deubiquitination of g-tubulin by bap1 prevents chromosome ...46 51 56 61 66 71 76 81 86 91 96 101...

Molecular and Cellular Pathobiology

Deubiquitination of g-Tubulin by BAP1 PreventsChromosome Instability in Breast Cancer Cells

Reihaneh Zarrizi1, Julien Albert Menard2, Mattias Belting2,3, and Ramin Massoumi1

AbstractMicrotubule nucleation requires the g-tubulin ring complex, and during the M-phase (mitosis) this complex

accumulates at the centrosome to support mitotic spindle formation. The posttranslational modification ofg-tubulin through ubiquitination is vital for regulating microtubule nucleation and centrosome duplication.Blocking the BRCA1/BARD1-dependent ubiquitination of g-tubulin causes centrosome amplification. In thecurrent study, we identifiedBRCA1-associated protein-1 (BAP1) as a deubiquitination enzyme for g-tubulin. BAP1was downregulated in metastatic adenocarcinoma breast cell lines compared with noncancerous human breastepithelial cells. Furthermore, low expression of BAP1 was associated with reduced overall survival of patientswith breast cancer. Reduced expression of BAP1 in breast cancer cell lines was associated with mitoticabnormalities. Importantly, rescue experiments including expression of full length but not the catalytic mutantof BAP1 reduced ubiquitination of g-tubulin and preventedmitotic defects. Our study uncovers a newmechanismfor BAP1 involved in deubiquitination of g-tubulin, which is required to prevent abnormal mitotic spindleformation and genome instability. Cancer Res; 74(22); 6499–508. �2014 AACR.

IntroductionPosttranslational modification of proteins by covalent

attachment of ubiquitin controls many essential cellular pro-cesses by targeting the proteins for assembly into complexes,transport, and degradation (1, 2). The type of polyubiquitinchains, including lysine 48 (K48) versus lysine 63 (K63), ormonoubiquitination can induce different signaling pathwaysthat regulate different aspects of cellular functions. The mostprominent function of polyubiquitination chains through K48linkage degrades the substrate through the mechanism of theproteasome, whereas K63-linked ubiquitin chains add newfunctional properties to the modified protein, except forproteosomal degradation (1, 2). Monoubiquitination or mul-tiple monoubiquitination generally leads to transcription reg-ulation as well as DNA repair and cell-cycle progression (1, 2).The action of ubiquitin ligation enzymes responsible for theattachment of the ubiquitin moieties to the substrate isreversed by the action of deubiquitination enzymes (DUB;ref. 3). The human genome encodes a large number of putativeDUBs, and accumulating evidence indicates that most of these

enzymes regulate a limited number of proteins (4). The dysre-gulation of DUB function is highly related to human diseases,especially cancer, with examples of hyperactivation or inacti-vation due to genetic or epigenetic changes. In this aspect, it isvital to identify substrate-specific DUBs to understand thesignaling pathways that are dysregulated in cancer cells.Although it is becoming increasingly clear that DUBs playcritical roles in various cellular pathways, physiologic roles formost of the human DUBs remain unknown.

BRCA1-associated protein-1 (BAP1) is a nuclear-localizedDUB that was originally identified as a BRCA1-binding proteinin a yeast two-hybrid screen (5). BAP1 is ubiquitouslyexpressed in different human tissues, especially in the testis,placenta, and ovaries, with varying levels detected in othertissues, including human breast. In mice, BAP1 expression isupregulated in the breast during puberty and pregnancy (6).BAP1 belongs to the ubiquitin C-terminal hydrolase (UCH)family of DUBs, and the activity of BAP1 is restricted to the N-terminal UCH domain, which executes binding and cleavage ofthe ubiquitin isopeptide bond (5). The BAP1 gene is frequentlymutated in lung, breast, melanoma, and renal cell carcinoma(5, 7) and overexpression of BAP1 in lung cancer cells reducesboth tumor formation inmice and cell growth in culture. It hasthus been suggested that BAP1 acts as a tumor suppressor gene(6). The tumor suppressor property of BAP1 is dependent on itsnuclear localization and deubiquitin activity (8). In MCF7breast cancer cell lines, it was shown that the binding of BAP1to breast cancer type 1 susceptibility protein (BRCA1)enhances BRCA1-mediated growth suppression (5). However,in contrast to this result, it has been shown that growthsuppression is independent of BRCA1 and that BAP1 disruptsthe BRCA1/BRCA1-associated ring domain 1 (BARD1) hetero-dimer, but cannot reverse its autoubiquitination (6, 9, 10).

1Department of LaboratoryMedicine, Translational Cancer Research, LundUniversity, Medicon Village, Lund, Sweden. 2Department of ClinicalSciences,SectionofOncology-Pathology, LundUniversity, Lund,Sweden.3Department of Oncology, Ska

�ne University Hospital, Lund, Sweden.

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

Corresponding Author: Ramin Massoumi, Department of LaboratoryMedicine, Molecular Tumor Pathology, Lund University, Medicon village,Scheelev€agen 8, SE-223 63 Lund 33381, Sweden. Phone: 0046-4622-26430; Fax: 0046-4033-7063; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-14-0221

�2014 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 6499

on December 29, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst September 16, 2014; DOI: 10.1158/0008-5472.CAN-14-0221

http://cancerres.aacrjournals.org/

Therefore, the exact role of BAP1 in BRCA1-mediated signalingand functions remain unknown. Host cell factor C1 (HCF-1) isanother binding partner for BAP1. HCF-1 is a transcriptionalcofactor found in a number of important regulatory complexesand it localizes to chromatin. BAP1 was shown to regulateHCF-1 protein expression, leading to changes in gene expres-sion involved in the G1–S transition and cell proliferation (11).

The eukaryotic function of g-tubulin is highly conserved andincludes regulating microtubule nucleation and centrosomeduplication (12). Ubiquitination of g-tubulin at lysines 48 and344, achieved by the activity of the heterodimeric tumorsuppressor complex BRCA1/BARD1, was shown to be impor-tant for microtubule nucleation at the centrosome and cen-trosomal amplification (13–16). As BRCA1/BARD1 promotesubiquitination of g-tubulin, the DUBs responsible for reversingthis process are not known.

In the current study, we identify BAP1 as a specific DUB forg-tubulin. Deubiquitination of g-tubulin is an important eventin mitosis that regulates mitotic spindle organization andprevents genomic instability.

Materials and MethodsCell culture

The human breast cell lines were cultured for 5 days at37�C and 5% CO2 as follows: MCF10A cells were cultured inDMEM/F12 (GIBCO, Life Technologies), supplemented with5%horse serum (SigmaAldrich), 20 ng/mLEGF (SigmaAldrich),100 ng/mL cholera toxin (Sigma Aldrich), 0.01 mg/mL insulin,500 ng/mL hydrocortisone (Sigma Aldrich), and 1% penicillin/streptomycin (GIBCO, Life Technologies). The breast cancercells (MCF7, MDA-MB-231, andMDA-MB-468) were cultured inRPMI (HyClone, Thermo Scientific), supplemented with 10%FBS (SigmaAldrich), 0.1% sodiumpyruvate (SigmaAldrich), and0.1% penicillin/streptomycin (GIBCO, Life Technologies).

