KDM4C (GASC1) lysine demethylase is associated with mitotic chromatin and regulates chromosome segregation during mitosis

Abstract

Various types of human cancers exhibit amplification or deletion of KDM4A-D members, which selectively demethylate H3K9 and
H3K36, thus implicating their activity in promoting carcinogenesis. On this basis, it was hypothesized that dysregulated expression
of KDM4A-D family promotes chromosomal instabilities by largely unknown mechanisms. Here, we show that unlike KDM4A-B, KDM4C
is associated with chromatin during mitosis. This association is accompanied by a decrease in the mitotic levels of H3K9me3.
We also show that the C-terminal region, containing the Tudor domains of KDM4C, is essential for its association with mitotic
chromatin. More specifically, we show that R919 residue on the proximal Tudor domain of KDM4C is critical for its association
with chromatin during mitosis. Interestingly, we demonstrate that depletion or overexpression of KDM4C, but not KDM4B, leads
to over 3-fold increase in the frequency of abnormal mitotic cells showing either misaligned chromosomes at metaphase, anaphase–telophase
lagging chromosomes or anaphase–telophase bridges. Furthermore, overexpression of KDM4C demethylase-dead mutant has no detectable
effect on mitotic chromosome segregation. Altogether, our findings implicate KDM4C demethylase activity in regulating the
fidelity of mitotic chromosome segregation by a yet unknown mechanism.

Full text

6168–6182 Nucleic Acids Research, 2014, Vol. 42, No. 10 Published online 11 April 2014
doi: 10.1093/nar/gku253
KDM4C (GASC1) lysine demethylase is associated
with mitotic chromatin and regulates chromosome
segregation during mitosis
Ilana Kupershmit†, Hanan Khoury-Haddad†, Samah W. Awwad, Noga Guttmann-Raviv and
Nabieh Ayoub∗
Department of Biology, Technion, Israel Institute of Technology, Haifa 3200003, Israel
Received September 18, 2013; Revised March 11, 2014; Accepted March 15, 2014
ABSTRACT
Various types of human cancers exhibit amplifica-
tion or deletion of KDM4A-D members, which selec-
tively demethylate H3K9 and H3K36, thus implicat-
ing their activity in promoting carcinogenesis. On
this basis, it was hypothesized that dysregulated ex-
pression of KDM4A-D family promotes chromosomal
instabilities by largely unknown mechanisms. Here,
we show that unlike KDM4A-B, KDM4C is associ-
ated with chromatin during mitosis. This associa-
tion is accompanied by a decrease in the mitotic lev-
els of H3K9me3. We also show that the C-terminal
region, containing the Tudor domains of KDM4C,
is essential for its association with mitotic chro-
matin. More specifically, we show that R919 residue
on the proximal Tudor domain of KDM4C is criti-
cal for its association with chromatin during mito-
sis. Interestingly, we demonstrate that depletion or
overexpression of KDM4C, but not KDM4B, leads
to over 3-fold increase in the frequency of abnor-
mal mitotic cells showing either misaligned chromo-
somes at metaphase, anaphase–telophase lagging
chromosomes or anaphase–telophase bridges. Fur-
thermore, overexpression of KDM4C demethylase-
dead mutant has no detectable effect on mitotic chro-
mosome segregation. Altogether, our findings impli-
cate KDM4C demethylase activity in regulating the
fidelity of mitotic chromosome segregation by a yet
unknown mechanism.
INTRODUCTION
Nucleus of the eukaryotic cells is composed of DNA and
proteins that are organized in higher-order structure termed
chromatin (1). One main function of chromatin is to al-
low sophisticated packaging of the DNA into the eukary-
otic nucleus. The second function is to provide a dynamic
platform that regulates the execution of diverse processes
such as replication, gene expression, DNA repair and re-
combination (2–6). Dysregulation of these processes or mu-
tations affecting chromatin-remodeling complexes has been
linked to many multi-system disorders and cancer develop-
ment (3,7,8).
Chromatin structure is regulated by the dynamic in-
terplay between histone-interacting proteins and post-
translational modications (PTMs) of histones (5,9,10).
This regulation is capable of altering the chromatin and
therefore modulates DNA accessibility (10,11). Histones
are subjected to a variety of PTMs such as acetyla-
tion, methylation, phosphorylation, ubiquitylation, sumoy-
lation, ADP-ribosylation, deimination and proline isomer-
ization (12). These PTMs have multiple functions includ-
ing the regulation of gene transcription (2,13). Accordingly,
PTMs regulate the establishment and maintenance of tran-
scriptionally active euchromatin and the condensed form of
chromatin named heterochromatin (13–17).
Lysine methylation is one of the most common modi-
cations of histone tails, which acts as a platform for chro-
matin modier proteins and leads to either gene activa-
tion or repression. Tri-methylation of histone H3 lysine
9 (H3K9me3) is enriched in condensed pericentric hete-
rochromatin, while di-methylation of histone H3 lysine 9
(H3K9me2) and mono-methylation of histone H3 lysine 9
(H3K9me1) are associated with transcriptionally silent re-
gions within euchromatin (18–20). Further, lysine methyla-
tion has been implicated in multiple cellular processes such
as regulation of gene expression, DNA replication, recom-
bination, repair, heterochromatin formation and mainte-
nance, mitosis and genomic imprinting. Aberrant histone
methylation has also been linked to different human dis-
eases such as cancer (21–31).
Two families of lysine demethylases (KDM) have been
identied, conrming that histone methylation is a re-
versible and dynamically regulated process (32–34). One
∗To whom correspondence should be addressed. Tel: +972 4 8294232; Fax: +972 4 8225153; Email: ayoubn@technion.ac.il
†Equal contribution.
C
The Author(s) 2014. Published by Oxford University Press on behalf of Nucleic Acids Research.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/), which
permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.

Nucleic Acids Research, 2014, Vol. 42, No. 10 6169
family is referred to as the Jumonji C (JmjC)-domain-
containing proteins. This class contains the conserved JmjC
catalytic domain, which belongs to the cupin superfamily of
metalloenzymes (35). The crystal structure of the JmjC cat-
alytic domain was solved and found to form an enzymati-
cally active pocket that coordinates the two co-factors that
are needed for the radical-based oxidative demethylation re-
action, ferrous oxide (Fe(II)) and ␣-ketoglutarate (35–37).
The human KDM4 family consists of four members,
KDM4A-D (also known as JMJD2A-D). These enzymes
specically catalyze the demethylation of H3K9me2/me3,
H3K36me2/me3 and H1.4K26me2/me3 in a Fe2+ and ␣-
KG-dependent manner (37). Besides the catalytic JmjC
domain, KDM4 demethylases contain the JmjN domain,
which is also required for the demethylase activity (38–
40). In addition, all KDM4 members, except the shortest
KDM4D protein, contain two Plant homeodomain (PHD)
and two Tudor domains. It was shown that PHD and Tu-
dor domains are not required for KDM4 enzymatic activ-
ity (39,41). On the other hand, it was found that deletion
of JmjN, PHD or Tudor domains resulted in both nuclear
and cytoplasmic localization of KDM4A and KDM4C pro-
teins, suggesting that these domains are essential for the
nuclear localization of KDM4 proteins (39,41). In agree-
ment with this, the Tudor domains of KDM4A isoform
bind H4K20me2, H4K20me3, H3K4me3 and H3K9me3,
while the PHD domains were found to bind H3K4me3 (41–
44). Thus, binding of KDM4A and KDM4C isoforms to
chromatin via their Tudor and PHD domains may assist
with their localization in the nucleus (41). Notably, the Tu-
dor domain is a conserved protein structural motif, which
is also present in many proteins that function in RNA
metabolism, histone modication, DNA damage response,
development, differentiation and cell division (45–49).
KDM4 proteins are involved in a plethora of cellular pro-
cesses including gene expression regulation (50–54), DNA
replication (21,55), DNA damage response (56–58), worm
development and germ cell apoptosis (34), renewal of em-
bryonic stem cells (59), and male lifespan in Drosophila
(60). Interestingly, various types of human cancer show
misregulated expression of KDM4A-D members (61–63).
For example, KDM4C (also known as GASC1) is am-
plied in esophageal squamous carcinomas, medulloblas-
tomas and breast cancer. The depletion of KDM4C in-
hibits proliferation of several tumor cell lines, while overex-
pression of the protein induces transformed phenotypes in
mammary epithelial cells (24,50,64–69). Likewise, KDM4B
overexpression promotes gastric tumorigenesis and its de-
pletion leads to cell-cycle arrest and apoptosis of gastric
cancer cell lines. Additionally, KDM4B depletion causes
a decrease in colony formation and cell proliferation of
estrogen-receptor-positive breast cancer cells and impairs
therefore the development of normal breast tissues in vivo
(70–72). Altogether, these observations suggest that overac-
tivity of KDM4 proteins have a causative role in tumorigen-
esis, and therefore understanding the mechanisms by which
KDM4 proteins promote chromosomal instability becomes
critical.
Here, we determined KDM4A-C subcellular localiza-
tion during mitosis. We reveal a previously unrecognized
differential mitotic localization of KDM4A-C members.
While KDM4C is associated with mitotic chromatin,
KDM4A-B proteins are excluded from chromatin through-
out prometaphase–telophase. Also, we show that KDM4C
Tudor domains are essential for its association with mitotic
chromatin, thus characterizing a novel function of the Tu-
dor domains in recruiting protein to mitotic chromatin. In
addition, we demonstrate that dysregulation of KDM4C
demethylase activity, but not KDM4B, promotes chromo-
some instability (CIN) by impairing the delity of mitotic
chromosome segregation. These observations suggest that
CIN, found in cancer cells driven by KDM4C dysregula-
tion, could result from mitotic chromosome missegrega-
tion.
MATERIALS AND METHODS
Plasmid construction and domain swapping
The constructions of all plasmids used in this study are de-
scribed in Table 1. All point mutations were introduced us-
ing the QuickChange site-directed mutagenesis kit (Strata-
gene). A complete list of all primers and their sequences is
described in Table 2. Domain swapping was performed as
described in Table 1. All constructs used in this study were
veried by nucleotide sequencing or restriction digestion.
Generation of stable cell lines
U2OS-TetON stable cell lines that conditionally express
the fusions EGFP-KDM4B and EGFP-KDM4C were es-
tablished as previously described (73). Cell line expressing
EGFP-KDM4A was generated as follows. A fragment in-
cluding EGFP-KDM4A was subcloned into pTRE2-puro
(Clontech). The resulting pTRE2-puro-Hs-KDM4A vec-
tor was transfected into U2OS-Tet-ON cells (Clontech).
Puromycin-resistant clones (0.6 ␮g/ml Puromycin) were
selected and tested for doxycycline-induced expression of
EGFP using uorescence microscopy. Clones that showed
EGFP expression only after the addition of 1 ␮Mdoxycy-
cline (Sigma, D9891) were selected for further characteriza-
tions.
Cell lines and growth conditions
All cells lines were supplemented with 10% heat-inactivated
fetal bovine serum, 2 mM L-glutamine, 100 ␮g/ml peni-
cillin, 100 ␮g/ml streptomycin and grown in a humidi-
ed incubator containing 5% CO2at 37◦C. U2OS cells
were grown in DMEM. U2OS-Tet-ON cells were grown
in Dulbecco’s modied Eagle’s medium (DMEM) in the
presence of 200 ␮g/ml Geneticin (G418). U2OS-Tet-ON-
EGFP-KDM4A-C cells were grown in DMEM contain-
ing 200 ␮g/ml G418 and either 0.6 ␮g/ml Puromycine (for
KDM4A-B) or 200 ␮g/ml Hygromycin B (for KDM4C).
Transient transfection
Cell transfections with plasmid DNA or siRNA were per-
formed using Poly Jet (Bio-Consult) and Lipofectamine
2000 (Invitrogen), respectively, following the manufac-
turer’s instructions. siRNAs used in this study include
Stealth KDM4B-C siRNA (Invitrogen) and Stealth RNAi

