Identification and Optimization of Novel Small c-Abl Kinase Activators Using Fragment and HTS Methodologies

Article abstract of DOI:10.1021/acs.jmedchem.8b01872

Abelson kinase (c-Abl) is a ubiquitously expressed, nonreceptor tyrosine kinase which plays a key role in cell differentiation and survival. It was hypothesized that transient activation of c-Abl kinase via displacement of the N-terminal autoinhibitory "myristoyl latch", may lead to an increased hematopoietic stem cell differentiation. This would increase the numbers of circulating neutrophils and so be an effective treatment for chemotherapy-induced neutropenia. This paper describes the discovery and optimization of a thiazole series of novel small molecule c-Abl activators, initially identified by a high throughput screening. Subsequently, a scaffold-hop, which exploited the improved physicochemical properties of a dihydropyrazole analogue, identified through fragment screening, delivered potent, soluble, cell-active c-Abl activators, which demonstrated the intracellular activation of c-Abl in vivo.

Full text of DOI:10.1021/acs.jmedchem.8b01872

Subscriber access provided by EKU Libraries
Article
Identification and Optimization of Novel Small c-Abl
Kinase Activators using Fragment and HTS Methodologies
Graham L Simpson, Sophie Marie Bertrand, Jennifer A Borthwick, Nino Campobasso,
Julien Chabanet, Julia Coggins, Joshua E. Cottom, Helen C Dawson, Helen Evans, Andrew
N Hobbs, Xuan Hong, Biju Mangatt, Jordi Munoz-Muriedas, Allen L Oliff, Donghui Qin,
Paul T Scott-Stevens, Paris Ward, Yoshiaki Washio, Jingsong Yang, and Robert J Young
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01872 • Publication Date (Web): 28 Jan 2019
Downloaded from http://pubs.acs.org on January 28, 2019
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 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 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.
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 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Identification and Optimization of Novel Small
c-Abl Kinase Activators using Fragment and
HTS Methodologies
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
§
┴
§
§
Graham L. Simpson , Sophie M. Bertrand* , Jennifer A. Borthwick , Nino
‡
§
†
§
‡
Campobasso , Julien Chabanet , Susan Chen , Julia Coggins , Josh Cottom , Siegfried
†
§
§ǁ
§
‡
B. Christensen , Helen C. Dawson , Helen L. Evans , Andrew N. Hobbs , Xuan Hong ,
‡
§
‡
§
Biju Mangatt , Jordi Munoz-Muriedas , Allen Oliff , Donghui Qin , Paul Scott-
§
‡
§
‡
§
Stevens , Paris Ward , Yoshiaki Washio , Jingsong Yang , Robert J. Young .
§GlaxoSmithKline R&D, Medicines Research Centre, Gunnels Wood Road, Stevenage,
Hertfordshire, SG1 2NY, UK.
‡GlaxoSmithKline R&D, 1250 South Collegeville Road, Collegeville, PA 19426, US.
†Author deceased.
ABSTRACT
Abelson kinase (c-Abl) is a ubiquitously expressed, non-receptor tyrosine kinase which
plays a key role in cell differentiation and survival. It was hypothesised that transient
activation of c-Abl kinase via displacement of the N-terminal autoinhibitory “myristoyl
latch,” may lead to increased hematopoietic stem cell differentiation. This would
increase the numbers of circulating neutrophils and so be an effective treatment for
chemotherapy-induced neutropenia. This paper describes the discovery and
optimisation of a thiazole series of novel small-molecule c-Abl activators, initially
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 53
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
identified by high throughput screening (HTS). Subsequently, a scaffold-hop, which
exploited the improved physicochemical properties of a dihydropyrazole analogue
identified through fragment screening, delivered potent, soluble, cell-active c-Abl
activators, which demonstrated intracellular activation of c-Abl in vivo.
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
INTRODUCTION
Kinases are key nodes in signal transduction pathways, modulating the phosphorylation
1
of pertinent hydroxyl residues on target proteins. Their activity is tightly regulated
in vivo by a combination of mechanisms, including phosphorylation, sub-cellular
2
,3
trafficking and conformational control. Whilst considerable research efforts have led
4
to the discovery of small molecule kinase inhibitors (over 28 approved), lesser
progress has been made with discovery of small molecule activators of protein kinases.
This mechanism has been shown to be able to modulate activity in several biologically
relevant targets, leading to increased downstream signalling and potentially
therapeutically useful pharmacological modulation.5
Abelson kinase (c-Abl) is a ubiquitously expressed, non-receptor tyrosine kinase which
6
plays a key role in cell differentiation and survival. It is well known that patients with
ACS Paragon Plus Environment