Immunofluorescence and confocal microscopyCellswere cultured on coverslips in 6-well plates for 24 hours

and then fixed in paraformaldehyde fixation (4% for 15 min-utes). Next, the cells were permeabilized with 0.3% Triton X-100 solution, after which they were blocked with 1% BSA and5% goat serum for 30 minutes to prevent nonspecific antibodybinding, and subsequently incubated for 1 hour with primaryantibodies against BAP1 (Santa Cruz Biotechnology), a-tubu-lin (SigmaAldrich), g-tubulin (SigmaAldrich), andUCHL1 (CellSignaling Technology). After washing coverslips, fluorescentantibodies (Alexa 488 goat anti-mouse or Alexa 546 goat anti-rabbit; Invitrogen) were applied and then the coverslips werewashed and mounted in Vectashield with 40,6-diamidino-2-phenylindole (DAPI; Vector Laboratories). Fluorescence imagewere captured and processed using a confocal microscope.

Cell-cycle synchronizationSynchronized cells were obtained by double treatment with

2mmol/L thymidine (Sigma-Aldrich) in tissue culturemediumfor 12 to 16 hours. Cells were washed and incubated in normalmedium for 8 to 10 hours to release them from the block. Cellstreated in this manner were further enriched for mitotic phaseby treatment with 40 ng/mL nocodazole (Sigma Aldrich) for 12

to 16 hours.Mitotic cells werewashed in PBS and released fromthe nocodazole block by addition of fresh medium (0–180minutes). Cells were fixed and stained with DAPI.

Retrovirus production and transductionRetroviral vector alone or vector encoding wild-type FLAG-

HA–tagged BAP1 (plasmid 22539) carrying a puromycin-resis-tant gene for the selection of stably transduced cells wereobtained fromAddgene. Retrovirus production was performedby a research engineer at Vector Unit in LundUniversity (Lund,Sweden). MDA-MB-468 cells were seeded in 6-well plateswith 40% confluency in RPMI medium (HyClone, ThermoScientific,) supplemented with 10% FBS, 2 mmol/L L-gluta-mine, 100 U/mL penicillin, and 100 mg/mL streptomycin(Sigma Aldrich). Next day, medium was changed with 1 mLmedium containing hexadimethrine bromide (Sigma Aldrich;final concentration 8 mg/mL) to each well. Ten microliters ofretroviral particles were added to appropriate wells and cellswere incubated for 24 hours at 37�C in a humidified incubatorin an atmosphere of 5%–7% CO2. Medium was changed with 1mL RPMI containing puromycin (Sigma Aldrich). The trans-duction efficiency was initially evaluated and checked usingWestern blot anaylsis and immunofluorescence staining. Thecells were useddirectly for experimentswithout picking clones.

In vitro deubiquitination assayIn vitro deubiquitination assay was carried out by immu-

noprecipitation of wild-type or catalytic inactive mutant ofBAP1 (BAP1-C91A) using anti-FLAGM2-agarose beads (SigmaAldrich). Beads were incubated with cell extracts for overnightunder constant rotation. Before elution, beads were washedfour times with TBS-T. The amount of BAP1 protein in thecrude extracts was estimated by immunoblot using anti-FLAGantibody (Sigma Aldrich). The Ub-g-tubulin was incubated in areaction buffer (50 mmol/L Tris pH 7.5, 150 mmol/L NaCl, 2mmol/L EDTA pH 8.0, 2 mmol/L dithiothreitol) with elutedBAP1 at 37�C for 2 hours. Products were resolved by SDS-PAGEand analyzed by immunoblotting with ubiquitin antibody.

siRNA DUB screenMCF10A cells were grown in 6-well plates to 30%–50%

confluency and then transfected with human DUBs siRNAlibrary (Thermo Scientific) at 2 mmol/L using interferin (Poly-plus Transfection). After 72 hours, cells were lysed with radio-immunoprecipitation buffer. Lysates were subjected to SDS-PAGE followed by immunoblotting with indicated antibodies.

BAP1 mutagenesisActive-site (C91A) mutation was introduced into BAP1-

FLAG by using QuikChange Site-Directed Mutagenesis (Stra-tagene) and the following primers: C91A-BAP1-forward, 50-CACCAGCTGATACCCAACTCTGCTGCAACTCATGCCTTGC-TGAG-30 and reverse, 50-CTCAGCAAGGCATGAGTTGCAGCA-GAGTTGGGTATCAGCTGGTG-30.

Live cell imagingCell divisions were followed live by real-time scanning laser

microscopy. Cells expressing GFP-tagged H2B protein were

Zarrizi et al.

Cancer Res; 74(22) November 15, 2014 Cancer Research6500




seeded in 4-well or 8-well chamber slides (Lab-Tek, ThermoScientific) in phenol red–free DMEM (HyClone, Thermo Sci-entific) supplemented with 10% FBS, 2 mmol/L L-glutamine,100 U/mL penicillin, and 100 mg/mL streptomycin (SigmaAldrich). Cells were allowed to attach overnight in a humidifiedincubator (37�C, 5% CO2) and analyzed the next day using aZeiss LSM 710 confocal scanning microscope (Carl Zeiss) in atemperature (37�C) and atmosphere (21% O2, 5% CO2) con-trolled chamber, on a heated stage insert (37�C). Live imagingof division stages were followed using 488 nm laser excitation,493–598 emission filters, and captured with an EC Plan-Neo-fluar 40�/1.30 oil DIC M27 objective (Carl Zeiss) scanningevery 30 seconds over the division time.

Quantitative PCRThe cells were rinsed in cold PBS and total RNA extracted

using the Perfect Pure RNA tissue Kit (5 PRIME) according tothemanufacturer's instruction. The purity of RNAwas analyzedand quantified by a NanoDrop spectrophotometer (SaveenWerner) and used for cDNA synthesis according to the man-ufacturer's instruction (QPCR cDNA Synthesis Kit, Stratagene).PCR runs were performed in the Mx3005P real-time thermo-cycler (Stratagene) with the following program; 2 minutes in50�C, 10 minutes in 95�C followed by 40 three-step cyclesconsisting of 95�C for 20 seconds and 60�C for 30 seconds and72�C for 1 minute. The following primers were used:

BAP1 Forward 50-GACCCAGGCCTCTTCACC-30; BAP1 Reverse50-AGTCCTTCATGCGACTCAGG-30; GAPDH Forward 50-TG-CACCACCAACTGCTTAGC-30; GAPDH Reverse 50-GGCATGG-ACTGTGGTCATGAG-30.

