Early Drug-Discovery Efforts towards the Identification of EP300/CBP Histone Acetyltransferase (HAT) Inhibitors

Article abstract of DOI:10.1002/cmdc.202000007

EP300 and CBP (KAT3A/3B) are two highly homologous, multidomain, epigenetic coregulators that play central roles in transcription through the acetylation of lysine residues on histones and other proteins. Both enzymes have been implicated in human diseases, especially cancer. From a high-throughput screen of 191 000 compounds searching for EP300/CBP histone acetyltransferase (HAT) inhibitors, 18 compounds were characterized by a suite of biochemical enzymatic assays and biophysical methods, including X-ray crystallography and native mass spectrometry. This work resulted in the discovery of three distinct mechanistic classes of EP300/CBP HAT inhibitors, including two classes not previously described. The profiles of an example of each class of inhibitor are described in detail. A subsequent medicinal chemistry effort led to the development of a novel class of orally bioavailable AcCoA-competitive EP300/CBP HAT inhibitors with in vivo activity. We believe that this work will prove to be a useful guide for other groups interested in the development of HAT inhibitors.

Full text of DOI:10.1002/cmdc.202000007

Accepted Article
Title: Early Drug Discovery Efforts Towards the Identification of EP300/
CBP Histone Acetyltransferase (HAT) Inhibitors
Authors: Jonathan Edward Wilson, Annissa Huhn, Anna S. Gardberg,
Florence Poy, Francois Brucelle, Valerie Vivat, Gaurav Patel,
Chirag Patel, Richard Cummings, Robert Sims, Julian Levell,
James E. Audia, Archana Bommi-Reddy, and Nico Cantone
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemMedChem 10.1002/cmdc.202000007
Link to VoR: http://dx.doi.org/10.1002/cmdc.202000007
A Journal of
www.chemmedchem.org

10.1002/cmdc.202000007
ChemMedChem
COMMUNICATION
Early Drug Discovery Efforts Towards the Identification of
EP300/CBP Histone Acetyltransferase (HAT) Inhibitors
Annissa J. Huhn[a], Anna S. Gardberg[a], Florence Poy[a], Francois Brucelle[b], Valerie Vivat[a], Nico
Cantone[a], Gaurav Patel[c], Chirag Patel[c], Richard Cummings[a], Robert Sims[d], Julian Levell[a], James
E. Audia,[d] Archana Bommi-Reddy*[a], and Jonathan E. Wilson*[a]
Abstract: EP300 and CBP (KAT3A/3B) are two highly homologous,
multidomain, epigenetic coregulators that play central roles in
transcription via acetylation of lysine residues on histones and other
proteins. Both enzymes have been implicated in human diseases,
catalytic acetylation activity of the enzyme, which is the primary
mechanism for EP300/CBP-dependent transcription.[4] Because
dysregulation of both enzymes has been implicated in a wide
variety of human diseases, such as cancer,[5] fibrosis,[6]
inflammation,[7] and neurodegeneration,[8] and the primary
transcriptional effects of EP300 and CBP can be attributed to the
histone acetyltransferase activity, we initiated a program aimed
at the development of orally active, small molecule EP300/CBP
HAT inhibitors.
especially cancer. From
a high-throughput screen of 191,000
compounds searching for EP300/CBP HAT inhibitors, 18
compounds were characterized by a suite of biochemical enzymatic
assays and biophysical methods, including X-ray crystallography
and native mass spectrometry. This work resulted in the discovery of
three distinct mechanistic classes of EP300/CBP HAT inhibitors,
including two classes not previously described. The profiles of an
exemplar of each class of inhibitor are described in detail. A
subsequent medicinal chemistry effort led to the development of a
novel class of orally bioavailable AcCoA-competitive EP300/CBP
HAT inhibitors with in vivo activity. We believe that this work will
prove to be a useful guide for other groups interested in the
development of histone acetyltransferase inhibitors.
Histone acetyltransferases
O
(HATs)
H
O
N
H
AcCoA
N
H
N
O
Histone deacetylases
(HDACs)
NH3
Me
HO
O
O
O
P
O
O
O
O
S
N
P
N
N
O
H
H
H2N
N
HO
O
Me
OH
EP300 and CBP are two highly homologous, multidomain,
epigenetic regulatory enzymes that play central roles in
transcription.[1] The HAT domain of EP300 and CBP catalyze the
acetylation of lysine residues on histones, notably H3K18 and
H3K27, using the cofactor AcCoA (acetyl coenzyme A) (Figure
1).[2] The acetylation of histone proteins is thought to promote
transcription by weakening the interaction of positively charged
lysine side-chains of histones with the negatively charged
phosphodiester backbone of DNA.[3] In addition to the
transcriptional effects that are mediated by the modification of
chromatin, EP300/CBP can modulate the activity of non-histone
proteins through posttranslational acetylation of lysine residues.
