Design of a Chemical Probe for the Bromodomain and Plant Homeodomain Finger-Containing (BRPF) Family of Proteins

Article abstract of DOI:10.1021/acs.jmedchem.7b00611

The bromodomain and plant homeodomain finger-containing (BRPF) family are scaffolding proteins important for the recruitment of histone acetyltransferases of the MYST family to chromatin. Here, we describe NI-57 (16) as new pan-BRPF chemical probe of the bromodomain (BRD) of the BRPFs. Inhibitor 16 preferentially bound the BRD of BRPF1 and BRPF2 over BRPF3, whereas binding to BRD9 was weaker. Compound 16 has excellent selectivity over nonclass IV BRD proteins. Target engagement of BRPF1B and BRPF2 with 16 was demonstrated in nanoBRET and FRAP assays. The binding of 16 to BRPF1B was rationalized through an X-ray cocrystal structure determination, which showed a flipped binding orientation when compared to previous structures. We report studies that show 16 has functional activity in cellular assays by modulation of the phenotype at low micromolar concentrations in both cancer and inflammatory models. Pharmacokinetic data for 16 was generated in mouse with single dose administration showing favorable oral bioavailability.

Full text of DOI:10.1021/acs.jmedchem.7b00611

Subscriber access provided by UNIV OF NEWCASTLE
Article
Design of a chemical probe for the Bromodomain and Plant
Homeodomain Finger-containing (BRPF) family of proteins
Niall Igoe, Elliott D Bayle, Cynthia Tallant, Oleg Fedorov, Julia C Meier, Pavel Savitsky,
Catherine Rogers, Yannick Morias, Sarah Scholze, Helen Boyd, Danen Cunoosamy, David M.
Andrews, Anne Cheasty, Paul E Brennan, Susanne Müller, Stefan Knapp, and Paul V. Fish
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b00611 • Publication Date (Web): 17 Jul 2017
Downloaded from http://pubs.acs.org on July 17, 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 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Design of a chemical probe for the Bromodomain and Plant Homeodomain Finger-containing
(BRPF) family of proteins
a
a
b
b
b
b
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
Niall Igoe, Elliott D. Bayle, Cynthia Tallant, Oleg Fedorov, Julia C. Meier, Pavel Savitsky,
b
c
c
c
c
Catherine Rogers, Yannick Morias, Sarah Scholze, Helen Boyd, Danen Cunoosamy, David M.
d
e
b
b,f
b,f,g
a,
Andrews, Anne Cheasty, Paul E. Brennan, Susanne Müller, Stefan Knapp,
and Paul V. Fish *
a
UCL School of Pharmacy, University College London, 29/39 Brunswick Square, London, WC1N
1AX, UK
b
c
d
e
f
Nuffield Department of Clinical Medicine, Structural Genomics Consortium, University of Oxford, Old
Road Campus Research Building, Roosevelt Drive, Oxford, OX3 7DQ, UK
AstraZeneca, Innovative Medicines & Early Development, Pepparedsleden 1, 431 83 Mölndal,
Sweden
AstraZeneca Discovery Sciences, Darwin Building, Cambridge Science Park, Cambridge CB4 0FZ,
UK
CRT Discovery Laboratories, Jonas Webb Building, Babraham Research Campus, Cambridge CB22
3AT, UK
Buchmann Institute for Molecular Life Sciences, MaxꢀvonꢀLaueꢀStrasse 15, Dꢀ60438 Frankfurt am
Main, Germany
g
Institute for Pharmaceutical Chemistry, Johann Wolfgang GoetheꢀUniversity, MaxꢀvonꢀLaueꢀStrasse
9, Dꢀ60438 Frankfurt am Main, Germany
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 51
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
ABSTRACT
The bromodomain and PHD fingerꢀcontaining (BRPF) family are scaffolding proteins important for the
recruitment of histone acetyltransferases of the MYST family to chromatin. Here we describe NI-57
(
16) as new panꢀBRPF chemical probe of the bromodomain (BRD) of the BRPFs. Inhibitor 16
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
preferentially bound the BRD of BRPF1 and BRD1 over BRPF3, whereas binding to BRD9 was
weaker. 16 has excellent selectivity over nonꢀclass IV BRD proteins. Target engagement of BRPF1B
and BRD1 with 16 was demonstrated in nanoBRET and FRAP assays. The binding of 16 to BRPF1B
was rationalized through an Xꢀray coꢀcrystal structure determination which showed a flipped binding
orientation when compared to previous structures. We report studies that show 16 has functional
activity in cellular assays by modulation of the phenotype at low ꢀmolar concentrations in both cancer
and inflammatory models. Pharmacokinetic data for 16 was generated in mouse with single dose
administration showing favourable oral bioavailability.
KEYWORDS
BRPF1, BRPF2, BRD1, BRPF3, BRD7, BRD9, ATAD2, BET, bromodomain, chemical probe, NIꢀ57
ACS Paragon Plus Environment

