Structure-guided design of a series of MCL-1 inhibitors with high affinity and selectivity

Article abstract of DOI:10.1021/jm501258m

Myeloid cell leukemia 1 (MCL-1) is a BCL-2 family protein that has been implicated in the progression and survival of multiple tumor types. Herein we report a series of MCL-1 inhibitors that emanated from a high throughput screening (HTS) hit and progressed via iterative cycles of structure-guided design. Advanced compounds from this series exhibited subnanomolar affinity for MCL-1 and excellent selectivity over other BCL-2 family proteins as well as multiple kinases and GPCRs. In a MCL-1 dependent human tumor cell line, administration of compound 30b rapidly induced caspase activation with associated loss in cell viability. The small molecules described herein thus comprise effective tools for studying MCL-1 biology.

Full text of DOI:10.1021/jm501258m

Subscriber access provided by CAL STATE UNIV POMONA
Article
Structure-guided design of a series of MCL-1
inhibitors with high affinity and selectivity
Milan Bruncko, Le Wang, George S. Sheppard, Darren C. Phillips, Stephen K. Tahir, John Xue,
Scott Erickson, Steve Fidanze, Elizabeth Fry, Lisa Hasvold, Gary J. Jenkins, Sha Jin, Russell
A. Judge, Peter J. Kovar, David Madar, Paul Nimmer, Chang Park, Andrew M. Petros, Saul H.
Rosenberg, Morey L. Smith, Xiaohong Song, Chaohong Sun, Zhi-Fu Tao, Xilu Wang, Yu Xiao,
Haichao Zhang, Chris Tse, Joel D. Leverson, Steve W. Elmore, and Andrew James Souers
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm501258m • Publication Date (Web): 13 Feb 2015
Downloaded from http://pubs.acs.org on February 18, 2015
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 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Structureꢀguided design of a series of MCLꢀ1
inhibitors with high affinity and selectivity
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
‡
‡
Milan Bruncko , Le Wang , George S. Sheppard, Darren C. Phillips, Stephen K. Tahir, John
Xue, Scott Erickson, Steve Fidanze, Elizabeth Fry, Lisa Hasvold, Gary J. Jenkins, Sha Jin,
Russell A. Judge, Peter J. Kovar, David Madar, Paul Nimmer, Chang Park, Andrew M. Petros,
Saul H. Rosenberg, Morey L. Smith, Xiaohong Song, Chaohong Sun, Zhi-Fu Tao, Xilu Wang, Yu
Xiao, Haichao Zhang, Chris Tse, Joel D. Leverson, Steven W. Elmore, and Andrew J. Souers*
AbbVie Inc., 1 North Waukegan Rd, North Chicago, IL 60064 USA
‡
These authors contributed equally.
Keywords: MCLꢀ1, apoptosis, BCLꢀ2 family proteins, navitoclax
Abstract: Myeloid cell leukemia 1 (MCLꢀ1) is a BCLꢀ2 family protein that has been implicated
in the progression and survival of multiple tumor types. Herein we report a series of MCLꢀ1
inhibitors which emanated from a high throughput screening (HTS) hit and progressed via
iterative cycles of structureꢀguided design. Advanced compounds from this series exhibited subꢀ
nanomolar affinity for MCLꢀ1, and excellent selectivity over other BCLꢀ2 family proteins as
well as multiple kinases and GPCRs. In a MCLꢀ1 dependent human tumor cell line,
administration of compound 30b rapidly induced caspase activation with associated loss in cell
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 55
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. The small molecules described herein thus comprise effective tools for studying MCLꢀ
1
biology.
Introduction: Apoptosis (programmed cell death) is a highly conserved cellular process that 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
required for embryonic development and normal tissue homeostasis. Aberrant expression or
function of the proteins responsible for apoptotic signaling contributes to numerous human
1
ꢀ3
pathologies including autoimmunity, neurodegeneration and cancer. Intrinsic apoptosis is
governed by the dynamic binding interactions between opposing proꢀ and antiꢀapoptotic BCLꢀ2
family proteins. This family of apoptosis regulators share regions of sequence homology termed
BCLꢀ2 homology (BH) domains. The multiꢀdomain proꢀapoptotic BCLꢀ2 proteins BAX and
BAK are essential mediators that, once activated, cause mitochondrial outer membrane
permeabilization, cytochrome c release, caspase activation, and ultimately, cell death. The proꢀ
apoptotic BH3ꢀonly proteins (e.g., BIM, BAD, BID, PUMA, NOXA) act upstream in response to
a variety of cellular stimuli to promote the activation of BAX and BAK. The antiꢀapoptotic BCLꢀ
2
family members, which contain up to four BH domains (BH1ꢀ4) and include BCLꢀ2, BCLꢀXL,
BCLꢀW and MCLꢀ1, can bind to and sequester their proꢀapoptotic counterparts, thus preventing
BAX/BAK activation and promoting cell survival. By overexpressing one or more of these proꢀ
survival proteins, cancer cells can evade elimination by normal physiological processes and thus
3
,4
gain a survival advantage.
Amongst the proꢀsurvival BCLꢀ2 family members, MCLꢀ1 has the distinct trait of being
5
essential for embryonic development as well as the survival of all hematopoietic lineages and
6
ꢀ8
progenitor populations. MCLꢀ1 is also one of the most highly amplified genes present in
9
human cancers and its role in tumor initiation and maintenance has been demonstrated in
2
ACS Paragon Plus Environment

