Oncogene (2011) 30, 3649–3660 & 2011 Macmillan Publishers Limited All rights reserved 0950-9232/11 www.nature.com/onc ORIGINAL ARTICLE YB-1 evokes susceptibility to cancer through cytokinesis failure, mitotic dysfunction and HER2 amplification

AH Davies1, I Barrett2, MR Pambid1,KHu1, AL Stratford1, S Freeman3, IM Berquin4, S Pelech5,6, P Hieter2, C Maxwell7 and SE Dunn1

1Laboratory of Oncogenomic Research, Departments of Pediatrics and Experimental Medicine, Child and Family Research Institute, University of British Columbia, Vancouver, British Columbia, Canada; 2Michael Smith Laboratories, University of British Columbia, Vancouver, British Columbia, Canada; 3Department of Microbiology and Immunology, University of British Columbia, Vancouver, British Columbia, Canada; 4Department of Cancer Biology, Wake Forest University School of Medicine, Winston-Salem, NC, USA; 5The Brain Research Centre, Division of Neurology, University of British Columbia, Vancouver, British Columbia, Canada; 6Kinexus Bioinformatics Corporation, Vancouver, British Columbia, Canada and 7Department of Pediatrics, University of British Columbia, Vancouver, British Columbia, Canada

Y-box binding -1 (YB-1) expression in the Introduction mammary gland promotes breast carcinoma that demon- strates a high degree of genomic instability. In the present Cancer arises from the progressive evolution of a cell study, we developed a model of pre-malignancy to from normalacy through intermediate pre-malignant characterize the role of this gene during breast cancer states to finally become invasive and metastasize initiation and early progression. Antibody microarray (Hanahan and Weinberg, 2000). Cells position them- technology was used to ascertain global changes in signal selves for this progression to malignancy by accumulat- transduction following the conditional expression of YB-1 ing genetic alterations that result in the activation of in human mammary epithelial cells (HMEC). Cell cycle- oncogenes and inactivation of tumor suppressors. associated were frequently altered with the most Studies of human mammary epithelial cells (HMEC) dramatic being LIM 1/2 (LIMK1/2). Conse- are beginning to provide key insight into these early quently, the misexpression of LIMK1/2 was associated genetic events to ascertain their role in fueling breast with cytokinesis failure that acted as a precursor to carcinogenesis (Romanov et al., 2001; Tlsty et al., 2004). centrosome amplification. Detailed investigation revealed Y-box binding protein-1 (YB-1) is a transcription/ that YB-1 localized to the centrosome in a phosphory- translation factor that is overexpressed in a plethora of lation-dependent manner, where it complexed with peri- cancers, including human breast carcinoma (40%) (Wu centrin and c-tubulin. This was found to be essential in et al., 2006; Habibi et al., 2008). A pro-tumorigenic role maintaining the structural integrity and microtubule for YB-1 is supported by its ability to directly bind Y-box nucleation capacity of the organelle. Prolonged exposure promoter elements of a variety of genes, notably epidermal to YB-1 led to rampant acceleration toward tumorigen- growth factor receptor (EGFR), ErbB2 (HER2), A esis, with the majority of cells acquiring numerical and and cyclin B1 (Jurchott et al., 2003; Wu et al., 2006; structural chromosomal abnormalities. Slippage through Stratford et al., 2007). Moreover, a function for YB-1 in the G1/S checkpoint due to overexpression of cyclin E regulating cell cycle progression is beginning to emerge promoted continued proliferation of these genomically through its ability to alter the expression of genes involved compromised cells. As malignancy further progressed, we at the G1/S boundary (Basaki et al.,2010;Yuet al., 2010). identified a subset of cells harboring HER2 amplification. The transcriptional activity of YB-1 is dependent upon Our results recognize YB-1 as a cancer susceptibility phosphorylation at its Ser-102 residue mediated by Akt/ gene, with the capacity to prime cells for tumorigenesis. PKB, and even more potently by p90 ribosomal S6 kinase Oncogene (2011) 30, 3649–3660; doi:10.1038/onc.2011.82; (RSK) (Sutherland et al., 2005; Stratford et al., 2008). published online 21 March 2011 To establish the importance of YB-1 in malignant trans- formation, transgenic mice were developed where expres- Keywords: YB-1; premalignancy; centrosome; cell cycle; sion was targeted to the lactating mammary gland HER2 amplification; breast cancer (Bergmann et al., 2005). The resulting mouse mammary tumors formed with 100% penetrance, and close examina- tion revealed substantial centrosome amplification and chromosomal instability (Bergmann et al., 2005). Given these findings, concurrent with the high prevalence of Correspondence: Dr SE Dunn, Departments of Pediatrics, University YB-1 in breast cancer, we hypothesized that it has an of British Columbia, 950 West 28th Avenue, Room 3083, Vancouver, essential role in breast tumorigenesis. British Columbia, Canada V5Z 4H4. Genomic instability, in the form of alterations to E-mail: [email protected] Received 10 October 2010; revised 4 February 2011; accepted 14 chromosome number and structure, is a characteristic February 2011; published online 21 March 2011 feature of almost all types of cancer (Nigg, 2002; YB-1 is a cancer susceptibility gene AH Davies et al 3650 Fukasawa, 2007; Holland and Cleveland, 2009; Nigg Table 1 Cell cycle-associated proteins putatively regulated by YB-1 and Raff, 2009). However, whether this represents Protein Antibody Localization % CFCa a cause or consequence of tumorigenesis remains mysterious. To address this contentious issue and begin LIMK1/2 Y507 þ T508/Y504 þ T505 Centrosome 365 to establish a paradigm for malignant transformation, ZAP70 Y315 þ Y319 Centrosome 230 it has become imperative to study cancer during the RSK1/2 S380/S386 Kinetochore 269 CDK1/2 Y15 Centrosome 161 earliest pre-malignant stages, which, to date, have Cdc34 Pan-specific Mitotic spindle 148 remained wholly uncharacterized. A large body of Cofilin Pan-specific Centrosome 135 evidence has recently been compiled indicating p53 Pan-specific Cytoplasm/nucleus 118 that amplification of centrosomes has the potential to CDK9 Pan-specific Nucleus 116 PP2A Pan-specific Centrosome 102 cause mitotic defects that lead to chromosomal insta- ZAP70 Y292 Centrosome 102 bility (Nigg, 2006; Basto et al., 2008; Ganem et al., CDK6 Pan-specific Centrosome 98 2009). According to this model, centrosome abnormal- PKA Pan-specific Centrosome À67 ities would need to emerge early during neoplastic progression. Ultimately, through Darwinian selection, a a%CFC refers to the percentage change from control (uninduced karyotype adept at enhancing tumor progression would HTRY cells). materialize and expand (Fujiwara et al., 2005; Shi and King, 2005). Of all sporadic breast cancer, 20–30% including HER2, strongly supported the fidelity of the exhibit amplification of HER2, prompting us to address screen. We prioritized subsequent analysis on the active if this is a common feature of pre-malignancy that arises LIM (pLIMK1/2T508/T505), as they exhibited the through targeted genomic instability preceding clonal most profound increase in level (365%) following YB-1 outgrowth (Slamon et al., 1987). induction. This correlation was validated both by In this study, we examined the role of YB-1 during immunoblotting (Figure 1a) and immunofluorescence pre-malignancy to uncover the molecular events that staining (Figure 1b). In a reciprocal experiment, define the earliest transitions in breast cancer initi- silencing YB-1 with small interfering RNA (siRNA) in ation and progression. A comprehensive understanding MDA-MB-231 and SUM149 breast cancer cell lines of these processes will usher the development of novel repressed the phosphorylation of LIMK1/2 at Thr-508/ therapeutics that target the process, rather than the Thr-505, with minimal effect on total LIMK1 expression consequences, of tumorigenesis. (Figure 1c). To confirm these results, we utilized a complimentary pharmacological approach for suppres- sion of YB-1 activity using a inhi- bitor to RSK. Our prior work indicated that inhibition Results of RSK directly impaired YB-1 phosphorylation and activity (Stratford et al., 2008). Treating MDA-MB-231 YB-1 alters the expression and activity of cell cells with the RSK inhibitor BI-D1870 (a gift cycle-associated proteins from Ching-Shih Chen, The Ohio State University, To address the potential contribution of YB-1 in the Columbus, OH, USA) yielded complete suppression of initiation of tumorigenesis, we engineered non-malignant pLIMK1/2T508/T505, highlighting that pYB-1S102 was H16N2 HMECs that conditionally expressed the gene necessary to promote LIMK1/2 activation (Figure 1d). under control of a tetracycline-inducible promoter These data were mirrored using siRNA against RSK1 (designated HTRY cells). HMECs containing an induci- and RSK2 (Supplementary Figure S2). Previous reports ble LacZ construct served as a matched control implicated LIMK1/2 as centrosomal proteins (Sumi (designated HTRZ cells). The ectopic expression level et al., 2006; Chakrabarti et al., 2007) and, accordingly, of YB-1 achieved in this model closely recapitulated that we wanted to examine the localization in our HTRY cell observed in established cancer cell lines (Supplementary model. At 96-h post-YB-1 induction, we observed Figure S1A). Further characterization revealed that both punctate pLIMK1/2T508/T505 staining that corresponded cell lines were karyotypically normal (data shown below) to the centrosome as demonstrated by co-localization and possessed similar activity deeming them with the centrosomal marker g-tubulin (Figure 1e). genetically stable, and thus amenable for investigating early transformation (Supplementary Figure S1B). To gain a global understanding into proteome Cytokinesis failure primes cells for pre-malignant remodeling following YB-1 induction, we utilized the transformation Kinex antibody microarray platform (Kinexus Bioinfor- With YB-1 regulating a myriad of cell cycle-associated matics Corporation, Vancouver, Canada), which al- genes, we wondered how cells from a non-malignant lowed us to probe the expression and activation status background would respond to expression of the gene. (levels of phosphorylation) of over 600 proteins concur- One of the earliest and most remarkable changes in the rently. From this unbiased protein array, we identified HTRY cells following YB-1 induction was the strikingly 56 proteins with altered expression (Supplementary high incidence of multinucleated cells (Figure 2a). At Table S1) many of which are fundamental in regulating 48 h following YB-1 induction, which corresponded centrosome dynamics and the cell cycle (Table 1). roughly to the doubling time of these cells (data not Identification of known YB-1 transcriptional targets, shown), 28% of HTRY cells were binucleate thus