Page 3 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Lysine acetylation is a key modification which, in combination with other posttranslational
modifications, constitutes the epigenetic code, a complex language that is interpreted by protein
interaction modules, so called reader domains that specifically interact with sequences harbouring
these modifications. Aberrant acetylation is a hallmark of many diseases and acetyl lysine reader
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
domains have therefore emerged as potential targets for treatment.
Lysine acetylation is specifically recognized by bromodomains (BRD), a family of 61 diverse protein
domains that all share a conserved helical fold comprising a leftꢀhanded bundle of four α helices (αZ,
3
αA, αB, αC), linked by loop regions of variable length (ZA and BC loops). Despite this structural
conservation, the sequence diversity within the BRD family is high facilitating the development of
specific inhibitors. Furthermore, the deep and mainly hydrophobic acetylꢀlysine binding sites in most
4
BRDs have good predicted druggability and this property has been validated experimentally by the
2, 5
recent development of a large diversity of BRD inhibitors.
The first potent BRD inhibitors have been developed against the BET family (BRD2, BRD3, BRD4,
6
ꢀ10
BRDT).
The disease modifying properties of these inhibitors and their general availability resulted
11
in a number of clinical studies, mainly in oncology. The success of preꢀclinical BET inhibitors
spurred the development of inhibitors outside this family which now covers most of the main BRD
5
12,13
14ꢀ16
17,18
19,20
21
families including BRDs of BAZ2A/B,
CREBBP/P300,
SMARCA2/4,
BRD9/7,
PCAF,
22
23
ATAD2 as well as promiscuous inhibitors called bromosporines. In addition, inhibitors for members
of the BRDs of the Bromodomain and Plant Homeodomain (PHD) Fingerꢀcontaining (BRPF) family of
24ꢀ30
proteins (BRPF1, BRPF2 (also termed BRD1) and BRPF3) have been identified.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 51
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 1. Cartoon representation of BRPFs as part of the tetrameric MYST scaffolding complex along
with ING5 and hEAF6. The BRPFs contain multiple chromatin reader domains.
BRPF family members are scaffolding proteins that harbour a diversity of protein interaction domains
2+
including two Nꢀterminal PHD domains linked by a Zn knuckle (PZP [PHD–Zn knuckle–PHD]
domain) (targeting unmodified sequences in histone H3), a BRD as well as a Cꢀterminal PWWP
31
domain binding to triꢀmethylated K36 in histone H3. The BRD in BRPF has been described to
32
interact with a number of histone acetylation sites such as H2AK5ac, H4K12ac and H3K14ac. BRPF
proteins are central scaffolds that recruit histone acetylꢀtransferases (HATs) of the MYST family
assembling a chromatin complex that also contains the ING tumour suppressor and hEAF6 (an EPCꢀ
33
associated protein) (Figure 1). The scaffolding function of BRPF proteins is required for substrate
specificity of MYST HATs. Alternative scaffolding proteins of the JADE family that share the BRPF Nꢀ
terminus but lack the Cꢀterminal bromo and PWWP domains direct MYST complexes towards the H4
34
tail whereas the BRPF containing complexes are targeted to histone H3. MYST/BRPF2 complexes
play important roles in the regulation of erythropoiesis. Deletion of the BRPF2 gene in mice for
3
5
instance results in severe anaemia due to impaired foetal liver erythropoiesis. We recently
demonstrated that the BRD in BRPF is required for differentiation of monocytes into bone resorbing
osteoclasts suggesting important function of the acetylꢀlysine reader function of BRPF/MYST
36
regulating bone turnover. Therapeutic targeting of BRPF1 by small molecules may also be valuable
3
0
in the myelomonocytic morphologic subtype of acute myeloid leukemia (AML).
ACS Paragon Plus Environment

Page 5 of 51
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
OF-1
PFI-4
GSK6853
pan BRPFs
BRPF1
BRPF1
IACS-9571
‘compound 34’
TRIM24ꢀBRPF1/2
dual TRIM24ꢀBRPF1
BAY-299
BRPF2 and TAF1
Chart 1. Chemical probes and inhibitors for the bromodomain of the BRPFs, TRIM24ꢀBRPFs and
BRPF2ꢀTAF1. The 1,3ꢀdimethyl benzimidazolone, which acts as a acetylꢀlysine mimic, is shown within
the blue box.
Recent studies have identified the 1,3ꢀdimethyl benzimidazolone template as binding to the BRD of
BRPF1 and this core has been optimised to give three chemical probes as either the panꢀBRPF BRD
24
25
26
inhibitor OFꢀ1, or as the selective BRPF1 BRD inhibitors PFIꢀ4 and GSK6853 (Chart 1). The 1,3ꢀ
dimethyl benzimidazolone template has also been used to identify TRIM24ꢀBRPF inhibitors such as
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 51
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
2
7
28
IACSꢀ9571 and others. BAYꢀ299 is a potent inhibitor of BRPF2 and the second BRD of TAF1 and
29
has been published with an inactive partner as a negative control. Although these inhibitors all share
some common structural features with respect to their acetylꢀlysine mimic, their pharmacology profiles
and biopharmaceutical properties are quite different.
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
The use of small molecule inhibitors as chemical probes to investigate the relationship between the
target and disease is significantly enhanced when two structurally orthogonal chemical probes are
employed. We reported the discovery of a BRD fragment hit which was optimized by structure guided
3
0
design to NI-42 (15) as an inhibitor of the BRD of the BRPF proteins (Table 1). Compound 15 has
an Nꢀmethylquinolinꢀ2ꢀone as the acylꢀlysine mimic which offers a different secondary pharmacology
fingerprint profile to the benzimidazolones. Quinolinꢀ2ꢀone 15 is a biased, potent inhibitor of the BRD
of the BRPFs with good selectivity over nonꢀclass IV BRD proteins.
Table 1. Selected properties of 15
NI-42 (15)
mw 353; clogP 2.5; tPSA 90
BRPF1 K
BRPF2 K
D
D
(ITC)
(ITC)
BRPF3 K (ITC)
40 nM
210 nM
940 nM
1130 nM
D
BRD9
D
K (ITC)
Selectivity vs nonꢀClass IV BRDs:
o
∆
T
m
shift at 10 ꢀM
< 1 C
Our aims were to further explore the quinolinꢀ2ꢀone series and to design new small molecule
inhibitors of the BRD of the BRPFs. Here we report detailed SAR studies that led to the discovery of a
second generation chemical probe for the BRPFs with improved potency and selectivity over 15 which
is suitable for cellular and in vivo studies.
ACS Paragon Plus Environment

