Regulation of cell cycle progression and gene
expression by H2A deubiquitination
Heui-Yun Joo1*, Ling Zhai1*, Chunying Yang1, Shuyi Nie2, Hediye Erdjument-Bromage3, Paul Tempst3,
Chenbei Chang2& Hengbin Wang1
Post-translational histone modifications have important regula-
tory roles in chromatin structure and function1–3. One example of
such modifications is histone ubiquitination, which occurs pre-
dominately on histone H2A and H2B. Although the recent iden-
tification of the ubiquitin ligase for histone H2A has revealed
important roles for H2A ubiquitination in Hox gene silencing4–6
in H2A deubiquitination and the function of H2A deubiquitina-
tion are not known. Here we report the identification and func-
tional characterization of the major deubiquitinase for histone
H2A, Ubp-M (also called USP16). Ubp-M prefers nucleosomal
substrates in vitro, and specifically deubiquitinates histone H2A
HeLa cells results in slow cell growth rates owing to defects in the
mitotic phase of the cell cycle. Further studies reveal that H2A
deubiquitination by Ubp-M isa prerequisite for subsequent phos-
phorylation of Ser10 of H3 and chromosome segregation when
ulates Hox gene expression through H2A deubiquitination and
that blocking the function of Ubp-M results in defective posterior
development in Xenopus laevis. This study identifies the major
deubiquitinase for histone H2A and demonstrates that H2A deu-
biquitination is critically involved in cell cycle progression and
To identify the deubiquitinase for histone H2A, we developed an
in vitro deubiquitination assay using ubiquitinated H2A (ubH2A)-
containing mononucleosomes as substrates (Fig. 1aand Supplemen-
tary Fig. 1a, b)6,9,10. When these nucleosomes were incubated with
HeLa nuclear protein fractions, a decrease of ubH2A levels and an
increase of the released intact ubiquitin were detected, suggesting
that an H2A-specific deubiquitinase is present in the corresponding
fractions (Fig. 1a, lanes 2–4)11,12. We first focused on the nuclear
extract P11 0.5M fraction, which contains the strongest activity.
Through a six-column purification scheme, we identified a polypep-
tide that seems to beresponsible for the H2A deubiquitination activ-
(middle panels) of fractions derived from the hydroxyapatite and
Superose 6 columns allowed us to correlate the deubiquitinase acti-
polyacrylamide gel electrophoresis (PAGE) assay (marked with an
putative H2A deubiquitinase was estimated to be approximately
500kDa, suggesting that it functions as a homo-tetramer (Fig. 1c).
Mass spectrometry analysis identified the protein as Ubp-M, a pre-
revealedthat theelutionprofiles ofUbp-Mcorrelatedwith H2Adeu-
biquitination activity (bottom panels in Fig. 1c and Supplementary
Fig. 2). To verify that Ubp-M is indeed responsible for the H2A deu-
biquitination activity, we performed an immunodepletion assay. As
an aliquot of the hydroxyapatite column fractions, the H2A deubi-
quitination activity was also depleted (compare lanes 1–3 and 4–6).
from Sf9 cells, also elutes at approximately 500kDa and is as active as
native Ubp-M in H2A deubiquitination (Supplementary Fig. 3a, b).
On the basis of these studies, we conclude that Ubp-M is a putative
To gain insight into the function of Ubp-M, we determined its
substrate preference and specificity. As shown in Fig. 2a, Ubp-M
could deubiquitinate nucleosomal ubH2A, with higher efficiency
for oligonucleosomes, but could not deubiquitinate histone form
ubH2A (compare lanes 1–4, 5–8 and 9–12)9. This result indicates
that Ubp-M is a nucleosome-specific deubiquitinase. To determine
the substrate specificity of Ubp-M, we reconstituted mononucleo-
of which contain about 10% ubiquitinated species, together with
other recombinant histones (Fig. 2b, top panels). As shown in
Fig. 2b, Ubp-M deubiquitinates ubH2A and results in an increase
of Flag–H2A levels in a time-dependent manner (bottom panels,
lanes 1–4). In contrast, Ubp-M fails to deubiquitinate ubH2B (ubi-
quitinated H2B) within the sametime period (Fig. 2b,bottom panel,
lanes 5–8). The failure to deubiquitinate ubH2B is not due to the
origin of the substrate (yeast H2B), as Ubp-M also fails to deubiqui-
tinate nucleosomal human H2B (Supplementary Fig. 4) and the nu-
clear pellet P11 0.5M fraction could deubiquitinate yeast ubH2B
efficiently (Fig. 2b, bottom panel, lane 9). Collectively, such results
indicate that Ubp-M is an H2A-specific deubiquitinase.