Oncogene YB-1 is a cancer susceptibility gene AH Davies et al 3651

Figure 1 YB-1 altered the activity of the centrosomal protein LIMK1/2. (a) Immunoblotting validated the efficacy of the HTRY cell line to express YB-1 following a 96-h induction with doxycycline. Concurrent with YB-1 expression was an increase in active pLIMK1/2T508/T505. The HTRZ cells did not express YB-1 and exhibited reduced pLIMK1/2T508/T505 expression. (b) These observations were recapitulated in immunofluorescence staining of HTRZ and HTRY cells with pLIMK1/2T508/T505 antibody (green). (c) siYB-1 was used to silence protein expression in MDA-MB-231 and SUM149 cells, and at 96-h post-transfection both total YB-1 and the pool of active pLIMK1/2T508/T505 were depleted, as demonstrated by immunoblotting. (d) Similarly, treatment of MDA-MB-231 cells for 24 h with BI-D1870 (1 or 10 mM) completely inhibited phosphorylation of Ser-102 of YB-1 and Thr-508/Thr-505 of LIMK1/2, as shown by immunoblotting. (e) It was demonstrated that pLIMK1/2T508/T505 (green) co-localized with the centrosomal marker g-tubulin (red) in HTRY cells, as visualized by immunofluorescence. indicating a failure of cytokinesis. This was compared formed a tight contractile ring with PLK1 characte- with only 5% of HTRZ cells (Figure 2b). ristically localized to the midzone between the dividing Previous work has demonstrated that LIMK1 cells. This supports our hypothesis that YB-1 promotes localizes to the cleavage furrow during cytokinesis cytokinesis failure through deregulation and subsequent (Sumi et al., 2006), and its overexpression is associated mislocalization of LIMK1/2. with polyploidy (Amano et al., 2002). Based on its role in modulating dynamics (Bernard, 2007), a likely explanation for the defect in cytokinesis might, there- Cell cycle checkpoint slippage potentiates centrosome fore, be that deregulation of LIMK1/2 perturbs the amplification leading to aneuploidy stability of the contractile ring. In cytokinetic HTRZ Cytokinesis failure can lead to both centrosome cells, pLIMK1/2T508/T505 was concentrated in the junction amplification and production of tetraploid cells, which between the two daughter cells as observed by immuno- could set the stage for the development of tumor cells fluorescence. Accordingly, F-actin was visualized, using (Fujiwara et al., 2005). We examined whether YB-1 phalloidin, along the cleavage furrow and at the nuclear expression allowed for cells arrested in cytokinesis periphery (Figure 2c). On the other hand, in cytokinetic to slip through cell cycle checkpoints and re-enter HTRY cells, pLIMK1/2T508/T505 was strongly expressed mitosis with multiple centrosomes and a compromised but remained diffused throughout the cytoplasm genome. At 96-h post-YB-1 induction, HTRY cells (Figure 2c). Consequently, the actin cytoskeleton failed demonstrated centrosome amplification leading to to reposition itself for cytokinesis. Another well-esta- multipolar spindle formation in mitosis (Figure 3a). blished protein at the cleavage furrow, polo-like kinase 1 Consequently, kinetochore bi-orientation was not estab- (PLK1), correctly localized in HTRY cells during lished resulting in failed segregation, which manifested telophase; however, the actomyosin contractile ring as lagging chromosomes at the metaphase plate and consistently failed to form in these cells (Supplementary micronuclei (Figure 3a). Quantifying the proportion of Figure S3). This was in contrast to HTRZ cells, which tetraploid cells (44N DNA content) and those with

