binding to Aurora-A alters how N-Myc interacts with SCFFbxW7 to disfavor the generation of Lys48-linked polyubiquitin chains.

Allosteric inhibition of Aurora-A kinase by a synthetic vNAR domain.

The vast majority of clinically approved protein kinase inhibitors target the ATP-binding pocket directly. Consequently, many inhibitors have broad selectivity profiles and most have significant off-target effects. Allosteric inhibitors are generally more selective, but are difficult to identify because allosteric binding sites are often unknown or poorly characterized. Aurora-A is activated through binding of TPX2 to an allosteric site on the kinase catalytic domain, and this knowledge could be exploited to generate an inhibitor. Here, we generated an allosteric inhibitor of Aurora-A kinase based on a synthetic, vNAR single domain scaffold, vNAR-D01. Biochemical studies and a crystal structure of the Aurora-A/vNAR-D01 complex show that the vNAR domain overlaps with the TPX2 binding site. In contrast with the binding of TPX2, which stabilizes an active conformation of the kinase, binding of the vNAR domain stabilizes an inactive conformation, in which the αC-helix is distorted, the canonical Lys-Glu salt bridge is broken and the regulatory (R-) spine is disrupted by an additional hydrophobic side chain from the activation loop. These studies illustrate how single domain antibodies can be used to characterize the regulatory mechanisms of kinases and provide a rational basis for structure-guided design of allosteric Aurora-A kinase inhibitors.

A closed conformation of the Caenorhabditis elegans separase-securin complex.

The protease separase plays a key role in sister chromatid disjunction and centriole disengagement. To maintain genomic stability, separase activity is strictly regulated by binding of an inhibitory protein, securin. Despite its central role in cell division, the separase and securin complex is poorly understood at the structural level. This is partly owing to the difficulty of generating a sufficient quantity of homogeneous, stable protein. Here, we report the production of Caenorhabditis elegans separase-securin complex, and its characterization using biochemical methods and by negative staining electron microscopy. Single particle analysis generated a density map at a resolution of 21-24 Å that reveals a close, globular structure of complex connectivity harbouring two lobes. One lobe matches closely a homology model of the N-terminal HEAT repeat domain of separase, whereas the second lobe readily accommodates homology models of the separase C-terminal death and caspase-like domains. The globular structure of the C. elegans separase-securin complex contrasts with the more elongated structure previously described for the Homo sapiens complex, which could represent a different functional state of the complex, suggesting a mechanism for the regulation of separase activity through conformational change.

Molecular mechanisms that underpin EML4-ALK driven cancers and their response to targeted drugs.

A fusion between the EML4 (echinoderm microtubule-associated protein-like) and ALK (anaplastic lymphoma kinase) genes was identified in non-small cell lung cancer (NSCLC) in 2007 and there has been rapid progress in applying this knowledge to the benefit of patients. However, we have a poor understanding of EML4 and ALK biology and there are many challenges to devising the optimal strategy for treating EML4-ALK NSCLC patients. In this review, we describe the biology of EML4 and ALK, explain the main features of EML4-ALK fusion proteins and outline the therapies that target EML4-ALK. In particular, we highlight the recent advances in our understanding of the structures of EML proteins, describe the molecular mechanisms of resistance to ALK inhibitors and assess current thinking about combinations of ALK drugs with inhibitors that target other kinases or Hsp90.

Mechanistic basis of Nek7 activation through Nek9 binding and induced dimerization.

Mitotic spindle assembly requires the regulated activities of protein kinases such as Nek7 and Nek9. Nek7 is autoinhibited by the protrusion of Tyr97 into the active site and activated by the Nek9 non-catalytic C-terminal domain (CTD). CTD binding apparently releases autoinhibition because mutation of Tyr97 to phenylalanine increases Nek7 activity independently of Nek9. Here we find that self-association of the Nek9-CTD is needed for Nek7 activation. We map the minimal Nek7 binding region of Nek9 to residues 810-828. A crystal structure of Nek7(Y97F) bound to Nek9(810-828) reveals a binding site on the C-lobe of the Nek7 kinase domain. Nek7(Y97F) crystallizes as a back-to-back dimer between kinase domain N-lobes, in which the specific contacts within the interface are coupled to the conformation of residue 97. Hence, we propose that the Nek9-CTD activates Nek7 through promoting back-to-back dimerization that releases the autoinhibitory tyrosine residue, a mechanism conserved in unrelated kinase families.

Molecular mechanisms of human IRE1 activation through dimerization and ligand binding.

IRE1 transduces the unfolded protein response by splicing XBP1 through its C-terminal cytoplasmic kinase-RNase region. IRE1 autophosphorylation is coupled to RNase activity through formation of a back-to-back dimer, although the conservation of the underlying molecular mechanism is not clear from existing structures. We have crystallized human IRE1 in a back-to-back conformation only previously seen for the yeast homologue. In our structure the kinase domain appears primed for catalysis but the RNase domains are disengaged. Structure-function analysis reveals that IRE1 is autoinhibited through a Tyr-down mechanism related to that found in the unrelated Ser/Thr protein kinase Nek7. We have developed a compound that potently inhibits human IRE1 kinase activity while stimulating XBP1 splicing. A crystal structure of the inhibitor bound to IRE1 shows an increased ordering of the kinase activation loop. The structures of hIRE in apo and ligand-bound forms are consistent with a previously proposed model of IRE1 regulation in which formation of a back-to-back dimer coupled to adoption of a kinase-active conformation drive RNase activation. The structures provide opportunities for structure-guided design of IRE1 inhibitors.

Hsp72 is targeted to the mitotic spindle by Nek6 to promote K-fiber assembly and mitotic progression.