Statistical analysisData are presented as mean � SD or �SEM. Statistical

comparisons were assessed by ANOVA or by Student t test (P <0.05).

ResultsIdentification of BAP1 in regulating g-tubulindeubiquitination

In the current study, we aimed to identify the specific DUBresponsible for deubiqitination of g-tubulin using the non-cancerous human breast epithelial cell line MCF10A.MCF10A cells were used because g-tubulin can bind andundergo ubiquitination via BRCA1 in breast cells (15);indeed, we found that BRCA1 was colocalized with g-tubulinduring prophase in MCF10A cells, but not during metaphase,anaphase, or telophase (Supplementary Fig. S1). Screening asiRNA library of DUBs, we identified BAP1 and UCHL1, byobserving an additional band corresponding to the possiblediubiquitination of g-tubulin (Fig. 1A and B). Double deple-tion of BAP1 and UCHL1 resulted in ubiquitination ofg-tubulin to the same extent as single depletion (Fig. 1B,

6

5

4

3

2

1

0

6

5

4

3

2

1

0

27

100

7055

40

BAPI

1 6 11 16 21 26 31 36 41siRNA library of human deubiquitinating enzymes

46 51 56 61 66 71 76 81 86 91 96 101

UCHL1

A

C

B

BAPI

γ-Tub

Actin

UCHL1

siC

trl

siCtrl

siC

OP

S5

siD

UB

1Asi

CO

XoR

F53

siB

AP

1

siBAP1

siU

CH

L1/B

AP

1si

UC

HL1

Ub-

γ-tu

bulin

rel

ativ

e de

nsity

Ub-

γ-tu

bulin

rel

ativ

e de

nsity

Figure 1. Identification of BAP1 in regulating g-tubulin deubiquitination using a siRNA screening approach. A, quantification of endogenous ubiquitinatedg-tubulin and total g-tubulin in MCF10A cells treated with a siRNA library of DUBs for 3 days. The data are expressed as mean � SD of two separateexperiments. B, Western blot analysis of MCF10A cells transfected with human DUBs siRNA (from DUBs siRNA library; Thermo Scientific) againstcontrol, COPS5, DUB1A, COXoRF53, BAP1, UCHL1, and double transfection with siRNA against BAP1 and UCHL1 for 3 days. The cells were lysed andsubjected to SDS-PAGE followed by immunoblotting with antibodies against g-tubulin, UCHL1, BAP1, and actin. C, quantification of endogenousubiquitinated g-tubulin and total g-tubulin in MCF10A cells treated for 3 days with siRNA oligos against BAP1 or a control (siCtrl). The data are expressed asmean � SD of four separate experiments. �, P < 0.05.

BAP1 Control Mitotic Spindle Organization

www.aacrjournals.org Cancer Res; 74(22) November 15, 2014 6501




last lane). In addition, we found that BAP1 but not UCHL1created a connection network with g-tubulin and BRCA1(Supplementary Fig. S2). Deconjugation of ubiquitin fromg-tubulin was also confirmed by siRNA oligonucleotidesagainst BAP1 (Fig. 1C).

Downregulation of BAP1 leads to increased g-tubulinubiquitination

Initially, we investigated the total levels of BAP1 in humanbreast cancer cells (MCF-7, MDA-MB-231, and MDA-MB-468) and noncancerous human breast epithelial cells(MCF10A). BAP1 was significantly downregulated at themRNA (Fig. 2A) and protein (Fig. 2B) levels in metastaticadenocarcinoma of the breast cell line MDA-MB-231, where-as another metastatic adenocarcinoma cell line MDA-MB-468 expressed even less BAP1 compared with nonmetastaticMCF7 or MCF10A cells. The subcellular distribution of

endogenous BAP1 by cellular fractionation and confocalmicroscopy showed both cytoplasmic and nuclear localiza-tion of BAP1 in MCF10A cells (Supplementary Fig. S3). Toinvestigate whether BAP1 expression is altered in breastcancer, we studied the association of BAP1 levels with breastcancer patient outcomes. Kaplan–Meier survival analysisusing a microarray dataset from 2,978 patients with breastcancer revealed that high BAP1 expression was significantlyassociated with better overall survival (P ¼ 3.4e�7; Fig. 2C).Furthermore, high BAP1 expression in tissue samples frombasal and luminal subtype tumors were significantly asso-ciated with prolonged progression-free survival (Fig. 2D–F).Together, in vitro data and patient tumor data suggest thatBAP1 expression is inversely correlated with tumor cellaggressiveness.

Next, we hypothesized that loss of BAP1 in MDA-MB-468leads to an increase in g-tubulin ubiquitination. Notably,

BA

P1

mR

NA

leve

ls(n

orm

aliz

ed v

s. G

AP

DH

leve

ls)

1

0.5

100 BAP1

Actin

Luminal A

Luminal B Basal-like

40

MC

F10

A

MC

F7

MD

A-M

B-2

31

MD

A-M

B-4

68

MC

F10

A

MC

F7

MD

A-M

B-2

31

MD

A-M

B-4

68

0

Low

0 5 10 15 20

Time (years)

0 5 10

A B

C D

E F

15

Time (years)

0 5 10 15 20

Time (years)

0 5 10 15

Time (years)

HR=0.72 (0.64–0.82)log-rank P = 3.4e–07

HR=0.69 (0.57–0.85)log-rank P = 0.00031

HR=0.81 (0.65–1)log-rank P = 0.049

HR=0.69 (0.52–0.92)log-rank P = 0.0098

HighLowHigh

LowHigh

LowHigh

Pro

gres

sion

-fre

e su

rviv

al

1.0

0.8

0.6

0.4

0.2

0.0

Pro

gres

sion

-fre

e su

rviv

al

1.0

0.8

0.6

0.4

0.2

0.0

Pro

gres

sion

-fre

e su

rviv

al

1.0

0.8

0.6

0.4

0.2

0.0

Pro

gres

sion

-fre

e su

rviv

al1.0

0.8

0.6

0.4

0.2

0.0

Figure 2. BAP1 expression inbreast cancer has a prognosticimplication. A, relativeBAP1mRNAexpression, measured usingquantitative real-time reversetranscription PCR (qRT-PCR) ofcDNA fromMCF10A, MCF7, MDA-MB-231, and MDA-MB-468 cells.The data are expressed as mean�SEM of four separate experiments.�,P < 0.05. B,Western blot analysisof BAP1 and actin expression inMCF10A, MCF7, MDA-MB-231,and MDA-MB-468 cells. C,Kaplan–Meier overall survivalanalysis of BAP1 mRNAexpression using a microarraydataset from 2,978 patients withbreast cancer from the publiclyavailable expression database forbreast cancer (GOBO; http://co.bmc.lu.se/gobo). D–F, high BAP1mRNA expression correlates withprolonged progression-freesurvival in luminal A (n ¼ 1,463; D),luminal B (n ¼ 805; E), and basal(n ¼ 402; F) breast tumor subtypesusing amicroarray dataset from thepublicly available expression database for breast cancer (GOBO;http://co.bmc.lu.se/gobo).