Oncogene YB-1 is a cancer susceptibility gene AH Davies et al 3652

Figure 2 Pre-malignancy was initiated by pLIMK1/2T508/T505 mislocalization leading to cytokinesis failure. (a) HTRZ and HTRY cells were induced with doxycycline for 48 h, and immunostained with a-tubulin antibody (green) to define cell boundaries. Visually, many HTRY cells were binucleate (arrows) and, thus, (b) we quantified the proportion of HTRZ and HTRY cells displaying this phenotype. In all, 200 cells were assessed in three independent experiments (**Po0.01). (c) Immunofluorescence staining was employed to evaluate the spatial localization of pLIMK1/2T508/T505 (red) and phalloidin (green) in cytokinetic HTRZ and HTRY cells following a 48-h induction with doxycycline. pLIMK1/2T508/T505 was concentrated in the cleavage furrow of HTRZ cells promoting stabilization of the actomyosin contractile ring (arrow). In HTRY cells, pLIMK1/2T508/T505 remained diffused throughout the cytoplasm, and the contractile ring failed to form (arrow).

supernumerary centrosomes (42 centrosomes) revealed into the 184-hTERT cell line. The parental cells, which a significant increase of 4.5-fold and 12.2-fold, respec- express no YB-1, have been extensively characterized tively, between the HTRY and HTRZ cells (Figure 3b). to be chromosomally stable with a nearly normal Deeper interrogation uncovered that the amplified karyotype (Raouf et al., 2005). Transient expression centrosomes contained an excess of mother, but not of YB-1 for 96 h was sufficient to promote polyploidy daughter, centrioles (Supplementary Figure S4A). and centrosome amplification (Figure 3c). We also Specifically, the 1:1 mother:daughter centriole ratio observed early indicators of genomic instability, includ- observed in the HTRZ cells approached 3:1 in the ing lagging chromosomes and micronuclei (Figure 3c). HTRY cells (Supplementary Figure S4B). Having The faithful recapitulation of phenotypes observed established the importance of pYB-1S102 at the centro- between the HTRY and 184-hTERT-YB-1 cells indi- some, it is not surprising that the described phenotype cates that YB-1 expression alone is sufficient to drive was contingent upon YB-1 Ser-102 phosphorylation, as genomic instability without the requirement for p53 transient expression of YB-1S102D in HTRZ cells could and Rb deregulation. recapitulate the phenotype, whereas YB-1S102A could One would expect that given the centrosome ampli- not (Supplementary Figure S5A). Moreover, the fication coupled with aneuploid DNA content that the incidence of aneuploidy and centrosome amplification YB-1 induced cells would be subject to cell cycle arrest. was most profound in YB-1S102D-overexpressing MDA- To our surprise, only 18% of HTRY cells were classified MB-231 cells by a significant margin (Supplementary as being in the G1-phase of the cell cycle based on DNA Figure S5B). content. This was in stark contrast to 66% of HTRZ Due to HPV-16 E6/E7 immortalization, the p53 and cells (Figure 3d), strongly supporting the notion that retinoblastoma (Rb) tumor suppressor genes are inacti- HTRY cells resist anti-proliferative signals and slip vated in the HTRZ and HTRY cells. To ascertain if through the G1/S checkpoint. Consistent with our this background was necessary to generate the abnormal findings of a cytokinesis defect, 62% of HTRY cells phenotypes observed in HTRY cells, we transfected YB-1 were in G2/M-phase compared with 14% of HTRZ cells

Oncogene YB-1 is a cancer susceptibility gene AH Davies et al 3653

Figure 3 Centrosome amplification and aneuploidy emerged as a consequence of cytokinesis failure and cell cycle checkpoint slippage. (a) At 96-h post-YB-1 induction, HTRY cells demonstrated an amplification of centrosomes promoting multipolar spindle formation during mitosis and, as a result, lagging chromosomes and micronuclei (arrows). 46-diamidino-2-phenyl indole (DAPI; blue) was used to visualize the nuclei, a-tubulin (green) labeled microtubules and pericentrin (red) or g-tubulin (red) served as markers for centrosomes. (b) At the same time point, we quantified the extent of polyploidy and centrosome amplification in 500 HTRZ and HTRY cells. DNA content was measured using ArrayScan VTI high-content screening (Thermo Fisher Scientific, Yokohama, Japan) while the number of centrosomes per cell was counted manually. The data represents a compilation of three independent experiments (**Po0.01). (c) Transient transfection of YB-1 into mammary epithelial 184-hTERT cells for 96 h promoted polyploidy and centrosome amplification as measured by nuclear size/DAPI intensity and pericentrin immunofluorescence, respectively. Data were acquired from six random microscope fields and are presented as the mean and standard deviation from three independent experiments (*Po0.05; **Po0.01). Representative images portrayed genomic instability, while immunoblotting confirmed FLAG:YB-1 transgene expression in the 184-hTERT cells. (d) The proportion of HTRZ and HTRY cells in each phase of the cell cycle was ascertained by assessing DNA content using ArrayScan VTI on asynchronous populations of cells induced for 96 h. Variation between three separate experiments is indicated (**Po0.01; ***Po0.001). (e) To determine if cells with G2/M DNA content (4N) were truly in M-phase, we quantified the proportion of HTRZ and HTRY cells positive for pHistone H3S10 immunofluorescence (green). Data were collected from five unique microscope fields and are presented as the mean and standard deviation from three independent experiments (***Po0.001). Representative images are shown. (f) In defining a mechanism for cell cycle checkpoint slippage, immunoblot analysis illustrated how YB-1 expression led to an enhancement in signal transduction through the RSK/p27Kip1 pathway, thereby promoting cyclin E/CDK2 overexpression.

(Figure 3d). To differentiate between cells arrested at inhibitory target of RSK and negative regulator of cyclin the G2 checkpoint and those that have progressed into E/CDK2, was suppressed (Figure 3f). These data demon- M-phase, we quantified pHistone H3S10 by immuno- strate that YB-1 deregulates the cell cycle by altering fluorescence. In agreement with increased DNA content, signal transduction to favor a proliferative program. 23% of HTRY cells were positive for pHistone H3S10, thus truly in mitosis, compared with 3% of HTRZ cells (Figure 3e). This implies that overcoming the cytokinesis Identification of YB-1 as a centrosomal protein defect maybe a rate-limiting step in tumorigenesis. The data described above establish the importance of We next assessed changes in signal transduction to YB-1 in regulating centrosomal proteins, with a role in define a mechanism to explain the observed slippage promoting amplification of the organelle during pre- through the G1/S checkpoint. Following 96 h of YB-1 malignancy. On this basis, we next explored whether induction, strong expression of HER2 was detected YB-1 was itself directly associated with the centrosome. correlative with RSK activation. Accordingly, p27Kip1,an Co-localization between pYB-1S102 and pericentrin, a