Page 3 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10,11
various genetic mouse models.
Further, its amplification is associated with both intrinsic and
acquired resistance to wide variety of antiꢀtumorigenic agents including microtubule binding
12ꢀ15
agents, gemcitabine, TRAIL and the BCLꢀ2/BCLꢀX dual inhibitor navitoclax.
Not only do
L
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
gene silencing approaches that specifically target MCLꢀ1 circumvent this resistance
1
3ꢀ15
phenotype,
but certain cancer cell types frequently undergo cell death in response to MCLꢀ1
1
6
silencing indicating a dependence on MCLꢀ1 for survival. Consequently, approaches that
inhibit MCLꢀ1 function are of considerable interest for cancer therapy.
To both further explore MCLꢀ1 biology and build a starting point for the generation of
clinically relevant MCLꢀ1 inhibitors, we set out to develop potent and selective compounds
capable of demonstrating onꢀtarget activity in a cellular context. Previously we reported the
generation of firstꢀinꢀclass inhibitors of the BCLꢀ2 family proteins, including the BCLꢀ2/BCLꢀ
17,18
X dual inhibitor navitoclax currently in Phase1/2 clinical trials
L
and the BCLꢀ2 selective
1
9
inhibitor ABTꢀ199 (GDCꢀ0199) currently in Phase III clinical trials. We also recently reported
2
0
the first highly potent and selective inhibitor of BCLꢀX that confers onꢀtarget in vivo activity.
L
In our experiences generating effective BCLꢀ2 protein family inhibitors, specific criteria must be
met in order to substantiate mechanismꢀbased activity in a cellular context. Subꢀnanomolar
binding affinity towards the target BCLꢀ2 family protein is essential to compete with the highꢀ
affinity endogenous ligands such as BIM. As with the discovery of inhibitors of BCLꢀ2 and
BCLꢀX , high selectivity for the target (MCLꢀ1) protein over other BCLꢀ2 family members and
L
other unrelated cellular proteins and receptors is an additional and critical requirement for an
effective tool molecule.
Despite intense interest in MCLꢀ1 inhibitors for cancer treatment, the difficulty in targeting
this protein is reflected by a lack of success in generating potent small molecule inhibitors. No
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 55
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
bona fide MCLꢀ1 inhibitor has reached the clinic, and only a limited number of small molecule
2
1ꢀ27
inhibitors of MCLꢀ1 have been disclosed.
Furthermore, in cases where cellular activity has
2
5ꢀ28
been reported,
the weak affinity of these compounds (Ki > 100 nM) confounds the
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
29ꢀ31
interpretation of cellular effects.
Herein we disclose the discovery of a series of small molecule MCLꢀ1 inhibitors that were
3
2
derived from the highꢀthroughput screening (HTS) hit 1. Iterative coꢀcrystal structure
generation of analogs bound to MCLꢀ1 allowed for the rational progression of this micromolar
hit compound into advanced analogs that show subꢀnanomolar and selective binding for the
target protein. Treatment of H929 and BxPCꢀ3 human tumor cell lines with 30b in either a single
agent context or in combination with navitoclax, respectively, provided a preliminary
demonstration of cellular activity.
Figure 1. HTS hit 1.
Chemistry
33
Compound 5 was prepared in three steps from the literature diester 2 (Scheme 1). Selective
borane reduction of 2 afforded the alcohol intermediate 3 directly and in good yield. Mitsunobu
reaction to provide napthyloxy intermediate 4 and subsequent ester hydrolysis with lithium
hydroxide afforded final compound 5.
4
ACS Paragon Plus Environment

Page 5 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
Scheme 1. Preparation of analog 5
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) BH ꢀTHF, ꢀ15 ºC, 77%; (b) DBAD, PPh ,THF, 1ꢀnaphthol in
3
3
THF, 15 ºC, 61%; (c) LiOH in dioxane/water (2/1), 100 ºC, 69%.
The C7ꢀsubstituted analogs 11a-d originated from the common 7ꢀbromo intermediate 7. This
3
3
compound, in turn, was generated via the JappꢀKlingemann reaction in moderate yield. Unlike
the preparation of compound 5, the selective oneꢀstep reduction of the alkyl ester in 7 was lowꢀ
yielding and the purification cumbersome. A twoꢀstep process was thus utilized, starting with
selective hydrolysis of the alkyl ester of 7 under acidic conditions with HCl in acetic acid to
afford intermediate 8 in excellent yield. Subsequent reduction of the alkyl acid via borane
furnished the Mitsunobu substrate 9 in 90% yield over the two steps. Installation of the
napthyloxy group under standard Mitsunobu conditions afforded common intermediate 10.
Finally, compounds 11a-d were prepared via concomitant SuzukiꢀMiyaura coupling and ester
hydrolysis in a one pot process under microwave conditions (Scheme 2).
a
Scheme 2. Preparation of C7ꢀsubstituted analogs 11a-d
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 55
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
Reagents and conditions: (a) NaNO /HCl, H SO , 2ꢀoxoꢀcyclopentanecarboxylic acid ethyl
2
2
4
ester, 40%; (b) AcOH, HCl, 80 ºC, 95%; (c) BH ꢀTHF, ꢀ15 ºC, 95%; (d) DBAD, PPh ,THF, 1ꢀ
3
3
naphthol in THF, 15 ºC, 61%; (e) boronic acid, LiOH, PdCl (PPh ) in DME/ ethanol/water
2
3 2
(
7/2/3), 150 ºC, 38ꢀ62%.
Nꢀsubstituted acid analog 14 was prepared in three steps from intermediate 10. Briefly, indole
nitrogen alkylation with 2ꢀchloroꢀ1ꢀmorpholinoethanone under basic conditions afforded alphaꢀ
indolyl amide 12 in good yield. Treatment with borane effected the chemoselective reduction of
the tethered amide to furnish amine 13, which was then subjected to concomitant Suzukiꢀ
Miyaura coupling and ester hydrolysis to generate final compound 14 (Scheme 3).
a
Scheme 3: Synthesis of N(1)ꢀsubstituted analog 14
6
ACS Paragon Plus Environment