The domains that flank the HAT domain of EP300 and CBP
participate in a variety of protein-protein interactions that serve
roles such as modulation of histone acetyltransferase activity
and recruitment of transcriptional machinery. Ultimately, many of
these interactions provide spatial and temporal control of the
N
N
OPO3H2
HO
AcCoA (acetyl coenzyme A)
Figure 1. Reversible lysine acetylation by histone acetyltransferases and
histone deacetylases and the cofactor AcCoA.
Our program was initiated with a high-throughput screen of
191,000 compounds in a TR-FRET assay designed to identify
inhibitors of the EP300 HAT domain-catalyzed acetylation of a
biotinylated H3(1-21) peptide substrate with acetyl coenzyme A.
[9]
The 2,646 compounds that exhibited ≥20% inhibition at a
concentration of 50 μM from the primary TR-FRET screen were
then retested at three concentrations of 100 μM, 50 μM, and 25
μM in the TR-FRET assay and in an orthogonal, confirmatory
EP300 HAT SPA assay which utilized the same biotinylated
H3(1-21) peptide substrate and 3H-AcCoA. Of the 482
compounds that showed a dose-response in the 3-point TR-
FRET assay and the 3-point SPA assay, 161 compounds
showed a confirmatory dose-response in a 10-point EP300 TR-
FRET assay. 96 compounds afforded Hill coefficients of ≥1.7,
almost certainly indicative of non-stoichiometric binding
interactions, in the 10-point EP300 HAT SPA assay and were
removed from the triage.[10]
The remaining 65 compounds were then subjected to a
final battery of triage assays to evaluate their potential
mechanisms of inhibition (MOI) and further validate each
compound as a legitimate starting point for a future medicinal
chemistry campaign. For triage two assays were employed: 1) a
counterscreen with the histone acetyltransferase PCAF to
assess for HAT selectivity[11] and 2) an EP300 HAT SPA assay
using 10-fold the amount of EP300 HAT enzyme to assess for
compound aggregation.[10] To broadly assign the MOI of each hit,
EP300 HAT SPA assays using either 10-fold more AcCoA (10X
[a]
Dr. A. Huhn, Dr. A. Gardberg, F. Poy, Dr. V. Vivat, N. Cantone, Dr.
R. Cummings, Dr. J. Levell, Dr. A. Bommi-Reddy, Dr. J. E. Wilson
Constellation Pharmaceuticals
215 First Street, Cambridge, MA 02142
E-mail: jonathan.wilson@constellationpharma.com,
archana.reddy@constellationpharma.com
Dr. F. Brucelle
Foghorn Therapeutics
100 Binney St., Cambridge, MA 02142
G. Patel, Dr. C. Patel
Piramal Enterprises Limited-Discovery Solutions
Ahmedabad, Gujarat 382 213
Dr. R. Sims, Dr. J. E. Audia
[b]
[c]
[d]
Third Rock Ventures
29 Newbury St. #301, Boston, MA 02116
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202000007
ChemMedChem
COMMUNICATION
Km; 10×-AcCoA SPA assay) or 30-fold more biotinylated H3(1-
21) peptide substrate (10× Km; 30×-peptide SPA assay) were
used to assess potential cofactor or peptide competition.
Compounds were removed from further consideration that
showed inhibitory activity in the PCAF HAT assay (ten
compounds), that showed reduced activity in the 10×-enzyme
SPA assay (nine compounds), or that did not show an apparent
MOI in the 10×-AcCoA or 30×-H3(1-21) peptide SPA assays.[12]
Using these criteria for compound triage, we chose 18
10×-AcCoA SPA assay and was classified as a peptide-
competitive EP300 inhibitor.
The third class of compounds was ambiguous with a
higher IC50 for both the 10×-AcCoA SPA assay and the 30×-
peptide SPA assay than in the EP300 HAT SPA assay although
to a lesser extent than the other two classes. Moreover, 3 was
the most potent validated screening hit and had shown robust
EP300 HAT binding by differential scanning fluorimetry.[16] At
this stage of the triage the MOI for this lead was not well defined
but became apparent with additional characterization (vide infra).
The inhibitory activity of the remaining compounds was
also examined against the full-length EP300 enzyme, because
up until this point in the triage an EP300 HAT domain construct
that lacked the some of the other domains of EP300 had been
compounds
for
further
biochemical
and
biophysical
characterization.