Oncogene YB-1 is a cancer susceptibility gene AH Davies et al 3654

Figure 4 pYB-1S102 localized to the centrosomes throughout the cell cycle. (a) Immunofluorescence staining with antibodies against pYB-1S102 (green) and pericentrin (red) demonstrated that pYB-1S102 was localized to the centrosomes in both interphase and mitotic HTRY cells (arrows). pYB-1S102 was predominately nuclear; however, it dissociated from DNA at metaphase and extended along the length of the mitotic spindle. (b) To evaluate the importance of YB-1 Ser-102 phosphorylation for its centrosomal localization, we generated MDA-MB-231 cells stably expressing FLAG:YB-1WT, FLAG:YB-1S102D and FLAG:YB-1S102A protein. Immunoblotting confirmed ectopic expression of the tagged-proteins. (c, d) Quantification of FLAG (green) and pericentrin (red) co-localization in 250 interphase cells revealed that FLAG:YB-1WT and FLAG:YB-1S102D proteins, but not FLAG:YB-1S102A protein, localized to the centrosomes (arrows). Data represents a compilation from three independent experiments (***Po0.001).

centrosomal marker, was observed in both interphase trafficking, retention and/or function of YB-1 at the and mitotic HTRY cells using immunofluorescence centrosome. (Figure 4a). Notably, during metaphase, pYB-1S102 To functionally examine the role of YB-1 at the was expressed along the entire length of the mitotic centrosome, we began by performing co-immunopreci- spindle. In MDA-MB-231 cells, pYB-1S102 was found to pitation experiments, which revealed physical associa- co-localize with pericentrin confirming its association tion between FLAG:YB-1 and the centrosomal proteins, with centrosomes was not unique to our inducible pericentrin and g-tubulin (Figure 5a). We further system, but rather extended to established cancer cell queried whether LIMK1 was part of a YB-1 centro- lines (Supplementary Figure S6A). To validate YB-1 somal complex; however, the two proteins failed to as a bona fide centrosomal protein, we mapped both co-precipitate (data not shown). Next, we silenced YB-1 FLAG:YB-1 (Supplementary Figure S6B) and GFP: in MDA-MB-231 cells using two independent siRNA YB-1 (Supplementary Figure S6C) to the centrosome sequences (siYB-1#1 and siYB-1#2), which in turn, was using antibody directed against the FLAG epitope and found to yield substantial enlargement and morpholo- direct immunofluorescence, respectively. gical changes of the centrosomes as visualized by To ascertain whether phosphorylation was a pre- immunofluorescence targeting pericentrin (Figure 5b). requisite for YB-1 centrosomal localization, we gene- Further investigation uncovered that these centrosomes rated MDA-MB-231 cells stably expressing FLAG: harbored large clusters of g-tubulin ring complexes YB-1WT, FLAG:YB-1S102D and FLAG:YB-1S102A protein (Figure 5b). We quantified a 3.3–4.1-fold increase in (Figure 4b). Double immunofluorescence using anti- centrosome area at 96-h post-siYB-1 transfection FLAG and anti-pericentrin antibodies revealed that relative to mock-transfected cells (Figure 5c). Impor- the centrosomal localization of YB-1 was contingent tantly, because Ser-102 phosphorylation was a necessity upon phosphorylation of the Ser-102 residue. FLAG: for YB-1 to localize at the centrosome, we wanted to YB-1S102D protein, which mimicked constitutively phos- assess if the mere inhibition of protein activity would phorylated YB-1, was detected in 93.5% of centro- be sufficient to alter centrosome structure. In agreement somes. In contrast, the non-phosphorylatable FLAG: with the YB-1 siRNA experiments, treating cells with 1 YB-1S102A protein failed to localize to the centrosomes or 10 mM BI-D1870 for 24 h prompted the emergence of (Figures 4c and d). Collectively, these data provide cells containing enlarged centrosomes with numerous insight into the dependence on phosphorylation for the g-tubulin ring complexes (Supplementary Figure S7A).

Oncogene YB-1 is a cancer susceptibility gene AH Davies et al 3655

Figure 5 YB-1 altered the architecture and microtubule nucleation capacity of centrosomes by directly binding pericentrin and g-tubulin. (a) Co-immunoprecipitation of FLAG:YB-1 with g-tubulin and pericentrin in MDA-MB-231 cells revealed direct association between the proteins. (b) To elucidate a role for YB-1 at the centrosomes, two unique siRNA oligos targeting YB-1 (siYB-1#1 and siYB-1#2) were transfected into MDA-MB-231 cells. At 96-h post-transfection cells were analyzed by immunofluorescence staining with pericentrin (red; upper panel) and g-tubulin (red; lower panel) to assess changes in centrosome organization. Immunoblotting confirmed YB-1 knockdown. (c) Centrosome area was measured using Image Pro Analyzer software (Media Cybernetics, Bethesda, MD, USA) and is represented relative to the non-transfected control. In all, 100 centrosomes from G1-phase cells were measured across three independent experiments (*Po0.05). (d) The function of YB-1 in microtubule nucleation at the centrosome was ascertained using the microtubule regrowth assay. Scrambled and siYB-1-transfected MDA-MB-231 cells were stained with a-tubulin (green) and g-tubulin (red) antibodies at 1, 5 or 10 min after regrowth.

Specifically, the centrosomes increased in area by up to amplification during pre-malignancy, we assessed meta- 2.9-fold relative to the dimethyl sulfoxide (DMSO)-treated phase chromosomes. The majority of uninduced HTRZ controls (Supplementary Figure S7B). In further support, and HTRY cells, as well as induced HTRZ cells, had a we observed increased centrosomal area in MDA-MB-231 normal diploid karyotype. A small subset was classified cells stably expressing YB-1S102A mutant protein (Supple- as ‘near diploid’ (40–52 chromosomes; Figure 6a). In mentary Figure S7C). Finally, to analyze if the changes to stark contrast, 94% of induced HTRY cells were centrosome structure correlated with altered function, we aneuploid (Figure 6a). Further to these numerical performed microtubule-regrowth assays to detect defects in abnormalities, structural chromosome aberrations were centrosome-mediated microtubule nucleation and anchor- readily detected in the induced HTRY cells. Dramatic ing. The assay was used to assess a fundamental parameter increases of X5.5-fold in the appearance of dicentric of centrosome function, that is, the ability to regrow chromosomes and double minutes were observed in microtubules following depolymerization. Displacement of the HTRY spreads compared with those from HTRZ YB-1 from the centrosome following siRNA silencing cells. Most strikingly, there was a 17.6-fold increase in clearly delayed microtubule-regrowth, as asters of short acentric pairs and 13.2-fold increase in acentric frag- microtubules only began to emerge after a 5 min regrowth ments between HTRY and HTRZ spreads, indicating as opposed to 1 min in mock-transfected cells (Figure 5d). that YB-1 promoted extensive chromosome breakage This clearly demonstrates that loss of YB-1 perturbs (Figure 6b). In addition, defective sister chromatid normal centrosome function. Taken together, these results cohesion was detected in the HTRY cells. Almost half provide strong evidence for a previously uncharacterized, of these cells exhibited a lack of primary constriction, yet essential, role for YB-1 at the centrosome. identified by primary constriction gaps (PCGs) between sister chromatids at metaphase, compared with 11% of HTRZ cells (Supplementary Figure S8). Genomic instability arises during pre-malignancy to To address whether genomic instability initiated by generate clones with strong tumorigenic potential YB-1 could promote an optimal karyotypic composition To better understand the aneuploidy and chromosomal for tumorigenesis, the frequency of HER2 amplification instability that emerge as a consequence of centrosome was measured using fluorescence in situ hybridization.

