Structure-Based Design, Synthesis, and Biological Evaluation of Highly Selective and Potent G Protein-Coupled Receptor Kinase 2 Inhibitors

Article abstract of DOI:10.1021/acs.jmedchem.5b02000

G protein-coupled receptors (GPCRs) are central to many physiological processes. Regulation of this superfamily of receptors is controlled by GPCR kinases (GRKs), some of which have been implicated in heart failure. GSK180736A, developed as a Rho-associated coiled-coil kinase 1 (ROCK1) inhibitor, was identified as an inhibitor of GRK2 and co-crystallized in the active site. Guided by its binding pose overlaid with the binding pose of a known potent GRK2 inhibitor, Takeda103A, a library of hybrid inhibitors was developed. This campaign produced several compounds possessing high potency and selectivity for GRK2 over other GRK subfamilies, PKA, and ROCK1. The most selective compound, 12n (CCG-224406), had an IC₅₀ for GRK2 of 130 nM, >700-fold selectivity over other GRK subfamilies, and no detectable inhibition of ROCK1. Four of the new inhibitors were crystallized with GRK2 to give molecular insights into the binding and kinase selectivity of this class of inhibitors.

Full text of DOI:10.1021/acs.jmedchem.5b02000

Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES
Article
Structure-Based Design, Synthesis and Biological Evaluation of Highly
Selective and Potent G Protein-Coupled Receptor Kinase 2 Inhibitors
Helen V Waldschmidt, Kristoff T Homan, Osvaldo Cruz-Rodriguez, Marilyn Claire Cato,
Jessica Waninger-Saroni, Kelly Marie Larimore, Alessandro Cannavo, Jianliang Song,
Joseph Y Cheung, Paul D. Kirchhoff, Walter J Koch, John J.G. Tesmer, and Scott D Larsen
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b02000 • Publication Date (Web): 06 Apr 2016
Downloaded from http://pubs.acs.org on April 7, 2016
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 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
StructureꢀBased Design, Synthesis and
Biological Evaluation of Highly Selective and
Potent G ProteinꢀCoupled Receptor Kinase 2
Inhibitors
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,b,d
b,d,†
AUTHOR NAMES: Helen V. Waldschmidt , Kristoff T. Homan , Osvaldo Cruz-
b,c,d
b,d
b,d,e
b,d
Rodríguez
, Marilyn C. Cato , Jessica Waninger-Saroni , Kelly L. Larimore ,
f
f
f
a,d
Alessandro Cannavo , Jianliang Song , Joseph Y. Cheung , Paul D. Kirchhoff , Walter J.
f
b,d
a,d
Koch , John J. G. Tesmer , and Scott D. Larsen *
a
AUTHOR ADDRESSES: Vahlteich Medicinal Chemistry Core, College of Pharmacy,
b
c
d
Life Sciences Institute, PhD Program in Chemical Biology, Departments of Medicinal
e
Chemistry, Pharmacology and Biological Chemistry, Medical Scientist Training
f
Program, University of Michigan, Ann Arbor, Michigan, 48109, Center for Translational
Medicine, Temple University, Philadelphia, Pennsylvania, 19140
KEYWORDS: G proteinꢀcoupled receptor kinases, G proteinꢀcoupled receptors, βꢀ
adrenergic receptors, protein crystallography, inhibitors, kinases, paroxetine,
Takeda103A, GSK180736A, CCGꢀ215022, CCGꢀ224406
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT:
G proteinꢀcoupled receptors (GPCRs) are central to many physiological processes.
Regulation of this superfamily of receptors is controlled by GPCR kinases (GRKs), some
of which have been implicated in heart failure. GSK180736A, developed as a Rhoꢀ
associated coiledꢀcoil kinase 1 (ROCK1) inhibitor, was identified as an inhibitor of
GRK2, and coꢀcrystallized in the active site. Guided by its binding pose overlaid with the
binding pose of a known potent GRK2 inhibitor, Takeda103A, a library of hybrid
inhibitors was developed. This campaign produced several compounds possessing high
potency and selectivity for GRK2 over other GRK subfamilies, PKA, and ROCK1. The
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
most selective compound, 12n (CCGꢀ224406), had an IC for GRK2 of 130 nM, greater
50
than 700ꢀfold selectivity over other GRK subfamilies, and no detectable inhibition of
ROCK1. Four of the new inhibitors were crystallized with GRK2 to give molecular
insights into the binding and kinase selectivity of this class of inhibitors.
INTRODUCTION:
Heart failure is characterized by an inability of the heart to produce a myocardial
contraction strong enough to effectively pump blood throughout the body. To compensate
for the reduced output, the sympathetic nervous system increases the levels of
1
ꢀ3
catecholamines such as norepinephrine and epinephrine. These catecholamines bind to
the βꢀadrenergic receptors (βARs) on the exterior of the cell, leading to an increase in
cyclic adenosine 3´, 5´ꢀmonophosphate (cAMP) inside the cell, which in turn mediates
4
, 5
the positive inotropic effect needed for enhanced contraction. Once stimulated, the
βARs are then desensitized via phosphorylation of serine and threonine residues on their
2
ACS Paragon Plus Environment

Page 3 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
6, 7
cytoplasmic loops or tails by Gꢀprotein coupled receptor kinases (GRKs). Chronic
8
heart failure leads to increased desensitization and uncoupling of the βARs, which can
9
, 10
ultimately lead to death.
Current therapeutics, such as βꢀblockers, target this
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
mechanism by antagonizing βARs and reducing blood pressure, but take four to six
1
1
weeks to ameliorate dysregulation of βARs and produce cardio protective effects.
Furthermore, long term use of βꢀblockers leads to a myriad of side effects resulting from
1
2
bradycardia such as dizziness, weakness, and low blood pressure.
Inhibition of GRKs is another route by which heart failure might be addressed.
The seven GRKs found in vertebrates are grouped into three subfamilies, represented by
GRK1, GRK2, and GRK4. GRK2, a member of the GRK2 subfamily, and GRK5, a
member of the GRK4 subfamily, are known to play significant roles in the heart and have
7, 13
shown promise as alternative therapeutic targets for heart failure.
GRK2 is
3
, 9, 13
overexpressed during, and can serve as a biomarker for, heart failure.
Inhibition of
GRK2 disrupts internalization of βARs, leading to an increase in activated βARs and
other GPCRs on the cell membrane that will allow the heart to remain responsive to the
4
, 5
sympathetic nervous system. The effectiveness of targeting GRK2 has been
demonstrated in mice wherein cardiacꢀspecific inhibition of GRK2 using adenoꢀ
associated viral delivery of βARK , a portion of the cꢀterminus of GRK2 that acts as a
ct
dominant negative, following myocardial infarction led to improvement in both cardiac
14
function and survival. Furthermore, βARK mediated GRK2 inhibition in several
ct
animal models, including swine, indicates that inhibition of GRK2 activity improves
9
contractile function, reduces catecholamine levels, and reverses cardiac remodeling.
Notably, both GRK2 and GRK5 also act in nonꢀGPCR pathways that are beneficial to the
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 62
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
15, 16
heart.
GRK2 influences cardiac glucose uptake leading to abnormal cardiac
1
6
metabolism when GRK2 is upregulated, effecting the growth of new cardiomyocytes.
Uniquely, GRK5 becomes localized to the cell nuclei of cardiomyocytes upon GPCR
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
17, 18
15
activation
where it phosphorylates histone deacetylase 5 ( HDAC5). This
phosphorylation event activates myocyte enhancer factorꢀ2, which regulates the stress
1
9, 20
and fetal gene response underlying maladaptive cardiac hypertrophy.
Thus, in
addition to GRK2, GRK5 may also serve as a promising target for heart failure treatment.
GRKs belong to the protein kinase A, G and C (AGC) family and their kinase
domains share ~32% sequence identity with that of protein kinase A (PKA). When
targeting GRKs, it is imperative to identify potent chemical probes that are highly
selective against other AGC kinases as they play key roles in regulating many vital
physiological processes, including inotropy. There is, however, precedence for both
potent and selective chemical probes. The natural product balanol (Figure 1) inhibits
1
7,
GRK2 with an IC of 50 nM (at 3 µM ATP), but also potently inhibits PKA and PKC.
5
0
18
Takeda Pharmaceuticals developed a series of compounds, including Takeda103A (1)
1
9, 20
(Figure 1), that are highly selective for GRK2.
Compound 1, under our assay
conditions, has an IC for GRK2 of 20 nM (at 5 µM ATP) –– 50ꢀfold more potent than
5
0
other tested AGC kinases. Despite their impressive in vitro profile, these compounds
never advanced to clinical trials, presumably due to poor bioavailability.
The selective serotonin reuptake inhibitor paroxetine (Figure 1) was later
identified as a modestly potent inhibitor of GRK2 with an IC of 1.1 µM (at 5 µM ATP)
50
2
1, 22
and greater than 6ꢀfold selectivity over other GRKs.
The fact that paroxetine is an
FDAꢀapproved drug allowed therapeutic potential to be readily assessed and it was
4
ACS Paragon Plus Environment

