GNE-781, A Highly Advanced Potent and Selective Bromodomain Inhibitor of Cyclic Adenosine Monophosphate Response Element Binding Protein, Binding Protein (CBP)

Article abstract of DOI:10.1021/acs.jmedchem.7b00796

Inhibition of the bromodomain of the transcriptional regulator CBP/P300 is an especially interesting new therapeutic approach in oncology. We recently disclosed in vivo chemical tool 1 (GNE-272) for the bromodomain of CBP that was moderately potent and selective over BRD4(1). In pursuit of a more potent and selective CBP inhibitor, we used structure-based design. Constraining the aniline of 1 into a tetrahydroquinoline motif maintained potency and increased selectivity 2-fold. Structure-activity relationship studies coupled with further structure-based design targeting the LPF shelf, BC loop, and KAc regions allowed us to significantly increase potency and selectivity, resulting in the identification of non-CNS penetrant 19 (GNE-781, TR-FRET IC₅₀ = 0.94 nM, BRET IC₅₀ = 6.2 nM; BRD4(1) IC₅₀ = 5100 nM) that maintained good in vivo PK properties in multiple species. Compound 19 displays antitumor activity in an AML tumor model and was also shown to decrease Foxp3 transcript levels in a dose dependent manner.

Full text of DOI:10.1021/acs.jmedchem.7b00796

Subscriber access provided by Gothenburg University Library
Article
GNE-781, A Highly Advanced Potent and Selective
Bromodomain Inhibitor of Cyclic Adenosine Monophosphate
Response Element Binding Protein, Binding Protein (CBP)
F. Anthony Romero, Jeremy M. Murray, Kwong Wah Lai, Vickie Tsui, Brian K. Albrecht, Le An, Maureen
H. Beresini, Gladys de Leon Boenig, Sarah M. Bronner, Emily W. Chan, Kevin X. Chen, Zhongguo
Chen, Edna F. Choo, Kyle Clagg, Kevin Clark, Terry D. Crawford, Patrick Cyr, Denise De Almeida
Nagata, Karen E. Gascoigne, Jane L. Grogan, Georgia Hatzivassiliou, Wei Huang, Thomas L. Hunsaker,
Susan Kaufman, Stefan G. Koenig, Ruina Li, Yingjie Li, Xiaorong Liang, Jiangpeng Liao, Wenfeng
Liu, Justin Q. Ly, Jonathan Maher, Colin Masui, Mark Merchant, Yingqing Ran, Alexander M. Taylor,
John S. Wai, Fei Wang, Xiaocang Wei, Dong Yu, Bing-Yan Zhu, Xiaoyu Zhu, and Steven R. Magnuson
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b00796 • Publication Date (Web): 11 Sep 2017
Downloaded from http://pubs.acs.org on September 11, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Wang, Fei; WuXi AppTec
Wei, Xiaocang; WuXi AppTec,
Yu, Dong; WuXi AppTec
Zhu, Bing-Yan; Genentech
Zhu, Xiaoyu; Wuxi AppTec Co., Ltd.
Magnuson, Steven; Genentech, Discovery Chemistry
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
GNE-781, A Highly Advanced Potent and
Selective Bromodomain Inhibitor of Cyclic
Adenosine Monophosphate Response Element
Binding Protein, Binding Protein (CBP)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
†
‡
†
§
F. Anthony Romero, * Jeremy Murray, Kwong Wah Lai, Vickie Tsui, Brian K. Albrecht, Le
†
†
†
†
†
An, Maureen H. Beresini, Gladys de Leon Boenig, Sarah M. Bronner, Emily W. Chan, Kevin
‡
‡
†
†
†
†
X. Chen, Zhongguo Chen, Edna F. Choo, Kyle Clagg, Kevin Clark, Terry D. Crawford,
†
†
†
†
Patrick Cyr, Denise de Almeida Nagata, Karen E. Gascoigne, Jane L. Grogan, Georgia
†
‡
†
†
†
Hatzivassiliou, Wei Huang, Thomas L. Hunsaker, Susan Kaufman, Stefan G. Koenig, Ruina
†
‡
†
‡
‡
†
†
Li, Yingjie Li, Xiaorong Liang, Jiangpeng Liao, Wenfeng Liu, Justin Ly, Jonathan Maher,
†
†
†
§
‡
‡
Colin Masui, Mark Merchant, Yingqing Ran, Alexander M. Taylor, John Wai, Fei Wang,
‡
‡
†
‡
†
Xiaocang Wei, Dong Yu, BingꢀYan Zhu, Xiaoyu Zhu, Steven Magnuson*
†
Genentech, Inc. 1 DNA Way, South San Francisco, California 94080, United States
Wuxi Apptec Co., Ltd., 288 Fute Zhong Road, Waigaoqiao Free Trade Zone, Shanghai
‡
2
00131, People’s Republic of China
§
Constellation Pharmaceuticals, Inc. 215 First Street, Suite 200, Cambridge, Massachusetts
2142, United States
0
ACS Paragon Plus Environment