Page 3 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the Philadelphia chromosome (Ph1) express an aberrant, constitutively active, fusion
protein of Abl, BCR-Abl, which results in uncontrolled neutrophil proliferation, the
clinical symptoms of chronic myelogenous leukaemia (CML).7, 8 These patients can be
treated effectively by the ATP-competitive kinase inhibitors Imatinib, Nilotinib or
Dasatinib (Figure 1).9,10
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
N
CH3
N
H
Cl O
NH
CH3
N
N
N
CH3
N
O
S
N
H C
3
N
H
N
N
NH
N
OH
N
Imatinib
Dasatinib
BCR-Abl/c-Abl inhibitor
binds to ATP-pocket
BCR-Abl/c-Abl inhibitor
binds to ATP-pocket
O
NH
HN
F
Cl
O
F
O
N
O
HN
N
F
N
N
H
N
N
OH
GNF5/ABL001
DPH (1)
BCR-Abl/c-Abl inhibitor
binds to allosteric
c-Abl activator
binds to allosteric
myristoyl pocket
myristoyl pocket
Figure 1. Published c-Abl kinase inhibitors and activators – (a) ATP-competitive inhibitors Imatinib and
Dasatinib; (b) allosteric inhibitor GNF5/ABL001; (c) allosteric activator DPH (1).
c-Abl (1b isoform) is mostly present in an autoinhibited, compact, inactive,
conformation maintained by binding of the N-terminal 1-82 residues “myristoyl latch”
to a lipid pocket in the kinase domain (Figure 2).11,12 Allosteric inhibitors of c-Abl
kinase,13 also active against BCR-Abl, GNF-2,14 GNF-5/ABL001,15,16 have been
described which bind at this myristoyl pocket, stabilising interaction with the kinase
domain helix αI’ and mimicking the autoinhibited conformation of c-Abl. (Figure 1).
Concurrently, our group have previously described the discovery of a small molecule
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 53
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
activator of c-Abl kinase, DPH (1), which binds at the same pocket, displacing the
1
7
myristoyl latch and leading to cellular activation (Figure 1). Activators of c-Abl
kinase have potential therapeutic applications in chemotherapy-induced
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
neutropenia,18,19 breast and prostate cancer and ischemic injury. Interestingly,
20
21
22
combination of c-Abl activator, DPH (1), and ATP-competitive BCR-Abl inhibitor
Imatinib, has been shown to be synergistic, hypersensitising BCR-Abl1 leukaemia cells
in CML.23
Activators of c-Abl were identified within GSK by high throughput screening and the
1
7
biology around small molecule activator DPH (1) was previously reported. This paper
describes the discovery of a series of pyrazoline activators of c-Abl kinase and their
medicinal chemistry optimisation as tool molecules for further in vivo investigations.
Figure 2. Activation of c-Abl kinase by relieving inactive, autoinhibited conformation. In the
autoinhibited form the tyrosine residues (Tyr245 and Tyr412) are inaccessible to phosphorylation. Upon
displacement of the N-terminal myristoyl latch from the hydrophobic pocket in the kinase domain, a
conformational change occurs allowing autophosphorylation of these residues and increased kinase
activation. (Abbreviations: KD=kinase domain; SH2/SH3=Src-homology domain 2/3; Y=tyrosine;
pY=phosphotyrosine; ATP=adenosine triphosphate).
ACS Paragon Plus Environment

Page 5 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
RESULTS AND DISCUSSION
Aminothiazoles 2 and 3 (Figure 3) were identified by high throughput screening, which
utilised an assay based on immobilized metal affinity for phosphochemicals (IMAP),24
configured to find molecules which increase the activation of c-Abl tyrosine kinase.
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
2
4
This was used to screen a diverse set of 1.3 million compounds. Additionally, a
complementary fluorescence polarisation (FP) assay monitored competitive
displacement of a fluorescent-tagged myristoylated peptide from the kinase-dead
mutant enzyme, N-terminal truncated c-Abl46-534(D382N) which was used for follow-
up screening. This peptide is known to bind into the kinase domain and its displacement
confirmed the allosteric mechanism of kinase activation. Cellular activity was assessed
by Western Blot detection of the phosphorylation of full-length c-Abl (pY245, pY412)
from HEK-MSRII cell lysates.24 In common with the previously disclosed c-Abl
activator, DPH (1), compounds in this paper were shown to be myristoyl-competitive
activators of full-length, autoinhibited c-Abl enzyme (c-Abl-1b1-531) in vitro (FP and
17
2
4
IMAP assays). The initial hits also displayed some cellular activation, although at a
low level which required optimisation. The key objectives of the medicinal chemistry
programme focussed on exploring the structure-activity relationships around the
template and improving the intracellular activation.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 53
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
O
Cl
Cl
NH
HN
O
Cl
Cl
N
F
HN
O
N
N
N
S
HN
O
S
N N
N
S
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
DPH (1)
FP Binding pIC : 6.4
IMAP Enzyme Act pEC : 6.6
2
3
FP Binding pIC : 6.5
FP Binding pIC : 6.1
50
50
50
IMAP Enzyme Act pEC : 5.6
IMAP Enzyme Act pEC : 5.0
50
50
50
Cell Act pEC : 6.1
Cell Act pEC : 4.5
Cell Act pEC : 4.7
50
50
50
Figure 3. Structure and c-Abl activation data for aminothiazoles 2 and 3 with comparison to previously
published pyrazole activator DPH (1).9
Initially, the SAR of amide substituents on the left-hand side of the molecule was
delineated; by maintaining the dichlorophenyl group identified in the hits 2 and 3. These
changes resulted in increased enzyme binding affinity and activation (Table 1 - 4-10).
As is often seen with small molecule-ligand interactions, increasing lipophilicity led to
an increase in potency (e.g. 4). Modification of the amide to a sulphonamide (10) led
to a reduction in activation. However, introduction of hydrogen-bond acceptors, was
favoured (e.g. 6-9). In general, the compounds showed poor cellular activation, with 5
and 6 showing the highest cellular activity of any of the compounds tested. Most of the
compounds displayed reasonable permeability in an artificial membrane assay (AMP)
but they also generally had poor physical properties, indicated by their low solubilities,
which likely contributed to their relatively poor cellular activity.
ACS Paragon Plus Environment

