A small molecule binding to the coactivator CREB-binding protein blocks apoptosis in cardiomyocytes

Article abstract of DOI:10.1016/j.chembiol.2010.12.021

As a master transcription factor in cellular responses to external stress, tumor suppressor p53 is tightly regulated. Excessive p53 activity during myocardial ischemia causes irreversible cellular injury and cardiomyocyte death. p53 activation is dependent on lysine acetylation by the lysine acetyltransferase and transcriptional coactivator CREB-binding protein (CBP) and on acetylation-directed CBP recruitment for p53 target gene expression. Here, we report a small molecule ischemin, developed with a structure-guided approach to inhibit the acetyl-lysine binding activity of the bromodomain of CBP. We show that ischemin alters post-translational modifications on p53 and histones, inhibits p53 interaction with CBP and transcriptional activity in cells, and prevents apoptosis in ischemic cardiomyocytes. Our study suggests small molecule modulation of acetylation-mediated interactions in gene transcription as a new approach to therapeutic interventions of human disorders such as myocardial ischemia.

Full text of DOI:10.1016/j.chembiol.2010.12.021

Chemistry & Biology
Article
A Small Molecule Binding to
the Coactivator CREB-Binding Protein
Blocks Apoptosis in Cardiomyocytes
Jagat C. Borah,1,3 Shiraz Mujtaba,1,3 Ioannis Karakikes,2 Lei Zeng,1 Michaela Muller,1 Jigneshkumar Patel,1
Natasha Moshkina,1 Keita Morohashi,1 Weijia Zhang,2 Guillermo Gerona-Navarro,1 Roger J. Hajjar,2
and Ming-Ming Zhou1,
*
1Department of Structural and Chemical Biology
2Department of Medicine
Mount Sinai School of Medicine, One Gustave L. Levy Place, New York, NY 10029, USA
3These authors contributed equally to this work
*Correspondence: Ming-Ming.Zhou@mssm.edu
DOI 10.1016/j.chembiol.2010.12.021
SUMMARY
Myocardial ischemia can also induce inflammatory responses
and cardiomyocyte necrosis, depending on the intensity and
As a master transcription factor in cellular responses duration of ischemia and reperfusion (Fliss and Gattinger,
1996; Kajstura et al., 1996; Lefer and Granger, 2000). Thus, it
appears that attenuation of myocyte apoptosis could potentially
to external stress, tumor suppressor p53 is tightly
regulated. Excessive p53 activity during myocardial
ischemia causes irreversible cellular injury and cardi-
omyocyte death. p53 activation is dependent on
lysine acetylation by the lysine acetyltransferase
and transcriptional coactivator CREB-binding
protein (CBP) and on acetylation-directed CBP
recruitment for p53 target gene expression. Here,
prevent myocyte loss, improve myocardial dysfunction, and
decrease mortality rate that is associated with myocardial
ischemia and reperfusion (Eefting et al., 2004). Previous studies
have shown that exposure of myocytes to hypoxia results in
increased p53 trans-activating activity and protein accumulation
along with the expression of p21/WAF-1/CIP-1, a well-charac-
terized target of p53 transactivation (Long et al., 1997). Although
we report a small molecule ischemin, developed
with a structure-guided approach to inhibit the
p53 activation has been recognized for therapeutic potential in
cancer treatments, its hyperactivation could also be detrimental
acetyl-lysine binding activity of the bromodomain of in both normal and ischemic conditions (Komarova and Gudkov,
1998, 2000). Therefore, in a different biological context, modula-
tion of p53 function as a transcriptional regulator, either activa-
CBP. We show that ischemin alters post-translational
modifications on p53 and histones, inhibits p53 inter-
action with CBP and transcriptional activity in cells,
and prevents apoptosis in ischemic cardiomyocytes.
Our study suggests small molecule modulation of
acetylation-mediated interactions in gene transcrip-
tion as a new approach to therapeutic interventions
of human disorders such as myocardial ischemia.
tion or inhibition, could present valid therapeutic opportunities.
Transcriptional coactivators CREB-binding protein (CBP) and
p300 (also known as KAT3B and KAT3A, respectively) (Giordano
and Avantaggiati, 1999; Imhof et al., 1997; Ogryzko et al., 1996;
Sterner and Berger, 2000) play a central role in regulating p53
stability and its function as a transcription factor in response to
genotoxic stress (Ferreon et al., 2009; Ott et al., 1999). The
main functions of CBP/p300 in gene transcription are dependent
on activities of both the lysine acetyltransferase (KAT) and the
INTRODUCTION
acetyl-lysine binding bromodomain (BRD). Like histones, lysine
acetylation of transcription factors facilitates the recruitment of
Cardiovascular diseases continue to be an epidemic in the BRD-containing cofactors required for chromatin structural
United States and the western world. Half of all hospital admis- change and transcriptional initiation and elongation (Dhalluin
sions are related to cardiovascular disorders; 80% of those et al., 1999; Mujtaba et al., 2007; Sanchez and Zhou, 2009).
who have fatal consequences are 65 years or older (Gillum The biochemical contribution of acetylation to p53 transcription
et al., 1998; Wei, 1992). The salient feature of cardiac ischemia, functions has been attributed to nuclear translocation, alteration
which is mainly due to coronary syndromes, includes lack of of DNA binding ability and enhancement of transcriptional poten-
oxygen and nutrition, which generates stress signals to activate tial (Prives and Manley, 2001). Previous studies from us and other
pathways leading to cardiac myocyte death. It has been reported laboratories have supported the role of p53 acetylation in
that ischemia-induced myocyte DNA damage results in promoting molecular interactions with transcriptional coregula-
enhanced transcriptional activity of the tumor suppressor p53 tors leading to target gene activation that ultimately determines
as well as p53-dependent cardiac myocyte apoptosis; the latter cellular responses to stress in the forms of senescence, cell
is a key feature in the progression of ischemic heart disease growth arrest, or apoptosis (Barlev et al., 2001; Mujtaba et al.,
(Gudkov and Komarova, 2005; Komarova and Gudkov, 1998). 2004; Sakaguchi et al., 1998; Zhao et al., 2006). Specifically,
Chemistry & Biology 18, 531–541, April 22, 2011 ª2011 Elsevier Ltd All rights reserved 531

Chemistry & Biology
A CBP Inhibitor Blocks Apoptosis in Cardiomyocytes
Figure 1. Target Structure-Guided Design of Chemical Probes for the CBP Bromodomain
(A) Superimposition of 2D 1H-15N-HSQC spectra of the 15N-labeled CBP BRD highlighting changes of protein NMR resonances on binding to a small molecule,
ischemin (black, free protein; red, in the presence of ischemin).
(B and C) Ribbon diagram of the lowest energy NMR structures of the CBP BRD bound to ischemin (left) versus to histone H4K20ac peptide (right), respectively.
Side chains of the key protein residues at the ligand-binding site are shown and color-coded by atom type. See also Figure S1.
we demonstrated that CBP acetylates p53 C-terminal lysine resi- cules that target the CBP BRD, and characterized their biochem-
dues on DNA damage, and subsequently the BRD of CBP binds ical and functional effects of modulating gene transcription in
to p53 at acetylated lysine 382 (p53K382ac), which is essential osteocarcinoma cells and primary cardiomyocytes where DNA
for p53-induced transcriptional activation of target genes such damage represents intracellular signaling during ischemia.
as cyclin-dependent kinase inhibitor 1A (CDKN1A/p21) (Mujtaba
et al., 2004, 2006).