Page 3 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Abstract
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Inhibition of the bromodomain of the transcriptional regulator CBP/P300 is an especially
interesting new therapeutic approach in oncology. We recently disclosed in vivo chemical tool 1
(GNEꢀ272) for the bromodomain of CBP that was moderately potent and selective over
BRD4(1). In pursuit of a more potent and selective CBP inhibitor we used structureꢀbased
design. Constraining the aniline of 1 into a tetrahydroquinoline motif maintained potency and
increased selectivity 2ꢀfold. Structureꢀactivity relationship studies coupled with further structureꢀ
based design targeting the LPF shelf, BC loop and KAc regions allowed us to significantly
increase potency and selectivity, resulting in the identification of nonꢀCNS penetrant 19 (GNEꢀ
7
81; TRꢀFRET IC50 = 0.94 nM, BRET IC50 = 6.2 nM, BRD4(1) IC50 = 5,100 nΜ) that
maintained good in vivo PK properties in multiple species. Compound 19 displays antiꢀtumor
activity in an AML tumor model and was also shown to decrease Foxp3 transcript levels in a
dose dependent manner.
Keywords: Epigenetics, CBP, CREBBP, P300, EP300, bromodomain, bromodomain inhibitor
Introduction
The histone acetyl transferase (HAT) cAMP response element binding protein, binding
protein (CREBBP, CBP) and highly homologous paralog, adenoviral E1A binding
protein of 300 kDa (EP300, P300) are large, multidomain proteins that each contain a
1ꢀ3
single bromodomain and contribute to transcriptional regulation. CBP/P300 (hereafter
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
together referred to as “CBP”) bind to chromatin via their bromodomain. Once associated
with chromatin, CBP modifies chromatin through its HAT activity, leading to the
recruitment of various transcriptional proteins to modulate gene expression. CBP has
been shown to bind nonꢀhistone proteins and to recognize acetylated p53 at K382
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
following DNA damage. There have been several studies that have implicated CBP in
4
ꢀ8
the development, maintenance or progression of cancer and tumor immunity. Studies
9
have shown that inhibition of the bromodomain of CBP regulates MYC (a regulator gene
that plays a role in cell cycle progression and that is amplified in many cancers). Other
studies have demonstrated that CBP bromodomain inhibition impairs differentiation of
1
0
regulatory T cells (Tregs) and immunosuppressive function. To help understand the
physiological implications of CBP in disease states, much work has centered around
11ꢀ19
discovering novel chemical tools that inhibit the bromodomain of CBP.
These studies
have revealed that inhibition of the CBP bromodomain is a promising new therapeutic
approach in oncology and immunoꢀoncology.
O
N
N
N
HN
O
N
N
F
1
CBP IC50 = 20 nM
BRD4(1) IC50 = 13,000 nM
BRD4(1) IC50 / CBP IC50 = 650
CBP BRET IC50 = 410 nM
MYC EC50 = 910 nM
Figure 1. In vivo chemical tool 1 and potency data.
ACS Paragon Plus Environment

Page 5 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
We recently reported an in vivo chemical tool (1, GNEꢀ272, Figure 1) as an inhibitor of
the bromodomain of CBP that was potent and 650ꢀfold selective over the bromodomain 1
of the bromodomainꢀcontaining protein 4 [BRD4(1)] (BET bromodomain surrogate). We
demonstrated through in vivo studies that 1 is able to modulate CBP MYC expression that
corresponds with antiꢀtumor activity in a MYCꢀdependent acute myeloid leukemia (AML)
model. While 1 is a promising in vivo tool to investigate CBP biology, we wished to
extend our efforts to develop a more drugꢀlike compound with improved cell potency and
increased bromodomain selectivity, particularly over the bromodomain and extra terminal
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
0
(
BET) family. Here we describe the design of 19 (GNEꢀ781; 3ꢀ[7ꢀ(difluoromethyl)ꢀ6ꢀ
(1ꢀmethylpyrazolꢀ4ꢀyl)ꢀ3,4ꢀdihydroꢀ2Hꢀquinolinꢀ1ꢀyl]ꢀNꢀmethylꢀ1ꢀtetrahydropyranꢀ4ꢀylꢀ
,7ꢀdihydroꢀ4Hꢀpyrazolo[4,3ꢀc]pyridineꢀ5ꢀcarboxamide), a highly potent and selective
6
CBP inhibitor that is efficacious in a MOLMꢀ16 AML xenograft model. Additionally, in
vitro studies show that 19 reduces FOXP3 (forkhead box P3) transcript levels, confirming
10
earlier studies with a less optimal chemical probe and suggesting that inhibition of the
CBP bromodomain may provide a novel small molecule therapeutic approach for cancer
immunotherapy.
Results and Discussion
Coꢀcrystal structures of 1 bound to both CBP and BRD4(1) provided insight towards
understanding the selectivity preference of this compound for CBP over BRD4(1) (Figure
1
2
2
). Previously we have shown that 1 binds to CBP and BRD4(1) in different
12
conformations. While 1 binds in a lower energy conformation in CBP, this compound
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
binds at a higher energy conformation in BRD4(1) due to a residue difference in the WPF
motif: Trp81, Pro82 and Phe83 for BRD4(1), and Leu1109, Pro1110 and Phe1111 for
2
1
CBP (Figure 2). Upon binding to BRD4(1), the fluorophenylpyrazole arm of 1 is
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
pushed into a strained conformation inꢀplane with the pyrazolopiperidine core to avoid
1
2
clashing with Trp81. With this in mind, we hypothesized that selectivity could be
increased further on this scaffold by constraining the aniline nitrogen of 1 in a bicycle,
thus limiting the ability of the fluorophenylpyrazole to become coꢀplanar with the
pyrazolopiperidine core in the BRD4(1) binding site. Accordingly, we explored three
saturated ringꢀsize analogues (5, 6 or 7ꢀmembered) of 1. While indoline 2 had a slight
decrease in CBP potency and selectivity compared to parent 1, tetrahydroquinoline
(THQ) 3 maintained CBP potency and improved selectivity by nearly 2ꢀfold. Expanding
the ring further to a sevenꢀmembered ring (4) resulted in a significant loss in CBP
potency indicating the preferred conformations of the 7ꢀmembered ring are not
accommodated in the binding site of CBP. We therefore used 3 as a template from which
to further improve potency and selectivity. Since inhibition of the BET family
bromodomains produces profound phenotypes we were particularly focused on
selectivity over BRD4(1) (our surrogate for BET bromodomains).
ACS Paragon Plus Environment

