Discovery of 2-((3-Acrylamido-4-methylphenyl)amino)-N-(2-methyl-5-(3,4,5-trimethoxybenzamido)phenyl)-4-(methylamino)pyrimidine-5-carboxamide (CHMFL-BMX-078) as a Highly Potent and Selective Type II Irreversible Bone Marrow Kinase in the X Chromosome (BMX) Kinase Inhibitor

Article abstract of DOI:10.1021/acs.jmedchem.6b01413

BMX is a member of TEC family nonreceptor tyrosine kinase and is involved in a variety of critical physiological and pathological processes. Through combination of irreversible inhibitor design and type II inhibitor design approaches, we have discovered a highly selective and potent type II irreversible BMX kinase inhibitor compound 41 (CHMFL-BMX-078), which exhibited an IC₅₀ of 11 nM by formation of a covalent bond with cysteine 496 residue in the DFG-out inactive conformation of BMX. It displayed a high selectivity profile (S score(1) = 0.01) against the 468 kinases/mutants in the KINOMEscan evaluation and achieved at least 40-fold selectivity over BTK kinase. Given the fact that BMX mediated signaling pathway is still not fully understood, compound 41 would serve as a useful pharmacological tool to elucidate the detailed mechanism of BMX mediated signaling pathways.

Full text of DOI:10.1021/acs.jmedchem.6b01413

Subscriber access provided by University of Newcastle, Australia
Article
Discovery of 2-((3-acrylamido-4-methylphenyl)amino)-N-(2-methyl-5-
3,4,5-trimethoxybenzamido)phenyl)-4-(methylamino)pyrimidine-5-
(
carboxamide (CHMFL-BMX-078) as a Highly Potent and Selective Type II
Irreversible Bone Marrow Kinase in the X Chromosome (BMX) Kinase Inhibitor
Xiaofei Liang, Fengchao Lv, Beilei Wang, Kailin Yu, Hong Wu, Ziping Qi, Zongru Jiang, Cheng Chen,
Aoli Wang, Weili Miao, Wenchao Wang, Zhenquan Hu, Juan Liu, Xiaochuan Liu, Zheng Zhao, Li
Wang, Shanchun Zhang, Zi Ye, Chu Wang, Tao Ren, Yinsheng Wang, Qingsong Liu, and Jing Liu
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b01413 • Publication Date (Web): 31 Jan 2017
Downloaded from http://pubs.acs.org on January 31, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Liu, Jing; High Magnetic Field Laboratory, Chinese Academy of Sciences,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 90
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
Discovery of 2ꢀ((3ꢀacrylamidoꢀ4ꢀ
methylphenyl)amino)ꢀNꢀ(2ꢀmethylꢀ5ꢀ(3,4,5ꢀ
trimethoxybenzamido)phenyl)ꢀ4ꢀ
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
(
methylamino)pyrimidineꢀ5ꢀcarboxamide (CHMFLꢀ
BMXꢀ078) as a Highly Potent and Selective Type II
Irreversible Bone Marrow Kinase in the X
Chromosome (BMX) Kinase Inhibitor
1
,2,8
1,3,8
1,3,8
1,3,8
1,2
1,2
Xiaofei Liang , Fengchao Lv , Beilei Wang , Kailin Yu , Hong Wu , Ziping Qi ,
1
,3
1,3
1,3
4
1,2
1,2
Zongru Jiang , Cheng Chen , Aoli, Wang , Weili Miao , Wenchao Wang , Zhenquan Hu ,
1,3
1,2
1,2
1,3
2,5
6
Juan Liu , Xiaochuan Liu , Zheng Zhao , Li Wang , Shanchuan Zhang , Zi Ye , Chu
6
7
4
1,2,3,7*
1,2*
Wang , Tao Ren , Yinsheng Wang , Qingsong Liu
, Jing Liu
1
. High Magnetic Field Laboratory, Chinese Academy of Sciences, Mailbox 1110, 350
Shushanhu Road, Hefei, Anhui 230031, P. R. China
2. CHMFLꢀHCMTC Target Therapy Joint Laboratory, 350 Shushanhu Road, Hefei, Anhui
30031, P. R. China
2
3
. University of Science and Technology of China, Hefei, Anhui 230036, P. R. China
ACS Paragon Plus Environment

