The BRCA1/BARD1 Heterodimer
Mitotic Spindle Assembly
Vladimir Joukov,1Aaron C. Groen,2Tatyana Prokhorova,3,4Ruth Gerson,1Erinn White,1Alison Rodriguez,1
Johannes C. Walter,3,* and David M. Livingston1,*
1Dana-Farber Cancer Institute and Harvard Medical School, 44 Binney Street, Boston, MA 02115, USA
2Department of Systems Biology
3Department of Biological Chemistry and Molecular Pharmacology
Harvard Medical School, 240 Longwood Avenue, Boston, MA 02115, USA
4Present address: The Molecular Endocrinology Unit, Medical Biotechnology Centre, University of Southern Denmark,
5000 Odense, Denmark.
*Contact: firstname.lastname@example.org (J.C.W.), email@example.com (D.M.L.)
The heterodimeric tumor-suppressor complex
BRCA1/BARD1 exhibits E3 ubiquitin ligase ac-
tivity and participates in cell proliferation and
chromosome stability control by incompletely
mammalian cells and Xenopus egg extracts,
BRCA1/BARD1 is required for mitotic spindle-
pole assembly and for accumulation of TPX2,
a major spindle organizer and Ran target, on
spindle poles. This function is centrosome inde-
pendent, operates downstream of Ran GTPase,
and depends upon BRCA1/BARD1 E3 ubiquitin
ligase activity. Xenopus BRCA1/BARD1 forms
endogenous complexes with three spindle-pole
proteins, TPX2, NuMA,andXRHAMM—aknown
XRHAMM function. These observations reveal
a previously unrecognized function of BRCA1/
BARD1 in mitotic spindle assembly that likely
contributes to its role in chromosome stability
control and tumor suppression.
suppression function remains incompletely understood.
to a BRCA1 role in the maintenance of genomic integrity
via participation in homologous-recombination-mediated
double-strand-break repair, the regulation of cell-cycle
In vivo, most BRCA1 molecules form heterodimers with
a structurally related protein, BARD1 (Wu et al., 1996).
BRCA1 and BARD1 each contain an N-terminal RING
domain and two C-terminal BRCT motifs. RING domains
catalyze ubiquitin transfer by interacting with ubiquitin-
conjugating enzymes, and BRCT domains can bind cer-
sequences (reviewed in Fang et al., 2003; Glover et al.,
2004). BRCA1/BARD1 heterodimers promote ubiquitin
transfer far more efficiently than either protein alone and
can catalyze autoubiquitination as well as the cell-free
ubiquitination of other proteins (Hashizume et al., 2001;
Mallery et al., 2002; Sato et al., 2004; Starita et al., 2004;
Yu et al., 2006). Whether any of these proteins is a physio-
logical BRCA1/BARD1 substrate is unknown.
BRCA1 and BARD1 are conserved in vertebrates,
plants, and worms but are absent from yeast (Boulton
et al., 2004; Joukov et al., 2001). Inactivation of BRCA1
and BARD1 in mice and frogs yields similar phenotypes,
with embryos dying early in embryogenesis. These em-
bryos also reveal marked chromosomal abnormalities
and a cell proliferation defect (Deng and Wang,2003; Jou-
kov etal., 2001;Ludwig etal.,1997; McCarthy et al.,2003;
Venkitaraman, 2002 and references therein). The mecha-
nism underlying these abnormalities is incompletely de-
fined, although accumulating DNA damage that in turn
activates cell-cycle checkpoints has been suggested
(reviewed in Deng and Wang, 2003; Venkitaraman, 2002).
Although it was initially believed that BRCA1 functions
largely in S phase (Scully et al., 1997; Venkitaraman,
2002), growing evidence suggests that it is also active in
mitosis. First, aneuploidy is common among BRCA1-
and BARD1-deficient cells (Joukov et al., 2001; McCarthy
et al., 2003; Xu et al., 1999). Second, mouse fibroblasts
that carry a biallelic hypomorphic BRCA1 mutation exhibit
mitotic defects (Xu et al., 1999). Third, BRCA1 binds to
tubulin and localizes in part at centrosomes and spindle
microtubules (Hsu and White, 1998). Fourth, steady-state
the protein is ubiquitinated and undergoes proteasome-
dependent degradation in G1 and S phase (Choudhury
Cell 127, 539–552, November 3, 2006 ª2006 Elsevier Inc. 539
et al., 2004). Finally, in contrast to normal proliferating so-
matic cells that are not viable without BRCA1 or BARD1,
trophoblast giant cells, which endoreduplicate their DNA
without intervening mitoses, remain unaffected when
depleted of either protein (Ludwig et al., 1997; McCarthy
et al., 2003).
We have examined the function of BRCA1/BARD1 het-
erodimers using Xenopus egg extracts as an experimental
system (Murray, 1991). This study demonstrates that
BRCA1/BARD1 ensures fidelity of mitosis and mitotic
dle assembly. BRCA1/BARD1 attenuates the activity of
XRHAMM (Xenopus receptor for hyaluronic-acid-medi-
ated motility) (Groen et al., 2004; Maxwell et al., 2003),
thereby permitting the normal concentration of TPX2
(Wittmann et al., 2000) on spindle poles and proper spin-
Cell-Cycle-Dependent Regulation of BRCA1/BARD1
Xenopus egg extracts were used to assess BRCA1/
BARD1 function because this system faithfully reca-
pitulates cellular processes in which BRCA1/BARD1 is
potentially involved. Importantly, these extracts allow one
to bypass the problem of nonviability of BRCA1- and
BARD1-deficient cells and embryos, which normally com-
plicates in vivo studies of these proteins. The levels of
BRCA1 and BARD1 were similar in extracts arrested in
interphase and meiotic metaphase (Figure 1A). A BRCA1
antibody (Ab) depleted ?98% of the ambient BRCA1
and BARD1 in both interphase- and metaphase-arrested
extracts, and a BARD1 Ab led to similar effects (Figure 1B
and data not shown). These results imply that nearly all of
the BRCA1 and BARD1 in Xenopus egg extracts exists
in a heterodimeric complex throughout the cell cycle. In
cycling egg extract that oscillates between S phase and
mitosis due to the periodic synthesis and degradation of
cyclin B (Murray, 1991), BRCA1 and BARD1 efficiently
bound to chromatin in interphase and largely dissociated
from it in mitosis (Figure 1C).
In cultured mammalian cells, BRCA1 formed character-
istic foci in a subset of interphase cells as reported (Scully
et al., 1997) (Figure 1D) and was diffusely distributed
throughout the cell and excluded from chromatin during
mitosis (Figure 1D, arrowhead). When soluble proteins
Figure 1. Cell-Cycle-Dependent Proper-
ties of BRCA1/BARD1
(B) BRCA1 and BARD1 associate with each
other in egg extract. Interphase egg extract
was depleted with XBRCA1- or XBARD1-spe-
cific Ab (I) or the corresponding preimmune
IgG (P). One microliter of each sample and
the indicated amount of untreated extract (%
of 1 ml) were analyzed by W blotting.
(C) Cell-cycle-dependent regulation of BRCA1/
BARD1 chromatin binding. Chromatin isolated
from cycling egg extract at the indicated time
points was analyzed by W blotting with the in-
dicated antibodies (upper panel). Aliquots of
extract were also analyzed for DNA replication
and chromatin morphology (lower panel and
data not shown).