Page 7 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. (A) A coꢀcrystal structure of 1 (green) in the CBP bromodomain (yellow, 1.23 Å
resolution, PDB code: 5KTX) highlighting hydrogen bond interactions between the ligand and
receptor. Water molecules are shown as small red spheres. (B) A coꢀcrystal structure of 1 (green)
and BRD4(1) (cyan, 1.14 Å resolution, PDB code: 5KU3) highlighting hydrogen bond
interactions between the ligand and receptor. Water molecules are shown as small red spheres.
Table 1. Constrained analogues of 1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
All IC values are reported as the geometric mean from at least two determinations. TRꢀFRET
5
0
b
assay with the isolated CBP or BRD4(1) bromodomain. Fold selectivity is defined by [BRD4(1)
IC (nM) / CBP IC (nM)]
5
0
50
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
We obtained a coꢀcrystal structure of 3 with the bromodomain of CBP (Figure 3). As
observed with 1, the amide of 3 binds in the acetylated lysine (KAc) binding site, with the
amide carbonyl making a canonical hydrogen bond to Asn1168 and additional waterꢀ
mediated hydrogen bond interactions to Tyr1125. The pyrazolopiperidine core is
sandwiched between the gatekeeper (Val1174) and Leu1120. The THQ ring makes van
der Waals interactions with Pro1110 and Leu1120. The Nꢀmethylpyrazole packs along
the lipophilic surface defined by the LPF shelf (Leu1109, Pro1110, Phe1111).
Examination of the 3ꢀCBP coꢀcrystal structure revealed several areas in the binding site
that could be explored to potentially improve potency. Initially, we examined structureꢀ
activity relationships (SAR) targeting Arg1173, the BC loop, and the KAc binding
pocket.
ACS Paragon Plus Environment

Page 9 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. A coꢀcrystal structure of 3 in the CBP (cyan) bromodomain (yellow, 1.6 Å resolution,
PDB code: 5W0F) displaying important contacts between the ligand and receptor. Water
molecules are shown as small red spheres.
We sought to improve the affinity of 3 by making favorable interactions with residues
in the LPF shelf and Arg1173 through the introduction of various substituents to the 7ꢀ
22
position of the THQ ring (Table 2). To explore potential hydrophobic interactions with
the methylene sideꢀchain of Arg1173 and to test the feasibility of disrupting the planarity
between the Nꢀmethyl pyrazole and the phenyl ring of the THQ, methyl (5), chloro (6)
and methoxy (7) substitutions were evaluated, resulting in each case in a 3.4–3.7ꢀfold
increase in CBP potency. The selectivity of 5 and 6 for CBP over BRD4(1) was
comparable to parent 3 while 7 had reduced (~3ꢀfold) selectivity for CBP. In order to
further increase van der Waals interactions with residues in the LPF shelf and potentially
interact with the guanidine atoms of Arg1173, fluorine atoms were added to the 7ꢀmethyl
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
of 5. Notably, difluoro 8 demonstrated an 11ꢀfold increase in CBP inhibitory potency
compared to 3 and exhibited selectivity (BRD4/CBP) substantially exceeding that of 3.
Additionally, 8 had significantly improved cell potency (CBP BRET IC50 = 37 nM).
Trifluoro analogue 9 was also quite potent, but had lower selectivity compared to 8.
While 9 was potentially an interesting compound itself to follow up, we choose 8 to
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
3
further examine SAR since it had a more favorable lipophilic ligand efficiency (cLogP
2
4
=
2.1; LLE = 6.9) compared to 9 (cLogP = 3.6; LLE = 5.4). To further understand the
increased CBP potency of 8 we obtained a coꢀcrystal structure of 8 and the CBP
bromodomain (Figure 4). The crystal structure revealed that while 8 had a similar binding
mode compared to 3, additional favorable dipolar interactions between the partial
negative charge on the fluorines of the 7ꢀdifluoromethyl and the positively charged
guanidine on Arg1173 were observed. Relative to 3, the Nꢀmethylpyrazole is rotated and
lies out of plane by ~30°, as expected due to the CF H substitution (see Supporting
2
Information for torsion scan plot). As a result, the Nꢀmethylpyrazole sits in a shallow
crevice on the surface of CBP that is formed by the sideꢀchains of Leu1109, Val1174,
Phe1177 of the LPF shelf, and the Cβ of Arg1173.
Table 2. Substitution at the 7ꢀposition of the THQ
ACS Paragon Plus Environment