Page 7 of 55
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) Cs CO , DMF, 2ꢀchloroꢀ1ꢀmorpholinoethanone, 86%; (b) BH ꢀ
2
3
3
THF, 81%; (c) oꢀtolylboronic acid, LiOH, PdCl (PPh ) in DME/ ethanol/water (7/2/3), 150 ºC,
2
3 2
8
5%.
Compound 17 was prepared via the key boronic ester 15, which was derived from common
intermediate 10 via palladium mediated boronylation in excellent yield. Suzuki coupling between
compound 15 and 4ꢀiodoꢀ1,3,5ꢀtrimethylꢀ1Hꢀpyrazole afforded the ester 16 in good yield.
Alkylation of the indole nitrogen under Mitsunobu conditions and subsequent ester hydrolysis
under basic conditions yielded compound 17 (Scheme 4).
a
Scheme 4: Synthesis of 17
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 55
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
Reagents and conditions: (a) PdCl (dppf), KOAc DMF, 60 ºC, 91%; (b) 4ꢀiodoꢀ1,3,5ꢀ
2
trimethylꢀ1Hꢀpyrazole, PdCl (dppf)ꢀCH Cl adduct, Cs CO , DMF, 100 ºC, microwave, 74%; (c)
2
2
2
2
3
2
ꢀmorpholinoethanol, PPh , DBAD, THF; (d) NaOH, THF, MeOH, 15% (2 steps).
3
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
Preparation of the side chain of compound 22 was initiated via the transformation of ester 18 to
alcohol 19 under reductive conditions. Iodination of 19 provided alcohol 20, which was
subsequently converted to the bromide 21 in good yield. This intermediate was then carried
through four steps starting with phenol displacement of the benzylic bromide and Suzuki
coupling with 15. Indole nitrogen alkylation with 4ꢀ(2ꢀchloroethyl)morpholine followed by ester
hydrolysis under basic conditions afforded target compound 22 in just under 10% yield for the
four step sequence (Scheme 5).
a
Scheme 5: Synthesis of analog 22
a
Reagents and conditions: (a) LiAlH , THF, 0 °C, 55%; (b) NIS, DMF, 75%; (c) PBr , DCM,
4
3
81%; (d) Cs CO , DMF, phenol; (e) 15, PdCl (dppf)ꢀCH Cl adduct, Cs CO , DMF, 100 ºC,
2
3
2
2
2
2
3
microwave; (f) 4ꢀ(2ꢀchloroethyl)morpholine hydrochloride, Cs CO ; (g) LiOH, THF/H O, 9% (4
2
3
2
steps).
The elaborated side chain of intermediate 28 originated from starting heterocycle 23. The acid
3 was reduced to afford the benzylic alcohol 24 in moderate yield, thus setting the stage for
2
Mitsunobu reaction with tertꢀbutyl 4ꢀ(4ꢀhydroxyphenyl)piperazineꢀ1ꢀcarboxylate to generate the
8
ACS Paragon Plus Environment

Page 9 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
sideꢀchain intermediate 25. Palladiumꢀmediated coupling with core intermediate 15 afforded 26
with retention of the ester. While Mitsunobu conditions failed to affect Nꢀalkylation, reaction
with 4ꢀ(2ꢀchloroethyl)morpholine hydrochloride under basic conditions afforded 27 in suitable
yields. Intermediate 28 was subsequently generated via acidꢀmediated deprotection of the BOC
group, and standard saponification then afforded final compound 29 (Scheme 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
a
Scheme 6: Synthesis of terminal piperazine analog 29
O
N
N
O
N
N
N
N
O
HO
HO
a
b
c
N
H
OEt
Br
BocN
N
N
O
Br
Br
24
26
N
O
Boc
23
25
N N
O
O
O
O
d
N
OEt
BocN
e
N
28: R = Et
29: R = OH
OR
HN
f
N
N
O
N
O
N
N
N
N
N
O
O
27
a
Reagents and conditions: (a) LiAlH , THF, 0 °C, 57%; (b) Tertꢀbutyl 4ꢀ(4ꢀ
4
hydroxyphenyl)piperazineꢀ1ꢀcarboxylate, THF, PPh , DBAD, 60%; (c) 15, PdCl (dppf), Cs CO ,
3
2
2
3
DMF, 100 ºC, microwave, 60%; (d) 4ꢀ(2ꢀchloroethyl)morpholine hydrochloride, TBAI, DMF,
Cs CO ; (e) TFA, DCM, 82%; (f) LiOH, THF/H O, 54%.
2
3
2
Intermediate 28 was used to generate final compounds 30a-f using slight variations of a twoꢀ
step one pot process. Terminal acetamide 30a was furnished via standard acetylation conditions
followed by ester hydrolysis at room temperature. The terminal sulfonamide (30b-d) and
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 55
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
sulfamide (30e, f) compounds were prepared via reaction with the corresponding electrophile
o
and ester hydrolysis at 145 C (Scheme 7).
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
Scheme 7: Synthesis of terminally substituted piperazine analogs 30a-f
a
Reagents and conditions: (a) acetyl chloride or RSO Cl, Et N, THF; (b) LiOH, THF/H O
2
3
2
o
(2/1), 40% (2 steps, 30a); (c) LiOH, THF/H O (2/1), 145 C, 10ꢀ35% (2 steps, 30b-f).
2
Results and Discussion
O
4
7
3
5
6
O
O
2
N
H
OH
N
OH
1
H
1
5
MCLꢀ1 FPA Ki = 4.4 ꢀM MCLꢀ1 FPA Ki = 0.34 ꢀM
Figure 2. Naphthyloxy ether 5 as a starting point for core optimization.
10
ACS Paragon Plus Environment

Page 11 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
A fluorescence polarization assay (FPA)ꢀbased highꢀthroughput screen was conducted to
assess the ability of compounds to disrupt the interaction between recombinant human MCLꢀ1
and a labeled NOXA peptide. The indole acid derivative 1 was identified as a single digit
micromolar inhibitor (Figure 1) in the HTS and subsequently verified in additional assays.
Incubation of 1 with MCLꢀ1 in which the methyl groups of isoleucine, leucine, valine and
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
13
13
methionine were labeled with C, resulted in large chemical shift perturbations in a CꢀHSQC
3
4
spectrum that were indicative of robust binding. We were intrigued by the low molecular
weight of acid 1. Other indole and structurally related heterocyclic compounds lacking a
carboxylate moiety were not active in the HTS, which suggested that the core 2ꢀcarboxyindole
was critical for activity. Consequently, we chose to retain this pharmacophore while initially
exploring the tethered cyclohexylꢀbutyl moiety at the 3 position of the 2ꢀcarboxyindole core.
While changes in tether length or the incorporation of polar moieties in the cyclohexyl region
were deleterious to MCLꢀ1 activity (data not shown), installation of a 4ꢀ(1ꢀnaphthoxypropyl)
group afforded an analog (5, Figure 2) with an approximately tenꢀfold increase in binding
affinity.
In order to progress this early lead, we obtained a coꢀcrystal structure of MCLꢀ1 bound to
compound 5. As shown in Figure 3A, the small molecule binds within the center of the BH3ꢀ
binding groove of MCLꢀ1 and is anchored by a hydrogen bond between the 2ꢀcarboxylate group
and Arg263 in helix 7 of the protein. The same arginine residue forms an essential electrostatic
3
5
interaction with a conserved aspartic acid residue in all BH3ꢀonly proteins, thus providing
rationale for the absence of binding by related indole cores lacking an appropriately positioned
acidic moiety. As demonstrated in the ribbon overlay of the bound conformations of compound 5
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 55
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
and a BIM peptide (Figure 3B) individually complexed to MCLꢀ1, the indole core of 5 occupies
the same space within the protein as the P2 bindingꢀleucine side chain of BIM. Interestingly, the
naphthyl moiety of 5 penetrates deeply into the helix 6, 7 region of the protein (Figure 3C),
where it packs tightly against a number of hydrophobic sidechains including Leu235, Leu246,
Met250, Tyr270, and Ile294. This is accommodated by a shift of helix 4 of MCLꢀ1 relative to the
bound conformation with BIM, thus demonstrating a distinct binding mode of the small molecule
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
compared to the peptide (Figure 3D) as recently reported with a structurally similar compound.
A
B
R263
P0 P1
P2 P3
P4
C
D
α4
α3
12
ACS Paragon Plus Environment