Page 7 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 1. Amide Modification on the Thiazole Series
Cl
Cl
R
HN
N
S
IMAPa
FPb
Cellularc
activation
Enzyme
activation
Myristoyl
binding
CLNDd
(µM)
AMPe
(nm/sec)
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
Cmpd
R
(pEC50
)
(pIC50
)
(pIC50
)
1f
2
4
5
6
7
-
6.6
6.4
6.1
>395
<1
3
350
196
227
<30
142
61
O
5.6
6.8
6.6
5.9
6.3
7.5
6.5
6.6
6.5
6.3
6.7
6.7
4.5
<4.3
5.3
O
F
O
5
F
O
N
N
5.3
57
2
O
<4.3
4.8
S
N
O2N
N
O
8
9
<1
260
O
6
.4
6.7
<4.3
<4.3
20
370
N
O
O
S
1
0
4.6
<4.3
429
220
a
b
n numbers ≥2. Assays performed using an end-point activation IMAP protocol; Fluorescence
c
Polarization (FP) competition binding assay; In-cell Western (ICW) assay using Bacmam transduced
HEK-MSRII cells.17 dCLND solubility refers to kinetic solubility measured using chemiluminescent
e
nitrogen detection - >100 ± 30 µM is defined as good solubility. AMP refers to artificial membrane
permeability measured using black-lipid membrane - >200 ± 30 nm/sec is defined as good permeability
f
17
(See Supporting Information). data have been reported previously.
To complement the knowledge gained from the chemical variation in the series, crystal
1
4
structures of allosteric inhibitor GNF-2 and activator 6, soaked into crystals of human
c-Abl kinase domain, were obtained. There were two molecules in the asymmetric unit
and both myristoyl pockets were fully occupied. As shown previously in the mouse
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 53
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
form of c-Abl,14,16 the αI’-helix was bent down towards the GNF-2 molecule (Figure 4).
In these structures, only one of the molecules in the asymmetric unit showed the αI’-
helix to be bent with GNF-2 whereas the other molecule had the αI’-helix involved in
a crystal contact and could not bend to interact with GNF-2. The crystal structure of
activator 6 showed that the molecule bound to the myristoyl pocket and pushed the αI’-
helix into an extended conformation away from the myristoyl pocket. There was also a
key 2.8Å hydrogen-bond formed with the backbone Ala452, at the C-terminus of αF-
helix, to Ala462 and to the guanidine of Arg351. The hydrogen-bond with Arg351 was
absent with DPH (1) highlighting the more efficient polar interactions in this
compound.17 The dichlorophenyl group showed good shape complementarity in a
lipophilic pocket defined by residues Leu359, Leu360, Ala363, Val487, Leu448, and
Phe512. To accommodate the larger dichlorophenyl ring, Leu360 and Phe512 adopted
alternative conformations compared to the bound structure of DPH (1). The
aminothiazole ring occupied the entrance area at the back of the myristoyl pocket,
sandwiched between the N-term of αE-helix and loop Pro380-Pro384 with no
observable hydrogen-bond interactions. The cyanomethyl picked up a strong hydrogen-
bond with the guanidine of Arg351, a common feature of inhibitors in this series. The
αI-helix, thought to be essential to maintain c-Abl in its inactive form was clearly
displaced. Superposition of aminothiazole 6 and pyrazole DPH (1) X-ray crystal
structures showed that the displaced Leu360 in the aminothiazole pushes the αI-helix
further out than in the pyrazole DPH (1) and disrupted the SH3-SH2 packing against
the kinase domain.
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
ACS Paragon Plus Environment

Page 9 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
a)
(b)

E

F
R351

I
L360
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

H

I
I451

I’
F512
1
4
Figure 4. (a) Overlay of c-Abl kinase domain crystal structures of GNF-2 (green – PDB code 3K5V)
and 6 (cyan – PDB code 6NPE) showing the difference between the positions of the αI’-helix. (b)
Overlay of X-ray crystal structure of 6 c-Abl (kinase domain) with DPH (1) (pink - PDB code 3PYY).
Leu360 and Phe512 are rendered as sticks and highlight the difference between respective lipophilic
pockets and the shift in αI-helix.
Next, the SAR around the right-hand side of the molecule, the 3,4-dichlorophenyl
functionality of 2, was investigated, through synthesis of thiazole analogues wherein
the left-hand side N-acetamide motif was maintained. The biochemical assay data and
physical measurement of these compounds are summarised in Table 2, wherein the 3,4-
dichlorophenyl group in 2 was found to be optimal for activity, as simple changes (H
1
1 and 12, F 13 and 14, Me 15) resulted in significant loss of activation. Other
modifications designed to improve solubility through the introduction of polar groups,
such as a hydroxyl in 16, were also not tolerated. Similarly, replacement of the phenyl
ring with heterocycles (17-19) gave a reduction in potency even though solubility was
increased in some cases (e.g. 19). Interestingly, compound 15 demonstrated a marked
difference between its binding and enzyme activation, suggesting that the modification
may possibly have led to a crossover between the activation and inhibition.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 53
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 2. Dichlorophenyl Moiety SAR
O
N
R
HN
S
IMAPa
FPb
Cellularc
activation
Enzyme
activation
Myristoyl
binding
CLNDd
(µM)
AMPe
(nm/sec)
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
Cmpd
R
(pEC50
)
(
pIC50
)
(pIC50
)
Cl
2
5.6
6.5
4.5
<1
1
196
250
340
350
94
Cl
Cl
11
12
4.5
5.1
5.0
<4.3
6.1
ND
ND
2
Cl
Cl
1
1
1
1
3
4
5
6
5.5
4.3*
<4.3*
<4.3
ND
5
F
F
4.4
5.6
10
1
Cl
Cl
4.5
6.1
140
270
280
210
250
<4.3
<4.0
4.9
4.9
103
1
OH
Cl
17
18
5.2
<4.3
<4.3
ND
Cl
Br
S
5.3
16
186
S
N
1
9
N
<4.0
<4.3
O
a
n numbers ≥2 unless stated *n=1. Assays performed using an end-point activation IMAP protocol;
b
c
Fluorescence Polarization (FP) competition binding assay; In-cell Western (ICW) assay using Bacmam
transduced HEK-MSRII cells.17 dCLND solubility refers to kinetic solubility measured using
e
chemiluminescent nitrogen detection - >100 ± 30 µM is defined as good solubility. AMP refers to
artificial membrane permeability measured using black-lipid membrane - >200 ± 30 nm/sec is defined
as good permeability.
ACS Paragon Plus Environment