Page 7 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
RESULTS AND DISCUSSION
Chemistry. The general methods for the synthesis of the quinolinꢀ2(1H)ꢀone sulfonamides 1-42 are
outlined in Schemes 1 and 2. The 6ꢀamino quinolinꢀ2(1H)ꢀones 43-55 were reacted with the
appropriate sulfonyl chlorides 56 in the presence of base to give the corresponding sulfonamides 1-9,
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
3
0
14-37 as previously described (Scheme 1). The amines 43ꢀ55 were generally prepared by published
methods or the application of know procedures. The sulfonyl chlorides 56 were purchased from
commercial sources or prepared by standard methods. The Nꢀalkyl sulfonamides 38-42 were
prepared from the corresponding NꢀH sulfonamides by deprotonation with NaH and reaction with the
appropriate alkyl iodide 57.
Scheme 1: General synthesis of sulfonamides 1ꢀ9, 14ꢀ42
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 51
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
Amines 43ꢀ55, with H unless otherwise stated:
1
4
4
4
4
4
4
4
5
5
52:
53:
54:
55:
3:
4:
5:
6:
7:
8:
9:
0:
1:
R = Me
1
3
R = R = Me
1
3
R = Me; R = Et
1
3
R = Me; R = iPr
1
4
R = Me; R = Me
1
5
R = Me; R = F
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
1
7
R = Me; R = OMe
1
7
R = Me; R = OEt
1
7
R = Me; R = OiPr
1
7
R = Me; R = F
1 3
7
R = Me; R = Me; R = OMe
1
3
8
R = Me; R = Me; R = F
1
3
R = R = Et
ACS Paragon Plus Environment

Page 9 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Reversed sulfonamides 10ꢀ13 were prepared by an analogous reaction of quinolinꢀ2ꢀoneꢀ6ꢀsulfonyl
chloride 59 with the appropriate anilines 60 (Scheme 2).
Scheme 2: Synthesis of 1ꢀmethylꢀ2ꢀoxoꢀ1,2ꢀdihydroquinolineꢀ6ꢀsulfonyl chloride (59) and 10-13.
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
Reagents and Conditions: (a) HSO
3
Cl, 95 C, 77 %; (b) ArNH
2
(60)(0.9 eq), pyridine (1.0 eq),
CH
CH
2
Cl
2
2
, rt, o/n, 39ꢀ79 %; (c) For 11: 2ꢀAminobenzonitrile (0.9 eq), pyridine (1.0 eq), DMAP (0.1 eq),
, reflux, o/n, 15 %;
2
Cl
Screening Strategy. Compounds were screened for BRPF affinity by two complementary methods to
develop SARs. Compounds were initially tested for BRPFs BRD binding (∆T ) with a differential
m
scanning fluorimetry (DSF) assay and then screened for inhibition (IC50) of BRPF BRD activity as
measured by the BROMOscan biochemical assay (Tables 2ꢀ4). BRD9 and BRD4(BD1) were routinely
used as BRD counter screens to establish evidence of selectivity in both the DSF and BROMOscan
assays. In general, there was a good correlation between the DSF and BROMOscan data for these
compounds with very similar trends in ranked order of affinities. Preferred compounds were then
evaluated for selectivity against a panel of 47 BRDs by DSF and inhibition of BRPF BRD activity was
confirmed by independent evaluation by an AlphaScreen biochemical assay (IC50). Finally, equilibrium
dissociation constants (K ) were determined by isothermal titration calorimetry (ITC).
D
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 51
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
Structure Activity Relationships. Earlier work. Previous SARs had identified phenyl sulfonamide 6
as having reasonable binding affinity for all three BRPFs with a preference for BRPF1 (IC50 65 nM).
These preliminary SARs were focussed on exploring three positions on this lead. The optimization of
substituents on the phenyl ring of 6 identified a modest preference for a 4ꢀCN group (7 << 6 ≈ 8 ≈ 9)
and then modification of the groups at the N1 and C3 positions of the quinolinꢀ2ꢀone core showed that
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
30
a 3ꢀMe gave a significant increase in BRPF affinity leading to the discovery of 15. Hence, there was
plenty of scope to further investigate this template with the aim of improving binding affinity for the
BRPFs whilst reducing affinity for other BRDs, especially BRD9.
Design principles. It was also important to retain the beneficial physicochemical properties of 6 (mw
314; cLogP 2.0; tPSA 66; LE 0.46; LLE 5.2) by moderating the size and lipophilicity of new target
compounds so as to be compatible with drugꢀlike space. Additional guiding design principles were to
avoid any reactive or potentially toxic groups, to have favourable biopharmaceutical properties, and to
be chemically enabled. Modulation of these properties was achieved by the application of the lighter
halogens (F), small alkyls (Me, Et, iPr), alkoxy (OMe, OEt, OiPr) and nitriles (CN) as substituents.
This new phase of SAR investigation was directed at: (1) broader range of substitution of the
sulfonamide, including reversing the connectivity (Table 2); (2) introduction of groups at each of the
available positions on the quinolinꢀ2ꢀone core (Table 3); and (3) substitution of the NH of the 6ꢀ
sulfonamide (Table 4). In addition to optimizing for BRPF activity, we also aimed to identify specific
points of SAR that would remove BRPF activity to generate a matched inactive control, and to provide
selective hits for other BRDs of the class IV proteins.
Simple sulfonamides. Simple alkyl sulfonamides 1ꢀ5 proved to be inferior to phenyl 6 with BRPF1
activity tracking with increased lipophilicity of the alkyl substituent. Reversing the sulfonamide
connection offered no advantage with a small set of benzonitriles 11ꢀ13 when compared to 7ꢀ9,
although the 2ꢀCN reversed sulfonamide 11 seemed to show a preference for BRD9 over BRPF1 in
the DSF assay albeit with relatively modest ∆T values (Table 2).
m
Sulfonamide substitution. Further exploration of the aryl sulfonamide ring when combined with the
1,3ꢀdimethylquinolinꢀ2ꢀone core identified our most potent BRPF inhibitors prepared so far (Table 3).
Previously it had been shown that introduction of a 2ꢀOMe 14 gave a potent inhibitor of BRPF1
although with some unwanted BRD4 activity. In addition, the 4ꢀCN group was a preferred single
ACS Paragon Plus Environment