From these remaining 18 screening hits, three types of
EP300 HAT inhibitors with distinct profiles across our panel of
triage assays were identified: 1) apparent AcCoA-competitive
inhibitors, 2) apparent peptide-competitive inhibitors, and 3) a
third, ambiguous, MOI. While multiple structurally distinct leads
of each class of inhibitor were identified during this effort, one
screening hit from each of the three classes will be used to
exemplify the distinct biochemical profile of each class. Further
kinetic analyses to unambiguously determine each MOI were not
utilized. While the AcCoA-competitive inhibitor
1 and the
peptide-competitive inhibitor 2 effectively inhibited the histone
acetyl transferase activity of the full-length EP300 protein, 3 had
no activity on the full-length enzyme.
In addition to the biochemical assays described above, we
also examined the lead compounds 1, 2, and 3 in a MSD (Meso
Scale Discovery) HCT116 cell assay which evaluated the ability
of compounds to inhibit the acetylation of H3K18 and H3K27,
primary histone acetylation targets of EP300 and CBP, in a
cellular context. Of the leads identified in the HTS, only the
AcCoA-competitive inhibitor 1 showed inhibitory activity in this
assay. While we were aware that Abbvie’s EP300/CBP HAT
inhibitor, A-485, also operated via an AcCoA-competitive
mechanism, we were intrigued by the peptide-competitive
mechanism of inhibition and decided to initiate medicinal
chemistry efforts on this class with the hope of improving the
biochemical potency to a level that would give rise to cellular
activity. Despite improving the EP300 HAT SPA potency of the
peptide-competitive series to 1.0 μM, inhibitory activity was not
observed with compounds from this class in the HCT116 cell
assay at concentrations up to 30 μM. However, it remains
unclear if the peptide-competitive inhibitors have a larger
biochemical-to-cellular assay potency shift than the AcCoA-
competitive inhibitors or if this mechanism of inhibition is not
viable in a cellular context. EP300 inhibitor 3 and its analogues
displayed no activity in this assay, in line with the absence of
inhibitory activity against full length EP300 described above.
Based on this final observation, we discontinued the
development of this class of inhibitors.
carried
out,
although
subsequent
biophysical
and
crystallographic studies supported these initial assignments
(vide infra).
The first class of inhibitors has previously been reported
and are represented by LysCoA, C646, and A-485 (Figure 2).[13]
LysCoA and A-485 have been shown to bind into the AcCoA
pocket by X-ray crystallography confirming an AcCoA-
competitive mechanism of inhibition for both compounds.[14, 15]
The screening hit 1 displayed an 14-fold shift in IC50 when
comparing the 10×-AcCoA SPA assay and EP300 HAT SPA
assay, had a modest shift in potency in the 30×-peptide SPA
assay, and was classified as an AcCoA-competitive EP300
inhibitor (Table 1).
F
NO2
O
NO2
CF3
Me
N
O
N
N
N
O
Me
O
O
O
O
O
Me
Me
N
H
H2N
MeHN
C-646
N
N
A-485
N
O
O
N
OH
O
S
O
O
N
N
N
H
P
H
H
OH
O
P
OPO3H2
HO
O
O
AcHN
NH2
Lys-CoA
Table 1. Key assays and mechanism of inhibition classification for three
representative hits from the EP300 HTS triage. Biochemical data are reported
as the mean (+SD) for 2 or more replicates.
CF3
Cl
O
H
O
N
N
SO2NH2
N
Me
Ph
N
N
H
Me
N
Me
H
Assay
1
2
3
H
1
2
3
EP300 HAT SPA IC50 (μM)
2.7 (+0.7)
3.8 (+0.6)
0.39 (+0.1)
10×-AcCoA SPA IC50 (μM)
30×-peptide SPA IC50 (μM)
Full length EP300 SPA IC50 (μM)
HCT116 MSD EC50 (μM)
37 (+9)
12 (+4)
0.62 (+0.1)
1.9
3.1 (+0.1)
94 (+50)
5.2 (+4)
>30
0.94 (+0.4)
5.0 (+1)
>160
Figure 2. Structures of AcCoA-competitive EP300/CBP HAT inhibitors C-646,
A-485, and Lys-CoA (upper panel) and screening hits 1, 2, and 3.
>30
The second class, peptide-competitive inhibitors, has not
previously been described. Compound 2 displays a 25-fold shift
in IC50 when comparing the 30×-peptide SPA assay and EP300
HAT SPA assay but shows essentially no shift in potency in the
AcCoA-
competitive
inhibitor
Peptide-
competitive
inhibitor
Preliminary MOI classification
Ambiguous
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202000007
ChemMedChem
COMMUNICATION
Additional support for the mechanistic classifications of
compounds 1, 2, and 3 came from obtaining X-ray cocrystal
structures with constructs of EP300.[17,18] An overlay of the
EP300 X-ray cocrystal structures of compounds 1 and 2 with the
EP300 X-ray cocrystal structure of LysCoA, a bi-substrate
inhibitor of EP300, nicely displays the observations made during
our biochemical studies. The AcCoA-competitive inhibitor 1
occupies the AcCoA-binding pocket of EP300 and overlays with
the CoA fragment of LysCoA, while the peptide-competitive
inhibitor 2 fills in the peptide substrate binding-channel and
overlays with the lysine fragment of LysCoA.