Page 3 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4
. Department of Chemistry, University of CaliforniaꢀRiverside, 900 University Ave.,
Riverside, CA 92521, USA
5
. Hefei Cosource Medicine Technology Co. LTD., 358 Ganquan Road, Hefei, Anhui
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
30031, P. R. China
6. Synthetic and Functional Biomolecules Center, Beijing National Laboratory for
Molecular Sciences, Key Laboratory of Bioorganic Chemistry and Molecular
Engineering of Ministry of Education, College of Chemistry and Molecular Engineering,
PekingꢀTsinghua Center for Life Sciences, Peking University, Beijing 100871, P. R.
China
7
. Precision Targeted Therapy Discovery Center, Institute of Technology Innovation, Hefei
Institutes of Physical Science, Chinese Academy of Sciences, Hefei, Anhui 230088, P. R.
China
8
. These authors contribute equally
ABSTRACT
BMX is a member of TEC family nonꢀreceptor tyrosine kinase and is involved in a variety of
critical physiological and pathological processes. Through combination of irreversible inhibitor
design and type II inhibitor design approaches, we have discovered a highly selective and potent
type II irreversible BMX kinase inhibitor compound 41 (CHMFLꢀBMXꢀ078), which exhibited
an IC of 11 nM by formation of a covalent bond with cysteine 496 residue in the DFGꢀout
5
0
inactive conformation of BMX. It displayed a high selectivity profile (S score(1)=0.01) against
the 468 kinases/mutants in the KINOMEscan evaluation and achieved at least 40ꢀfold selectivity
over BTK kinase. Given the fact that BMX mediated signaling pathway is still not fully
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 90
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
understood, compound 41 would serve as a useful pharmacological tool to elucidate the detailed
mechanism of BMX mediated signaling pathways.
INTRODUCTION
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
Bone marrow kinase in the X chromosome (BMX, also called ETK) is a nonꢀreceptor
tyrosine kinase belonging to the TEC family which also includes BTK, ITK, TEC, TXK
1
kinases. Varieties of biological evidence demonstrate that BMX is involved in tumorigenicity,
cell motility, adhesion, angiogenesis, proliferation and differentiation. BMX is expressed
primarily in bone marrow, but also wildly found in granulo/monocytic cells, epithelial and
2
ꢀ8
endothelial cells, as well as brain, prostate, heart and lung. BMX kinase contains a pleckstrin
homology domain (PH) that mediates membrane association, a Tec homology domain (TH) that
consists of a proline rich region, a SRC homology 3 domain (SH3) that is involved in inter and
intraꢀmolecular interactions, and a SRC homology 2 domain (SH2) that recognizes
phosphotyrosine containing peptides and proteins from the amino terminus, as well as the kinase
1
domain in the carboxyl terminus. BMX participates in signal transduction in response to
virtually all types of extracellular stimuli which are transmitted by growth factor receptors, Gꢀ
3
protein coupled receptors, cytokine receptors, antigen receptors and integrins. BMX is regulated
by a variety of nonꢀreceptor tyrosine kinases like SRC, JAK, SYK, FAK family kinases and lipid
kinase such as PI3K. In turn, BMX may be involved in regulation of several signaling pathways
3,9
including AKT, STAT3, PKC, PLCγ and p21ꢀactivated kinases under certain conditions.
Due to the potential pathological roles of BMX in malignant disorders, several small
9
molecule or peptide inhibitors of BMX were delvelped. However, most of these inhibitors are
multiple target compounds. Besides BMX kinase, they usually also potently inhibit other TEC
ACS Paragon Plus Environment

Page 5 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
9
10
family kinases or other protein kinases. For example, compound 1 (BMXꢀINꢀ1, Figure 1)
which covalently modified Cys496 displayed high selectivity profile against other protein
1
0
11
kinases but also strongly inhibited BTK kinase (IC = 10.4 nM). Compounds 2 (CIꢀ1033) , 3
50
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
12
13
14
(
PD168393) , 4 (LFMꢀA13) and 5 (PCIꢀ32765) have also been reported to bear BMX kinase
activity though their primary targets are other structurally related kinases such as EGFR and
BTK, etc. Given the fact that BMX mediated signaling pathway in the physiological and
pathological context is still not fully understood yet, selective BMX kinase inhibitors are highly
desired for mechanistic investigation and proofꢀofꢀconcept efficacy validation.
Figure 1. Chemical structures of representative inhibitors that bear BMX kinase activity.
Accumulated evidence have shown that irreversible kinase inhibitors, which possess a
number of potential advantages including high potency, prolonged pharmacodynamics,
suitability for rational design and ability to validate pharmacological specificity, sometimes may
15ꢀ18
exhibit distinct selectivity profile from conventional reversible inhibitors.
However, although
BMX kinase bears a cysteine 496 residue which is feasible for design of irreversible inhibitors,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 90
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
similar cysteine residues located at the identical position were also found in 11 other kinases
such as TEC kinase family (BMX, ITK, RLK, TEC), EGFR family (EGFR, HER2, HER4), as
1
9
well as BLK, JAK3 and MAP2K7 kinases. Therefore, in order to achieve a high selectivity,
other selectivity trimming factors need to be introduced. Type II kinase inhibitors, which are
featured by binding to the inactive conformation of kinase and occupying the nonꢀhighly
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
20, 21
conserved allosteric pocket generated by “DFGꢀout” motif flip,
sometimes can provide better
selectivity compared to type I inhibitors and have attracted extensive attention in the kinase
2
2,23
targeted drug discovery.
Therefore, we postulated that combining the irreversible inhibitor
design and type II inhibitor design approaches might provide better selectivity. Here, we
described our medicinal effort that led to the discovery of a highly selective type II irreversible
BMX kinase inhibitor compound 41 (CHMFLꢀBMXꢀ078), which bound to the “DFGꢀout”
inactive conformation of BMX kinase meanwhile took advantage of the cysteine 496 residue to
form a covalent bond.
RESULTS AND DISCUSSION
Structure-Activity Relation (SAR) Exploration
During search of BMX inhibitor with type II binding mode, we found that the
2
4
dihydropyrimidopyrimidine compound 6 (GNFꢀ7), which has been reported as a type II
inhibitor that was capable to inhibit the T315I “gatekeeper” point mutant of BCRꢀABL, could
potently inhibit BMX kinase at 10 ꢀM (98% inhibition). However, it also potently inhibited
other kinases such as JAK1, 2, 3, FGFR3, FLT3, PDGFR, TRKC, cꢀRAF, CSK, JNK, MKK, etc.
After docking compound 6 to the Xꢀray crystal structure of BMX (PDB ID: 3SXR), we found
that it adopted a canonical type II binding mode (Figure 2A). The 2ꢀamine pyrimidine moiety
ACS Paragon Plus Environment