Page 11 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
0
substituent or could be combined with a second substituent at either the 2ꢀ or 3ꢀposition. Combining
these points of SAR prompted the design and synthesis of the 2ꢀOMe,4ꢀCN analogue 16.
Sulfonamide 16 showed improved BRPF1 activity in both the DSF and BROMOscan assays (BRPF1:
o
∆
T
m
12.3 C; IC50 3.1 nM; mw 383; cLogP 2.1; tPSA 99; LE 0.44; LLE 6.4) and was superior to either
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
the monoꢀsubstituted analogues 2ꢀOMe (14) or 4ꢀCN (15). This result prompted the routine
exploration of both the 4ꢀCN and 2ꢀOMe,4ꢀCN benzene sulfonamide analogues in compound design,
and 16 was selected for further evaluation including BRPF1B coꢀcrystal structure determination.
The BRPF1B coꢀcrystal structure with 16 was solved at 2.3 Å (Figure 2). Quinolinꢀ2ꢀone 16 fitted in
the lysine acetylatedꢀbinding site of BRPF1B as expected, maintaining the conserved Hꢀbond
interaction with Asn708 and the water mediated Hꢀbond with Tyr665. Additional key contacts specific
of the core scaffold included a parallel and faceꢀcentered πꢀπ stacking interaction of Phe714 side
chain with the quinolinꢀ2ꢀone moiety at approximately 3.7 Å. Moreover, the phenyl rotamer was
pointing towards the benzonitrile ring at 3.9 Å forming a CHꢀπ interaction. Embedded into the
hydrophobic and nonꢀcharged binding site, the shape complementarity of 16 to the protein interface
was high (Figure 2a). Further compound stabilization was observed by means of a water mediated
bridge interaction between the sulfonamide and the 2ꢀMeO group of the benzonitrile. Interestingly, an
30
overlay of 16 with NI-48 (66) showed a flip of 180º of the quinolinꢀ2ꢀone core with the N1ꢀMe group
pointing towards Tyr707 (Figure 2d and S2). This new scaffold conformation presented the lowest Bꢀ
factor values during compound fitting and refinement. This flip forced the benzonitrile group to move
closer to the ZAꢀloop (Ile652).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 51
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 2. Xꢀray crystal structure of 16 (green) bound to BRPF1 (grey) (5MYG.pdb). (a) Electrostatic
potential surface represented at the bromodomain binding site. (b) Closeꢀup view of the interactions of
1
6 with the BRPF1 binding site displayed in cartoon and sticks. (c) 2Fo − Fc electron density map
contoured at 1σ showing electron density around 16. The figure shows the map for the fully refined
structure. (d) Structural superposition of 16 with 66 (purple; 5T4V.pdb) in complex with BRPF1B.
O
Me
H
S N
O
NC
MeO
N
O
Me
NI-48 (66)
Quinolinꢀ2ꢀone substitution. Introduction of a limited number of single groups at specific positions on
the quinolinꢀ2ꢀone core showed some advantage over hydrogen (e.g. 4ꢀMe (21); 7ꢀOMe (24) ≥ H (9))
but failed to improve BRPF1 activity to the same level as was achieved with a 3ꢀMe substituent (15)
(Table 3). Larger alkyls at C3 such as 3ꢀEt (17, 18) or 3ꢀiPr (19, 20) significantly reduced BRPF
activity whereas groups at the 5ꢀposition (5ꢀF (22); 5ꢀOMe) were tolerated. Small alkoxy groups at C7
ACS Paragon Plus Environment

Page 13 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
showed a modest beneficial effect with 23ꢀ26 and so a 7ꢀOMe was combined with the 3ꢀMe group on
the quinolinꢀ2ꢀone core to see if these effects were additive.
A set of 6 preferred sulfonyl chlorides 56 were coupled with 6ꢀaminoꢀ7ꢀmethoxyꢀ1,3ꢀdimethylquinolinꢀ
2(1H)ꢀone (53) to give the corresponding sulfonamides 28ꢀ33 (Table 3). These sulfonyl chlorides were
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
selected based on previous beneficial SAR and extrapolation to close analogues. Compounds 28ꢀ33
o
were all found to bind to BRPF1 with strong affinity as measured by DSF (∆T
m
>8 C) and 32 was
selected as a representative example from this set for further evaluation based on its excellent
o
potency (BRPF1: ∆T
m
12.7 C; IC50 12 nM).
Previously it has been shown that increasing the size of the alkyl groups flanking the quinolinone C2ꢀ
carbonyl would prevent binding to the BRPFs (and other BRDs) by disrupting the key recognition
interaction between the carbonyl with Asn708 and the Tyr665 water network. This was most
successfully achieved with ethyl groups at both the N1 and C3 positions which produced 36 as a
matched inactive control for 15. Applying this strategy to lead 16 by the introduction of the N1ꢀEt and
3ꢀEt groups gave 37 which still retained some modest BRPF1 activity (IC50 250 nM) (Table 3) with
only an 80ꢀfold drop in BRPF1 inhibition so as to be unsuitable as a negative control. This level of
BRPF1 activity for 37 was somewhat surprising to us and difficult to explain based on our current
understanding of the binding modes of these inhibitors (cf 16 and 66). Ultimately, this will need to be
resolved by the determination of a BRPF1ꢀ37 coꢀcrystal structure.
NꢀAlkyl sulfonamides. The Nꢀalkylation of the sulfonamide was detrimental to BRPF1 activity (Table
4). It was found across several structures that the introduction of a NꢀMe on the sulfonamide group
generally reduced BRPF1 activity by 20ꢀfold (38 vs 9, 39 vs 24). Applying this approach to 16 gave Nꢀ
Me analogue 40 which also showed a 15ꢀfold reduction in BRPF1 inhibition but 40 still showed
significant BRPF1 inhibition (IC50 52 nM). Increasing the size of the Nꢀalkyl group to NꢀEt (41) and Nꢀ
iPr (42) did reduce further BRPF1 activity by 35ꢀfold and 50ꢀfold respectively showing some additional
disruption to binding but again not to a level to provide a credible negative control.
Structural insights of 16 in complex to BRPF1B led us to understand the lower affinity for these Nꢀ
alkyl analogues for BRPF1. We examined all four molecules crystallized at the asymmetric unit of the
orthorhombic crystal form. Overall, the NH of the sulfonamide group is facing towards the Glu661
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 51
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
residue of the ZA loop. Distances from the NH group to the glutamic acid lipophilic side chain or the
carboxylic group ranged between 3.6 – 4.3 Å limiting the space to accommodate bulky substituents
(Figure S3). Additionally, Glu661 adopts a closed conformation as a result of a water mediated Hꢀ
bond interaction with a S(=O) of the sulfonamide group of 16.
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
Inactive control. The SARs for a partnered inactive control would suggest that further increasing the
size of the 3ꢀalkyl to a 3ꢀiPr (cf 16 vs 18 vs 20) in combination with a N1ꢀEt (or N1ꢀiPr) would further
reduce the BRPF1 activity. A complementary approach to reduce BRPF activity was by substitution of
the NH of the 6ꢀsulfonamide (cf 16 vs 40ꢀ42). However, these SARs seemed to converge on the
2
addition of two or more larger lipophilic groups at either N1, C3 and/or SO NH positions to completely
remove BRPF1 activity; these modifications were likely to yield compounds with minimal similarity to
the active partner 16 in terms of chemical structure, molecular and physicochemical properties, and
30
so were not pursued. On balance, compound 36 (NI-198) represents the best available matched
inactive control for both 15 and 16.
Preferred compounds. Completing the evaluation of preferred leads 16 and 32 against BRD1 and
BRPF3 showed both compounds had potent activity against all three BRPFs with 16 having a slight
advantage (Table 5). However, 16 showed a clear advantage over 32 when evaluated in the BRD9
and BRD4(BD1) selectivity counter screens. Compound 16 had good selectivity over both BRD9
(160ꢀfold) and BRD4 (1200ꢀfold) in the BROMOscan assay which was also superior to 15. In contrast,
32 had poor selectivity compared to 16 with only 14ꢀfold over BRD9 and 140ꢀfold over BRD4. Nonꢀ
BET BRD inhibitors need to demonstrate excellent selectivity over BRD4 because of the dominant
phenotype of this protein in cellular models and the narrow window of selectivity of 32 over BRD4
activity could potentially confound interpretation of biological results. The low level promiscuous
nature of 32 was further confirmed by a full BRD profile of 32 by DSF with 6 nonꢀclass IV proteins
o
having binding ∆T
m
in the range 2ꢀ3 C (10 ꢀM) including several members of the BET family (Table
S1). The erosion of selectivity for 32 was a combination of weaker activity for BRPF1 and increase
activity for the other BRDs.
Hence 16 emerged as the superior compound from this set suitable for further evaluation as a
potential chemical probe for the BRD of the BRPFs.
ACS Paragon Plus Environment