Page 11 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
From this SAR exploration, it was concluded that the dichlorophenyl group was likely
to be close to optimal, so it was decided to focus future efforts on the core of the
molecule. Thus, a range of 5-membered and 6-membered rings (20-27), which were
predicted to place the substituents along similar vectors to the aminothiazole template,
was investigated (Table 3).
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
Table 3. Thiazole Core Modification
IMAPa
Enzyme
activation
FPb
Myristoyl
binding
Cellularc
activation
CLND
(µM)
AMP
(nm/sec)
Cmpd
6
Structure
(pEC50
)
(pIC50
)
(pIC50
)
Cl
Cl
H
N
N
5.9
6.3
5.3
57
10
2
142
120
350
190
310
280
310
230
215
Cl
Cl
Cl
N
N
N
O
O
O
S
H
N
2
0
4.8
5.8
5.5
5.8
5.7
5.6
5.5
4.9
5.4
6.9
5.4*
6.2
5.6
5.8
6.0
5.8
<4.3
4.8
S
Cl
Cl
H
N
N
21
S
H
N
2
2
2
2
2
3
4
5
4.7
12
14
69
34
370
1
Cl
Cl
Cl
N
N
N
O
O
O
S
Cl
Cl
H
N
N
N
5.5
HN
H
N
4.8
N
S
Cl
Cl
H
N
N
5.1
Cl
Cl
N
N
N
O
O
H
N
N
26
27
5.0
S
N
O
N
Cl
Cl
N
N
<4.3
H
a
n numbers ≥2 unless stated *n=1. Assays performed using an end-point activation IMAP protocol;
b
c
Fluorescence Polarization (FP) competition binding assay; In-cell Western (ICW) assay using Bacmam
transduced HEK-MSRII cells.17 dCLND solubility refers to kinetic solubility measured using
e
chemiluminescent nitrogen detection - >100 ± 30 µM is defined as good solubility. AMP refers to
artificial membrane permeability measured using black-lipid membrane - >200 ± 30 nm/sec is defined
as good permeability.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 53
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
Disappointingly, the cellular activity and, with the exception of 1,2,4-thiadiazole 26,
solubility of the analogues were not improved across the board. At this point, it was
apparent that the chemistry strategy was largely pursuing compounds with rather poor
physical properties, especially low solubility (generally < 20 µM) associated with
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
relatively high lipophilicity and proportion of sp -hybridised heavy atoms. In Table 2
it was apparent that the dichlorophenyl motif was likely to be optimal and in Table 3
(compounds 20-27) the isosteric replacements for the 2-aminothiazole all maintained
the planarity of the molecule and resulted in generally poor translation of intrinsic
potency into cellular potency. These observations were consistent with the growing
body of evidence indicating the consequences of working with overtly lipophilic and/or
2
high sp /high aromatic ring count molecules, which are poorly soluble, may have poor
permeation and other developability risks. Lipophilicity is the antithesis of aqueous
solubility and is associated with non-specific entropically-driven binding.25,26 It has
been shown that in many developability assays aromatic ring count per se or the
2
proportion of sp atoms is indicative of similar risks and that, the former count has an
additive risk with lipophilicity values in many assays.27
Almost concurrently, synthetic concepts to address the planarity issues were
complemented by the output of a fragment screening initiative, which identified
pyrazoline 28 as a low micromolar activator of c-Abl (Table 4). Gratifyingly,
acetylation and cyanoacetylation to give compounds 29 and 30 showed increased
binding and effective activators were quickly identified. The generation of a X-ray
structure with 29 (PDB code 6NPU) showed a similar binding mode to the
aminothiazole series (Figure 5), but importantly displayed a conformational twist in the
2
core due to the smaller ring and the reduced sp character. Such modifications led to a
significant improvement in melting point (Table S1), solubility and cell potency, with
ACS Paragon Plus Environment

Page 13 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
28
much improved kinetic solubility. Reduction in aromaticity of drug molecules has
been shown to improve physicochemical properties such as solubility and has been
associated with a lower incidence of attrition during drug development.27,29,30 In the
pyrazoline series the improved physical properties were manifested through enhanced
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
2
solubility and permeability compared with the more sp -rich analogues, which led to a
lower drop-off in potency between the biochemical and cell activation assay. The
aminopyrazoline core was relatively stable to oxidation to the pyrazole with no
decomposition observed during storage in solution or as solid 28.31,32 Matched pair data
also showed that the pyrazole analogues had lower activities (e.g. 25 vs 30).
Table 4. Pyrazoline Series
IMAPa
Enzyme
activation
FPb
Myristoyl
binding
Cellularc
activation
CLNDd
(µM)
AMPe
(nm/sec)
Cmpd
28
Structure
(pEC50
)
(
pIC50
)
(pIC50
)
Cl
H2N
N
N
5.2
5.7
4.9
69
31
320
360
320
Cl
Cl
H
N
N
N
2
3
9
0
6.5
6.6
6.4
6.4
6.1
6.0
Cl
O
Cl
H
N
N
202
Cl
N
N
O
a
b
n numbers ≥2. Assays performed using an end-point activation IMAP protocol; Fluorescence
c
Polarization (FP) competition binding assay; In-cell Western (ICW) assay using Bacmam transduced
HEK-MSRII cells.17 dCLND solubility refers to kinetic solubility measured using chemiluminescent
e
nitrogen detection - >100 ± 30 µM is defined as good solubility. AMP refers to artificial membrane
permeability measured using black-lipid membrane - >200 ± 30 nm/sec is defined as good permeability.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 53
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
Reduced aromaticity
enhanced developability (lower PFI* score)
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
Ab-initio calculation
predicted pyramidal N
3
Sp centre to introduce shape
and break crystal packing
3
0
=
PFI Property Forecast Index
*
Figure 5. Aminopyrazoline series (29, left – PDB code 6NPU) has lower lipophilicity and increased
cellular activation compared to aminothiazoles (6, right – PDB code 6NPE). H-bond to backbone Ala452
is indicated by dashed lined and annotated with distance (Å).
Optimisation of the pyrazoline series. The similarity in binding mode with the
aminothiazole series allowed a rapid transfer of SAR. A range of substituted aromatic
and alkyl amides were well tolerated with the pyrazoline core (Table 5). Acetamide 29
demonstrated good affinity and efficacy in the enzyme assays although it displayed
poor aqueous solubility. To further improve the molecule’s overall physicochemical
profile, polar functional groups were introduced on the amide. The introduction of a
hydrogen-bond acceptor to interact with Arg351 was expected to improve potency and
solubility. Cyanoacetamide 30 achieved excellent solubility with a similar level of
potency to the acetamide. An acyclic acetamide analogue 31 (only S-enantiomer tested)
ACS Paragon Plus Environment