The X-ray cocrystal structure of inhibitor 3, which shows
that 3 binds to an allosteric site adjacent to the peptide binding-
channel and AcCoA-binding site but not directly in either of
these sites, providing insight into our preliminary MOI studies.
Interestingly, the amino acid residues that make up the allosteric
binding site of the EP300 HAT domain construct that was used
to generate the X-ray cocrystal structure with 3 are present in
the full-length protein. However, this site must not adopt the
same conformation, be sterically occluded, or have the same
affinity for compounds such as 3 in the context of the full-length
enzyme because we observe no inhibitory activity of 3 against
the full-length EP300 enzyme although well characterized
allosteric inhibitors have been described.[19]
compounds 1 and 3 in the absence of AcCoA, the peptide-
competitive compound 2 showed no detectable binding to
EP300 without AcCoA present (Figure 4C, 4G, and 4E). An
unexplained 1:2 EP300:inhibitor complex was also observed for
compound 1 in the absence of AcCoA. When EP300 was
incubated with AcCoA prior to the introduction of 1, the
EP300:AcCoA complex was observed but not the EP300:1
complex which is consistent with the AcCoA-competitive
classification and relatively weak activity of 1. Interestingly, a
more potent AcCoA-competitive inhibitor, 10, displaces
approximately 10% of the EP300:AcCoA complex to form a 1:1
EP300:10 complex in the same experiment (Figure 4J).
Morever, in the absence of AcCoA a 1:1 complex of EP300:10 is
predominantly formed, whereas in the similar experiment with 1,
a substantial amount of uncomplexed “apo”-EP300 remains
(Figure 4I vs. 4C). Binding of 2 to EP300 was only observed in
the presence of AcCoA as a 1:1:1 EP300:AcCoA:2 complex,
demonstrating that AcCoA stabilizes inhibitor binding (Figure 4F).
This observation and the significant shift in IC50 at higher
substrate concentration (Table 1) are consistent with the
previously determined Theorell-Chance mechanism where the
peptide substrate associates only transiently with the enzyme
and displays a mixed type inhibition with AcCoA and a
[14a]
competitive mechanism with substrate.
Interestingly, for 2,
which occupies the same site as substrate, the ternary complex
is clearly observed by native mass spectrometry. Whether there
are subtle differences in the binding of 2 relative to substrate or
as a consequence of the mass spectrometry conditions is
unknown. Attempts to detect the association of peptide with
EP300 in the presence or absence of AcCoA were unsuccessful
(data not shown). Finally, for compound 3, the ratio of the
EP300:AcCoA:3
complex
to
EP300:AcCoA
complex
(approximately 2:3) in the AcCoA-preincubation experiment is
lower than the ratio of EP300:3 to “apo”-EP300 (approximately
1:9). These results are consistent with our biochemical
experiments that the binding affinity of 3 to EP300 is reduced by
AcCoA (~ 3-fold) but not directly competitive with AcCoA (Figure
4G and 4H).
Based on our biochemical and biophysical characterization
of the three classes of EP300 HAT inhibitors, we elected to
focus our medicinal chemistry campaign on the optimization of
the AcCoA-competitive lead series. To optimize the potency of 1,
we first sought to identify the minimum pharmacophore
necessary for EP300/CBP HAT inhibition. Removal of the
sulfonamide provided compound 4, which resulted in minimal
loss of activity. Efforts were then made to identify alternatives for
the indole and two benzene moieties of the lead 1 to improve the
physicochemical properties. Although we explored a variety of
heterobicyclic rings and heterocycles as alternatives to the
indole system, suitable replacements were not identified.
Additionally, replacement of either of the two benzene rings of 4
with aliphatic groups or heterocycles led to a substantial loss of
Figure 3. Overlays of x-ray cocrystal structures of lysCoA (green), 1 (blue), 2
(yellow), and 3 (purple) with EP300.
More evidence for the mechanistic classifications was
obtained from native mass spectrometry experiments in which
an enzymatically-inactive EP300 construct was incubated with
compounds 1, 2, and 3 in the presence or absence of AcCoA
and the binding interactions were characterized by mass
spectrometry.[20] EP300 HAT was detected as a monomer
carrying primarily 10, 11, or 12 protons. EP300 HAT that was
preincubated with AcCoA shifted towards a lower protonation
state, mainly +9, +10 or +11, indicating a more compact
structure of the HAT domain bound to its cofactor (Figure 4A
and 4B). While 1:1 EP300:inhibitor complexes were observed for
activity. These studies defined
4
as the minimum
pharmacophore of 1 which was used as the starting point for
optimizing the potency of the series (Scheme 1).