6170 Nucleic Acids Research, 2014, Vol. 42, No. 10
Table 1. Plasmids constructed in this study
Chimera Plasmid name Vector backbone Insert
pEGFP-N1-KDM4C1-708aa pEGFP-N1 digested with NheI,
XhoI
KDM4C1-708aa was amplied from
pEGFP-N1-Hs-KDM4C-wt using F1
and R1
Chimera 1 pEGFP-N1-KDM4C-1-708-
KDM4A-707-1064
aa
pEGFP-N1-KDM4C1-708aa
digested with XhoI, SalI
KDM4A-707-1064 aa was amplied
from pEGFP-N1-Hs-KDM4A-wt
using F3 and R3
pEGFP-N1-KDM4A1-714aa pEGFP-N1 digested with SalI,AfeI KDM4A1-714aa was amplied from
pEGFP-N1-Hs-KDM4A-wt using F2
and R2
Chimera 2 pEGFP-N1-KDM4A-1-714-
KDM4C-695-1056
aa
pEGFP-N1-KDM4A1-714aa
digested with SalI, XmaI
KDM4C-695-1056 aa was amplied
from pEGFP-N1-Hs-KDM4C-wt
using F4 and R4
pEGFP-N1-Hs-KDM4C1-865aa pEGFP-N1 digested with NheI,
XhoI
KDM4C1-865aa was amplied from
pEGFP-N1-Hs-KDM4C-wt using F1
and R5
Chimera 3 pEGFP-N1-Hs-KDM4C1-865-
KDM4A886-1064aa
pEGFP-N1-Hs-KDM4C1-865aa
digested with XhoI, SalI
KDM4A886-1064aa was amplied
from pEGFP-N1-Hs-KDM4A-wt
using F6 and R3
pEGFP-N1-KDM4A1-885aa pEGFP-N1 digested with SalI,AfeI KDM4A1-885aa was amplied from
pEGFP-N1-Hs-KDM4A-wt using F2
and R6
Chimera 4 pEGFP-N1-KDM4A1-885-
KDM4C866-1056aa
pEGFP-N1-KDM4A1-885aa
digested with SalI, XmaI
KDM4C866-1056aa was amplied
from pEGFP-N1-Hs-KDM4C-wt
using F7 and R4
pEGFP-N1-KDM4C1-934 pEGFP-N1 digested with NheI,
XhoI
KDM4C1-934 was amplied from
pEGFP-N1-Hs-KDM4C-wt using F1
and R7
Chimera 5 pEGFP-N1-KDM4C1-934-
KDM4A935-1064aa
pEGFP-N1-KDM4C1-934 digested
with XhoI, SalI
KDM4A935-1064aa was amplied
from pEGFP-N1-Hs-KDM4A-wt
using F8 and R3
pEGFP-N1-KDM4A1-954 pEGFP-N1 digested with SalI,AfeI KDM4A1-954 was amplied from
pEGFP-N1-Hs-KDM4A-wt using F2
and R8
Chimera 6 pEGFP-N1-KDM4A1-954-
KDM4C955-1056aa
pEGFP-N1-KDM4A1-954 digested
with SalI and XmaI
KDM4C955-1056aa was amplied
from pEGFP-N1-Hs-KDM4C-wt
using F9 and R4
pEGFP-N1-KDM4C-RTDF-DNLY pEGFP-N1-KDM4C-WT PCR mutagenesis using primers F10
and R10
pEGFP-N1-KDM4A-DNLY-RTDF pEGFP-N1-KDM4A-WT PCR mutagenesis using primers F11
and R11
pEGFP-N1-KDM4C-R919D pEGFP-N1-KDM4C-WT PCR mutagenesis using primers F12
and R12
negative control. All constructs and siRNA sequences are
available upon request.
Western blotting
Protein lysates were prepared using two different method-
ologies. First, cells were lysed using NP40 lysis buffer (50
mM HEPES pH 7.4, 100 mM NaCl, 0.5% NP-40, 10 mM
EDTA, 20 mM ␤-glycerophosphate, 0.1 mg/ml PMSF, 1.2
mM NaVO4, 5 mM NaF, 1 mM DTT, protease inhibitor
cocktail and 25 ␮g/ml Benzonase (Novagen)) for 30 min on
ice, centrifuged at 14 000 rpm for 25 min at 4◦C, and super-
natant was recovered. Second, protein lysates were prepared
using hot-lysis buffer (1% SDS, 5mM EDTA, 50 mM Tris,
pH7.5), boiled for 15 min, sonicated with two 15 s pulses of
35% amplitude, and centrifuged at 14 000 rpm for 15 min at
room temperature; supernatant was then treated with ben-
zonaze for 30 min, centrifuged and recovered. Protein con-
centration was determined using bicinchoninic acid (BCA)
protein determination reagent (Sigma). Immunoblots were
performed using appropriate antibodies (see Table 3). Mem-
branes were developed using Quantum ECL detection kit
(K-12042-D20, Advansta).
Immunouorescence and Microscopy
Cells were grown on coverslips for 24–48 h before xa-
tion and then washed twice with PBSX1, xed with 4%
paraformaldehyde for 10 min, permeabilized with 0.15%
Triton-X-100 and 0.15% Tween-20 in PBSx1 for 10 min,
blocked with 3% BSA, 0.2% Tween-20 and 0.2% Triton-
X-100 for 1 h at RT, stained with the appropriate pri-
mary antibodies (see Table 3for a complete list of all
antibodies used in this study) for 3 h at 37◦C, washed
three times with wash buffer (0.2% Tween-20 and 0.2%
Triton-X-100 in PBSx1), stained with AlexaFluor488, Alex-
aFluor568 (Molecular Probes; 1:500) or DyLight 649 (Jack-
son ImmunoResearch; 1:500) secondary antibodies for 1 h
at RT in dark and then washed as above. Slides were then
mounted using VECTASHIELD mounting medium with
DAPI (VECTOR) and photographed using an inverted mi-
croscope Confocal Zeiss LSM 700 with 40X oil EC Plan
Neouar objective.

