I. Introduction

It is estimated that 60-90% of human cancer is due to the intricate and poorly understood interactions between exogenous environmental factors and multiple low penetrance genetic factors. For instance, no single gene defect has been identified that accounts for the initiation of sporadic ovarian cancer and only 10% of ovarian cancer patients inherit a familial predisposition. Even in this latter category, only 35-50% can be attributed to the inheritance of mutations in the BRCA1 and BRCA2 tumor suppressor genes (Gayther et al, 1999; Werness et al, 2000). Therefore, the underlying etiological bases of most sporadic tumors are largely unknown and additional unidentified gene/environment interactions must play a significant role in the etiology of cancer.

Several pieces of evidence have implicated the TACC family in processes underlying the development and progression of cancer (Still et al, 1999a,b; Conte et al, 2002, 2003; Line et al, 2002; Raff 2002; Lauffart et al, 2003; Stewart et al, 2004). Notably, in an immunohistochemical study of over 500 breast tumors, TACC3 (and TACC2) was identified as a member of a 21 protein set that could predict clinical outcome in the breast cancer patients (Jacquemier et al, 2005). TACC3 is also deregulated in multiple myeloma (Stewart et al, 2004), lung (Jung et al, 2006) and gastric cancer (Kim et al, 2005). TACC3 is normally expressed in the nucleus of normal ovarian surface epithelial cells (Aitola et al, 2003; Sadek et al, 2003; Lauffart et al, 2005), however, we have demonstrated that nearly 100% of dissected ovarian tumors show aberrations in the expression of TACC3 protein (Lauffart et al, 2005). This is manifested as loss of expression or mislocalization of the TACC3 protein from the nucleus to the cytoplasm. In addition, we have identified two constitutional mutations in the TACC3 gene specific to patients with ovarian cancer from the Gilda Radner Familial Ovarian Cancer Registry (Lauffart et al, 2005). These patients had previously tested negative for mutations in the BRCA1, BRCA2 and other previously identified predisposing genes. These findings indicate that functional deregulation/loss of nuclear TACC3 represents a common event underlying the development or progression of breast and ovarian tumors.

The exact mechanistic link between TACC function and development/progression of cancer is still poorly understood. A well-defined role of TACCs in the centrosome and formation of the mitotic spindle of species from yeast to man has been described (Reviewed in (Gergely 2002; Schneider et al, 2007)). Aurora kinase mediated phosphorylation of residues in the conserved N-terminal region of TACC3 is essential for recruitment of CKAP5/XMAP215 to the mitotic spindle and subsequent elongation of the spindle microtubules (Kinoshita et al, 2005). In vitro, this can result in the missegregation of metaphase chromosomes and subsequent aneuploidy in the daughter cells. However, in vivo, targeted disruption of the mouse TACC3 (and TACC2) largely fail to recapitulate such observations (Piekorz et al, 2002; Schuendeln et al, 2004). The TACC3 knockout is embryonic lethal with death occurring at E12-13, due to failure of differentiation events and increased apoptosis of neural and hematopoietic precursors (Piekorz et al, 2002). Intriguingly, the phenotype of the TACC3 knockout mouse partially resembles that of the DNA ligase IV deficient mice (Piekorz et al, 2002). Notably, similar to the BRCA1 knockout mice (Ludwig et al, 1997), the lethality of the TACC3 knockout in mice is partially rescued by a deficiency in p53 (Piekorz et al, 2002), raising the possibility that TACC3 genetically interacts with BRCA1- and p53-regulated pathways.