Zarrizi et al.





although total levels of g-tubulin were similar in the differentcell lines, a slow migrating band above the g-tubulin bandcould be observed only in the MDA-MB-468 cell line(Fig. 3A). Endogenous immunoprecipitation by using anti-bodies against g-tubulin (Fig. 3B) or ubiquitin (Fig. 3C)showed a physical association of ubiquitin with g-tubulin.To confirm the posttranslational modification of g-tubulin,we used urea-treated denatured cell extracts for g-tubulinimmunoprecipitation and detected ubiquitin by immuno-blotting (Fig. 3D).

Interaction between BAP1 and g-tubulin occurs duringmetaphase, anaphase, and telophase

To further investigate the specific cell-cycle stage where theinteraction betweenBAP1orUCHL1 and g-tubulin occurs, cellswere synchronized in G1, S, and G2–M cell-cycle phase. Endog-enous immunoprecipitation of g-tubulin in MCF10A cellsprecipitated BAP1 during mitosis and UCHL1 during G1 cell-cycle phase (Supplementary Fig. S4A). In addition, we foundthat BAP1 at the M-phase and UCHL1 at the G1 phase of cellcycle reduce the levels of ubiquitin-associated tubulin com-pared with S-phase (Supplementary Fig. S4A) in MCF10A cells.The interaction between BAP1 and g-tubulin at the M phasewas also confirmed in two other breast cancer cell lines,including CAMA and T47D (Supplementary Fig. S5A-B). Asexpected, downregulation of BAP1 in these cell lines elevatedthe levels of g-tubulin ubiquitination (Supplementary Fig. S5C).This suggests that BAP1 and UCHL1 both act as DUBs forg-tubulin but in different phases of the cell cycle. Furthermore,interphase- and mitosis-synchronized cells confirmed theinteraction between BAP1 and g-tubulin at the M-phase (Fig.4A). As the levels of BAP1 are low inMDA-MB-468 cells (Fig. 2Aand 2B), we virally infected these cells with BAP1 (Supplemen-tary Fig. S6). Consistent with the MCF10A data, immunopre-cipitation of overexpressed BAP1 in MDA-MB-468 cells pre-cipitated g-tubulin during the process ofmitosis (Fig. 4B) and invitro pull-down experiments using g-tubulin and BAP1 recom-binant proteins showed a direct interaction between these twoproteins (Fig. 4C). Furthermore, immunofluorescence stainingand confocalmicroscopy experiments showed colocalization ofendogenous g-tubulin and BAP1 during metaphase, anaphase,and telophase but not during prophase in MCF10A cells (Fig.4D).Overexpression ofBAP1 inMDA-MB-468 cells also resultedin a colocalization between exogenous BAP1 and g-tubulinduring metaphase, anaphase, and telophase (Fig. 4E). Consis-tent with the confocal microscopy results, endogenous BAP1(Fig. 4F) or overexpressed BAP1 (Fig. 4G) was efficiently coim-munoprecipitated with g-tubulin during metaphase (M1), ana-phase (M2), and telophase (M3) but not during prophase (M0)in cell extracts prepared from different stages of mitosis.Consistent with these results, g-tubulin ubiquitination wasonly detected during prophase (M0) and not duringmetaphase(M1), anaphase (M2), and telophase (M3) of cell cycle (Fig. 4H).As expected, we could not find any interaction between UCHL1and g-tubulin in MCF10A cell extracts prepared from differentstages of mitosis (Supplementary Fig. S4B). These resultstogether suggest that a direct interaction between BAP1 andg-tubulin aswell as deubiquitination of g-tubulin occurs duringmetaphase, anaphase, and telophase.

g-Tubulin is a substrate for the deubiquitinating enzymeBAP1

To analyze g-tubulin ubiquitination, we compared the over-all levels of ubiquitin-associated tubulin in BAP1- or emptyexpression plasmid-infected MDA-MB-468 cells. As controlcells showed ubiquitination of g-tubulin, BAP1-overexpressingcells significantly reduced this effect (Fig. 5A). Furthermore, weenriched and immunoprecipitated endogenous g-tubulin fromMDA-MB-468 cells and incubated that with recombinant BAP1

γ-Tub

γ-Tu

b

IgG

γ-Tu

b

IgG

IgG

Actin

Actin

γ-Tub55

Ub

γ-Tub

Ub

Ub

γ-Tub

Actin

γ-Tub

Actin

γ-Tub

IP:

IP:

70

55

70

55

40

55

40

70

70

55

IP:

70

55

4070

Total lysate

Total lysate

Total lysate

40

MC

F10

A

MC

F7

MD

A-M

B-2

31

MD

A-M

B-4

68A B

C D

Figure 3. Downregulation of BAP1 in MDA-MB-468 cells leads toincreased g-tubulin ubiquitination. A, the levels of g-tubulin in total celllysates of MCF10A, MCF7, MDA-MB-231, and MDA-MB-468 cellswere determined by Western blotting using anti-g-tubulin and -actinantibodies. The arrowhead points to a shift in size of g-tubulin in celllysates from MDA-MB-468 cells. B, immunoprecipitation of g-tubulin intotal cell lysates from MDA-MB-468 cells and a Western blot usingantibodies against g-tubulin and ubiquitin. The arrowheadpoints to a shiftin size of g-tubulin corresponding to ubiquitinated g-tubulin. The lysate(bottom) showsanequal amount of protein used for immunoprecipitation.C, immunoprecipitation of ubiquitin in total cell lysates from MDA-MB-468 cells and aWestern blot using antibodies against g-tubulin and actin.The arrowhead points to a shift in the size of g-tubulin corresponding toubiquitinated g-tubulin. The lysate (bottom) shows an equal amount ofprotein used for immunoprecipitation. D, immunoprecipitation ofg-tubulin in total cell lysates from MDA-MB-468 cells under denaturingconditions (8 mol/L urea) and a Western blot using antibodies againstg-tubulin and ubiquitin. The arrowhead points to a shift in the size ofg-tubulin corresponding to ubiquitinated g-tubulin. The lysate (bottom)shows an equal amount of protein used for immunoprecipitation.






protein. The reaction was carried out for the different timepoints in the presence or absence of DUB inhibitor N-ethyl-maleimide (NEM). This resulted in an increase of g-tubulin–ubiquitin conjugates, as measured by ubiquitin immunoblot-ting (Fig. 5B). Next, we constructed a catalytic inactive mutantof BAP1 by mutation of cysteine a.a. 91 to alanine in the UCHdomain of BAP1 (BAP1-C91A) and incubated purified wild-type (BAP1-Wt) or BAP1-C91A with g-tubulin isolated fromMDA-MB-468 cells. Wild-type, but not the catalytic inactivemutant of BAP1-C91A, resulted in significantly decreased

g-tubulin ubiquitination in vitro (Fig. 5C). These data togethersuggest that BAP1 is a specific DUB for g-tubulin.

