Fragment-based discovery of potent inhibitors of the anti-apoptotic MCL-1 protein

Article abstract of DOI:10.1016/j.bmcl.2014.02.010

Apoptosis is regulated by the BCL-2 family of proteins, which is comprised of both pro-death and pro-survival members. Evasion of apoptosis is a hallmark of malignant cells. One way in which cancer cells achieve this evasion is thru overexpression of the pro-survival members of the BCL-2 family. Overexpression of MCL-1, a pro-survival protein, has been shown to be a resistance factor for Navitoclax, a potent inhibitor of BCL-2 and BCL-XL. Here we describe the use of fragment screening methods and structural biology to drive the discovery of novel MCL-1 inhibitors from two distinct structural classes. Specifically, cores derived from a biphenyl sulfonamide and salicylic acid were uncovered in an NMR-based fragment screen and elaborated using high throughput analog synthesis. This culminated in the discovery of selective and potent inhibitors of MCL-1 that may serve as promising leads for medicinal chemistry optimization efforts.

Full text of DOI:10.1016/j.bmcl.2014.02.010

Bioorganic & Medicinal Chemistry Letters 24 (2014) 1484–1488
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Fragment-based discovery of potent inhibitors of the anti-apoptotic
MCL-1 protein
⇑
Andrew M. Petros , Steven L. Swann, Danying Song, Kerren Swinger, Chang Park, Haichao Zhang,
⇑
Michael D. Wendt, Aaron R. Kunzer, Andrew J. Souers, Chaohong Sun
Research & Development, AbbVie, 1 North Waukegan Road, North Chicago, IL 60064, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Apoptosis is regulated by the BCL-2 family of proteins, which is comprised of both pro-death and pro-sur-
vival members. Evasion of apoptosis is a hallmark of malignant cells. One way in which cancer cells
achieve this evasion is thru overexpression of the pro-survival members of the BCL-2 family. Overexpres-
sion of MCL-1, a pro-survival protein, has been shown to be a resistance factor for Navitoclax, a potent
inhibitor of BCL-2 and BCL-XL. Here we describe the use of fragment screening methods and structural
biology to drive the discovery of novel MCL-1 inhibitors from two distinct structural classes. Specifically,
cores derived from a biphenyl sulfonamide and salicylic acid were uncovered in an NMR-based fragment
screen and elaborated using high throughput analog synthesis. This culminated in the discovery of
selective and potent inhibitors of MCL-1 that may serve as promising leads for medicinal chemistry
optimization efforts.
Received 19 November 2013
Accepted 4 February 2014
Available online 14 February 2014
Keywords:
NMR
Fragment-based
FBDD
FBLD
Apoptosis
BCL
Ó 2014 Elsevier Ltd. All rights reserved.
MCL-1
Programmed cell death, or apoptosis, is vital for such critical
functions as normal embryonic development, tissue remodeling,
cellular homeostasis, and differentiation events. The ability to
evade apoptosis is one hallmark of cancer, and allows cancer cells
to survive stresses that would kill a normal cell.1 For many types of
cancer, resistance to apoptosis arises via overexpression of one or
more anti-apoptotic proteins of the BCL-2 family, whose members
include BCL-2, BCL-XL, BCL-W, A1, and MCL-1. Consequently, there
has been substantial interest in targeting these proteins. This strat-
egy has been recently validated through the clinical activity of
Navitoclax, a potent inhibitor of both BCL-2 and BCL-XL,2 and
ABT-199, a selective inhibitor of BCL-2.3
Navitoclax has shown preclinical efficacy against a number of
different cancer models that are dependent on BCL-2, BCL-XL, or
both.2,4 Additionally, Navitolcax has shown clinical efficacy in
patients with hematological malignancies that are believed to be
dependent upon BCL-2 for survival. However, MCL-1, for which
Navitoclax has poor affinity, has been demonstrated to act as a
resistance factor in several tumor types.5 Additionally, MCL-1 plays
a vital role in the maintenance of several types of cancer. Therefore,
development of a potent MCL-1 inhibitor could provide an alterna-
tive route for apoptosis-inducing cancer therapy. Toward this end,
we employed a fragment-based strategy to identify two novel
series of MCL-1 binders. These molecules were optimized into
nanomolar inhibitors via structure-based design and high through-
put analog synthesis, and now constitute attractive leads for
further medicinal chemistry efforts.
An initial NMR-based fragment screen was carried out follow-
ing the standard method we currently employ for protein-based
NMR screening.6 For this method, recombinant, human, MCL-1
protein, labeled with 13C at the methyl groups of isoleucine (d
only), leucine, valine, and methionine was expressed in Escherichia
coli and purified as described in Supporting information. Screening
was then carried out by monitoring changes in the 13C-HSQC spec-
trum of this methyl-labeled protein upon addition of approxi-
mately 17,000 potential binders, which were initially added in
mixtures of 30. The hit rate from this screen was high (ꢀ0.6%) sug-
gesting that MCL-1 is amenable to small molecule inhibition. The
affinity of the top NMR binders was measured in a fluorescence
polarization assay (FPA, Supporting information) and ranked for
binding efficiency, (BE) as shown in Table 1. Based on the criteria
of both binding efficiency and synthetic tractability, the aryl
sulfonamide (1) and salicylic acid (2) were chosen for further study
and elaboration.
High resolution crystal structure information is widely recog-
nized as the critical piece of information necessary to help drive
rapid and efficient synthetic elaboration of fragment ligands. How-
ever, we were unable to obtain co-crystals of MCL-1 with either of
⇑
Corresponding authors. Tel.: +1 847 937 1922 (A.M.P.), +1 847 935 4730 (C.S.).
E-mail addresses: andrew.petros@abbvie.com (A.M. Petros), chaohong.sun@
abbvie.com (C. Sun).
http://dx.doi.org/10.1016/j.bmcl.2014.02.010
0960-894X/Ó 2014 Elsevier Ltd. All rights reserved.