Page 13 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 3. (A) Crystal structure of 5 bound to MCLꢀ1, 2.25 Å. (B) Overlay of 5 with bound
conformation of BIM peptide. (C) Solventꢀaccessible surface for complex of MCLꢀ1 with 5.
(D) Backbone overlay of MCLꢀ1 protein bound to 5, grey ribbon, with MCLꢀ1 protein bound to
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
BIM peptide, orange ribbon. BIM peptide removed for clarity.
The coꢀcrystal structure of MCLꢀ1 with compound 5 suggested that vectors from the C4ꢀ6
positions on the indole core would be limited by the tight van der Waals interactions within the
corresponding region of the protein (Figure 3C). The C7 position of the bound indole, in
contrast, was solvent exposed yet in proximity to the hydrophobic surface partially comprised by
the amino acid sidechains of Ala227 and Met231 in helix 3. This observation suggested that
affinity gains could be made with the appropriate substitution at the C7 position. To rapidly
explore this vector, we prepared a simple phenylꢀsubstituted compound (11a) along with analogs
containing a methyl group at either the ortho, meta or para position of the C7ꢀphenyl moiety. As
shown in Table 1, the orthoꢀtolyl substituent in 11b conferred an order of magnitude
improvement in binding affinity relative to parent 5 while the phenyl analog 11a and analogs 11c
and 11d all demonstrated an approximately twoꢀfold boost in affinity.
O
O
N
OH
R
2
R
1
Table 1: Structure activity relationship (SAR) of 2ꢀindole carboxylic acids by FPA.
Compounda
R
1
R
2
K
b
(µM)
K (µM)
b
i
i
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 55
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
1
% HS
5
H
H
H
H
H
H
0.34 ± 0.06
0.13 ± 0.02
0.035 ± 0.013
0.16
7.3 ± 1.8
5.3 ± 1.1
8.6
1
1a
1b
Ph
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
o-tolyl
m-tolyl
p-tolyl
11c
8.2
1
1d
0.15
9.1
14
o-tolyl
N
0.009 ± 0.004*
0.47 ± 0.18*
O
a
1
b
All compounds were > 95% pure by HPLC and characterized by H NMR and MS. Values are mean ± standard deviation for
two experiments run in duplicate. Asterisk indicates geometric mean
± standard deviation for greater than three experiments run
in duplicate. Values without error are single experiments run in duplicate.
The presence of a hydrophobic core tethered to a carboxylate group suggested that human
serum albumin (HSA) binding would be substantial for the pharmacophore exemplified by
3
6
1
1b. This hypothesis was confirmed upon FPA assessment in the presence of small amounts of
human serum (HS), where all analogs showed substantially decreased apparent K values (Table
i
1
). To gain further insight into the SAR of albumin binding, compound 11b and analogs bearing
systematic deletions were assessed in an assay measuring the displacement of a labeled probe
3
7
from recombinant domain III of HSA (supplementary information). As shown in Figure 4, 11b
bound to domain III with a high affinity, thus reaffirming the FPA data in the presence of 1%
HS. Removal of the 7ꢀoꢀtolyl group (5) decreased the domain III affinity by at least an order of
magnitude, an effect similar to the difference in MCLꢀ1 affinity between 11b and 5. Retention of
the 7ꢀoꢀtolyl group and removal of the tethered naphthyl group (31) afforded greater than 150ꢀ
14
ACS Paragon Plus Environment

Page 15 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
fold diminution in domain III binding as well as an expected loss of MCLꢀ1 affinity. Finally,
deletion of both naphthyl and 7ꢀoꢀtolyl moieties (32) depleted all measurable binding to domain
III as well as MCLꢀ1.
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
5
O
CLogP = 4.8
Ki = 0.40 ꢀM
OH
O
N
H
OH
O
O
OH
N
OH
N
OH
1b
CLogP = 6.9
Ki < 0.03 ꢀM
H
H
1
32
O
CLogP = 0.80
Ki > 100 ꢀM
No MCLꢀ1 binding
31
N
H
OH
CLogP = 2.9
Ki = 5.5 ꢀM
No MCLꢀ1 binding
Figure 4. Brief SAR study of human serum albumin domain III.
This quick analysis indicated both that the substituents critical for MCLꢀ1 and domain III
affinity were inseparable, and that the current pharmacophore would be hampered by extreme
affinity shifts in the presence of HSA. As MCLꢀ1 protein levels decline following serum
deprivation, the use of HSA or other proteinꢀrich media is critical for studying MCLꢀ1 at a
cellular level. The effects of high affinity binding to the abundant albumin protein would also be
expected to alter the distribution, excretion and other parameters in vivo. The identification of
pharmacophore modifications that could increase MCLꢀ1 affinity while alleviating albumin
binding was thus deemed critical at this point in the optimization process.
Because the N(1) position of the indole core comprised a solvent vector with no closely
available hydrophobic interactions (Figure 3), we reasoned that a polar substituent could be
3
6,37
incorporated at this position to alleviate albumin binding.
To test this hypothesis, an analog
of 11b possessing an N(1)ꢀtethered morpholinoethyl substituent was prepared. The resultant
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 55
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
compound (14) showed modestly improved MCLꢀ1 affinity relative to 11b despite the putatively
solventꢀexposed nature of the tethered amine moiety. Additionally, the diminution of binding
affinity in the presence of 1% HS was substantially less for 14 as compared to 11b. This result
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
was corroborated in the recombinant domain III binding assay where the K of 14 (360 nM) was
i
at least ten fold lower than that of 11b, thus consistent with previous reports demonstrating
3
6,37
reduced affinity of albumin for zwitterionic species.
We subsequently obtained a crystal structure of 14 bound to MCLꢀ1. This structure (Figure
5B, C) confirmed the projection of the N(1) substituent into the solvent space along with the
maintenance of the overall pharmacophore binding motif. The 7’ꢀoꢀtolyl methyl group was
oriented towards the P3 and P4 pockets of the BH3 binding groove, where it sat in close
proximity to the P3ꢀbinding isoleucine residue of the BIM peptide as shown in the ribbon
overlay diagram (Figure 5D).
16
ACS Paragon Plus Environment