BAP1 is necessary for proper mitotic spindleorganization and chromosome abnormalities

As an interaction was observed between BAP1 and g-tubulinduring mitosis (see Fig. 4), we next investigated whether BAP1expression can affect proper mitosis. The number of cellsharboring mitotic abnormalities observed in control MDA-MB-468 cells (Fig. 6A–C) was reduced when the cells were

γ-Tub

IgG

BA

P1

IgG

BA

P1

IgG

BA

P1

IgG

IgG

IgG

BA

P1

BA

P1

BA

P1

IgG

IgG

IgG

IgG

IgG

IgG

IgG

IgG

BA

P1

BA

P1

BA

P1

BA

P1

BAP1

γ-Tub

BAP1

γ-Tub

BAP1

BAP1 Merge

Actin

γ-TubDAPI

Prophase

Metaphase

Anaphase

Telophase

HA Mergeγ-TubDAPI

Prophase

Metaphase

Anaphase

Telophase

IP:

IP:

IgG

BA

P1

IgG

BA

P1

GS

T

GS

T-γ-

Tub

GST

GST-γ-Tub

His-BAP1

IP:

55

100

40

55

Interphase Mitosis Interphase Mitosis

100

55

100 100

70

30

7055

70

Total lysate

γ-Tub

BAP1

Actin

γ-Tub

BAP1

Actin γ-Tub

BAP1

Actin

55

100

40

55

100

40 55100

40

Total lysateTotal lysate

Total lysate

γ-Tub

γ-Tu

b

γ-Tu

b

γ-Tu

b

γ-Tu

b

BAP1

γ-Tub

BAP1

γ-Tub γ-Tub

Ub

BAP1

Actin

M0 M1 M2 M3 M0 M1 M2 M3

M0 M1 M2 M3

55

100

55

100

IP: IP:

55

100

40

Total lysate

A B C D

E F G H

Figure 4. BAP1 colocalizes and interacts with g-tubulin during metaphase, anaphase, and telophase. A, immunoprecipitation of endogenous BAP1 fromsynchronized MCF10A cells and immunoblotting using anti-BAP1 and -g-tubulin antibodies. MCF10A cells were synchronized by double thymidine blockrelease and collected in mitosis by mitotic shake-off. Both asynchronous and synchronous cell extracts were used for immunoprecipitation. Thelysate (bottom) shows an equal amount of protein used for immunoprecipitation. B, immunoprecipitation of HA-tagged BAP1 from synchronized MDA-MB-468 cells stably transduced with viral vectors encoding BAP1 (BAP1-MDA-MB-468) and immunoblotting using BAP1 and g-tubulin antibodies. BAP1-MDA-MB-468 cells were synchronized by double thymidine block release and collected in mitosis by mitotic shake-off. Both asynchronous and synchronouscell extracts were used for immunoprecipitation. The lysate (bottom) shows an equal amount of protein used for immunoprecipitation. C, purified GSTor GST-g-tubulin (0.5 mg) was incubated in vitrowith His-BAP1 (0.5 mg) for 30 minutes, followed by a GST pull-down assay using glutathione-agarose beads.The complexes were recovered from the beads and analyzed by means of immunoblotting against BAP1 for detection of His-BAP1 and GST for detection ofGST-g-tubulin. The arrowhead indicates the band corresponding to g-tubulin. D, confocal plane of MCF10A cells stained for g-tubulin (green), BAP1 (red),and DAPI (blue) as indicated. DAPI staining shows different stages of mitosis including prophase, metaphase, anaphase, and telophase. Bars, 10 mm.E, confocal plane of MDA-MB-468 cells stably transduced with viral vectors encoding HA-tagged BAP1 (BAP1-MDA-MB-468) and stained for g-tubulin(green), HA (red), and DAPI (blue) as indicated. DAPI staining shows different stages of mitosis including prophase, metaphase, anaphase, and telophase.Bars, 10 mm. F, MCF10A cells were synchronized by a double thymidine block. Cells in mitotic phase were obtained by nocodazole treatment (M0) andreleased from nocodazole after 45 minutes (M1), 90 minutes (M2), and 180 minutes (M3). Cell extracts were immunoprecipitated with BAP1 antibody andanalyzed by Western blotting using antibodies against BAP1, g-tubulin, and actin. The lysate (bottom) shows an equal amount of protein used forimmunoprecipitation. G, MDA-MB-468 cells stably transduced with viral vectors encoding HA-tagged BAP1 (BAP1-MDA-MB-468) were synchronized by adouble thymidine block. Cells in mitotic phase were obtained by nocodazole treatment (M0) and released from nocodazole after 45 minutes (M1), 90minutes(M2), and 180 minutes (M3). Cell extracts were immunoprecipitated with BAP1 antibody and analyzed by Western blotting using antibodies against BAP1,g-tubulin, and actin. The lysate (bottom) shows an equal amount of protein used for immunoprecipitation. H, MDA-MB-468 cells stably transduced withviral vectors encoding HA-tagged BAP1 (BAP1-MDA-MB-468) were synchronized by a double thymidine block. Cells in mitotic phase were obtainedby nocodazole treatment (M0) and released from nocodazole after 45 minutes (M1), 90 minutes (M2), and 180 minutes (M3). Cell extracts wereimmunoprecipitated with g-tubulin antibody and analyzed by Western blotting using antibodies against g-tubulin, ubiquitin, and actin. The lysate (bottom)shows an equal amount of protein used for immunoprecipitation.

Zarrizi et al.





stably expressing BAP1 (Fig. 6D). These abnormalities includechromosomes lagging during metaphase, uncondensed mitot-ic chromosomes, anaphase bridges, telophase bridges, andtripolar mitotic spindles, which are signs of chromosomeaberrations (Fig. 6A–6D). Figure 6E and F summarizes thepercentage of these abnormalities comparing BAP1- withempty expression plasmid-infected cells. Importantly, BAP1-C91A–expressing MDA-MB-468 cells did not show any differ-ences in the percentage of different mitotic abnormalitiescompared with empty expression plasmid-transfected cells(Fig. 6E–G), suggesting that the deubiquitin activity of BAP1is necessary for preventing chromosome abnormalities. Wecould not find any differences in microtubule nucleation andmicrotubule regrowth in control and BAP1-expressing cells,suggesting that deubiquitination of g-tubulin by BAP1 is notessential for mitotic spindle formation (Fig. 6H). Properlyanchored and organized microtubules at centrosomes andspindles are prerequisites for proper chromosome segregation;therefore, we investigated whether loss of BAP1 expressioncauses defects at these stages of mitosis in malignant breastepithelial cells. Immunofluorescent staining and statisticalanalysis showed significant differences in the number of

control cells exhibiting bipolar spindles with misaligned anduncondensed chromosomes with abnormal mitotic spindlescompared with BAP1-expressing cells (Fig. 7A and B). Theseresults suggest that lack of BAP1 expression causes mitoticspindle abnormalities that might contribute to improper chro-mosome segregation.