(D and E) Immunofluorescence microscopic
images showing localization of BRCA1 in mi-
totic and S phase HeLa cells. Cells were fixed
in methanol/acetone (D) or paraformaldehyde
following permeabilization with digitonin (E)
and stained with the indicated antibodies and
540 Cell 127, 539–552, November 3, 2006 ª2006 Elsevier Inc.
were eluted from cells by digitonin permeabilization prior
to fixation, residual BRCA1 was detected in metaphase
cells in foci distributed around, but not on, chromatin or
the mitotic spindle (Figure 1E).
The electrophoretic mobility of BRCA1 and BARD1
decreased as a result of specific phosphorylation during
mitosis (Figure 1A and data not shown), when the hetero-
dimer is largely excluded from chromatin (Figure 1C).
Thus, in egg extract, the heterodimer accumulates in the
nucleus and binds to interphase chromatin. In mitosis,
most BRCA1/BARD1 is phosphorylated and excluded
BARD1 islocalized in foci thatsurround chromatin and the
BRCA1/BARD1 Is Required for Proper Nuclear
Assembly in Postmitotic Interphase
BRCA1- and BARD1-specific antibodies (Joukov et al.,
2001) were used to deplete the heterodimer from extract
(Figure 1B), and chromatin dynamics during interphase
and mitosis was analyzed. When demembranated sperm
chromatin was added to BRCA1/BARD1-depleted inter-
phase extract, chromatin decondensation, nuclear-enve-
lope formation, and the rate of DNA replication were the
same as in mock-treated extract (Figures 2A and 2B).
Thus, BRCA1/BARD1 is dispensable for S phase progres-
sion in egg extract. Similarly, when sperm chromatin was
added to mock-treated or BRCA1/BARD1-depleted cy-
cling extract, it efficiently decondensed and formed nuclei
(Figure 2C, 25 min and 55 min). Both extracts subse-
quently entered mitosis as seen by nuclear-envelope
breakdown and chromatin condensation (Figure 2C, 85
min). The extracts exited mitosis and continued to cycle
nearly synchronously. Upon mitotic exit, nuclei of rela-
tively uniform size formed in the mock-treated extract
(Figure 2C, upper row, 130 min and 225 min). In contrast,
postmitotic nuclei in BRCA1/BARD1-depleted extract
were heterogeneous, with the majority being 3 to 10 times
depleted extract during the first, premitotic interphase
Postmitotic interphase can be also generated by prein-
cubating sperm chromatin in metaphase-arrested (also
referred to as cytostatic factor [CSF]-arrested) egg extract
followed by release into interphase (detailed in the Sup-
plemental Experimental Procedures in the Supplemental
Data available with this article online). In such settings,
uniform nuclei were observed in mock-treated extract. In
contrast, nuclei that formed in BRCA1/BARD1-depleted
extract upon release from the metaphase arrest varied in
size, being 3 to 10 times smaller compared to the control
nuclei (Figure S1A). Importantly, this defect was reversed
by supplementing the depleted extract with immunoaffin-
ity purified, recombinant Xenopus BRCA1/BARD1 hetero-
dimer (rBRCA1/BARD1, detailed below) (Figures S1C and
S1D; see also Supplemental Results).
Taken together, these results demonstrate that, in
Xenopus egg extract, BRCA1/BARD1 is dispensable for
DNA replication. However, passage of chromatin through
mitosis establishes a requirement for BRCA1/BARD1 for
proper nuclear assembly.
BRCA1/BARD1 Is Required for Proper Mitotic
Given BRCA1/BARD1’s involvement in postmitotic inter-
phase, it was important to determine whether the hetero-
dimer is also required for the execution of mitosis itself.
To this end, we examined the effect of BRCA1/BARD1
depletion on metaphase spindle assembly. A standard
approach that includes replication of sperm-chromatin
DNA in interphase extract followed by the addition of
Figure 2. BRCA1/BARD1 Is Required for
Postmitotic Nuclear Assembly
(A and B)BRCA1/BARD1is dispensable fornu-
clear assembly and DNA replication in inter-
phase egg extract. Aliquots of mock-treated
and BRCA1/BARD1-depleted interphase ex-
tracts were withdrawn 90 min (A) or at the indi-
cated times (B)after addition ofsperm chroma-
tin and analyzed for nuclear morphology (A)
and DNA replication (B).
(C) Cycling extracts were supplemented with
sperm chromatin. Aliquots were removed at
the indicated times and analyzed for chromatin
Cell 127, 539–552, November 3, 2006 ª2006 Elsevier Inc. 541
CSF-arrested extract to induce the metaphase state was
employed (Desai et al., 1999). In the mock-treated extract,
most spindles were largely bipolar with focused spindle
poles, and chromosomes properly congressed at the
metaphase plate (Figures 3Aa and 3Ab). In contrast, most
spindles in the BRCA1/BARD1-depleted extract con-
tained unfocused spindle poles and were more rounded.
They also exhibited a higher density of microtubules and
a failure of chromosome congression at the metaphase
plate (Figures 3Ad and 3Ae). To confirm that the spindle
Figure 3. BRCA1/BARD1 Is Required for Proper Mitotic Spindle Assembly
(A and B) Sperm chromatin was replicated in mock-treated and BRCA1/BARD1-depleted, CSF-arrested extracts supplemented with rhodamine-
labeled tubulin and, where indicated, rBRCA1/BARD1. The extractswere driven into metaphase and were analyzed 1 hr later for spindle morphology.
(A) Representative metaphase spindles.
(B) Spindle structures were categorized based on the degree of chromosome alignment (top panel) and were quantified (bottom panel).
(C–E) Mitotic defects in BRCA1/BARD1 siRNA-transfected HeLa cells.
(C) Representative images of normal (top panels) and disorganized (bottom panels) metaphase spindles assembled in mock-treated and BRCA1/
BARD1-depleted cells, respectively.
(D) Quantitative analysis of normal and abnormal mitotic spindle structures that appeared in mock-treated versus BRCA1/BARD1-depleted cells.
Examples are shown in Figure S3B.
(E) Representative examples of chromosome segregation defects in BRCA1/BARD1-deficient cells (arrowheads). Note inefficient focusing of micro-
tubules into spindle poles in anaphase (DBRCA1/BARD1, ‘‘Tubulin’’ panel). In all images, microtubules are in red and chromatin is pseudocolored
542 Cell 127, 539–552, November 3, 2006 ª2006 Elsevier Inc.
outcome of eliminating BRCA1/BARD1, a recombinant
heterodimer was produced by coexpression of Xenopus
BRCA1 and BARD1 in cultured insect cells followed by
immunoaffinity purification. Xenopus rBRCA1/BARD1
(Figure S2A, lane 2), like its human counterpart (Mallery
etal.,2002),bothdisplayed autoubiquitination andubiqui-
tinated certain histones, implying that it is biologically
notype was substantially alleviated by supplementing
fect of BRCA1/BARD1 depletion and rescue on chromo-
some alignment is quantified in Figure 3B.