A. M. Petros et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1484–1488
1485
Table 1
MCL-1 fragments
O
OEt
NH2
O
OEt
O
OH
H
H
O
S
O
S
N
N
a
b
Y
Y
No.
Structure
IC50
(l
M)
BE
17
10O
O
R = H
O
6
11
OH
R
X
X
H
N
1
5
R = Br
c
S
O
O
O
OEt
O
OEt
O
OH
a
b
S
H
H
O
S
O
O
S
O
NH2
N
N
Y
Y
2
3
160
21
18
HO
7
9
X
8
X
X
O
OH
F
F
F
Scheme 1. Synthesis of 2-Aminobenzothiazole Library: (a) Y(SO2)Cl, DMF, DIEA,
25 °C; (b) 1 N NaOH, 1,4-dioxane; (c) Pd(dppf)-CHCl3-adduct, 1,4-dioxane, 2 N
Cs2CO3(aq), boronate ester.
70
O
N
N
H
tide binding groove. Briefly, ethyl 2-amino-4-bromobenzoate (6)
was first reacted with a small library of aryl bornonates (Scheme 1)
to provide the primary point of diversity. Reaction with phenyl sul-
fonyl chloride and subsequent hydrolysis to the carboxylic acid
yielded the aryl sulfonamide analogs shown in Table 2. The
most potent substitution at this position of the molecule was a
4-methyl naphthyl ring (15), which yielded an IC50 of 400 nM
against MCL-1.
O
OH
4
5
130
280
21
19
N
N
HO
The solvent-exposed sulfonamide moiety was chosen as the
second point of exploration as it was oriented parallel with the
groove. To this end, the para-substituted benzoate (7) shown in
Scheme 1 was reacted with a small library of diverse sulfonyl chlo-
rides, again followed by acid formation via hydrolysis of the corre-
sponding ethyl ester. The resulting compounds are shown in
Table 3, the most potent of which (24) contains a tethered pyrazole
ring and yielded an IC50 of 30 nM.
To better understand the key binding elements for this series of
compounds, a crystal structure of (17) was obtained at 2.5 A reso-
lution (PDB ID 4OQ5). The acid moiety of this ligand interacts with
Arg 263 of the protein in a similar fashion to that observed for the
conserved Asp of the amphipathic peptides from the pro-apoptotic
proteins Bim and Noxa (Fig. 2A,B).9 The naphthyl moiety of (17)
occupies the leucine ‘hot spot’ in the MCL-1 groove. This is the sub-
site of the groove which is occupied by the highly-conserved leu-
cine residue of the pro-apoptotic peptides (Bim, Noxa, Puma)
the fragment hits. Thus, as a secondary approach, we focused on
elucidating the binding mode for the aryl sulfonamide compound
via nuclear Overhauser effect (NOE) driven docking of this frag-
ment onto the crystal structure of human MCL-1. In order to facil-
itate this docking, the backbone resonances of MCL-1 were
assigned using the standard suite of triple-resonance experi-
ments.7 These backbone assignments were then extended to
approximately 80% of the sidechain resonances, including many
in the peptide binding groove of MCL-1.7 Intermolecular NOEs, ob-
served in a 3D-13C-edited,12C-filtered NOESY spectrum8 were then
used to dock compound (1) into the binding groove of MCL-1, des
Bim peptide.9 This NOE-derived model is shown in Figure 1.
Based on the binding mode of our NMR-derived model, two
positions on the scaffold were chosen for synthetic elaboration,
each of which provided a vector for key interactions with the
protein. First, modification of the para position of the A-ring was
explored to further probe the hydrophobic pocket of the Bim-pep-
Table 2
Elaboration of aryl sulfonamide core (part 1)
No.
10
X
Y
IC50
3.4
(lM)
11
37
N
12
13
4.9
1.8
A
B
14
15
1.0
0.4
Figure 1. NOE-derived model of (1) bound to MCL-1.