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202000007
ChemMedChem
COMMUNICATION
potency. Methylation of the phenethylamine nitrogen or the
methine carbon adjacent to the carbonyl resulted in complete
loss of activity. The aza-indole and pyridine variants of 4, where
one of the benzenes is replaced with a pyridine, were also
explored but in general these nitrogen analogues were less
potent than 4 or had other undesirable properties that led to
retaining the indole and benzene moieties of 4.[21]
Define Minimum
Pharmacophore
O
H
O
H
N
N
SO2NH2
Ph
Ph
N
N
H
H
1
4
= 4.2
= 2.4
= 1.9
µ
µ
M
SPA IC50
MSD EC50
M
M
SPA IC50
= n.a.
µ
MSD EC50
phenethylamine
Establish Minimum Pharmacore's SAR
indole
O
N
C-6
Chlorine and Nitrogen "Walk"
N
para-position
Methyl "Walk"
C-2
central benzene
O
H
O
H
N
N
Cl
Cl
SO2NH2
Ph
Ph
N
N
Me
H
H
5
= 2.7
6
O
H
= 0.87
= n.a.
µ
SPA IC50
M
µ
N
M
= n.a.
SPA IC50
MSD EC50
MSD EC50
Cl
Ph
N
H
Indole C-6 SAR
Phenethyl SAR
7
>50 Analogues
-
-
-
>30 Analogues
Limited SAR
-
-
= 0.17
= 0.14
µ
SPA IC50
MSD EC50
M
M
Many groups tolerated
Vector to solvent front
µ
Combine SAR
SAR is additive
O
H
O
H
-
-
-
Me
N
N
N
N
Good potency
poor solubility
CN
SO2NH2
Ph
Ph
N
N
H
H
9
8
= 0.19
= 0.23
µ
M
= 0.48
µ
SPA IC50
MSD EC50
SPA IC50
MSD EC50
M
M
= 0.043
µ
M
µ
O
H
Me
N
N
N
CN
Ph
N
Improve solubility
by 10-fold
H
10
= 0.041
= 0.037
µ
SPA IC50
M
M
µ
Me
MSD EC50
O
H
O
H
N
O
N
O
CN
CN
Ph
EtHN
Figure 4. Native mass spectrometry experiments with EP300 and compounds
1 (MW = 433.53), 2 (MW = 270.38), 3 (MW = 310.7) and 10 (MW =459.55).
The ratios above each peak are representative of the ratio of EP300:inhibitor
for binary complexes or as the ratio of EP300:AcCoA:inhibitor for ternary
complexes. The red arrow indicates the change in mass for complex formation.
A. MS of EP300 showing the charge distribution of “apo”-EP300. B. MS of
EP300 in the presence of AcCoA showing the charge distribution of the
complex. C. MS of EP300 with 1. D. MS of EP300 preincubated with AcCoA
then treated 1. E. MS of EP300 with 2. F. MS of EP300 preincubated with
AcCoA then treated with 2. G. MS of EP300 with 3. H. MS of EP300
preincubated with AcCoA then treated with 3. I. MS of EP300 with 10. J. MS of
EP300 preincubated with AcCoA then treated with 10.
Ph
EtHN
N
H
N
H
Improve Potency
by 2-fold
11
12
= 0.024
= 0.036
= 0.0088
= 0.019
µ
µ
M
µ
M
SPA IC50
MSD EC50
M
M
SPA IC50
MSD EC50
µ
Scheme 1. Medicinal chemistry efforts toward the development of potent and
orally bioavailable AcCoA-competitive EP300/CBP HAT inhibitors.[22]
Based on the chlorine “walk” study described above, the
SAR of indole C-6 position and para-position of the
phenethylamine side chain were explored in more depth.
Because the indole C-6 substituent points towards the solvent
front, many substituents were tolerated at this position. Of the
more than 50 groups examined, a pyrazole, as in 8, was found
to have the largest impact on potency. As for the para-position of
the phenethylamine side chain, the nitrile analogue 9 provided
the largest improvement in potency from the more than 30
groups that were surveyed. These two findings led to the
synthesis of 10, which demonstrated that the indole C-6 SAR
and the phenethylamine SAR were additive. Compound 10 also
displayed modest levels of bioavailability in mice, which enabled
early in vivo experiments for the program. However, 10 was
hampered by poor solubility which limited oral exposure in vivo.