Nucleic Acids Research, 2014, Vol. 42, No. 10 6171
Table 2. Primers used in this study
# Name Sequence
F1 SpeI-KDM4C-F1 GTACTAGTATGGAGGTGGCCGAGGTGGAA
R1 XhoI-KDM4C-R2121 ACTTGTTCCATCCTCCTCGAGGAAGGCATTGGGTGGAG
F2 Eco47III-D2A-F GAGCCTCAGCGCTATGGCTTCTGAGTCTGAAACTCTGAAT
R2 SalI-KDM4A-R2150 CTGCAGCCAGTCGACGTGAAGCACATTTCTGGAATC
F3 XhoI-KDM4A-F2181 TCTACTCCTTATCTCGAGGAGGATGGCACCAGCATAC
R3 SalI-D2A-R CATGTCGACCGCTCCATGATGGCCCGGTATAGTGCAG
F4 SalI-KDM4C-F2083 GCGGTCGACCGAAGAAAATATAGAATATTCTCCACCCAATG
R4 XmaI-D2C-R ATCCCGGGTCTGTCTCTTCTGGCACTTCTTCTGGAAA
F5 SalI-KDM4C-F2083 GCGGTCGACCGAAGAAAATATAGAATATTCTCCACCCAATG
R5 XhoI-KDM4C-R2595 GGCTCGAGCTTATGTCGAAAGCATGTAATGTTCACCAC
F6 XhoI-KDM4A-F2656 GACTCGAGATTCCTAATTTGGAGCGTGCC
R6 SalI-KDM4A-R2655 AGGGTCGACCTTGTGCCGAAAGCAGGTAATGAAG
F7 SalI-KDM4C-F2598 ACGGTCGACGACAACCCCAACGTGAAGTCCAAGG
R7 XhoI-KDM4C-R2802 GGCTCGAGCAGCTTCAGACAGTCTCGGCTCACGATATC
F8 XhoI-KDM4A-F2865 GACTCGAGACTCCTCCTGCTGAAGGGGAAGT
R8 SalI-KDM4A-R2862 AGGGTCGACAAACTGGAGACAGTCCTGGCTCACTA
F9 SalI-KDM4C-F2805 ACGGTCGACGACCCACCTGCTGAGGGAGAAGTCGTC
F10 BamHI-KDM4C-RDTF-
DNLY-F
GTTTGATGATGGATCCTTTAGCGATAACTTATATCCTGAGGATATCGTG
R10 BamHI-KDM4C-RDTF-
DNLY-R
CACGATATCCTCAGGATATAAGTTATCGCTAAAGGATCCATCATCAAAC
F11 BamHI-KDM4A-DNLY-
RTD F - F
ACTTTGATGATGGATCCTTCAGCCGCGATACCTTTCCTGAGGACATAGT
R11 BamHI-KDM4A-DNLY-
RTD F - R
ACTATGTCCTCAGGAAAGGTATCGCGGCTGAAGGATCCATCATCAAAGT
F12 BamHI-KDM4C-R919D-F TGTTTGATGATGGATCCTTTAGCGACGACACATTTCCTGAGGAT
R12 BamHI-KDM4C-R919D-R ATCCTCAGGAAATGTGTCGTCGCTAAAGGATCCATCATCAAACA
R13 EcoR571-KDM4C-S198M GCATGGCACACTGAAGACATGGACCTCTATATGATTAATTATCTCCAC
Table 3. Antibodies used in this study
Name Source
Dilution for western
blot Dilution for IF
Primary antibodies
Anti-H3K9me3 Abcam ab8898 1:2000 1:500
Anti-␤-actin SIGMA #A5441 1:15 000
Anti-H3K36me3 Abcam ab9050 1:3000
Anti-H3K4me3 Abcam ab8580 1:3000
Anti-H3 Abcam ab1791 1:10 000
Anti-KDM4C Santa Cruz #sc-98678 1:1000
Anti-KDM4C Novus NBP149600 1:200
Anti-JMJD2B (KDM4B) Santa Cruz sc-67192 1:1000 1:400
Anti-KDM4A Abcam ab104831 1:250
Anti-GFP Abcam ab290 1:1500
Anti-␣-tubulin Santa Cruz #sc-23948 1:500
Anti-Pericentrin Abcam ab4448 1:500
Secondary antibodies
Donkey anti-mouse-Alexa
Flour R
488
Invitrogen #A21202 1:500
Donkey anti-rabbit-Alexa Flour R
488 Invitrogen #A21206 1:500
Donkey anti-mouse-Alexa
Flour R
568
Invitrogen #A10037 1:500
Donkey anti-rabbit-Alexa Flour R
568 Invitrogen #A10042 1:500
Donkey anti-rabbit DyLightTM649 Jackson ImmunoResearch 1:500
Anti-mouse(IgG)-HRP Amersham 1:10 000
Anti-rabbit(IgG)-HRP Jackson ImmunoResearch
#111-035-003
1:20 000
RESULTS
KDM4C, but not KDM4A-B, protein is associated with mi-
totic chromatin
Revealing the subcellular localization of proteins is vital
for understanding their biological function(s). We sought
therefore to determine the localization of KDM4A-C mem-
bers, which share common domain architecture consisting
of JmjN, JmjC, two PHD and two Tudor domains (Fig-
ure 1A). Toward this, we established U2OS-TetON cell lines
expressing comparable protein levels of EGFP-KDM4A-C
fusions upon the addition of doxycycline (Dox). As shown
in Figure 1B, addition of Dox induces the expression of
EGFP-KDM4A-C fusions, which leads to a severe reduc-
tion in H3K9me3 and H3K36me3 levels. On the other hand,

6172 Nucleic Acids Research, 2014, Vol. 42, No. 10
Figure 1. Differential localization of EGFP-KDM4A-C fusions during mitosis. (A) Schematic structure representation of the KDM4A-D histone demethy-
lases. The KDM4 family consists of four members: KDM4A, KDM4B, KDM4C and KDM4D. All members, except KDM4D, share, in addition to JmjC
and JmjN, two PHD and two Tudor domains. KDM4A-C schematic was built using the SMART software (http://smart.embl-heidelberg.de). (B) West-
ern blot analysis shows that U2OS-TetON cells express comparable levels of KDM4A-C proteins that demethylate H3K9me3 and H3K36me3 but not
H3K4me3. Cells were treated with Dox for 36 h and protein lysates were prepared using hot-lysis and immunoblotted with the indicated antibodies. (C)
Representative images showing the localization of EGFP-KDM4A-C fusions from prometaphase to telophase. U2OS-Tet-ON-EGFP-KDM4A-C cells
were treated with Dox to induce the expression of EGFP-KDM4A-C fusions (green). Cells were stained with DAPI (blue). KDM4A and KDM4B are
excluded from chromatin while KDM4C is associated with mitotic chromatin. Twenty to thirty cells were counted for each of the mitotic stages expressing
either KDM4A or KDM4B. 0ver 100 cells were counted for mitotic cells expressing EGFP-KDM4C. (D) Overexpression of EGFP-KDM4C, but not
EGFP-KDM4A, leads to a severe reduction in H3K9me3 levels. U2OS-Tet-ON-EGFP-KDM4A and U2OS-Tet-ON-EGFP-KDM4C cells were treated
with Dox for 36 h, xed and subjected to immunouorescence analysis using H3K9me3 antibody (yellow). DNA is stained with DAPI (blue) and EGFP-
KDM4A and EGFP-KDM4C are in green. Results shown in (C) and (D) are typical of at least two independent experiments. (E) Representative images
showing the localization of the endogenous KDM4A-C proteins from prometaphase to telophase. MCF7 cells were xed and subjected to immunouo-
rescence analysis using KDM4A-C antibodies.

Nucleic Acids Research, 2014, Vol. 42, No. 10 6173
the H3K4me3 levels were not affected by overexpression of
KDM4A-C proteins.
To determine the localization of EGFP-KDM4A-C fu-
sions during the different stages of mitosis, U2OS-TetON-
EGFP-KDM4A-C cells were treated with Dox for 36 h,
xed and stained with DAPI to visualize mitotic cells. Re-
sults show that KDM4C protein is localized to mitotic chro-
matin from prometaphase to telophase. In striking contrast
to KDM4C, EGFP-KDM4A-B fusions are excluded from
mitotic chromatin (Figure 1C). Interestingly, the levels of
H3K9me3 on mitotic chromatin are severely reduced in cells
overexpressing EGFP-KDM4C comparing to cell overex-
pressing EGFP-KDM4A fusion (Figure 1D).
Next, to assess whether the mitotic localization of the
endogenous KDM4A-C proteins is similar to their over-
expressed EGFP-fused forms, we rst tested the suitabil-
ity of commercial KDM4A-C antibodies to detect the na-
tive forms of KDM4A-C proteins by immunouorescence
analysis. U2OS cells were transfected with expression con-
structs expressing EGFP-KDM4A-C fusions (green) and
immunostained with KDM4A-C antibodies (red). Results
show that the intensity of the red signal in cells express-
ing the EGFP-KDM4A-C fusions is much higher than
the untransfected cells (Supplementary Figure S1). This re-
sult conrms that these antibodies detect KDM4A-C pro-
teins in cells and can be used for immunouorescence-
based studies. MCF7 cells were then immunostained us-
ing KDM4A-C antibodies to detect their mitotic local-
ization. Results show that, similar to the localization of
EGFP-KDM4A-C fusions, the endogenous KDM4A-B are
excluded from mitotic chromatin, while KDM4C protein
is associated with chromatin during the different mitotic
stages (Figure 1E). Altogether, these observations demon-
strate for the rst time that, unlike KDM4A and KDM4B,
KDM4C is associated with mitotic chromatin and triggers
the demethylation of H3K9me3 mark.
The Tudor domains of KDM4C mediate its localization to
mitotic chromatin
To map KDM4C region that mediates its localization to mi-
totic chromatin, we performed domain-swapping analysis
between KDM4A and KDM4C proteins. First, we swapped
the regions containing the two PHD and the two Tudor do-
mains between KDM4A and KDM4C proteins. As a re-
sult, two chimeras were produced (Table 1): chimera1 en-
codes the rst 708 amino acids of KDM4C protein, which
includes the JmjN and JmjC domain, fused to the last 357
amino acids of the KDM4A containing the two PHD and
the two Tudor domains. Chimera2 is the reciprocal chimera,
which encodes the rst 714 amino acids of KDM4A pro-
tein and the last 361 amino acids of the KDM4C pro-
tein containing the two PHD and the two Tudor domains.
U2OS cells were transfected with expression vectors encod-
ing chimera1 and 2 and the mitotic localization was deter-
mined as described in Figure 1B. Results show that while
chimera1 shows chromatin-excluded localization (Figure
2A), chimera2 exhibits chromatin-bound localization dur-
ing mitosis (Figure 2B). Together, these results suggest that
the C-terminal region containing the two PHD and the two
Figure 2. The C-terminal region of KDM4C mediates its association with
mitotic chromatin. Panels (A)–(D) show that the C-terminus of KDM4C,
containing the two Tudor domains, is essential and sufcient for its asso-
ciation with mitotic chromatin. Panels (E)–(F) show that the distal Tudor
domain is essential but not sufcient for the localization of KDM4C at
mitotic chromatin. In panels (A)–(F), U2OS cells were transfected with
expression constructs encoding the indicated chimeras fused to EGFP
(green). DNA is stained with DAPI (blue). Results are typical of 2–3 differ-
ent experiments and each image represents at least 10 different cells. Image
acquisition and scoring were performed by a student who was blind of the
experimental condition.
Tudor domains mediate the distinct mitotic localization of
KDM4A and KDM4C proteins.
To identify the domain that mediates the association of
KDM4C with mitotic chromatin, we repeated the domain-
swapping analysis and exchanged the regions contain-
ing only the two Tudor domains between KDM4A and
KDM4C proteins. This analysis produced chimera3 and
chimera4 (Table 1). Chimera3 encodes the rst 865 amino
acids of KDM4C protein and the last 178 amino acids of
KDM4A that contain the two Tudor domains. Chimera4
encodes the rst 885 amino acids of KDM4A protein and
the last 190 of KDM4C protein. We observed that chimera3
shows chromatin-excluded localization (Figure 2C) and
chimera4 has chromatin-bound localization during mito-