Page 11 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
b
All IC values are reported as the geometric mean from at least two determinations. BRET
5
0
assay in HEK293 cells transiently transfected with CBPꢀNanoluc and Histone H3.3ꢀHalo tag
c
constructs. Fold selectivity is defined by [BRD4(1) IC (nM) / CBP IC (nM)].
5
0
50
Figure 4. (A) Coꢀcrystal structure of 8 (orange) in the CBP bromodomain (yellow, 1.43 Å
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
resolution, PDB code: 5W0I) displaying important contacts between the ligand and receptor.
Water molecules are shown as small red spheres. (B) The solvent accessible surface of the CBP
bromodomain colored to highlight hydrophobicity and charge where hydrocarbon groups are
white; atoms with a partial negative charge are red; atoms with partial positive charge are blue.
Bound ligand 8 is shown in stick representation (orange). Water molecules are shown as small
red spheres.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Further inspection of the coꢀcrystal structure of 8 and CBP suggested that we could explore N1ꢀ
substitution of the pyrazolopiperidine might increase hydrophobic interactions with residues in
the BC and ZA loops. Figure 4B displays the surface of the protein around the THF moiety of 8
and indicates that expanding the THF to a 6ꢀmembered ring could increase van der Waals
interactions with Leu1120 and Ile1122. Tetrahydropyran (THP) 10 resulted in an increase in both
biochemical and cell potency (Table 3; it should be noted that the effective lower limit of the
CBP TRꢀFRET assay is 1 nM, as the CBP concentration is 2 nM). Additionally, the stability of
1
0 in liver microsomes (LM) increased compared to 8. A coꢀcrystal structure of 10 with CBP
shows similar interactions to those described for 3 and 8 with the exception of the THP ring
Figure 5). Rather than pointing towards Arg1173 as observed with the THF moiety, the oxygen
atom of the THP pointed out towards solvent in the BC loop region and suggested that this
(
2
3
region would enable us to tune physicochemical properties. To reduce the cLogP of 10 we
investigated the addition of polar substituents to this area of the binding site. The structure
suggested that these modifications would have minimal effect on the biochemical potency and
potentially would increase LM stability. Removing the oxygen atom of the THP ring of 10
afforded cyclohexane 11 that had decreased CBP inhibitory potency. To further investigate polar
ACS Paragon Plus Environment

Page 13 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
atoms in this area, the oxygen atom of 10 was replaced with a nitrogen to make piperidines 12
and 13. While the biochemical potency of 12 only decreased slightly compared to 10, there was a
significant decrease in cell potency presumably that may indicate poor permeability of this
compound. The Nꢀmethyl piperidine 13 rescues the cell potency of 12, but leads to increased
cLogP. The decrease in LM stability observed for 13 may be attributed to the increased cLogP.
To lower cLogP and increase LM stability, we evaluated analogues 14, 15 and 16. This afforded
compounds with reduced cellular potency compared to 10, but limited increase in LM stability.
Ultimately, 10 (GNEꢀ049) was selected for further profiling as it had the best balance of LM
stability, selectivity and cellular potency (Figure 6). Furthermore, 10 demonstrated acceptable
PK in mouse, rat, dog and monkey. Determination of potency versus a selection of
bromodomains revealed that this compound is selective for CBP/P300 (full bromodomain
selectivity profile provided in Supporting Information) and, importantly, quite selective (3,820ꢀ
fold) over BRD4(1). The compound had excellent potency in the BRET cellular assay and, in an
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9
orthogonal measure of the target engagement, 10 was shown to inhibit the expression of MYC
(
MV4ꢀ11 cell line) with an EC of 14 nM. Compound 10 was further evaluated in a rat single
50
dose (30–250 mg/kg QD) toxicokinetic study (data not shown). Adverse central nervous system
CNS)ꢀrelated signs (e.g., marked hyperactivity and vocalization) were observed in several of the
(
rats at the 250 mg/kg dose level. Furthermore, at the 250 mg/kg dose level, the ratio of the
unbound drug concentration in brain to unbound drug concentration in plasma (Kp,uu) 3 hours
post dose was determined to be 0.43 indicating that compound was penetrating into the CNS, and
potentially resulting in the observed toxicity.
Table 3. Exploration of SAR in the BC loop region of CBP
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
R
N
N
F
N
F
N
O
N
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
N
CBPa
BRD4(1)a CBP BRETa
IC50 (nM) IC50 (nM) Selectivityb M / R / H
LM Clhepc
Fold
Cmpd
R
IC50 (nM)
cLogP
2.1
8
1.4
5,600
4,200
9,500
3,100
37
4,000
3,820
3,070
2,070
55 / 24 / 13
29 / 21 / 9.4
62 / 45 / 19
O
10
11
12
O
1.1
3.1
1.5
12
2.5
4.0
45
NH
N
690
<18 / <7.7 / <4.8 2.2
13
14
2.0
1.0
1.2
1.2
4,200
4,100
4,000
5,400
98
53
46
41
2,100
4,100
3,300
4500
54 / 22 / 17
39 / <6.1 / 10
25 / 23 / 6.2
31 / 18 / 5.3
2.8
2.1
1.7
0.99
b
O
N
O O
1
5
6
N S
O
1
S
O
a
All IC values are reported as the geometric mean from at least two determinations. Fold
5
0
c
selectivity is defined by [BRD4(1) IC (nM) / CBP IC (nM)]. Mouse (M), rat (R) and human
5
0
50
(
H) liver microsomeꢀpredicted hepatic clearance (mL/min/kg).
ACS Paragon Plus Environment