Activation of p53 in cancer is not restricted to the killing of
RESULTS
tumor cells (Rogel et al., 1985). p53-dependent apoptosis also Target Structure-Guided Lead Discovery
occurs in sensitive tissues shortly after gamma irradiation We conducted nuclear magnetic resonance (NMR)-based
(Komarova et al., 1997a), and p53-deficient mice survive higher screening of 3000 diversity chemical compounds selected
doses of gamma irradiation than do wild-type animals (Westphal from our in-house collection of 115,000 compounds to identify
et al., 1997, 1998). These studies highlight the toxic nature of p53 small molecules for the bromodomain of CBP. Ligand binding
after chemo- and radiation-therapies on normal tissues that to the protein was detected by monitoring changes of backbone
ameliorate the disease condition and extend the recovery in amide resonances of 15N-labeled CBP BRD in 2D 1H-15N-HSQC
cancer patients. Therefore, suppression of p53 functions has NMR spectra on addition of the chemicals as mixtures of 30.
been suggested as a therapeutic strategy to prevent damage Subsequently, the positive mixtures with estimated affinity better
of normal tissues (Komarova and Gudkov, 2001). Consequently, than Kd of about 300 mM were deconvoluted to identify individual
Pifithrin was identified from cell based chemical library small molecules that bind CBP BRD at the acetyl-lysine binding
screening, which is probably the first p53 inhibitor reported site (Figure 1A). This process was guided by our NMR resonance
(Komarov et al., 1999). In addition to inhibiting p53-dependent assignment of the protein. This target structure-based screening
apoptosis, Pifithrin has been shown to reduce the activation of yielded 10 promising hits for the CBP BRD, several of which
p53-regulated target genes, including cyclin G, p21/WAF-1/ share the azobenzene scaffold, as exemplified by 4-hydroxyphe-
CIP-1, and MDM2 (Komarov et al., 1999). Anti-apoptotic proper- nylazo-benzenesulfonic acid MS456 (Table 1). We solved the 3D
ties of Pifithrin has also been demonstrated in cellular and mouse structure of MS456 bound to the CBP BRD with NMR spectros-
models of genotoxic stress (Komarov et al., 1999; Liu et al., copy, which reveals that this compound binds at the entrance of
2004), dopamine-induced apoptosis in neurons (Culmsee the acetyl-lysine binding pocket in the BRD, thereby blocking the
et al., 2001, 2003), cisplatin-induced apoptosis in cochlear and protein from interacting with lysine-acetylated binding partners
vestibular hair cells (Zhang et al., 2003), as well as endotoxin- (see Figure S1A available online; Table 2).
induced apoptosis in liver tissue (Schafer et al., 2003). Despite
these studies, Pifthrin’s mechanism of action, however, is not Synthesis and Structure-Activity Relationship Analysis
well understood (Murphy et al., 2004).
of Chemical Analogs
Because activated p53 is associated with CBP in an acetyla- To improve lead compound binding affinity and selectivity for
tion-sensitive manner, we postulated that small molecules the CBP BRD, we synthesized a series of azobenzene analogs
that inhibit the acetyl-lysine binding activity of the BRD of CBP by introducing various chemical functional groups on the
would inhibit the ability of p53 to direct its target gene activation. aromatic rings, as well as by changing the sulfonate group
Toward this goal, in this study we have developed small mole- from para- to meta- position with respect to the azo linkage on
532 Chemistry & Biology 18, 531–541, April 22, 2011 ª2011 Elsevier Ltd All rights reserved

Chemistry & Biology
A CBP Inhibitor Blocks Apoptosis in Cardiomyocytes
Table 1. Structure-Activity Relationships of Azobenzene Compounds in p53 Inhibition
Compound
MS456
MS450
MS113
MS451
MS110
MS111
MS105
MS101
MS103
MS100
R1
R2
H
R3
R4
H
R5
R6 R7
R8
R9 R10 Inhibition (%)a Kd (mM)b
OH
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
SO3H
SO3H
SO3H
SO3H
SO3H
SO3H
SO3H
SO3H
SO3H
SO3H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Cl
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
4.6
OH CH3
H
H
H
85.6
25.7
87.4
86.2
22.8
26.4
82.9
38.9
36.4
76
OH CH3CH2
OH CH3
H
H
H
H
H
CH3
CH3
CH3
(CH3)2CH
H
44
OH CH2CHCH2
OH (CH3)3C
OH (CH3)2CH
OH CH3
H
H
104
H
H
H
H
H
CH3
OH CH3
H
CH3 CH3
OH
H
H
H
H
H
H
CH3
CH3
H
H
Ischemin/MS120 OH
NH2
CH3
SO3H CH3
SO3H CH3
SO3H CH3
SO3H CH3
SO3H OH
SO3H CH3
SO3H CH3
SO3H CH3
SO3H CH3
SO3H CH3
SO3H CH3
SO3H CH3
SO3H CH3
SO3H CH3
SO3H CH3
SO3H CH3
SO3H CH3
CH3 104.5
CH3 12.8
CH3 13.1
CH3 54.0
19
MS148
MS155
MS119
MS153
MS131
MS124
MS126
MS127
MS109
MS130
MS129
MS128
MS154
MS135
MS118
MS146
OH
OH
OH
OH
CH3CH2(NH2)CH
H
CH3
CO2H
CH3
CH3
H
H
CH3
CH3
CH3
H
H
H
H
39.0
OH Cl
H
CH3 49.7
CH3 25.0
CH3 93.5
CH3 86.8
CH3 60.1
CH3 40.4
CH3 44.6
CH3 47.2
CH3 6.9
OH CH3CH2
OH CH3CH2CH2
OH CH3CH2CO
OH CH3
H
H
H
H
H
H
97
H
H
H
385
H
H
CH3
OH CH2CHCH2
OH (CH3)2CH
OH (CH3)2CH
OH Cl
H
H
CH3
H
H
CH3
H
H
(CH3)2CH
H
CH3
CH3
CH3
H
H
H
OH NH2
H
CH3 54.8
CH3 49.7
CH3 30.8
OH CH3
H
OH CH3
H
CH3 CH3
All compounds were used at 50 mM concentration. Percent inhibition was calculated by [1 ꢁ (A/B)] 3 100, where A is the difference of luciferase activity
measured between cells treated with a compound and doxorubicin and the negative control, and B is the difference of luciferase activity between cells
treated with and without doxorubicin.
a Compounds shown 80% + inhibition of p53 activity are highlighted in blue.
b Kd (mM) was determined in a titration of protein tryptophan fluorescence change as a function of ligand concentration.
the benzensulfonate (Table 1). These compounds were synthe- dent p21 luciferase activity by almost 100% as compared to
sized using a two-step reaction scheme (Scheme S1). Specifi- the controls. Using a tryptophan fluorescence binding assay
cally, the synthesis starts with treatment of a substituted sulfa- (Sachchidanand et al., 2006), we determined the binding affinity
nilic acid with concentrated HCL, and resulting amine is of MS120 to the CBP BRD to be Kd = 19 mM (Figure S2B), which
diazotized by addition of sodium nitrite to produce diazonium is consistent with its cellular p21 inhibition activity as compared
salt. The latter is added to a solution containing a substituted to those analogs that exhibited >80% inhibition of doxorubicin-
phenol to yield a corresponding diazobenzene compound. induced p53 activation in the p21 luciferase assay (Table 1).
Toward the objective of developing small molecules that could Given its effective cellular protective activity against myocardial
be used to study effects of inhibiting the stress-induced p53/ ischemia (see below), we named this compound MS120 as
CBP interaction in cells, we systematically evaluated structure- ischemin. We further determined IC50 = 5 mM for ischemin in
activity relationship (SAR) of these azobenzene analogs in their inhibition of p53-induced p21 activation using the luciferase
ability to inhibit p53 activation in human osteosarcoma (U2OS) activity assay, which is markedly stronger than an analog
cells. The p53 activation was induced by DNA damage on treat- MS119 with IC50 = ꢀ100 mM (Figure 2A). This difference agrees
ment of doxorubicin, and effects of compound treatment were with the difference in their inhibition activities observed in the
measured by p53-dependent p21 luciferase activity (Table 1; p21 luciferase activity measurement (Table 1; Figure S2A).