We asked whether the mitotic spindle defects seen
in depleted Xenopus egg extracts could be observed in
BRCA1/BARD1-deficient cells. HeLa cells were trans-
fected with control or BRCA1- and BARD1-specific siRNA
oligonucleotides (Figure S3A), and mitotic spindle mor-
phology was assessed. The mitotic figures were catego-
rized based on the stage of mitosis in which they were
detected (i.e., prophase, metaphase, anaphase, and telo-
phase) and the extent to which they were defective
(Figure S3B). The percentage of mitotic figures in each
category was calculated (Figure 3D). BRCA1/BARD1 de-
pletion did not affect the morphology or proportion of cells
in prophase. However, it reduced from 35% to 10% the
abundance of normal metaphase cells with bipolar spin-
dles and properly aligned chromosomes. Accordingly,
BRCA1/BARD1-deficient cells displayed a higher propor-
tion of disorganized mitotic spindles (34% versus 14%).
egg extracts were remarkably similar (Figure 3C versus
exhibited a severe defect in chromosome segregation
during anaphase, revealing chromosomal bridges and
lagging chromosomes (Figures 3D and 3E, arrowheads
in the ‘‘Anaphase’’ panel). At telophase, some lagging
chromosomes became enclosed in nuclear envelopes,
giving rise to micronuclei (Figure 3E, arrowheads in ‘‘Telo-
Taken together, these results indicate that BRCA1/
assembly and at the mitosis-to-interphase transition for
proper chromosome segregation and nuclear assembly.
BRCA1/BARD1 Regulates Ran-Driven
We observed that BRCA1/BARD1-depleted extracts and
cells share major phenotypic features with cells in which
the Ran pathway or certain downstream targets of Ran-
GTP are disrupted. These features include mitotic spindle
defects, chromosome missegregation, and micronucleus
formation (Compton and Cleveland, 1993; Merdes and
Cleveland, 1998; Moore et al., 2002; O’Brien and Wiese,
2006; Wang et al., 2004). We therefore asked whether
BRCA1/BARD1 is involved in Ran-dependent spindle as-
sembly. Addition of Ran-GTP to Xenopus egg extract is
sufficient to cause the formation of spindle-related struc-
tures, called asters and pseudospindles, in the absence
and Zheng, 1999). This phenomenon imitates chromatin-
driven spindle assembly and is likely dependent upon
achieving high local concentrations of spindle assembly
factors (SAFs) (e.g., NuMA and TPX2) following their re-
lease by Ran-GTP from inhibitory binding by the importin
a/b heterodimer (reviewed in Fant et al. 2004; Hetzer
et al., 2002; Quimby and Dasso, 2003). When a constitu-
tively active Ran mutant defective in GTP hydrolysis
(Ran(Q69L)-GTP) was added to mock-treated extract,
asters with radially oriented microtubules and sharply
focused poles formed (Figure 4Aa), as reported (Carazo-
Salas et al., 1999). In contrast, asters assembled in
BRCA1/BARD1-depleted extract appeared larger in size
and contained dense, disoriented microtubules with
most severecases, asters lacked defined centers (Figures
4Ac and 4Bd). Importantly, the number of Ran-induced
asters was not affected by BRCA1/BARD1 depletion
(data not shown), implying that BRCA1/BARD1 is not
essential for microtubule (MT) assembly per se. Addition
of wild-type (WT) rBRCA1/BARD1 to depleted extract sig-
nificantly restored aster MT organization in terms of both
qualitative appearance (Figure 4B) and absolute size (Fig-
et al., 2003; Figure S2A, lane 3; Figure S2B, lanes 3 and 5
versus lanes 2 and 4) was significantly less efficient than
its WT counterpart in rescuing these defects (Figures
4B–4D). These results indicate that BRCA1/BARD1 and
its E3 ubiquitin ligase activity participate in MT organiza-
tion and spindle-pole assembly downstream of Ran-GTP.
BRCA1/BARD1 Controls Targeting of TPX2
to Spindle Poles
affected by BRCA1/BARD1 depletion was investigated
next. NuMA and TPX2 both participate in spindle-pole as-
sembly and are targets of Ran during mitosis (Fant et al.,
2004; Hetzer et al., 2002; Merdes et al., 2000; Wittmann
et al., 2000). In addition, XRHAMM was recently impli-
cated in chromatin-driven MT nucleation and spindle-
pole formation (Groen et al., 2004). During mitosis, when
the nuclear envelope disassembles, XRHAMM binds to
microtubules and, in association with g-TuRC and TPX2,
facilitates Ran-dependent MT nucleation and concentra-
tion of TPX2 on spindle poles via a currently unknown
mechanism (Groen et al., 2004; Maxwell et al., 2003). In
mock-treated extract, NuMA, g-tubulin, XRHAMM, and
TPX2 efficiently bound to microtubules and concentrated
on aster poles (Figures 5A and 5B). In BRCA1/BARD1-
depleted extract, these proteins also bound to microtu-
bules: NuMA accumulated on the aster poles almost as
efficiently as in mock-treated extracts (Figure 5A, row 2
on aster poles, although in a more diffuse and less orderly
manner compared to mock-treated extract (Figure 5A,
Cell 127, 539–552, November 3, 2006 ª2006 Elsevier Inc. 543
row 4 versus 3). In contrast, TPX2 was diffusely localized
along the length of microtubules and failed to concentrate
on aster poles (Figure 5B, row 2 versus 1). Addition of WT
rBRCA1/BARD1 restored both aster MT organization and
the concentration of TPX2 on aster poles, whereas
rBRCA1(I26A)/BARD1 was less efficient compared to the
WT heterodimer in rescuing both defects (Figure 5B, row
3 versus 2 and 4).
As stated above, Ran-GTP-induced aster formation
utilizes the chromatin-driven/anastral pathway of spindle
assembly. This pathway operates in cells that lack a de-
and vertebrates and cells of higher plants. In contrast, in
mostsomatic cells,which containcentrosomes, the anas-
in the presence of centrosomes, BRCA1/BARD1 is also
needed for efficient TPX2 accumulation on spindle poles.
Notably, each sperm pronucleus contains a centrosome
attached to its surface. Thus, MT architecture and TPX2
localization were compared in asters induced by sperm
chromatin in mock-treated and BRCA1/BARD1-depleted
CSF-arrested extracts. Although both extracts supported
formation of microtubular structures around chromatin
with similar efficiency, there was a profound difference in
theirarchitecture.Asters and spindles inthe mock-treated
Figure 4. BRCA1/BARD1 Regulates Ran-
Dependent MT Organization
(A) Fluorescence micrographs (top panels)
and 3D surface plots (bottom panels) of the
sembled in mock-treated and BRCA/BARD1-
depleted extracts. Scale bar = 10 mm.
(B–D) Rescue of the MT aster structures with
(B) Ran-GTP-induced MT asters assembled in
mock-treated and BRCA1/BARD1-depleted
extract were supplemented with buffer or with
recombinant WT or enzymatically deficient
(I26A dimer) BRCA1/BARD1. Asters were cate-
gorized based on the degree of aster-pole
focusing (upper panel), and each category was
quantified (lower panel).
(C) Average diameter of asters in each group.
Error bars represent standard deviations (n =
(D) W blot analysis of the extracts.
544 Cell 127, 539–552, November 3, 2006 ª2006 Elsevier Inc.
with TPX2 concentrated at the center of each pole (Fig-
tract, chromatin-induced spindles exhibited disorganized
poles, and TPX2 diffusely bound to microtubules (Fig-
ure 5C, row 3; for more examples, see Figure S4).