Page 15 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. (A) Coꢀcrystal structure of 10 (purple) in the CBP bromodomain (yellow, 1.55 Å
resolution, PDB code: 5W0L) displaying important contacts between the ligand and receptor.
Water molecules are shown as small red spheres.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
b
All IC values are reported as the geometric mean from at least two determinations. MDCK
5
0
−
6
−1 c
cell line; apicalꢀtoꢀbasolateral; units = ×10 cm s . Mouse (M), rat (R), dog (D), cynomolgus
d
monkey (C) and human (H) plasma protein binding. Liver microsomeꢀpredicted hepatic
e
f
clearance (mL/min/kg). Hepatocyteꢀpredicted hepatic clearance (mL/min/kg). Compound was
dosed i.v. (1 mg/kg) as PEG400/H O (35/65) solution and p.o. (5 mg/kg) as an aqueous
2
g
suspension with 0.5% methylcellulose and 0.2% Tween 80. Compound was dosed i.v. (1 mg/kg)
as EtOH/PEG400/H O (20/15/65) solution and p.o. (5 mg/kg) as an aqueous suspension with
2
0
.5% methylcellulose and 0.2% Tween 80.
Figure 6. Summary of potency, selectivity and DMPK properties of 10.
In pursuit of a more drugꢀlike CBP inhibitor, we sought a compound with a similar profile to 10,
but with limited CNS penetration. To do this we targeted a molecule with increased total polar
ACS Paragon Plus Environment

Page 17 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
surface area (tPSA) (10 tPSA = 68) and containing an additional Hꢀbond donor. Previously, as
part of our screening efforts, we had run a highꢀconcentration fragment screen attempting to find
alternative and novel Asn1168 binding chemotypes (not shown) and identified a highly ligand
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
5
efficient fragment, 17 (CBP IC = 18 ꢀM, LE = 0.41), that displayed >6ꢀfold selectivity for
5
0
CBP over BRD4 (Figure 7A). A coꢀcrystal structure of this compound bound to the CBP
bromodomain (Figure 7B) revealed the methyl urea moiety to be occupying the KAc binding
pocket. The NH of the urea makes a hydrogen bond with the backbone of Pro1110, and the
methyl group of the urea making van der Waals interactions with Phe1111. Interestingly, the
phenyl ring of Phe1111 is rotated slightly such that the Cδ1 and Cε1 atoms are 0.8 Å and 0.9 Å
from their positions when bound to 10 (Figure 7C). From the structure of 10 overlaid with 17, it
appeared that the key Asnꢀbinding acetamide of 10 could be replaced with a urea moiety. We
therefore explored a small set of substituted urea analogues of 10 (Table 4). Primary urea 18
matched the biochemical potency of 10 and displayed a remarkable 9,600ꢀfold selectivity for
CBP over BRD4(1). Methyl urea 19, the same group present in fragment hit 17, afforded
subnanomolar biochemical potency, 5,425ꢀfold selectivity for CBP over BRD4(1) and <10 nM
cellular potency. To further probe the KAc pocket we made ethyl urea 20 and N,Nꢀdimethyl urea
2
1, but these resulted in 33ꢀfold and 330ꢀfold losses in potency, respectively, compared to 19.
The significant loss in potency observed with 21 was expected since it was assumed that the NH
of the urea was participating in a Hꢀbond with the backbone of Pro1110 similar to that observed
with the 17ꢀCBP crystal structure.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 7. (A) CBPꢀselective fragment screening hit 17. (B) Coꢀcrystal structure of 17
(aqua) and the CBP bromodomain (yellow, 1.7 Å resolution; PDB code: 5W0Q).
Residues engaging in important contacts between the ligand and receptor are shown in
stick representation. Water molecules are shown as small red spheres. (C) Compound 17
(aqua) overlaid with 10 (purple) in the CBP binding site (yellow).
Table 4. Exploring the KAc binding site with urea substitution
ACS Paragon Plus Environment

Page 19 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
b
All IC values are reported as the geometric mean from at least two determinations. Fold
5
0
selectivity is defined by [BRD4(1) IC (nM) / CBP IC (nM)]
50
50
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To further understand the selectivity of 19 we obtained coꢀcrystal structures of this compound
with CBP and BRD4(1) (Figure 8). In the 19ꢀCBP coꢀcrystal structure, similar interactions with
the binding site are observed as seen in the 10ꢀCBP coꢀcrystal crystal structure, with an
additional interaction provided by the methyl urea of 19 (Figure 8A). As observed in the 17ꢀCBP
coꢀcrystal structure, the urea NH of 19 makes a hydrogen bond with the backbone of Pro1110
and the methyl group of the urea makes van der Waals interactions with Phe1111.
Unsurprisingly, the Cδ1 and Cε1 atoms of Phe1111 shifted by 0.8 Å and 1.0 Å, respectively,
relative to the 10 structure, consistent with the shifts observed in the 17ꢀCBP coꢀcrystal structure.
In BRD4, Trp81 from the WPF shelf served as a lipophilic “protrusion” compared to CBP’s LPF
“wall” (Figure 8B). As a result, the THQ of 19 is rotated in BRD4(1) such that the phenyl ring
makes van der Waals contacts with Leu92 from the BC loop while the Nꢀmethylpyrazole is
completely solventꢀexposed. The ligandꢀreceptor complementarity of 19 in BRD4 is less optimal
relative to 19 bound to CBP, providing an explanation for the selectivity.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 8. (A) Coꢀcrystal structure of 19 (pink) and the CBP bromodomain (yellow, 1.41
Å resolution; PDB code: 5W0E). Residues engaging in important contacts between the
ligand and receptor are shown in stick representation. Water molecules are shown as
small red spheres. (B) Coꢀcrystal structure of 19 (pink) and the BRD4(1) bromodomain
(cyan, 1.71 Å resolution; PDB code: 5VZS). Residues engaging in important contacts
between the ligand and receptor are shown in stick representation. Water molecules are
shown as small red spheres.
Based on the cell potency and selectivity of 19, we further profiled this compound (Figure 9).
Compound 19 showed moderate to low clearance in vivo in all species evaluated, with
acceptable oral bioavailability. Examination of a subset of bromodomains revealed that this
compound is exquisitely selective for CBP/P300 (full bromodomain selectivity profile provided
in Supporting Information) and is remarkably selective for CBP (5,425ꢀfold) and P300 (4,250ꢀ
fold). When tested (full details in Supporting Information) in an Invitrogen kinase panel (1
ꢀM,
2
20 kinases) 19 did not inhibit any target at >10%, and in a Cerep offꢀtarget screening panel (10
ꢀM, 43 receptors), 19 did not inhibit any target at >39%. In addition, 19 did not inhibit (>10 ꢀM,
ACS Paragon Plus Environment