Page 7 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
“hinge binding” part in Figure 3) formed hydrogen bond in the hinge region. The
trifluoromethylbenzene moiety (“tail” part in Figure 3) occupied the allosteric hydrophobic
pocket generated by DFGꢀout shifting. The amide moiety formed two hydrogen bonds with
Glu460 and Asp554, respectively (Figure 2B). The methylpyridine moiety (“Head” part in
Figure 3) oriented to the adjacent area of the hingeꢀbinding region. Interestingly, the distance
between the N atom of pyridine and the cysteine 496 residue was about 5.3Å, which was feasible
for irreversible inhibitor design (Figure 2A). Virtually attaching an acrylamide to the N position
in the head moiety of compound 6 (compound 7) did show a covalent bond formation,
meanwhile still kept type II binding in the modeling (Figure 2CꢀD). Thus, we proposed that by
installing a suitable electrophile at the methylpyridine N position meanwhile exploring other key
binding elements of compound 6 and taking advantage of the nonꢀhighly conserved DFGꢀout
hydrophobic pocket might provide a new highly selective type II irreversible BMX kinase
inhibitor, and these efforts finally resulted in the discovery of compound 41 (Figure 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
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 90
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. Modeling of interactions of dihydropyrimidopyrimidine compounds 6 and 7 with
BMX kinase (PDB code: 3SXR). (A) The Cys 496 is 5.3 Å from the N atom in the head moiety
of compound 6. (B) The covalent bond formed between the acrylamide moiety of compound 7
and Cys 496. (C) Demonstration of canonical hydrogen bonds in the type II binding mode
between compound 6 and DFGꢀout inactive conformation of BMX kinase. (D) Demonstration of
canonical hydrogen bonds in the type II binding mode between compound 7 and DFGꢀout
inactive conformation of BMX kinase.
ACS Paragon Plus Environment

Page 9 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Schematic illustration of the discovery of compound 41.
For SAR exploration, we used the recombinant BMX kinase protein to test compounds’
enzymatic inhibition capability (IC ). In addition, to avoid the problems of different ATP
5
0
concentrations used in the enzymatic assay (uM ranges) and cellular assay (mM ranges), which
usually makes the compounds displaying potent activity in vitro lose the activity in cell, we also
used BaF3 cells transformed with TEL fused BMX kinase whose proliferation depend on BMX
kinase to monitor the compound’s cellular activity (GI ). Parental BaF3 cells were used to
5
0
monitor the compound’s general selectivity. Ideally in this assay the highly selective and potent
compound would show strong antiꢀproliferative efficacy against BaF3ꢀTELꢀBMX cell but not
parental BaF3 cell. To test our design, we first prepared compound 7 and found that compared to
compound 6, it retained high potency against BaF3ꢀTELꢀBMX cell (GI : <0.0005 ꢁM, IC : 8
5
0
50
nM) meanwhile showed potent inhibitory activity against parental BaF3 cell (GI : 0.74 ꢀM) like
5
0
6
, which indicated that its selectivity still needed to be improved (Table 1). Since compound 7
still potently inhibited parental BaF3 cells, to detect its real activity against BMX kinase, we then
examined 7’s activity against BMX autoꢀphosphorylation with purified BMX wt kinase and the
C496S mutant. Interestingly, compound 7 exhibited an EC of 18 nM against BMX wt and an
5
0
EC of 477 nM against C496S mutant (Figure 4). This indicated that 7 might inhibit BMX
50
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 90
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
kinase through irreversible binding mode. Encouraged by this finding, we decided to keep the
head moiety as the acrylamide substituted methyl aniline like 7 and further modify the linker and
tail parts in order to achieve better selectivity.
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 4. Compound 7’s effects against the autoꢀphosphorylation of purified BMX wt and
C496S mutant proteins.
We first fixed the linker part and varied the tail moiety which occupied the hydrophobic
pocket generated by the “DFG” motif shift in the inactive conformation. Switching the
trifluoromethyl substituent (7) to oꢀmethoxy (8) or pꢀmethoxy group (9) resulted in significant
activity loss against BMX (IC : 591 nM and 1535 nM; GI : 3.01 ꢁM and 1.07 ꢁM,
5
0
50
respectively) (Table 1). Similarly, introduction of one or two more methyloxy groups to the
benzene ring in the tail part (10-14) all led to activity loss compared to 7. Interestingly,
cyclization of 3, 4ꢀdimethyoxy group to form fiveꢀmembered ring (15) started to gain back the
BMX activity (IC : 382 nM; GI : 0.23 ꢁM). Although it was still about 500ꢀfold less active
5
0
50
than compound 7 in terms of GI , it started to show the selectivity against parental BaF3 cell
5
0
(
GI : >10 ꢁM). However, cyclization of 3,4ꢀdimethyl group to form sixꢀmembered ring (16)
50
ACS Paragon Plus Environment