This issue was addressed by exploring alternative groups at C-6,
such as amides, and resulted in the discovery of 11, which
maintained the potency and mouse PK properties of 10 while
improving the solubility. Incorporation of a methyl group in the
Each carbon and nitrogen atom of 1 or 4 was then
systematically substituted with a chlorine atom (chlorine ‘walk’)
or a methyl group (methyl ‘walk’) to probe for vectors that would
allow for the introduction of substituents to optimize the potency
and physical properties. Introduction of a chloride substituent at
the indole C-6 position, as in 5, or at the para-position of the
phenethylamine side chain, as in 6, resulted in approximately a
2-fold and 5-fold improvement in potency, respectively. The
central benzene ring did not tolerate a chloride substituent at
any position. While methylation was tolerated to varying degrees
on the indole nitrogen, the indole C-2 position, and on both
carbon atoms of the aliphatic portion of the phenethylamine
sidechain, only methylation at the benzylic position on the
phenethylamine side-chain, as in 7, led to an improvement in
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202000007
ChemMedChem
COMMUNICATION
phenethylamine side chain, as in 7, resulted in 12, which further
improved the biochemical and cellular potency while retaining
the other favorable attributes of 11. Compounds 10, 11, and 12
all showed activity against both full-length EP300 and CBP.
Additionally, the potency of 10, 11, and 12 compare favorably
with A-485 in our assays. Finally, the enantiomer of 10, the
enantiomer of 11, and the other three stereoisomers of 12 were
much less active in the EP300 HAT SPA assay.[23, 24]
preliminary
mechanistic
classification
in
biochemical
experiments, in native mass spectrometry and fully corroborated
by X-ray crystallography.. This work led us to deemphasize the
peptide-competitive and allosteric inhibitors and focus our
subsequent medicinal chemistry campaign on the development
of the AcCoA-competitive lead class. This campaign involved a
detailed SAR study that enabled the development of HTS hit 1
into 12 which had the requisite potency, physicochemical
properties, and pharmacokinetics to inhibit EP300 HAT activity
in vivo. In addition, these studies revealed two previously
undescribed modes of EP300 HAT inhibition, with inhibitors
occupying the peptide binding site or a proximal allosteric site.
While it was determined that the well-established AcCoA-
competitive mechanism of inhibition holds the most promise for
the development of EP300/CBP HAT inhibitors with in vivo
activity, we believe that this work will prove to be a useful guide
for other groups interested in the development of inhibitors of
other members of the histone acetyltransferase family. Using
this work as a foundation, we have successfully developed
Table 2. Mouse PK, ADME, and enzymatic activity of 10, 11, and 12.
10
3.5
11
3.0
2.4
79.8
26
12
Mouse PK
Cl (L/h/kg)
1/2 (h)
T
1.4
Oral AUC/dose
Bioavailability (%)
Cl (mL/min/kg)
121
42
Mouse liver microsomes
Plasma protein binding
Solubility
46.6
99.1
<1.0
2.5
123
98.1
13
150
98.3
14
% unbound
pH = 7.4 (μM)
Enzymatic activity
EP300 SPA IC50 (nM)
CBP SPA IC50 (nM)
1.3
16
0.71
8.7
EP300/CBP
HAT
inhibitors
with
further
improved
<3.2
pharmacokinetic and physicochemical properties, and this work
will be the subject of a future report.
Compound 12 was then tested in PK/PD assay that
evaluated the compounds’ ability to inhibit the histone
acetyltransferase activity of EP300 and CBP in vivo. In this
experiment, 12 was orally administered to C57BL/6 mice at
three doses, 50 mg/kg, 20 mg/kg, and 7.5 mg/kg, and the
reduction of H3K27 acetylation in PBMCs and the reduction of
H3K18 acetylation in spleen tissue was monitored by FACS and
MSD, respectively. Compound 12 reduced acetylation at H3K27
in plasma and H3K18 in spleen in a dose- and time-dependent
manner, consistent with our expectations based on its in vitro
profile.
Experimental Section
Experimental details for biochemical assays, cellular assays, X-ray
crystallography, compound synthesis, and in vivo experiments can be
found in the accompanying supporting information documents.
Acknowledgements
The support of the WuXi AppTec HTS group, particularly Liu
Yingjie, and Zhu Ghuozen, is gratefully acknowledged.
Additionally, native mass spectrometry studies were carried out
on instrumentation at the UMass Amherst Mass Spectrometry
Center under the direction of Dr. Stephen Eyles who also
provided helpful technical advice and direction.
Keywords: EP300, CBP, histone acetyltransferase, epigenetics
[1]
[2]
B. M. Dancy, P. A. Cole, Chem. Rev. 2015, 115, 2419-2452.
a) V. V. Ogryzko, R. L. Schiltz, V. Russanova, B. H. Howard, Y.
Nakatani, Cell 1996, 87, 953. b) A. J. Bannister, T. Kouzarides, Nature
1996, 384, 641.
Figure 5. Oral administration of 12 to C57BL/6 mice leads to a dose- and
time-dependent reduction of histone acetylation in vivo. A) Reduction of
H3K27 acetylation in PBMCs as measured by FACS (blue bars) and plasma
levels of 12 (orange dots) over time. B) Reduction of H3K18 acetylation in
spleen tissue as measured by MSD (blue bars) and associated levels of 12
(orange dots) in spleen tissue over time.