1486
A. M. Petros et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1484–1488
Table 3
Elaboration of aryl sulfonamide core (Part 2)
No.
16
X
Y
IC50
1.2
(lM)
O
17
18
19
20
21
22
23
0.5
nPentyl
0.3
0.2
N
O
Figure 3. Model of (2) bound to MCL-1.
H
N
H
N
In contrast to the NOE-guided docking that was utilized for the
aryl sulfonamide fragment, the salicylic acid fragment was simply
docked into the groove with the single restraint consisting of an
electrostatic interaction between the acid moiety of the fragment
and Arg 263 of the protein (Fig. 3). From the resultant model, the
4 position appeared promising for further elaboration. To probe
this hypothesis, 4-bromo methylsalicylate (25) was reacted with
a small library of boronic esters (Scheme 2) to yield biaryl ester
intermediates, which were subsequently saponified to yield the
target compounds (27). The resultant compounds, with IC50 values,
are shown in Tables 4 and 5.
A cocrystal structure of salicylate 36 bound to MCL-1 was
subsequently generated (PDB ID 4OQ6). As predicted from the
initial model, the phenyl ring of the salicylate moiety occupies
the leucine sub-pocket of the groove while the carboxylate forms
a salt bridge with Arg263 of the protein (Fig. 4A,B). The distal
phenyl ring occupies the P1 position (an isoleucine in the Bim
peptide) while the n-propyl moiety extends even further down
the groove making Van der Waals contact with the sidechains of
Leu 235 and Val249 (Fig. 4C).
0.15
0.11
0.08
0.06
O
O
N
O
N
O
O
N
N
O
N
24
0.03
upon binding to MCL-1. However, in contrast to what is observed
for the binding of the pro-apoptotic peptides to MCL-1, the
naphthyl moiety penetrates very deeply into the groove (Fig. 2C).
This is facilitated by an outward movement of helix 4 which opens
up a hydrophobic pocket ideally sized to accommodate this moi-
ety. The sulfonamide forms a relatively tight turn which serves
to orient the bi-aryl ether portion along the groove and puts the
distal phenyl ring in the same position as occupied by an isoleucine
of the Bim peptide (Fig. 2C).
O
OH
OH
O
R
O
O
O
OH
OH
b
a
R = Br
R
R
25
26
27
Scheme 2. Synthesis of salicylate library: (a) Pd(dppf)-CHCl3-adduct, 1,4-dioxane,
2 N Cs2CO3(aq), boronate ester; (b) 1 N NaOH, 1,4-dioxane.
A
B
C
P1
P3
P2
P4
Figure 2. Crystal structure of (17) bound to MCL-1. (A) Protein backbone shown in ribbon representation. (B) Protein shown with solvent-accessible surface. Oxygen atoms in
red, nitrogen atoms in blue, sulfur atoms in yellow and carbon atoms in orange. (C) Side view of (17) bound to MCL-1 overlaid with the bound Bim peptide.