Page 11 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
caused complete activity loss to BMX (IC : 1263 nM; GI : >10 ꢁM). While 5,6,7,8ꢀ
5
0
50
tetrahydronaphthalene (17) in the tail part gained back a moderate activity (IC : 813 nM; GI :
50
50
0.86 ꢁM), introduction of another methyl group (18) or trifluoromethyl group (19) to compound
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
generated moderate activities against BMX (IC : 2468 nM and 1544 nM; GI : 1.24 ꢁM and
5
0
50
0.78 ꢁM). Replacement of 4ꢀmethoxy substituent (9) with N, Nꢀdimethylamine (20) did not
affect the potency too much. Interestingly, although 2ꢀfluoroꢀ6ꢀmethyl benzene (21) only
displayed moderate activity against BMX (IC : 480 nM; GI : 2.78 ꢁM), shifting the methyl
5
0
50
group from 6ꢀ to 5ꢀpostion (22) dramatically increased its growth inhibition activity against
BaF3ꢀTELꢀBMX (IC : 120 nM; GI : 0.031ꢁM) meanwhile displayed a good selectivity to the
5
0
50
parental BaF3 cell (GI : >10 ꢁM). Replacement of the 3ꢀ(trifluoromethyl)benzoic acid group in
50
the tail moiety of 7 with more bulky amino acids, i.e., phenylanaline (23) and tryptophan (24)
both completely abolished the activity to BMX. A more flexible leucine moiety (25) exhibited
slightly better activity against BMX (IC : 33 nM; GI : 0.13 ꢁM) meanwhile kept a good
50
50
selectivity against parental BaF3 cells (GI : >10 ꢁM). However, the more bulky Bocꢀprotected
5
0
glutamic acid moiety (26) decreased the activity again (IC : 4620 nM; GI : 8.35 ꢁM). Further
5
0
50
replacement in the tail moiety of 7 with different paraꢀsubstituted cyclohexane derivatives (27-
1) did not show increased activity against BMX.
3
a
Table 1. SAR Exploration Focused on the R1 Moiety
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 90
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
BMX Kinase BaF3ꢀTELꢀBMX ParentalꢀBaF3
Compd
R1
ꢀ
R2
ꢀ
(
IC : nM)
(GI : ꢁM)
(GI : ꢁM)
5
0
50
50
6
7
8
9
8±3
8±3
<0.0005
<0.0005
0.12±0.008
0.74±0.01
>10
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
591±165
3.01±0.03
1.07±0.16
0.24±0.07
0.69±0.19
2.74±0.04
3.00±0.72
5.93±0.22
0.23±0.05
>10
1535±642
637±210
1147±437
927±376
1300±405
>10000
>10
O
O
O
1
1
1
1
1
1
1
1
0
1
2
3
4
5
6
7
8.13±1.40
>10
>10
>10
>10
382±134
1263±400
813±200
>10
>10
0.86±0.12
>10
ACS Paragon Plus Environment

Page 13 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
18
19
20
21
2468±853
1544±768
5160±1979
480±177
1.24±0.26
0.78±0.17
1.03±0.14
2.78±0.06
>10
>10
>10
>10
>10
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
2
2
3
120±45
>10000
0.031±0.006
>10
>10
2
2
2
4
5
6
2372±772
33±14
>10
>10
>10
0.13±0.01
8.35±1.12
0.64±0.19
5.28±1.29
5.9±0.22
2.28±0.95
4620±1298
3025±1057
>10000
>10
27
28
29
30
7.4±0.35
>10
2823±956
5472±1747
>10
8.81±1.28
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 90
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
31
4017±1142
2.49±0.40
5.08±1.62
a
All GI values were obtained by triplet testing.
5
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
We then turned our attention to the linker part for further SAR study. Considering that the
cyclic urea ring was rigid, we thought that opening the ring would presumably generate more
flexibility of the scaffold. Replacement of the cyclic urea to amide moiety (32) exhibited
moderate biochemical inhibition (IC : 37 nM), great antiꢀproliferative efficacy against BaF3ꢀ
5
0
TELꢀBMX cell (GI : 0.013 ꢁM) as well as good selectivity over parental BaF3 cell (GI : >10
5
0
50
ꢁ
M). Interestingly, saturation of the acrylamide to propinamide (33) displayed about 6ꢀfold
selectivity against BaF3ꢀTELꢀBMX cell (IC : 102 nM; GI : 0.085 ꢁM) compared to 32,
5
0
50
meanwhile kept the selectivity window to parental BaF3 cells (GI : >10 ꢁM). Encouraged by
5
0
these results, we decided to keep this core linker scaffold and explore more on the tail moiety.
Replacement of the 3ꢀtrifluoromethyl group with 3ꢀmethoxy group on the benzene ring (34)
dramatically decreased the potency against BMX kinase (IC : 2380 nM) and BaF3ꢀTELꢀBMX
5
0
cell (GI : 6 ꢁM). Neither shifting the 3ꢀmethoxy group to the 2ꢀpostion (35), 4ꢀposition (36), nor
5
0
the 3,4ꢀdimethyoxy (37) or the [1,3]dioxo (38) substituents on the tail benzene ring retained the
activity against BaF3ꢀTELꢀBMX cell. The 2,4ꢀdimethoxy group (39) and [1,4] dioxine (40)
substituents on the tail benzene ring displayed about 7ꢀfold enzymatic inhibition loss (IC : 225
5
0
nM and 279 nM) and 10ꢀfold growth inhibition loss against BaF3ꢀTELꢀBMX cell (GI : 0.119
5
0
ꢁ
M and 0.155 ꢁM) compared to 32. Interestingly, the 3,4,5ꢀtrimethoxy substituted benzene ring
(
41) retained the antiproliferation efficacy against BMX (IC : 11 nM; GI : 0.016 ꢁM).
5
0
50
Saturation of the acrylamide to propinamide (42) resulted in about 20ꢀfold (IC ) and 40ꢀfold
5
0
(
GI ) activity loss against BMX, indicating an irreversible binding. The 2ꢀfluoroꢀ5ꢀmethyl
50
ACS Paragon Plus Environment