6174 Nucleic Acids Research, 2014, Vol. 42, No. 10
sis (Figure 2D). These results show that the replacement
of KDM4C Tudor domains with those of KDM4A protein
leads to its exclusion from mitotic chromatin. In addition,
KDM4C C-terminal region, consisting of the two Tudor
domains, leads to the association of KDM4A with mitotic
chromatin. Collectively, we concluded that the KDM4C C-
terminus, containing the two Tudor domains, is essential
and sufcient for its association with mitotic chromatin.
To determine whether both Tudor domains are re-
quired for KDM4C mitotic localization, we swapped the
C-terminus region containing the distal Tudor domain be-
tween KDM4C and KDM4A (Table 1). Chimera5, which
encodes the rst 934 amino acids of KDM4C fused with
the last 129 amino acid containing the distal Tudor do-
main of KDM4A, is excluded from mitotic chromatin (Fig-
ure 2E). On the other hand, chimera6 that encodes the rst
954 amino acids of KDM4A fused to 101 amino acids of
KDM4C, which includes its distal Tudor domain, remains
excluded from chromatin (Figure 2F). We concluded there-
fore that the C-terminus of KDM4C containing the distal
Tudor domain is essential but not sufcient for its mitotic
chromatin localization.
Mapping candidate residues within KDM4C Tudor domains
that regulate its localization to mitotic chromatin
Domain-swapping analyses suggest that the localization of
KDM4C at mitotic chromatin is mediated by its Tudor do-
mains (Figure 2). To map residues within the Tudor do-
mains of KDM4C that regulate its association with mitotic
chromatin, we performed sequence alignment of KDM4A-
C proteins using MUSCLE software and searched for
residues that are conserved between KDM4A and KDM4B
proteins but not in KDM4C (Figure 3A). Noticeably, com-
parison of the amino acid sequences shows that the proxi-
mal Tudor domain of KDM4A and KDM4B contains four
identical amino acids, DNLY, which appear as RDTF in
KDM4C protein (corresponds to 919–922 amino acids).
On this basis, we speculated that these four residues might
be implicated in regulating KDM4C localization at mitotic
chromatin. Site-directed mutagenesis was used to substi-
tute RDTF residues of KDM4C with DNLY, and the mi-
totic localization of KDM4CRDTF/DNLY mutant was deter-
mined. Results show that EGFP-KDM4CRDTF/DNLY mu-
tant is excluded from mitotic chromatin (Figure 3B). We
concluded therefore that the RDTF residues are critical
for KDM4C association with mitotic chromatin. Next, we
sought to address whether RDTF residues are sufcient
for KDM4C localization at mitotic chromatin. To do so,
we substituted the DNLY (corresponds to 939–942 amino
acids) of KDM4A with RDTF. Results show that, similar
to the wild-type KDM4A, KDM4ADNLY/RDTF mutant re-
mains excluded from mitotic chromatin (Figure 3C). Alto-
gether, these observations suggest that RDTF residues are
required, but not sufcient, for the localization of KDM4C
at mitotic chromatin.
Interestingly, it was recently shown that KDM4AD939
residue is essential for the binding of the KDM4A Tudor
domain with methylated H4K20me2 (56). Therefore, we
sought to address whether this residue is also involved in
regulating KDM4A and KDM4C association with mitotic
chromatin. Toward this, we generated KDM4CR919D.Re-
sults show that this mutant is excluded from mitotic chro-
matin (Figure 3D). This observation suggests that, simi-
lar to the RDTF residues, KDM4CR919 residue is essen-
tial for KDM4C localization at mitotic chromatin. Collec-
tively, our data strongly suggest that sequences at the two
Tudor domains are required for the distinct localization of
KDM4C at mitotic chromatin.
Dysregulation of KDM4C expression promotes mitotic chro-
mosome missegregation
The localization of KDM4C on mitotic chromatin raises a
possibility that it might be implicated in regulating chro-
mosome segregation. To assess this possibility, we looked
at four abnormal mitotic phenotypes in cells overexpress-
ing or depleted of either KDM4B or KDM4C. The abnor-
mal phenotypes include misaligned chromosomes during
metaphase (Figure 4A), lagging chromosomes, anaphase–
telophase bridges (Figure 4B and C) and multiple centro-
somes (Figure 4D). To deplete KDM4B-C, U2OS cells were
transfected with KDM4B-C siRNA sequences (KDM4C
siRNA #46, #58 and #59; KDM4B #06). Western blot re-
veals that all siRNA sequences targeting KDM4B-C show
severe reduction in the protein levels compared to control
siRNA (Figure 4E).
KDM4C-depleted cells were subjected to immunouo-
rescence analysis using ␣-tubulin and ␣-Pericentrin anti-
bodies and DNA was stained with DAPI to allow the iden-
tication of mitotic cells. Metaphase cells with misaligned
chromosomes were counted for each KDM4C siRNA se-
quences and divided by the total number of metaphase cells
(n=448 total metaphases). Results show that KDM4C
depletion using either one of the three siRNA sequences
leads to over 3-fold increase in percentage of metaphase
cells with misaligned chromosomes compared to cell trans-
fected with control siRNA (n=107 metaphase cells) (Fig-
ure 4F). Similar increase was also obtained in anaphase–
telophase cells with lagging chromosomes or anaphase–
telophase bridges (n=272 anaphase–telophase cells) com-
pared to control cells (n=101) (Figure 4G). Next, we
sought to address whether KDM4C depletion affects cen-
trosome number. Mitotic cells were analyzed based on ␣-
tubulin and Pericentrin staining. Results show no detectable
changes in the percentage of mitotic cells with multiple spin-
dle poles between control and KDM4C-depleted cells (mul-
tipolar spindle pole formation was present in 0.5±0.4% of
KDM4C-depleted cells and 0.4±0.3% of control cells). In-
terestingly, KDM4B-depleted cells show no signicant in-
crease in the percentage of abnormal mitotic cells (Figure
4F and G). Altogether, these observations show for the rst
time that depletion of KDM4C, but not KDM4B, affects
the delity of mitotic chromosome segregation, therefore
suggesting that CIN in cancers lacking KDM4C can result
in part from mitotic chromosome missegregation.
Importantly, KDM4B-C members are overexpressed in
several types of human cancer and its depletion impairs can-
cer cell proliferation (24,50,61,64–69,71,74,75). We sought
therefore to address whether KDM4B-C overexpression
also affects mitotic chromosome missegregation. To do so,
U2OS-TetON cells were treated with Dox for 72 h to in-

Nucleic Acids Research, 2014, Vol. 42, No. 10 6175
Figure 3. KDM4C-R919 residue is essential but not sufcient for KDM4C association with mitotic chromatin. (A) Multiple sequence alignment of the
KDM4A-C Tudor domains was performed using the MUSCLE software. The color-coding is based on the Zapo color scheme. Black arrowhead indicates
amino acid residues that are conserved in KDM4A and KDM4B isoforms, and are distinct in KDM4C. (B)–(D) U2OS cells were transfected with expression
constructs encoding the indicated KDM4A and KDM4C mutants fused to EGFP. Results are typical of two independent experiments and each image
represents at least 10 different cells.