Page 5 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
shown to improve left ventricular activity as well as decrease cardiac remodeling in a
2
3
mouse model of myocardial infarction. Most recently, GSK180736A (2), a compound
structurally similar to paroxetine that was developed as a Rhoꢀassociated, coiledꢀcoil
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
24
containing kinase 1 (ROCK1) inhibitor (Figure 1), was shown to be an even more
25
potent and selective inhibitor of GRK2 with an IC of 0.77 ꢁM and >100ꢀfold
50
selectivity over other GRKs. It is a weak inhibitor of PKA with an IC of 30 µM, but
5
0
highly potent against ROCK1 (IC = 100 nM). Therefore its use as a GRK2 inhibitor is
5
0
2
4
limited by its lack of selectivity versus ROCK1 as well as its poor bioavailability.
Structural and functional analysis of the GRK2·1 complex (PDB entry 3PVW)
suggests that the selectivity of compound 1 for GRK2 results from its ability to recognize
1
9
a conformation that is more readily accessible to GRK2 than to the other GRKs.
Docking of 1 into the inactive conformations of GRK1 and GRK6 predicts steric clashes
of the terminal difluorophenyl ring of 1 with residues that form what is referred to as the
hydrophobic subsite (Figure 2), a pocket positioned adjacent to the γ phosphate in the
1
9
ATP binding site. Alignment of the crystal complexes of 1, balanol, and 2 in GRK2
(Figure 2) revealed that although the compounds are chemically unrelated, they occupy
the adenine, ribose, and Pꢀloop subsites of the ATPꢀbinding pocket in remarkably similar
ways. However, unlike compound 1 and balanol, 2 does not have a substituent that can
25
occupy the hydrophobic subsite. We hypothesized that extending the scaffold of 2 into
the hydrophobic subsite would increase potency and, potentially, selectivity against other
2
6
AGC kinases such as PKA and ROCK1. Utilizing a hybrid approach, we added
moieties to the position ortho to the fluorine of the C ring of 2 through an amide linker
appending a fourth, soꢀcalled D ring (Figure 1). We have thereby successfully generated
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 62
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
more potent inhibitors that are highly selective for GRK2 as well as inhibitors that are
potent for both GRK2 and GRK5. Crystal structures of four representative compounds in
complex with GRK2 give insights into the molecular basis for the improved potency and
selectivity of these compounds.
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
RESULTS AND DISCUSSION:
Chemistry. Hybrid analogs 12aꢀr and 13aꢀc were prepared as described in Scheme 1.
Synthesis began with Fischer esterification of 4ꢀfluoroꢀ3ꢀmethylbenzoic acid 3 followed
by benzylic bromination under radical conditions to give the methyl ester 4. After
reduction of the ester, displacement of the bromine using sodium cyanide afforded the
nitrile 5. The nitrile was hydrolyzed under basic conditions to yield the carboxylic acid 6.
Oxidation of the benzylic alcohol of 6 using 2ꢀiodoxybenzoic acid or under Swern
conditions proved unsuccessful. Fortunately, Parikh Doering oxidation yielded the
aldehyde 7 cleanly and in high yield. In a converging pathway, treatment of 5ꢀ
aminoindazole 8 with 2,2,6ꢀtrimethylꢀ4Hꢀ1,3ꢀdioxinꢀ4ꢀone resulted in the acetylated
27
compound 9. Through a Biginelli cyclization, catalyzed by ytterbium triflate, either 2ꢀ
fluoroꢀ5ꢀformylbenzoic acid or intermediate 7 was condensed with urea and 9 to yield 10
2
8
or 11, respectively. Derivatives were then introduced via amidation resulting in the final
compounds 12aꢀr and 13aꢀc (summarized in Table 1).
Synthesis of the Nꢀmethylated hybrid molecule 20 commenced with acylation of
29
5
ꢀaminoindazole 8 using trifluoroacetic anhydride to give 14. Attempts at selective
methylation of the acetylated amine through the use of varying basic conditions led to
competing Nꢀmethylation of the indazole. Therefore, protection of the indazole nitrogen
was necessary, and was accomplished using Bocꢀanhydride to give two regioisomers of
6
ACS Paragon Plus Environment

Page 7 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
5. Methyl iodide in the presence of sodium hydride then furnished the methylated amide
6. Removal of the trifluoroacetate and Boc under refluxing basic conditions afforded the
monomethylated amine 17 which was then acylated with 2,2,6ꢀ trimethylꢀ4Hꢀ1,3ꢀdioxinꢀ
ꢀone to give 18. Final analogue 20 was completed using Biginelli cyclization followed
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
4
by amide coupling as described previously in Scheme 1.
Incorporation of the benzodioxole ring was undertaken as shown in Scheme 3.
Synthesis began with alkylation of 3,4ꢀ(methylenedioxy)aniline through heating with
ethyl acetoacetate and pyridine to give ketoamide 22. Biginelli cyclization followed by
HATUꢀmediated amidation furnished the hybrid compound 24.
Structure–Activity Relationships. Our hybrid design utilized an amide linker
between the template of 2 and the new substituents intended to fill the hydrophobic
subsite of GRK2. Although it was hypothesized that addition of a methyl amide group
alone would establish an extra hydrogen bond interaction with the Pꢀloop of GRK2
similar to that observed in the crystal complex with 1, prototype compound 12a exhibited
a 5ꢀfold loss in potency against GRK2. However, the compound maintained selectivity
against GRK1 and GRK5. It is possible that the additional hydrogen bond formed by the
amide with the enzyme cannot overcome a desolvation penalty or an entropic penalty for
fixing the conformation of the Pꢀloop. Notably, incorporation of the methyl amide did
result in a favorable loss in potency against PKA (IC > 100 ꢁM). The carboxylic acid
5
0
derivative 10, on the other hand, gave a dramatic 20ꢀfold decrease in potency for GRK2
and only a 2ꢀfold loss against PKA.
As both 1 and balanol place an aromatic ring into the hydrophobic pocket of
GRK2, we introduced a benzyl amide (12b), which was essentially equipotent with the
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 62
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
17,
lead (2) with respect to GRK2 and ROCK1, but gained substantial potency for GRK5.
1
9
Adding a single fluorine atom in the meta position of the benzyl (12c) resulted in a
“pan” GRK inhibitor, in that it inhibited GRK1, 2 and 5 but lacked PKA activity. The
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
results with both 12b and 12c suggest that addition of lipophilicity to the amide
appendage is a path to higher potency, but not necessarily GRK2 selectivity.
Incorporation of the 2,6ꢀdifluorobenyzl amide (12d), the same D ring as in Takeda103A,
did not improve upon the GRK2 potency of the 3ꢀfluorobenzyl amide, but restored
selectivity for GRK2 versus GRK1 and GRK5, providing a clue that the size of the amide
substituent may be an important selectivity determinant among GRKs.
Next, the effect of adding polar groups to the Dꢀring extension was explored as
the Dꢀring of balanol has two polar groups in the 2 and 6 positions. The 2ꢀmethoxy
benzyl amide 12e gave a 13ꢀfold increase in potency against GRK2 (IC = 60 nM) with
5
0
selectivity of 270ꢀ and 38ꢀfold over GRK1 and GRK5, respectively. Unfortunately, 12c
retained potency against ROCK1 (IC = 57 nM). Movement of the methoxy substituent
5
0
around the benzyl ring to the meta (12f) and para (12g) positions resulted in a significant
loss of potency compared to the 2ꢀmethoxy benzyl amide (12e) for GRK2 and GRK1, but
was more tolerated in GRK5 and ROCK1. Further investigation of the ortho position of
the benzyl ring led to the 2ꢀpyridylmethyl amide analog 12h which showed little
selectivity among the GRKs and ROCK1 (thus a panꢀinhibitor) having subꢀµM affinity
for all 3 GRKs and exhibiting a 10ꢀfold increase in potency over 2 for ROCK1.
Extending the linker between the amide bond and the terminal pyridine ring by one
carbon (12i) substantially improved selectivity in comparison to 12h against GRK1 and
GRK5 but somewhat reduced inhibition of GRK2. The 4ꢀpyridyl ethyl amide (12j) was
8
ACS Paragon Plus Environment