Previous studies indicated that overexpression of Ubp-M reduces
the levels of cellular ubH2A; however, the physiological relevance of
these observations was not clear16. To determine whether Ubp-M
Flag–H2B, respectively (Fig. 2c). Because the levels of ubH2B are
low, we co-transfected haemagglutinin-tagged ubiquitin to observe
ubH2B after immunoprecipitation. Transfection of Ubp-M siRNA
reduces the levels of Ubp-M significantly, and results in a clear
increase in the levels of ubiquitinated histones (Fig. 2c, top three
panels, lanes 3, 6). To determine whether the increase of ubiquiti-
nated histones was due to ubH2A and/or ubH2B, we performed
denaturing immunoprecipitation assays17,18. As indicated in Fig. 2c,
Ubp-M knockdown results in a specific increase in the levels of
*These authors contributed equally to this work.
1Department of Biochemistry and Molecular Genetics, University of Alabama at Birmingham, Kaul Human Genetics Building 402A, 720 South 20th Street, Birmingham, Alabama
Kettering Cancer Center, 1275 York Avenue, New York, New York 10021, USA.
Vol 449|25 October 2007|doi:10.1038/nature06256
ubH2A but not ubH2B (Fig. 2c, fourth panel, and Supplementary
Fig. 5). On the basis of these experiments, we conclude that Ubp-M
specifically deubiquitinates histone H2A in vivo.
To determine the function(s) of H2A deubiquitination in vivo, we
generated HeLa celllines withstable knockdown ofUbp-M4.Knock-
down of Ubp-M (.90%, protein level) results in a 50620%
(mean6s.d. of five cell lines) increase of ubH2A levels (Fig. 3a, left
panels). Notably, Ubp-M knockdown cells exhibit slow proliferation
rates compared with control cells (threefold slower, Fig. 3a, right
panel). Further analyses indicate that Ubp-M knockdown results in
a decrease in the proportion of cells at G2/M but has no effect on
apoptosis (Supplementary Figs 6 and 7). To substantiate this
M-phase effect, control and Ubp-M knockdown cells were synchro-
media, cells were fixed at different time points and stained with
the anti-phospho-histone H3 antibody to monitor mitotic cell pro-
gression, and with propidium iodide to monitor DNA contents.
Fluorescence-activated cell sorting (FACS) analyses revealed that
the M-phase cell population was reduced almost twofold in Ubp-
M knockdown cells (Fig. 3b). To determine whether this effect is
linked to H2A deubiquitination, we analysed the levels of ubH2A
throughout the cell cycle (Fig. 3c). In control cells, ubH2A levels
when cells are in M phase (9h), start to recover as cells exit from M
phase (10–11h), and are restored to normal levels when cells enter
G1/S (12–18h; Fig. 3c, first panel, and Supplementary Fig. 8a). In
as cells enter M phase, and progression through M phase is delayed
(Fig. 3c, fourth panel, and Supplementary Fig. 8b). Intriguingly,
fluctuations in ubH2A levels during the cell cycle are inversely cor-
related with the fluctuations in phospho-histone H3 levels (Fig. 3c,
second and fifth panels, and Supplementary Fig. 8a, b). Immuno-
ubH2A and H3 phosphorylated at Ser10 (H3S10ph; Supplementary
Fig. 9). These results suggest that histone ubiquitination may inter-
fere with histone H3 phosphorylation.