6176 Nucleic Acids Research, 2014, Vol. 42, No. 10
Figure 4. KDM4C, but not KDM4B, depletion increases chromosomalsegregation errors during mitosis. (A)–(D) Representative images showing normal
and defective mitotic U2OS cells depleted of KDM4C. Cells were subjected to immunouorescence analysis using Pericentrin (green) and ␣-tubulin
antibodies (red). DNA is stained with DAPI (blue). (A) Normal and abnormal metaphase with misaligned chromosome (indicated by white arrow). (B)
Normal and abnormal anaphase with either lagging chromosomes or anaphase bridge (indicated by white arrows). (C) Abnormal telophase with either
lagging chromosomes or telophase bridge (indicated by white arrows). (D) Multipolar metaphase. (E) KDM4B and KDM4C knockdown by western
blotting. U2OS cells were transfected with either control or different sequences of KDM4B and KDM4C Stealth siRNA (Invitrogen). Protein extracts
were prepared 72 h after transfection and immunoblotted with KDM4B and KDM4C antibody. ␤-Actin is used as a loading control. (F) A histogram
showing the percentage of metaphases with misaligned chromosomes 72 h after transfection with control and different KDM4B-C siRNA sequences.
KDM4C, but not KDM4B, depletion increases the frequency of metaphase cells with misaligned chromosomes. n, number of metaphase cells counted.
Error bars represent standard deviation from two independent experiments. (G) KDM4C, but not KDM4B, depletion increases the frequency of anaphase–
telophase cells with either lagging chromosomes or anaphase–telophase bridges. As in (F), except that the histogram shows the percentage of defective
anaphase–telophase cells.
duce the expression of EGFP-KDM4C or EGFP-KDM4B
fusions. Cells were then xed and stained with ␣-tubulin,
␣-Pericentrin and DAPI to determine the frequency of ab-
normal mitotic cells. Results show that Dox treatment of
U2OS-TetON-EGFP-KDM4C cells leads to 3.8-fold in-
crease in metaphase cells with misaligned chromosomes
(Figure 5A) compared to Dox-untreated cells (Figure 5B).
Likewise, the frequency of cells with lagging chromosomes
and anaphase–telophase bridges (Figure 5A) was increased
by 3-fold following the expression of EGFP-KDM4C fu-
sion (Figure 5C). We concluded that similar to KDM4C
knockdown, upregulation of KDM4C promotes mitotic

Nucleic Acids Research, 2014, Vol. 42, No. 10 6177
Figure 5. KDM4C, but not KDM4B, overexpression promotes mitotic chromosome missegregation. (A) Representative images of defective mitotic cells
overexpressing EGFP-KDM4C fusion. U2OS-TetON-EGFP-KDM4C cells were treated with Dox for 72 h to induce the expression of EGFP-KDM4C
(green). Cells were then stained for Pericentrin (red) and ␣-tubulin antibodies (gray). DNA is stained with DAPI (blue). (B)and(C) Histograms showing
the percentage of metaphases with misaligned chromosomes (B) and anaphase–telophase cells that exhibit lagging chromosomes or anaphase–telophase
bridges (C). Untreated and Dox-treated U2OS-TetON-EGFP-KDM4C and U2OS-TetON-EGFP-KDM4B cells were subjected to immunouorescence
and mitotic cells were acquired using confocal microscope. n, number of mitotic cells counted. Error bars represent standard deviation from three and two
independent experiments of cells expressing KDM4C and KDM4B, respectively.
chromosome missegregation that can potentially lead to
CIN found in cancer driven by KDM4C overexpression.
Similar to KDM4C depletion, no signicant changes were
observed in the frequency of multipolar mitotic cells after
the addition of Dox (Dox-untreated cells show 0.5±0.3%
and Dox-treated cells show 0.5±0.4%). Importantly, the
increase in the percentage of abnormal mitotic cells was
not observed in cells overexpressing EGFP-KDM4B fusion
(Figure 5B and C). Collectively, our data suggest that both
up- and downregulation of KDM4C have similar effect on
the delity of mitotic chromosome segregation.
The demethylase activity and the mitotic localization of
KDM4C inuence the integrity of mitotic chromosome seg-
regation
To gain further insights into KDM4C mitotic function,
we sought to address whether its demethylase activity is
implicated in regulating chromosome segregation. Toward

6178 Nucleic Acids Research, 2014, Vol. 42, No. 10
this, we generated KDM4C ‘demethylase-dead’ mutant.
As we have previously reported (76), Ser198Met mutation,
within the JmjC of KDM4C, is expected to abolish an ex-
isting hydrogen-bond network, disrupting the coordination
of ␣-KG within the catalytic site and consequently abro-
gating the demethylase activity. Indeed, western blot and
immunouorescence analysis show that overexpression of
KDM4C-S198M in U2OS cells has no effect on H3K9me3
levels, whereas overexpression of wild-type KDM4C leads
to a sever decrease in the levels of H3K9me3 mark (Figure
6A and B). Next, we looked at abnormal mitosis in cells
overexpressing EGFP-KDM4C-S198M demethylase-dead
mutant. Results show no detectable effect on the percentage
of cells showing abnormal chromosome segregation (Figure
6C and D). This observation suggests that dysregulation of
KDM4C demethylase activity disrupt the delity of mitotic
chromosome segregation.
Next, we sought to address whether the defective chromo-
some segregation in cells overexpressing KDM4C is due to
its localization on mitotic chromatin. To this end, we over-
expressed EGFP-KDM4C-R919D mutant, which is not as-
sociated with mitotic chromatin (Figure 3D), and deter-
mined the percentage of abnormal mitotic cells. Results
show that EGFP-KDM4C-R919D overexpression has no
signicant increase in the percentage of metaphases with
misaligned chromosomes (Figure 6C). On the other hand,
it led to 1.8-fold increase in the percentage of anaphase–
telophase cells showing lagging chromosomes or anaphase–
telophase bridge (Figure 6D). Altogether, we made two in-
teresting conclusions. The rst is that the defective mitotic
chromosome segregation in cells overexpressing or depleted
of KDM4C is due to alteration in KDM4C demethylase
activity. The second is that part of the abnormal mitotic
phenotype (lagging chromosomes and anaphase–telophase
bridges) is independent of its localization on mitotic chro-
matin.
DISCUSSION
This study shows a unique localization of KDM4C on mi-
totic chromatin that differs from the closely related mem-
bers, KDM4A and KDM4B, which are excluded from chro-
matin during mitosis (Figure 1). This differential mitotic
localization of KDM4A-C suggests that they might have
distinct functions during mitosis. Similar to our observa-
tions, it was reported that the three different isoforms of the
heterochromatin protein (HP1␣,HP1␤and HP1␥), which
share the same domain architecture, are localized differ-
ently during interphase and mitosis. HP1␣is the only iso-
form that remains associated with chromatin during mito-
sis, while HP1␤and HP1␥showed different mitotic distri-
bution compared to their nuclear localization during inter-
phase (77,78). In agreement with this, HP1␣isoform plays a
central role in mitotic chromosome segregation by regulat-
ing Aurora B activation and dissociation from chromosome
arms (79). Furthermore, protein phosphatase-1 (PP-1) iso-
forms ␣,␥1and␦, which share nearly identical catalytic
domains, localize differently during interphase and mitosis
suggesting unique roles for each of the PP-1isoforms during
the different cell-cycle stages (80).
Our data suggest that the two Tudor domains are in-
volved in the regulation of KDM4C localization during mi-
tosis. Interestingly, the Tudor domain was shown to me-
diate the subcellular localization of proteins. For example,
the Tudor domain of TDRD3 is both required and suf-
cient for its localization to stress granules, which are cyto-
plasmic structures involved in RNA metabolism (81). Sim-
ilarly, Tudor domain-containing protein, Yb, which is re-
quired for the primary processing of piRNAs and transpo-
son repression, localizes to a cytoplasmic structure called
the Yb body via its Tudor domain (49). Also, the Tudor do-
main mediates the recruitment of various proteins to chro-
matin. For instance, it was found that foci formation of
53BP1 protein after DNA damage is mediated by the bind-
ing of its Tudor domain to H4K20me2 mark (56,82). More-
over, it was shown that KDM4A is guided by its Tudor do-
main to H3K4me3 and H4K20me3 regions to demethylate
H3K9me3 and H3K36me3 methyl marks (83,84). Here, we
further expand the function of the Tudor domain by charac-
terizing a previously unrecognized role of the Tudor domain
in recruiting KDM4C to mitotic chromatin. Future work
will be required to identify the mechanism by which the Tu-
dor domains regulate KDM4C association with chromatin
during mitosis.
In addition, we reveal a novel role of KDM4C in
regulating mitotic chromosome segregation. Our results
show that the levels of KDM4C protein are critical for
the proper chromosome segregation. KDM4C, but not
KDM4B, knockdown or overexpression increases the fre-
quency of abnormal mitotic cells showing misaligned chro-
mosomes during metaphase, anaphase bridge and chromo-
some lagging. Further, overexpression of KDM4C-S198M
demethylase-dead mutant has no detectable effect on the -
delity of chromosome segregation. These results imply that
CIN can result from loss or gain of KDM4C demethylase
activity. In accord with this, recent analysis of the Cancer
Genome Atlas revealed that KDM4C is lost in some cancer
types and overexpressed in others (62). It should be noted
however that we cannot exclude the existence of additional
yet unknown mechanisms that contribute to CIN found in
cancer driven by either lack or overactivity of KDM4C ly-
sine demethylase. In this regard, it was recently shown that
KDM4A overexpression induces copy number gains of spe-
cic genomic regions which are known to contain onco-
genes (62).
Determining the cellular localization of protein is often
a crucial step toward understanding its biological functions
(85). Our data showing that KDM4C is localized at mitotic
chromatin and promotes chromosome segregation is in ac-
cordance with the mitotic localization pattern of other pro-
teins, which are associated with mitotic chromatin and play
a role in chromosome condensation and sister chromatid
separation. For example, the Tudor-domain protein EKL-1
is localized at mitotic chromatin and promotes chromosome
segregation (86). Likewise, Aurora B kinase, which is associ-
ated with mitotic chromosomes, is involved in chromosome
segregation and cytokinesis (87,88).
How does KDM4C misregulation disrupt the delity
of mitotic chromosome segregation? There are three main
possibilities which are not necessarily mutually exclusive.
First, KDM4C serves as a scaffold protein for recruit-