Page 15 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
only demonstrated moderate potency, in contrast to a cyclic lactam analogue 33 (tested
as racemate), which achieved an excellent potency in the enzyme efficacy assay. It was
postulated that the limited rotation of the cyclic amide constrained the vector for
hydrogen bonding towards Arg351. Solubility may have been improved due to
decreased propensity to form infinite ladders of intermolecular hydrogen-bond
interactions between cis-amides in the solid state.16
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
Aromatic amide derivatives were explored in more detail. The 3- and 4-pyridyl amide
analogues, 34 and 35, showed significantly improved activation in the enzyme assay
compared to isoelectronic 2-pyridyl amide, 36. Further exploration of this result
suggested that introduction of nitrogen atoms at the 2-position of the ring reduced
potency even if the 3- or 4-position nitrogen was retained (37, 38). These results were
reinforced with the pyridazine 39 and pyrimidine 40. A CH in the 2- and 6-position of
the ring was expected to cause a slightly twisted conformation to avoid steric clash
between the amide NH and aromatic CH, enhancing the hydrogen-bond with Arg351
as was seen with the difluorophenyl amide 5 in the aminothiazole series. In addition,
an intramolecular hydrogen-bond between the 2-pyridyl nitrogen and the amide was
predicted to disrupt the hydrogen-bond of the amide NH to the enzyme pocket, thereby
introducing a planar conformation with a reduction in cell activity. The retention of
enzyme potency with 2,6-difluorophenyl derivative 41 would suggest a significant
contribution of a conformational preference, although a slightly lower cell potency was
observed. Based on these results, the 2-fluoro-3-pyridyl (42), 3-fluoro-4-pyridyl (43)
and 4-methyl-5-pyrimidyl (44) analogues were synthesised and 44 was found to have
the highest potency to date (pEC = 7.4).
50
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 53
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 5. Amide Modification on the Pyrazoline Series
Cl
Cl
H
N
N
N
R
IMAPa
FPb
Cellularc
activation
CLNDd
(µM)
AMPe
(nm/sec)
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
Enzyme
activation
Myristoyl
binding
Cmpd
R
(pEC50
)
(pIC50
)
(pIC50
)
O
2
3
3
3
9
0
1
2
6.5
6.4
6.1
31
202
58
39
416
4
360
320
340
355
250
305
303
350
ND
320
360
350
ND
570
280
340
O
N
6.6
5.8
6.6
6.4
6.8
6.7
5.7
5.3
4.8
6.6
6.6
6.2
6.9
6.9
7.4
6.4
5.9
6.5
6.3
6.6
6.5
5.7
6.1
6.0
6.3
6.3
6.4
6.4
6.7
6.8
6.0
5.1
6.1
5.6
6.2
5.8
O
H
N
O
O
O
N
H
N
O
33
O
3
3
3
3
4
5
6
7
N
O
O
O
O
9
N
N
<4.3
11
67
13
23
8
N
N
<4.3*
N
N
38
4.4*
6.3
6.2
5.5
5.9
5.8
6.5
O
O
3
4
4
4
4
9
0
1
2
3
N
N
N
N
F
O
ND
13
36
12
F
F
F
O
N
O
O
N
N
44
N
ACS Paragon Plus Environment

Page 17 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
n numbers ≥2 unless stated *n=1. Assays performed using an end-point activation IMAP protocol;
b
c
Fluorescence Polarization (FP) competition binding assay; In-cell Western (ICW) assay using Bacmam
transduced HEK-MSRII cells.17 dCLND solubility refers to kinetic solubility measured using
e
chemiluminescent nitrogen detection - >100 ± 30 µM is defined as good solubility. AMP refers to
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
artificial membrane permeability measured using black-lipid membrane - >200 ± 30 nm/sec is defined
as good permeability.
Based on computational modelling (multi-component solvent search, MCSS) it was
proposed to exploit a small hydrophobic pocket from the substituent positions on the
33
pyrazoline ring. To probe this binding site, substitution of the pyrazoline ring with a
methyl group was investigated. Four possible methyl substituted isomers at the 4- or 5-
position of the pyrazoline ring were prepared (racemate 45, (R)-5-Me 48, (S)-5-Me 49,
(R)-4-Me 46, (S)-4-Me 47). As predicted by molecular modelling, only the (S)-4-Me
analogue 47 demonstrated a small improvement in potency (pIC = 6.9). To attempt to
5
0
introduce further hydrogen-bond interactions, the backbone NH of Ala356 was targeted
by building from the 4-(S)-methyl group. Calculation of the distance between NH and
CH suggested a hydroxyethyl group at the 4-position could make a positive hydrogen
3
bonding interaction whilst minimising the increase of lipophilicity. Hydroxyethyl
analogues, 50 and 51 both demonstrated a similar level of potency in the enzyme assay
as the racemate with significantly improved solubility but unfortunately led to lower
cellular activation (Table 6). The X-ray crystallography of 51 confirmed that the
terminal OH successfully made a hydrogen-bond interaction with backbone NH via a
water molecule (Figure 6). Although in vitro enzyme activation had been maintained it
did not translate to the cellular assay. Combination of the optimal (S)-methylated core
with a preferred pyrimidine amide group afforded 52 with an enzyme potency of
pEC = 7.7 and a cellular efficacy of pEC = 7.0, representing the most active
5
0
50
compound synthesised.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 53
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 6. Exploration of Pyrazoline Core Substitution
IMAPa
FPb
Cellularc
activation
Enzyme
activation
Myristoyl
binding
(pIC50
CLNDd
(µM)
AMPe
(nm/sec)
Cmpd
29
Structure
(pEC50
)
(pIC50
)
)
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
Cl
H
N
N
N
6.5
6.4
6.1
31
360
Cl
O
Cl
H
3
3
4
4
5
5
N
N
N
N
N
6.8
6.7
6.6
6.6
6.5
6.1
6.2
5.8
6.3
4
9
305
303
405
Cl
Cl
N
N
O
O
Cl
H
N
Cl
Cl
Cl
Cl
H
N
N
N
N
N
Cl
49
N
O
O
O
O
H
N
4
4
6
7
Cl
Cl
Cl
5.6
6.9
5.4
6.4
4.7
6.2
36
32
380
420
N
N
N
H
N
H
N
4
4
8
9
5.7
6.3
5.8
6.4
5.0
5.0
173
125
410
420
Cl
H
N
N
Cl
N
O
Cl
H
N
N
N
N
Cl
Cl
N
N
5
0
6.9
6.6
5.1
192
190
O
HO
Cl
H
N
N
N
5
5
1
2
7.0
7.7
6.8
6.5
5.5*
7.0
167
9
335
420
O
HO
N
Cl
H
N
N
Cl
N
O
a
n numbers ≥2 unless stated *n=1. Assays performed using an end-point activation IMAP protocol;
b
c
Fluorescence Polarization (FP) competition binding assay; In-cell Western (ICW) assay using Bacmam
transduced HEK-MSRII cells.17 dCLND solubility refers to kinetic solubility measured using
e
chemiluminescent nitrogen detection - >100 ± 30 µM is defined as good solubility. AMP refers to
artificial membrane permeability measured using black-lipid membrane - >200 ± 30 nm/sec is defined
as good permeability.
ACS Paragon Plus Environment