To determine the interplay between these two modifications, we
reconstituted mononucleosomes containing ubH2A and ubH2B
(Supplementary Fig. 10), and used them in Aurora-B-mediated
kinase reactions. As indicated in Fig. 3d, histone ubiquitination
reduced H3S10 phosphorylation (middle panel, compare lanes 4, 9
with 2, 7). Importantly, Ubp-M-mediated H2A deubiquitination
restores H3S10 phosphorylation (compare lane 5 with 4), suggesting
that histone deubiquitination may be a prerequisite for H3S10 phos-
phorylation. To determine whether histone ubiquitination affects
Aurora B association with nucleosomes, we performed nucleosome
pull-down assays17. As shown in Fig. 3e, histone ubiquitination
panel, compare lanes 2, 4 with 1, 3). Because the level of endogenous
ubH2B is much lower than that of ubH2A (Supplementary Fig. 5),
these results strongly suggest that Ubp-M-mediated H2A deubi-
quitination constitutes the primary mechanism that regulates the
BC300 BC500 BC1000
In 29 32 35 38 41 44 47 50 53 56 59 62 65 68 71 74 77 80 83
443 kDa 220 kDa
In Ft B In Ft B
C 1 2 3 4 5 6
Figure 1 | Purification of the H2A-specific deubiquitinase. a, H2A
deubiquitination assay with HeLa nuclear extracts (NE) and nuclear pellet
(NP) proteins fractionated on DE52 and P11 columns. b, Schematic
representation of the steps used for the purification. Numbers representthe
salt concentrations (mM) at which the deubiquitinase activity elutes from
the columns. c, Silver staining (top panel), H2A deubiquitination assay
(middle panel) and western blot analysis (bottom panel) of the Superose 6
column fractions. The candidate band is indicated by an asterisk. d, Silver
staining (top panel), western blot analysis (middle panel) and H2A
deubiquitination assay (bottom panel) of input (In), flow through (Ft) and
bound (B) with purified IgG and Ubp-M antibody. HC, heavy chain; LC,
light chain; NS, nonspecific binding. The protein size markers (kDa) are
shown on the left side of c and d.
NATURE|Vol 449|25 October 2007
association ofAurora Bkinase withnucleosomes andthesubsequent
Previous studies indicated that H2A ubiquitination regulates Hox
gene silencing4–6. To determine the roles of H2A deubiquitination in
this process, we measured the expression of selected Hox genes with
semi-quantitative polymerase chain reaction with reverse transcrip-
in Fig. 4a and Supplementary Fig. 11, knockdown of Ubp-M results
in a specific decrease of HOXD10 gene expression, which could be
rescued by transfection of wild type but not enzymatically inactive
(C205S) Ubp-M (Supplementary Fig. 12)15. To determine the role of
H2A deubiquitination in Hox gene expression, we measured the
binding of Ubp-M and Bmi1 across the HOXD10 gene (see diagram
in Fig. 4b)6. In control cells, Ubp-M and Bmi1 bound to two regions
of HOXD10, one localized in the promoter region (overlapping with
transcription start sites) and the other localized in 59 regulatory
regions (1.7kb upstream of the transcription start site; Fig. 4b, top
two panels, lanes 1–8). In Ubp-M KD1 knockdown cells, the binding
of Ubp-M to both regions of HOXD10 was markedly decreased
whereas the binding of Bmi1 was not changed (Fig. 4b, top two
panels, lanes 9–16). Chromatin double immunoprecipitations
(anti-Flag immunoprecipitation followed by anti-HA immuno-
precipitation) revealed that the levels of ubH2A at Ubp-M binding
H2A deubiquitination regulates gene expression.
Because Ubp-M regulates Hox gene expression, we investigated
whether Ubp-M regulates body patterning in Xenopus laevis. The
Xenopus homologue of Ubp-M is recognized by our antibody and
has H2A deubiquitination activity (Supplementary Fig. 13a, b).
100 bp M.
0 10 30 60
0 10 30 60
1 0.4 0.2 0.08
1 1.61 1.8
Figure 2 | Substrate preference and specificity of Ubp-M. a, Ubp-M-
mediated deubiquitination reaction with substrates as indicated at the top.
deubiquitination reaction with nucleosomes reconstituted with Flag–H2A
and Flag–H2B as substrates (bottom panels). Top panels show nucleosome
reconstitution with Flag–H2A and Flag–H2B. The protein size markers
(kDa) are shown on the left side of the panels. c, Ubp-M is specific for H2A
deubiquitination in vivo. HeLa cells stably expressing Flag–H2A (lanes 1–3)
and Flag–H2B (lanes 4–6) were transfected with the indicated plasmid and
as quantified by Image J Software.