Figure S2A). All compounds were tested at concentration of
50 mM in U2OS cells.
Molecular Basis of Lead Recognition by the CBP BRD
Of the diazobenzene analogs, 5-(2-amino-4-hydroxy-5- To understand the molecular basis of CBP BRD recognition of
methyl-phenylazo)-2,4-dimethyl-benzenesulfonic acid, MS120, the diazobenzenes, we determined the 3D structure of the ische-
exhibited most potent cellular activity in inhibiting p53-depen- min/CBP BRD complex by using NMR (Figure 1B and Table 2;
Chemistry & Biology 18, 531–541, April 22, 2011 ª2011 Elsevier Ltd All rights reserved 533

Chemistry & Biology
A CBP Inhibitor Blocks Apoptosis in Cardiomyocytes
tent with its higher affinity (Figure S1). Ischemin binds across the
entrance of the acetyl-lysine binding pocket in an extended
conformation with its phenoxyl group forming a hydrogen bond
Table 2. NMR Statistics of the Structures of the CBP BRD/Ligand
Complexes
NMR Distance and Dihedral
Constraints
CBP BRD/
MS456
CBP BRD/
Ischemin
˚
(ꢀ2.8 A) to the amide nitrogen of Asn1168 in CBP. The latter is
a highly conserved residue in the BRDs whose amide nitrogen
is hydrogen-bonded to the acetyl oxygen of the acetyl-lysine in
a biological binding partner as seen with acetylated-lysine 20
of histone H4 recognition by the CBP BRD (Figure 1B versus Fig-
ure 1C). The sulfonate group forms electrostatic interactions with
guanidinium group of Arg1173 in the BC loop and possibly also
with side chain amide of Gln1113 in the ZA loop.
Distance constraints
Total NOE
2888
1155
1714
560
654
500
50
2805
1131
1674
554
665
455
50
Intraresidue
Interresidue
Sequential (ji – jj = 1)
Medium range (ji – jj < 4)
Long range (ji – jj > 5)
Hydrogen bonds
Intermolecular constraints
Total dihedral angle restraints
f
Ischemin in the acetyl-lysine binding pocket is sandwiched
through hydrophobic and aromatic interactions between the di-
azobenzene and Leu1109, Pro1110, and Val1174 on one side,
Leu1120 and Ile1122 in the ZA loop on the other. Because all
the diazobenzenes contain a para-phenoxyl group, a hydrogen
bond between the phenoxyl with Asn1168 is likely present in all
the compounds when bound to the CBP BRD. As such, this
structure explains our SAR data (Table 1). For instance, with
a para-sulfonate in the diazonbenzene, ortho- but not meta-
substitution of methyl groups on the phenol ring results in
a marked increase in the lead’s ability to inhibit p53-dependent
p21 luciferase activity, e.g., MS450, MS451, and MS101 versus
MS110. Ortho-substitution of a larger alkyl group such as ethyl
(MS113), isopropyl (MS105), or t-butyl (MS111) showed reduced
activity on p21 inhibition as compare to that of ortho-methyl. The
preferred small hydrophobic group at ortho-position is due to its
possible interaction with a small hydrophobic cavity formed with
Ile1122, Tyr1125, and Tyr1167 that is positioned next to the
conserved Asn1168 in the acetyl-lysine binding pocket (see
below).
53
25
81
81
81
81
j
Structure statistics
Violations (mean SD)
˚
Distance constraints (A)
0.078 0.0072
0.50 0.046
0.069 0.0064
0.49 0.19
Dihedral angle
constraints (ꢂ)
Maximum dihedral
angle violation (ꢂ)
0.58
0.63
1.03
Maximum distance
˚
constraint violation (A)
0.092
Deviations from idealized geometry
˚
Bond lengths (A)
Bond angles (ꢂ)
Impropers (ꢂ)
0.0078 0.00014 0.0076 0.00014
0.86 0.011
0.47 0.0051
0.84 0.012
0.48 0.0084
When resided at meta-position in diazobenzene, sulfonate
establishes electrostatic interactions with guanidinium side
chain of Arg1173 (Figure 1B); this alters CBP preference for
substitutions on the aromatic ring. For instance, inhibition of
p21 expression seems less sensitive to variations of size and
position of hydrophobic substituent groups on the phenol.
Nevertheless, ortho-propyl (MS126) and ortho-ethyl-keto
(MS127) substituted diazobenzenes exhibit 93.5% and 86.8%
inhibition activity, respectively. This preferred ortho-substituent
likely interacts with side chains of Ile1122, Tyr1125, and
Average pairwise rmsda (A)
˚
Heavy protein
0.76 0.088
0.34 0.074
0.71 0.12
Backbone protein
0.32 0.093
Ramachandran map analysisa (%)
Most favorable regions
78.6
76.2
18.7
3.6
Additionally allowed regions 17.1
Generously allowed regions 2.7
Disallowed regions
1.6
1.5
BRD, bromodomain; CBP, CREB-binding protein; NMR, nuclear
magnetic resonance; rmsd, root-mean-square deviation; SD, standard
deviation.
a Rmsd calculations were performed for 20 final NMR structures of the
CBP bromodomain consisting of residues 11–120. Pairwise rmsd was
calculated among the 20 refined structures.
Tyr1167,
a small hydrophobic pocket embedded in the
acetyl-lysine binding site (Figure 1B). With a meta-amino
substituent, which electron-donating functionality may aid
formation of a hydrogen bond between the phenoxyl in the
diazobenzene and side chain amide of Asn1168 of the protein,
ischemin nearly completely suppresses the p21 expression.
Although many ischemin binding residues in the acetyl-lysine
also see Figure S1B). The overall position and orientation of binding pocket are conserved among human BRDs (Sanchez
ischemin bound to CBP BRD is similar to that of the initial hit and Zhou, 2009), we observed that ischemin exhibits up to
MS456 (Figure S1). It is worth noting that binding ischemin five-fold selectivity for the CBP BRD over several other human
caused severe line broadening of several protein residues at BRDs including PCAF, BRD41, BAZ1B, and BAZ2B as deter-
the ligand-binding site, which include Pro1110, Phe1111, mined by an in vitro tryptophan fluorescence binding assay
Ile1122, Tyr1125, Ile1128, and Tyr1167. The ligand binding (Figure S2C). This level of selectivity may attribute to several
induced line-broadening resulted in a fewer number of intermo- ischemin binding residues in CBP such as Pro1110, Gln1113,
lecular NOE-derived distance constraints used for the ischemin- and Arg1173 that are not conserved in other human BRDs
bound structure determination than that for MS456, i.e., (Zeng et al., 2008). Collectively, the new structure provides
25 versus 53, respectively (see Table 2). Nevertheless, the ische- the detailed molecular basis of ischemin recognition by the
min/CBP BRD structure is better defined than the latter, consis- CBP BRD.
534 Chemistry & Biology 18, 531–541, April 22, 2011 ª2011 Elsevier Ltd All rights reserved

Chemistry & Biology
A CBP Inhibitor Blocks Apoptosis in Cardiomyocytes
Figure 2. Functional Characterization of CBP BRD Chemical Modulators in Transcription
(A) Dose-dependent inhibition of p21 luciferase activity in U2OS cells on treatment of ischemin or MS119. The luciferase activity was normalized to Renilla
luciferase as a control. The IC50 was calculated using PRISM software. Error bars represent the data obtained from the experiments repeated three times or more.