Page 15 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Improved Synthesis of 16. Once 16 had been identified as a potentially superior compound, the
need for larger amounts of material to support SAR and biological studies prompted the development
30
of an improved synthesis of both 6ꢀaminoꢀ1,3ꢀdimethylquinolinꢀ2(1H)ꢀone (44) and 4ꢀcyanoꢀ2ꢀ
37
methoxybenzenesulfonyl chloride (65) (Scheme 3). A reliable synthesis of 65 initially proved to be
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
somewhat problematic as, to date, there had been no published preparations for 65 and our original
approaches to the synthesis of 65 were flawed. Only the diazotization of 4ꢀaminoꢀ3ꢀ
38
methoxybenzonitrile then treatment with aqueous SO ꢀHCl gave modest yields of 65.
2
We developed an efficient synthesis of sulfonyl chloride 65 from 4ꢀhydroxyꢀ3ꢀmethoxybenzonitrile (61)
39
utilising a thermal Newark–Kwart rearrangement (62 to 63) and then KOH hydrolysis of the
carbamothioate 63 to thiol 64, followed by a zirconium(IV) chloride promoted oxidative chlorination
40
37
reaction as the key steps. High purity 65 could be obtained by recrystallization from a mixed
solvent system of tolueneꢀhexane, however the crude material obtained directly from the reaction was
found to be sufficiently pure to form sulfonamides in high yields. Simply filtering the acetonitrile
solution of crude 65 directly from the reaction flask into a second receiving flask containing a solution
of amine 44 and DABCO in DMF produced 16 in a good yield comparable to the 2ꢀstep process with
recrystallized 65.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 51
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
Scheme 3: Synthesis of 65 and 16.
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
o
Reagents and Conditions: (a) ClC(S)NMe
2
(1.0 eq), DABCO (1.05 eq), DMF, 50 C, 4 h, 76 %; (b)
o
o
200 C, 3 h, 94 %; (c) KOH (3.0 eq), MeOH, 30 C, 12 h, 95 %; (d) ZrCl (1.0 eq), 30 % H O (3.0
4
2
2
o
eq), MeCN, 0 C, 0.5 h, 92 %; (e) 6ꢀaminoꢀ1,3ꢀdimethylquinolinꢀ2(1H)ꢀone (44) (0.95 eq), DABCO (1.0
eq), DMF, rt, 6 h, 76 %.
ACS Paragon Plus Environment