Viable cell number (×105)
3697.510.5 12 17
Mitotic cells (%)
Time after release (h)
Time point (h)
Figure 3 | Ubp-M-mediated H2A deubiquitination is required for cell cycle
M phase progression. a, Ubp-M knockdown of HeLa cells results in an
increase of cellular ubH2A levels (left) and slow cell growth rate (right
panel). Numbers representthe relative protein levels as quantified by Image
M knockdown decreases the M-phase cell population. c, Western blot
analyses of ubH2A and H3S10ph levels in synchronized control (top three
panels) and Ubp-M KD1 (bottom three panels) cells throughout the cell
cycle. Experiments were repeated three times and a representative result is
shown. d, H2A ubiquitination inhibits H3S10 phosphorylation. Aurora B
along the top of the panel. e, Histone ubiquitination interferes with the
associationbetween AuroraB andnucleosomes. Nucleosomes,asindicated,
were incubated with Flag-tagged Aurora B kinase and the nucleosome-
associated Aurora B was determined by western blot assay.
NATURE|Vol 449|25 October 2007
When equal amounts of purified IgG and Ubp-M antibody were
injected into two-cell-stage Xenopus embryos, Ubp-M antibody,
but not IgG, induced a homeotic transformation (enlarged cement
gland and reduced posterior structures; Fig. 4c, middle panels, and
Supplementary Fig.14).Thephenotype is specific, aspre-incubation
of Ubp-M antibody with the Ubp-M antigen abolishes these mor-
phological changes (Fig. 4c, bottom panels). In situ hybridization
is greatly reduced when Ubp-M activity is blocked by injection of
the antibody (Fig. 4d). Taken together, we conclude that Ubp-M
lation of Hox gene expression.
Our data reveal two independent functions of Ubp-M-mediated
and (2) to facilitate cell cycle M-phase progression. Although we do
not know how Ubp-M is recruited to specific regions and how Ubp-
M is activated during M phase, we found that Ubp-M is phosphory-
lated and that the phosphorylation status of Ubp-M correlates with
tary Fig. 15)15. Further investigations will determine whether phos-
phorylation contributes to the regulation of Ubp-M. In addition,
H2Aubiquitination hasbeenimplicatedinX-chromosome inactiva-
tion7,8, and further experiments will determine whether Ubp-M is
involved in this process. Nonetheless, the identification of the major
H2A deubiquitinase should greatly improve our understanding of
histone ubiquitination in chromatin regulation.
graphy and identified by matrix-assisted laser desorption/ionization reflectron
time-of-flight (MALDI-reTOF) mass spectrometry. Recombinant Ubp-M was
purified by a combination of immunoaffinity and gel filtration. In vitro histone
deubiquitination activity was determined by western blot analysis. Native
ubH2A-containing mononucleosomes were purified by tandem immunopreci-
histone octamers were purified from these nucleosomes with a hydroxyapatite
column9. Mononucleosomes were reconstituted by serial salt dilutions17. Flag-
ubH2B by electro-elution. Ubp-M knockdown was achieved by RNA interfer-
ence4,17. Effects of Ubp-M knockdown on histone ubiquitination and H3S10ph
were determined by western blot analysis. Effects of Ubp-M knockdown on cell
and H3S10ph was revealed by immunofluorescence staining. Cell synchroniza-
tion was achieved by double-thymidine treatment. The expression levels of Hox
genes were determined by semi-quantitative RT–PCR6. The pattern of Ubp-M,
Bmi1 and ubH2A binding to the HOXD10 gene was determined by ChIP assays.
Effects of Ubp-M on body transformation were determined by injection of IgG
and Ubp-M antibody into Xenopus laevis embryos at the two-cell stage.
Expression of the Hoxd10 gene and posterior and anterior molecular markers
was revealed by in situ hybridization.
Full Methods and any associated references are available in the online version of
the paper at www.nature.com/nature.
Received 23 May; accepted 11 September 2007.
Published online 3 October 2007.
1. Martin,C.&Zhang, Y.Thediversefunctionsofhistonelysine methylation. Nature
Rev. Mol. Cell Biol. 6, 838–849 (2005).
Jenuwein, T. & Allis, C. D. Translating the histone code. Science 293, 1074–1080
Peterson, C. L. & Laniel, M.-A. Histones and histone modifications. Curr. Biol. 14,
Wang, H. et al. Role of histone H2A ubiquitination in Polycomb silencing. Nature
431, 873–878 (2004).