Depending upon the tissue examined, the TACCs can be found solely in the cytoplasm, solely in the nucleus, or distributed throughout the cell (Aitola et al, 2003; Sadek et al, 2003; Lauffart et al, 2005; Lauffart et al, 2006). However, in contrast to the concentration of research on the role of the TACCs in the formation of the mitotic spindle, the functions of the TACCs in the interphase cell have received less attention (Sadek et al, 2000; McKeveney et al, 2001; Conte et al, 2002; Angrisano et al, 2006; Lauffart et al, 2002, 2007). In part, clues to the function of TACC3 in the nucleus have come from large scale proteomics that aim to build binary protein-protein interaction maps (an interactome) for the invertebrate proteome (Walhout et al, 2000; Still et al, 2004). This approach has centered on the systematic high-throughput use of the yeast two-hybrid system to generate a protein-protein interaction map for these organisms (Walhout et al, 2000; Bader et al, 2003). Using "phylogenetic profiling" (Walhout and Vidal, 2001) and “cluster analysis”, this has led to the development of the first draft human protein-protein interaction database (Lehner and Fraser, 2004). Intriguingly, the interaction networks indicate that the single TACC protein in both C. elegans and D. melanogaster can indirectly complex with DNA damage response/repair proteins (Walhout et al, 2000; Still et al, 2004) (Figure 1). Notably, the ceTAC protein, which primarily consists of the unique TACC domain, directly binds the C. elegans orthologue of the BRCA1-associated protein, BARD1. The mammalian BARD1 is a known mediator of proapoptotic stress and p53-dependent cell death (Irminger-Finger et al, 2001), and in conjunction with BRCA1 contributes to the repair of double-stranded DNA breaks (Reviewed in (Irminger-Finger and Leung 2002)). These functions are also evident in C. elegans, with siRNA-mediated downregulation of the C. elegans TACC and BARD genes in whole larvae further implicating both proteins in BRCA1-mediated pathways and DNA repair (Boulton et al, 2004).

Thus genetic data from mammalian systems and proteomic data from invertebrate model organisms suggests that one or more TACC proteins functionally interact with DNA damage response/ repair pathways. However, to date, no physical or functional interactions between the TACCs and components of these pathways in humans have been described. In this manuscript, we demonstrate that TACCs are indeed components of DNA damage response/repair complexes in humans. Furthermore, siRNA-mediated downregulation of TACC3 sensitizes p53 wild type A2780 ovarian cancer cells to the DNA topoisomerase inhibitor adriamycin. This suggests that defects in the TACC3 protein in mammary and ovarian surface epithelial cells and/or their malignant derivatives may effect changes in the response to environmental DNA damaging agents and determine resistance to certain forms of chemotherapy.

II. Materials and Methods

A. Cell Lines and immunologicals

MCF7, T47D, A2780 and SkOV3 were obtained from ATCC and maintained in DMEM +10% fetal calf serum containing the antimycotic/antibiotic normocin (Invivogen, USA). α-TACC, α-pCAF, and α-β-Tubulin antibodies were as previously described (Gangisetty et al, 2004). Commercial antibodies were obtained from the following companies: mouse α-BRCA1 (#OP92T) from Calbiochem, rabbit α-BARD1 antibody #BL518 (Bethyl labs. Inc.), goat α-BARD1 (N-19), mouse α-p53 (DO1), normal IgG and α-rabbit-horseradish peroxidase conjugates were purchased from Santa Cruz Biotechnology, and mouse α-Ku70 (#611893) from BD Transduction laboratories. Secondary antibodies and conjugates were obtained from Jackson Laboratories.

B. Coimmunoprecipitation and immunoblotting

Human breast and ovarian cancer cells were washed with ice-cold 1xPBS. Nuclear extracts were then prepared according to the protocol of Schreiber et al (Schreiber et al, 1989), and diluted 1:2 in 50mM Tris-HCl pH7.4/0.5% v/v Nonidet-NP40 to reduce the salt concentration to 130mM. For coimmunoprecipitation 500mg of the diluted nuclear extract was precleared with IgG for 30 min. The extract was centrifuged for 5 min. at 4000g at 4˚C, and the cleared supernatant incubated with primary antibody (5mg) or matched IgG control (5mg) overnight at 4°C. Appropriate secondary antibodies were then added and immunoprecipitated an additional hour at 4°C. Immune complexes were pelleted by centrifugation (1000g, 5 min. at 4˚C), washed three times with binding buffer, and immunoprecipitated proteins eluted by boiling with 2x Laemmli buffer (125mM Tris-HCl (pH6.8 at 25˚C), 4% w/v SDS, 20% v/v glycerol, 10% v/v β-mercaptoethanol, 0.004% w/v bromophenol blue). Purity of nuclear and cytoplasmic extracts was verified by immunoblot analysis of cytoplasmic and nuclear for pCAF and β-Tubulin. Cell lysates and eluted complexes were separated by 8% w/v SDS-PAGE and immunoblotted with respective antibodies as described in (Lauffart et al, 2002).