We next compared the localization of TPX2 on meta-
phase spindles of HeLa cells transfected with either con-
trol or BRCA1/BARD1-specific siRNAs. In control cells,
TPX2 tightly concentrated in the vicinity of spindlepoles of
all metaphase spindles analyzed. In contrast, in ?20% of
BRCA1/BARD1 siRNA-treated metaphase cells, TPX2
was diffusely localized along the length of spindle micro-
tubules and failed to concentrate on spindle poles (Fig-
ure 5D, compare panels in row 2).
These observations demonstrate that BRCA1/BARD1
and its E3 ubiquitin ligase activity control spindle-pole
assembly by facilitating targeting of TPX2 to spindle poles
independent of centrosomes.
BRCA1/BARD1 Associates with
To test whether BRCA1/BARD1 associates with spindle-
pole-organizing proteins, we analyzed BRCA1 immuno-
precipitates (IPs) for the presence of TPX2, NuMA, and
Figure 5. BRCA1/BARD1 Is Required for Concentrating TPX2 on Spindle Poles
(A) IF images of Ran-GTP-induced MT asters assembled in mock-treated and BRCA1/BARD1-depleted extracts after methanol fixation followed by
staining with antibodies directed against NuMA and a-tubulin (rows 1 and 2) or g-tubulin and XRHAMM (rows 3 and 4).
(Band C) Representative Ran-GTP-induced (B) and chromatin-induced (C) asters assembled in mock-treated and BRCA1/BARD1-depleted extracts
supplemented with rhodamine-labeled tubulin, Alexa Fluor 488-labeled anti-TPX2 Ab, and the indicated components.
Cell 127, 539–552, November 3, 2006 ª2006 Elsevier Inc. 545
XRHAMM. Both NuMA and XRHAMM were significantly
enriched in these fractions compared to control IPs
(Figure 6A, lane 3 versus 2). Although TPX2 was not
enriched in anti-BRCA1 IPs (Figure 6A), an identical ex-
periment performed with TPX2 Ab led to specific co-
immunoprecipitation of BRCA1/BARD1 (Figure 6B, lanes
5 and 6 versus 1 and 2). An analogous result was obtained
with XRHAMM- and NuMA-specific antibodies (Fig-
ure 6B, lanes 3 and 4 versus 1 and 2; Figure 6C, lane 3
An association between XRHAMM and TPX2 requires
additional factors present in extract (Groen et al., 2004).
One could envision that BRCA1/BARD1 regulates the
XRHAMM-TPX2 interaction by ubiquitination. However,
this scenario seems unlikely since the XRHAMM-TPX2
interaction was not affected by BRCA1/BARD1 depletion
(data not shown) or by supplementing an extract with
ubiquitin aldehyde, a potent inhibitor of multiple deubiqui-
depletion did not affect the electrophoretic mobility or
abundance of XRHAMM, TPX2, NuMA, or g-tubulin (Fig-
ure 6D and data not shown). The latter outcome likely
siderably less abundant than SAFs and therefore interacts
with only a small fraction of these proteins. Indeed, no
more than 5% of NuMA or XRHAMM was found to associ-
ate with BRCA1/BARD1 (Figure 6A).
Given the specific association of BRCA1/BARD1 with
the aforementioned SAFs, binding of BRCA1/BARD1 to
microtubules was also tested in extracts and mammalian
cells. Using multiple BRCA1- and BARD1-monospecific
antibodies, we did not detect clear colocalization of
BRCA1 and/or BARD1 with spindle microtubules (Figures
1D and 1E and data not shown). These results demon-
strate that BRCA1/BARD1 physically interacts with SAFs
participating in the processes of Ran-dependent MT poly-
merization and spindle-pole assembly (i.e., with NuMA,
XRHAMM, and TPX2). Thus far, there is no evidence sup-
porting the notionthat BRCA1and BARD1 arethemselves
MT-associated proteins (MAPs) or are involved in MAP
BRCA1/BARD1 Regulates Mitotic MT Organization
in a XRHAMM-Dependent Manner
While analyzing the localization of XRHAMM on micro-
tubules, a surprising observation was made. Supplemen-
tation of the BRCA1/BARD1-depleted extract with an
a C-terminal segment of XRHAMM (a-XRHAMM) led to
rescue of the MT aster defects (data not shown). Rescue
was also achieved by supplementing BRCA1/BARD1-
depleted extract with small amounts (5–20 ng/ml) of un-
labeled a-XRHAMM, but not with nonimmune rabbit IgG
(Figure 7B, column 2 versus 1). An identical concentration
of a-XRHAMM had no effect on MT asters in the mock-
treated extract (Figure 7A, column 2 versus 1). Moreover,
MT aster assembly was not affected when either mock-
mented withsimilar amounts ofAb directed against NuMA
or TPX2 (data not shown). a-XRHAMM (20 ng/ml) also effi-
ciently rescued defects in asters and spindles assembled
around sperm chromatin in BRCA/BARD1-depleted ex-
tract (Figure 5C, row 4 versus 3).
is targeted by a-XRHAMM, we added to extracts a C-ter-
minal fragment of XRHAMM (aa 1038–1175; XRHAMM-
CWT) that had been used as immunogen during Ab devel-
opment. Interestingly, on its own, XRHAMM-CWT (1.25–4
mM) disrupted aster-pole structure and prevented efficient
(Figure 7A, column 3 versus 1). Surprisingly, addition of
this peptide to the BRCA1/BARD1-depleted extract had
an even more dramatic effect. MT asters became severely
disorganized, and TPX2 was abnormally bound along the
length of thick, disoriented MT fibers (Figure 7B, column 3
Figure 6. BRCA1/BARD1 Interacts with Spindle-Pole-Orga-
(A–C) CSF-arrested extracts were incubated for 1 hr at 21?C followed
by IP with nonimmune Ig (Control) or the indicated antibodies. The IPs
were analyzed by W blotting with the indicated antibodies. Where indi-
cated, the extracts were supplemented with 6 mM ubiquitin aldehyde
(+) prior to incubation (B).
XRHAMM-codepleted CSF-arrested extracts were analyzed by W
blotting with the indicated antibodies.
546 Cell 127, 539–552, November 3, 2006 ª2006 Elsevier Inc.
versus 1). XRHAMM-CWT contains a leucine zipper motif,
and its amino acid sequence exhibits high interspecies
conservation (Groen et al., 2004; Maxwell et al., 2003).
This XRHAMM motif appears to be functionally important
because a C-terminal fragment in which three conserved
leucines of the leucine zipper were replaced by arginines
(XRHAMM-CR3) was inactive in disrupting aster structure
and TPX2 targeting to aster poles (Figure 7A, column 4
versus 3). It also failed to generate the ‘‘extreme’’ aster
phenotype produced by the WT peptide in BRCA1/
BARD1-depleted extract (Figure 7B, column 4 versus 3).
Nevertheless, XRHAMM-CR3retained the ability to bind
a-XRHAMM as efficiently as XRHAMM-CWT (Figure 7C,
lane 5 versus 4). Indeed, when ithad fully titrated the avail-
able a-XRHAMM, it abrogated the rescue effect of this Ab
in BRCA1/BARD1-depleted extract (Figure 7B, column 5
versus 2). This result indicates that the effect of the Ab
on aster structure is specific to XRHAMM.