The mitotic defects observed in MDA-MB-468 cells withlow levels of BAP1 might be indicative of a premature exitfrom mitosis. This was tested by performing live-cell micros-copy using control and BAP1 stably expressing MDA-MB-468cells, which were transfected with the histone marker GFP-H2B. Control (Fig. 7C; Supplementary Video S1), but notBAP1-expressing cells, showed defects in chromosome con-densation and lagging of chromosomes that never correctlyprogressed to the metaphase plate (Fig. 7D, top; Supplemen-tary Video S2), incorrect chromosome segregation resultingin the formation of anaphase bridges (Fig. 7D, middle;Supplementary Video S3), and finally skipped metaphaseand subsequent aberrant anaphase (Fig. 7D, bottom; Sup-plementary Video S4). These results together suggest thatlow expression of BAP1 in breast cancer cells contributes tochromosome instability.

γ-Tub

γ-Tub

γ-Tu

b

γ-Tu

b

IgG

IgG

Actin

HA

Ub

γ-Tub

γ-Tub

Ub

1 2 3 4

Time: 1 h 4 h 1 h 4 hγ-Tub

Actin

Flag100

BAP1

γ-Tub

IP:

70

55

70

55

40

10040

100

70

70

IP: FLAG

His-BAP1BAP1-WtBAP1-C91A

BA

P1-

Wt

BA

P1-

C91

A

+ + + +

+

+–

– +++– ––

–+ –––– +–

– +NEM

55

70 55

70

55

70Total lysate

Total lysate

BAP1-MDA-MB-468

A B C

Figure 5. BAP1 is a specific DUB for g-tubulin. A, immunoprecipitation of g-tubulin in total cell lysates from MDA-MB-468 cells transduced with viral vectorsencoding HA-tagged BAP1 (BAP1-MDA-MB-468, right lane) or the control (empty expression plasmid, mid lane) and a Western blot using antibodiesagainst g-tubulin. The lysate (bottom) shows an equal amount of protein used for immunoprecipitation. B, g-tubulin isolated from MDA-MB-468 cellswas incubated with His-BAP1 for the indicated time in the presence or absence of NEM. The samples were analyzed byWestern blotting using anti-g-tubulinand -ubiquitin antibodies. C, BAP1 or a catalytically inactive mutant of BAP1 were purified from MDA-MB-468 cells transiently transfected with wild-typeFlag-tagged BAP1 (Flag-BAP1) or Flag-BAP1-C91A and incubated with endogenous g-tubulin immunoprecipitated from MDA-MB-468 cells. Thereaction mixture was incubated at 37�C for 1 hour and examined by Western blotting using anti-g-tubulin antibody. Densitometric analysis shows that 34%(comparing lane 2 and 3) and 36% (comparing lane 2 and 4) of total ubiquitin (lane 1) is modified via ubiquitination. The lysate (bottom) shows an equalamount of protein used for immunoprecipitation.






Discussiong-Tubulin is a member of the tubulin superfamily and is

required for nucleating the polymerization of microtubules.g-Tubulin ismainly located at the centrosome, and the amountof g-tubulin at the centrosome increases dramatically at thebeginning of mitosis and decreases at the end of mitosis (12).Posttranslational modification of g-tubulin by ubiquitinationis mediated via the BRCA1/BARD1 complex, which was shownto regulate the centrosome number and aster formation (15).During mitosis, BRCA1 localizes to centrosomes (17, 18) andthe BRCA1/BARD1 complex localizes to centrosomes through-

out the cell cycle, however, at very low levels during mitosis(19). Furthermore, disruption of the BRCA1 gene in mice led tocentrosome amplification and aneuploidy (20). Loss of BRCA1activity, as occurs in breast cancer cells, results in unrestrainedcentrosomes. This leads to hyperactive and overduplicatedcentrosomes often observed in breast cancer cells (15).

In the current study, we aimed to identify the enzyme(s)responsible for reversing ubiquitination of g-tubulin in breastepithelial cells. We screened a siRNA library for DUBs andidentified BAP1 and UCHL1. Genetic studies have uncoveredthat BAP1 mutations are related to different tumor types,

α-Tub MergeDAPI

α-Tub MergeDAPI

α-Tub MergeDAPI

α-Tub MergeDAPI

α-Tub MergeDAPI

α-Tub MergeDAPI

BAP1-C91A MergeDAPI

BAP1-C91A MergeDAPI

α-Tub MergeDAPI

n = 200Mitotic cells

Ctrl BAP1-C91A

Ctrl BAP1-MDA-MB-468

Laggingchromosomes

Noncondensedmetaphase

Anaphasebridge

Telophasebridge

BAP1

Ctrl

BAP1-C91

ABAP1

Ctrl

BAP1-C91

ABAP1

Ctrl

BAP1-C91

ABAP1

Ctrl

BAP1-C91

ABAP1

Cel

ls w

ith m

itotic

defe

cts

(%)

Cel

ls w

ith m

itotic

erro

rs (

%)

α-Tub MergeDAPI

α-Tub MergeDAPI

α-Tub MergeDAPI

2520151050

8

6

4

2

0

A D G

H

E

F

B

C

Figure6. BAP1expression rescues chromosomesegregation defects associatedwith abnormal spindles inMDA-MB-468 cells. A–C, representative imagesofchromosomal misalignment including chromosomal lagging (A), mitotic bridges (B), and noncondensed metaphase (C) observed in MDA-MB-468 cellsusing immunofluorescence staining and confocal microscopy. The arrowheads indicate chromosome segregation defects observed in MDA-MB-468 cells.Bars, 10 mm. D, representative images of proper alignment of chromosomes observed in MDA-MB-468 cells stably transduced with viral vectors encodingFlag-HA-tagged BAP1 (BAP1-MDA-MB-468) using immunofluorescence staining and confocal microscopy. Bars, 10 mm. E, quantification and percentagesof mitotic defects observed in empty expression plasmid (Ctrl) or BAP1 virally infected MDA-MB-468 cells in 50 mitotic cells per experiment from fourindependent experiments. The data are expressed as mean � SD. ��, P < 0.001. F, quantification of mitotic defects including chromosome lagging,noncondensed metaphase, anaphase, and telophase bridge in empty expression plasmid (Ctrl), BAP1, and BAP1-C91A–expressing cells. The data indicatethe percentage of 50 mitotic cells per experiment from four independent experiments expressed as mean � SD. �, P < 0.05. G, representative imagesof chromosomal abnormalities induced in cells transfectedwith a catalytically inactivemutant of Flag-tagged BAP1 (BAP1-C91A) using immunofluorescencestaining and confocal microscopy. The arrowheads indicate chromosome lagging (top) and anaphase bridge (bottom) observed in MDA-MB-468 cells.Bars, 10 mm. H, microtubule nucleation in MDA-MB-468 inducibly expressing BAP1 or empty expression plasmid (Ctrl) monitored by anti-g-tubulin stainingafter nocodozole exposure and washout. Bars, 10 mm.