Page 17 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
A
B
R263
O
4
7
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
5
6
O
2
1
A227
N
OH
N
M231
O
1
4
C
D
P3, P4
P0 P1
P2 P3 P4
P0, P1
Figure 5. (A) Structure of 14 with numbered carbons. (B) Crystal structure of 14 bound to
MCLꢀ1, 1.95 Å. (C) Solventꢀaccessible surface for 14 bound to MCLꢀ1. (D) Overlay of 14
with the bound conformation of the BIM peptide.
As the 6’ position on the C7’ꢀoꢀtolyl ring of 14 was oriented towards the hydrophobic P1
pocket, we envisioned that elaboration towards this vector with small nonꢀpolar substitution
could achieve further affinity gains. It also appeared that an appropriately disubstituted 5ꢀ
membered ring could afford the optimal geometry for engaging the P1 and P3/P4 pockets given
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 55
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
the greater bond angle relative to an ortho,orthoꢀdisubstituted phenyl moiety (Figure 6). Finally,
the use of a moderately polar heterocycle in place of the 7ꢀsubstituted phenyl group could further
3
6
alleviate protein binding. These design hypotheses culminated in the preparation of an analog
containing the readily available (1,3,5)ꢀtrimethylꢀ1Hꢀpyrazole moiety at the 7ꢀindole position
while retaining the N(1) 2ꢀmorpholinoethyl group (17, Figure 6). Subsequent testing in the FPA
revealed binding affinity below the practical detection limit based on the concentration of MCLꢀ
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
protein and NOXA probe. In the presence of 1% HS, the apparent K value was dramatically
i
improved relative to compound 14, providing initial validation of the overall design hypothesis.
O
O
O
O
O
O
O
O
N
OH
N
OH
20 o
C
N
OH
N
OH
1
47 o
C
1
R
2
R2
N
N
N
N
N N
O
O
O
O
1
4
17
MCLꢀ1 FPA Ki = 0.009 ꢀM
Ki 1% HS = 0.47 ꢀM
MCLꢀ1 FPA Ki < 0.01 ꢀM
Ki 1% HS = 0.004 ꢀM
Figure 6. Putative disubstituted 6 and 5ꢀmembered rings targeting both P1 and P3/P4 pockets
and the design of analog 17. Bond angles calculated using the Chem3D Pro v12.0 program from
CambridgeSoft. Binding affinity of 17 (FPA) without and with 1% HS.
At this point in our efforts it was clear that the lower limit of detection in the FPA was
preventing the assessment of intrinsic MCLꢀ1 affinity. While the presence of HS in the FPA
afforded a vantage point for comparing relative activity of analogs, we elected to develop a timeꢀ
18
ACS Paragon Plus Environment

Page 19 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
resolved fluorescence energy transfer (TRꢀFRET) assay which measured the ability of
2
0
compounds to inhibit the interaction between MCLꢀ1 and a labeled BAK peptide. The ability to
detect these interactions at low probe concentrations enabled accurate and reproducible
measurement of compound affinity to a lower limit of 10 picomolar. The HS content was
increased to 10% to further differentiate the binding of compounds to MCLꢀ1 in a more
physiologically relevant setting.
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
As shown in Table 2, 17 achieved an over tenꢀfold gain in affinity relative to analog 14 in the
TRꢀFRET assay. The presence of 10% HS shifted the apparent K of 17 to a lesser degree than
i
1
4, which was consistent with a lower affinity for HSA domain III (K = 1.45 µM). Encouraged
i
by these results, efforts were then focused on further engaging the P3/P4 pocket region of MCLꢀ
1. This extended region remained shallow yet exposed upon binding to 14 (Figure 5BꢀD),
suggesting that affinity gains could be obtained through extended orthoꢀsubstitution on the 7ꢀ
heterocycle. The ability to routinely install phenolic moieties via Mitsunobu reaction lead to the
design and synthesis of hydrophobic analog 22 as well as the extended 4ꢀpiperazinylꢀphenoxy
analogs 28 and 30a. While compounds 22 and 28 showed similar binding affinity relative to 14,
Nꢀacetylpiperazine analog 30a showed a dramatic boost in affinity. This subꢀnanomolar affinity
of MCLꢀ1 translated well in the presence of 10% HS, where an approximately tenꢀfold rightward
shift still afforded single digit nanomolar binding.
Table 2: MCLꢀ1 binding affinity of 14, 17 and orthoꢀextended C7ꢀpyrazole analogs 22, 28, and
30a-e (TRꢀFRET).
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 55
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
Compounda
R
K
(nM)b
K
i
(nM) 10% HSb
i
1
4
21 ± 7.9
1230 ± 132
17
1.5 ± 0.63
35 ± 7.4
114 ± 55*
12 ± 0.49
N
N
2
2
2
8
2.3 ± 1.2*
2.0 ± 0.7
3
0
0.24 ± 0.19
0.43 ± 0.34
0.20 ± 0.04
2.7 ± 1.4
6.0 ± 2.0
4.9 ± 1.6
3
0b
3
0c
20
ACS Paragon Plus Environment