Page 19 of 53
Journal of Medicinal Chemistry
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 6. Crystal structure of c-Abl kinase domain with 51 showing a bridging water molecule between
hydroxyl group and amide backbone – PDB code 6PNV.
In Vivo c-Abl Activation. In order to evaluate if small molecule c-Abl activators were
able to activate endogenous c-Abl in vivo, comparative studies of three equipotent
compounds in the cellular activation assay (1, 32 and 45) from different series were
1
7
undertaken in mice. The microsomal intrinsic clearance (mL/min/mg protein) was
determined in pooled CD1 mouse liver microsomes and compared to measurement of
compound concentrations in terminal mouse plasma samples at two different timepoints
(40 min and 180 min), following Intraperitoneal injection (100 mg/kg, 10 mL/kg, n=4
mice per compound) of formulated compounds. Concurrently phosphorylation of c-Abl
substrate, Crk-L, in the blood samples was measured as an indication of the level of
c-Abl activation in vivo. Crk-L is a well-characterized cellular substrate of active c-Abl.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 53
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 7. c-Abl Activator mouse plasma concentrations and increase in c-Abl phosphorylation of
substrate Crk-L after
In Vivo Plasma
Fold Increase
phCRKL Vs Total
Crk-L (normalised)
Concentration
a
b
(ng/mL)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In Vitro
Clearance
mL/ min/ mg
protein)
cellular
activation
ICW EC50
c
Cmpd
Structure
40 min
180 min
40 min
180 min
(
(
)
O
NH
HN
O
F
1
±
5100
1150
1050
± 595
1
6.1
0.266
13x
28x
N
N
Cl
H
N
N
5540
± 233
2540
± 143
4
5
2
Cl
6.3
6.1
0.408
1x
5x
N
O
Cl
N
H
N
N
15600
17600
± 3530
Cl
3
N
< 0.01
19x
35x
±
1700
O
a
b
Terminal Plasma concentrations (Mean ± SEM) based on n=4 mice/time point; The phCrk antibody
utilized recognizes pY221 of Crk and pY207 of Crk-L, a Crk isoform and values are quoted as a ratio of
c
pCrk-L to total Crk, normalised to background ratio (phCrk-L/Crk = 0.001); Intrinsic clearance
determined in CD1 mouse liver microsomes (final protein concentration 0.5 mg/mL, drug concentration
0
.5µM, 1 h incubation).
Control compound DPH (1) was metabolised rapidly in vitro. Compound 45 had a
significantly higher turnover in the in vitro microsomal assay (0.408 mL/min/mg
protein) showing a clear metabolic liability from the introduction of the methyl group
into the pyrazoline core. In contrast, compound 32 showed good metabolic stability
over 1 hour in vitro. This trend was borne out in vivo with compound 45 showing only
moderate plasma levels at the 40-minute time-point, whereas compounds 32 and 1
showed ~ 3-fold higher plasma exposure. At the 180-minute time-point, both
compounds 1 and 45 were reduced to low levels (<5000 ng/mL) with compound 32
showing sustained high levels (17600 ng/mL).
ACS Paragon Plus Environment

Page 21 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Gratifyingly the plasma drug levels correlated well with the observed activation of
c-Abl confirming target engagement in vivo. Compounds 1 and 32 both demonstrated
increased activation of c-Abl kinase resulting in higher levels of the product phCrk-L
at both the 40-minute and 180-minute time-points (13-fold to 35-fold over vehicle).
Interestingly, phCrk-L levels were still increased at the 180-minute time point even
with low drug levels of compound 1 (1050 ng/mL), showing the potential for catalytic
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
3
4
autoactivation of c-Abl in vivo. Further investigation of compound 32 in a chronic
dosing in vivo model (oral, 100 mg/kg BID for 10 days) was attempted to understand
the effect of increasing c-Abl activity on neutrophil levels as a potential treatment for
chemotherapy-induced neutropenia. Unfortunately, the study was halted due to
significant levels of haemolysis being observed before the efficacy could be assessed.
Although 3-aminopyrazoline templates have been exemplified as anti-inflammatory
compounds in the literature,35-37 there are reports of a similar compound showing
induction of methemoglobinemia, and haemolytic anaemia in rats at higher doses,38
which may suggest compound rather than mechanism-related toxicity. Further studies
would be required to fully elucidate the mechanism of toxicity.
In this work, biochemical, biophysical and cellular assays were configured to identify
a novel, myristoyl competitive series of activators of c-Abl which led to increased
phosphorylation in cellular systems and in vivo.17,24 X-ray crystallography indicates that
these molecules bind in the myristoyl pocket of c-Abl-1b, displacing the myristoyl tail,
forcing the αI-helix of the C-terminus. This contrasts with discovered allosteric
inhibitors of BCR-Abl, GNF-2 and GNF-5,14,39 which bind in the same myristoyl
pocket but stabilise the inactive conformation of BCR-Abl by forming key H-bonding
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 53
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
interactions with the αI-helix of the C-terminus. This mechanism, first demonstrated in
c-Abl has so far not been shown to operate in other kinases.
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
CONCLUSIONS
In summary, a novel aminopyrazoline series of direct small molecule activators of
autoinhibited c-Abl kinase has been discovered, which may be developed as treatments
for chemotherapy induced neutropenia. SAR was optimised using crystallography to
guide synthesis leading to potent cell-active compounds with a balance of
physicochemical properties. By focussing on physicochemical properties, improved
solubility and cell potency the aminodihydropyrazole template was discovered which
showed favourable binding kinetics at the target and higher cellular potency. The
compounds showed in vivo target engagement and pathway activation although could
not be developed further into oral medicines due to in vivo toxicity. These molecules
may be useful as tools and to develop further as therapeutics for chemotherapy induced
neutropenia.
ACS Paragon Plus Environment