Page 9 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
not well tolerated by any of the GRKs, suggesting the importance of the 2ꢀpyridyl
nitrogen to binding.
We next investigated the potential of increasing the size of the pyridine of 12h to
an isoquinoline (12k), which improved selectivity against GRK5 6.4ꢀfold (versus 2.5ꢀfold
for 12h). However, there was no change in ROCK1 inhibition between 12h and 12k
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
(both ~10 nM). Interestingly, 12k achieved the highest potency and selectivity for GRK1
with an IC of 100 nM – a 50ꢀfold increase in comparison to the 2ꢀpyridylmethyl analog
5
0
12h and a greater than 100ꢀfold increase in comparison to the parent compound 2.
As the 2,6ꢀdifluorobenzyl (12d) showed marginally improved selectivity in
comparison to the unsubstituted benzyl 12a, the potential for achieving greater selectivity
for GRK2 by further increasing the size of the Dꢀring amide substituents was explored.
Both the 2,6ꢀdichloro (12l) and the 2,6ꢀdimethyl (12m) resulted in dramatic
improvements in GRK2 selectivity. Both analogs had over 770ꢀfold selectivity against
GRK1 and GRK5, as well as 50ꢀfold (12l) and 80ꢀfold (12m) selectivity against ROCK1.
Thus, for the first time we observed that ROCK inhibition can be mitigated via the
introduction of steric bulk into the hydrophobic subsite. Extending this line of reasoning
to the larger dimethoxy (12n) and diꢀtrifluoromethyl (12o) analogs essentially eliminated
all kinase inhibitory activity except for GRK2. The dimethoxy analog 12n retains
excellent potency against GRK2 compared to the parent compound (0.13 versus 0.77
µM) and is our most selective GRK2 inhibitor to date. Compared to the 2ꢀmethoxy
benzylamide analogue 12e, addition of the second methoxy in 12n was clearly a key step
in building in selectivity for GRK2 (over 175ꢀfold decrease in potency at ROCK1, 6ꢀfold
at GRK1, and 43ꢀfold at GRK5). Moving the added bulk to the meta positions (with the
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 62
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
3
,5ꢀbis(trifluoromethyl)benzyl amide 12p) produced a profound loss in GRK2 potency
revealing that the 2,6 disubstitution pattern of the D ring is critical for potency and
selectivity. The remarkable impact of sterics on GRK2 selectivity is further highlighted
by comparing 2ꢀpyridyl amide 12h with the corresponding orthoꢀmethyl analog 12q. The
latter compound, although less potent at GRK2, had increased selectivity versus GRK1
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
(>50ꢀfold) and GRK5 (36ꢀfold) versus 12h (33ꢀfold and 2ꢀfold, respectively).
In general, homologating the amide substituent by insertion of a methylene
between the amide and the aromatic ring led to decreases in potency and selectivity. This
was particularly evident with 13a and 13b. In each case, GRK2 potency dropped
precipitously and, oddly, subꢀµM ROCK1 activity returned. Homologation of 12n (13c)
resulted in a less dramatic decrease in GRK2 potency, but also restored ROCK1
inhibition. This interesting pattern was also observed with ethyl homolog 12r, which has
decreased GRK2 potency and increased ROCK1 potency versus the shorter 12m.
We briefly explored replacing the indazole moiety of 12h, which binds to the
22
hinge of the kinase domain, with the benzodioxole group of paroxetine (24, Table 2).
Because paroxetine has excellent selectivity for GRK2 versus ROCK1, we anticipated
that the benzodioxole might increase selectivity. Despite the potent panꢀkinase inhibition
of 12h, the corresponding benzodioxole analog 24 was completely inactive. Several other
benzodioxole analogs of other hybrids were similarly inactive (data not shown).
In a final attempt to further design out ROCK1 affinity, the nitrogen of the amide
bond linking the indazole and the dihydropyrimidone core was methylated. Prior
crystallization of a close analog of 2 into ROCK1 revealed that the amide nitrogen forms
24
a waterꢀmediated hydrogen bond to ROCK1ꢀAsp216, and therefore, methylation of the
10
ACS Paragon Plus Environment

Page 11 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
amide would be expected to weaken its affinity. The methylated amide of potent hybrid
1
2m was thus synthesized. The resulting compound 20 however showed ROCK1 potency
similar to 12m having an IC of 2.1 µM, and a nearly 70ꢀfold decrease in potency at
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
GRK2, discouraging us from pursuing additional Nꢀmethyl analogues.
GRK2–Gβγ·Inhibitor Crystal Structures: To gain insight into the molecular
basis for the relative potency and selectivity of this series of compounds, 12h, 12k, 12n,
and 12r were each coꢀcrystallized with GRK2–Gβγ and their resulting Xꢀray crystal
structures were determined (Figure 3, Supplemental Table 1). All four complexes
crystallized in space group C222 with nearly identical unit cell constants with resolution
1
ranges of 2.56 Å – 3.26 Å. In comparison to the previously reported structure of 2 in
2
5
complex with GRK2–Gβγ (PDB entry 4PNK) , all four inhibitor complexes exhibit
nearly identical kinase domain conformations with differences in relative rotations
between the large and small lobes no greater than 1˚. This result is consistent with the
idea that the indazole, which occupies the adenine subsite and forms two hydrogen bonds
with the hinge of the kinase domain, dictates the overall conformation of the large and
25, 26
small lobes.
As expected, the four inhibitors bind in the ATP pocket of GRK2 in essentially
the same manner as the parent compound 2, with the exception that the amide bond
connecting the indazole and the dihydropyrimidine is flipped relative to the model of 2
(Figure 3). However, the electron density for this amide in the complex with 2 is
ambiguous and hence the linker may adopt multiple configurations in the previous
structure. As noted above, the indazole rings bind in the adenine subsite forming two
hydrogen bonds with backbone atoms of the hinge residues Asp272 and Met274, and the
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 62
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
dihydropyrimidine and fluorophenyl rings fill the ribose and polyphosphate subsites,
respectively. However, the presence of the Dꢀrings, and presumably their interactions in
the hydrophobic subsite, seems to alter the conformation of the AꢀC rings to some extent
among the four complexes (Figure 4). Compared to the GRK2–Gβγ·2 complex, the 12h,
12k, and 12n compounds form additional hydrogen bonds in the ribose subsite with
Asp335 and Asn322. As predicted, their variable amideꢀlinked Dꢀrings occupy the
hydrophobic subsite of GRK2, and the carbonyls of the amide bond linker ortho to the
fluorine atom in the Cꢀring form a hydrogen bond with the backbone nitrogen of Phe202
in the Pꢀloop. The Dꢀring of 12r, however, flips out of the hydrophobic site towards the
solvent, and there is no interpretable electron density beyond the amide linker.
The largest conformational changes induced by the various inhibitors occur in the
Pꢀloop (Figure 4). Relative to 2, each of the four inhibitors causes the Pꢀloop to shift
away from the polyphosphate subsite as if to accommodate the terminal aromatic rings.
Compounds 12h, 12n, and 12r each have a maximum Pꢀloop shift of 2.2 Å at the Gly201
Cα relative to the parent structure (2). Compound 12k demonstrates the largest Pꢀloop
shift of 3.0 Å, possibly because it has the bulkiest substituent. In addition, the benzene
ring of Phe202 rotates to allow space for the terminal aromatic substituents depending on
their orientation (Figure 3). Notably, AST loop residues 487ꢀ493, which are typically
ordered in active conformations of AGC kinases, but are disordered in most GRK2 kinase
domain structures to date, are visible in the 12n electron density map and pack on top of
the Pꢀloop. The reason for these residues being more ordered in the 12n complex relative
to the others is unclear, but the density may simply reflect the higher quality of this
particular crystallographic data set (Supplemental Table 1).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
12
ACS Paragon Plus Environment

Page 13 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Molecular Basis for Potency and Selectivity: Analysis of these four crystal
structures in comparison to structures of other AGC kinase domains provides insight into
the molecular basis for their relative potencies and selectivities. Consistent with previous
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
30ꢀ32
studies,
the number of hydrogen bonds does not correlate well with inhibitor binding
affinity in AGC kinases (Figure 5). For example, compound 12r has the second highest
affinity of the four crystallized compounds from this study (12h, 12k, 12n, 12r) and the
parent compound 2, but only forms three hydrogen bonds. Plotting the K of 10 potent
i
GRK2 inhibitors (calculated from the IC values using the ChengꢀPrusoff
5
0
transformation) versus both the number of hydrogen bonds and the buried accessible
surface area (Figure 5) exhibited no correlation between the number of hydrogen bonds
and K but did show a correlation between buried surface area and K . The buried
i
i
accessible surface area is the most consistent determinants of inhibitor potency, as
3
2
previously noted for GRK2.
Interestingly, neither 2 nor any of its derivative compounds reported here inhibit
PKA below concentrations of 30 ꢁM despite its kinase domain being highly homologous
to those of GRKs and ROCK1. Comparison to both the PKA·AMPPNP substrate analog
structure (PDB entry 4HPT) and the PKAꢀbalanol complex (PDB entry 1BX6) indicates
that the derivatives (of compound 2) should not be sterically blocked from binding in the
PKA binding site. However, the hinge in the structures of PKA is shifted 1.5ꢀ1.7 Å away
from the adenine subsite relative to the position of the hinge in both GRK2 and GRK5
(
Figure 6). This difference may prevent the formation of favorable contacts between
hinge backbone atoms and the indazole ring common to 2 and all of its derivatives.
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 62
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
Compounds 2, 12n, and 12r do not bind tightly to either GRK1 or GRK5, but 12h
and 12k inhibit these kinases with potencies in the low µM or nM range. To explore the
molecular basis of this selectivity among GRK subfamilies, the structures of GRK5·12h
and GRK2·12h were compared (Figure 7). The hydrophobic subsite in GRK5 is deeper
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
and narrower than that of GRK2. Thus, one would predict that compounds with larger
Dꢀring substituents would tend to be excluded from the hydrophobic subsite of GRK5. In
the GRK5·12h complex, a hydrogen bond is formed between the catalytic lysine (Lys215
in GRK5) and the pyridine nitrogen of the inhibitor. In the crystal structure of
GRK2·12h, the pyridine nitrogen most likely faces the solvent. The fact that 12h buries
2
nearly identical surface area in each complex (only 5 Å additional buried surface area
when bound to GRK5) is consistent with its similar potencies against GRK2 and GRK5
and its behavior as a panꢀGRK inhibitor.
Based on the hydrophobic subsite hypothesis above, the isoquinoline ring of 12k
was predicted to select against GRK5 and its close homolog GRK1. However, it was an
efficacious inhibitor of all three GRKs as well as ROCK1, with 56ꢀfold and 1000ꢀfold
increases in potency against GRK5 and GRK1 relative to the parent compound 2. The
crystal structure shows that the isoquinoline reorganizes local structure in the
hydrophobic subsite of GRK2, in particular the Pꢀloop, such that its benzyl ring forms a
πꢀπ stacking interaction with the side chain of Phe202 (Figure 3b). Superimposing 12k
from the GRK2ꢀGβγ structure in the active sites of GRK1 and GRK5 illustrates that the
isoquinoline group would be compatible with their hydrophobic subsites as well,
provided their Pꢀloops can similarly reorganize. This modeling exercise also suggests that
the nitrogen of the isoquinoline ring could form an additional hydrogen bond with the
14
ACS Paragon Plus Environment