Page 17 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 2. BRPF1 SAR of the 6-sulfonamide.
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
1-5
6-9
10-13
Compound
R or X
BRPF1
a
∆
1
T
m
shift at
IC50 (nM)
o
0 ꢀM ( C)
1
2
3
4
5
Me
Et
iPr
cPr
cHex
1.4
1.8
2.2
1.9
3.9
2200
1200
1100
520
380
6
7
8
9
H
5.3
2.4
4.4
5.2
65
430
56
2ꢀCN
3ꢀCN
4ꢀCN
43
1
1
0
H
3.5
2.6
2.3
3.1
1000
100
b
1
2ꢀCN
3ꢀCN
4ꢀCN
1
1
2
3
a
BROMOscan BRD IC50 values are geometric means of two experiments
b
(n=2). Differences of <2ꢀfold should not be considered significant.
o
BRD9, ∆T
m
(10 ꢀM) = 3.9 C.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 51
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
Table 3. BRPF1 SAR of substitution on the quinolin-2-one.
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
14-35
36, 37
x
Compound
R
X
BRPF1
m
T shift at IC50 (nM)
a
∆
o
10 ꢀM ( C)
14
15
16
17
3ꢀMe
3ꢀMe
3ꢀMe
3ꢀEt
2ꢀOMe
4ꢀCN
2ꢀOMe,4ꢀCN
4ꢀCN
10.2
9.7
12.3
5.6
9.0
b
7.9
3.1
180
93
870
170
24
18
19
20
21
22
23
24
25
26
27
3ꢀEt
2ꢀOMe,4ꢀCN
4ꢀCN
2ꢀOMe,4ꢀCN
4ꢀCN
8.0
3ꢀiPr
3ꢀiPr
4ꢀMe
5ꢀF
7ꢀOMe
7ꢀOMe
7ꢀOEt
7ꢀOiPr
7ꢀF
4.3
2.7
6.3
5.6
4ꢀCN
H
4ꢀCN
4ꢀCN
4ꢀCN
4ꢀCN
29
53
55
39
3.8
28
29
3ꢀMe,7ꢀOMe
3ꢀMe,7ꢀOMe
3ꢀMe,7ꢀOMe
3ꢀMe,7ꢀOMe
3ꢀMe,7ꢀOMe
3ꢀMe,7ꢀOMe
3ꢀMe, 8ꢀF
H
9.4
11.0
8.8
9.9
12.7
9.5
17
15
12
2ꢀOMe
4ꢀCN
2ꢀMe,4ꢀCN
2ꢀOMe,4ꢀCN
2ꢀOEt,4ꢀCN
4ꢀCN
30
31
32
33
34
35
7.1
6.8
3ꢀMe, 8ꢀF
2ꢀOMe,4ꢀCN
12
3
3
6
7
4ꢀCN
2ꢀOMe,4ꢀCN
>100,000
250
a
BROMOscan BRD IC50 values are geometric means of two experiments (n = 2).
b
Differences of <2ꢀfold should not be considered significant. n = 4.
ACS Paragon Plus Environment

Page 19 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 4. BRPF1 SAR at the N-sulfonamide position.
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
38-42
N
3 or 7
Compound
R
R
X
BRPF1
shift at
0 ꢀM ( C)
a
∆
1
T
m
IC50 (nM)
o
38
39
40
41
42
Me
Me
Me
Et
H
4ꢀCN
4ꢀCN
1.8
910
1000
52
110
160
7ꢀOMe
3ꢀMe
3ꢀMe
3ꢀMe
2ꢀOMe,4ꢀCN
2ꢀOMe,4ꢀCN
2ꢀOMe,4ꢀCN
7.1
iPr
a
BROMOscan BRD IC50 values are geometric means of two experiments (n = 2).
Differences of <2ꢀfold should not be considered significant.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 51
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
Table 5:
BRD selectivity profiles of N-methylquinolin-2(1H)-ones
a
IC50 (nM)
BRPF3
Compound
BRPF1
BRPF2
BRD9
2200
BRD4(BD1)
1700
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
(BRD1)
6
7
8
9
65
430
56
43
1000
100
9.0
7.9
3.1
870
170
1400
1700
270
200
7800
150
130
48
7600
1000
>100,000
2200
910
9800
2800
37,000
10
13
14
600
4500
3700
b
15
16
19
260
140
16,000
4600
310
520
46
14,000
3300
150
380
520
190
100
80
20
21
23
24
28
30
32
35
9300
2700
24
29
53
17
15
2200
2200
590
280
190
>10,000
12
12
170
1700
510
>100,000
4400
10,000
>100,000
4600
c
3
6
>100,000
250
>100,000
>100,000
37
38
910
6700
a
BROMOscan BRD IC50 values are geometric means of two experiments (n = 2). Differences of <2ꢀfold should not be considered significant.
b
c
n = 4
BRD7, ATAD2A, ATAD2B, CREBBP, PCAF, BAZ2A, TRIM33, CECR2: BROMOscan BRD IC50 >100,000 nM. SMARCA2A: IC50 77,000 nM. See, reference
30.
ACS Paragon Plus Environment

Page 21 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Potency and selectivity of 16. Completing the evaluation of lead compound 16 against all members
of the class IV family of BRDs in the BROMOscan assay identified affinity for BRD7 but was
completely selective over the ATAD2s (Table 6; Figure S4). 16 was confirmed as a potent inhibitor of
BRPF1 by independent evaluation in an AlphaScreen biochemical assay and with a very similar
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
36
relative rank order of activities to other screening formats (Table 6).
Compound 16 was screened for BRD binding by DSF against a panel of 47 BRDs and showed
o
excellent selectivity. All activity was confined to the class IV family of BRDs with ∆Tm < 1
C
stabilization for all nonꢀclass IV BRDs except TRIM24 (2.0), CREBBP (1.1) and EP300 (1.1) (Figure 3;
Table S1). 16 was also tested for BRD selectivity in the BROMOscan panel of 46 binding assays and
showed all potent activity was confined to the class IV family (Table S2). Once again, TRIM24 was
identified as the nonꢀclass IV BRD with most affinity (IC50 1600 nM) but still with a 500ꢀfold window of
selectivity.
It is important to note that we did not observe any significant binding or inhibition of BRD4 across
multiple orthogonal screening platforms (DSF, BROMOscan, AlphaScreen, FP) (Table 6). These
29
results have been independently corroborated by a group at Bayer in a TRꢀFRET assay.
Compound 16 was then evaluated in a number of assays to establish the broader pharmacological
selectivity and biopharmaceutical properties. 16 was screened for offꢀtarget pharmacology against a
panel of 55 representative GPCRs, ion channels and enzymes (Eurofins, ExpresSProfile) and was
36
found to have minimal inhibition (<40 % @ 10 ꢀM). Any cardiovascular risk was low as 16 showed
minimal activity against a panel of 8 cardiac ion channels (<40% @ 30 ꢁM) (Table S3). 16 has
reasonable aqueous solubility and good cell permeability as measured by transit performance in the
Caco2 cell line but with some evidence of efflux (Table 6). 16 has good metabolic stability in both
MLM and HLM consistent with predicted low clearance, and weak CYP450 enzyme inhibition (Table
S4).
ITC data showed that 16 bound the BRD of BRPF1 with a K
D
of 31 ± 2 nM, BRD1 with a K
D
of 110 ±
1
3 nM and BRPF3 with a K of 410 ± 47 nM, whereas binding to BRD9 was weaker (K
D
D
1000 ± 130
3
6
nM) (Table 6). These data showed that 16 was more potent for the BRDs of BRPF1/2/3 and slightly
more selective over BRD9 than 15.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 51
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
o
Figure 3. Selectivity profile of 16 assessed against 47 BRDs by thermal shift, ∆T ( C), at 10 ꢀM as
m
measured by DSF (n = 2).
ACS Paragon Plus Environment