C. Indirect immunofluorescence microscopy

Cells were prepared as previously described (Lauffart et al, 2002). Asynchronous cell populations on coverslips were incubated with the rabbit α-TACC3 (1/500 dilution) or rabbit α-TACC2 (1/100 dilution) polyclonal antibody together with either goat α-BARD1 (4mg/ml), mouse α-BRCA1 (2.5mg/ml) or mouse α-Ku70 (5mg/ml) antibody for 1 h at room temperature. The coverslips were washed with PBS (3 x 5 min.) and incubated with a mixture of 5mg/ml rhodamine red-X-conjugated α-rabbit IgG (to detect TACC2 or TACC3) and 7.5 mg/ml FITC-conjugated α-goat IgG (to detect BARD1) or 6 mg/ml FITC-conjugated α-mouse IgG (for BRCA1) for 30 min. For colocalization studies with Ku70, the TACC3 primary antibody was detected with FITC-conjugated α-rabbit IgG (5mg/ml), and the Ku70 primary antibody detected with rhodamine red-X-conjugated α-mouse IgG (3mg/ml). Finally, the coverslips were washed first with PBS containing DAPI (4',6'-diamidino-2-phenylindole hydrochloride) (1mg/ml) and then twice with PBS, mounted with AquaPolyMount (Polysciences Inc., Warrington, PA, USA) and examined at 60x magnification (oil).

D. siRNA-mediated downregulation of TACC3 and DNA damage survival analysis

The 21 nucleotide siRNA used to target human TACC3 corresponds to nt83-103 of the TACC3 open reading frame (identical to that previously used in (Gergely et al, 2003)), with a dTdT overhang. This siRNA was synthesized and HPLC purified by Integrated DNA technologies, MD, USA. siRNA controls were obtained from Dharmacon, USA. All siRNAs were manipulated according to manufacturer’s instructions.

For DNA damage analysis, cells were plated to 60% confluency in triplicate, into 24 well plates, 24h prior to transfection. Duplicated sets of controls were similarly plated for RNA and protein analysis. Cells were transfected with 100nM siRNA to TACC3 or 100nM control siRNA (Dharmacon) in the presence of 5nM siGLO (Dharmacon) using Oligofectamine (Invitrogen) according to manufacturer’s instructions. 48h post transfection, cells were exposed to 2mg/ml of adriamycin or dimethyl sulphoxide vehicle control. After a further 24h, culture medium was removed and viability of 200 cells per well measured using Trypan blue exclusion. Percentage survival was determined relative to the siRNA (siControl or siTACC3) treated cells incubated with DMSO. This experiment was repeated twice and differences in survival analyzed using one way ANOVA followed by Bonferroni’s Multiple Comparison Post Test (Graphpad Prism Version 3.0, Graphpad Prism Software Inc.). For each experiment, the duplicate sets of cell controls were harvested for RNA and protein as previously described (Still and Cowell 1998, Lauffart et al, 2006). cDNA synthesis was performed using iScript (Biorad, USA), according to manufacturer’s instructions and used for subsequent PCR to confirm downregulation of TACC3 in the absence of activation of 2’5’-oligoadenylate synthetase (data not shown). Efficiency of reduction of TACC3 protein expression in siRNA treated cells was determined by immunoblot analysis relative to b-tubulin as a loading control, as described above.

III. Results

Based upon the data from the large-scale proteomic databases, we predicted that one or more human TACC orthologues would interact with BARD1, as well as other associated DNA repair complexes (summarized in Figure 1). To test this hypothesis, we performed coimmunoprecipitation experiments using well-characterized antibodies to human BARD1 in T47D breast cancer cells and A2780 serous ovarian cancer cells that

TACC2/3 with BRCA1/BARD (Figure 2) and the predicted interaction between TACC and double-stranded DNA break repair complexes (Figure 1), we next tested, whether the TACCs, and TACC3 in particular, are intimately involved with DNA repair. First we investigated a potential interaction between the TACCs and the Ku70/Ku80 heterodimer, a major player in the recognition of double-stranded DNA breaks and the recruitment of the DNA-dependent protein kinase prior to repair (Koike, 2002). Both TACC2 and TACC3 colocalize by indirect immunohistochemistry (Figure 3A) and are specifically immunoprecipitated by the Ku70 component of DNA-dependent protein kinase (Figure 3B).