Page 23 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CHEMISTRY
The compounds reported in this paper were prepared by a variety of methods from
commercially available starting materials.
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
Compounds 2 and 4-10 were synthesised following Scheme 1. The 4-(3,4-
dichlorophenyl)thiazol-2-amine intermediate 54 was prepared via the Hantzsch thiazole
4
0
condensation of the bromoketone 53 with thiourea. The amide formation was carried
4
1
out using the in situ generated acid chloride or HATU-mediated conditions to afford
analogues 2 and 4-9 (Scheme 1). The sulphonamide analogue 10 was prepared using
reaction of methylsulfonyl chloride. Isolated yields were low but acceptable for initial
investigation of SAR.
O
R
X
55a-g
Cl
Cl
Cl
Cl
Cl
Cl
a
b, c or d
N
H
N
N
NH2
Br
R
S
S
O
O
53
54
2, 4-9
e
Cl
Cl
N
H
N
S
S
O
O
10
Scheme 1. Amide and Sulphonamide Formations using Aminothiazole Core
Reagents and conditions: (a) thiourea, EtOH, rt, 10 min, 85%; (b) HATU, DIPEA, DMF, 50 °C, 2-65 h,
9-33%; (c) acid chloride, DIPEA, THF, rt, overnight, 60%; (d) (i) (1-chloro-2-methyl-1-propen-1-
yl)dimethylamine, THF, rt, 1 h; (ii) DIPEA, THF, 40 °C, overnight, 9-33%; (e) methylsulfonyl chloride,
DCM, 0 °C to rt, 24 h, 16%.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 53
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
Syntheses of the acetyl analogues (11, 13-14, 16, 18) were secured by the Hantzsch
4
0
thiazole condensation of bromoketone derivatives 56a-e with acetylthiourea 57 to
afford the desired analogues in good yield. Alternatively, Suzuki-Miyaura cross-
coupling of 4-bromo-2-acylaminothiazole 59 with boronic acid 58 using standard
palladium cross-coupling conditions afforded the desired product 15 in modest yield
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
(Scheme 2). Analogues (12, 17, 19) were obtained from commercial suppliers.
S
O
R
R
a
N
H
Br
H N
2
N
N
H
O
S
O
56a-e
57
11, 13-14, 16, 18
Cl
Cl
Br
N
b
H
N
N
H
OH
N
B
S
S
OH
O
O
58
15
59
Scheme 2. Modification of Aryl Thiazole
Reagents and conditions: (a) EtOH, microwave, 100-120 °C, 15-20 min or rt, 2.5 h, 6-74%; (b)
Pd(dppf)Cl , K CO , iPrOH, microwave, 100 °C, 1 h, 25%.
2
2
3
Heterocycles analogues (20-27) were synthesised using standard heterocyclic synthesis
methods (see Supporting information for detailed scheme).
The pyrazoline cores 28, 62 and 65 were assembled by cyclisation of (3,4-
dichlorophenyl)hydrazine 60 with the corresponding substituted acrylonitriles under
basic conditions. Coupling with a range of acid chloride or carboxylic acid derivatives
ACS Paragon Plus Environment

Page 25 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
delivered the substituted pyrazolines 29-49, 52 in acceptable yields for biological
evaluation (Scheme 3-5).
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
R
X
6
1a-p
Cl
Cl
Cl
Cl
Cl
Cl
b or c
a
N
H
NH2
N
N
N
N
N
NH2
R
H
O
60
28
29-44
Scheme 3. Modification of the Amide Substituents on the Pyrazoline Core
Reagents and conditions: (a) acrylonitrile, EtOH, NaOEt, 80 °C, nitrogen atm., 19 h, 66%; (b) acid
chloride, pyridine, DCM, rt, overnight to 2 days, 2-4%; (c) carboxylic acid, HATU, DIPEA, DCM or
DMF, rt, 1 h to 2 days, 1-80%.
Cl
Cl
Cl
Cl
Cl
b
a
N
H
N
Cl
N
NH2
N
N
N
H
NH2
O
60
62
45
chiral separation
c
46-47
Cl
Cl
N
H
N
N
N
N
O
63
chiral separation
2 and 64
5
Scheme 4. Substitution of the Pyrazoline Core
Reagents and conditions: (a) 1. NaOEt, EtOH, 80 °C, 5 min, 2. 2-methyl-2-propenenitrile, 85 °C, 2 h,
2%; (b) acetal chloride, pyridine, DCM, rt, 16 h, 30%; (c) 5-pyrimidinecarboxylic acid, HATU, DIPEA,
DCM, rt, 2 h, 7%.
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 53
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
Cl
Cl
Cl
Cl
Cl
Cl
b
a
N
H
NH2
N
N
N
N
N
NH2
H
O
6
0
65
66
chiral separation
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
4
8-49
Scheme 5. Substitution of the Pyrazoline Core
Reagents and conditions: (a) crotonitrile, EtOH, NaOEt in EtOH (21 w/w%), reflux, overnight, 36%; (b)
acetal chloride, pyridine, DCM, rt, 2 days, sealed vial, 36%.
Compounds 50-51 were synthesised following Scheme 6. Condensation of compound
6
7 with ethyl cyanoacetate gave intermediate 68 in moderate yield. Reduction of the
ester group using lithium borohydride gave the alcohol intermediate 69 in good yield.
Mesylation of the alcohol using methanesulfonyl chloride and triethylamine followed
by elimination using DBU gave the acrylonitrile intermediate 70 in excellent yield. The
aminopyrazoline core 71 was assembled by cyclisation of (3,4-
dichlorophenyl)hydrazine 60 with acrylonitrile 70 under basic conditions. Amide
coupling using HATU with the appropriate carboxylic acid followed by acid-catalysed
deprotection of the THP group gave the final compounds 50 and 51 and good yields.
ACS Paragon Plus Environment