Page 15 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
active site lysine as 12h does in GRK5. A greater degree of kinase domain closure in
GRK1 and GRK5 relative to GRK2 may account for why enhancement of potency is
higher for GRK1 and GRK5.
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 dimethoxybenzyl Dꢀring of the most selective inhibitor, 12n, binds snugly in
the hydrophobic subsite of GRK2 with one methoxy group packing deep in the pocket,
and the other projecting towards solvent (Figures 3c, 8a). Docking the compound in the
active site of GRK5 (Figure 8b) demonstrates that the 2,6ꢀdimethoxybenzyl substituent of
12n would collide with the DFG loop, which is shifted towards the hydrophobic subsite
due to a greater degree of kinase domain closure in GRK5 than in GRK2. This collision
also likely explains the GRK2 selectivity seen with the similar, but not quite as bulky,
hybrids 12l and 12m. The packing of 12n is likely mimicked by the analogous potent
ortho methoxy hybrid 12e. Movement of the methoxy substituent to the meta and para
positions (as in 12f and 12g) would cause the methoxy to collide with the αCꢀhelix or
DFG loop, respectively, explaining their less favorable binding to GRK2.
Of the four inhibitors structurally characterized here, only 12n achieves
selectivity over both the GRK1 and GRK4 subfamilies and ROCK1. Upon superimposing
these inhibitor structures on ROCK1 in complex with an indazole derivative similar to 2
3
3
(
3V8S) , it is apparent that ROCK1ꢀPhe120 (Leu235 in GRK2) sterically clashes with
each of the four characterized inhibitor Dꢀrings. However, ROCK1ꢀPhe120 must be able
to adopt a rotamer more similar to that of GRK2ꢀLeu235 because 12h, 12k, and 12r all
inhibit ROCK1 with high potency. Compound 12n, however, also sterically clashes with
the backbone carbonyl of ROCK1ꢀGly218 (Gly337 in GRK2, Figure 8c) in the DFG
loop, which is shifted towards the hydrophobic subsite by 2.9 Å in ROCK1 relative to
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 62
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
GRK2. The other inhibitors avoid generating this collision but 12n cannot, as a
consequence of its two methoxy substituents, which greatly restrict its ability to alter its
conformation within the hydrophobic subsite. The analogous ethylene linked inhibitor
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
3c has low nanomolar affinity for ROCK1 despite having a bulky 2,6ꢀdimethoxyphenyl
substituent. It is likely that the extra carbon extension allows the inhibitor to pack outside
of the hydrophobic pocket to avoid steric clashes, as does 12r.
Electron density for the 2,6ꢀdimethyl phenethyl moiety of 12r is not evident after
the amide linker (Figure 3d), suggesting that this arm of the inhibitor is flexible while
bound to GRK2. This flexibility likely results from the extra degree of rotational freedom
conferred by the longer ethylene linker and its inability to be accommodated within the
hydrophobic subsite. For the same reasons, the Dꢀring of 12r likely would not be able to
occupy the hydrophobic subsites of GRK1 and GRK5. Thus, the compound would bind
no better than 2 alone, consistent with the relatively poor binding of 12r to GRK1 and
GRK5 (Table 1).
Contractility in Mouse Cardiomyocytes: To ascertain the ability of these
inhibitors to produce a myocardial effect, they were incubated with cardiomyocytes
isolated from adult mice, and then stimulated with the βAR agonist isoproterenol as
2
6
described previously. In the presence of a GRK2 inhibitor, the strength of the resulting
contraction should increase as there is more βAR signaling. Previously, it was shown that
paroxetine induces a significant increase in contractility in comparison to the DMSO
22
control when dosed at 10 ꢁM. In comparison, 2 showed similar efficacy as paroxetine
when dosed at 1 ꢁM (Figure 9), consistent with the increased potency of this compound
of GRK2 relative to paroxetine.
16
ACS Paragon Plus Environment

Page 17 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Four hybrid molecules were evaluated for their effects on cardiomyocyte
contractility: 12d, 12h, 12m, and 12n (Figure 9). Two of these were their highly selective
for GRK2 (12m and 12n) while the other two were more balanced GRK2/GRK5
inhibitors (12d and 12h). All of the hybrids proved to be equivalent to or slightly more
potent than the lead 2 at stimulating cardiomyocyte contraction. Unfortunately, it is not
possible to draw any firm conclusions from this limited study regarding whether different
levels of selectivity for GRK2 vs GRK5 lead to greater or lesser efficacy in the
cardiomyocytes. The GRK2 selectivities in this group ranged from 2ꢀfold (12h) to >700ꢀ
fold (12m and 12n), but no clear differences in cardiomyocyte potency are apparent.
Interpretation of these results is further complicated by unknown differences in cell
permeability of the hybrid analogs. Future studies may require the careful determination
of EC values with probes of known equivalent cell permeabilities to provide a definitive
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
5
0
answer to what GRK inhibition selectivity profile is optimum for achieving maximum
efficacy in cardiomyocytes.
CONCLUSIONS:
Utilizing a hybrid approach we envisioned development of selective and potent
GRK2 inhibitors constructed from the potent ROCK1 inhibitor 2 and the selective GRK2
inhibitor 1. Overlaying the binding poses of the two compounds (Figure 1) in the active
site of GRK2 revealed that it may be possible to utilize the hydrophobic binding site of
GRK2 to achieve selectivity over other AGC kinases. A small library of compounds was
thus synthesized in which a variety of amide substituents were appended to the fluoroꢀ
aromatic ring of 2 to mimic the difluorobenzyl amide of 1. The major SAR findings from
this study were that small benzyl amide substituents lead to significant inhibition of
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 62
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
GRK5 in addition to GRK2, and that addition of steric bulk to the benzyl amides favors
GRK2 selectivity over all of the other AGC kinases tested. This ultimately led to the
identification of 12n, a 2,6ꢀdimethoxybenzylamide with remarkable potency against and
selectivity for GRK2 and complete loss of the ROCK inhibitory activity of the lead 2. We
also identified some potent panꢀinhibitors of GRKs 1, 2 and 5, including 2ꢀpyridyl 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
1
2h. Collectively these new hybrid analogs represent an important start in the design of
small molecule probes that can be used to explore the physiological roles of the
individual GRKs. .
Coꢀcrystal structures of several representative analogs bound to GRK2 were
solved to help define the molecular basis for their differing selectivities. Comparison of
the GRK2ꢀ12h structure with the previously determined GRK5ꢀ12h crystal structure
showed that the hydrophobic subsite of GRK5 is much narrower and deeper than the
shallower, wider subsite of GRK2, helping to rationalize the increased GRK2 selectivity
seen with bulkier Dꢀring appendages. In addition, three other GRK2 inhibitor complexes
were crystallized: GRK2ꢀ12k, GRK2ꢀ12n, and GRK2ꢀ12r. Overall, the structures
confirm that the size and shape of the hydrophobic subsite, which is in part dependent on
the degree of kinase domain closure, puts constraints on the chemical nature of the Dꢀring
amide substituent that can be accommodated. Larger substituents, such as the 2,6ꢀ
dimethoxy benzyl of 12n, can fit into GRK2 more readily than the other AGC kinases
tested in this study. These conclusions were supported by overlays of the bound poses of
our analogs in GRK2 with other published kinase structures. Based on the published
GRK5ꢀ12h structure and the selectivity data we observed, we also conclude that smaller
18
ACS Paragon Plus Environment