Cao, R., Tsukada, Y.-i. & Zhang, Y. Roleof Bmi-1 andRing1A inH2A ubiquitylation
and Hox gene silencing. Mol. Cell 20, 845–854 (2005).
silencing. J. Biol. Chem. 281, 22537–22544 (2006).
Fang, J., Chen, T., Chadwick, B., Li, E. & Zhang, Y. Ring1b-mediated H2A
ubiquitination associates with inactive X chromosomes and is involved in
initiation of X inactivation. J. Biol. Chem. 279, 52812–52815 (2004).
de Napoles, M. et al. Polycomb group proteins Ring1A/B link ubiquitylation of
histone H2A to heritable gene silencing and X inactivation. Dev. Cell 7, 663–676
Fang, J., Wang, H. & Zhang, Y. Purification of histone methyltransferases from
HeLa cells. Methods Enzymol. 377, 213–226 (2004).
10. Wang, H. et al. Purification and functional characterization of a histone H3-lysine
4-specific methyltransferase. Mol. Cell 8, 1207–1217 (2001).
11.Henry, K. W. et al. Transcriptional activation via sequential histone H2B
ubiquitylation and deubiquitylation, mediated by SAGA-associated Ubp8. Genes
Dev. 17, 2648–2663 (2003).
12. Daniel, J. A. et al. Deubiquitination of histone H2B by a yeast acetyltransferase
complex regulates transcription. J. Biol. Chem. 279, 1867–1871 (2004).
A B C D E
11 12 13
Figure 4 | Ubp-M-mediated H2A deubiquitination regulates Hox gene
expression. a, Semi-quantitative RT–PCR analysis of HOXD10 expression
in HeLa cells as indicated above. GAPDH was used as a control. b, ChIP
assays of HOXD10 gene binding in control (lanes 1–8) and Ubp-M
knockdown (lanes 9–16) cells. A diagram of the HOXD10 gene is illustrated
indicated. c, Control uninjected Xenopus embryos and embryos injected
with the indicated reagents are shown (side view) with the head to the left.
panel) or in embryos injected with IgG (top right panel) or anti-Ubp-M
antibody (bottom panels).
NATURE|Vol 449|25 October 2007
13. Erdjument-Bromage, H. et al. Examination of micro-tip reversed-phase liquid
chromatographic extraction of peptide pools for mass spectrometric analysis.
J. Chromatogr. A. 826, 167–181 (1988).
14. Winkler, G. et al. Isolation and mass spectrometry of transcription factor
complexes. Methods 26, 260–269 (2002).
15. Cai, S. Y., Babbitt, R. & Marchesi, V. A mutant deubiquitinating enzyme (Ubp-M)
USA 96, 2828–2833 (1999).
nucleosomal histone H2A induced by diverse apoptogenic stimuli. Cell Death
Differ. 8, 1182–1196 (2001).
ligase facilitates cellular response to DNAdamage. Mol. Cell22, 383–394 (2006).
18. Shiio, Y. &Eisenman, R. N. Histone sumoylation is associated with transcriptional
repression. Proc. Natl Acad. Sci. USA 100, 13225–13230 (2003).
in yeast. Science 287, 501–504 (2000).
Supplementary Information is linked to the online version of the paper at
Acknowledgements We thank S. L. Berger and M. A. Osley for yeast strains;
J. J. Hayes for XP-10 plasmid; and J. Wei for the initial cloning of Ubp-M. We also
thank D. Crawford and W. S. Brooks for suggestions on cell cycle and apoptosis
analysis; T. Townes and W. S. Brooks for critical reading of the manuscript;
E. F. Keyser for assistance with FACS analysis; and A. Nazarian for help with mass
NCI Cancer Center Support Grant (to P.T.).
Author Contributions H.W. designed the experimental strategy, performed parts
of the purification and wrote the paper; H.-Y.J. performed most of the purification,
determined the substrate preference and investigated the role of Ubp-M in cell
Ubp-M and performed the in vitro kinase reaction and nucleosome pull-down
experiments; C.Y. cloned the Xenopus Hoxd10 gene and generated the
RNAi-resistant Ubp-M constructs; S.N. and C.C. performed the Xenopus injection
and in situ hybridization; H.E.-B. and P.T. performed mass spectrometric analysis.
Author Information Reprints and permissions information is available at
www.nature.com/reprints. Correspondence and requests for materials should be
addressed to H.W. (email@example.com).