Based on the presence of TACC3 in BRCA1/BARD and Ku70 containing complexes in the nucleus of ovarian cancer cell lines, we next examined the effect of reducing TACC3 levels on the survival of ovarian cancer cell lines in response to double-strand DNA breaks caused by the topoisomerase II inhibitor adriamycin. siRNA-mediated repression of TACC3 significantly increased the sensitivity of A2780 to the chemotherapeutic agent compared to cells exposed to the vehicle (DMSO) control (Figure 3C). However, no effect was observed in the adriamycin-resistant, p53-null ovarian cancer cell line SkOV3. As indicated in Figure 3C, these correlations may have a mechanistic basis, because we were able to detect TACC3 (and TACC2), but not TACC1 in p53 immunoprecipitates from A2780 (Figure 3D) and MCF7 cells (data not shown). Thus, a physical interaction between normal TACC3 and normal p53 may be an important mediator for establishing cell cycle arrest and repair of DNA damage in ovarian cancer cells.

IV. Discussion

Tumorigenesis is a complex, multistage process. In general, unbiased systematic analysis of tumors by high throughput techniques at the level of the genome, transcriptome and more importantly the proteome is demonstrating that molecular mechanisms behind the development and progression of tumors are many and varied. While key highly penetrant genetic factors have been identified from analysis of familial and hereditary cancer, most tumors are classified as sporadic. These latter tumors result from an initially random accumulation of environmentally caused genetic damage that eventually mutate or deregulate the pathways governing the normal growth/differentiation of tissues (Anderson et al, 2001; Feki and Irminger-Finger, 2004; Olivier et al, 2004; Turner et al, 2004; Gullo et al, 2006).

The consistent observation of loss or exclusion of the TACC3 protein from the nucleus in resected ovarian tumors (Lauffart et al, 2005) suggested that normal nuclear TACC3 function is compromised during the transition from normal to malignant cells in vivo. Although most investigations have centered on the role of TACC3 during mitosis, the interphase role of TACC3 in mammals is only gradually being elucidated (Sadek et al, 2000; Gangisetty et al, 2004; Garriga-Canut and Orkin 2004; Angrisano et al, 2006; Lauffart et al, 2007). In this report, we extrapolated proteomic data from the high throughput analysis of model organisms, and demonstrated for the first time the presence of TACC2 and TACC3 in nuclear localized complexes containing BARD1, BRCA1 and Ku70. Each of these proteins is a targets for functional inactivation in tumors (Turner et al, 2004; Gullo et al, 2006). Mutations, aberrant pre-mRNA splicing and abnormal nuclear-cytoplasmic translocation of BARD1 are also observed in breast, ovarian and lung cancer (Wu et al, 2006). Similarly, TACC proteins have been variably associated with tumor status in many different types of cancer (Conte et al, 2002; Line et al, 2002; Lauffart et al, 2005; Jung et al, 2006), although as with most of the proteins involved in DNA damage and repair, TACC mutation/deregulation is more likely to contribute to tumor progression, than be solely responsible for tumor initiation. In light of the interactions between TACC3 and BARD1, it is particularly intriguing that both proteins are frequently translocated from the nucleus to the cytoplasm in tumor cells in vivo (Lauffart et al, 2005; Jung et al, 2006; Wu et al, 2006). The nuclear-cytoplasmic shuffling of BARD1 is linked to its BRCA1-independent proapoptotic activities (Jefford et al, 2004). Recently, we have demonstrated that overexpression of TACC1 can block the translocation of phospho-ERK to the nucleus of serum stimulated cells, suggesting that the TACCs are directly or indirectly involved in nuclear-cytoplasmic shuttling (Lauffart et al, 2007). Thus, TACC3 may be involved in the correct partitioning of key proteins to specific subcellular locations in normal and malignant cells.

We next examined the potential functional role of these physical interactions by determining the effect of siRNA-mediated knockdown of TACC3 on the survival of ovarian cancer cell lines in response to double-strand DNA breaks caused by the topoisomerase II inhibitor adriamycin. Similar to the previously noted association of loss of function of BRCA1 and increased chemosensitivity to adriamycin (Kennedy et al, 2004), siRNA-mediated repression of TACC3 significantly increased the sensitivity of A2780 to the chemotherapeutic agent compared to cells exposed to the vehicle (DMSO) control (Figure 3C). This is consistent with the observation that siRNA-mediated downregulation of the ceTAC gene produces a him phenotype, characteristic of a defect in normal DNA repair (Boulton et al, 2004). For TACC3, this sensitization effect seemed to be dependent on functional p53 as no significant effect was observed in the adriamycin-resistant, mutant p53 ovarian cancer cell line SkOV3. Previous analysis of the TACC3 knockout mouse (Piekorz et al, 2002) had also indicated that TACC3 is genetically linked to p53-mediated apoptosis. This was suggested by the upregulation of p21 in the fetal liver of TACC3 knockout embryos, and the partial rescue of the TACC3 embryonic lethality in a p53 heterozygous background. Furthermore, correlative immunohistochemical analysis of non-small cell lung carcinomas suggested that TACC3-negative tumor cells might initiate a p53-mediated apoptotic signal (Jung et al, 2006). As indicated in Figure 3, these correlations may have a mechanistic basis based on the presence of TACC3 in p53 immunoprecipitates from A2780 (Figure 3D) and