Page 15 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
substituted benzene ring (43) also kept the inhibitory activity against BaF3ꢀTELꢀBMX (GI :
5
0
0.044 ꢁM) in the similar range to 32. Reverting the amide linkage in compound 32 (44) caused
increased potency against BMX (IC : 86 nM; GI : 0.008 ꢁM) meanwhile gained activity to
5
0
50
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
parental BaF3 cell (GI : 4.9 ꢁM), which indicated offꢀtarget activities. Reverting the amide
5
0
linkage in compound 41 (45) led to about 6ꢀfold growth inhibition activity loss against BaF3ꢀ
TELꢀBMX cell (GI : 0.098 ꢁM versus 0.016 ꢁM) and 50ꢀfold enzymatic inhibition activity loss
5
0
(
IC : 540 nM versus 11 nM).
50
a
Table 2. SAR Exploration Focused on the R3/R4 Moieties
BMX Kinase BaF3ꢀTELꢀBMX ParentalꢀBaF3
Compd
R3
R4
(
IC : nM)
(GI : ꢁM)
(GI : ꢁM)
5
0
50
50
3
2
37±10
0.013±0.004
0.085±0.002
6.00±0.19
0.67±0.17
1.10±0.02
>10
>10
>10
>10
>10
33
34
35
36
102±17
2380±839
2877±750
2427±934
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 90
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
37
38
>10000
1.0±0.2
>10
>10
2562±794
1.08±0.14
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
9
0
225±69
279±86
0.119±0.013
0.155±0.049
>10
>10
4
41
11±0.7
230±14
0.016±0.003
0.61±0.12
>10
>10
42
43
44
45
133±53
86±34
0.044±0.002
0.008±0.004
0.098±0.009
>10
4.9
540±180
>10
a
All GI values were obtained by triplet testing.
5
0
Since compounds 32 and 41 exhibited the best potencies against BaF3ꢀTELꢀBMX cells and
best selectivity over parental BaF3 cells, we then examined them together with their reversible
version compounds 33 and 42 in a panel of TEL transfused isogenic BaF3 cells to further explore
their selectivity profiles (Table 3). The results demonstrated that compound 32 displayed
relatively strong potency against PDGFRα (GI : 0.099 ꢀM) and moderate activity to PDGFRβ
5
0
ACS Paragon Plus Environment

Page 17 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
GI : 0.13 ꢀM), ABL (GI : 0.118 ꢀM), cKIT (GI : 0.38 ꢀM) but no apparent activity against
5
0
50
50
FLT3 kinase (GI : >10ꢀM). By comparison, compound 41 displayed better selectivity than 32.
5
0
It only moderately inhibited PDGFRα/β (GI : 0.67 ꢀM, 0.91 ꢀM) but not other kinases.
5
0
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
Interestingly, both compounds 32 and 41 had no apparent activities against EGFR, JAK3 or BLK
kinases, which bear similar cysteine residue at the identical position. In addition, the potent
inhibitory activity against BMX kinase of both compounds 32 and 41 could be rescued by ILꢀ3,
which is the natural growth stimulator of isogenic BaF3 cells, indicating that their
antiproliferative activity against BaF3ꢀTELꢀBMX cells mainly came from the BMX activity
inhibition. Since compound 41 exhibited better selectivity among these structurally similar
kinases, we decided to further characterize its biological profiles.
Table 3. Antiproliferative Effects of Compounds 32-33 and 41-42 against a Panel of Isogenic
a
BaF3 Cell Lines
Compd. 32
GI : ꢀM)
Compd. 33
Compd. 41
Compd. 42
Cell line
Parental BaF3
(
(GI : ꢀM)
(GI : ꢀM)
(GI : ꢀM)
5
0
50
50
50
>10
>10
>10
2.56±0.07
2.01±0.05
0.67±0.06
0.91±0.05
>10
>10
2.51±0.24
2.80±0.02
0.37±0.02
0.79±0.05
>10
BaF3ꢀTELꢀABL
BaF3ꢀTELꢀcKIT
BaF3ꢀTELꢀPDGFRa
BaF3ꢀTELꢀPDGFRb
BaF3ꢀTELꢀFLT3
BaF3ꢀTELꢀEGFR
BaF3ꢀTELꢀJAK3
0.118±0.056
0.38±0.03
0.099±0.014
0.13±0.07
>10
0.044±0.001
0.17±0.02
0.045±0.005
0.059±0.006
8.6±0.24
>10
>10
>10
>10
>10
8.3±0.16
>10
>10
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 90
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
BaF3ꢀTELꢀBLK
5.9±0.7
0.013±0.004
>10
6.5±0.9
0.085±0.002
>10
>10
0.016±0.003
>10
>10
0.61±0.12
>10
BaF3ꢀTELꢀBMX
BaF3ꢀTELꢀBMX+IL3
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
All GI values were obtained by triple testing.
5
0
Compound 41’s Selectivity Profiling
We next examined compound 41 against a panel of 468 kinases/mutants in the DiscoverRx’s
KINOMEscan profiling platform. At the concentration of 1 ꢁM, compound 41 demonstrated
high selectivity with a S score(1)=0.01 (Figure 5AꢀB and Supplemental Table 1)．The data
indicated that compound 41 had strong binding (with a percent control number less than 1)
against DDR1, P38α and ABL1 kinases. Interestingly, 41 did not show strong binding to BMX
in KINOMEscan assay (percent control number 14). This might due to the reason that BMX
kinase did not prefer the DFGꢀout inactive conformation under this assay condition. This
phenomenon was also observed for ABL1 kinase, e.g. compound 41 displayed a binding factor
of 0.7 against ABL1ꢀnonphosphoryalted (inactive conformation) but 6.5 against ABL1ꢀ
phosphoryalted (active conformation). To further confirm this hypothesis, we then used the
recombinant inactive (unphosphoylated) and active (phosphorylated) BMX kinase proteins
respectively to test compound 41’s binding Kd values with MST (microscale thermophoresis)
technology (Figure 5C). The results demonstrated that for the inactive state of BMX kinase,
compound 41 displayed a binding Kd of 81 nM, while for the active state of BMX kinase that
was generated by preincubation of the kinase with 1 ꢀm ATP overnight, 41 exhibited a binding
Kd of 10200 nM. The over 120ꢀfold difference of these Kd values further suggested that 41
preferred to bind to the inactive conformation of the kinase with a type II binding mode. Given
ACS Paragon Plus Environment