Page 21 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
30d
0.31 ± 0.17
0.45 ± 0.40
8.8 ± 3.2
7.9 ± 2.0
12.3 ± 6.8
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
0e
3
0f
0.48 ± 0.30
a
1
b
All compounds were > 95% pure by HPLC and characterized by H NMR and MS. Values are geometric mean ± standard
deviation for at least three experiments run in triplicate. Asterisk indicates geometric mean ± standard deviation for two or more
experiments run in triplicate.
During the process of exploring more routine SAR of this analog, we obtained a coꢀcrystal
structure of related compound 33 with MCLꢀ1 (Figure 7, see supplementary information for
synthesis route and characterization). As shown in Figure 7BꢀD, the extended Nꢀacetylpiperazine
moiety of 33 binds along the P4 pockets of the protein, thus validating the design of a single
orthoꢀextension targeting this domain. The 5ꢀsubstituted methyl group projected into the P1
pocket as expected, where it was oriented towards the Met231 residue of helix 2. The remaining
portions of the pharmacophore bind to the protein in the same manner as compound 14.
The substantial increase in MCLꢀ1 binding of acetylꢀcapped analog 30a relative to
unsubstituted piperazine analog 28 indicated a preference for the neutral amide versus a charged
amine in the P4 pocket. The extent to which the methyl terminus of 30a was engaging the
protein was not clear from the Xꢀray structure of MCLꢀ1 bound to analog 33 (Figure 7).
Consequently, we decided to probe this region of the molecule with substitutions that retained
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 55
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
the sp ꢀhybridized piperazine nitrogen while presenting different termini to the P4 pocket.
Specifically, we prepared analogs with simple sulfonamide or sulfamoyl groups that would allow
for a quick survey of hydrophobic and moderately hydrophilic substitution at this region (30b-f).
All analogs showed similar subꢀnanomolar binding affinity to MCLꢀ1, indicating a degree of
tolerance within the P4 pocket region of MCLꢀ1 (Table 2).
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
O
R263
O
N
OH
O
N N
N
3
3
N
MCLꢀ1 FRET Ki = 1.8 nM
Ki 10% HS = 98 nM
N
A227
M231
O
C
D
P0 P1
P2 P3 P4
22
ACS Paragon Plus Environment

Page 23 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 7. (A) Structure and MCLꢀ1 affinity of compound 33 (TRꢀFRET). (B) Crystal structure
of 33 bound to MCLꢀ1, 2.31 Å. (C) Solventꢀaccessible surface for 33 bound to MCLꢀ1. (D)
Overlay of 33 with bound conformation of BIM peptide.
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
Having achieved picomolar MCLꢀ1 inhibition, compound 30b was selected for additional
profiling. Routine screening against the proꢀsurvival BCLꢀ2 family proteins revealed excellent
selectivity (FRET K > 0.66 µM for BCLꢀ2, BCLꢀX , BCLꢀW and A1). Additionally, 30b
i
L
showed little to no binding below 10 micromolar in panels of kinases and GPCRs (see
Supplementary information). In the MCLꢀ1ꢀdependent multiple myeloma cell line H929,
3
8
treatment with 30b induced caspaseꢀactivation (Figure 8a) with associated loss in cell viability
Figure 8b) within 4 hours. These apoptotic effects were dependent upon media FBS
(
concentration, reflecting the significant protein binding properties of this chemical class.
Furthermore, as MCLꢀ1 has been identified as a resistance factor for the dual BCLꢀ2/BCLꢀXL
15
13,39
inhibitor navitoclax, we treated the human pancreatic tumor cell line (BxPC3)
with
navitoclax in the absence or presence of 30b (10% FBS). While navitoclax induced cell killing
with an EC of just under 1 micromolar, addition of 30b shifted the EC by over 45ꢀfold in the
5
0
50
presence of 10% FBS (Figure 8). These data provide preliminary evidence that 30b can inhibit
MCLꢀ1 in a cellular context. Micromolar concentrations of 30b were required for cellular effects
despite subꢀnanomolar binding affinity; this difference is at least partially attributable to high
protein binding as reflected by the rightꢀshift in FRET and viability assays, as well as the low
4
0
free fraction (f ) of 30b in human plasma (f < 0.01). Additionally, the zwitterionic
u
u
41,42
pharmacophore may have inherent limitations in passive diffusion.
The ability to penetrate
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 55
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
cells and induce caspase activation within 4 hours, however, indicates that permeability may not
be the major contributor to the disparity between biochemical affinity and cellular potency.
A
B
C
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. H929 cells were treated with increasing concentrations of 30b for 4 hours in media
containing 0.1%, 1% or 10% FBS and caspaseꢀ3/ꢀ7 (A) or cell viability (B) determined using
caspaseꢀGlo® 3/7 or CellTiterꢀGlo® respectively. Alternatively, BxPC3 cells were treated with
increasing doses of navitoclax in the presence or absence of 30b (20 µM) for 72 hours in media
containing 10% FBS and cell viability determined with CellTiterꢀGlo® (C). Data are presented
as the mean ± s.e.m of at least three independent experiments.
The identification of compounds that directly bind to MCLꢀ1 and demonstrate selective and
onꢀtarget activity is of critical importance for both advancing the understanding of MCLꢀ1
biology as well as laying the foundation for drug discovery efforts targeting this protein. Herein
we describe the identification of an HTS hit with micromolar affinity to MCLꢀ1 and its
subsequent progression into a novel series of small molecule antagonists that exhibit a unique
binding mode relative to the BH3ꢀonly proteins. The generation of multiple Xꢀray coꢀcrystal
structures allowed for the rational design of binding elements that target key hotꢀspots within
MCLꢀ1 and ultimately allowed for the generation of advanced compounds exemplified by 30b.
This small molecule exhibits picomolar affinity for MCLꢀ1 along with selectivity over other proꢀ
24
ACS Paragon Plus Environment