Zarrizi et al.





including malignant mesothelioma (21, 22), melanocytictumor, uveal and cutaneous melanoma (7, 23), as well as lungcancer (24, 25) and renal carcinoma (26). Restoration of BAP1inhibits cell growth of non–small lung cell carcinoma cell lineNCI-H226 that lacks BAP1 by inducing cell death (6). In MCF7breast cancer cell lines, it was shown that the binding of BAP1to BRCA1 enhances BRCA1-mediated growth suppression,whereas other studies demonstrated that growth suppressionof cells is independent of BRCA1; instead, BAP1 disrupts theBRCA1/BARD1 heterodimer but cannot reverse its autoubi-quitination (5, 6, 9, 10). Therefore, the exact role of BAP1 inbreast cancer is unknown.To investigate whether BAP1 expression in breast cancer has

a prognostic implication, we performed Kaplan–Meier survivalanalysis using a microarray dataset from 2,978 patients withbreast cancer. High BAP1 expression was associatedwith betteroverall survival, and more specifically, high BAP1 expression intissue samples from basal and luminal subtype tumors wassignificantly associated with prolonged progression-free surviv-al. The regulatory mechanism that mediates the downregula-tion of BAP1 in metastatic breast cancer cells needs to beassessed in future studies. In MDA-MB-468 cells, we observedg-tubulin ubiquitination in nonsynchronized cells, whereasoverexpression of BAP1 in these cells significantly reduced thelevels of ubiquitin conjugates to g-tubulin. In addition,we foundthat the association between BAP1 and g-tubulin occurs duringM-phase, including metaphase, anaphase, and telophase, butnot during prophase of the cell cycle. The expression of BAP1

but not the catalytic inactive mutant of BAP1- or emptyexpression plasmid-infected cells reduced the percentage ofmitotic abnormalities and showed normal bipolar microtubuleorganization.

The MT nucleation rate usually increases at the entry intomitosis. The centrosomal concentration of g-tubulin reached atmetaphase and centrosome size increase through anaphase,whereas nucleation remains high through telophase, implyingthe presence of additional regulatory processes (27). Further-more, during mitosis there is a sudden increase of the g-tubulincontent in the centrosome (28), andmitotic centrosomes exhibitenhanced MT nucleation capacity (29). Therefore, the g-tubulincontent in thecentrosomeregulatesMTnucleation function (14).Previous studies found that ubiquitination of g-tubulin byBRCA1/BARD1 during the S and G2 phases (13, 14) has asignificant role in regulating MT nucleation, which further pre-vents aberrant reduplication of the centrosomes (15). This couldimply that g-tubulin ubiquitination directly inhibits centrosomemicrotubule nucleation function. In addition, during early mito-sis, it has been shown that Aurora A kinase (AURKA) localizes tothe centrosome and mediates microtubule nucleation activitythrough phosphorylation and inactivation of BRCA1 (14).

Deubiquitination enzymes and E3 ligases are often found ina complex together. These interactions allowing the E3 ligaseto regulate the target and its deubiquitination enzyme simul-taneously or deubiquitination enzyme/E3 ligase interactionsconfer target specificity, which is strictly dependent on the E3ligase for recognition of the substrate (4). The current study

α-Tub MergeDAPI

α-Tub MergeDAPI

α-Tub MergeDAPI

α-Tub MergeDAPI

α-Tub MergeDAPI

α-Tub MergeDAPI

n = 200Mitotic cells

Ctrl

Ctrl

BAP1-MDA-MB-468

BAP1

Ctrl BAP1

Cel

ls w

ith a

bnor

mal

mito

tic s

pind

le(%

)

50

40

30

20

10

0

A

D

B

C

Figure 7. BAP1 is required formitotic spindle organization. A, representative images ofmitotic spindles observed inMDA-MB-468 cells stably transducedwithviral vectors encoding empty expression plasmid (Ctrl) or Flag-HA-tagged BAP1 using immunofluorescence staining and confocal microscopy. Bars, 10 mm.B, cumulative data were obtained from four independent experiments, with at least 50 mitotic cells counted per experiment. Data are presented as thepercentage of cells with abnormal mitotic spindle expressed as mean � SD. ��, P < 0.001. C and D, images from live cell microscopy of MDA-MB-468inducibly expressing FLAG-HA-BAP1 or empty expression plasmid (Ctrl) and transiently transfected with H2B-GFP. The arrowhead shows mitotic controlcells harboring chromosome lagging (D; top), anaphase bridge (D; middle), and aberrant anaphase (D; bottom).






suggests that the complex interaction between BAP1, BRCA1,and g-tubulin is essential for deubiquitination of g-tubulin toallow proper mitotic progression.

In summary, our present findings together with previousobservations suggest that in the M-phase when BRCA1becomes inactive through AURKA phosphorylation, BAP1colocalizes with g-tubulin and removes ubiquitin from g-tubu-lin. Removal of ubiquitin from g-tubulin by BAP1might inducerecruitment of unmodified g-tubulin to the centrosome. Deu-biquitination of g-tubulin by BAP1 can further control themetaphase to anaphase transition and mitotic spindle orga-nization as well as preventing genomic instability by inhibitingaberrations in mitotic spindles. Our findings may have impor-tant implications for tumor progression, as chromosomalinstability and aneuploidy are associated with spindle defectsin several tumor types, including breast carcinoma.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: R. MassoumiDevelopment of methodology: M. Belting, R. MassoumiAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): R. Zarrizi, J.A. Menard, R. MassoumiAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): R. Zarrizi, J.A. Menard, M. Belting, R. MassoumiWriting, review, and/or revision of the manuscript: R. Zarrizi, M. Belting,R. MassoumiStudy supervision: R. Massoumi

Grant SupportThis work was supported by the Swedish Cancer Foundation, the Royal

Physiographic Society in Lund, the Magnus Bergvall Foundation, the GunnarNilsson Cancer Foundation, the Gyllenstiernska Krapperup Foundation, theStrategic Cancer Research Program BioCARE, and by funding from the EuropeanResearch Council (ERC), under the European Union's Seventh FrameworkProgramme for Research and Technology Development, grant agreement no.(260460; R.R.Massoumi).

The costs of publication of this article were defrayed in part by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received January 24, 2014; revised August 27, 2014; accepted September 9,2014; published OnlineFirst September 16, 2014.

References1. Pickart CM. Mechanisms underlying ubiquitination. Annu Rev Bio-

chem 2001;70:503–33.2. Weissman AM Themes and variations on ubiquitylation. Nature Rev

Mol Cell Biol 2001;2:169–78.3. Massoumi RUbiquitin chain cleavage: CYLD at work. Trends Biochem

Sci 2010;35:392–9.4. Nijman S, Luna-Vargas M, Velds A, Brummelkamp TR, Dirac AM,

SixmaTK, et al. A genomic and functional inventory of deubiquitinatingenzymes. Cell 2005;123:773–86.