Page 19 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the fact that KINOMEscan is a binding assay and cannot fully reflect the compound’s inhibitory
activity, we then further confirmed these potential targets with the Invitrogen SelectScreen
biochemical assay (Figure 5D). Compound 41 exhibited an IC of 11 nM against BMX, while
5
0
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
the reversible version compound 42 (IC : 230 nM) was about 21ꢀfold less active, indicating an
5
0
irreversible binding mode. For the structurally very similar BTK kinase, compound 41 displayed
an IC50 of 437 nM, which was about 40ꢀfold less active compared to BMX. In addition,
compound 41 also displayed about 10ꢀfold selectivity against p38α kinase (IC : 106 nM) and
5
0
2
4ꢀfold selectivity against DDR1 kinase (IC : 265 nM) compared to BMX.
50
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 90
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
Figure 5. Kinomeꢀwide selectivity profiling of compound 41 with DiscoverRx KINOMEscan
technology and biochemical IC50 values of compounds 41 and 42 measured with Invitrogen
SelectScreen Technology. (A) Treespot demonstration of 41’s selectivity in 468 kinase targets.
Measurements were performed at a concentration of 1 ꢀM of the compound. The affinity was
defined with respect to DMSO control. (B) Other targets that demonstrated strong binding to 41
with a percent control number less than 1. (C) Binding Kd values of compounds 41 and 42 with
purified recombinant inactive and active BMX kinase proteins measured by MST (microscale
thermophoresis) technology. (D) Biochemical IC values of compounds 41 and 42 measured
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
50
with Invitrogen SelectScreen Technology.
Compound 41’s Binding Mode Exploration
To better understand the binding mode of compound 41, we then docked it into the BMX Xꢀ
ray crystal structure (PDB ID: 3SXR) (Figure 6A). The modeling result demonstrated that
compound 41 did prefer to adopt a type II binding mode and form a covalent bond with Cys 496
as expected. The aminopyrimidine moiety formed two hydrogen bonds with Met 492 in the
hinge area. The ꢀOH side chain of “gatekeeper” residue Thr 489 formed a hydrogen bond with
the oxygen atom of amide group. The Glu 460 residue in the cꢀhelix and Asp 554 residue in the
DFG motif formed two typical hydrogen bonds with the amide group near the tail moiety. The
backbone of Leu 423 and Cys 496 residues also formed two hydrogen bonds with the amide
group in the acrylamide moiety. In addition, the 3,4,5ꢀtrimethoxylbenzene in the tail moiety
occupied the hydrophobic pocket formed by “DFG” motif through Van der Waals interactions.
The side chain of Phe 555 interacted with the aminomethyl group in the linker moiety through pꢀ
π Van der Waals.
ACS Paragon Plus Environment