Nucleic Acids Research, 2014, Vol. 42, No. 10 6179
Figure 6. KDM4C demethylase activity and mitotic localization affect the delity of chromosome segregation during mitosis. (A)and(B) show the effect
of S198M mutation on the demethylase activity of KDM4C protein. (A) western blot analysis showing that overexpression of EGFP-KDM4C-S198M has
no detectable effect on the levels of H3K9me3. Protein extracts were prepared from U2OS-TetON cells expressing either EGFP-KDM4C-WT or EGFP-
KDM4C-S198M and immunoblotted using the indicated antibodies. (B) Immunouorescence analysis of U2OS-TetON expressing either EGFP-KDM4C-
WT (bottom) or EGFP-KDM4C-S198M (top). Cells were stained for H3K9me3 (red). DNA is stained with DAPI (blue), and the EGFP-KDM4C is in
green. (C)and(D) Histograms showing the percentage of metaphases with misaligned chromosomes (B) and anaphase–telophase cells that exhibit lagging
chromosomes or anaphase–telophase bridges (C). U2OS-TetON cells expressing EGFP-KDM4C-S198M or EGFP-KDM4C-R919D were subjected to
immunouorescence and mitotic cells were acquired using confocal microscope. n, number of mitotic cells counted. Error bars represent standard deviation
from two independent experiments.
ing other proteins that are required for proper chromo-
some segregation. In accord with this, cells overexpress-
ing KDM4C-R919D mutant (which does not localize to
mitotic chromatin) show no increase in the percentage of
metaphases with misaligned chromosome (Figure 6C). Sec-
ond, KDM4C regulates the activity of non-histone proteins,
which are involved in regulation of chromosome segrega-
tion, through demethylating their lysine residues. In sup-
port of this, it was shown that the polycomb protein, Pc2,
is a substrate of KDM4C. Interestingly, Pc2 is SUMO E3
ligase that promotes sumoylation of multiple proteins (89),
a modication which is essential for proper chromosomes
segregation (90). Third, by demethylating KDM4C histone
substrates such as methylated H3K9 or H3K36 residues. In
agreement with this, it was shown that the level of H3K9me3
mark decreases as cells exit mitosis, during the period be-
tween anaphase and cytokinesis. This decrease is essential
for chromosome congression and segregation. It was also

6180 Nucleic Acids Research, 2014, Vol. 42, No. 10
found that H3K9me3-decient cells exhibit a wide range
of abnormal mitotic phenotypes such as an increase in
misaligned and lagging chromosomes, which leads to ane-
uploidy, nondisjunction and the appearance of micronu-
clei at cytokinesis or early G1 (91). Likewise, it was previ-
ously shown that loss of H3K9 methyltransferase, Suv39h,
or overexpression of H3K9 demethylase KDM4B leads to
CIN (18,92). Collectively, our data suggest that denite
methylation levels of KDM4C substrates might be required
to ensure proper chromosome segregation. Nonetheless,
further studies will be required to address how alterations
in KDM4C levels promote chromosome missegregation.
SUPPLEMENTARY DATA
Supplementary Data are available at NAR Online.
ACKNOWLEDGEMENTS
We are grateful to Fabian Glaser for his help in preparing
Figure 3A and to Arnon Henn for critical reading of the
manuscript. We thank Maayan Duvshani-Eshet and Nitzan
Dahan from the Life Sciences and Engineering Infrastruc-
ture Unit at the Technion, for their help in the microscopy.
FUNDING
Israel Cancer Research Fund [2015211]; Israel Science
Foundation [2014673]; Israel Cancer association [2015573];
Rubin Scientic and Medical Research Fund [2018025].
Funding for open access charge: Israel Cancer Research
Fund; Israel Science Foundation; Israel Cancer association;
Rubin Scientic and Medical Research Fund.
Author contributions. I. K. performed most of the exper-
iments and analyzed the data. H. K. performed the experi-
ments shown in Figure 1E, Figures 2E–F, Figures 3B–D and
part of the data of Figures 4E–F, helped in writing the Ma-
terials and Methods and proofreading the manuscript. S. W.
A. performed the experiments shown in Figure 1B and part
of the data shown in Figure 5B–C. N. G. R. helped in ana-
lyzing and interpreting the data and reading the manuscript.
N. A. conceived the study, planned the experiments and
wrote the paper.
Conict of interest statement. None declared.
REFERENCES
1. Flemming,W. (1882) Zellsubstanz, Kern und Zelltheilung.W.Vogel,
Leipzig.
2. Berger,S.L. (2007) The complex language of chromatin regulation
during transcription. Nature,447, 407–412.
3. Skene,P.J. and Henikoff,S. (2013) Histone variants in pluripotency
and disease. Development,140, 2513–2524.
4. Zentner,G.E. and Henikoff,S. (2013) Regulation of nucleosome
dynamics by histone modications. Nat.Struct.Mol.Biol.,20,
259–266.
5. Ray-Gallet,D. and Almouzni,G. (2010) Nucleosome dynamics and
histone variants. Essays Biochem.,48, 75–87.
6. Li,B., Carey,M. and Workman,J.L. (2007) The role of chromatin
during transcription. Cell,128, 707–719.
7. Neely,K.E. and Workman,J.L. (2002) The complexity of chromatin
remodeling and its links to cancer. Biochim. Biophys. Acta.,1603,
19–29.
8. Wang,G.G., Allis,C.D. and Chi,P. (2007) Chromatin remodeling and
cancer, Part II: ATP-dependent chromatin remodeling. Trends Mol.
Med.,13, 373–380.
9. Badeaux,A.I. and Shi,Y. (2013) Emerging roles for chromatin as a
signal integration and storage platform. Nat.Rev.Mol.CellBiol.,14,
211–224.
10. Kouzarides,T. (2007) Chromatin modications and their function.
Cell,128, 693–705.
11. Johnson,D.G. and Dent,S.Y. (2013) Chromatin: receiver and
quarterback for cellular signals. Cell,152, 685–689.
12. Yun,M., Wu,J., Workman,J.L. and Li,B. (2011) Readers of histone
modications. Cell Res., 21, 564–578.
13. Martin,C. and Zhang,Y. (2005) The diverse functions of histone
lysine methylation. Nat.Rev.Mol.CellBiol.,6, 838–849.
14. Grunstein,M. (1997) Histone acetylation in chromatin structure and
transcription. Nature,389, 349–352.
15. Li,F., Huarte,M., Zaratiegui,M., Vaughn,M.W., Shi,Y.,
Martienssen,R. and Cande,W.Z. (2008) Lid2 is required for
coordinating H3K4 and H3K9 methylation of heterochromatin and
euchromatin. Cell,135, 272–283.
16. Margueron,R. and Reinberg,D. (2010) Chromatin structure and the
inheritance of epigenetic information. Nat. Rev. Genet.,11, 285–296.
17. Richards,E.J. and Elgin,S.C. (2002) Epigenetic codes for
heterochromatin formation and silencing: rounding up the usual
suspects. Cell,108, 489–500.
18. Peters,A.H., O’Carroll,D., Scherthan,H., Mechtler,K., Sauer,S.,
Schofer,C., Weipoltshammer,K., Pagani,M., Lachner,M.,
Kohlmaier,A. et al. (2001) Loss of the Suv39h histone
methyltransferases impairs mammalian heterochromatin and genome
stability. Cell,107, 323–337.
19. Cheutin,T., McNairn,A.J., Jenuwein,T., Gilbert,D.M., Singh,P.B. and
Misteli,T. (2003) Maintenance of stable heterochromatin domains by
dynamic HP1 binding. Science,299, 721–725.
20. Schotta,G., Lachner,M., Sarma,K., Ebert,A., Sengupta,R.,
Reuter,G., Reinberg,D. and Jenuwein,T. (2004) A silencing pathway
to induce H3-K9 and H4-K20 trimethylation at constitutive
heterochromatin. Genes Dev., 18, 1251–1262.
21. Black,J.C., Van Rechem,C. and Whetstine,J.R. (2012) Histone lysine
methylation dynamics: establishment, regulation, and biological
impact. Mol. Cell,48, 491–507.
22. Zhang,X., Wen,H. and Shi,X. (2012) Lysine methylation: beyond
histones. Acta Biochim. Biophys. Sin (Shanghai),44, 14–27.
23. Greer,E.L. and Shi,Y. (2012) Histone methylation: a dynamic mark in
health, disease and inheritance. Nat. Rev. Genet.,13, 343–357.
24. Northcott,P.A., Nakahara,Y., Wu,X., Feuk,L., Ellison,D.W.,
Croul,S., Mack,S., Kongkham,P.N., Peacock,J., Dubuc,A. et al.
(2009) Multiple recurrent genetic events converge on control of
histone lysine methylation in medulloblastoma. Nat. Genet., 41,
465–472.
25. Pedersen,M.T. and Helin,K. (2010) Histone demethylases in
development and disease. Trends Cell Biol., 20, 662–671.
26. Hake,S.B., Xiao,A. and Allis,C.D. (2007) Linking the epigenetic
‘language’ of covalent histone modications to cancer. Br.J.Cancer,
96 Suppl, R31–R39.
27. Hess,J.L. (2004) MLL: a histone methyltransferase disrupted in
leukemia. Trends Mol. Med.,10, 500–507.
28. Okada,Y., Feng,Q., Lin,Y., Jiang,Q., Li,Y., Cofeld,V.M., Su,L.,
Xu,G. and Zhang,Y. (2005) hDOT1L links histone methylation to
leukemogenesis. Cell,121, 167–178.
29. Schneider,R., Bannister,A.J. and Kouzarides,T. (2002) Unsafe SETs:
histone lysine methyltransferases and cancer. Trends Biochem. Sci.,
27, 396–402.
30. Varambally,S., Dhanasekaran,S.M., Zhou,M., Barrette,T.R.,
Kumar-Sinha,C., Sanda,M.G., Ghosh,D., Pienta,K.J., Sewalt,R.G.,
Otte,A.P. et al. (2002) The polycomb group protein EZH2 is involved
in progression of prostate cancer. Nature,419, 624–629.
31. Fraga,M.F., Ballestar,E., Villar-Garea,A., Boix-Chornet,M.,
Espada,J., Schotta,G., Bonaldi,T., Haydon,C., Ropero,S., Petrie,K.
et al. (2005) Loss of acetylation at Lys16 and trimethylation at Lys20
of histone H4 is a common hallmark of human cancer. Nat. Genet,
37, 391–400.
32. Shi,Y., Lan,F., Matson,C., Mulligan,P., Whetstine,J.R., Cole,P.A.,
Casero,R.A. and Shi,Y. (2004) Histone demethylation mediated by
the nuclear amine oxidase homolog LSD1. Cell,119, 941–953.