(B) Effects of the CBP BRD ligands on BRDU incorporation in U2OS cells on doxorubicin treatment. The data showing that ischemin or MS119 prevents
a doxorubicin-induced decrease of BRDU incorporation. Error bars represent the data obtained from the experiments repeated three times or more.
See also Figure S2.
Ischemin Inhibits p53 Activation on DNA Damaging
Stress
histone H3 are associated with downregulation of p21, PUMA
and 14-3-3, but not the controls of actin, histone H3, and lamin
DNA damage induced by doxorubicin leads to p53 stimulated B. Noticeably, ischemin treatment did not affect the level or func-
cellular responses including cell cycle arrest, damage repair, tional phosphorylation state of ATM and CHK1, which are the
and apoptosis (Prives and Hall, 1999; Vogelstein et al., 2000; upstream signal transducers of p53 (Figure 3B). Collectively,
Woods and Vousden, 2001). To determine the effect of ischemin these results suggest that ischemin inhibits doxorubicin-induced
on dividing U2OS cells, we treated U2OS cells with 5-bromo-2- p53 activation and transcriptional functions by altering post-
deoxyuridine (BRDU) and measured the incorporated BRDU translational modification states on p53 and histones.
during in DNA synthesis using an ELISA assay. The result
We further investigated whether ischemin downregulates p53
showed that doxorubicin treatment of U2OS cells resulted in by blocking p53 binding to CBP. We overexpressed hemagglu-
a 45% decrease of BRDU incorporation, indicative of doxoru- tinin-tagged CBP (HA-CBP) and Flag-tagged p53 (Flag-p53) in
bicin induced cell cycle arrest. However, the presence of ische- human embryonic kidney (HEK) 293T cells. Treatment of the
min or MS119 (50 mM) almost completely prevented U2OS cells 293T cells with ischemin in the presence or absence of doxoru-
from undergoing doxorubicin-induced cell cycle arrest (Fig- bicin did not affect the expression of HA-CBP or Flag-p53, or
ure 2B). Note that these results also indicate that ischemin is acetylation and phosphorylation levels on p53 as assessed by
not toxic to the cells at this concentration.
immunoprecipitation with anti-Flag antibody followed by
We then examined biochemical effects of ischemin on p53 western blot analysis using specific antibodies (Figure 3C). The
stability and function as transcription factor. We incubated results reveal that ischemin was clearly capable of inhibiting in
U2OS cells in the presence of doxorubicin with or without ische- a dose-dependent manner p53 binding to CBP, particularly on
min at concentration of 50 or 100 mM for 24 hr. Subsequently, under doxorubicin treatment (Figure 3C, lanes 8 and 9 versus
cellular proteins were subjected to western blot analysis. As lane 7). Note that p53 associated with HA-CBP is phosphory-
shown in Figure 3A, the doxorubicin-induced increased levels lated on Ser15, indicating that p53 is transcriptionally active.
of p53 protein, its Ser15-phosphorylated (p53S15p) and These results confirm that ischemin inhibits p53-induced p21
Lys382-acetylated (p53K382ac) forms underwent marked activation on doxorubicin exposure by blocking p53 recruitment
reduction in the presence of ischemin as assessed by direct of CBP, which is required for p53 target gene activation (Mujtaba
western blots of cell lysate or after immunoprecipitation. Further, et al., 2004).
we observed that p53 directed expression of its target genes
p21, PUMA and 14-3-3 s induced by doxorubicin retreatment Ischemin Inhibits p53 Cellular Signaling Pathways
was significantly decreased in the presence of ischemin whereas We next investigated selectivity of ischemin in transcription inhi-
the level of actin remained the same.
bition of p53 target genes using a RT-PCR array analysis of RNA
As a transcription factor, p53 ability to activate gene expres- isolated from biological samples of U2OS cells (Figure 4A). The
sion is also dependent on chromatin modifications. Because results show that doxorubicin treatment upregulated p53 target
CBP acetylates both histones and p53, we evaluated possible genes that include CCNB2, CCNH, CDC25C, and CDK4, but did
changes of epigenetic marks on p53 and global histones in pres- not affect housekeeping genes GAPDH, b-2 microglobulin (B2M)
ence of ischemin. The western blot analysis of the nuclear and actin (ACTB) (Figure 4B; also see Figure S3). On the other
extracts from U2OS cells revealed that p53 inhibition by ische- hand, ischemin can differentially reduce doxorubicin-induced
min is associated with an increase in histone H3 phosphorylation expression of p53 target genes CCNE2, CCNG2, CDC2,
at Ser10 and a decrease in H3 acetylation at Lys9 (Figure 3B). CDC25A, CDKN1A, CDKN2A (p21), GADD45A, E2F1, E2F3,
These changes of posttranslational modifications on p53 and PCNA, SESN1, and SESN2 (Figure 4B). These gene products
Chemistry & Biology 18, 531–541, April 22, 2011 ª2011 Elsevier Ltd All rights reserved 535

Chemistry & Biology
A CBP Inhibitor Blocks Apoptosis in Cardiomyocytes
Figure 3. Effects of Ischemin on p53 Activation Induced by DNA Damage
(A) Immunoblots showing ischemin effects on levels of endogenous p53, p53 phosphorylation on serine 15, p53 acetylation on lysine 382, as well as p53 target
genes.
(B) Immunoblots showing ischemin effects on levels of correlated H3K9 acetylation and H3S10 phosphorylation, and unaffected upstream kinases CHK1 and
ATM on doxorubicin treatment.
(C) Inhibition of overexpressed HA-tagged CBP and flag-tagged p53 interaction in 293T cells by ischemin in a concentration-dependent manner under doxo-
rubicin-induced DNA damaging condition. An arrow indicates the expressed Flag-tagged p53 in the HEK293T cells.
are known to participate in different cellular pathways driven by effectively inhibit doxorubicin-induced apoptosis in the cardio-
p53, of which the best known is CDKN1A (p21) that functions as myocytes (Figure 5A). Further, similar to U2OS cells, we
an inhibitor for cell cycle progression. Taken together, these confirmed that ischemin was able to inhibit doxorubicin induced
results confirm our hypothesis that small-molecule inhibition of p53 activation in the primary neonatal rat cardiomyocytes, but
the acetyl-lysine binding activity of the CBP BRD could downre- did not alter H2AX phosphorylation at Ser139 (Figure 5B). The
gulate p53 activation and its ability to activate its target genes latter argues that ATM is active in presence of ischemin, which
under stress conditions.
is consistent with our analysis using western blots (Figure 3B).
We further examined and concluded that ischemin likely blocks
apoptosis in cardiomyocytes by inhibiting caspase 3/7 activity in
a dose-dependent manner (Figure 5C). Finally, we ruled out that
Ischemin Functions as a Cellular Protective Agent
against Myocardial Ischemic Stress
We next investigated whether ischemin could inhibit apoptosis in ischemin’s ability to directly inhibit the lysine acetyltransferase
cardiomyocytes under DNA damage stress. Primary neonatal rat activity of CBP/p300 toward a histone H3 peptide substrate in
cardiomyocytes were isolated and maintained in culture, then, a fluorescence-based assay (data not shown). Taken together,
treated with doxorubicin for 24 hr to induce DNA damage in these results strongly demonstrate that ischemin is cell perme-
the presence or absence of ischemin. The DNA damage induced able and capable of functioning as a cellular protective agent
by apoptosis was analyzed by the TUNEL (terminal deoxynu- against myocardial damage by downregulating p53-induced
cleotidyl transferase dUTP nick and end labeling) assay, in which apoptosis under the stress conditions.
a terminal deoxynucleotidyl transferase was used to identify
30-OH of DNA generated by DNA fragmentation resulting from DISCUSSION
apoptosis, and then labels it with biotinylated dUTP. The latter
was then detected with avidin-conjugated FITC for specific Cardiac dysfunction results from the loss of myocyte mass via
staining. Using this TUNEL assay we observed that doxorubicin necrosis and apoptosis (Joaquin and Gollapudi, 2001; Kajstura
treatment induces apoptosis in the cardiomyocytes (Figure S4), et al., 1996; Vanden Hoek et al., 2003). Necrosis is a passive
and observed that ischemin, which has no toxicity of its own, can process that causes injury to cellular membranes resulting in
536 Chemistry & Biology 18, 531–541, April 22, 2011 ª2011 Elsevier Ltd All rights reserved

Chemistry & Biology
A CBP Inhibitor Blocks Apoptosis in Cardiomyocytes
Figure 4. Microarray Analysis of Effects of Ischemin on p53 Target Gene Expression, as Assessed by Using Ingenuity System
(A) Heat-map of the microarray data of selected genes in U2OS cells on doxorubicin treatment with and without ischemin. Changes of gene expression are
highlighted in red and green as for up- or downregulation as compared to the control, respectively.