Figure 3. Coimmunoprecipitation analysis of native TACCs and native DNA response/ repair proteins in ovarian cancer cell lines. A: Colocalization of TACC2 and TACC3 with Ku70 in the nucleus of MCF7 and A2780 respectively. Top panels: Indirect immunofluorescence using the TACC2 antibody and a rhodamine red-X-labeled secondary antibody (red) shows that this protein is predominantly located in the nucleus of theses interphase cells. Similarly Ku70 (green) is concentrated in the nucleus of MCF7. Bottom panels: Endogenous TACC3 is detected with fluorescein isothiocyanate (FITC)-conjugated secondary antibody in the nucleus of A2780, with some staining of the cytoplasm. A similar distribution of Ku70 is detected with the rhodamine red-X-conjugated secondary antibody. Colocalization of Ku70 and TACC2/3 is revealed by the yellow/green signal in the nuclei of MCF7 and A2780 cells. MCF7 and A2780 cells were prepared as previously described in (Gangisetty et al, 2004). B: Extracts were taken from asynchronous cultures and coimmunoprecipitations performed under stringent conditions as described above. TACC2 and TACC3 are able to interact with Ku70 in asynchronous A2780 cells. Although TACC1 is expressed in these cells, it does not seem to be in a stable complex with Ku70. NUC: nuclear extract, IP: specific antibody recognizing target protein used for immunoprecipitation, WB: western blot antibody. Immunoprecipitation of the original target protein was confirmed by Western blot, and the purity of nuclear vs. cytoplasmic extracts verified (see Figure 2B). C: (i) Sensitization of ovarian cancer cells to adriamycin (ADR) in absence of TACC3. A2780 and SkOV3 were treated with 100nM siRNA to TACC3 (siT3) or 100nM control siRNA (siC) (Dharmacon, USA) in the presence of 5nM siGLO (transfection efficiency control) for 48hrs prior to treatment with 2 mg/ml adriamycin (ADR) or DMSO (vehicle control). Viability was determined after 24 hours by Trypan blue exclusion with percentage survival determined relative to the paired siRNA-treated, vehicle control. Knockdown of TACC3 sensitized p53 wild type A2780 cells to DNA damage by adriamycin relative to the DMSO control (p<0.01). This effect was not noted in SkOV3. Values show mean ± standard error from three experiments (ii): Confirmation of knockdown of TACC3 in siT3-treated cells, compared to control siC-treated cells. SiT3 siRNA significantly reduced expression of the TACC3 protein in A2780 and SkOV3 cells. No effect on TACC3 expression was noted in siC-treated cells. b-tubulin expression was used as a loading control. D: Whole cell extracts were prepared from asynchronous cultures and coimmunoprecipitations performed under stringent conditions as described above. TACC2 and TACC3 are able to interact with p53 in asynchronous A2780 cells. TACC1 is not apparent in p53 immunoprecipitates. NUC: nuclear extract, IP: specific antibody recognizing target protein used for immunoprecipitation, WB: western blot antibody. Immunoprecipitation of the original target protein was confirmed by Western blot.

MCF7 cells (data not shown). Thus, a physical interaction between normal TACC3 and normal p53 may be an important mediator for establishing cell cycle arrest and repair of DNA damage. If TACC3 is lost in the presence of normal p53, however, cell death can be triggered, preventing an accumulation of genetic errors. Recently, Schneider et al (Schneider et al, 2007) have shown that siRNA-mediated depletion of TACC3 can also overcome activation of antiapoptotic pathways induced by the mitotic spindle poison paclitaxel, triggering apoptosis in NIH3T3 cells (Schneider et al, 2007). However, our observation in SkOV3 cells indicates that loss of TACC3 does not appear to increase cell death in response to DNA damage in cells without normal p53. Thus, these data suggest that functional loss of TACC3 may promote tumor initiation/progression by altering DNA damage responses in addition to those caused by TACC3-mediated defects in centrosomal duplication and function.