Page 21 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
top concentration) several cytochrome P450s (3A4, 1A2, 2C9, 2C19, 2D6). The compound had
excellent and comparable potency (<10 nM) between the BRET and MYC cellular assays.
Furthermore, when 19 was dosed in a rat at 250 mg/kg, the Kp,uu (3 hours post dose) was
determined to be 0.02 (compared to 0.43 for 10), indicating that there was little brain penetration
of this compound, and no adverse CNSꢀrelated signs were observed.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
N
F
H
N
N
N
F
N
O
N
Selectivity
1
9
N
IC50 (nM)a
Bromodomain
CBP
P300
CBP BRET
IC50 (nM)a
Myc
BAZ2B
BPTF
BRD4(1)
BRD4(2)
BRD8(1)
BRD9
BRPF1
CECR2
GCN5L
PCAF
TAF1(1)
TAF1(2)
>18,000
>20,000
5,100
IC50 (nM)a
IC50 (nM)a
EC50 (nM)a
0.94
1.2
6.2
6.6
12,000
>20,000
>20,000
4,600
>20,000
>20,000
>20,000
>20,000
>20,000
In vitro stability & binding
PPB (%)c
LM Clhepd
M/R/D/C/H
Hep Clhepe
M/R/D/C/H
Papp
A to B)b
(
M/R/D/C/H
1
9
94/94/77/78/84 35/18/17/23/10 <21/<10/<7.8/24/<6.2
PK Parameters
CL
t1/2
(h)
F
Vss
Cmax
( M)
AUClast
(hr* M)
Species
(mL/min/kg)
(%)
(L/kg)
f
Mouse
5.2
3.8
4.8
18
1.4
3.7
4.7
73
37
100
0.60
0.64
1.5
23
6.5
16
44
f
cLogP: 2.4
Kinetic Solubility: 95 M
Rat
Dog
4.1
8.3
1.9
g
Cynog
0.72 28
1.0
2.5
a
b
All IC values are reported as the geometric mean from at least two determinations. MDCK
5
0
−6
−1 c
cell line; apicalꢀtoꢀbasolateral; units = ×10 cm s . Mouse (M), rat (R), dog (D), cynomolgus
d
monkey (C) and human (H) plasma protein binding. Liver microsomeꢀpredicted hepatic
e
f
clearance (mL/min/kg). Hepatic clearance (mL/min/kg). Compound was dosed i.v. (1 mg/kg) as
PEG400/H O (35/65) solution and p.o. (5 mg/kg) as an aqueous suspension with 0.5%
2
g
methylcellulose and 0.2% Tween 80. Compound was dosed i.v. (1 mg/kg) as
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
EtOH/PEG400/H O (20/15/65) solution and p.o. (5 mg/kg) as an aqueous suspension with 0.5%
2
methylcellulose and 0.2% Tween 80.
Figure 9. Summary of potency, selectivity and DMPK properties of 19.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
We next determined the effect of 19 in an in vivo PK/PD experiment using a MOLMꢀ16 (adult
AML cell line) xenograft mouse model. Single doses of 19 were given at dose levels between 3
and 30 mg/kg in MOLMꢀ16 tumorꢀbearing animals, and samples were collected at time points
covering 2 to 24 hours. Tumor RNA was generated and used to assess MYC transcript by
quantitative RTꢀPCR relative to vehicleꢀtreated animals. Suppression of MYC was observed at
doses as low as 3 mg/kg at 2 and 8 h, with maximal suppression observed at 10 and 30 mg/kg at
2 hours (87% and 88% inhibition, respectively; Figure 10A). It should be noted that suppression
of MYC at the 3, 10 and 30 mg/kg doses was seen without tumor or plasma levels of 19 reaching
the BRD4(1) IC50 (5.1 ꢀM) and thus results observed are clearly not from BET inhibition.
12
Consistent with previous results, inhibition of MYCꢀtranscript was transient over the 24ꢀhour
window with an apparent rebound effect at the 3 mg/kg dose level, rising to about 150% of
vehicle control at the 16 h time point. Increasing suppression of MYCꢀtranscript level correlated
with increasing plasma and tumor drug levels (Figure 10A).
To evaluate the in vivo efficacy of 19, MOLMꢀ16 AML xenografts were established in SCID
beige mice. Upon tumor establishment, dosing of compound 19 was initiated with p.o. doses of
3
–30 mg/kg, twice daily (BID). Singleꢀagent efficacy was observed at all doses, as evidenced by
inhibition of MOLMꢀ16 tumor growth (Figure 10B). Tumor growth inhibition (%TGI) was 73%,
7
1%, and 89% at 3, 10 and 30 mg/kg, respectively. All doses of 19 were well tolerated over the
2
1ꢀday dosing window, with a maximal body weight loss of 3.7%. There was a lack of dose
ACS Paragon Plus Environment