(B) Schematic illustration of selected genes associated with p53 signaling pathways that underwent change of expression levels on doxorubicin treatment with
(left panel) and without (right panel) the presence of ischemin. Gene products are color-coded in red (increase in expression), green (decrease in expression), and
white (no significant change in expression).
Figure S3 depicts a more complete signaling diagram of the genes subject to the microarray analysis. A, activation; E, expression; EC, phosphorylation; I,
inhibition; M, modification; PP, protein-protein interaction; RE, reaction.
the release of intracellular contents and in the marked inflamma- is not clearly understood. In this study, we have rationally
tory responses. In contrast, apoptosis is an active, gene- designed and developed a new p53 inhibitor, ischemin, which
directed, and energy-requiring process, in which cells initiate targets p53 interaction with coactivator CBP that is required
their own death in response to either internal or external stimuli for p53 transcriptional activity. We have demonstrated that
(Coopersmith et al., 1999; Freude et al., 2000; Misao et al., ischemin can effectively inhibit transcription functions of p53
1996). The p53-dependent apoptosis is a key mechanism by on DNA damage and block cardiac myocyte apoptosis during
which the host eliminates unwanted or damaged cells thereby ischemia conditions.
ensuring the overall genome integrity (Schafer et al., 2003).
As a master regulator of transcriptional program p53 controls
Several published studies indicated that apoptosis is the major genomic stability by inducing growth arrest or apoptosis in
form of cell loss and that p53-dependent apoptosis plays a crit- response to either genotoxic stress or activation of oncogenes,
ical role in myocardial infarction and reperfusion injury and also thereby protecting normal cell populations from potential
in doxorubicin induced cardiomyopathy (Grazette et al., 2004; tumor progenitors (Vousden, 2000). p53 is frequently mutated
Komarov et al., 1999). Pifthrin-a, probably the only known p53 in tumors (Soussi and Beroud, 2001). p53-deficient tumors
inhibitor, has been shown to inactivate p53 function during are usually sensitive to ectopically expressed p53 (May and
cardiac ischemia (Gudkov and Komarova, 2005; Komarov May, 1999) and considered to be targets for p53 gene
et al., 1999; Liu et al., 2004). However, its mechanism of action therapy (Fang and Roth, 2003; Willis and Chen, 2002).
Chemistry & Biology 18, 531–541, April 22, 2011 ª2011 Elsevier Ltd All rights reserved 537

Chemistry & Biology
A CBP Inhibitor Blocks Apoptosis in Cardiomyocytes
Figure 5. Ischemin Functions as a Cellular Protective Agent against Myocardial Ischemic Stress
(A) TUNEL assay showing ischemin inhibition of doxorubicin-induced apoptosis in rat neonatal cardiomyocytes.
(B) Evaluation of ischemin effects in U2OS cells and cardiomyocytes. The immunoblots show downregulation of doxorubicin-induced activated p53 in both cell
types in the presence of ischemin, whereas levels of H2XS139p remained the same.
(C) Inhibition of doxorubicin-induced caspase 3/7 activation in cardiomyocytes by ischemin. Error bars represent the data obtained from the experiments
repeated three times or more.
See also Figure S4.
However, p53-mediated apoptosis has also been reported to transcriptional machinery for productive gene transcriptional
occur in gastrointestinal tract and hemopoietic cells during activation and elongation. Although the coactivator CBP/p300
radio and/or chemotherapy, which can result in severe and has been implicated to participate in all these steps of p53-
well-known side effects of cancer treatment (Chow et al., induced gene activation, the detailed underlying molecular
2000; Cui et al., 1995; Komarova et al., 1997a, 1997b; Song mechanism is yet to be fully elucidated. Other bromodomain-
and Lambert, 1999). Therefore, balancing p53 activation in containing proteins could also participate in the process. Given
a temporal manner likely holds therapeutic benefits. The the modest affinity of ischemin to CBP BRD, we cannot rule
primary challenge of this strategy is to block activation of out ischemin binding to possibly other bromodomain proteins
downstream targets of p53 without affecting the upstream in cells. Nevertheless, additional chemical modifications of the
pathway of p53 (such as ATR, ATM and ChK kinases). As diazobenzene scaffold could be explored, which can be guided
shown in this study, ischemin can selectively downregulate by our new insights into the molecular basis of ligand recognition
p53 down-stream target genes without affecting p53 upstream by the target protein. Furthermore, it is conceivable that the
effector proteins and protect the primary rat cardiomyocyte diazo moiety could be replaced by carbon-carbon connectivity,
cells from ischemic apoptosis through inhibiting caspase 3/7 which could result in an improvement of chemical stability of lead
activity.
As shown in previously reported studies by us and the others
compounds in cells.
In summary, the new small molecule probes, as described
(Barlev et al., 2001; Mujtaba et al., 2004, 2006), bromodomain/ in this study, lay an important foundation to facilitate future
acetyl-lysine mediated protein-protein interactions likely play functional investigation of endogenous p53 under biologically
an important role in different steps in stress-induced activation relevant conditions. Further, we expect that combined use of
of p53 in gene transcription. These include: (1) coactivator additional small molecule probes that are high affinity and selec-
CBP/p300 or PCAF binding to p53 leading to acetylation of tive for bromodomains from other chromosomal proteins such as
multiple lysine residues on p53; (2) coactivator recruitment by PCAF and BRD4 will further help us dissect detailed mechanistic
acetylated p53 to its target gene loci to increase histone lysine insights into the complex biological functions of p53 in gene tran-
acetylation; (3) recruitment of chromatin remodeling complexes scriptional control in response to cellular stress in human biology
to facilitate chromatin opening; and (4) assembly of the active of health and diseases such as myocardial ischemia.
538 Chemistry & Biology 18, 531–541, April 22, 2011 ª2011 Elsevier Ltd All rights reserved

Chemistry & Biology
A CBP Inhibitor Blocks Apoptosis in Cardiomyocytes
After 3–4 hr of transfection, cell were treated with compounds for overnight,
and then exposed to 300 ng of doxorubicin for next 24 hr. In these experi-
ments, DMSO, transfected cells with empty vector and cell without doxoru-
bicin were used as controls. DMSO concentration is maintained at 0.01%.
Transfected cells with doxorubicin treatment were used as positive control.
The luciferase activity was estimated by following the manufacturer’s instruc-
tion (Promega) in a luminometer. Both active and passive lysis of cells yielded
consistent results. The inhibitory activity (IC50) of a small molecule on p21 lucif-
erase activity was obtained from the average of three biological replicates
using PRISM software.
SIGNIFICANCE
Posttranslational modifications of proteins are essential
in cellular control of physiological processes; their dysregu-
lation has been recognized as the cause of human diseases.