NATURE|Vol 449|25 October 2007
Substrate preparation and in vitro histone deubiquitination assay. H2A-
containing mononucleosomes were purified by tandem immunoprecipitation6.
were prepared as described10. Flag–H2A was purified from HeLa stable cell lines
and Flag–H2B was purified from the T85 yeast strain and the YKH045 yeast
strain with anti-Flag immunoprecipitation under denaturing conditions17,20,21.
Recombinant H3, H4, H2A and H2B were purified from Escherichia coli22. To
purify individual H2A, ubH2A, H2B and ubH2B, gel strips containing the
corresponding histones were excised and proteins were electro-eluted out and
further passed through a hydroxyapatite column to remove the excess SDS.
Histone octamer refolding and nucleosome reconstitution were performed as
Histone deubiquitination reactions were performed as follows. 1.5mg of
ubH2A-containing mononucleosomes, histone octamers, oligonucleosomes,
or 0.8mg of reconstituted nucleosomes were incubated with protein fractions
or recombinant Ubp-M in deubiquitination reaction buffer (100mM Tris-HCl,
pH8.0, 1mM EDTA, 0.1mM PMSF, 1mM dithiothreitol (DTT), 1mgml21
aprotinin, 1mgml21leupeptin and 1mgml21pepstatin A) at 37uC for 45min,
or as indicated. The reaction was terminated by the addition of SDS–PAGE
sample loading buffer and proteins were resolved on SDS–PAGE and blotted
with the anti-HA antibody.
Purification of native and recombinant Ubp-M. HeLa nuclear proteins were
separated into nuclear extracts and nuclear pellets using a previously described
procedure24. Nuclear extract (6g) was loaded onto a 700ml P11 column equili-
brated with buffer C (20mM Tris-HCl, pH7.9, 0.2mM EDTA, 1mM DTT,
0.1mM PMSF, 0.025% NP-40, 10% glycerol) containing 100mM KCl
(BC100). Proteins that bound to the column were step eluted with BC300,
BC500 and BC1000. The BC500 fraction was dialysed against buffer D
(40mM HEPES-KOH, pH7.9, 0.2mM EDTA, 1mM DTT, 0.1mM PMSF,
0.025% NP-40, 10% glycerol) containing 20mM ammonium sulphate (BD20)
andthenloadedontoa 45mlHPLC-DEAE-5PW column(TOSOHBioscience).
Bound proteins were eluted with an 8-column-volume linear gradient from
BD20 to BD500. Fractions that contain H2A-deubiquitination activity were
adjusted to BD600 with saturated ammonium sulphate and loaded onto a
were eluted with a 12-column-volume linear gradient from BD600 to BD0. The
active fractions were then applied to a Sephacryl S-300 column (Amersham
Biosciences). The H2A-deubiquitination activity eluted between 443–670kDa.
These fractions were dialysed against buffer P (5mM HEPES-KOH, pH7.5,
(Bio-rad). The column was eluted with a 20-column-volume linear gradient
from BP10 to BP600. Fractions that contained H2A-deubiquitination activity
wereloadedonto aSuperose6column(AmershamBiosciences)that wasequili-
bratedwithBC500.Toidentify the120-kDaproteinthat co-eluteswiththeH2A
deubiquitination activity, fractions 58–60 of the Superose 6 column were con-
centrated and resolved on SDS–PAGE. The band that contains the 120-kDa
protein was excised and subjected to in-gel trypsin digestion. The resulting
peptides were analysed by matrix-assisted laser desorption/ionization reflectron
time-of-flight (MALDI-reTOF) mass spectrometry (MS) (UltraFlex TOF/TOF;
BRUKER) for peptide mass fingerprinting13,14. Recombinant Ubp-M was puri-
fied from Sf9 cells infected with baculovirus encoding Flag–Ubp-M as
Ubp-M knockdown, RT–PCR and ChIP assays. Stable cell lines with knock-
For transient knockdown, siRNA oligonucleotides against Ubp-M were pur-
chasedfrom Invitrogenina purified, annealedduplex formand transfected into
cells with Lipofectamine 20006,17. Semi-quantitative RT–PCR was performed as
ChIP and double ChIP were performed as described except that cells were
lysed in buffer containing 1% SDS and diluted 10 times before immuno-
precipitation6. All PCR reactions were performed within the range of linear
amplification. The sequences for RT–PCR and ChIP assays are available on
The DNA sequence encoding the stem hairpin RNA for Ubp-M was 59-
(KD1) and 59-ACCAGTGCTTAGAGAACTATTCAAGAGATAGTTCTCTAA-
GCACTGGT-39 (KD2). Underlined letters represent targeting sequences of
purchased from Invitrogen in a purified, annealed duplex form. The sequences
for these siRNA are as follows: siRNA1 59-CCUCCUGUUCUUACUCUU-
CAUUUAA-39 and 59-UUAAAUGAAGAGUAAGAACAGGAGG-39; siRNA2
59-CCGGAAAUCUUAGAUUUGGCUCCUU-39 and 59-AAGGAGCCAAAUC-
UAAGAUUUCCGG-39. siRNA1 and siRNA2 were mixed at a 1:1 ratio and
transfected into cells.