Nucleic Acids Research, 2014, Vol. 42, No. 10 6181
33. Klose,R.J., Kallin,E.M. and Zhang,Y. (2006)
JmjC-domain-containing proteins and histone demethylation. Nat.
Rev. Genet.,7, 715–727.
34. Whetstine,J.R., Nottke,A., Lan,F., Huarte,M., Smolikov,S., Chen,Z.,
Spooner,E., Li,E., Zhang,G., Colaiacovo,M. et al. (2006) Reversal of
histone lysine trimethylation by the JMJD2 family of histone
demethylases. Cell,125, 467–481.
35. Tsukada,Y., Fang,J., Erdjument-Bromage,H., Warren,M.E.,
Borchers,C.H., Tempst,P. and Zhang,Y. (2006) Histone
demethylation by a family of JmjC domain-containing proteins.
Nature,439, 811–816.
36. Shi,Y. and Whetstine,J.R. (2007) Dynamic regulation of histone
lysine methylation by demethylases. Mol. Cell,25, 1–14.
37. Chen,Z., Zang,J., Whetstine,J., Hong,X., Davrazou,F.,
Kutateladze,T.G., Simpson,M., Mao,Q., Pan,C.H., Dai,S. et al.
(2006) Structural insights into histone demethylation by JMJD2
family members. Cell,125, 691–702.
38. Fodor,B.D., Kubicek,S., Yonezawa,M., O’Sullivan,R.J., Sengupta,R.,
Perez-Burgos,L., Opravil,S., Mechtler,K., Schotta,G. and Jenuwein,T.
(2006) Jmjd2b antagonizes H3K9 trimethylation at pericentric
heterochromatin in mammalian cells. Genes Dev., 20, 1557–1562.
39. Klose,R.J., Yamane,K., Bae,Y., Zhang,D., Erdjument-Bromage,H.,
Tempst,P., Wong,J. and Zhang,Y. (2006) The transcriptional repressor
JHDM3A demethylates trimethyl histone H3 lysine 9 and lysine 36.
Nature,442, 312–316.
40. Tan,H., Wu,S., Wang,J. and Zhao,Z.K. (2008) The JMJD2 members
of histone demethylase revisited. Mol. Biol. Rep. ,35 ,551–556 .
41. Shin,S. and Janknecht,R. (2007) Diversity within the JMJD2 histone
demethylase family. Biochem. Biophys. Res. Commun.,353, 973–977.
42. Hillringhaus,L., Yue,W.W., Rose,N.R., Ng,S.S., Gileadi,C.,
Loenarz,C., Bello,S.H., Bray,J.E., Schoeld,C.J. and Oppermann,U.
(2012) Structural and evolutionary basis for the dual substrate
selectivity of human KDM4 histone demethylase family. J. Bio l .
Chem.,286, 41616–41625.
43. Horton,J.R., Upadhyay,A.K., Qi,H.H., Zhang,X., Shi,Y. and
Cheng,X. (2010) Enzymatic and structural insights for substrate
specicity of a family of jumonji histone lysine demethylases. Nat.
Struct.Mol.Biol.,17, 38–43.
44. Yu,L., Wang,Y., Huang,S., Wang,J., Deng,Z., Zhang,Q., Wu,W.,
Zhang,X., Liu,Z., Gong,W. et al. (2010) Structural insights into a
novel histone demethylase PHF8. Cell Res.,20, 166–173.
45. Chen,C., Nott,T.J., Jin,J. and Pawson,T. (2011) Deciphering arginine
methylation: Tudor tells the tale. Nat.Rev.Mol.CellBiol.,12,
629–642.
46. Huyen,Y., Zgheib,O., Ditullio,R.A. Jr, Gorgoulis,V.G.,
Zacharatos,P., Petty,T.J., Sheston,E.A., Mellert,H.S., Stavridi,E.S.
and Halazonetis,T.D. (2004) Methylated lysine 79 of histone H3
targets 53BP1 to DNA double-strand breaks. Nature,432, 406–411.
47. Lasko,P. (2010) Tudor domain. Curr. Biol.,20, R666–667.
48. Siomi,M.C., Mannen,T. and Siomi,H. (2010) How does the royal
family of Tudor rule the PIWI-interacting RNA pathway? Genes
Dev., 24, 636–646.
49. Pek,J.W., Anand,A. and Kai,T. (2012) Tudor domain proteins in
development. Development,139, 2255–2266.
50. Wissmann,M., Yin,N., Muller,J.M., Greschik,H., Fodor,B.D.,
Jenuwein,T., Vogler,C., Schneider,R., Gunther,T., Buettner,R. et al.
(2007) Cooperative demethylation by JMJD2C and LSD1 promotes
androgen receptor-dependent gene expression. Nat.CellBiol.,9,
347–353.
51. Shin,S. and Janknecht,R. (2007) Activation of androgen receptor by
histone demethylases JMJD2A and JMJD2D. Biochem. Biophys. Res.
Commun.,359, 742–746.
52. Zhang,D., Yoon,H.G. and Wong,J. (2005) JMJD2A is a novel
N-CoR-interacting protein and is involved in repression of the
human transcription factor achaete scute-like homologue 2
(ASCL2/Hash2). Mol. Cell Biol.,25, 6404–6414.
53. Mallette,F.A. and Richard,S. (2012) JMJD2A promotes cellular
transformation by blocking cellular senescence through
transcriptional repression of the tumor suppressor CHD5. Cell Rep.,
2, 1233–1243.
54. Kim,T.D., Oh,S., Shin,S. and Janknecht,R. (2012) Regulation of
tumor suppressor p53 and HCT116 cell physiology by histone
demethylase JMJD2D/KDM4D. PLoS One,7, e34618.
55. Black,J.C., Allen,A., Van Rechem,C., Forbes,E., Longworth,M.,
Tschop,K., Rinehart,C., Quiton,J., Walsh,R., Smallwood,A. et al.
(2010) Conserved antagonism between JMJD2A/KDM4A and
HP1gamma during cell cycle progression. Mol. Cell,40, 736–748.
56. Mallette,F.A., Mattiroli,F., Cui,G., Young,L.C., Hendzel,M.J.,
Mer,G., Sixma,T.K. and Richard,S. (2012) RNF8- and
RNF168-dependent degradation of KDM4A/JMJD2A triggers
53BP1 recruitment to DNA damage sites. Embo. J., 31, 1865–1878.
57. Palomera-Sanchez,Z., Bucio-Mendez,A., Valadez-Graham,V.,
Reynaud,E. and Zurita,M. (2010) Drosophila p53 is required to
increase the levels of the dKDM4B demethylase after UV induced
DNA damage to demethylate histone H3-lysine 9. J. Bi o l.
Chem,285, 31370–31379.
58. Young,L.C., McDonald,D.W. and Hendzel,M.J. (2013) Kdm4b
histone demethylase Is a DNA damage response protein and confers
a survival advantage following gamma-Irradiation. J. Biol. Chem.,
288, 21376–21388.
59. Loh,Y.H., Zhang,W., Chen,X., George,J. and Ng,H.H. (2007) Jmjd1a
and Jmjd2c histone H3 Lys 9 demethylases regulate self-renewal in
embryonic stem cells. Genes Dev.,21, 2545–2557.
60. Lorbeck,M.T., Singh,N., Zervos,A., Dhatta,M., Lapchenko,M.,
Yang,C. and Elefant,F. (2010) The histone demethylase
Dmel\Kdm4A controls genes required for life span and male-specic
sex determination in Drosophila. Gene,450, 8–17.
61. Labbe,R.M., Holowatyj,A. and Yang,Z.Q. (2013) Histone lysine
demethylase (KDM) subfamily 4: structures, functions and
therapeutic potential. Am.J.Transl.Res.,6, 1–15.
62. Black,J.C., Manning,A.L., Van Rechem,C., Kim,J., Ladd,B., Cho,J.,
Pineda,C.M., Murphy,N., Daniels,D.L., Montagna,C. et al. (2013)
KDM4A lysine demethylase induces site-specic copy gain and
rereplication of regions amplied in tumors. Cell,154, 541–555.
63. Berry,W.L. and Janknecht,R. (2013) KDM4/JMJD2 histone
demethylases: epigenetic regulators in cancer cells. Cancer Res., 73,
2936–2942.
64. Cloos,P.A., Christensen,J., Agger,K., Maiolica,A., Rappsilber,J.,
Antal,T., Hansen,K.H. and Helin,K. (2006) The putative oncogene
GASC1 demethylates tri- and dimethylated lysine 9 on histone H3.
Nature,442, 307–311.
65. Yang,Z.Q., Imoto,I., Fukuda,Y., Pimkhaokham,A., Shimada,Y.,
Imamura,M., Sugano,S., Nakamura,Y. and Inazawa,J. (2000)
Identication of a novel gene, GASC1, within an amplicon at
9p23–24 frequently detected in esophageal cancer cell lines. Cancer
Res., 60, 4735–4739.
66. Ehrbrecht,A., Muller,U., Wolter,M., Hoischen,A., Koch,A.,
Radlwimmer,B., Actor,B., Mincheva,A., Pietsch,T., Lichter,P. et al.
(2006) Comprehensive genomic analysis of desmoplastic
medulloblastomas: identication of novel amplied genes and
separate evaluation of the different histological components. J.
Pathol., 208, 554–563.
67. Luo,W., Chang,R., Zhong,J., Pandey,A. and Semenza,G.L. (2012)
Histone demethylase JMJD2C is a coactivator for hypoxia-inducible
factor 1 that is required for breast cancer progression. Proc. Natl.
Acad.Sci.U.S.A.,109, E3367–3376.
68. Liu,G., Bollig-Fischer,A., Kreike,B., van de Vijver,M.J., Abrams,J.,
Ethier,S.P. and Yang,Z.Q. (2009) Genomic amplication and
oncogenic properties of the GASC1 histone demethylase gene in
breast cancer. Oncogene,28, 4491–4500.
69. Berdel,B., Nieminen,K., Soini,Y., Tengstrom,M., Malinen,M.,
Kosma,V.M., Palvimo,J.J. and Mannermaa,A. (2012) Histone
demethylase GASC1–a potential prognostic and predictive marker in
invasive breast cancer. BMC cancer,12, 516.
70. Li,W., Zhao,L., Zang,W., Liu,Z., Chen,L., Liu,T., Xu,D. and Jia,J.
(2011) Histone demethylase JMJD2B is required for tumor cell
proliferation and survival and is overexpressed in gastric cancer.
Biochem. Biophys. Res. Commun.,416, 372–378.
71. Kawazu,M., Saso,K., Tong,K.I., McQuire,T., Goto,K., Son,D.O.,
Wakeham,A., Miyagishi,M., Mak,T.W. and Okada,H. (2011) Histone
demethylase JMJD2B functions as a co-factor of estrogen receptor in
breast cancer proliferation and mammary gland development. PLoS
One,6, e17830.
72. Shi,L., Sun,L., Li,Q., Liang,J., Yu,W., Yi,X., Yang,X., Li,Y., Han,X.,
Zhang,Y. et al. (2011) Histone demethylase JMJD2B coordinates
H3K4/H3K9 methylation and promotes hormonally responsive
breast carcinogenesis. Proc. Natl. Acad. Sci. U. S. A.,108, 7541–7546.