[3]
[4]
a) V. G. Allfrey, R. Faulkner, A. E. Mirsky, Proc. Natl. Acad. Sci. U.S.A.
1964, 51, 786. b) T. R. Hebbes, A. W. Thorne, C. Crane-Robinson,
EMBO J. 1988, 7, 1395.
B. T. Weinert, T. Narita, S. Satpathy, B. Srinivasan, B. K. Hansen, C.
Scholz, W. B. Hamilton, B. E. Zucconi, W. W. Wang, W. R. Liu, J. M.
Brickman, E. A. Kesicki, A. Lai, K. D. Bromberg, P. A. Cole, C.
Choudhary, Cell 2018, 174, 1-14.
In summary, a high-throughput screen for inhibitors of
EP300 HAT activity provided 18 hits which were profiled in
biochemical assays and validated by biophysical methods.
Three classes of compounds, AcCoA-competitive inhibitors,
peptide-competitive inhibitors, and a third class of inhibitors with
distinct mechanisms of EP300 HAT inhibition were identified.
The behaviour of each class of inhibitor was consistent with its
[5]
a) N. Attar, S. K. Kurdistani, Cold Spring Harbor Perspect. Med. 2017, 7,
a026534. b) A. Farria, W. Li, S. Y. R. Dent, Oncogene 2015, 34,
4901−4913.
[6]
[7]
A. K. Ghosh, A. K. J. Varga, J. Cellular Physiology 2007, 12, 663-671.
M. Ghizzoni, H. J. Haisma, H. Maarsingh, F. J. Dekker, Drug Discovery
Today 2011, 16, 504-511.
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202000007
ChemMedChem
COMMUNICATION
[8]
a) L. M. Valor, J. Viosca, J. P. Lopez-Atalaya, A. Barco Current
Pharmaceutical Design 2013, 19, 5051-5064. b) S. W. Min, X. Chen, T.
E. Tracy, Y. Li, Y. Zhou, C. Wang, K. Shi-rakawa, S. S. Minami, E.
Defensor, S. A. Mok, P. D. M. Sohn, B. Schilling, X. Cong, L. Ellerby, B.
W. Gibson, J. Johnson, N. Krogan, M. Shamloo, J. Gestwicki, E.
Masliah, E. Verdin, L. Gan, Nature Medicine 2015, 21, 1154-1162.
The EP300 HAT domain construct used for screening was prepared as
in: P. R. Thompson, D. Wang, L. Wang, M. Fulco, N. Pediconi, D.
Zhang, W. An, Q. Ge, R. G. Roeder, J. Wong, M. Levrero, V. Sartorelli,
R. J. Cotter, P. A. Cole, Nature Stuctural Mol. Biol. 2004, 11, 308-315.
[16] A. S. Gardberg, A. J. Huhn, R. Cummings, A. Bommi-Reddy, F. Poy, J.
Setser, V. Vivat, F. Brucelle, J. E. Wilson, Structural Dynamics 2019, 6,
054702.
[17] For compound 1 a semisynthetic EP300 construct was used, as
described in reference 14a. For compounds 2 and 3 a catalytically-
inactive EP300 construct, EP300 HAT Y1467F, that omits the
autoacetylation-prone autoinhibitory loop was used to improve
homogeneity. See the supporting information and reference 16 for
additional details.
[9]
[18] The X-ray cocrystal structures of compounds 1, 2, and 3 have been
deposited in RSC PDB under the codes 6v8b, 6v8k, and 6pf1,
respectively.
[10] B. K. Schoichet, J. Med. Chem. 2006, 49, 7274-7277.
[11] a) X.-J. Yang, V. V. Ogryzko, J. Nishikawa, B. H. Howard, Y. Nakatani,
Nature, 1996, 382, 319-324. b) K. G. Tanner, M. R. Langer, and J. M.
Denu, Biochemistry 2000, 39, 11961-11969.
[19] C. Milite, A. Feoli, K. Sasaki, V. La Pietra, A.L. Balzano, L. Marinelli, A.
Mai, E. Novellino, S. Castellano, A. Tosco, G. Sbardella, J. Med. Chem.
2015, 58, 2779-2798.
[12] Y. Cheng, W. H. Prusoff. Biochem. Pharmacol. 1973, 22(23), 3099-
3108.
[20] The EP300 HAT Y1467F construct that used for x-ray crystallography
was also used for MS experiments. For a reviews of native mass
spectrometry, see: a) V. Vivat, H. C Atmanene, D. Zeyer, A. Van
Dorsselaer, S. Sanglier-Cianférani. Future Med. Chem. 2010, 2, 35-50.
b) A. C. Leney, A. J. R. Heck, J. Am. Mass Spectrom. 2017, 28, 5-13.