Page 21 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
We also performed the in vitro kinase assay with BMX wt and C496S mutant in order to
further confirm the irreversible binding mode of compound 41. The results showed that
compound 41 was about 80ꢀfold more potent against BMX wt (EC : 5.8 nM) than C496S
5
0
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
mutant (EC : 459 nM) for the inhibition of BMX autoꢀphosphorylation (Figure 6B). Meanwhile,
50
the reversible version compound 42 inhibited the wildꢀtype BMX autoꢀphosphorylation with an
EC of 197 nM, which was about 34ꢀfold less active than compound 41. To further confirm the
50
covalent binding, we then performed the mass spectrometry experiments and the results
demonstrated that upon incubation, C496 in the BMX kinase was indeed modified by compound
41 (Figure 6CꢀD and Supplemental Figure 1). These data, together with the antiproliferation
results observed from BaF3ꢀTELꢀBMX cells suggested that compound 41 indeed adopted an
irreversible binding mode through forming a covalent bond with Cys 496 of BMX protein.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 90
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. Binding mode illustration of compound 41 with BMX kinase. (A) Putative binding
model of 41 with BMX inactive conformation (PDB ID: 3SXR). (B) Effects of compounds 41
and 42 on BMX wt and C496S mutant’s totalꢀtyrosine phosphorylation. Purified Hisꢀtag labeled
recombinant full length of the inactive state of BMX wt and C496S mutant were first treated
with the inhibitors and then incubated with 100 µM ATP before the total protein phosphorylation
was detected with westernꢀblot. (C, D) LCꢀMS/MS in the MRM mode revealed the compound
41ꢀconjugated Cys496ꢀcontaining tryptic peptide derived from recombinant BMX after reaction
with 41. Shown in (C) are the selectedꢀion chromatograms (SICs) for monitoring the indicated
ACS Paragon Plus Environment

Page 23 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3+
transitions arising from the fragmentations of the [M+3H] ions of the unmodified (top) and
compound 41ꢀmodified (bottom) Cys496ꢀcontaining peptides. Displayed in (D) are the MS/MS
3
+
for the [M+3H] ions of the unmodified (top) and compound 41ꢀmodified (bottom) cysteineꢀ
containing peptides averaged from the peak at around 69.6 min. From the peak areas found in
the SICs, it can be estimated that approximately 23% of Cys496 was modified.
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
Compound41’s Effect in Cells
Since BMX’s role in pathology has not been fully validated yet and currently there is no BMX
dependent disease model, we chose a panel of established cancer cell lines including prostate
cancer, bladder cancer and renal cancer which had been reported to express BMX kinases to test
the antiproliferative effects of compound 41 (Table 4). Not surprisingly, compound 41 only
exhibited moderated growth inhibition efficacies against most of these cell lines (subꢀmicomolar
GI ). However, it didn’t show apparent inhibitory activities against normal mouse cells CHO
5
0
and CHL (GI : >10 ꢁM), which suggested no general toxicity.
5
0
a
Table 4. Antiproliferative Effects of Compounds 41 against a Panel of Cancer Cell Lines
Cell line
RVꢀ1
DUꢀ145
PCꢀ3
EJ
Cell origin
GI (ꢀM)
5
0
Prostate cancer
Prostate cancer
Prostate cancer
Bladder cancer
Bladder cancer
Bladder cancer
3.45±1.20
7.89±1.10
3.45±0.45
>10
Hbꢀc
8.87±0.87
5.78±0.64
J82
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 90
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
T24
Bladder cancer
Bladder cancer
8.98±2.40
>10
5637
769ꢀP
Renal cancer
>10
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
786ꢀO
Renal cancer
>10
Aꢀ498
ACHN
CHL
Renal cancer
>10
Renal cancer
4.93±0.65
>10
Normal lung of mouse
Normal ovary of mouse
CHO
>10
a
All GI values were obtained by triple testing.
5
0
We then chose a prostate cancer cell line DU145 that overexpressed FLAG labeled BMX
1
0
kinase to confirm compound 41’s effect on cells (Figure 7). Given the fact that currently there
is no workable specific phosphorylated BMX antibody, we decide to examine the overall
tyrosine phosphorylation with general pꢀtyrosine antibodies. The results showed that 41 did
potently inhibited overall BMX kinase tyrosine phosphorylation starting from the concentration
of 1 ꢀM.
ACS Paragon Plus Environment

Page 25 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 7. Effect of compound 41 on inhibition of total tyrosine phosphorylation in DUꢀ145 cells
with overexpressed FLAG labeled full length BMX kinase.
Compound 41’s In Vivo Pharmacokinetic Property Evaluation.
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
Next, we evaluated the pharmacokinetic property of compound 41 in rats (table 5). The data
showed that 41 exhibited a short halfꢀlife (T = 0.80±0.34 h) in i.v. injection. It is noteworthy
1
/2
that for irreversible inhibitors, a short halfꢀlife is actually preferred to avoid potential offꢀtarget
1
7,25
toxicities and drugꢀdrug interactions, which can lead to increased therapeutic margins.
Compound 41 also displayed an acceptable Cmax (13565.23 ng/mL) and AUC(0ꢀt) (1386.41
ng/mL*h ) in i.v. injection. However, 41 was not absorbed by oral administration indicating that
this compound could be administrated through i.v. or i.p. injection when used as a research tool.
Table 5. Pharmacokinetic Property Evaluation of Compound 41 on Sprague Dawley Rats.
AUC(0ꢀt)
AUC(0ꢀ∞)
MRT(0ꢀt)
T1/2
Tmax
Cmax
F
Compd.41
(
ng/mL*h)
( ng/mL*h)
(h)
(h)
(h)
(ng/mL)
(%)
iv (1mg/kg)
Mean±SD
1386.41±133.93 1388.25±132.98 0.224±0.01 0.80±0.34
0.017
13565.23±1627.96
5.58±1.8
/
/
po (10mg/kg)
Mean±SD
1
3.39±4.72
27.63±2.50
2.76±0.82
6.11±2.35 0.19±0.09
CHEMISTRY
The general synthetic procedure was depicted in Schemes 1ꢀ3. The intermediate 49 was
2
6
prepared following the previously reported route. Briefly, substitution of ethyl 4ꢀchloroꢀ2ꢀ
(methylthio)pyrimidineꢀ5ꢀcarboxylate (46) with methylamine afforded methylamine pyrimidine
47, then LAH reduction of the ethyl ester to alcohol (48) followed by oxidation reaction
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 90
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
furnished the aldehyde 49 (Scheme 1). Reductive amination of 49 with tertꢀbutyl 3ꢀaminoꢀ4ꢀ
methylphenylcarbamate provided the intermediate 50, which was then converted to the cyclic
urea 51 with triphosgene. Exposure of the methyl sulfide 51 to mꢀCPBA generated the
corresponding sulfoxide 52, which was transformed into the Boc protected aniline 53. Removal
of the Boc protection group (54) followed by amide coupling reaction provided the amide
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
6
product 55, which was then acylated with acryloyl chloride to gave target compounds 7-31.
a
Scheme 1. Synthesis of Compounds 7−31
ACS Paragon Plus Environment