Use of small molecules targeting enzymatic activities
responsible for such amino acid modifications is an
accepted approach in drug discovery as exemplified by the
development of kinase inhibitors for cancer treatments
(Chico et al., 2009). We argue, however, that small molecule
targeting of modification-mediated protein-protein interac-
tions offers unique advantages, particularly in regulation of
gene transcription in chromatin. This is because protein
functional domains likely have a fewer number of cellular
binding partners than that of a given chromatin modifying
enzymes. These protein domains have also been shown to
bind to modified amino acids of biological partners with
high specificity but modest affinity (1–100 mM) (Sanchez
and Zhou, 2009; Zeng et al., 2008), which allows necessary
dynamic regulation of cellular biology such as gene
transcription (Huang et al., 2009). As illustrated in this study,
inhibition of acetylation-mediated protein-protein interac-
tions via the bromodomain with selective small molecules
can provide a temporal modulation of gene transcription in
a chromatin context. Therefore, our study highlights an
alternative approach to the future development of new
therapeutic strategies to ameliorate stress-induced human
cardiovascular disorders such as myocardial ischemia.
BRDU Cell Cycle Analysis
BRDU incorporation assay for cell cycle evaluation was performed in 96-well
plates using calorimetric based kit from Calbiochem (catalog #QiA58). One
hundred microliters of 1 3 105/ml cells were plated in DMEM media (Medi-
atech) with 10% fetal bovine serum (FBS). After 12 hr, cells were treated
with compounds ischemin and MS119 (50 mM) with or without doxorubicin
treatment (5 mM). The controls were DMSO and untreated cells. BRDU was
added for 24 hr treatment. After 24 hr, cells were fixed and treated with anti-
BRDU antibody. After washing, the wells were incubated with peroxidase.
After final wash, the color was developed using TMB as substrate and the
reaction was stopped with stop solution and optical density was estimated
at 450 nm.
HA-CBP and Flag-p53 Pull-Down Assay
HA-CBP and Flag-p53 were transfected into human embryonic kidney (HEK)
293T cells with recommended amount of Fugene 6 (Roche). After transfection,
the HA-CBP and Flag-p53 cotransfected cells were treated with ischemin in
the presence or absence of doxorubicin. To test the inhibitory potential of
ischemin against CBP and p53 association, CBP was first immunoprecipitated
by pulling-down with HA-agarose beads (Sigma) and its association with p53
was then determined with western blot using anti-Flag antibody (Sigma).
Microarray Analysis
EXPERIMENTAL PROCEDURES
Effects of ischemin on transcription inhibition of p53 target genes were as-
sessed using a RT-PCR array study. The array was performed on RNA isolated
from three different biological repeats in U2OS cells using a set of primers
selected for a group of genes that are known to be associated within p53
signaling pathways. The differentially expressed genes in treated related to
untreated groups, i.e., doxorubicin treated versus untreated, or doxorubicin
plus ischemin versus doxorubicin alone, were subjected to pathway analysis
by using the Ingenuity System software (http://www.ingenuity.com/). The
fold changes of these genes were converted to log2Ratio and then imported
into IPA tool along with gene symbols. The enriched pathways in the gene
list were identified by Fisher exact test at p value of 0.05 and visualized in
Canonical pathway explorer.
Chemical Synthesis
All compounds were synthesized using commercially available starting mate-
rials without further purification unless otherwise stated. See Supplemental
Experimental Procedures for details.
Cell Lines, Plasmids, and Reagents
U2OS cells were grown in Eagle’s minimal essential medium (DMEM) (Medi-
atech) supplemented with 10% fetal bovine serum (Invitrogen) and antibiotics
(Invitrogen). For p53 activation, doxorubicin (Sigma) was used. The compounds
were dissolved in DMSO (Sigma). The antibodies used for immunoprecipitation
and western blot are p53 (sc-6243), p21 (sc-397), 14-3-3 (sc-7683), lamin B
(sc-6215) from Santa Cruz Biotech; p53Ser15p (9282), p53K382ac (2525),
ATM (2873), ATMp1981 (4526), CHK (2345), CHKp (2341), and PUMA (4976)
from Cell Signaling Technology; H3 (ab1791), H3KS10p (ab14955), H3K9ac
(ab4441) from ABCAM; and Actin (A4700) from Sigma.
Cardiac Myocyte Isolation
Neonatal rat ventricular myocytes were isolated by enzymatic dissociation of
cardiac ventricle from 1–2-day-old Sprague-Dawley pups using the Worthing-
ton neonatal cardiomyocyte isolation system (Worthington). Briefly, the pups
were anesthetized and their hearts were excised. The ventricular tissues
were minced in ice cold HBSS and then digested with trypsin overnight at
4ꢂC followed by collagenase treatment for 45 min at 37ꢂC. Cells were collected
by centrifugation at 800 rpm for 5 min and subsequently underwent two rounds
of preplating on culture dishes to minimize nonmyocyte contamination. The
enriched cardiomyocytes were cultured in DMEM/F12 nutrient mixture (Invitro-
gen) with 10% horse serum and 5% fetal calf serum (Invitrogen). After 48 hr, the
medium was changed to DMEM/F12 containing 1% insulin, transferrin, and
selenium media supplement (ITS; Invitrogen) and 0.1% BSA.
Western Blotting
U2OS cells were harvested cells and lysed in lysis buffer (20 mM Tris (pH 8.0),
150 mM NaCl, 1 mM EGTA, 1% Triton X-100, and 50 mM NaF) containing
protease inhibitor cocktail (Sigma). The cells were sonicated and spun down
at 14,000 rpm for 30 min at 4ꢂC. After protein estimation, 30–50 mg of lysates
were subjected to SDS-PAGE, transferred onto nitrocellulose membranes,
blocked with 5% milk/PBS and blotted with a primary antibody. Horse radish
peroxidase-labeled secondary antibodies (goat anti-mouse or anti-rabbit)
were added for 60 min at room temperature, and the blots were washed
with TBS (20 mM Tris, 150 mM NaCl, and 0.05% Tween-20) and subjected
to autoradiography after development of reaction by ECL (GE Health Care).
Apoptosis Assays in Cardiomyocytes
Caspase 3/7 and TUNEL assays were performed to assess inhibition of
apoptosis by ischemin. Caspase assay and TUNEL assays were performed
using Caspase-Glo 3/7 and DeadEnd kits from Promega. Caspase assay
was performed on live cardiomyocytes in 96-well plates on 3 different days.
Similarly, TUNEL assay was performed in triplicate on 3 different days. For
caspase assay 7500 cardiomyocytes were plated in 96-well plates. After
Luciferase Assay
U2OS Cells were transfected with p21 luciferase (1 mg) and Renilla luciferase
(100 ng) vectors in 6-well plate format using Fugene 6 (Roche). Briefly, total
of 1.1 mg of vector was incubated with 3 ml of Fugene 6 reagent for 30 min.
Chemistry & Biology 18, 531–541, April 22, 2011 ª2011 Elsevier Ltd All rights reserved 539

Chemistry & Biology
A CBP Inhibitor Blocks Apoptosis in Cardiomyocytes
treatment with compounds overnight and then doxorubicin for 24 hr, the inten-
sities of luminescence were read. Similarly, the TUNELassay was performed on
cardiomyocytes attached on coverslips. Briefly, cells were fixed with 4% para-
formaldehyde in phosphate buffer saline and permeabilized with 0.5% Tween
20 for 10 min. The TUNEL reaction was performed on cells with nucleotide
labeled with FITC by following manufacturer’s instruction.
Fang, B., and Roth, J.A. (2003). Tumor-suppressing gene therapy. Cancer Biol.