In summary, we have identified new connections between the TACCs and key genetic players in the etiology of cancer. This provides an additional role of the TACCs, and TACC3, in particular in the maintenance of a stable genome, beyond their role in centrosome maturation and mitotic spindle assembly. As sporadic cancer is due to the interplay between low penetrance genetic factors and exogenous environmental factors, further investigation of the TACCs in the pathways initiated by DNA damaging, mutagenic environmental agents may be worthwhile.

Acknowledgements

This work was supported in part by developmental funds support from Arkansas Tech University, Russellville, AR and the Roswell Park Cancer Institute, Buffalo, NY and by US Army Medical Research grant DAMD17-01-1-0208. The authors have no competing financial interests.

References

Aitola M, Sadek CM, Gustafsson JA, Pelto-Huikko M (2003) Aint/Tacc3 is highly expressed in proliferating mouse tissues during development, spermatogenesis, oogenesis. J Histochem Cytochem 51, 455-469.

Anderson GR, Stoler DL, Brenner BM (2001) Cancer: the evolved consequence of a destabilized genome. Bioessays 23, 1037-1046.

Angrisano T, Lembo F, Pero R, Natale F, Fusco A, Avvedimento VE, Bruni CB, Chiariotti L (2006) TACC3 mediates the association of MBD2 with histone acetyltransferases and relieves transcriptional repression of methylated promoters. Nucleic Acids Res 34, 364-372.

Bader GD, Betel D, Hogue CW (2003) BIND: the Biomolecular Interaction Network Database. Nucleic Acids Res 31, 248-250.

Boulton SJ, Martin JS, Polanowska J, Hill DE, Gartner A, Vidal M (2004) BRCA1/BARD1 Orthologs Required for DNA Repair in Caenorhabditis elegans. Curr Biol 14, 33-39.

Conte N, Charafe-Jauffret E, Delaval B, Adelaide J, Ginestier C, Geneix J, Isnardon D, Jacquemier J, Birnbaum D (2002) Carcinogenesis and translational controls: TACC1 is down-regulated in human cancers and associates with mRNA regulators. Oncogene 21, 5619-5630.

Conte N, Delaval B, Ginestier C, Ferrand A, Isnardon D, Larroque C, Prigent C, Seraphin B, Jacquemier J, Birnbaum D (2003) TACC1-chTOG-Aurora A protein complex in breast cancer. Oncogene 22, 8102-8116.

Feki A and Irminger-Finger I (2004) Mutational spectrum of p53 mutations in primary breast and ovarian tumors. Crit Rev Oncol Hematol 52, 103-116.

Gangisetty O, Lauffart B, Sondarva GV, Chelsea DM, Still IH (2004) The transforming acidic coiled coil proteins interact with nuclear histone acetyltransferases. Oncogene 23, 2559-2563.

Garriga-Canut M and Orkin SH (2004) Transforming acidic coiled-coil protein 3 (TACC3) controls friend of GATA-1 (FOG-1) subcellular localization and regulates the association between GATA-1 and FOG-1 during hematopoiesis. J Biol Chem 279, 23597-23605.

Gayther SA, Russell P, Harrington P, Antoniou AC, Easton DF, Ponder BA (1999) The contribution of germline BRCA1 and BRCA2 mutations to familial ovarian cancer: no evidence for other ovarian cancer-susceptibility genes. Am J Hum Genet 65, 1021-1029.

Gergely F (2002) Centrosomal TACCtics. Bioessays 24, 915-925.

Gergely F, Draviam VM, Raff JW (2003) The ch-TOG/XMAP215 protein is essential for spindle pole organization in human somatic cells. Genes Dev 17, 336-341.

Gullo C, Au M, Feng G, Teoh G (2006) The biology of Ku and its potential oncogenic role in cancer. Biochim Biophys Acta 1765, 223-234.

Irminger-Finger I and Leung WC (2002) BRCA1-dependent and independent functions of BARD1. Int J Biochem Cell Biol 34, 582-587.

Irminger-Finger I, Leung WC, Li J, Dubois-Dauphin M, Harb J, Feki A, Jefford CE, Soriano JV, Jaconi M, Montesano R, Krause KH (2001) Identification of BARD1 as mediator between proapoptotic stress and p53-dependent apoptosis. Mol Cell 8, 1255-1266.