These experiments revealed that a-XRHAMM and the
ated exhibited opposite effects on MT asters in BRCA1/
BARD1-depleted extract: The Ab rescued, while the pep-
tide aggravated, the MT aster phenotype (Table S1). Two
interpretations of these results were considered: (1)
XRHAMM function is inhibited in the absence of BRCA1/
BARD1; the Ab stimulates, while the peptide further in-
hibits XRHAMM function, or (2) XRHAMM is hyperactive
in the absence of BRCA1/BARD1; the Ab downregulates
XRHAMM function, and the peptide activates it.
To distinguish between these possibilities, we com-
pared the effects of adding a-XRHAMM and XRHAMM-
CWT on the structure of asters and TPX2 MT localization
in an extract that had been partially depleted of XRHAMM.
Depleting extract of XRHAMM by 90%–95% significantly
inhibited both the abundance and size of asters (Figures
7D and 7E, column 2 versus 1). Although XRHAMM was
previously shown to be required for TPX2 concentration
on spindle poles (Groen et al., 2004), in this setting,
TPX2 still concentrated at the centers of faint MT asters,
suggesting that the amount of residual XRHAMM in the
extract (5% to 10%) was sufficient to perform certain
key XRHAMM functions, albeit inefficiently. In keeping
with this notion, the effect of adding a-XRHAMM to this
extract was additive with XRHAMM depletion—i.e., the
Ab further decreased the efficiency of aster formation as
well as the intensity of tubulin and TPX2 staining at aster
centers (Figures 7D and 7E, column 3 versus 2). By con-
trast, addition of XRHAMM-CWT to the extract, which
was partially depleted of XRHAMM, led to a substantial
rescue of the aster formation defect: The asters were
larger and TPX2 was concentrated on aster centers,
assembled in mock-treated extract (Figure 7E, column 4
versus 2). The peptide, however, failed to significantly in-
crease the abundance of asters formed (Figure 7D). These
results indicate that a-XRHAMM inhibits XRHAMM MT-
organizing function, while XRHAMM-CWT can partially
compensate for the loss of XRHAMM function.
Because abnormalities associated with BRCA1/BARD1
depletion were rescued by a-XRHAMM and because
this reagent appears to inhibit XRHAMM function, the
data suggest that XRHAMM is hyperactive in the absence
of BRCA1/BARD1, a condition that is deleterious for spin-
dle function. As a further test of this hypothesis, we asked
whether the spindle-pole defects caused by BRCA1/
BARD1 depletion could be rescued by partial elimination
of XRHAMM from extract. Such treatment indeed rescued
aster architecture and localization of TPX2 on aster poles
to an extent similar to addition of Ab (Figure 7F, column 4
versus columns 2 and 3). Results of the experiments in-
volving a-XRHAMM and XRHAMM-CWT are summarized
in Table S1.
Taken together, these observations indicate that
attenuating the otherwise excessive activity of XRHAMM
BRCA1/BARD1 Controls Ran-Dependent Mitotic
This study demonstrates a critical role for BRCA1/BARD1
in mitotic MT organization and spindle-pole assembly in
both Xenopus eggextractsand culturedmammaliancells.
Spindle poles form by concentrating MT minus ends at
their centers, a process that does not require centro-
somes but rather relies on the activity of various noncen-
trosomal MAPs as well as plus- and minus-end-directed
motor proteins (reviewed in Fant et al., 2004). Two MAPs
critical for spindle-pole assembly, NuMA and TPX2, are
transported to spindle poles by the minus-end-directed
motor complex, dynein/dynactin (Merdes et al., 2000;
Wittmann et al., 2000). NuMA remained at the centers of
unfocused aster poles that assembled in BRCA1/BARD1-
depleted extract, suggesting that BRCA1/BARD1 is not
required for the dynein/dynactin-dependent transport
per se. TPX2, however, was not effectively targeted to
spindle poles in this setting, providing evidence for a spe-
cific line of communication between BRCA1/BARD1 and
a key step in spindle-pole formation.
spindle-assembly checkpoint defect in cells expressing
a hypomorphic mutant BRCA1 allele (Wang et al., 2004;
Xu et al., 1999). However, the fidelity of mitotic exit was
compromised, as evidenced by the appearance of chro-
mosome segregation defects and micronucleus formation
in BRCA1/BARD1-siRNA-treated cells (Figure 3E) and the
postmitotic nuclear-assembly defect in BRCA1/BARD1-
depleted egg extract (Figure 2C and Figure S1A). Of note,
multiple nuclei and micronuclei were previously observed
in BRCA1-deficient cells, and this defect was thought to
result from multipolar spindle formation due to abnormal
centrosome amplification in these cells (Xu et al., 1999).
We speculate that both spindle-pole abnormalities and
postmitotic nuclear-assembly defects that develop in
Cell 127, 539–552, November 3, 2006 ª2006 Elsevier Inc. 547
Figure 7. BRCA1/BARD1 Regulates MT Organization in a XRHAMM-Dependent Fashion
Prior to assaying, egg extracts were supplemented with rhodamine-labeled tubulin and Alexa Fluor 488-labeled anti-TPX2 Ab; MT asters were in-
duced by the addition of Ran(Q69L)-GTP (except in [C]).
(A and B) Representative asters assembled in mock-treated (A) and BRCA1/BARD1-depleted (B) extracts supplemented with the indicated compo-
(C) Disruption of its leucine zipper does not affect XRHAMM-C interaction with the corresponding specific Ab. Equal amounts of recombinant
XRHAMM-CWT (wt) and XRHAMM-CR3(R3) were immunoprecipitated with the affinity-purified a-XRHAMM (lanes 4 and 5, respectively) or nonim-
mune IgG (lanes 3 and 6, respectively) and analyzed by SDS-PAGE followed by staining of the membrane with Ponceau S (IgG) or S protein-HRP
(XRHAMM-C). Note that XRHAMM-CR3exhibits slightly slower electrophoretic mobility than XRHAMM-CWT.
(D and E) Quantitative analysis of asters (D) and representative structures (E) assembled in the mock-treated and XRHAMM-depleted extracts sup-
plemented with the indicated components. Values in (D) represent means ± standard deviations of two independent measurements.
548 Cell 127, 539–552, November 3, 2006 ª2006 Elsevier Inc.
BRCA1/BARD1-deficient settings can be attributed to de-
regulation of TPX2. In this regard, Ran-GTP and its target
proteins (and BRCA1/BARD1 partners) TPX2 and NuMA
play a role in both spindle formation/function and post-
mitotic nuclear assembly (Compton and Cleveland, 1993;
Hetzer et al., 2002; Merdes and Cleveland, 1998; O’Brien
and Wiese, 2006). Spindle-pole disorganization due to
mitotic TPX2 dysfunction may lead to inefficient chromo-
some tethering to spindle poles in anaphase/telophase,
followed by enclosure of the resulting loose/lagging chro-
mosomes by the nuclear envelope. Such a mechanism
has been proposed as the reason for similar abnormalities
seen in cells with perturbed NuMA function (Merdes and
Cleveland, 1998). In addition, TPX2 has been implicated in
nuclear assembly and nuclear-envelope growth indepen-
dent of its mitotic function (O’Brien and Wiese, 2006). The
evidence for a biochemical interaction of BRCA1/BARD1
with TPX2, XRHAMM, and NuMA in egg extract supports
the existence of a functional link between these proteins.