Ther. 2, S115–S121.
Ferreon, J.C., Martinez-Yamout, M.A., Dyson, H.J., and Wright, P.E. (2009).
Structural basis for subversion of cellular control mechanisms by the adeno-
viral E1A oncoprotein. Proc. Natl. Acad. Sci. USA 106, 13260–13265.
Fliss, H., and Gattinger, D. (1996). Apoptosis in ischemic and reperfused rat
myocardium. Circ. Res. 79, 949–956.
ACCESSION NUMBERS
Freude, B., Masters, T.N., Robicsek, F., Fokin, A., Kostin, S., Zimmermann, R.,
Ullmann, C., Lorenz-Meyer, S., and Schaper, J. (2000). Apoptosis is initiated
by myocardial ischemia and executed during reperfusion. J. Mol. Cell.
Cardiol. 32, 197–208.
Coordinates for the 3D NMR solution structures of CBP bromodomain bound
to ischemin and MS456 are deposited in the Protein Data Bank with accession
numbers of PDB ID 2L84 and 2L85, and RCSB ID RCSB102075 and
RCSB102076.
Gillum, B.S., Graves, E.J., and Wood, E. (1998). National hospital discharge
survey. Vital Health Stat. 13 133, 1–51.
Giordano, A., and Avantaggiati, M.L. (1999). p300 and CBP: partners for life
SUPPLEMENTAL INFORMATION
and death. J. Cell. Physiol. 181, 218–230.
Grazette, L.P., Boecker, W., Matsui, T., Semigran, M., Force, T.L., Hajjar, R.J.,
and Rosenzweig, A. (2004). Inhibition of ErbB2 causes mitochondrial dysfunc-
tion in cardiomyocytes: implications for herceptin-induced cardiomyopathy.
J. Am. Coll. Cardiol. 44, 2231–2238.
Supplemental Information includes Supplemental Experimental Procedures
and four figures and can be found with this article online at doi:10.1016/j.
chembiol.2010.12.021.
ACKNOWLEDGMENTS
Gudkov, A.V., and Komarova, E.A. (2005). Prospective therapeutic applica-
tions of p53 inhibitors. Biochem. Biophys. Res. Commun. 331, 726–736.
We acknowledge the use of the NMR facility at the New York Structural Biology
Center. G.G.-N. is recipient of the Research Supplements to Promote Diversity
in Health-Related Research Program from the National Cancer Institute. This
work was support by research grants from the National Institutes of Health
(M.-M.Z., S.M., R.J.H.).
Huang, B., Yang, X.-D., Zhou, M.-M., Ozato, K., and Chen, L.-F. (2009). Brd4
coactivates transcriptional activation of NFkB via specific binding to acety-
lated RelA. Mol. Cell. Biol. 29, 1375–1387.
Imhof, A., Yang, X.J., Ogryzko, V.V., Nakatani, Y., Wolffe, A.P., and Ge, H.
(1997). Acetylation of general transcription factors by histone acetyltrans-
ferases. Curr. Biol. 7, 689–692.
Received: September 1, 2010
Revised: December 18, 2010
Accepted: December 29, 2010
Published: April 21, 2011
Joaquin, A.M., and Gollapudi, S. (2001). Functional decline in aging and
disease: a role for apoptosis. J. Am. Geriatr. Soc. 49, 1234–1240.
Kajstura, J., Cheng, W., Reiss, K., Clark, W.A., Sonnenblick, E.H., Krajewski,
S., Reed, J.C., Olivetti, G., and Anversa, P. (1996). Apoptotic and necrotic my-
ocyte cell deaths are independent contributing variables of infarct size in rats.
Lab. Invest. 74, 86–107.
REFERENCES
Barlev, N.A., Liu, L., Chehab, N.H., Mansfield, K., Harris, K.G., Halazonetis,
T.D., and Berger, S.L. (2001). Acetylation of p53 activates transcription
through recruitment of coactivators/histone acetyltransferases. Mol. Cell 8,
1243–1254.
Komarova, E.A., and Gudkov, A.V. (1998). Could p53 be a target for thera-
peutic suppression? Semin. Cancer Biol. 8, 389–400.
Komarova, E.A., and Gudkov, A.V. (2000). Suppression of p53:
a new
approach to overcome side effects of antitumor therapy. Biochemistry
Chico, L.K., Van Eldik, L.J., and Watterson, D.M. (2009). Targeting protein
kinases in central nervous system disorders. Nat. Rev. Drug Discov. 8,
892–909.
(Mosc.) 65, 41–48.
Komarova, E.A., and Gudkov, A.V. (2001). Chemoprotection from p53-depen-
dent apoptosis: potential clinical applications of the p53 inhibitors. Biochem.
Pharmacol. 62, 657–667.
Chow, B.M., Li, Y.Q., and Wong, C.S. (2000). Radiation-induced apoptosis
in the adult central nervous system is p53-dependent. Cell Death Differ. 7,
712–720.
Komarova, E.A., Chernov, M.V., Franks, R., Wang, K., Armin, G., Zelnick, C.R.,
Chin, D.M., Bacus, S.S., Stark, G.R., and Gudkov, A.V. (1997a). Transgenic
mice with p53-responsive lacZ: p53 activity varies dramatically during normal
development and determines radiation and drug sensitivity in vivo. EMBO J.
16, 1391–1400.
Coopersmith, C.M., O’Donnell, D., and Gordon, J.I. (1999). Bcl-2 inhibits
ischemia-reperfusion-induced apoptosis in the intestinal epithelium of trans-
genic mice. Am. J. Physiol. 276, G677–G686.
Cui, Y.F., Zhou, P.K., Woolford, L.B., Lord, B.I., Hendry, J.H., and Wang, D.W.
(1995). Apoptosis in bone marrow cells of mice with different p53 genotypes
after gamma-rays irradiation in vitro. J. Environ. Pathol. Toxicol. Oncol. 14,
159–163.
Komarova, E.A., Zelnick, C.R., Chin, D., Zeremski, M., Gleiberman, A.S.,
Bacus, S.S., and Gudkov, A.V. (1997b). Intracellular localization of p53 tumor
suppressor protein in gamma-irradiated cells is cell cycle regulated and deter-
mined by the nucleus. Cancer Res. 57, 5217–5220.
Culmsee, C., Zhu, X., Yu, Q.S., Chan, S.L., Camandola, S., Guo, Z., Greig,
N.H., and Mattson, M.P. (2001). A synthetic inhibitor of p53 protects neurons
against death induced by ischemic and excitotoxic insults, and amyloid
beta-peptide. J. Neurochem. 77, 220–228.
Komarov, P.G., Komarova, E.A., Kondratov, R.V., Christov-Tselkov, K., Coon,
J.S., Chernov, M.V., and Gudkov, A.V. (1999). A chemical inhibitor of p53 that
protects mice from the side effects of cancer therapy. Science 285, 1733–
1737.
Culmsee, C., Siewe, J., Junker, V., Retiounskaia, M., Schwarz, S., Camandola,
S., El-Metainy, S., Behnke, H., Mattson, M.P., and Krieglstein, J. (2003).
Reciprocal inhibition of p53 and nuclear factor-kappaB transcriptional activi-
ties determines cell survival or death in neurons. J. Neurosci. 23, 8586–8595.
Lefer, D.J., and Granger, D.N. (2000). Oxidative stress and cardiac disease.
Am. J. Med. 109, 315–323.
Liu, X., Chua, C.C., Gao, J., Chen, Z., Landy, C.L., Hamdy, R., and Chua, B.H.
(2004). Pifithrin-alpha protects against doxorubicin-induced apoptosis and
acute cardiotoxicity in mice. Am. J. Physiol. Heart Circ. Physiol. 286, H933–
H939.