Page 27 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
Reagents and conditions: (a) NH Me.HCl, TEA, THF, 0 °C to rt; (b) LiAlH , THF, 0 °C to rt;
2
4
(c) MnO , DCM, rt; (d) tertꢀbutyl 3ꢀaminoꢀ4ꢀmethylphenylcarbamate, MeOH/AcOH, reflux;
2
then NaBH CN, rt; (e) triphosgene, DIPEA, dioxane, 0 °C to rt; (f) mCPBA, DCM, 0°C to rt; (g)
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
tertꢀbutyl 5ꢀaminoꢀ2ꢀmethylphenylcarbamate, LHMDS, THF, ꢀ78 °C to rt; (h) 4 M HCl in
MeOH, rt; (i) RꢀCOOH, HATU, DIPEA, DMF; (j) for the synthesis of 7-22 and 27-31: acryloyl
chloride, THF, DIPEA, rt; for the synthesis of 23-26: acryloyl chloride, THF, DIPEA, rt; then
HCl in MeOH, rt.
Compounds 32ꢀ43 were prepared following the procedure shown in Scheme 2. Compound
47 was hydrolyzed to the corresponding carboxylic acid 56, which was coupled with tertꢀbutyl 3ꢀ
aminoꢀ4ꢀmethylphenylcarbamate to afford the amide intermediate 57. Deprotection of the Boc
group (58) followed by acylation with substituted acyl chloride derivatives furnished the sulfide
intermediate 59, which was then subjected to oxidation, amination, reduction, and acryloylation
or propionylation to furnish the desired compounds 32ꢀ43.
a
Scheme 2. Synthesis of Compounds 32-43
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 90
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) NaOH, EtOH, H O, rt; (b) tertꢀbutyl 3ꢀaminoꢀ4ꢀ
2
methylphenylcarbamate, HATU, DIPEA, DMF; (c) 4 M HCl in MeOH, rt; (d) R’COCl, THF,
DIPEA, 0 °C; (e) mꢀCPBA, DCM, 0 °C; (f) 4ꢀmethylꢀ3ꢀnitroaniline, dioxane, TFA, 120 °C; (g)
SnCl , MeOH, reflux; (h) for the synthesis of 32, 34-41 and 43: acryloyl chloride, THF, DIPEA,
2
rt; for the synthesis of 33 and 42: propionyl chloride, THF, DIPEA, rt.
As shown in Scheme 3, compounds 44ꢀ45 were prepared starting from carboxylic acid 46.
Amide coupling reaction with methyl 3ꢀaminoꢀ4ꢀmethylbenzoate afforded methyl ester 63,
which was then hydrolyzed to carboxylic acid 64. Another amide coupling with corresponding
ACS Paragon Plus Environment

Page 29 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
aniline derivatives generated the sulfide intermediate 65, which was then subjected to oxidation,
amination, reduction, and acryloylation to offer the final compounds 44 and 45.
a
Scheme 3. Synthesis of Compounds 44-45
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) methyl 3ꢀaminoꢀ4ꢀmethylbenzoate, HATU, DIPEA, DMF, rt; (b)
NaOH, MeOH, H O, rt; (c) RꢀNH , HATU, DIPEA, DMF, rt; (d) mꢀCPBA, DCM, 0 °C to rt; (e)
2
2
4ꢀmethylꢀ3ꢀnitroaniline, LiHMDS, THF, ꢀ78 °C to rt; (f) SnCl , MeOH, reflux; (g) acryloyl
2
chloride, THF, DIPEA, rt.
CONCLUSIONS
Starting from a dihydropyrimidopyrimidine scaffold based multiꢀtarget compound 6 bearing
high BMX potency, through combination of irreversible inhibitor design and type II inhibitor
design approaches, we discovered a highly selective type II irreversible BMX inhibitor 41 via 8
steps of chemical synthesis (over 9% overall yield). Compound 41 exhibited great selectivity
ACS Paragon Plus Environment