Page 23 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
proportionality observed between the 3 mg/kg and 10 mg/kg doses for 19, resulting in
comparable exposures and suppression of MYC transcript levels, corresponding to comparable
efficacy for these two doses in the efficacy study. At a dose of 30 mg/kg we observed higher
exposures of 19 resulting in more durable suppression of MYC transcript levels, which
corresponded to an increase in antiꢀtumor activity. These data suggest that there is a doseꢀ
proportional PKꢀPDꢀefficacy relationship where more durable suppression of MYC transcript
results in better suppression of MOLMꢀ16 tumor growth. Previously we had evaluated 1 in this
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
2
same xenograft model. Compound 19, possessing greater cell potency than 1, exhibited
approximately the same %TGI at 3 mg/kg as 1 does at 25 mg/kg (65% TGI). Taken together
these results demonstrate that 19 is highly active in vivo and capable of modulating CBPꢀ
dependent MYC transcription, which corresponds with antiꢀtumor activity in a MYCꢀdependent
AML tumor model.
Figure 10. In vivo evaluation of 19 (A) Effect on 19 treatment on MYC transcript levels in
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
MOLMꢀ16 xenograft tumors with single doses of 19 (n=3/group; mean ± SEM plotted). Plasma
and tumor levels were adjusted for PPB and tissue binding, respectively, and mean plasma and
tumor unbound drug levels (± SD) are displayed. (B) Antiꢀtumor efficacy of 19 at 3, 10 and 30
mg/kg in the MOLMꢀ16 AML xenograft tumor model and effect on body weight (n=10/group;
tumor volume ꢀ mean ± SEM plotted; body weight – group fitted % body weight loss plotted).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In addition to studying the effect of 19 in a hematologic antiꢀtumor xenograft model, we also
explored the ability of this compound to impair Treg differentiation and immunosuppressive
26
function. CBP can interact with multiple transcription factors. One of them is FOXP3, the
master transcription factor defining Tregs. FOXP3 is essential for immune homeostasis,
preventing autoimmunity, limiting chronic inflammation, and hampering effective immunity
27ꢀ30
against tumors.
Inhibition of the histone acetyltransferase domain of CBP has been shown to
decrease the expression of FOXP3, leading to reduced tumor growth in mouse models of
8
,31
cancer.
More recently, inhibition of the CBP bromodomain was also reported to impair
FOXP3 expression and Treg function, suggesting that bromodomain inhibition could impact
1
0
optimal acetylation of chromatin associated transcription factors, including FOXP3 itself. To
determine whether 19 would affect in vitro generation of inducible Tregs (iT ) and FOXP3
regs
expression, human naïve CD4+T cells were differentiated towards Tregs using antiꢀCD3, antiꢀ
CD28, rILꢀ2 and rTGFβ in the presence of increasing concentrations of 19 (0.001 to 2 ꢀM),
using DMSO as a control. At day 4 postꢀdifferentiation, cells were harvested, mRNA was
extracted, and FOXP3 gene expression was assessed by quantitative reverse transcriptase
polymerase chain reaction (qꢀRTꢀPCR). Figure 11A shows that 19 decreases FOXP3 transcripts
in a doseꢀdependent manner (IC = 17 nM) consistent with the cellular potency in our MYC RNA
5
0
ACS Paragon Plus Environment

Page 25 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
assay. Moreover the reduction in FOXP3 mRNA corresponded to a reduction in FOXP3 protein
+
+
as determined in fixed and permeabilized CD4 CD25 cells by fluorescenceꢀactivated cell
sorting. Compound 19 did not affect cell viability (Figure 11B), and a doseꢀdependent decrease
in FOXP3 staining was observed (Figure 11C), corroborating the gene expression data. Taken
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
0
together with our previous work, these data suggest that inhibition of the CBP bromodomain
may provide a novel small molecule approach for cancer immunotherapy.
Figure 11. Compound 19 decreased the generation of iTregs in vitro without affecting cell
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
viability. Human iT cells were differentiated in vitro from naïve CD4+T cells derived from
reg
Peripheral blood mononuclear cells (PBMCs) of a healthy donor in the presence of 19 or DMSO
as control. On day 4, cells were harvested. (A) FOXP3 expression was assessed by qꢀRTꢀPCR
and normalized with B2M. (B) Cell viability was assessed by flow cytometry using a fixable
viability dye. (C) Treg cells were also fixed, permeabilized and intracellular FOXP3 protein
assessed in CD4+CD25+ by flow cytometry. Graph shows % FOXP3+iTregs cells out of total
CD4+T cells. Data are representative of three independent experiments.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Chemistry
Intermediates 27–35 were prepared according to Scheme 1. Conjugate addition of ethyl 3ꢀ
aminopropanoate (22) to acrylonitrile followed by Boc protection yielded 23. Compound 23 was
treated with sodium hydride and cyclized to give 24. The pyrazolopiperidine core (25) was
formed by treatment of 24 with hydrazine, and subsequent conversion of the amino function of
2
5 to the bromide under Sandmeyer conditions afforded 26. Alkylation of the pyrazole 26 with
the appropriate alkyl methanesulfonate yielded 27–31. Deprotection of the Boc group with TFA
followed by acylation with acetic anhydride afforded 32–34. Intermediate sulfone 35 was formed
by oxidation of thioether 31 with Oxone®.
a
Scheme 1. Synthesis of Intermediates 27–35
ACS Paragon Plus Environment