Page 23 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 6: Summary of physicochemical properties, in vitro bromodomain
inhibition, ITC screening and cell permeability of 16.
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
16: NI-57
Physicochemical properties
mw
ClogP
tPSA (Å )
383
2.1
99
2
Bromodomain inhibition
BRPF1/2/3
BRD7/9
ATAD2A/2B
BRD4(BD1)
TRIM24
BROMOscan IC50 (nM)
3.1; 46; 140
140; 520
>100,000; >100,000
3700
BROMOscan IC50 (nM)
BROMOscan IC50 (nM)
BROMOscan IC50 (nM)
BROMOscan IC50 (nM)
1600
a
BRPF1/2/3
BRD9
BRD4(BD1)
TRIM24
AlphaScreen IC50 (nM)
AlphaScreen IC50 (nM)
AlphaScreen IC50 (nM)
AlphaScreen IC50 (nM)
114; 619; 1010
b
4400
a
>10,000
a
>10,000
c
BRD4(BD1)
FP assay
IC50 (nM)
>33,000
Bromodomain ITC screening
BRPF1/2/3
BRD9
ITC
ITC
K (nM)
K
31; 110; 410
1000
D
D
(nM)
Solubility and cell permeability
d
Aqueous solubility
moderate (100 ꢀM min)
11/52
ꢀ
6
ꢀ1
Caco2 transit AB/BA, Papp (x10 cms )
a
See reference 36.
b
This work.
c
See Figure S5.
d
Maximum concentration used in screening assays created by serial dilution of a 100 mM stock
solution of 16 in DMSO.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 51
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
Evidence of cellular activity. Cellular ‘onꢀtarget’ activity for 16 was assessed in a BRPF1Bꢀ
Histone3.3 nanoBRET assay which showed doseꢀdependent displacement of BRPF1B (but not
BRPF1A) from Histone H3.3 with an estimated IC50 of 0.07 ± 0.0034 ꢀM. These results were verified
in a FRAP assay where 16 displaced a triple BRDꢀBRPF1BꢀGFP fusion construct, as well as full
length BRD1ꢀGFP, which was demonstrated by significant reduction in recovery times in the presence
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
of the inhibitor. Hence, the nanoBRET and FRAP experiments confirm 16 inhibit BRPF BRDs in the
nucleus.
41
Activity in phenotypic screens. BRD containing proteins are frequently dysregulated in cancer.
Previously, compound 15 has been evaluated in a panel of cancer cell lines and showed a modest
inhibition of proliferation of certain AML lines, particularly those lines exhibiting monocytic lineage
30
differentiation. Here we report additional studies that show 16 has functional activity in cellular
assays by modulation of the phenotype (or markers) at low ꢀmolar concentrations in both cancer and
inflammatory models.
Compound 16 was evaluated in two indexed panels of cancer cell lines: the NIH National Cancer
42
43
Institute NCIꢀ60 and AstaZeneca’s CLIMB panel. Screening 16 in the NCIꢀ60 panel showed
minimal effects on inhibition of proliferation over 48 h with typically ≤ 25 % inhibition at 10ꢀM in these
lines. We subsequently treated the CLIMB panel with 16 for 72 h and observed modest and selective
inhibition of proliferation of certain cell lines (Table 7). It is encouraging that, for these lines, both 15
and 16 show a similar effect on inhibition of proliferation. The different outcomes in the two panel
screens are not inconsistent as the responding cell lines in the CLIMB panel were not present in the
NCIꢀ60. These results highlight the value of wide profile screening in phenotypical assays in
conjunction with hypothesis driven screening strategies. In addition, these BRPF BRD inhibitors 15
and 16 had minimal effect on the proliferation of many cancer cell lines in the CLIMB panel (and other
30, 36
cell lines) suggesting they are not cytotoxic.
ACS Paragon Plus Environment

Page 25 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
Table 7. Examples of inhibition of proliferation of cancer cell lines (GI ) by 15 and 16.
50
GI50 (ꢀM)
Cancer cell line
Tissue
15
16
NCIꢀH1703
DMS114
HRAꢀ19
lung
lung
colon
lung
>30
10.4
14.7
15.6
16.6
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
16.6
15.6
17.1
RERFꢀLCꢀSq1
b
~ 150 cell lines
various
>30
a
AstaZeneca CLIMB panel includes 211 cancer cell lines. 72 h proliferation assay; 9 point
doseꢀresponseꢀcurve with a highest concentration of 10 or 30 ꢀM (n = 2). For an example of
b
the use of oncology cell panels in mechanistic and pathways analysis, see reference 43.
For a more complete profile of 15 in this panel, see reference 30.
Small molecule chemical probes that modulate BRD proteins are being employed to reveal the role of
epigenetics in guiding cell fate decisions during immune responses and have highlighted the potential
4
4
of these approaches for the treatment of inflammatory and immuneꢀrelated disease. Human
macrophages are characterized by their remarkable plasticity and can be skewed towards a proꢀ
inflammatory phenotype upon LPS/IFNγ stimulation (100ng/ml), resulting in the production of proꢀ
inflammatory cytokines such as TNF (Figure 4A). On the opposite end of the macrophage activation
spectrum, ILꢀ4 (200ng/ml) induces a Type II inflammation phenotype characterized by the production
45,46
of CCLꢀ22 (Figure 4A).
BRPF1 gene expression was significantly increased in ILꢀ4ꢀstimulated
human alveolar macrophages (AM) in comparison to LPS/IFNγꢀstimulated AM with a 1.7 ± 0.5 fold
increase (Figure 4B). Blocking BRPF1 in the ILꢀ4ꢀstimulated AM with 16 (10 µM) decreased the gene
expression of CCLꢀ22 by 27.7 ± 9.4% (Figure 4C).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 51
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
*
100
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
50
untreated control
0
LPS/IFN
IL-4
*
100
50
untreated control
0
LPS/IFN
IL-4
B
C
ACS Paragon Plus Environment