Page 25 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
survival BCLꢀ2 family proteins as well as multiple kinases and GPCRs. A preliminary
demonstration of the cellular activity of 30b has been disclosed, and a thorough mechanistic
3
9
study of this chemical series will be reported in an accompanying manuscript. Compounds
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
exemplified by 30b thus represent effective tools for studying MCLꢀ1 biology.
ASSOCIATED CONTENT
Supporting Information. Biological assay protocols, synthetic procedures and analytic data
including NMR, MS and crystallographic methods. This material is available free of charge via
the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*
Email: andrew.souers@abbvie.com Tel : (847) 937ꢀ5312.
Present Addresses
SE, 2025 Gabriel Avenue, Zion, IL 60099.
EF, 2011 w. Khaibar Place, Tucson, AZ 85704.
Notes
MB, LW, GS, AP, HZ, PN, DCP, SE, SF, EF, LH, SJ, RJ, PJK, DM, CP, SHR, XS, CS, SKT, ZT,
XW, YX, GJ, CT, JDL, SWE, and AJS are employees of AbbVie or were employed by AbbVie or
Abbott Laboratories at the time the research was conducted. This work was sponsored by
AbbVie, Inc.
ABBREVIATIONS
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 55
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
MCLꢀ1, Myeloid cell leukemia sequence 1; BCLꢀ2, Bꢀcell lymphoma 2; BCLꢀX , Bꢀcell
L
lymphomaꢀextra large; HTS, highꢀthroughput screen; HS, human serum; HSA, human serum
albumin; GPCRs, Gꢀprotein coupled receptors; FPA, fluorescence polarization assay.
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
Experimental Section
All commercially available chemical reagents were used without further purification unless
otherwise noted. Flash chromatography was carried out using an ISCO Companion flash system.
Proton NMR data were recorded using a Varian Mercury 300 spectrometer or a Varian UNITY
4
00/500 spectrometer. Mass spectral data were collected using a Finnigan SSQ7000 GC/MS
(ESI) or a Finnigan DCI/MS SSQ7000. Preparative HPLC was performed on a ZorBAX RXꢀC18
column (250x21.2 mm, 7 µm particle size) using a gradient of 10% to 100% acetonitrile: 0.1%
aqueous TFA over 40 minutes at a flow rate of 15 mL/min. Analytical LCꢀMS was performed
on a Finnigan Navigator mass spectrometer and Agilent 1100 HPLC system running Xcalibur
1
.2, OpenꢀAccess 1.3, and custom login software. The mass spectrometer was operated under
positive APCI ionization conditions. The HPLC system comprised an Agilent Quaternary pump,
degasser, column compartment, autosampler and diodeꢀarray detector, with a Polymer Labs
ELSꢀ2100 evaporative lightꢀscattering detector. The column used was a Phenomenex Luna
CombiꢀHTS C8(2) 5ꢁm 100Å (2.1 mm × 50 mm), at a temperature of 55 °C. A gradient of 5ꢀ
1
00% acetonitrile and 0.1% trifluoroacetic acid in water was used, at a flow rate of 2.0 mL/min
(
0ꢀ0.1 min 5% acetonitrile, 0.1ꢀ2.6 min 5ꢀ100% acetonitrile, 2.6ꢀ2.9 min 100% acetonitrile, 2.9ꢀ
3.0 min 100ꢀ5% acetonitrile. 0.5 min postꢀrun delay). A Biotage Initiator was used for
microwave reactions. All test compounds were ≥ 95% purity as assessed by HPLC or HPLC and
NMR.
26
ACS Paragon Plus Environment