Page 27 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
O
O
H
O
NC
CN
N
N
O
a
b
c
d
N
H2N
HCl
O
O
N
O
O
N
NH2
O
O
O
R
22
23
24
25
O
N
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
R
N
N
H
O
27, 32 R =
28, 33 R =
30 R =
31 R =
35 R =
N
N
N
f
N
( )
e
O
N
O
N
N
S
Br
Br
Br
O
O
O
O
26
27 31
32 34
g
O
O
O
S
2
9, 34 R =
a
Reagents and conditions: (a) i. NaOH, acrylonitrile, MeOH, 70 °C ii. diꢀtertꢀbutyl dicarbonate, 0
C, 99%; (b) NaH, toluene, 100 °C, 100%; (c) hydrazine monohydrate, EtOH, 60 °C, 70%; (d)
isopentyl nitrite, MeCN, CuBr , 60 °C, 34%; (e) RꢀOMs, Cs CO , DMF, 80 °C, 28–71%; (f) i.
°
2
2
3
TFA, DCM, rt, ii. Ac O, TEA, DMF, rt, 77–87%; (g) Oxone, THF, H O, rt, 98%.
2
2
The synthesis of 2 and 3 is outlined in Scheme 2. Suzuki coupling of 1ꢀmethylꢀ4ꢀ(4,4,5,5ꢀ
tetramethylꢀ1,3,2ꢀdioxaborolanꢀ2ꢀyl)ꢀ1Hꢀpyrazole with commerciallyꢀavailable 36 or 37
afforded 38 and 39, respectively. Buchwald coupling of racemate 32 and 38 followed by chiral
chromatography yielded 2. (We have observed from evaluating analogues from the synthesis of
enantiopure starting material that the (R)ꢀisomer is always less potent than the (S)ꢀisomer.
Compound 2 was assumed to be the (S)ꢀisomer as this was more potent than the (R)ꢀisomer.)
Buchwald coupling of optically pure 33 with 39 afforded 3.
a
Scheme 2. Synthesis of Compounds 2 and 3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
H
H
N
N
a
n
Br
n
N
N
3
6, 37
38, 39
O
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
36, 38, 2 n = 1
37, 39, 3 n = 2
N
N
b or c
N
N
O
n
N
N
2, 3
a
Reagents and conditions: (a) 1ꢀmethylꢀ4ꢀ(4,4,5,5ꢀtetramethylꢀ1,3,2ꢀdioxaborolanꢀ2ꢀyl)ꢀ1Hꢀ
pyrazole, Pd(dppf)Cl , Na CO or K CO , dioxane, H O, 120 °C, 47–86%; (b) 32, RuPhos Pd
2
2
3
2
3
2
G1, RuPhos, tꢀBuONa, dioxane, 120 °C, Supercritical Fluid Chromatography (SFC) chiral
resolution, 16%; (c) 33, RuPhos Pd G1, RuPhos, tꢀBuONa, dioxane,120 °C, 34%
Buchwald coupling of commerciallyꢀavailable 40 or 41 with intermediate 33 yielded 42 and 43,
respectively (Scheme 3). Bromination of 42 and 43 with Nꢀbromosuccinimide (NBS) gave
bromides 44 and 45, respectively. Subsequent Suzuki coupling with 1ꢀmethylꢀ4ꢀ(4,4,5,5ꢀ
tetramethylꢀ1,3,2ꢀdioxaborolanꢀ2ꢀyl)pyrazole provided analogues 4 and 7, respectively.
a
Scheme 3. Synthesis of Compounds 4 and 7
ACS Paragon Plus Environment

Page 29 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
Reagents and conditions: (a) 33, RuPhos Pd G1, RuPhos, tꢀBuONa, dioxane, 120 °C, 26–39%;
(b) NBS, DCM, 0 °C to rt 90–97%; (c) 1ꢀmethylꢀ4ꢀ(4,4,5,5ꢀtetramethylꢀ1,3,2ꢀdioxaborolanꢀ2ꢀ
yl)ꢀ1Hꢀpyrazole, Pd(dppf)Cl , K CO , dioxane, H O, 110 °C, 23–24%.
2
2
3
2
Scheme 4 shows the synthesis of compounds 5, 6, 8 and 9. Appropriately substituted THQs (46–
4
8) were treated with NBS yielding the brominated intermediates 49–51. Suzuki coupling of 49–
1 with 1ꢀmethylꢀ4ꢀ(4,4,5,5ꢀtetramethylꢀ1,3,2ꢀdioxaborolanꢀ2ꢀyl)ꢀ1Hꢀpyrazole provided 52–54.
5
Buchwald coupling of intermediate 33 with 52 or 54 provided compounds 6 and 9, respectively.
Buchwald coupling of racemic intermediate 32 with 53 followed by chiral chromatography
yielded 8. Chloride 6 was converted directly to methyl analogue 5 via palladium coupling with
potassium methyltrifluoroborate.
a
Scheme 4. Synthesis of Compounds 5, 6, 8 and 9
ACS Paragon Plus Environment