Oncogene YB-1 is a cancer susceptibility gene AH Davies et al 3656

Figure 6 Numerical and structural chromosomal aberrations materialized as a consequence of YB-1 expression. (a, b) Metaphase spreads were performed on HTRZ and HTRY cells following a 96-h induction with doxycycline. Uninduced cells served as controls. (a) The number of chromosomes and (b) structural chromosomal abnormalities were evaluated in 200 metaphase spreads from both HTRZ and HTRY cells. Three independent experiments were performed (*Po0.05; ***Po0.001). Representative images are shown. (c) To determine if a recurrent pattern of amplification at the HER2 locus could be detected, we performed fluorescence in situ hybridization analysis on interphase cells. Following a 96-h induction, HTRZ and HTRY cells were hybridized with HER2 (red) and CEP17 (green) probes. In all, 100 cells were assessed for low-copy and high-copy gene amplification. Representative images are shown.

We uncovered that 11% of HTRY cells were positive for HER2 amplification (HER2:CEP17 42.2). No HTRZ cells exhibited HER2 amplification (Figure 6c). Upon a more rigorous assessment, we noted that 20% of the HTRY cell population contained low-level HER2 ampli- fication (HER2:CEP17 between 1.5 and 2.2; Figure 6c). The HER2 amplification in HTRY cells was largely undetected until they reached tetraploid DNA content. At this time there was an apparent relaxation in the mechanisms safeguarding genomic stability, and the number of HER2 signals began to exceed the number of centromeres, a hallmark of gene amplification (Supple- mentary Figure S9A and S9B). Collectively, these data indicate that a subset of pre-malignant cells have an amplification at the HER2 locus that could enhance their tumorigenic potential. We conclude from our data that YB-1 is instrumental in activating a tumorigenic program that manifests as a cytokinesis defect, and progresses toward the emergence of HER2-positive cancer (Figure 7). From this, we have proposed a model of pre-malignant progression. Figure 7 Proposed model for how YB-1 instigates pre-malig- nancy. Targeted YB-1 expression in non-tumorigenic HMECs prompted a strong enhancement in LIMK1/2 activity that resulted in a cytokinesis defect. Concurrently, YB-1 altered signal Discussion transduction allowing cells to slip through the G1/S checkpoint. The resulting centrosome amplification led to multipolar spindles during mitosis, which promoted aneuploidy. With sustained YB-1 In the present study, we propose a distinctive model of expression, a population of cells containing HER2 amplification breast cancer pre-malignancy whereby YB-1 enables the emerged.

Oncogene YB-1 is a cancer susceptibility gene AH Davies et al 3657 evolution of HMECs toward a tumorigenic fate. A genome. In support of this, deregulation of centro- cytokinesis defect acted as the initial trigger for trans- somal proteins including Aurora A, BRCA1 and PLK1 formation promoting centrosome amplification and all promote genomic instability with eventual cellular aneuploidy, which were potentiated by cell cycle transformation (Scully, 2000; Takai et al., 2005; Wang checkpoint slippage. In turn, we identified a small et al., 2006). population of cells harboring amplification at the HER2 Studies of HMECs have proven effective in providing locus. These studies provide significant insight into the key insights into the early genetic events that fuel breast process of tumor initiation, and demonstrate how YB-1 carcinogenesis (Elenbaas et al., 2001; Romanov et al., alone can initiate a program that primes cells for 2001; Tlsty et al., 2004; Dimri et al., 2005; Dumont tumorigenesis. et al., 2009). In this study, we report that YB-1 expres- Although YB-1 upregulation is well characterized in sion in this model leads to catastrophic genetic changes, breast cancer cell lines and advanced stage primary which if left unchecked could allow for the replication tumors (Janz et al., 2002; Kohno et al., 2003; Habibi of cells containing vast chromosomal amplifications et al., 2008; To et al., 2010), a role for the gene in tumor and rearrangements. Because YB-1-expressing HTRY initiation and pre-malignant progression is unknown. cells fail to arrest following genomic destabilization, Chromosomal aberrations observed in YB-1 transgenic it suggests that mutant cells are able to escape the mice prompted us to address if ectopic YB-1 expression necessary checkpoints needed to eliminate such rene- in genetically stable HMECs acted to directly destabilize gade cells. Permissiveness through the cell cycle could the genome as a prelude to malignancy (Bergmann relate to direct YB-1 transcriptional targets, such as et al., 2005). In this study, we have demonstrated that CCNB1, CDC6, PCNA, and TOPO2 (Shibao et al., expression of the gene promoted gross alterations to the 1999; Jurchott et al., 2003; Basaki et al., 2010; Yu et al., centrosomal milieu and, ultimately, led to centrosome 2010). We chose to address the possibility that YB-1 amplification. Most notably, a strong activation of permits the expansion of cells harboring specific LIMK1/2 was detected at the centrosomes. Likewise, in amplifications common to breast cancer. Notably, we prostate cancer LIMK is expressed at the centrosome describe HER2 as being amplified in a small subset and has been linked to chromosomal instability and of HTRY cells. We speculate that over time this metastasis (Yoshioka et al., 2003; Chakrabarti et al., population of cells would clonally expand due to 2007; Davila et al., 2007). An important finding of this the distinct survival advantage brought about by study was that cytokinesis failure is the predominant HER2 overexpression. Moreover, this study furthers mechanism for the amplification of centrosomes during our understanding of the relationship between YB-1 pre-malignancy. We identified that sustained upregula- and HER2 as previously described by our laboratory tion and mislocalization of active LIMK1/2 by YB-1 (Wu et al., 2006; Lee et al., 2008; Dhillon et al., 2010). was sufficient to induce a cytokinesis defect. This could Our previous studies show that YB-1 directly binds to be attributed to enhanced actin polymerization at the the HER2 promoter in cells where the gene is known cleavage furrow (Yang et al., 2004). Given these results, to be amplified (Wu et al., 2006). One could envisage we propose that YB-1 causes early changes in cytokin- that YB-1 is permissive for allowing cells with HER2 esis and centrosomal architecture that lead to eventual amplification to slip through the cell cycle checkpoints. chromosomal instability. Following this, YB-1 is poised to increase the expression In this work, we have established YB-1 as a of HER2 by binding directly to its promoter. This too centrosomal protein. This was found to be contingent may explain why YB-1 is highly expressed in B65% upon phosphorylation of the Ser-102 residue in the cold of HER2-positive breast tumors (Habibi et al., 2008). shock domain, implying that this domain is minimally Future work will focus on identifying additional geno- required for centrosomal trafficking. Especially interest- mic rearrangements that frequently materialize during ing is the fact that the cold shock domain is necessary pre-malignancy. for binding oligonucleotides, including RNA due to the We report that increased expression of YB-1 is a presence of two RNP motifs (Bouvet et al., 1995). As the single event sufficient to uncouple genomic integrity and centrosome contains an intrinsic compliment of RNA cell cycle progression during breast cancer pre-malig- (Alliegro et al., 2006), it is possible that YB-1 is involved nancy. In summary, our findings argue that YB-1 has in regulating their translation. YB-1 has already been a principal role in the early evolution of cancer, and shown to induce cap-dependent translation of RNA thus represents a promising biomarker and therapeutic giving credence to this hypothesis (Evdokimova et al., target. 2009). A second function for YB-1 at the centrosome may be to mediate protein stability via physical association. It has been demonstrated that YB-1 interacts with a myriad of proteins, including PCNA, Materials and methods MSH2 and DNA d, via B/A repeats residing Cell culture and drug treatments in the C-terminal domain (Ise et al., 1999; Gaudreault H16N2 HMEC with tetracycline-inducible YB-1 (HTRY) et al., 2004). It is tempting to speculate that centro- or LacZ (HTRZ) were generated using the T-Rex system somal proteins may represent an underappreciated (Invitrogen, Burlington, ON, Canada), as previously described pool with strong capacity to promote tumorigenesis, (Band et al., 1990; Berquin et al., 2005). The cells were cultured by their inherent ability to directly interface with the in Ham’s F12 media, and induction was achieved through the