Dhalluin, C., Carlson, J., Zeng, L., He, C., Aggarwal, K., and Zhou, M. (1999).
Structure and ligand of a histone acetyltransferase bromodomain. Nature
399, 491–496.
Long, X., Boluyt, M.O., Hipolito, M.L., Lundberg, M.S., Zheng, J.S., O’Neill, L.,
Cirielli, C., Lakatta, E.G., and Crow, M.T. (1997). p53 and the hypoxia-induced
apoptosis of cultured neonatal rat cardiac myocytes. J. Clin. Invest. 99, 2635–
2643.
Eefting, F., Rensing, B., Wigman, J., Pannekoek, W.J., Liu, W.M., Cramer,
M.J., Lips, D.J., and Doevendans, P.A. (2004). Role of apoptosis in reperfusion
injury. Cardiovasc. Res. 61, 414–426.
540 Chemistry & Biology 18, 531–541, April 22, 2011 ª2011 Elsevier Ltd All rights reserved

Chemistry & Biology
A CBP Inhibitor Blocks Apoptosis in Cardiomyocytes
May, P., and May, E. (1999). Twenty years of p53 research: structural and func-
Schafer, T., Scheuer, C., Roemer, K., Menger, M.D., and Vollmar, B. (2003).
Inhibition of p53 protects liver tissue against endotoxin-induced apoptotic
and necrotic cell death. FASEB J. 17, 660–667.
tional aspects of the p53 protein. Oncogene 18, 7621–7636.
Misao, J., Hayakawa, Y., Ohno, M., Kato, S., Fujiwara, T., and Fujiwara, H.
(1996). Expression of bcl-2 protein, an inhibitor of apoptosis, and Bax, an
accelerator of apoptosis, in ventricular myocytes of human hearts with
myocardial infarction. Circulation 94, 1506–1512.
Song, S., and Lambert, P.F. (1999). Different responses of epidermal and hair
follicular cells to radiation correlate with distinct patterns of p53 and p21 induc-
tion. Am. J. Pathol. 155, 1121–1127.
Soussi, T., and Beroud, C. (2001). Assessing TP53 status in human tumours to
Mujtaba, S., He, Y., Zeng, L., Yan, S., Plotnikova, O., Sachchidanand,
Sanchez, R., Zeleznik-Le, N.J., Ronai, Z., and Zhou, M.M. (2004). Structural
mechanism of the bromodomain of the coactivator CBP in p53 transcriptional
activation. Mol. Cell 13, 251–263.
evaluate clinical outcome. Nat. Rev. Cancer 1, 233–240.
Sterner, D.E., and Berger, S.L. (2000). Acetylation of histones and transcrip-
tion-related factors. Microbiol. Mol. Biol. Rev. 64, 435–459.
Mujtaba, S., Zeng, L., and Zhou, M.M. (2006). Modulating molecular functions
Vanden Hoek, T.L., Qin, Y., Wojcik, K., Li, C.Q., Shao, Z.H., Anderson, T.,
Becker, L.B., and Hamann, K.J. (2003). Reperfusion, not simulated ischemia,
initiates intrinsic apoptosis injury in chick cardiomyocytes. Am. J. Physiol.
Heart Circ. Physiol. 284, H141–H150.
of p53 with small molecules. Cell Cycle 5, 2575–2578.
Mujtaba, S., Zeng, L., and Zhou, M.M. (2007). Structure and acetyl-lysine
recognition of the bromodomain. Oncogene 26, 5521–5527.
Vogelstein, B., Lane, D., and Levine, A.J. (2000). Surfing the p53 network.
Murphy, P.J., Galigniana, M.D., Morishima, Y., Harrell, J.M., Kwok, R.P.,
Ljungman, M., and Pratt, W.B. (2004). Pifithrin-alpha inhibits p53 signaling
after interaction of the tumor suppressor protein with hsp90 and its nuclear
translocation. J. Biol. Chem. 279, 30195–30201.
Nature 408, 307–310.
Vousden, K.H. (2000). p53: death star. Cell 103, 691–694.
Wei, J.Y. (1992). Age and the cardiovascular system. N. Engl. J. Med. 327,
Ogryzko, V.V., Schiltz, R.L., Russanova, V., Howard, B.H., and Nakatani, Y.
(1996). The transcriptional coactivators p300 and CBP are histone acetyltrans-
ferases. Cell 87, 953–959.
1735–1739.
Westphal, C.H., Rowan, S., Schmaltz, C., Elson, A., Fisher, D.E., and Leder, P.
(1997). ATM and p53 cooperate in apoptosis and suppression of tumorigen-
esis, but not in resistance to acute radiation toxicity. Nat. Genet. 16, 397–401.
Ott, M., Schnolzer, M., Garnica, J., Fischle, W., Emiliani, S., Rackwitz, H., and
Verdin, E. (1999). Acetylation of the HIV-1 Tat protein by p300 is important for
its transcriptional activity. Curr. Biol. 9, 1489–1492.
Westphal, C.H., Hoyes, K.P., Canman, C.E., Huang, X., Kastan, M.B., Hendry,
J.H., and Leder, P. (1998). Loss of ATM radiosensitizes multiple p53 null
tissues. Cancer Res. 58, 5637–5639.
Prives, C., and Hall, P.A. (1999). The p53 pathway. J. Pathol. 187, 112–126.
Prives, C., and Manley, J.L. (2001). Why is p53 acetylated? Cell 107, 815–818.
Willis, A.C., and Chen, X. (2002). The promise and obstacle of p53 as a cancer
therapeutic agent. Curr. Mol. Med. 2, 329–345.
Rogel, A., Popliker, M., Webb, C.G., and Oren, M. (1985). p53 cellular tumor
antigen: analysis of mRNA levels in normal adult tissues, embryos, and tumors.
Mol. Cell. Biol. 5, 2851–2855.
Woods, D.B., and Vousden, K.H. (2001). Regulation of p53 function. Exp. Cell
Res. 264, 56–66.
Sachchidanand, Resnick-Silverman, L., Yan, S., Mutjaba, S., Liu, W.J., Zeng,
L., Manfredi, J.J., and Zhou, M.M. (2006). Target structure-based discovery of
small molecules that block human p53 and CREB binding protein association.
Chem. Biol. 13, 81–90.
Zeng, L., Zhang, Q., Gerona-Navarro, G., Moshkina, N., and Zhou, M.M.
(2008). Structural basis of site-specific histone recognition by the bromodo-
mains of human coactivators PCAF and CBP/p300. Structure 16, 643–652.
Zhang, M., Liu, W., Ding, D., and Salvi, R. (2003). Pifithrin-alpha suppresses
p53 and protects cochlear and vestibular hair cells from cisplatin-induced
apoptosis. Neuroscience 120, 191–205.
Sakaguchi, K., Herrera, J.E., Saito, S., Miki, T., Bustin, M., Vassilev, A.,
Anderson, C.W., and Appella, E. (1998). DNA damage activates p53 through
a phosphorylation-acetylation cascade. Genes Dev. 12, 2831–2841.
Zhao, Y., Lu, S., Wu, L., Chai, G., Wang, H., Chen, Y., Sun, J., Yu, Y., Zhou, W.,
Zheng, Q., et al. (2006). Acetylation of p53 at lysine 373/382 by the histone
deacetylase inhibitor depsipeptide induces expression of p21(Waf1/Cip1).
Mol. Cell. Biol. 26, 2782–2790.
Sanchez, R., and Zhou, M.M. (2009). The role of human bromodomains in
chromatin biology and gene transcription. Curr. Opin. Drug Discov. Devel.
12, 659–665.
Chemistry & Biology 18, 531–541, April 22, 2011 ª2011 Elsevier Ltd All rights reserved 541