Cell cycle and FACS analysis. Cells were synchronized by double thymidine
(Sigma) treatment25. After release into fresh medium, cells were harvested at the
Triton X-100 and subsequently incubated with anti-phospho-histone H3 anti-
body (Ser10; Upstate), APC-conjugated anti-rabbit antibody (Invitrogen) and
propidium iodide/RNase A. Mitotic cell distribution was analysed by FACS.
Alternatively, acid-soluble proteins were extracted from control and Ubp-M
knockdown cells and levels of ubH2A and H3S10ph were analysed by western
injected into both blastomeres of two-cell-stage embryos. The embryos were
collected at tailbud stages for in situ hybridization with Hoxd10, Sox2, Otx2,
En2 and Krox20 antisense RNA probes, or at tadpole stages for morphological
assessment of the effects. For specificity control, an equal amount of Ubp-M
antigen was mixed with the antibody on ice for at least 30min before they were
Constructs, antibodies, kinase reactions and immunofluorescence staining.
The cDNA for Ubp-M was provided by V. T. Marchesi15. To generate antibody
against Ubp-M, the amino terminus of Ubp-M (1–257 amino acids) was cloned
into pGEX-KG and the purified recombinant proteins were used to immunize
rabbits. To construct RNAi-resistant Ubp-M constructs, a point mutation was
(Gibco) with a Flag tag at its N terminus. Recombinant baculovirus expressing
Flag–Ubp-M was generated and amplified following the manufacturer’s pro-
tocol. Kinase assay was performed as described27. Aurora B kinase was purified
by immunoprecipitation from HeLa cells that stably express Flag–Aurora B
and the phosphorylation of H3 was determined by western blot with anti-
phospho-H3S10 antibody (Upstate, 06-570). Immunofluorescence staining
was performed essentially as described28,29.
20. Henry, K. W. et al. Transcriptional activation via sequential histone H2B
ubiquitylation and deubiquitylation, mediated by SAGA-associated Ubp8. Genes
Dev. 17, 2648–2663 (2003).
22. Luger, K., Rechsteiner, T. J. & Richmond, T. J. Preparation of nucleosome core
particle from recombinant histones. Methods Enzymol. 304, 3–19 (1999).
23. Dyer, P. N. et al. Reconstitution of nucleosome core particles from recombinant
histones and DNA. Methods Enzymol. 375, 23–44 (2004).
24. Dignam, J. D., Martin, P., Shastry, B. & Roeder, R. Eukaryotic gene transcription
with purified components. Methods Enzymol. 101, 582–598 (1983).
25. Fischle, W. et al. Regulation of HP1-chromatin binding by histone H3 methylation
and phosphorylation. Nature 438, 1116–1122 (2005).
26. Harland, R. M. In situ hybridization: an improved whole-mount method for
Xenopus embryos. Methods Cell Biol. 36, 685–695 (1991).
27. Bolton, M. A. et al. Aurora B kinase exists in a complex with survivin and INCENP
and its kinase activity is stimulated by survivin binding and phosphorylation. Mol.
Biol. Cell 13, 3064–3077 (2002).
28. Perez-Burgos, L. et al. Generation and characterization of methyl-lysine histone
antibodies. Methods Enzymol. 376, 234–254 (2003).
29. Song, L. et al. Ser-10 phosphorylated histone H3 is involved in cytokinesis as a
chromosomal passenger. Cell Biol. Int. 31, 1184–1190 (2007).
Page 7 Download full-text