A. M. Petros et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1484–1488
1487
M)
Table 4
Table 5
Elaboration of salycilate core (part 1)
Elaboration of salycilate core (part 2)
No.
R
IC50
190
(
lM)
No.
R
IC50
>25
(l
28
NH2
O
39
HN
S
29
30
73
45
O
O
HN
S
40
41
>25
7.2
O
O
31
32
34
31
O
O
N
33
34
22
22
42
43
44
0.78
0.67
0.57
Cl
35
36
6.2
3.3
37
38
2.2
1.1
Bn
O
O
O
O
F
CF3
In conclusion, an NMR-based fragment screen against the
anti-apoptotic protein MCL-1 yielded several novel and efficient
cores, two of which were chosen for synthetic elaboration. The
elaboration was based on models of how these compounds bind
into the hydrophobic groove of MCL-1 and were constructed
both with, in the case of the biphenyl sulfonamide, and without,
in the case of the salicylate, NMR-derived NOE restraints. The
elaboration of the biphenyl sulfonamide core led to a 167-fold
gain in potency while elaboration of the salicylate core led to
a 280-fold gain in potency. The resulting compounds are poised
for further elaboration and thus serve as attractive leads for a
drug discovery effort.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2014.02.010.
A
B
C
P1 P2 P3
P4
Figure 4. Crystal structure of (36) bound to MCL-1. (A) Protein backbone shown in ribbon representation. (B) Protein shown with solvent-accessible surface. Oxygen atoms in
red, nitrogen atoms in blue, sulfur atoms in yellow and carbon atoms in orange. (C) Side view of (36) bound to MCL-1 overlaid with the bound Bim peptide.

1488
A. M. Petros et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1484–1488
4. Ackler, S.; Mitten, M. J.; Chen, J.; Clarin, J.; Foster, K.; Jin, S.; Phillips, D. C.;
References and notes
Schlessinger, S.; Wang, B.; Leverson, J. D.; Boghaert, E. R. Br. J. Pharmacol. 2012,
167, 881.
5. Tahir, S. K.; Yang, X.; Anderson, M. G.; Morgan-Lappe, S. E.; Sarthy, A. V.; Chen, J.;
Warner, R. B.; Ng, S. C.; Fesik, S. W.; Elmore, S. W.; Rosenberg, S. H.; Tse, C. Cancer
Res. 2007, 67, 1176.
6. Petros, A. M.; Dinges, J.; Augeri, D. J.; Baumeister, S. A.; Betebenner, D. A.; Bures,
M. G.; Elmore, S. W.; Hajduk, P. J.; Joseph, M. K.; Landis, S. K.; Nettesheim, D. G.;
Rosenberg, S. H.; Shen, W.; Thomas, S.; Wang, X.; Zanze, I.; Zhang, H.; Fesik, S. W.
J. Med. Chem. 2006, 49, 656.
1. Hanahan, D.; Weinberg, R. A. Cell 2000, 100, 5770.
2. Tse, C.; Shoemaker, A. R.; Adickes, J.; Anderson, M. G.; Chen, J.; Jin, S.; Johnson,
E. F.; Marsh, K. C.; Mitten, M. J.; Nimmer, P.; Roberts, L.; Tahir, S. K.; Xiao, Y.;
Yang, X.; Zhang, H.; Fesik, S.; Rosenberg, S. H.; Elmore, S. W. Cancer Res. 2008,
68, 3421.
3. Souers, A. J.; Leverson, J. D.; Boghaert, E. R.; Ackler, S. L.; Catron, N. D.; Chen, J.;
Dayton, B. D.; Ding, H.; Enschede, S. H.; Fairbrother, W. J.; Huang, D. C.;
Hymowitz, S. G.; Jin, S.; Khaw, S. L.; Kovar, P. J.; Lam, L. T.; Lee, J.; Maecker, H. L.;
Marsh, K. C.; Mason, K. D.; Mitten, M. J.; Nimmer, P. M.; Oleksijew, A.; Park, C. H.;
Park, C. M.; Phillips, D. C.; Roberts, A. W.; Sampath, D.; Seymour, J. F.; Smith, M.
L.; Sullivan, G. M.; Tahir, S. K.; Tse, C.; Wendt, M. D.; Xiao, Y.; Xue, J. C.; Zhang, H.;
Humerickhouse, R. A.; Rosenberg, S. H.; Elmore, S. W. Nat. Med. 2013, 19,
202.
7. Clore, G. M.; Gronenborn, A. M. Methods Enzymol. 1994, 239, 349.
8. Petros, A. M.; Fesik, S. W. Methods Enzymol. 1994, 239, 717.
9. (a) Czabotar, P. E.; Lee, E. F.; van Delft, M. F.; Day, C. L.; Smith, B. J.; Huang, D. C.;
Fairlie, W. D.; Hinds, M. G.; Colman, P. M. Proc. Natl. Acad. Sci. 2007, 104, 6217;
(b) Fire, E.; Gulla, S. V.; Grant, R. A.; Keating, A. E. Protein Sci. 2010, 19, 507.