Page 19 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
hydrogenꢀbonding benzyl amides can interact effectively with the hydrophobic subsite of
GRK5, conferring potent inhibitory activity to the GRK5ꢀinactive lead 2.
Finally, evaluation of selected inhibitors in cardiomyocytes suggests that panꢀ
GRK inhibition or dual GRK2/ GRK5 inhibition is just as effective as GRK2 inhibition
alone at stimulating contraction in the failing heart, but further studies will be needed to
confirm this hypothesis. In the future we plan to evaluate the pharmacokinetics of these
lead compounds with an eye towards their future development as potential heart failure
therapeutics.
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
Experimental Section:
Chemistry. All reagents were used without further purification as received from
commercial sources unless noted otherwise. 1H NMR and 13C NMR spectra were taken
in DMSOꢀd6 or MeOD at room temperature on Varian Inova 400 MHz or Varian Inova
5
00 MHz instruments. Reported chemical shifts for the 1H NMR and 13C NMR spectra
were recorded in parts per million (ppm) on the δ scale from an internal standard of
residual tetramethylsilane (0 ppm). Mass spectrometry data was measured using a Waters
Corporation LCT. HPLC was used to determine purity of biologically tested compounds
on an Agilent 1100 series with an Agilent Zorbax Eclipse Plus−C18 column. A gradient
of 10ꢀ90% acetonitrile/water over 13 min was used with detection at 254 nm. All tested
compounds had purity >95%. Solvent abbreviations used: MeOH (methanol), DCM
(dichloromethane), EtOAc (ethyl acetate), hex (hexanes), DMSO (dimethylsulfoxide),
DMF (dimethylformamide), H O (water), THF (tetrahydrofuran). Reagent abbreviations
2
used: DIBAL (diisobutylaluminum hydride), HATU (1ꢀ[Bis(dimethylamino)methylene]ꢀ
1
Hꢀ1,2,3ꢀtriazolo[4,5ꢀb]pyridinium 3ꢀoxid hexafluorophosphate), DIEA
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 62
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
(
diisopropylethylamine), MgSO (magnesium sulfate), NaHCO (sodium bicarbonate),
4
3
Na CO (sodium carbonate), NH Cl (ammonium chloride), NaCl (sodium chloride),
2
3
4
K CO (potassium carbonate), H SO (sulfuric acid), and HCl (hydrogen chloride).
2
3
2
4
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
Methyl 4ꢀfluoroꢀ3ꢀmethylbenzoate (3b). To a 100 mL round bottom flask with 40 mL
of methanol was added 4ꢀfluoroꢀ3ꢀmethyl benzoic acid (4.0 g, 25.96 mmol) followed by
concentrated H SO (1.0 mL). The reaction vessel was equipped with a condenser and
2
4
refluxed overnight at 65 °C. Cooled down reaction and diluted with ether and
NaHCO .The organic layer was washed with brine (3x), dried over MgSO , and
3
4
concentrated at room temperature under reduced pressure to give 3b as a clear liquid
1
(98% yield, 4.38g). H NMR (400 MHz, DMSOꢀd ) δ 7.90 (dd, J = 7.5, 2.2 Hz, 1H), 7.83
6
(ddd, J = 7.9, 5.0, 2.3 Hz, 1H), 7.27 (t, J = 9.1 Hz, 1H), 3.83 (s, 3H), 2.28 (s, 3H). HPLC
purity: 100%.
Methyl 3ꢀ(bromomethyl)ꢀ4ꢀfluorobenzoate (4). To a 250 mL closed pressure vessel
was added 3 (4.02g, 23.91 mmol) dissolved in anhydrous carbon tetrachloride (40 mL).
Nꢀbromosuccinimide (3.83g, 21.50 mmol) was then added followed by benzoyl peroxide
o
(0.290g, 1.20 mmol). The reaction stirred for 1 hour at 65 C then cooled and filtered
through celite with dichloromethane and concentrated. Purified using 0%ꢀ1%
EtOAc/Hexanes using flash chromatography to give 4 as a clear oil (82% yield, 4.31g).
1
H NMR (400 MHz, DMSOꢀd ) δ 8.19 (dd, J = 7.4, 2.3 Hz, 1H), 8.00 (ddd, J = 8.7, 5.0,
6
2
.3 Hz, 1H), 7.41 (dd, J = 9.8, 8.7 Hz, 1H), 4.79 (d, J = 1.1 Hz, 2H), 3.87 (s, 3H).
(3ꢀ(Bromomethyl)ꢀ4ꢀfluorophenyl)methanol (4b) To a dry 250 mL flask was added
toluene (30 mL) followed by 1M DIBAL in toluene (23.16 mL, 23.16mmol). The flask
was then cooled to 0 °C and then 4 (2.73 g, 11.05 mmol) in toluene (5 mL) was added.
20
ACS Paragon Plus Environment

Page 21 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The reaction stirred for 2 hours and was then quenched (on ice) with 2N HCl to a pH of
2
. The reaction was extracted with ethyl acetate (2x), the organic layers were combined
and washed with NaCl (2x), dried over MgSO , and concentrated to give 4b as a white
4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
solid (94% yield, 2.09g). H NMR (500 MHz, DMSOꢀd ) δ 7.46 (dd, J = 7.5, 2.2 Hz,
6
1H), 7.32 (ddd, J = 7.8, 5.1, 2.2 Hz, 1H), 7.18 (dd, J = 10.0, 8.4 Hz, 1H), 5.28 (s, 1H),
4
.69 (s, 2H), 4.46 (s, 2H).
2
ꢀ(2ꢀFluoroꢀ5ꢀ(hydroxymethyl)phenyl)acetonitrile (5). To a 250 mL round bottom
flask was added 4b (2.06 g, 10.26 mmol) dissolved in DMSO (40 mL) followed by
sodium cyanide (1.51 g, 30.77 mmol). After stirring at room temperature for 2 hours the
reaction was diluted with H O and ethyl acetate. The layers were separated and the
2
aqueous layer was further extracted with ethyl acetate (2x). The organic layers were
combined, washed 3x with NaCl, dried over MgSO , and concentrated. Using flash
4
chromatography (15% ꢀ 30% EtOAc/Hexanes) the residue was purified to obtain 5 as a
1
clear liquid (1.50 g, 99% yield). H NMR (400 MHz, DMSOꢀd ) δ 7.42 (dd, J = 7.5, 2.1
6
Hz, 1H), 7.32 (ddd, J = 7.7, 5.1, 2.1 Hz, 1H), 7.21 (dd, J = 9.9, 8.4 Hz, 1H), 5.31 (t, J =
5.7 Hz, 1H), 4.48 (d, J = 5.6 Hz, 2H), 4.06 (s, 2H).
2
ꢀ(2ꢀFluoroꢀ5ꢀ(hydroxymethyl)phenyl)acetic acid (6). To a 100 mL round bottom flask
was added 5 (0.881 g, 5.33 mmol) followed by 2N NaOH (20 mL). The reaction vessel
was equipped with a condenser and refluxed overnight. After cooling the reaction down
HCl was added to reach a pH of 3. The reaction was extracted with ethyl acetate (2x).
The organic layers were combined, washed with NaCl (2x), dried over MgSO , and
4
1
concentrated in vacuo to give 6 as a white solid (0.969 g, 99% yield). H NMR (400
MHz, DMSOꢀd ) δ 12.45 (s, 1H), 7.28 – 7.20 (m, 2H), 7.11 (dd, J = 9.7, 8.5 Hz, 1H),
6
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 62
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
5
.22 (s, 1H), 4.45 (s, 2H), 3.60 (s, 2H).
ꢀ(2ꢀFluoroꢀ5ꢀformylphenyl)acetic acid (7). To a 100 mL round bottom flask at room
2
temperature was added 6 (1.26g, 7.58 mmol), triethylamine (8.46 mL, 60.66 mmol), and
DMSO (20 mL). Sulfur trioxide pyridine complex (4.83 g, 30.33 mmol) in DMSO (20
mL) was then added to the reaction and the reaction was stirred for 20 minutes. The
reaction was diluted with Na CO then extracted with ether. To the aqueous layer 2N HCl
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
3
was added to give a pH of 3. The aqueous layer was then extracted with ethyl acetate
(2x). The ethyl acetate layers were combined and washed with NaCl (2x), dried over
1
MgSO , and concentrated in vacuo to give 7 as a white solid (1.24g, 100% yield). H
4
NMR (400 MHz, DMSOꢀd ) δ 12.49 (s, 1H), 9.96 (s, 1H), 7.98 – 7.85 (m, 2H), 7.43 (dd,
6
J = 9.6, 8.3 Hz, 1H), 3.76 (d, J = 1.4 Hz, 2H).
Nꢀ(1HꢀIndazolꢀ5ꢀyl)ꢀ3ꢀoxobutanamide (9): To a 50 mL pressure vessel was added 1Hꢀ
indazolꢀ5ꢀamine (0.800 g, 6.00 mmol), 2,2,6ꢀtrimethylꢀ4Hꢀ1,3ꢀdioxinꢀ4ꢀone (1.57 g,
1
2.00 mmol), sodium acetate (0.492 g, 6.00 mmol), and acetonitrile (5 mL). The reaction
was stirred overnight at 100˚C. The reaction was diluted with citric acid (10%) and
EtOAc. After separating, the organic layer was washed 2x with NaCl, dried over MgSO4,
and concentrated to give a brown residue. The residue was then triturated with 2:1
dichloromethane:ethyl acetate and filtered off to give a light brown solid. The solid was
then purified on flash chromatography in a gradient of 0% to 3% MeOH/DCM to afford 9
1
as an off white solid (0.901 g, 69% yield). H NMR (400 MHz, DMSOꢀd6) δ 12.99 (s,
1
H), 10.08 (s, 1H), 8.12 (s, 1H), 8.02 (s, 1H), 7.49 (d, J = 8.9 Hz, 1H), 7.38 (d, J = 9.2
Hz, 1H), 3.56 (s, 2H), 2.23 (d, J = 1.6 Hz, 3H).
22
ACS Paragon Plus Environment