Oncogene YB-1 is a cancer susceptibility gene AH Davies et al 3658 addition of 1 mg/ml doxycycline (Calbiochem, Gibbstown, NJ, Dulbecco’s modified Eagle’s medium. At 1, 5 and 10 min after USA). The human mammary epithelial 184-hTERT cell line regrowth, the cells were fixed with 100% methanol and stained (a gift from Dr J. Carl Barrett, National Institutes of Health, with a-tubulin and g-tubulin antibodies. Bethesda, MD, USA) was cultured in supplemented HuMEC media (Invitrogen). LCC6, MDA-MB-231 (American Tissue Cell cycle analysis culture collection, Manassas, VA, USA) and SUM149 (Asterand, HTRZ and HTRY cells were seeded in 96-well plates and Detroit, MI, USA) breast cancer cell lines were cultured as induced for 96 h. The cells were subsequently fixed using 2% recommended. paraformaldehyde and stained with Hoechst 33 342 (1 ug/ml; For treatment with BI-D1870, MDA-MB-231 cells were 5 Sigma) for 30 min. Based on total nuclear Hoechst intensity, seeded at a density of 3 Â 10 cells in a six-well plate. Subse- the proportion of cells in each stage of the cell cycle was quently, cells were treated with a DMSO vehicle or BI-D1870 analyzed by Cell Cycle Bioapplication software on a high (1 or 10 mM) for 24 h. content screening instrument (ArrayScan VTI, Thermo Fisher Scientific, Yokohama, Japan). Kinexus Kinex antibody mircoarray HTRY cells were induced for 96 h. Comparisons were made to Chromosome spreads cells not treated with doxycycline. Protein was sent to Kinexus Following a 96-h induction, HTRZ and HTRY cells were Bioinformatics Corporation (Vancouver, BC, Canada) for treated with 0.1 mg/ml colcemid (Invitrogen) for 2 h. Mitotic hybridization and analysis using the Kinex KAM-1.1 antibody chromosomes were resuspended in hypotonic solution (75 mM microarray. KCl) for 20 min and fixed using methanol:glacial acetic acid (3:1), as previously described (Barber et al., 2008). Metaphase Immunoblotting, immunoprecipitation and immunofluorescence chromosomes were imaged using a Zeiss Axioplan digital Immunoblotting, immunoprecipitation and immunofluores- imaging microscope and analyzed with Metamorph imaging cence were performed as described previously (Wu et al., 2006; software (Universal Imaging Corp., Downingtown, PA, Stratford et al., 2008; Finkbeiner et al., 2009). The origin and USA). For analysis, we assessed chromosomal abnormalities dilutions of all antibodies used in this study are detailed in based on their incidence with ‘mild’ referring to less than Supplementary Table S2. For immunoprecipitation, 500 mgof 5 occurrences in a spread, ‘moderate’ between 5 and 20, cell lysate was pre-cleared with 35 ml of protein G agarose and ‘severe’ greater than 20. PCGs were defined as a (Sigma, St Louis, MO, USA) before overnight antibody clear separation between DAPI-stained sister chromatids. incubation. The proteins were retrieved through the addition The severity ranged from only 1 or 2 chromosomes in a of protein G agarose for 2 h and eluted in 5 Â SDS-sample- spread exhibiting a gap (PCGI), to between 3 and 10 loading buffer heated to 100 1C for 5 min. For immuno- chromosomes (PCGII), to no semblance of cohesion (PCGIII). fluorescence staining, antibodies were diluted in ICC buffer (10% bovine serum albumin, 2% goat serum and 1% saponin HER2 fluorescence in situ hybridization in phosphate-buffered saline), and all incubations were carried Asynchronous HTRZ and HTRY cells were prepared out at room temperature for 1 h with three washes in phosphate- for chromosome analysis as described above. Interphase cells buffered saline following each of the incubations. Cells were were hybridized with LSI HER2 and CEP17 probe using mounted using ProLong Gold antifade reagent containing the PathVysion HER2 DNA Probe Kit at the Center DAPI (Invitrogen). Images were acquired using an Olympus for Translational and Applied Genomics (Vancouver, BC, FV1000 laser scanning confocal microscope (Olympus, Center Canada). Analysis of fluorescence in situ hybridization signals Valley, PA, USA), a DeltaVision personalDV live cell imaging was performed in 100 randomly selected cells. HER2 ampli- microscope (Applied Precision, Issaquah, WA, USA) or an fication was defined as a HER2:CEP17 ratio of greater than Olympus BX61 epifluorescence microscope, and analyzed 2.2. A HER2:CEP17 ratio o1.5 was considered negative for with ImageJ 1.43 (National Institutes of Health, Bethesda, HER2 amplification, whereas a ratio near the cut-off (1.5–2.2) MD, USA). was interpreted as intermediate amplification.