Page 27 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
N
N
a
b
Br
O
O
O
O
O
OH
O
O
O
6
7
68
69
N
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
c
O
O
70
Cl
Cl
Cl
Cl
N
N
H
N
N
NH2
N
Cl
Cl
R
d
e
O
NH2
N
H
O
O
60
O
O
7
1
72-73
Cl
Cl
f
N
H
N
N
R
O
OH
5
0-51
Scheme 6. Substitution of the Pyrazoline Core
Reagents and conditions: (a) ethyl cyanoacetate, DMF, 70 °C, 2 h followed by rt, 3 days, 50%; (b)LiBH
THF, 0 °C then rt, overnight, 75%; (c) 1. methanesulfonyl chloride, Et N, THF, rt, 1 h, 2. DBU, 60 °C,
h followed by rt, 3 h, 96%; (d) 1.NaOEt in EtOH (21 w/w%), EtOH, 80 °C, 5 min, 2. Intermediate 70
4
,
3
1
in EtOH dropwise, reflux, 3 h followed by rt, overnight, 51%; (e) carboxylic acid, HATU, DIPEA, DCM,
40 °C, overnight; (f) acetic acid, THF/water, 40 °C, 2 days, 51-61% over two steps.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 53
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
EXPERIMENTAL SECTION
Chemistry. General. All solvents and reagents, unless otherwise stated, were
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
commercially available from regular suppliers such as Sigma-Aldrich and Fluorochem
1
and were used without further purification. Proton nuclear magnetic resonance ( H
NMR) spectra were recorded using a Bruker DPX250, DPX400, AV400 or AVIII600
(with cryoprobe) in the indicated solvent. Chemical shifts  are reported in parts per
million (ppm) relative to tetramethylsilane and are internally referenced to the residual
solvent peak. Coupling constants (J) are given in hertz (Hz) to the nearest 0.1 Hz. The
following abbreviations are used for multiplicities: s=singlet; br.s=broad singlet;
d=doublet; t=triplet; m=multiplet; dd=doublet of doublets; ddd=double double doublet.
Liquid Chromatography-Mass Spectroscopy (LCMS) was conducted on either a
Acquity UPLC BEH C18 column (50 mm x 2.1 mm i.d. 1.7µm packing diameter) at
o
40 C or a HPLC XBridge or Sunfire C18 column (50 mm x 4.6 mm i.d. 3.5 µm packing
o
diameter) at 30 C eluting with either 10 mM ammonium bicarbonate in water adjusted
to pH 10 with ammonia solution (solvent A) and acetonitrile (solvent B) or 0.1% v/v
solution of formic acid in water (solvent A) and 0.1% v/v solution of formic acid in
acetonitrile (solvent B). The UV detection is a summed signal from wavelength of
2
10 nm to 350 nm. The mass spectra were recorded on a Waters ZQ spectrometer using
electrospray positive and negative mode. LCMS methods are detailed in the Supporting
Information. High resolution mass spectra (HRMS) were obtained on a Micromass Q-
Tof 2 hybrid quadrupole time-of-flight mass spectrometer, equipped with a Z-spray
interface ESI (+), over a mass range of 100 – 1100 Da, with a scan time of 0.9 s and an
+
interscan delay of 0.1 s. Reserpine was used as the external mass calibrant ([M+H] =
6
09.2812 Da). The Q-Tof 2 mass spectrometer was operated in W reflectron mode to
ACS Paragon Plus Environment

Page 29 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
give a resolution (FWHM) of 16000-20000. Ionisation was achieved with a spray
voltage of 3.2 kV, a cone voltage of 50V, with cone and desolvation gas flows of 10-
20 and 600 L/h respectively. The source block and desolvation temperatures were
maintained at 120 ºC and 250 ºC respectively. The elemental composition was
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
+
calculated using MassLynx v4.1 for the [M+H] and the mass error quoted as ppm. An
Agilent 1100 Liquid Chromatograph equipped with a model G1367A autosampler, a
model G1312A binary pump and a HP1100 model G1315B diode array detector was
used. The method used was generic for all experiments. All separations were achieved
using a Phenomenex Luna C18(2) reversed phase column (100 x 2.1 mm, 3µm particle
size). Gradient elution was carried out with the mobile phases as (A) water containing
0
.1% v/v formic acid and (B) acetonitrile containing 0.1% v/v formic acid. The
conditions for the gradient elution were initially 5% B, increasing linearly to 100% B
over 6 minutes, remaining at 100% B for 2.5 minutes then decreasing linearly to 5% B
over 1 minute followed by an equilibration period of 2.5 minutes prior to the next
injection. The flow rate was 0.5 mL/min, temperature controlled at 35ºC with an
injection volume of between 2 to 5 µL. All samples were diluted with DMSO (99.9%)
prior to LCMS analysis. Analytical chiral chromatography was conducted on a
Chiralpak IA, col.no.IAOOCE-MC024 (25 cm) at room temperature eluting with 5%
ethanol/heptane (1 mL/min). The UV detection was at 215 nm. Preparative chiral
chromatography was conducted on a Chiralpak IA.col.no.IAOOCE-KF008 (25 cm x
2
cm) at room temperature eluting with 5% ethanol/heptane (15mL/min). The UV
detection was at 215 nm. Preparative HPLC using a mass directed auto purification
(MDAP) was conducted on a Waters MassLynx system comprising of a Waters 515
pump with extended pump heads, Waters 2767 autosampler, Waters 996 photodiode
array detector and Gilson 202 fraction collector on a XBridge or Sunfire C18 column
ACS Paragon Plus Environment