Page 23 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5
ꢀ(5ꢀ((1HꢀIndazolꢀ5ꢀyl)carbamoyl)ꢀ6ꢀmethylꢀ2ꢀoxoꢀ1,2,3,4ꢀtetrahydropyrimidinꢀ4ꢀ
yl)ꢀ2ꢀfluorobenzoic acid (10). To a 100 mL pressure vessel equipped with a stir bar was
added 2ꢀfluoroꢀ5ꢀformylbenzoic acid (0.682 g, 4.054 mmol), 9 (0.801 g, 3.686 mmol),
urea (0.332 g, 5.529 mmol), ytterbium trifluormethanesulfonate (0.229 g, 0.3686 mmol)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
o
and acetonitrile (15.0 mL). The reaction mixture was heated to 100 C and stirred. The
reaction mixture went from a white suspension to a clear solution, and then back to a
white suspension. After four hours, the reaction was diluted with 2 mL of water then
ethyl acetate (~60 mL). The resulting white precipitate was filtered off and washed with
1
ethyl acetate followed by diethyl ether yielding 10 as a white powder (84% yield). H
NMR (400 MHz, DMSOꢀd ) δ 13.13 (s, 1H), 9.59 (s, 1H), 8.79 (s, 1H), 7.98 (t, J = 2.1
6
Hz, 2H), 7.88 – 7.78 (m, 1H), 7.64 (s, 1H), 7.50 (s, 1H), 7.46 – 7.33 (m, 2H), 7.33 – 7.22
(m, 1H), 5.42 (s, 1H), 2.06 (s, 3H).
2
4
ꢀ(5ꢀ(5ꢀ((1HꢀIndazolꢀ5ꢀyl)carbamoyl)ꢀ6ꢀmethylꢀ2ꢀoxoꢀ1,2,3,4ꢀtetrahydropyrimidinꢀ
ꢀyl)ꢀ2ꢀfluorophenyl)acetic acid (11). Compound 11 was prepared as described for
1
compound 10, replacing 2ꢀfluoroꢀ5ꢀformylbenzoic acid with 7 (72% yield). H NMR
(400 MHz, DMSOꢀd ) δ 12.80 (s, 1H), 12.53 (s, 1H), 9.54 (s, 1H), 8.69 (d, J = 2.0 Hz,
6
1
H), 7.99 – 7.92 (m, 2H), 7.58 (s, 1H), 7.42 – 7.33 (m, 2H), 7.26 – 7.16 (m, 2H), 7.11 (t,
J = 9.1 Hz, 1H), 5.38 (s, 1H), 3.56 (s, 2H), 2.04 (s, 3H).
4ꢀ(4ꢀFluoroꢀ3ꢀ(methylcarbamoyl)phenyl)ꢀNꢀ(1Hꢀindazolꢀ5ꢀyl)ꢀ6ꢀmethylꢀ2ꢀoxoꢀ
1
,2,3,4ꢀtetrahydropyrimidineꢀ5ꢀcarboxamide (12a) : Compound 10 (0.100 g, 0.244
mmol), HATU (0.185 g, 0.489 mmol), DIEA (0.085 mL, 0.489 mmol), and 2N
methylamine (0.245 mL, 0.489 mmol) were added to 5 mL of DMF in a 25 mL round
bottom flask and allowed to stir overnight at room temperature. The reaction was diluted
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 62
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
with water and ethyl acetate giving a white suspension in the organic layer. The layers
were separated and the organic layer was washed once with Na CO and twice with
2
3
NaCl. The organic suspension was then filtered off and washed with water giving a white
solid. The solid was then washed with dichloromethane and purified using flash
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
chromatography (0ꢀ15% MeOH/DCM) yielding 12a as a white solid: 80.7 mg (78%). H
NMR (500 MHz, DMSOꢀd ) δ 12.93 (s, 1H), 9.58 (s, 1H), 8.75 (d, J = 1.9 Hz, 1H), 8.20
6
(t, J = 1.3 Hz, 1H), 7.98 (d, J = 7.1 Hz, 2H), 7.63 – 7.57 (m, 2H), 7.44 – 7.35 (m, 3H),
7.24 (dd, J = 10.3, 8.6 Hz, 1H), 5.43 (d, J = 2.9 Hz, 1H), 3.17 (d, J = 4.6 Hz, 1H), 2.75
13
(
d, J = 4.6 Hz, 3H), 2.07 (s, 3H). C NMR (126 MHz, DMSO) δ 164.92, 163.71, 152.24,
1
40.48, 138.29, 136.77, 133.18, 131.90, 130.01, 128.19, 123.70, 122.50, 120.85, 115.97,
115.79, 110.25, 109.72, 104.96, 54.32, 26.17, 16.98. HPLC purity: 96%; MS (ESI+) m/z:
4
4
1
45.1 (M+1).
ꢀ(3ꢀ(Benzylcarbamoyl)ꢀ4ꢀfluorophenyl)ꢀNꢀ(1Hꢀindazolꢀ5ꢀyl)ꢀ6ꢀmethylꢀ2ꢀoxoꢀ
,2,3,4ꢀtetrahydropyrimidineꢀ5ꢀcarboxamide (12b): Compound 12b was prepared as
described for 12a, replacing methylamine with benzylamine. Yield: 21.0 mg, 17% yield).
1
H NMR (500 MHz, DMSOꢀd ) δ 12.93 (s, 1H), 9.59 (s, 1H), 8.83 (td, J = 5.4, 1.5 Hz,
6
1
H), 8.77 (d, J = 1.4 Hz, 1H), 7.98 (d, J = 18.6 Hz, 2H), 7.65ꢀ7.60 (m, 2H), 7.44 – 7.37
(m, 3H), 7.32 – 7.20 (m, 6H), 5.45 (d, J = 2.9 Hz, 1H), 4.45 (d, J = 6.1 Hz, 2H), 2.08 (s,
1
3
3H). C NMR (126 MHz, DMSO) δ 164.92, 163.46, 159.13, 157.15, 152.26, 140.51,
1
1
39.09, 138.40, 136.78, 133.18, 131.91, 130.12, 128.14, 126.95, 126.62, 123.71, 123.59,
22.51, 120.85, 116.06, 115.87, 110.26, 109.72, 104.91, 54.28, 42.52, 16.99. HPLC
purity: 95%; MS (ESI+) m/z: 499.1 (M+1).
ꢀ(4ꢀFluoroꢀ3ꢀ((3ꢀfluorobenzyl)carbamoyl)phenyl)ꢀNꢀ(1Hꢀindazolꢀ5ꢀyl)ꢀ6ꢀmethylꢀ2ꢀ
4
24
ACS Paragon Plus Environment