Page 27 of 51
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 4. Alveolar Macrophages (AM) were isolated from lung resections of nonꢀsmoker, exꢀsmoker
and smoker subjects. Isolated AM were LPS/IFNγꢀstimulated (100ng/ml) or ILꢀ4ꢀstimulated (200ng/ml)
during 24h together with DMSO (●, ●, respectively). Subsequently, TNF and CCLꢀ22 (A) or BRPF1
(B) gene expression were measured and plotted. Gene expression was relativised to untreated
macrophages. (C) ILꢀ4 stimulated alveolar macrophages were treated with DMSO or 16 (●) and CCLꢀ
22 gene expression was measured and plotted. Gene expression was relativised to DMSO treated
macrophages. (* = p <0.05, Wilcoxon comparison test)
M2ꢀactivated alveolar macrophages have been shown to play a role in the resolution of many lung
inflammatory conditions, but can also contribute to different pulmonary diseases. Eosinophilic
pneumonia and asthma have been associated with an increased CCLꢀ22 production by alveolar
4
7,48
macrophages.
The observation that BRPF1 is involved in inducing CCLꢀ22 production by alveolar
macrophages could be of importance for the treatment of these pulmonary diseases.
Taken together, these results provide evidence that 16 is functionally active in cellular assays although
we have yet to prove that these effects are the direct consequences of the inhibition of the BRD of the
BRPFs. Additional studies will be required to more fully establish the role of 16 and the BRPFs in
cancer and inflammation including demonstration of onꢀtarget effects and target confirmation in
disease relevant cell models. These studies are ongoing by this group and will be reported in due
course.
In an associated Article, we explore functional consequences of inhibitors targeting BRDs present in
BRPFs in bone maintenance. We report three complementary potent and selective inhibitors: one with
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 51
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
2
5
24
high selectivity for the BRPF1B isoform, and two structurally different panꢀBRPF BRD inhibitors.
The inhibitors impaired RANKLꢀinduced differentiation of primary murine bone marrow cells and
human primary monocytes into bone resorbing osteoclasts by repressing transcriptional programs.
The data suggest a key role of BRPF in regulating gene expression during osteoclastogenesis. These
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
studies will be reported separately.
In vivo Pharmacokinetics. PK data for 16 was generated in vivo in mouse (Table S4). Following
single intravenous administration of 16, clearance was low relative to liver blood flow and volume of
distribution was moderate resulting in an elimination halfꢀlife of 1.2 h. Single oral administration of 16
as a suspension showed favourable oral bioavailability of 29 %. The fraction absorbed was calculated
to be 35 % based on the measured blood clearance and oral bioavailability values.
ACS Paragon Plus Environment

Page 29 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CONCLUSION
We have designed and evaluated a number of new quinolinꢀ2ꢀone derivatives as inhibitors of BRPF
BRD activity. Previous SARs had identified phenyl sulfonamide 6 as having reasonable binding
affinity for all three BRPFs with a preference for BRPF1 (IC50 65 nM). This new phase of detailed SAR
studies showed that a broader range of substitution of the phenyl ring of the sulfonamide could be
accommodated with 4ꢀCN and 2ꢀOMe,4ꢀCN groups preferred. Introduction of groups at each of the
available positions on the quinolinꢀ2ꢀone core showed some advantage over hydrogen but failed to
improve BRPF1 activity to the same level as had been achieved with a 3ꢀMe substituent (15). The
combination of the preferred substituents on the sulfonamide ring with the 1,3ꢀdimethylquinolinꢀ2ꢀone
core gave 16 as our most potent BRPF1 inhibitor (IC50 3.1 nM). Additional screening of 16 showed
potent binding affinity for all three BRPFs with good selectivity over BRD9 and BRD4. The binding of
16 to BRPF1 was rationalized through an Xꢀray coꢀcrystal structure determination which showed a
flipped binding orientation of the quinolinꢀ2ꢀone in the AcꢀLys pocket when compared to previous
structures.
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
Compound 16 was then evaluated in a number of assays to establish the broader pharmacological
selectivity. 16 showed excellent selectivity in panels of BRDs with all potent activity confined to the
class IV family of BRDs and minimal offꢀtarget pharmacology against a panel of representative
GPCRs, ion channels and enzymes. ITC data confirmed that 16 preferentially bound the BRD of
BRPF1 and BRD1 over BRPF3 whereas binding to BRD9 was weaker. Based on these results, 16
was identified as a potent inhibitor of the BRD of the BRPFs with excellent selectivity over nonꢀclass
IV BRD proteins. BRPF1 target engagement in living cells with 16 was demonstrated in both
nanoBRET and FRAP assays.
Evaluation of 16 in a panel of cancer cell lines showed a modest (GI50 10ꢀ16 ꢀM) and selective
inhibition of proliferation of certain lines. In addition, 16 decreased the expression of M2ꢀassociated
genes, such as CCLꢀ22, in human alveolar macrophages. These results provide clear evidence that
16 has activity in cellular assays. However, additional studies will be required to fully establish the role
of 16 and the BRPFs in disease models.
ACS Paragon Plus Environment