Moreover, in human cell extracts, BRCA1/BARD1 also
coexists in a complex with NuMA (R. Greenberg, B. Sob-
hian, and D.M.L., unpublished data). Thus, in both mam-
malian cells and frog egg extracts, there is evidence for
One wonders whether NuMA and TPX2, which, like
BRCA1/BARD1, are localized in the interphase nucleus
(Compton and Cleveland, 1993; Wittmann et al., 2000),
also engage in certain postdamage S phase functions
of BRCA1/BARD1 (e.g., repair of double-strand breaks,
checkpoint activation, etc.). In this regard, there is a grow-
that perform seemingly unrelated functions during mitosis
and interphase (reviewed in Hetzer et al., 2005; Quimby
and Dasso, 2003).
BRCA1/BARD1 Regulates the MT-Organizing
Function of XRHAMM
This study demonstrates that BRCA1/BARD1 facilitates
TPX2 targeting to spindle poles by downmodulating
XRHAMM function. Data presented here also implicate
the highly conserved, leucine-zipper-bearing C-terminal
domain of XRHAMM as a contributor to those aspects of
MT-organizing function that are regulated by BRCA1/
BARD1. Given that the leucine zipper is a potential pro-
tein-interacting motif, we speculate that the C-terminal
domain of XRHAMM is involved in the formation of
XRHAMM homodimers or heterodimers with other SAFs
and that these interactions are important for targeting of
TPX2 to spindle poles. In a similar vein, one way of
explaining the XRHAMM-agonistic effect of XRHAMM-
CWT is to propose that it interferes with the homo- and/or
heterodimerization of endogenous XRHAMM.
The E3 ubiquitin ligase activity of BRCA1/BARD1 ap-
an enzymatically deficient heterodimer, rBRCA1(I26A)/
BARD1, was considerably less efficient than its WT coun-
terpart in reversing an abnormal aster phenotype. Con-
ceivably, BRCA1/BARD1 ubiquitinates TPX2, XRHAMM,
and/or NuMA, and this modification is required for the tar-
geting of TPX2 to spindle poles. If BRCA1/BARD1 does
ubiquitinate any of these proteins, the modification might
be transient and/or sensitive to the action of certain deu-
biquitinating enzymes, one of which, BAP1, is known to
associate with BRCA1/BARD1 (Jensen et al., 1998). Alter-
natively, BRCA1/BARD1 might regulate XRHAMM/TPX2
indirectly, via ubiquitination of additional protein partners.
Further studies addressing the functional link between
BRCA1/BARD1 and XRHAMM/TPX2 will be essential for
understanding the mechanisms of spindle-pole assembly.
A New Pathway for BRCA1-Mediated
There are reasons to believe that the newly uncovered
function of BRCA1/BARD1 in the control of Ran-depen-
dent MT and spindle-pole assembly is related to the ac-
knowledged role of BRCA1 in the maintenance of genome
stability and tumor suppression. Failure to properly form
spindle apparatus and can lead to chromosome segrega-
tion defects and aneuploidy, abnormalities that are char-
acteristic of both BRCA1/BARD1-deficient cellsandmany
tumor cells(Fant etal.,2004; Xuet al.,1999).Furthermore,
our study implicates BRCA1 in the regulation of SAFs that
have been previously linked to cancer in their own right
(Figure S5). Aberrant expression of RHAMM and TPX2, as
well as Aurora A, a mitotic kinase whose localization and
activity are regulated by TPX2 (reviewed in Crane et al.,
2003), were linked to malignant transformation as well as
progression of certain human tumors (Crane et al., 2003;
Maxwell et al., 2005; Smith et al., 2006 and references
therein). Moreover, RHAMM and TPX2 are considered
candidate oncoproteins (Hall et al., 1995; Maxwell et al.,
2005; Smith et al., 2006), and Aurora A is a likely oncopro-
tein given that its gene is amplified and its mRNA is over-
expressed in multiple human cancers. Furthermore,
ectopic overexpression of Aurora A is sufficient to trans-
form certain cell types (reviewed in Crane et al., 2003). In
this regard, we have also found that TPX2 mislocalization
in BRCA1-deficient cells leads to mislocalization of Aurora
A (data not shown).
In keeping with our observations, a recent independent
study based on a ‘‘breast cancer network’’ model of
mammalian functional genomic and protein interaction
parameters has suggested a functional link between
(F) Representative asters assembled in the mock-treated (column 1), BRCA1/BARD1-depleted (columns 2 and 3), or BRCA1/BARD1 +
XRHAMM-codepleted (column 4) extracts supplemented with the indicated components. W blots of the corresponding extracts are shown in
Cell 127, 539–552, November 3, 2006 ª2006 Elsevier Inc. 549
BRCA1, RHAMM, and Aurora A (M.A. Pujana et al., un-
published data). In addition, Aurora A has been shown to
ble feedback connection between this kinase and its reg-
ulators (Crane et al., 2003; Ouchi et al., 2004). Whether
certain aspects of a BRCA1?/?breast or ovarian cancer
phenotype are a product of dysfunction of XRHAMM,
TPX2, and/or Aurora A remains to be determined.
Finally, another breast and ovarian tumor suppressor,
BRCA2, which interacts physically with BRCA1, has re-
cently been implicated in the control of cytokinesis (Dan-
iels et al., 2004; Venkitaraman, 2002). It will be interesting
to learn whether the cytokinesis function of BRCA2 is re-
lated to the BRCA1/BARD1 mitotic/MT-organizing func-
tion and, if so, whether a defect in this complex set of
events contributes to a breakdown in BRCA1 and/or
BRCA2 tumor-suppression function.
Recombinant Proteins and Antibodies
The recombinant FLAG-BRCA1/HA-BARD1 heterodimers were pro-
duced using the Bac-to-Bac Baculovirus Expression System (Gibco
BRL) and doubly immunoaffinity purified prior to use. In vitro analysis
of Xenopus BRCA1/BARD1 E3 ligase activity was carried out as previ-
ously described for the human heterodimer (Mallery et al., 2002). Plas-
mid construction, protein expression and purification, and antibodies
Xenopus Egg Extracts
Crude egg extracts were prepared as described (Murray, 1991). Meta-
phase extracts were released into interphase by addition of 0.5 mM
CaCl2. Immunodepletions were carried out using specific antibodies
bound to protein A-Sepharose. Incubations of extracts were carried
out at 21?C unless indicated otherwise. For more details on extracts
and immunodepletions, see the Supplemental Experimental Proce-
Analysis of Chromatin, Mitotic Spindles, and Asters
in Egg Extracts
DNA replication was analyzed by measuring the incorporation of
[a-32P]dATP into DNA as described (Dasso and Newport, 1990). Anal-
ysis of chromatin in egg extract is detailed in the Supplemental Exper-
imental Procedures. Metaphase bipolar spindles were assembled in
egg extract supplemented with 75 mg/ml of rhodamine tubulin (Cyto-
skeleton) as described (Desai et al., 1999). MT asters were induced
by supplementing CSF-arrested extract with 15 mM Ran(Q69L)-GTP.
For IF analysis of NuMA, XRHAMM, and g-tubulin localization, MT
spindle and aster structures were isolated by centrifugation through
a glycerol cushion, fixed, and stained with the corresponding anti-
bodies as described (Desai et al., 1999). TPX2 localization on MT as-
ters and spindles was analyzed by direct IF microscopy of extracts
supplemented with anti-TPX2 Ab (5 ng/ml). The antibody was labeled
with Alexa Fluor 488 carboxylic acid, succinimidyl ester (Groen et al.,
Immunoprecipitations from Xenopus egg extracts were performed as
detailed in the Supplemental Experimental Procedures.