Jacquemier J, Ginestier C, Rougemont J, Bardou VJ, Charafe-Jauffret E, Geneix J, Adelaide J, Koki A, Houvenaeghel G, Hassoun J, Maraninchi D, Viens P, Birnbaum D, Bertucci F (2005) Protein expression profiling identifies subclasses of breast cancer and predicts prognosis. Cancer Res 65, 767-779.

Jefford CE, Feki A, Harb J, Krause KH, Irminger-Finger I (2004) Nuclear-cytoplasmic translocation of BARD1 is linked to its apoptotic activity. Oncogene 23, 3509-3520.

Jung CK, Jung JH, Park GS, Lee A, Kang CS, Lee KY (2006) Expression of transforming acidic coiled-coil containing protein 3 is a novel independent prognostic marker in non-small cell lung cancer. Pathol Int 56, 503-509.

Kennedy RD, Quinn JE, Mullan PB, Johnston PG, Harkin DP (2004) The role of BRCA1 in the cellular response to chemotherapy. J Natl Cancer Inst 96, 1659-1668.

Kim JM, Sohn HY, Yoon SY, Oh JH, Yang JO, Kim JH, Song KS, Rho SM, Yoo HS, Kim YS, Kim JG, Kim NS (2005) Identification of gastric cancer-related genes using a cDNA microarray containing novel expressed sequence tags expressed in gastric cancer cells. Clin Cancer Res 11, 473-482.

Kinoshita K, Noetzel TL, Pelletier L, Mechtler K, Drechsel DN, Schwager A, Lee M, Raff JW, Hyman AA (2005) Aurora A phosphorylation of TACC3/maskin is required for centrosome-dependent microtubule assembly in mitosis. J Cell Biol 170, 1047-1055.

Koike M (2002) Dimerization, translocation and localization of Ku70 and Ku80 proteins. J Radiat Res (Tokyo) 43, 223-236.

Lauffart B, DiMatteo A, Vaughan MM, Cincotta MA, Black JD, Still IH (2006) Temporal and spatial expression of TACC1 in the mouse and human. Dev Dyn 235, 1638-1647.

Lauffart B, Gangisetty O, Still IH (2003) Molecular cloning, genomic structure and interactions of the putative breast tumor suppressor TACC2. Genomics 81, 192-201.

Lauffart B, Howell SJ, Tasch JE, Cowell JK, Still IH (2002) Interaction of the transforming acidic coiled-coil 1 (TACC1) protein with ch-TOG and GAS41/NuBI1 suggests multiple TACC1-containing protein complexes in human cells. Biochem J 363, 195-200.

Lauffart B, Sondarva GV, Gangisetty O, Cincotta M, Still IH (2007) Interaction of TACC proteins with the FHL family: implications for ERK signaling. J Cell Communication and Cell Signaling 1, 5-15.

Lauffart B, Vaughan MM, Eddy R, Chervinsky D, Dicioccio RA, Black JD, Still IH (2005) Aberrations of TACC1 and TACC3 are associated with ovarian cancer. BMC Womens Health 5, 8.

Lehner B, Fraser AG (2004) A first-draft human protein-interaction map. Genome Biol 5, R63

Line A, Slucka Z, Stengrevics A, Li G, Rees RC (2002) Altered splicing pattern of TACC1 mRNA in gastric cancer. Cancer Genet Cytogenet 139, 78-83.

Ludwig T, Chapman DL, Papaioannou VE, Efstratiadis A (1997) Targeted mutations of breast cancer susceptibility gene homologs in mice: lethal phenotypes of Brca1, Brca2, Brca1/Brca2, Brca1/p53, Brca2/p53 nullizygous embryos. Genes Dev 11, 1226-1241.

McKeveney PJ, Hodges VM, Mullan RN, Maxwell P, Simpson D, Thompson A, Winter PC, Lappin TR, Maxwell AP (2001) Characterization and localization of expression of an erythropoietin- induced gene, ERIC-1/TACC3, identified in erythroid precursor cells. Br J Haematol 112, 1016-1024.

Olivier M, Hussain SP, Caron dF, Hainaut P, Harris CC (2004) TP53 mutation spectra and load: a tool for generating hypotheses on the etiology of cancer. IARC Sci Publ 247-270.