siRNA and plasmid transfections Telomerase assay Cells were transfected with 20 nM of siRNA to RSK1, RSK2, The telomerase activity in 1 mg of cell lysate from HTRZ and YB-1 or scrambled control using Lipofectamine RNAiMAX HTRY cells was measured using the Quantitative Telomerase (Invitrogen). The siRNA target sequences are provided in Detection Kit (Allied Biotech, Vallejo, CA, USA), following WT S102A Supplementary Table S3. The empty vector, YB-1 , YB-1 the manufacturer’s instructions. Each sample was analyzed S102D and YB-1 constructs have previously been described in triplicate. A no template control and cell lysate from (Sutherland et al., 2005; Wu et al., 2006; Finkbeiner et al., telomerase-positive cells (MDA-MB-231) were included in 2009). Plasmid transfections were performed using 4 mgof each experiment. DNA and carried out with Lipofectamine 2000 (Invitrogen). Stable transfectants were generated and selected in G418 (400 mg/ml; Invitrogen). The GFP:YB-1 construct (Guay Statistical analysis et al., 2006) was transfected into MDA-MB-231 cells by Data from at least three independent experiments are reported electroporation with Amaxa Nucleofactor Kit V, using as mean±standard deviation. Significance was examined using the manufacturer’s recommendations (Lonza, Walkersville, a paired Student’s t-test, where *Po0.05, **Po0.01 and MD, USA). ***Po0.001.

Microtubule regrowth assay SiRNA-transfected MDA-MB-231 cells were treated with Conflict of interest 5 mM nocodazole for 1 h to depolymerize all microtubules. Nocodazole was then removed by washing twice with The authors declare no conflict of interest.

Oncogene YB-1 is a cancer susceptibility gene AH Davies et al 3659 Acknowledgements was supported by the National Institutes of Health RO1 CA114017 (SED, IMB), the Michael Smith Foundation for We thank Ching-Shih Chen (The Ohio State University) for Health Research (AHD), the Canadian Institutes of Health providing BI-D1870 and Michel Lebel (l’Universite Laval) Research (AHD, SED) and the Canadian Breast Cancer for providing plasmid encoding GFP-tagged YB-1. This study Foundation (ALS).

References

Alliegro MC, Alliegro MA, Palazzo RE. (2006). Centrosome-associated Finkbeiner MR, Astanehe A, To K, Fotovati A, Davies AH, Zhao Y RNA in surf clam oocytes. Proc Natl Acad Sci USA 103: 9034–9038. et al. (2009). Profiling YB-1 target genes uncovers a new mechanism Amano T, Kaji N, Ohashi K, Mizuno K. (2002). Mitosis-specific for MET receptor regulation in normal and malignant human activation of LIM motif-containing and roles of mammary cells. Oncogene 28: 1421–1431. cofilin phosphorylation and dephosphorylation in mitosis. J Biol Fujiwara T, Bandi M, Nitta M, Ivanova EV, Bronson RT, Pellman D. Chem 277: 22093–22102. (2005). Cytokinesis failure generating tetraploids promotes tumor- Band V, Zajchowski D, Kulesa V, Sager R. (1990). Human papilloma igenesis in p53-null cells. Nature 437: 1043–1047. virus immortalize normal human mammary epithelial cells Fukasawa K. (2007). Oncogenes and tumour suppressors take on and reduce their growth factor requirements. Proc Natl Acad Sci centrosomes. Nat Rev Cancer 7: 911–924. USA 87: 463–467. Ganem NJ, Godinho SA, Pellman D. (2009). A mechanism linking Barber TD, McManus K, Yuen KW, Reis M, Parmigiani G, Shen D extra centrosomes to chromosomal instability. Nature 460: 278–282. et al. (2008). Chromatid cohesion defects may underlie chromosome Gaudreault I, Guay D, Lebel M. (2004). YB-1 promotes strand instability in human colorectal cancers. Proc Natl Acad Sci USA separation in vitro of duplex DNA containing either mispaired 105: 3443–3448. bases or cisplatin modifications, exhibits endonucleolytic activities Basaki Y, Taguchi K, Izumi H, Murakami Y, Kubo T, Hosoi F et al. and binds several DNA repair proteins. Nucleic Acids Res 32: (2010). Y-box binding protein-1 (YB-1) promotes cell cycle 316–327. progression through CDC6-dependent pathway in human cancer Guay D, Gaudreault I, Massip L, Lebel M. (2006). Formation of a cells. Eur J Cancer 46: 954–965. nuclear complex containing the p53 tumor suppressor, YB-1, and Basto R, Brunk K, Vinadogrova T, Peel N, Franz A, Khodjakov A the werner syndrome gene product in cells treated with UV light. et al. (2008). Centrosome amplification can initiate tumorigenesis in Int J Biochem Cell Biol 38: 1300–1313. flies. Cell 133: 1032–1042. Habibi G, Leung S, Law JH, Gelmon K, Masoudi H, Turbin D et al. Bergmann S, Royer-Pokora B, Fietze E, Jurchott K, Hildebrandt B, (2008). Redefining prognostic factors for breast cancer: YB-1 is a Trost D et al. (2005). YB-1 provokes breast cancer through the stronger predictor of relapse and disease-specific survival than estrogen induction of chromosomal instability that emerges from mitotic receptor or HER-2 across all tumor subtypes. Breast Cancer Res 10:R86. failure and centrosome amplification. Cancer Res 65: 4078–4087. Hanahan D, Weinberg RA. (2000). The hallmarks of cancer. Cell 100: Bernard O. (2007). Lim kinases, regulators of actin dynamics. Int J 57–70. Biochem Cell Biol 39: 1071–1076. Holland AJ, Cleveland DW. (2009). Boveri revisited: chromosomal Berquin IM, Pang B, Dziubinski ML, Scott LM, Chen YQ, Nolan GP instability, aneuploidy and tumorigenesis. Nat Rev Mol Cell Biol et al. (2005). Y-box-binding protein 1 confers EGF independence to 10: 478–487. human mammary epithelial cells. Oncogene 24: 3177–3186. Ise T, Nagatani G, Imamura T, Kato K, Takano H, Nomoto M et al. Bouvet P, Matsumoto K, Wolffe AP. (1995). Sequence-specific RNA (1999). Transcription factor Y-box binding protein 1 binds preferen- recognition by the xenopus Y-box proteins. An essential role for the tially to cisplatin-modified DNA and interacts with proliferating cell cold shock domain. J Biol Chem 270: 28297–28303. nuclear antigen. Cancer Res 59: 342–346. Chakrabarti R, Jones JL, Oelschlager DK, Tapia T, Tousson A, Janz M, Harbeck N, Dettmar P, Berger U, Schmidt A, Jurchott K Grizzle WE. (2007). Phosphorylated LIM kinases colocalize with et al. (2002). Y-box factor YB-1 predicts drug resistance and patient gamma-tubulin in centrosomes during early stages of mitosis. Cell outcome in breast cancer independent of clinically relevant tumor Cycle 6: 2944–2952. biologic factors HER2, uPA and PAI-1. Int J Cancer 97: 278–282. Davila M, Jhala D, Ghosh D, Grizzle WE, Chakrabarti R. (2007). Jurchott K, Bergmann S, Stein U, Walther W, Janz M, Manni I et al. Expression of LIM kinase 1 is associated with reversible G1/S phase (2003). YB-1 as a cell cycle-regulated transcription factor facili- arrest, chromosomal instability and prostate cancer. Mol Cancer 6: tating cyclin A and cyclin B1 gene expression. J Biol Chem 278: 40. 27988–27996. Dhillon J, Astanehe A, Lee C, Fotovati A, Hu K, Dunn SE. (2010). Kohno K, Izumi H, Uchiumi T, Ashizuka M, Kuwano M. (2003). The The expression of activated Y-box binding protein-1 serine 102 pleiotropic functions of the Y-box-binding protein, YB-1. Bioessays mediates trastuzumab resistance in breast cancer cells by increasing 25: 691–698. CD44+ cells. Oncogene 29: 6294–6300. Lee C, Dhillon J, Wang MY, Gao Y, Hu K, Park E et al. (2008). Dimri G, Band H, Band V. (2005). Mammary epithelial cell Targeting YB-1 in HER-2 overexpressing breast cancer cells induces transformation: insights from cell culture and mouse models. Breast apoptosis via the mTOR/STAT3 pathway and suppresses tumor Cancer Res 7: 171–179. growth in mice. Cancer Res 68: 8661–8666. Dumont N, Crawford YG, Sigaroudinia M, Nagrani SS, Wilson MB, Nigg EA. (2006). Origins and consequences of centrosome aberrations Buehring GC et al. (2009). Human mammary cancer progression in human cancers. Int J Cancer 119: 2717–2723. model recapitulates methylation events associated with breast Nigg EA. (2002). Centrosome aberrations: cause or consequence of premalignancy. Breast Cancer Res 11: R87. cancer progression? Nat Rev Cancer 2: 815–825. Elenbaas B, Spirio L, Koerner F, Fleming MD, Zimonjic DB, Nigg EA, Raff JW. (2009). Centrioles, centrosomes, and cilia in health Donaher JL et al. (2001). Human breast cancer cells generated by and disease. Cell 139: 663–678. oncogenic transformation of primary mammary epithelial cells. Raouf A, Brown L, Vrcelj N, To K, Kwok W, Huntsman D et al. Genes Dev 15: 50–65. (2005). Genomic instability of human mammary epithelial cells Evdokimova V, Tognon C, Ng T, Ruzanov P, Melnyk N, Fink D et al. overexpressing a truncated form of EMSY. J Natl Cancer Inst 97: (2009). Translational activation of snail1 and other developmentally 1302–1306. regulated transcription factors by YB-1 promotes an epithelial- Romanov SR, Kozakiewicz BK, Holst CR, Stampfer MR, Haupt LM, mesenchymal transition. Cancer Cell 15: 402–415. Tlsty TD. (2001). Normal human mammary epithelial cells