5. Jensen DE, Proctor M, Marquis ST, Gardner HP, Ha SI, Chodosh LA,et al. BAP1: a novel ubiquitin hydrolase which binds to the BRCA1RING finger and enhances BRCA1-mediated cell growth suppression.Oncogene 1998;16:1097–112.

6. Ventii KH, Devi NS, Friedrich KL, Chernova TA, Tighiouart M, Van MeirEG, et al. BRCA1-associated protein-1 is a tumor suppressor thatrequires deubiquitinating activity and nuclear localization. Cancer Res2008;68:6953–62.

7. Harbour JW,OnkenMD,RobersonED,DuanS,CaoL,Worley LA, et al.Frequent mutation of BAP1 in metastasizing uveal melanomas. Sci-ence 2010;330:1410–3.

8. Eletr ZM, Wilkinson KD. An emerging model for BAP1's role in regu-lating cell cycle progression. Cell Biochem Biophys 2011;60:3–11.

9. Mallery DL, Vandenberg CJ, Hiom K. Activation of the E3 ligasefunction of the BRCA1/BARD1 complex by polyubiquitin chains.EMBO J 2002;21:6755–62.

10. Nishikawa H, Wu W, Koike A, Kojima R, Gomi H, Fukuda M, et al.BRCA1-associated protein 1 interferes with BRCA1/BARD1 RINGheterodimer activity. Cancer Res 2009;69:111–9.

11. Dey A, SeshasayeeD, Noubade R, French DM, Liu J, ChaurushiyaMS,et al. Loss of the tumor suppressor BAP1 causes myeloid transfor-mation. Science 2012;337:1541–6.

12. Kollman JM, Merdes A, Mourey L, Agard DA. Microtubule nucleationby g-tubulin complexes. Nature Rev Mol Cell Biol 2011;12:709–21.

13. Sankaran S, Starita LM, Groen AC, Ko MJ, Parvin JD. Centrosomalmicrotubule nucleation activity is inhibited by BRCA1-dependentubiquitination. Mol Cell Biol 2005;25:8656–68.

14. Sankaran S, Crone DE, Palazzo RE, Parvin JD. BRCA1 regulatesg-tubulin binding to centrosomes. Cancer Biol Ther 2007;6:1853–7.

15. Starita LM, Machida Y, Sankaran S, Elias JE, Griffin K, Schlegel BP,et al. BRCA1-dependent ubiquitination of g-tubulin regulates centro-some number. Mol Cell Biol 2004;24:8457–66.

16. Parvin JD. The BRCA1-dependent ubiquitin ligase, g-tubulin, andcentrosomes. Environ Mol Mutagen 2009;50:649–53.

17. Hsu LC, White RL. BRCA1 is associated with the centrosome duringmitosis. Proc Natl Acad Sci U S A 1998;95:12983–8.

18. Lotti LV, Ottini L, D'Amico C, Gradini R, Cama A, Belleudi F, et al.Subcellular localization of the BRCA1 gene product in mitotic cells.Genes Chromosomes Cancer 2002;35:193–203.

19. Sankaran S, Starita LM, Simons AM, Parvin JD. Identification ofdomains of BRCA1 critical for the ubiquitin-dependent inhibition ofcentrosome function. Cancer Res 2006;66:4100–7.

20. Xu X, Weaver Z, Linke SP, Li C, Gotay J, Wang XW, et al. Centrosomeamplification and a defective G2-M cell cycle checkpoint inducegenetic instability in BRCA1 exon 11 isoform-deficient cells. Mol Cell1999;3:389–95.

21. Testa JR, Cheung M, Pei J, Below JE, Tan Y, Sementino E, et al.GermlineBAP1mutations predispose tomalignantmesothelioma. NatGenet 2011;43:1022–5.

22. BottM,BrevetM, Taylor BS, ShimizuS, Ito T,WangL, et al. The nucleardeubiquitinase BAP1 is commonly inactivated by somatic mutationsand 3p21.1 losses in malignant pleural mesothelioma. Nat Genet2011;43:668–72.

23. Wiesner T, Obenauf AC, Murali R, Fried I, Griewank KG, Ulz P, et al.Germline mutations in BAP1 predispose to melanocytic tumors. NatGenet 2011;43:1018–21.

24. Abdel-Rahman MH, Pilarski R, Cebulla CM, Massengill JB, Christo-pher BN, Boru G, et al. Germline BAP1 mutation predisposes to uvealmelanoma, lung adenocarcinoma, meningioma, and other cancers.J Med Genet 2011;48:856–9.

25. Fan LH, Tang LN, Yue L, Yang Y, Gao ZL, Shen Z. BAP1 is a goodprognostic factor in advanced non-small cell lung cancer. Clin InvestMed 2012;35:E182–9.

26. Pena-Llopis S, Vega-Rubin-de-Celis S, Liao A, Leng N, Pavia-JimenezA,WangS, et al. BAP1 loss defines a newclass of renal cell carcinoma.Nat Genet 2012;44:751–9.

27. Piehl M, Tulu US, Wadsworth P, Cassimeris L. Centrosome matura-tion: measurement of microtubule nucleation throughout the cell cycleby usingGFP-taggedEB1. ProcNatl AcadSci USA2004;101:1584–8.

28. Khodjakov A, Rieder CL. The sudden recruitment of g-tubulin to thecentrosome at the onset ofmitosis and its dynamic exchange through-out the cell cycle, do not require microtubules. J Cell Biol 1999;146:585–96.

29. Kuriyama R, Borisy GG. Microtubule-nucleating activity of centro-somes in Chinese hamster ovary cells is independent of the cen-triole cycle but coupled to the mitotic cycle. J Cell Biol 1981;91:822–6.


Zarrizi et al.




2014;74:6499-6508. Published OnlineFirst September 16, 2014.Cancer Res Reihaneh Zarrizi, Julien Albert Menard, Mattias Belting, et al. Instability in Breast Cancer Cells

-Tubulin by BAP1 Prevents ChromosomeγDeubiquitination of

Updated version

10.1158/0008-5472.CAN-14-0221doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2014/09/16/0008-5472.CAN-14-0221.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/74/22/6499.full#ref-list-1

This article cites 29 articles, 13 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/74/22/6499.full#related-urls

This article has been cited by 11 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/74/22/6499To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-14-0221

http://cancerres.aacrjournals.org/content/suppl/2014/09/16/0008-5472.CAN-14-0221.DC1

http://cancerres.aacrjournals.org/content/74/22/6499.full#ref-list-1

http://cancerres.aacrjournals.org/content/74/22/6499.full#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/74/22/6499


deubiquitination of g-tubulin by bap1 prevents chromosome ...46 51 56 61 66 71 76 81 86 91 96 101...

Documents