Hsp70 proteins represent a family of chaperones that regulate cellular homeostasis and are required for cancer cell survival. However, their function and regulation in mitosis remain unknown. In this paper, we show that the major inducible cytoplasmic Hsp70 isoform, Hsp72, is required for assembly of a robust bipolar spindle capable of efficient chromosome congression. Mechanistically, Hsp72 associates with the K-fiber-stabilizing proteins, ch-TOG and TACC3, and promotes their interaction with each other and recruitment to spindle microtubules (MTs). Targeting of Hsp72 to the mitotic spindle is dependent on phosphorylation at Thr-66 within its nucleotide-binding domain by the Nek6 kinase. Phosphorylated Hsp72 concentrates on spindle poles and sites of MT-kinetochore attachment. A phosphomimetic Hsp72 mutant rescued defects in K-fiber assembly, ch-TOG/TACC3 recruitment and mitotic progression that also resulted from Nek6 depletion. We therefore propose that Nek6 facilitates association of Hsp72 with the mitotic spindle, where it promotes stable K-fiber assembly through recruitment of the ch-TOG-TACC3 complex.

Microtubule association of EML proteins and the EML4-ALK variant 3 oncoprotein require an N-terminal trimerization domain.

Proteins of the echinoderm microtubule (MT)-associated protein (EMAP)-like (EML) family contribute to formation of the mitotic spindle and interphase MT network. EML1-4 consist of Trp-Asp 40 (WD40) repeats and an N-terminal region containing a putative coiled-coil. Recurrent gene rearrangements in non-small cell lung cancer (NSCLC) fuse EML4 to anaplastic lymphoma kinase (ALK) causing expression of several oncogenic fusion variants. The fusions have constitutive ALK activity due to self-association through the EML4 coiled-coil. We have determined crystal structures of the coiled-coils from EML2 and EML4, which describe the structural basis of both EML self-association and oncogenic EML4-ALK activation. The structures reveal a trimeric oligomerization state directed by a conserved pattern of hydrophobic residues and salt bridges. We show that the trimerization domain (TD) of EML1 is necessary and sufficient for self-association. The TD is also essential for MT binding; however, this property requires an adjacent basic region. These observations prompted us to investigate MT association of EML4-ALK and EML1-ABL1 (Abelson 1) fusions in which variable portions of the EML component are present. Uniquely, EML4-ALK variant 3, which includes the TD and basic region of EML4 but none of the WD40 repeats, was localized to MTs, both when expressed recombinantly and when expressed in a patient-derived NSCLC cell line (H2228). This raises the question of whether the mislocalization of ALK activity to MTs might influence downstream signalling and malignant properties of cells. Furthermore, the structure of EML4 TD may enable the development of protein-protein interaction inhibitors targeting the trimerization interface, providing a possible avenue towards therapeutic intervention in EML4-ALK NSCLC.

Crystal structure of EML1 reveals the basis for Hsp90 dependence of oncogenic EML4-ALK by disruption of an atypical ß-propeller domain.

Proteins of the echinoderm microtubule-associated protein (EMAP)-like (EML) family contribute to formation of the mitotic spindle and interphase microtubule network. They contain a unique hydrophobic EML protein (HELP) motif and a variable number of WD40 repeats. Recurrent gene rearrangements in nonsmall cell lung cancer fuse EML4 to anaplastic lymphoma kinase (ALK), causing expression of several fusion oncoprotein variants. We have determined a 2.6-Å crystal structure of the representative ∼70-kDa core of EML1, revealing an intimately associated pair of ß-propellers, which we term a TAPE (tandem atypical propeller in EMLs) domain. One propeller is highly atypical, having a discontinuous subdomain unrelated to a WD40 motif in place of one of its blades. This unexpected feature shows how a propeller structure can be assembled from subdomains with distinct folds. The HELP motif is not an independent domain but forms part of the hydrophobic core that joins the two ß-propellers. The TAPE domain binds α/ß-tubulin via its conserved, concave surface, including part of the atypical blade. Mapping the characteristic breakpoints of each EML4-ALK variant onto our structure indicates that the EML4 TAPE domain is truncated in many variants in a manner likely to make the fusion protein structurally unstable. We found that the heat shock protein 90 (Hsp90) inhibitor ganetespib induced degradation of these variants whereas others lacking a partial TAPE domain were resistant in both overexpression models and patient-derived cell lines. The Hsp90-sensitive EML4-ALK variants are exceptions to the rule that oncogenic fusion proteins involve breakpoints in disordered regions of both partners.

Coordination of adjacent domains mediates TACC3-ch-TOG-clathrin assembly and mitotic spindle binding.

A complex of transforming acidic coiled-coil protein 3 (TACC3), colonic and hepatic tumor overexpressed gene (ch-TOG), and clathrin has been implicated in mitotic spindle assembly and in the stabilization of kinetochore fibers by cross-linking microtubules. It is unclear how this complex binds microtubules and how the proteins in the complex interact with one another. TACC3 and clathrin have each been proposed to be the spindle recruitment factor. We have mapped the interactions within the complex and show that TACC3 and clathrin were interdependent for spindle recruitment, having to interact in order for either to be recruited to the spindle. The N-terminal domain of clathrin and the TACC domain of TACC3 in tandem made a microtubule interaction surface, coordinated by TACC3-clathrin binding. A dileucine motif and Aurora A-phosphorylated serine 558 on TACC3 bound to the "ankle" of clathrin. The other interaction within the complex involved a stutter in the TACC3 coiled-coil and a proposed novel sixth TOG domain in ch-TOG, which was required for microtubule localization of ch-TOG but not TACC3-clathrin.