Piekorz RP, Hoffmeyer A, Duntsch CD, McKay C, Nakajima H, Sexl V, Snyder L, Rehg J, Ihle JN (2002) The centrosomal protein TACC3 is essential for hematopoietic stem cell function and genetically interfaces with p53-regulated apoptosis. EMBO J 21, 653-664.

Raff JW (2002) Centrosomes and cancer: lessons from a TACC. Trends Cell Biol 12, 222-225.

Sadek CM, Jalaguier S, Feeney EP, Aitola M, Damdimopoulos AE, Pelto-Huikko M, Gustafsson J (2000) Isolation and characterization of AINT: a novel ARNT interacting protein expressed during murine embryonic development. Mech Dev 97, 13-26.

Sadek CM, Pelto-Huikko M, Tujague M, Steffensen KR, Wennerholm M, Gustafsson JA (2003) TACC3 expression is tightly regulated during early differentiation. Gene Expr Patterns 3, 203-211.

Schneider L, Essmann F, Kletke A, Rio P, Hanenberg H, Schulze-Osthoff K, Nurnberg B, Piekorz RP (2007) TACC3 depletion sensitizes to paclitaxel-induced cell death and overrides p21(WAF)-mediated cell cycle arrest. Oncogene Jun 25 [Epub ahead of print].

Schreiber E, Matthias P, Muller MM, Schaffner W (1989) Rapid test for in vivo stability and DNA binding of mutated octamer binding proteins with 'mini-extracts' prepared from transfected cells. Nucleic Acids Res 17, 6419-

Schuendeln MM, Piekorz RP, Wichmann C, Lee Y, McKinnon PJ, Boyd K, Takahashi Y, Ihle JN (2004) The centrosomal, putative tumor suppressor protein TACC2 is dispensable for normal development, deficiency does not lead to cancer. Mol Cell Biol 24, 6403-6409.

Stewart JP, Thompson A, Santra M, Barlogie B, Lappin TR, Shaughnessy J, Jr. (2004) Correlation of TACC3, FGFR3, MMSET and p21 expression with the t(4;14)(p16.3;q32) in multiple myeloma. Br J Haematol 126, 72-76.

Still IH and Cowell JK (1998) The t(8;13) atypical myeloproliferative disorder: further analysis of the ZNF198 gene and lack of evidence for multiple genes disrupted on chromosome 13 [letter]. Blood 92, 1456-1458.

Still IH, Hamilton M, Vince P, Wolfman A, Cowell JK (1999a) Cloning of TACC1, an embryonically expressed, potentially transforming coiled coil containing gene, from the 8p11 breast cancer amplicon. Oncogene 18, 4032-4038.

Still IH, Vettaikkorumakankauv AK, DiMatteo A, Liang P (2004) Structure-function evolution of the transforming acidic coiled coil genes revealed by analysis of phylogenetically diverse organisms. BMC Evol Biol 4, 16

Still IH, Vince P, Cowell JK (1999b) The third member of the transforming acidic coiled coil-containing gene family, TACC3, maps in 4p16, close to translocation breakpoints in multiple myeloma, is upregulated in various cancer cell lines. Genomics 58, 165-170.

Turner N, Tutt A, Ashworth A (2004) Hallmarks of 'BRCAness' in sporadic cancers. Nat Rev Cancer 4, 814-819.

Walhout AJ, Sordella R, Lu X, Hartley JL, Temple GF, Brasch MA, Thierry-Mieg N, Vidal M (2000) Protein interaction mapping in C. elegans using proteins involved in vulval development. Science 287, 116-122.

Walhout AJ and Vidal M (2001) Protein interaction maps for model organisms. Nat Rev Mol Cell Biol 2, 55-62.

Werness BA, Ramus SJ, Whittemore AS, Garlinghouse-Jones K, Oakley-Girvan I, Dicioccio RA, Tsukada Y, Ponder BA, Piver MS (2000) Histopathology of familial ovarian tumors in women from families with and without germline BRCA1 mutations. Hum Pathol 31, 1420-1424.

Wu JY, Vlastos AT, Pelte MF, Caligo MA, Bianco A, Krause KH, Laurent GJ, Irminger-Finger I (2006) Aberrant expression of BARD1 in breast and ovarian cancers with poor prognosis. Int J Cancer 118, 1215-1226.

From left to right: Omkaram Gangisetty, Ivan Still, Brenda Lauffart

Research Article

C. Indirect immunofluorescence microscopy

Acknowledgements

References