[21] Biochemical data for additional analogues from the chlorine, methyl,
and nitrogen “walk” studies can be found in the supporting information.
[22] The data represents the average of three or more separate duplicate
runs. Standard deviations and standard error of the mean can be found
in the supporting information.
[13] a) O. D. Lau, T. K. Kundu, R. E. Soccio, S. Ait-Si-Ali, E. M. Khalil, A.
Vassilev, A. P. Wolffe, Y. Nakatani, R. G. Roeder, P. A. Cole, Mol. Cell
2000, 5, 589. b) E. M. Bowers, G. Yan, C. Mukherjee, A. Orry, L. Wang,
M. A. Holbert, N. T. Crump, C. A. Hazzalin, G. Liszczak, H. Yuan, C.
Larocca, S. A. Saldanha, R. Abagyan, Y. Sun, D. J. Meyers, R.
Marmorstein, L. C. Mahadevan, R. M. Alani, P. A. Cole, Chem. Biol.
2010, 17, 471. c) R. M. Michaelides, A. Kluge, M. Patane, J. H. Van
Drie, C. Wang, T. M. Hansen, R. M. Risi, R. Mantei, C. Hertel, K.
Karukurichi, A. Nesterov, F. McElligott, P. de Vries, J. William, J. W.
Langston, P. A. Cole, R. Marmorstein, H. Liu, L. Lasko, K. Bromberg, K.
A. Lai, E. A. Kesick, ACS Med. Chem. Lett. 2018, 9, 28-33.
[14] a) X. Liu, L. Wang, K. Zhao, P. R. Thompson, Y. Hwang, R.
Marmorstein, P. A. Cole, Nature 2008, 451, 846-850. b) L. M. Lasko, C.
S. Jakob, R. P. Edalji, W. Qiu, D. Montgomery, E. L. Digiammarino, T.
M. Hansen, R. M. Risi, R. Frey, V. Manaves, B. Shaw, M. Algire, P.
Hessler, L. T. Lam, T. Uziel, E. Faivre, D. Ferguson, F. G. Buchanan, R.
L. Martin, M. Torrent, G. G. Chiang, K. Karukurichi, J. W. Langston, B.
T. Weinert, C. Choudhary, P. DeVries, J. H. VanDrie, D. McElligott, E.
Kesicki, R. Marmorstein, C. Sun, P. A. Cole, S. Rosenberg, M. R.
Michaelides, A. Lai, K. D. Bromberg, Nature 2017, 550, 128−132.
[15] A-485 was tested in the EP300 HAT SPA, EP300 SPA, and CBP SPA
assays. The IC50s for A-485 in our assays were 0.30 µM, 0.010 µM, and
0.034 µM, respectively. LysCoA was evaluated in the EP300 HAT SPA
assay; IC50 = 0.20 µM.
[23] The absolute stereochemistry of the compounds in Scheme 1 was
assigned by analogy to X-ray cocrystal structures of an EP300 HAT
construct with compound 1 and two related CBP/p300 HAT inhibitors
not described in this manuscript. These structures have been deposited
in the PDB under the respective codes 6v8b, 6v90, and 6v8n. See the
supporting information for additional details.
[24] The IC50s of the stereoisomers of 10, 11, and 12 in the EP300 HAT
SPA assay are as follows: enantiomer of 10 (2.9 µM), the enantiomer of
11 (0.93 µM), and the three stereoisomers of 12 (0.34 µM, 0.45 µM,
and 0.72 µM).
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202000007
ChemMedChem
COMMUNICATION
Entry for the Table of Contents (Please choose one layout)
Layout 1:
COMMUNICATION
Dysregulation of the histone
acetyltransferases, EP300 and
CBP, have been implicated in a
wide variety of human diseases.
Three distinct mechanistic classes
of EP300/CBP HAT inhibitors
have been characterized by a
variety of biochemical and
Annissa J. Huhn, Anna S.
Gardberg, Florence Poy, Francois
Brucelle, Valerie Vivat, Nico
Cantone, Gaurav Patel, Chirag
Patel, Richard Cummings, Robert
Sims, Julian Levell, James E. Audia,
Archana Bommi-Reddy*, and
Jonathan E. Wilson*
biophysical methods. This work
resulted in the discovery of a
structurally novel and highly
potent EP300/CBP HAT inhibitor
that displays good activity in vivo.
Page No. – Page No.
Early Drug Discovery Efforts
Towards the Identification of
EP300/CBP Histone
Acetyltransferase (HAT)
Inhibitors
Layout 2:
COMMUNICATION
Author(s), Corresponding Author(s)*
((Insert TOC Graphic here))
Page No. – Page No.
Title
Text for Table of Contents
This article is protected by copyright. All rights reserved.