Page 27 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Ethyl 3-(3-hydroxypropyl)-1H-indole-2-carboxylate (3). To a solution of ethyl 3ꢀ(3ꢀethoxyꢀ
3
3
3
ꢀoxopropyl)ꢀ1Hꢀindoleꢀ2ꢀcarboxylate (2.82 g, 9.75 mmol) in THF (40 mL) cooled to ꢀ15 °C
was added a solution of boraneꢀTHF (1M, 40 mL, 40 mmol) dropwise. The reaction mixture was
slowly warmed to ambient temperature, stirred for 16 h and then slowly quenched with methanol
(100 mL). The resulting mixture was concentrated and the residue purified by flash
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
chromatography (5 ꢀ 25 % EtOAc/hexanes) to provide 1.85 g (77%) of the title compound. H
NMR (500 MHz, DMSO) δ 11.45 (s, 1H), 7.67 (d, J = 8.09 Hz, 1H), 7.41 (d, J = 8.28 Hz, 1H),
7.29 – 7.18 (m, 1H), 7.11 – 7.00 (m, 1H), 4.42 (t, J = 5.18 Hz, 1H), 4.34 (q, J = 7.10 Hz, 2H),
3
.43 (dd, J = 6.61, 11.88 Hz, 2H), 3.10 – 3.01 (m, 2H), 1.80 – 1.69 (m, 2H), 1.36 (t, J = 7.11 Hz,
+
+
3
H). MS (ESI ) m/z 247.9 (M+H) .
Ethyl 3-(3-(naphthalen-1-yloxy)propyl)-1H-indole-2-carboxylate (4). A solution of 1ꢀ
naphthol (4.45 g, 30.9 mmol), alcohol 3 (7.27 g, 29.4 mmol), diꢀtertꢀbutyl azodicarboxylate
7.78 g, 33.8 mmol), and triphenylphosphine (8.87 g, 33.8 mmol) in THF (100 mL) was stirred at
(
ambient temperature for 24 h. The mixture was concentrated and the product was purified by
flash chromatography (0 ꢀ 10% EtOAc/hexanes) followed by crystallization from hexanes to
1
provide 6.74 g (61%) of the title compound. H NMR (300 MHz, DMSO) δ 11.53 (s, 1H), 8.22
(
dd, J = 3.57, 6.18 Hz, 1H), 7.90 – 7.85 (m, 1H), 7.68 (d, J = 8.07 Hz, 1H), 7.57 – 7.48 (m, 2H),
.41 (dt, J = 8.33, 15.91 Hz, 3H), 7.27 – 7.18 (m, 1H), 6.97 (dd, J = 4.01, 11.13 Hz, 1H), 6.88 (d,
J = 6.77 Hz, 1H), 4.31 (q, J = 7.09 Hz, 2H), 4.17 (t, J = 6.08 Hz, 2H), 3.33 (t, J = 8.34 Hz, 2H),
7
+
+
2
.25 – 2.15 (m, 2H), 1.31 (t, J = 7.10 Hz, 3H). MS (ESI ) m/z 374.1 (M+H) .
-(3-(Naphthalen-1-yloxy)propyl)-1H-indole-2-carboxylic acid (5). A suspension of ester 4
0.33 g, 0.88 mmol) and LiOHꢂH O (0.1 g, 2.4 mmol) in MeOH/water (12 mL, 3/1) was heated
3
(
2
to 150 °C for 10 min on a Biotage Initiator microwave reactor. The reaction mixture was
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 55
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
neutralized with HCl (1M, 3 mL) and the product was extracted with EtOAc (3 x 10 mL). The
combined organic layers were dried over Na SO , filtered and concentrated. The product was
2
4
purified by reverse phase preparative HPLC (C18, gradient 30 to 100 % of MeCN/water/0.1%
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
TFA) to provide 0.21 g (69%) of 5. H NMR (500 MHz, DMSO) δ 12.96 (d, J = 0.49 Hz, 1H),
11.44 (s, 1H), 8.27 – 8.21 (m, 1H), 7.86 (dd, J = 2.40, 6.83 Hz, 1H), 7.66 (d, J = 8.07 Hz, 1H),
7
.56 – 7.48 (m, 2H), 7.45 (d, J = 8.24 Hz, 1H), 7.38 (dd, J = 8.45, 16.41 Hz, 2H), 7.24 – 7.18 (m,
H), 6.99 – 6.93 (m, 1H), 6.87 (d, J = 7.49 Hz, 1H), 4.16 (t, J = 6.11 Hz, 2H), 3.36 – 3.29 (m,
1
+
+
2H), 2.27 – 2.14 (m, 2H). MS (ESI ) m/z 346.1 (M+H) .
Ethyl 7-bromo-3-(3-ethoxy-3-oxopropyl)-1H-indole-2-carboxylate (7). 2ꢀBromoaniline
(34.2 g, 200 mmol) was dissolved in an aqueous HCl solution (50 mL cc HCl, 320 mL water,
6
00 mmol) and the reaction solution cooled to 0 °C. 2.5 M NaNO (80 mL, 200 mmol) in water
2
was added dropwise at 0 °C. After the addition was complete, 4.5 M NaOAc (250 mL, 1125
mmol) in water was added, followed by 2ꢀoxoꢀcyclopentanecarboxylic acid ethyl ester (3) (30
mL, 200 mmol). The reaction mixture was stirred at 0 °C for 15 min, then warmed to 19 °C over
2 h. The reaction mixture was then extracted with dichloromethane (250 mL x 3), dried over
anhydrous MgSO , filtered and concentrated under reduced pressure. The residue was dissolved
4
in H SO /EtOH (230 mL, 30 mL:200 mL) and refluxed overnight. The reaction mixture was then
2
4
cooled to ambient temperature and quenched with water (1.5 L) to produce a solid. The solid was
collected by filtration. The crude product was filtered through silica gel, eluting with
dichloromethane. The product was crystallized from hexane (250 mL) to afford 32.5 g (40 %) of
1
the title compound. H NMR (300 MHz, DMSO) δ 11.40 (s, 1H), 7.73 (d, J = 8.04 Hz, 1H),
7.48 (t, J = 14.56 Hz, 1H), 7.04 (t, J = 7.78 Hz, 1H), 4.36 (q, J = 7.11 Hz, 2H), 3.99 (q, J = 7.11
28
ACS Paragon Plus Environment

Page 29 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Hz, 2H), 3.29 (t, J = 7.94 Hz, 2H), 2.59 (t, J = 7.66 Hz, 2H), 1.36 (t, J = 7.10 Hz, 3H), 1.11 (t, J
+
+
=
7.11 Hz, 3H). MS (ESI ) m/z 367.9 (M+H) .
-(7-Bromo-2-(ethoxycarbonyl)-1H-indol-3-yl)propanoic acid (8). A suspension of diester
(16.1 g, 43.7 mmol) in AcOH (150 mL) was heated to 80 °C until all starting material
3
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
7
dissolved. Concentrated HCl (12N) (20 mL) was added and the reaction mixture was stirred at
0 °C for 3 h. Upon formation of a precipitate, the reaction mixture was cooled to ambient
8
temperature and water (250 mL) was added. The precipitate was then collected by filtration,
washed with water (2 x 150 mL) and dried under reduced pressure to furnish 14.2 g (95%) of the
1
title compound. H NMR (300 MHz, DMSO) δ 12.08 (s, 1H), 11.39 (s, 1H), 7.74 (d, J = 8.03
Hz, 1H), 7.51 (dd, J = 0.76, 7.53 Hz, 1H), 7.13 – 6.94 (m, 1H), 4.36 (q, J = 7.09 Hz, 2H), 3.27 (t,
+
J = 7.81 Hz, 2H), 2.52 (t, J = 7.81 Hz, 2H), 1.36 (t, J = 7.10 Hz, 3H). MS (ESI ) m/z 340.0
+
(
M+H) .
Ethyl 7-bromo-3-(3-hydroxypropyl)-1H-indole-2-carboxylate (9). A solution of acid 8
(
14.2 g, 41.6 mmol) in THF (100 mL) was cooled to – 15 °C under N . A boraneꢀTHF solution
2
(1M, 46 mL, 1.1 eq.) was then added dropwise and the reaction mixture allowed to warm to
ambient temperature and stirred for 16 h. After this time the reaction mixture was quenched with
methanol (50 mL), stirred for 30 min and concentrated. The residue was purified by flash
1
chromatography (5 ꢀ 25 % EtOAc/hexanes) to afford 12.9 g (95 %) of the title compound. H
NMR (300 MHz, DMSO) δ 11.26 (s, 1H), 7.72 (dd, J = 0.85, 8.06 Hz, 1H), 7.50 (dd, J = 0.72,
7.49 Hz, 1H), 7.14 – 6.87 (m, 1H), 4.44 (t, J = 5.16 Hz, 1H), 4.35 (q, J = 7.08 Hz, 2H), 3.42 (dd,
J = 6.58, 11.83 Hz, 2H), 3.11 – 2.98 (m, 2H), 1.81 – 1.66 (m, 2H), 1.37 (t, J = 7.10 Hz, 3H). MS
+
+
(
ESI ) m/z 325.9 (M+H) .
29
ACS Paragon Plus Environment