Analysis of BRCA1/BARD1 Mitotic Function in HeLa Cells
HeLa cells were cultivated in DMEM/10% fetal calf serum. Cells were
seeded on coverslips in a 6-well plate and were transfected 24 hr later
with a mixture of hBRCA1 and hBARD1 SMARTpool siRNAs (100 nM
each) or with an equal amount of the control nontargeting siRNAs
(Dharmacon). Transfections were carried out using Oligofectamine re-
agent (Invitrogen) according to the manufacturer’s instructions.
Twenty-four hours after the first transfection, a second, identical,
transfection was carried out. Thirty-six hours after the second trans-
fection, cells were washed in PBS, fixed in methanol/acetone (7:3 mix-
ture) at ?20?C, and immunostained. Alternatively, cells were permea-
bilized with digitonin and fixed with formaldehyde as described
(Joseph et al., 2002). Coverslips were mounted over DAPI-containing
VECTASHIELD stain (Vector Laboratories).
Fluorescence Microscopy and Image Analysis
Fluorescence microscopy of chromatin and spindle structures in egg
extract was carried out using an Eclipse E600 (Nikon) equipped with
a SPOT camera (Diagnostic Instruments) and an Axioskop 2 (Zeiss).
Fluorescence microscopy of HeLa cells was performed using the
Axioskop 2. Images were obtained and analyzed using Spot RT Soft-
ware v3.0 (Diagnostic Instruments) and AxioVision software (Zeiss).
Three-dimensional surface plots of MT asters were generated using
the program, ImageJ 1.34s (http://rsb.info.nih.gov/ij/).
Supplemental Data include Supplemental Results, Supplemental Ex-
perimental Procedures, Supplemental References, five figures, and
one table and can be found with this article online at http://www.cell.
We thank M. Dasso, T. Hirano, D. Compton, and T. Mitchison for gen-
erous gifts of reagents and R. Ohi, J.S. Stanford, E. Arias, R. Green-
berg, B. Sobhian, S. Ganesan, A. DeNicolo, and other members of
the Livingston, Walter, and Mitchison laboratories for many helpful dis-
cussions and for reagents. We also wish to thank J.B.A. Green for con-
siderable advice andinsightand W.Luo and R.S.Gelmanfor statistical
analysis of the data. We apologize to authors whose work could not be
directly cited owing to space constraints. This work was supported by
grants from the National Cancer Institute to D.M.L.; a Stewart Trust
grant, NIH grant GM62267, and ACS grant 106201 to J.C.W., and
Department of Defense Breast Cancer Research Program award
W81XWH-04-1-0524 to V.J. D.M.L. is a grantee of and consultant to
the Novartis Institute for Biomedical Research.
Received: March 28, 2005
Revised: June 23, 2006
Accepted: August 31, 2006
Published: November 2, 2006
Boulton, S.J., Martin, J.S., Polanowska, J., Hill, D.E., Gartner, A., and
Vidal, M. (2004). BRCA1/BARD1 orthologs required for DNA repair in
Caenorhabditis elegans. Curr. Biol. 14, 33–39.
Brzovic, P.S., Keefe, J.R., Nishikawa, H., Miyamoto, K., Fox, D., 3rd,
Fukuda, M., Ohta, T., and Klevit, R. (2003). Binding and recognition
in the assembly of an active BRCA1/BARD1 ubiquitin-ligase complex.
Proc. Natl. Acad. Sci. USA 100, 5646–5651.
Carazo-Salas, R.E., Guarguaglini, G., Gruss, O.J., Segref, A., Karsenti,
E., and Mattaj, I.W. (1999). Generation of GTP-bound Ran by RCC1 is
required for chromatin-induced mitotic spindle formation. Nature 400,
Choudhury, A.D., Xu, H., and Baer, R. (2004). Ubiquitination and pro-
teasomal degradation of the BRCA1 tumor suppressor is regulated
during cell cycle progression. J. Biol. Chem. 279, 33909–33918.
550 Cell 127, 539–552, November 3, 2006 ª2006 Elsevier Inc.
Compton, D.A., and Cleveland, D.W. (1993). NuMA is required for the
proper completion of mitosis. J. Cell Biol. 120, 947–957.
Crane, R., Gadea, B., Littlepage, L., Wu, H., and Ruderman, J.V.
(2003). Aurora A, meiosis and mitosis. Biol. Cell. 96, 215–229.
Daniels, M.J., Wang, Y., Lee, M., and Venkitaraman, A.R. (2004). Ab-
normal cytokinesis in cells deficient in the breast cancer susceptibility
protein BRCA2. Science 306, 876–879.
Dasso, M.,and Newport, J.W. (1990). Completion of DNA replication is
monitored by a feedback system that controls the initiation of mitosis
in vitro: studies in Xenopus. Cell 61, 811–823.
Deng, C.X., and Wang, R.H. (2003). Roles of BRCA1 in DNA damage
repair: a link between development and cancer. Hum. Mol. Genet.
Desai, A., Murray, A., Mitchison, T.J., and Walczak, C.E. (1999). The
use of Xenopus egg extracts to study mitotic spindle assembly and
function in vitro. Methods Cell Biol. 61, 385–412.
Fang, S., Lorick, K.L., Jensen, J.P., and Weissman, A.M. (2003). RING
finger ubiquitin protein ligases: implications for tumorigenesis, metas-
tasisand formolecular targets incancer.Semin.Cancer Biol. 13,5–14.
Fant, X., Merdes, A., and Haren, L. (2004). Cell and molecular biology
of spindle poles and NuMA. Int. Rev. Cytol. 238, 1–57.
Glover, J.N., Williams, R.S., and Lee, M.S. (2004). Interactions be-
tween BRCT repeats and phosphoproteins: tangled up in two. Trends
Biochem. Sci. 29, 579–585.
Groen, A.C., Cameron, L.A., Coughlin, M., Miyamoto, D.T., Mitchison,
T.J., and Ohi, R. (2004). XRHAMM functions in ran-dependent micro-
Curr. Biol. 14, 1801–1811.
Hall, C.L., Yang, B., Yang, X., Zhang,S., Turley, M., Samuel, S., Lange,
L.A., Wang, C., Curpen, G.D., Savani, R.C., et al. (1995). Overexpres-
sion of the hyaluronan receptor RHAMM is transforming and is also
required for H-ras transformation. Cell 82, 19–26.
Hashizume, R., Fukuda, M., Maeda, I., Nishikawa, H., Oyake, D.,
Yabuki, Y., Ogata, H., and Ohta, T. (2001). The RING heterodimer
BRCA1-BARD1 is a ubiquitin ligase inactivated by a breast cancer-
derived mutation. J. Biol. Chem. 276, 14537–14540.
Hetzer, M., Gruss, O.J., and Mattaj, I.W. (2002). The Ran GTPase as
a marker of chromosome position in spindle formation and nuclear en-
velope assembly. Nat. Cell Biol. 4, E177–E184.
Hetzer, M.W., Walther, T.C., and Mattaj, I.W. (2005). Pushing the enve-
lope: structure, function, and dynamics of the nuclear periphery. Annu.