6182 Nucleic Acids Research, 2014, Vol. 42, No. 10
73. Ipenberg,I., Guttmann-Raviv,N., Khoury,H.P., Kupershmit,I. and
Ayoub,N. (2013) Heat shock protein 90 (Hsp90) selectively regulates
the stability of KDM4B/JMJD2B histone demethylase. J. Bio l .
Chem.,288, 14681–14687.
74. Coffey,K., Rogerson,L., Ryan-Munden,C., Alkharaif,D., Stockley,J.,
Heer,R., Sahadevan,K., O’Neill,D., Jones,D., Darby,S. et al. (2013)
The lysine demethylase, KDM4B, is a key molecule in androgen
receptor signalling and turnover. Nucleic Acids Res., 41, 4433–4446.
75. Gaughan,L., Stockley,J., Coffey,K., O’Neill,D., Jones,D.L.,
Wade,M., Wright,J., Moore,M., Tse,S., Rogerson,L. et al. (2013)
KDM4B is a master regulator of the estrogen receptor signalling
cascade. Nucleic Acids Res., 41, 6892–6904.
76. Khoury,H., Guttmann-Raviv,N., Ipenberg,I., Huggins,D.,
Jeyasekharan,A. D. and Nabieh Ayoub,. (2014) PARP1-dependent
recruitment of KDM4D histone demethylase to DNA damage sites
promotes double-strand break repair. PNAS Plus,111,E728–E737.
77. Minc,E., Allory,Y., Worman,H.J., Courvalin,J.C. and Buendia,B.
(1999) Localization and phosphorylation of HP1 proteins during the
cell cycle in mammalian cells. Chromosoma,108, 220–234.
78. Fischle,W., Tseng,B.S., Dormann,H.L., Ueberheide,B.M.,
Garcia,B.A., Shabanowitz,J., Hunt,D.F., Funabiki,H. and Allis,C.D.
(2005) Regulation of HP1-chromatin binding by histone H3
methylation and phosphorylation. Nature,438, 1116–1122.
79. Nozawa,R.S., Nagao,K., Masuda,H.T., Iwasaki,O., Hirota,T.,
Nozaki,N., Kimura,H. and Obuse,C. (2010) Human POGZ
modulates dissociation of HP1alpha from mitotic chromosome arms
through Aurora B activation. Nat.CellBiol.,12, 719–727.
80. Andreassen,P.R., Lacroix,F.B., Villa-Moruzzi,E. and Margolis,R.L.
(1998) Differential subcellular localization of protein phosphatase-1
alpha, gamma1, and delta isoforms during both interphase and
mitosis in mammalian cells. J. Cell Biol.,141, 1207–1215.
81. Goulet,I., Boisvenue,S., Mokas,S., Mazroui,R. and Cote,J. (2008)
TDRD3, a novel Tudor domain-containing protein, localizes to
cytoplasmic stress granules. Hum. Mol. Genet.,17, 3055–3074.
82. Botuyan,M.V., Lee,J., Ward,I.M., Kim,J.E., Thompson,J.R., Chen,J.
and Mer,G. (2006) Structural basis for the methylation state-specic
recognition of histone H4-K20 by 53BP1 and Crb2 in DNA repair.
Cell,127, 1361–1373.
83. Huang,Y., Fang,J., Bedford,M.T., Zhang,Y. and Xu,R.M. (2006)
Recognition of histone H3 lysine-4 methylation by the double tudor
domain of JMJD2A. Science,312, 748–751.
84. Lee,J., Thompson,J.R., Botuyan,M.V. and Mer,G. (2008) Distinct
binding modes specify the recognition of methylated histones H3K4
and H4K20 by JMJD2A-tudor. Nat. Struct. Mol. Biol.,15, 109–111.
85. Scott,M.S., Calafell,S.J., Thomas,D.Y. and Hallett,M.T. (2005)
Rening protein subcellular localization. PLoS Comput. Biol.,1, e66.
86. Claycomb,J.M., Batista,P.J., Pang,K.M., Gu,W., Vasale,J.J., van
Wolfswinkel,J.C., Chaves,D.A., Shirayama,M., Mitani,S.,
Ketting,R.F. et al. (2009) The Argonaute CSR-1 and its 22G-RNA
cofactors are required for holocentric chromosome segregation. Cell,
139, 123–134.
87. Kallio,M.J., McCleland,M.L., Stukenberg,P.T. and Gorbsky,G.J.
(2002) Inhibition of aurora B kinase blocks chromosome segregation,
overrides the spindle checkpoint, and perturbs microtubule dynamics
in mitosis. Curr. Biol.,12, 900–905.
88. Richie,C.T. and Golden,A. (2005) Chromosome segregation: Aurora
B gets Tousled. Curr. Biol., 15, R379–382.
89. Kagey,M.H., Melhuish,T.A. and Wotton,D. (2003) The polycomb
protein Pc2 is a SUMO E3. Cell,113, 127–137.
90. Nacerddine,K., Lehembre,F., Bhaumik,M., Artus,J.,
Cohen-Tannoudji,M., Babinet,C., Pandol,P.P. and Dejean,A. (2005)
The SUMO pathway is essential for nuclear integrity and
chromosome segregation in mice. Dev. cell,9, 769–779.
91. McManus,K.J., Biron,V.L., Heit,R., Underhill,D.A. and
Hendzel,M.J. (2006) Dynamic changes in histone H3 lysine 9
methylations: identication of a mitosis-specic function for dynamic
methylation in chromosome congression and segregation. J. Biol .
Chem.,281, 8888–8897.
92. Slee,R.B., Steiner,C.M., Herbert,B.S., Vance,G.H., Hickey,R.J.,
Schwarz,T., Christan,S., Radovich,M., Schneider,B.P.,
Schindelhauer,D. et al. (2012) Cancer-associated alteration of
pericentromeric heterochromatin may contribute to chromosome
instability. Oncogene,31, 3244–3253.