Oncogene YB-1 is a cancer susceptibility gene AH Davies et al 3660 spontaneously escape senescence and acquire genomic changes. located in the cold shock domain and affects the anchorage- Nature 409: 633–637. independent growth of breast cancer cells. Oncogene 24: 4281–4292. Scully R. (2000). Role of BRCA gene dysfunction in breast and Takai N, Hamanaka R, Yoshimatsu J, Miyakawa I. (2005). Polo-like ovarian cancer predisposition. Breast Cancer Res 2: 324–330. kinases (plks) and cancer. Oncogene 24: 287–291. Shi Q, King RW. (2005). Chromosome nondisjunction yields tetra- Tlsty TD, Crawford YG, Holst CR, Fordyce CA, Zhang J, ploid rather than aneuploid cells in human cell lines. Nature 437: McDermott K et al. (2004). Genetic and epigenetic changes in 1038–1042. mammary epithelial cells may mimic early events in carcinogenesis. Shibao K, Takano H, Nakayama Y, Okazaki K, Nagata N, Izumi H J Mammary Gland Biol Neoplasia 9: 263–274. et al. (1999). Enhanced coexpression of YB-1 and DNA topo- To K, Fotovati A, Reipas KM, Law JH, Hu K, Wang J et al. (2010). II alpha genes in human colorectal carcinomas. Int J Y-box binding protein-1 induces the expression of CD44 and CD49f Cancer 83: 732–737. leading to enhanced self-renewal, mammosphere growth, and drug Slamon DJ, Clark GM, Wong SG, Levin WJ, Ullrich A, McGuire resistance. Cancer Res 70: 2840–2851. WL. (1987). Human breast cancer: correlation of relapse and Wang X, Zhou YX, Qiao W, Tominaga Y, Ouchi M, Ouchi T et al. survival with amplification of the HER-2/neu oncogene. Science (2006). Overexpression of aurora kinase A in mouse mammary 235: 177–182. epithelium induces genetic instability preceding mammary tumor Stratford AL, Fry CJ, Desilets C, Davies AH, Cho YY, Li Y et al. formation. Oncogene 25: 7148–7158. (2008). Y-box binding protein-1 serine 102 is a downstream target of Wu J, Lee C, Yokom D, Jiang H, Cheang MC, Yorida E et al. (2006). p90 ribosomal S6 kinase in basal-like breast cancer cells. Breast Disruption of the Y-box binding protein-1 results in suppression of Cancer Res 10: R99. the epidermal growth factor receptor and HER-2. Cancer Res 66: Stratford AL, Habibi G, Astanehe A, Jiang H, Hu K, Park E et al. 4872–4879. (2007). Epidermal growth factor receptor (EGFR) is transcription- Yang X, Yu K, Hao Y, Li DM, Stewart R, Insogna KL et al. (2004). ally induced by the Y-box binding protein-1 (YB-1) and can be LATS1 tumour suppressor affects cytokinesis by inhibiting LIMK1. inhibited with iressa in basal-like breast cancer, providing a Nat Cell Biol 6: 609–617. potential target for therapy. Breast Cancer Res 9: R61. Yoshioka K, Foletta V, Bernard O, Itoh K. (2003). A role for Sumi T, Hashigasako A, Matsumoto K, Nakamura T. (2006). LIM kinase in cancer invasion. Proc Natl Acad Sci USA 100: Different activity regulation and subcellular localization of 7247–7252. LIMK1 and LIMK2 during cell cycle transition. Exp Cell Res Yu YN, Yip GW, Tan PH, Thike AA, Matsumoto K, Tsujimoto M 312: 1021–1030. et al. (2010). Y-box binding protein 1 is up-regulated in proliferative Sutherland BW, Kucab J, Wu J, Lee C, Cheang MC, Yorida E et al. breast cancer and its inhibition deregulates the cell cycle. Int J Oncol (2005). Akt phosphorylates the Y-box binding protein 1 at Ser102 37: 483–492.

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc)

Oncogene