Rev. Cell Dev. Biol. 21, 347–380.
Hsu, L.C., and White, R.L. (1998). BRCA1 is associated with the cen-
trosome during mitosis. Proc. Natl. Acad. Sci. USA 95, 12983–12988.
Jensen, D.E., Proctor, M., Marquis, S.T., Gardner, H.P., Ha, S.I.,
Chodosh, L.A., Ishov, A.M., Tommerup, N., Vissing, H., Sekido, Y.,
et al. (1998). BAP1: a novel ubiquitin hydrolase which binds to the
BRCA1 RING finger and enhances BRCA1-mediated cell growth sup-
pression. Oncogene 16, 1097–1112.
Joseph, J., Tan, S.H., Karpova, T.S., McNally, J.G., and Dasso, M.
(2002). SUMO-1 targets RanGAP1 to kinetochores and mitotic spin-
dles. J. Cell Biol. 156, 595–602.
Joukov, V., Chen, J., Fox, E.A., Green, J.B., and Livingston, D.M.
(2001). Functional communication between endogenous BRCA1 and
its partner, BARD1, during Xenopus laevis development. Proc. Natl.
Acad. Sci. USA 98, 12078–12083.
Ludwig, T., Chapman, D.L., Papaioannou, V.E., and Efstratiadis, A.
(1997). Targeted mutations of breast cancer susceptibility gene homo-
logs in mice: lethal phenotypes of Brca1, Brca2, Brca1/Brca2, Brca1/
p53, and Brca2/p53 nullizygous embryos. Genes Dev. 11, 1226–1241.
Mallery, D.L., Vandenberg, C.J., and Hiom, K. (2002). Activation of the
E3 ligase function of the BRCA1/BARD1 complex by polyubiquitin
chains. EMBO J. 21, 6755–6762.
Maxwell, C.A., Keats, J.J., Crainie, M., Sun, X., Yen, T., Shibuya, E.,
Hendzel, M., Chan, G., and Pilarski, L.M. (2003). RHAMM is a centro-
somal protein that interacts with dynein and maintains spindle pole
stability. Mol. Biol. Cell 14, 2262–2276.
Maxwell, C.A., Keats, J.J., Belch, A.R., Pilarski, L.M., and Reiman, T.
(2005). Receptor for hyaluronan-mediated motility correlates with cen-
trosome abnormalities in multiple myeloma and maintains mitotic in-
tegrity. Cancer Res. 65, 850–860.
McCarthy, E.E., Celebi, J.T., Baer, R., and Ludwig, T. (2003). Loss of
Bard1, the heterodimeric partner of the Brca1 tumor suppressor, re-
sults in early embryonic lethality and chromosomal instability. Mol.
Cell. Biol. 23, 5056–5063.
Merdes, A., and Cleveland, D.W. (1998). The role of NuMA in the inter-
phase nucleus. J. Cell Sci. 111, 71–79.
Merdes, A., Heald, R., Samejima, K., Earnshaw, W.C., and Cleveland,
D.W. (2000). Formation of spindle poles by dynein/dynactin-depen-
dent transport of NuMA. J. Cell Biol. 149, 851–862.
Miki, Y., Swensen, J., Shattuck-Eidens, D., Futreal, P.A., Harshman,
K., Tavtigian, S., Liu, Q., Cochran, C., Bennett, L.M., and Ding, W.
(1994). A strong candidate for the breast and ovarian cancer suscep-
tibility gene BRCA1. Science 266, 66–71.
Moore, W., Zhang, C., and Clarke, P.R. (2002). Targeting of RCC1 to
chromosomes is required for proper mitotic spindle assembly in
human cells. Curr. Biol. 12, 1442–1447.
Murray, A.W. (1991). Cell cycle extracts. Methods Cell Biol. 36, 581–
clear assembly in cell-free Xenopus laevis egg extracts. J. Cell Biol.
Ouchi, M., Fujiuchi, N., Sasai, K., Katayama, H., Minamishima, Y.A.,
Ongusaha, P.P., Deng, C., Sen, S., Lee, S.W., and Ouchi, T. (2004).
BRCA1 phosphorylation by Aurora-A in the regulation of G2 to M tran-
sition. J. Biol. Chem. 279, 19643–19648.
Quimby, B.B., and Dasso, M. (2003). The small GTPase Ran: interpret-
ing the signs. Curr. Opin. Cell Biol. 15, 338–344.
Sato, K., Hayami, R., Wu, W., Nishikawa, T., Nishikawa, H., Okuda, Y.,
Ogata, H., Fukuda, M., and Ohta, T. (2004). Nucleophosmin/B23 is
a candidate substrate for the BRCA1-BARD1 ubiquitin ligase. J. Biol.
Chem. 279, 30919–30922.
T., and Livingston, D.M. (1997). Association of BRCA1 with Rad51 in
mitotic and meiotic cells. Cell 88, 265–275.
Smith, L.T., Mayerson, J., Nowak, N.J., Suster, D., Mohammed, N.,
Long, S., Auer, H., Jones, S., McKeegan, C., Young, G., et al. (2006).
20q11.1 amplification in giant-cell tumor of bone: Array CGH, FISH,
and association with outcome. Genes Chromosomes Cancer 45,
Starita, L.M., Machida, Y., Sankaran, S., Elias, J.E., Griffin, K., Schle-
gel, B.P., Gygi, S.P., and Parvin, J.D. (2004). BRCA1-dependent ubiq-
uitination of gamma-tubulin regulates centrosome number. Mol. Cell.
Biol. 24, 8457–8466.
Venkitaraman, A.R. (2002). Cancer susceptibility and the functions of
BRCA1 and BRCA2. Cell 108, 171–182.
Wang, R.H., Yu, H., and Deng, C.X. (2004). A requirement for breast-
cancer-associated gene 1 (BRCA1) in the spindle checkpoint. Proc.
Natl. Acad. Sci. USA 101, 17108–17113.
Wilde, A., and Zheng, Y. (1999). Stimulation of microtubule aster for-
mation and spindle assembly by the small GTPase Ran. Science
Wittmann, T., Wilm, M., Karsenti, E., and Vernos, I. (2000). TPX2,
a novel Xenopus MAP involved in spindle pole organization. J. Cell
Biol. 149, 1405–1418.
Cell 127, 539–552, November 3, 2006 ª2006 Elsevier Inc. 551
Wu, L.C., Wang, Z.W., Tsan, J.T., Spillman, M.A., Phung, A., Xu, X.L.,
Yang, M.C., Hwang, L.Y., Bowcock, A.M., and Baer, R. (1996). Identi-
ficationof aRING protein that can interact invivo withthe BRCA1 gene
product. Nat. Genet. 14, 430–440.
Xu, X., Weaver, Z., Linke, S.P., Li, C., Gotay, J., Wang, X.W.,
Harris, C.C., Ried, T., and Deng, C.X. (1999). Centrosome amplifi-
cation and a defective G2-M cell cycle checkpoint induce genetic
instability in BRCA1 exon 11 isoform-deficient cells. Mol. Cell 3,
Yu, X., Fu, S., Lai, M., Baer, R., and Chen, J. (2006). BRCA1 ubiquiti-
nates its phosphorylation-dependent binding partner CtIP. Genes
Dev. 20, 1721–1726.
552 Cell 127, 539–552, November 3, 2006 ª2006 Elsevier Inc.