Page 25 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
oxoꢀ1,2,3,4ꢀtetrahydropyrimidineꢀ5ꢀcarboxamide (12c): Compound 12c was prepared
as described for 12a, replacing methylamine with 3ꢀfluorobenzylamine. Yield: 46.8 mg
1
(
37%). H NMR (500 MHz, DMSOꢀd ) δ 12.90 (s, 1H), 9.58 (s, 1H), 8.88 (s, 1H), 8.76
6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
s, 1H), 7.97 (d, J = 18.0 Hz, 2H), 7.63 (s, 2H), 7.46 – 7.31 (m, 4H), 7.27 (t, J = 9.4 Hz,
1H), 7.12 (dd, J = 15.2, 9.0 Hz, 2H), 7.08 – 7.02 (m, J = 11.8, 5.3 Hz, 1H), 5.44 (s, 1H),
1
3
4
.46 (d, J = 5.8 Hz, 2H), 2.07 (s, 3H). C NMR (126 MHz, DMSO) δ 171.71, 165.49,
1
64.12, 163.62, 161.69, 159.73, 157.75, 152.83, 142.48, 141.11, 139.00, 132.47, 130.85,
130.67, 128.82, 124.02, 123.50, 123.06, 121.44, 116.57, 114.18, 113.97, 110.85, 110.36,
05.45, 54.85, 42.65, 17.56. HPLC purity: 99%; MS (ESI+) m/z: 516.9 (M+1), 538.8
M+Na+).
1
(
4ꢀ(3ꢀ((2,6ꢀDifluorobenzyl)carbamoyl)ꢀ4ꢀfluorophenyl)ꢀNꢀ(1Hꢀindazolꢀ5ꢀyl)ꢀ6ꢀ
methylꢀ2ꢀoxoꢀ1,2,3,4ꢀtetrahydropyrimidineꢀ5ꢀcarboxamide (12d): Compound 12d
was prepared as described for 12a, replacing methylamine with 2,6ꢀdifluorobenzylamine.
1
Yield: 60.1 mg (31%). H NMR (400 MHz, DMSOꢀd ) δ 12.91 (s, 1H), 9.55 (s, 1H), 8.74
6
(s, 1H), 8.70 (s, 1H), 7.96 (d, J = 10.5 Hz, 2H), 7.59 (s, 1H), 7.52 (d, J = 6.6 Hz, 1H),
7.42ꢀ7.35 (m, 4H), 7.19 (t, J = 9.5 Hz, 1H), 7.05 (t, J = 7.8 Hz, 2H), 5.40 (s, 1H), 4.47 (d,
1
3
J = 4.9 Hz, 2H), 2.05 (s, 3H). C NMR (126 MHz, DMSO) δ 164.90, 163.22, 152.25,
40.41, 138.43, 136.78, 133.17, 131.89, 130.11, 129.72, 128.12, 123.47, 122.50, 120.87,
115.96, 115.78, 113.79, 111.43, 111.23, 110.28, 109.72, 109.42, 104.88, 54.23, 31.19,
1
1
4
2
6.99. HPLC purity: 98%; MS (ESI+) m/z: 534.9 (M+1), 556.8 (M+Na+).
ꢀ(4ꢀFluoroꢀ3ꢀ((2ꢀmethoxybenzyl)carbamoyl)phenyl)ꢀNꢀ(1Hꢀindazolꢀ5ꢀyl)ꢀ6ꢀmethylꢀ
ꢀoxoꢀ1,2,3,4ꢀtetrahydropyrimidineꢀ5ꢀcarboxamide (12e): Compound 12e was
prepared as described for 12a, replacing methylamine with 2ꢀmethoxybenzylamine.
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
Yield: 120 mg (46%). H NMR (400 MHz, DMSOꢀd ) δ 12.94 (s, 1H), 9.60 (s, 1H), 8.78
6
(
s, 1H), 8.64 (s, 1H), 8.01 (s, 1H), 7.97 (s, 1H), 7.67 – 7.61 (m, 2H), 7.45 – 7.36 (m, 3H),
7
.31 – 7.17 (m, 3H), 6.98 (d, J = 8.0 Hz, 1H), 6.88 (t, J = 7.4 Hz, 1H), 5.45 (s, 1H), 4.42
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
13
(
d, J = 6.0 Hz, 2H), 3.81 (s, 3H), 2.08 (s, 3H). C NMR (126 MHz, DMSO) δ 164.92,
163.45, 163.44, 157.21, 156.46, 152.27, 140.52, 138.43, 136.78, 133.18, 131.91, 130.14,
1
1
28.26, 127.88, 127.05, 126.29, 123.63, 122.51, 120.86, 119.99, 115.96, 110.34, 109.57,
04.91, 55.20, 54.28, 37.75, 16.99. HPLC purity: 95%; MS (ESI+) m/z: 528.8 (M+1),
550.8 (M+Na+).
4
2
ꢀ(4ꢀFluoroꢀ3ꢀ((3ꢀmethoxybenzyl)carbamoyl)phenyl)ꢀNꢀ(1Hꢀindazolꢀ5ꢀyl)ꢀ6ꢀmethylꢀ
ꢀoxoꢀ1,2,3,4ꢀtetrahydropyrimidineꢀ5ꢀcarboxamide (12f): Compound 12f was
prepared as described for 12a, replacing methylamine with (3ꢀmethoxyphenyl)
1
methanamine. Yield: 74.3 mg (58%). H NMR (400 MHz, DMSOꢀd ) δ 12.93 (s, 1H),
6
9
.58 (s, 1H), 8.82 (broad s, 1H), 8.76 (s, 1H), 7.98 (d, J = 16 Hz, 2H), 7.65 – 7.57 (m,
H), 7.46 – 7.34 (m, 3H), 7.32 – 7.17 (m, 2H), 6.91 – 6.84 (m, 2H), 6.80 (dt, J = 8.2, 1.4
2
Hz, 1H), 5.45 (d, J = 2.8 Hz, 1H), 4.43 (d, J = 6.1 Hz, 2H), 3.72 (s, 3H), 2.07 (s, 3H).
HPLC purity: 95%; MS (ESI+) m/z: 529.1 (M+1), 551.1 (M+Na+).
4
2
ꢀ(4ꢀFluoroꢀ3ꢀ((4ꢀmethoxybenzyl)carbamoyl)phenyl)ꢀNꢀ(1Hꢀindazolꢀ5ꢀyl)ꢀ6ꢀmethylꢀ
ꢀoxoꢀ1,2,3,4ꢀtetrahydropyrimidineꢀ5ꢀcarboxamide (12g): Compound 12g was
prepared as described for 12a, replacing methylamine with 4ꢀmethoxybenzylamine.
1
Yield: 10.7 mg (8%). H NMR (400 MHz, DMSOꢀd ) δ 12.94 (s, 1H), 9.60 (s, 1H), 8.77
6
(
s, 2H), 7.98 (d, J = 14.1 Hz, 2H), 7.63 (s, 1H), 7.60 (dd, J = 6.9, 2.3 Hz, 1H), 7.44 – 7.35
(
m, 3H), 7.27 (d, J = 10.2 Hz, 1H), 7.22 (d, J = 8.5 Hz, 2H), 6.86 (d, J = 8.7 Hz, 2H),
13
5
.44 (s, 1H), 4.37 (d, J = 6.0 Hz, 2H), 3.71 (s, 3H), 2.07 (s, 3H). C NMR (101 MHz,
26
ACS Paragon Plus Environment

Page 27 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
DMSO) δ 165.47, 163.86, 158.60, 157.43, 152.83, 138.93, 137.32, 133.74, 132.46,
1
1
1
31.57, 128.91, 128.73, 125.28, 124.22, 123.06, 121.40, 116.62, 116.39, 114.09, 110.80,
10.29, 105.47, 55.47, 54.81, 42.53, 17.55. HPLC purity: 99%; MS (ESIꢀ) m/z: 527.0 (Mꢀ
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
).
4ꢀ(4ꢀFluoroꢀ3ꢀ((pyridinꢀ2ꢀylmethyl)carbamoyl)phenyl)ꢀNꢀ(1Hꢀindazolꢀ5ꢀyl)ꢀ6ꢀ
methylꢀ2ꢀoxoꢀ1,2,3,4ꢀtetrahydropyrimidineꢀ5ꢀcarboxamide (12h): Compound 12h
was prepared as described for 12a, replacing methylamine with 2ꢀ(aminomethyl)pyridine.
1
Yield: 41.2 mg (20%) H NMR (400 MHz, DMSOꢀd ) δ ppm 12.93, 9.59, 8.87, 8.77,
6
1
8
.50, 7.98, 7.73, 7.69, 7.63, 7.46ꢀ7.37, 7.33ꢀ7.24, 5.46, 4.56, 2.08. H NMR (400 MHz,
DMSOꢀd ) δ 12.96 (s, 1H), 9.64 (s, 1H), 8.91 (q, J = 5.3 Hz, 1H), 8.76 (d, J = 2.0 Hz,
6
1H), 8.48 (d, J = 4.3 Hz, 1H), 7.99 (s, 1H), 7.94 (s, 1H), 7.71 (td, J = 7.7, 1.8 Hz, 1H),
7
.66 (dd, J = 6.9, 2.4 Hz, 1H), 7.62 (t, J = 2.4 Hz, 1H), 7.47 – 7.36 (m, 3H), 7.33 – 7.20
1
3
(
m, 3H), 5.45 (d, J = 2.9 Hz, 1H), 4.53 (d, J = 5.9 Hz, 2H), 2.06 (s, 3H). C NMR (126
MHz, DMSO) δ 164.89, 164.024, 158.56, 157.90, 152.83, 149.28, 141.14, 138.93,
37.17, 133.74, 132.47, 130.94, 128.94, 123.77, 123.65, 122.55,121.43, 121.12, 116.71,
116.52, 110.83, 110.30, 105.49, 54.88, 45.15, 17.55. HPLC purity: 97%; MS (ESI+) m/z:
1
5
00.3 (M+1), 522.2 (M+Na+).
4
ꢀ(4ꢀFluoroꢀ3ꢀ((2ꢀ(pyridinꢀ2ꢀyl)ethyl)carbamoyl)phenyl)ꢀNꢀ(1Hꢀindazolꢀ5ꢀyl)ꢀ6ꢀ
methylꢀ2ꢀoxoꢀ1,2,3,4ꢀtetrahydropyrimidineꢀ5ꢀcarboxamide: Compound 12i was
prepared as described for 12a, replacing methylamine with 2ꢀpyridinꢀ2ꢀylethylamine.
1
1
Yield: 38 mg (20%). H NMR (400 MHz, DMSO) δ ppm H NMR (400 MHz, DMSOꢀ
d6) δ 12.92 (s, 1H), 9.58 (s, 1H), 8.77 (s, 1H), 8.47 (d, J = 4.0 Hz, 1H), 8.37 (dd, J = 8.2,
5
.1 Hz, 1H), 7.97 (d, J = 13.9 Hz, 2H), 7.67 (td, J = 7.6, 1.8 Hz, 1H), 7.62 (s, 1H), 7.57
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 62
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
(
dd, J = 6.9, 2.3 Hz, 1H), 7.43 – 7.34 (m, 3H), 7.26ꢀ7.15 (m 3H), 5.42 (d, J = 2.1 Hz,
1
3
1
H), 3.57 (dd, J = 13.2, 6.9 Hz, 2H), 2.94 (t, J = 7.2 Hz, 2H), 2.05 (s, 3H). C NMR (126
MHz, DMSO) δ 164.91, 163.16, 159.33, 158.86, 156.86, 152.26, 148.93, 140.49, 138.32,
36.75, 136.34, 133.18, 131.90, 130.03, 128.17, 123.05, 122.50, 121.40, 120.84, 116.02,
110.238, 109.74, 104.93, 54.31, 36.90, 16.99. HPLC purity: 95%; MS (ESI+) m/z: 514.0
M+1), 536.0 (M+Na+).
ꢀ(4ꢀFluoroꢀ3ꢀ((2ꢀ(pyridinꢀ4ꢀyl)ethyl)carbamoyl)phenyl)ꢀNꢀ(1Hꢀindazolꢀ5ꢀyl)ꢀ6ꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
(
4
methylꢀ2ꢀoxoꢀ1,2,3,4ꢀtetrahydropyrimidineꢀ5ꢀcarboxamide: Compound 12j was
prepared as described for 12a, replacing methylamine with 2ꢀ(4ꢀpyridyl)ethyl amine.
1
Yield: 95.8 mg (51%). H NMR (400 MHz, DMSO) δ 12.94 (s, 1H), 9.59 (s, 1H), 8.78 (s,
1H), 8.45 (d, J = 5.5 Hz, 2H), 8.38 (s, 1H), 7.99 (d, J = 13.9 Hz, 2H), 7.63 (s, 1H), 7.55
(d, J = 4.7 Hz, 1H), 7.47 – 7.34 (m, 3H), 7.29 – 7.18 (m, J = 7.5 Hz, 3H), 5.44 (s, 1H),
1
3
3
.50 (dd, J = 12.8, 6.6 Hz, 2H), 2.82 (t, J = 7.0 Hz, 2H), 2.07 (s, 3H). C NMR (126
MHz, DMSO) δ 165.48, 163.94, 152.83, 149.87, 148.68, 141.05, 138.95, 137.34, 133.75,
32.47, 130.63, 128.61, 124.50, 123.07, 121.42, 116.56, 116.37, 114.22, 110.82, 110.30,
105.48, 54.86, 39.47, 34.50, 17.55. HPLC purity: 99%; MS (ESI+) m/z: 514.1 (M+1),
1
5
36.1 (M+Na+).
4
ꢀ(4ꢀFluoroꢀ3ꢀ((isoquinolinꢀ1ꢀylmethyl)carbamoyl)phenyl)ꢀNꢀ(1Hꢀindazolꢀ5ꢀyl)ꢀ6ꢀ
methylꢀ2ꢀoxoꢀ1,2,3,4ꢀtetrahydropyrimidineꢀ5ꢀcarboxamide (12k): Compound 12k
was prepared as described for 12a, replacing methylamine with isoquinolin1ꢀ
1
ylmethylamine. Yield: 5.3 mg (4%). H NMR (400 MHz, DMSOꢀd ) δ ppm 12.92 (s,
6
1
H), 9.58 (s, 1H), 8.87 (d, J = 5.7 Hz, 1H), 8.77 (s, 1H), 8.42 (d, J = 5.7 Hz, 1H), 8.29 (d,
J = 8.4 Hz, 1H), 8.00 (d, J = 7.3 Hz, 2H), 7.95 (s, 1H), 7.83ꢀ7.73 (m, 3H), 7.69 (t, J = 7.6
28
ACS Paragon Plus Environment

Page 29 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Hz, 1H), 7.63 (s, 1H), 7.47ꢀ7.34 (m, 3H), 7.28 (t, J = 9.5 Hz, 1H), 5.46 (s, 1H), 5.13 (d,
J = 5.0 Hz, 2H), 2.06 (s, 3H). MS (ESI+) m/z: 550.1 (M+1). HPLC purity: 98%.
4
ꢀ(3ꢀ((2,6ꢀDichlorobenzyl)carbamoyl)ꢀ4ꢀfluorophenyl)ꢀNꢀ(1Hꢀindazolꢀ5ꢀyl)ꢀ6ꢀ
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
methylꢀ2ꢀoxoꢀ1,2,3,4ꢀtetrahydropyrimidineꢀ5ꢀcarboxamide (12l): Compound 12l was
prepared as described for 12a, replacing methylamine with 2,6ꢀdichlorobenzylamine.
1
Yield: 68.0 mg (49%). H NMR (500 MHz, DMSOꢀd ) δ 12.91 (s, 1H), 9.56 (s, 1H),
6
8
.74 (d, J = 2.0 Hz, 1H), 8.55 (t, J = 5.3 Hz, 1H), 7.98 (s, 1H), 7.95 (s, 1H), 7.61 – 7.59
(m, 1H), 7.52 (dd, J = 6.7, 2.4 Hz, 1H), 7.45 (s, 1H), 7.43 (s, 1H), 7.41 – 7.35 (m, 3H),
7
.33 (dd, J = 8.5, 7.6 Hz, 1H), 7.18 (dd, J = 10.0, 8.7 Hz, 1H), 5.41 (d, J = 2.2 Hz, 1H),
1
3
4
.70 – 4.61 (m, 2H), 2.05 (s, 3H). C NMR (126 MHz, DMSO) δ 164.90, 163.37,
152.28, 140.30, 138.50, 136.78, 135.55, 133.18, 132.88, 131.91, 130.07, 128.38, 128.02,
1
1
4
23.83, 123.71, 122.51, 120.88, 115.92, 115.73, 110.28, 109.71, 104.88, 54.20, 38.91,
6.99. MS (ESI+) m/z: 568.1 (M+1). HPLC purity: 97%;
ꢀ(3ꢀ((2,6ꢀDimethylbenzyl)carbamoyl)ꢀ4ꢀfluorophenyl)ꢀNꢀ(1Hꢀindazolꢀ5ꢀyl)ꢀ6ꢀ
methylꢀ2ꢀoxoꢀ1,2,3,4ꢀtetrahydropyrimidineꢀ5 carboxamide (12m): Compound 12m
was prepared as described for 12a, replacing methylamine with 2,6ꢀ
1
dimethylbenzylamine. Yield: 82.9 mg (65%). H NMR (500 MHz, DMSOꢀd ) δ 12.93 (s,
6
1
H), 9.58 (s, 1H), 8.76 (d, J = 1.6 Hz, 1H), 8.41 (t, J = 4.6 Hz, 1H), 7.99 (d, J = 18.7 Hz,
2H), 7.51 (dd, J = 6.7, 2.3 Hz, 1H), 7.43 – 7.34 (m, 3H), 7.19 (dd, J = 9.9, 8.7 Hz, 1H),
.10 – 7.03 (m, 1H), 6.99 (d, J = 7.5 Hz, 2H), 5.43 (d, J = 2.4 Hz, 1H), 4.43 (t, J = 4.7
Hz, 2H), 2.32 (s, 6H), 2.07 (s, 3H). HPLC purity: 96%; MS (ESI+) m/z: 527.1 (M+1).
ꢀ(3ꢀ((2,6ꢀDimethoxybenzyl)carbamoyl)ꢀ4ꢀfluorophenyl)ꢀNꢀ(1Hꢀindazolꢀ5ꢀyl)ꢀ6ꢀ
methylꢀ2ꢀoxoꢀ1,2,3,4ꢀtetrahydropyrimidineꢀ5ꢀcarboxamide (12n): Compound 12n
7
4
29
ACS Paragon Plus Environment