Structural Basis for EGFR Mutant Inhibition by Trisubstituted Imidazole Inhibitors

Article abstract of DOI:10.1021/acs.jmedchem.0c00200

Acquired drug resistance in epidermal growth factor receptor (EGFR) mutant non-small-cell lung cancer is a persistent challenge in cancer therapy. Previous studies of trisubstituted imidazole inhibitors led to the serendipitous discovery of inhibitors that target the drug resistant EGFR(L858R/T790M/C797S) mutant with nanomolar potencies in a reversible binding mechanism. To dissect the molecular basis for their activity, we determined the binding modes of several trisubstituted imidazole inhibitors in complex with the EGFR kinase domain with X-ray crystallography. These structures reveal that the imidazole core acts as an H-bond acceptor for the catalytic lysine (K745) in the "αC-helix out" inactive state. Selective N-methylation of the H-bond accepting nitrogen ablates inhibitor potency, confirming the role of the K745 H-bond in potent, noncovalent inhibition of the C797S variant. Insights from these studies offer new strategies for developing next generation inhibitors targeting EGFR in non-small-cell lung cancer.

Full text of DOI:10.1021/acs.jmedchem.0c00200

Subscriber access provided by University of Victoria Libraries
Article
Structural Basis for EGFR Mutant Inhibition
by Trisubstituted Imidazole Inhibitors
David E Heppner, Marcel Günther, Florian Wittlinger, Stefan A. Laufer, and Michael J. Eck
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.0c00200 • Publication Date (Web): 03 Apr 2020
Downloaded from pubs.acs.org on April 5, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
Structural Basis for EGFR Mutant Inhibition by Trisubstituted
Imidazole Inhibitors
7
8
9
David E. Heppner,a,b Marcel Günther,c Florian Wittlinger,c Stefan A. Laufer,c* Michael J. Eck.a,b*
aDepartment of Cancer Biology, Dana-Farber Cancer Institute, Boston, MA 02215, USA.
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
bDepartment of Biological Chemistry and Molecular Pharmacology, Harvard Medical School,
Boston, MA 02115, USA.
cInstitute for Pharmaceutical Sciences, Eberhard Karls Universität Tübingen, Tübingen 72076,
Germany.
Keywords: Epidermal growth factor receptor, mutant targeting, imidazole, non-small cell lung
cancer.
Abstract
Acquired drug resistance in epidermal growth factor receptor (EGFR) mutant non-small
cell lung cancer is a persistent challenge in cancer therapy. Previous studies of trisubstituted
imidazole inhibitors lead to the serendipitous discovery of inhibitors that target the drug resistant
EGFR(L858R/T790M/C797S) mutant with nanomolar potencies in a reversible binding
mechanism. To dissect the molecular basis for their activity, we determined the binding modes of
several trisubstituted imidazole inhibitors in complex with the EGFR kinase domain with X-ray
crystallography. These structures reveal that the imidazole core acts as an H-bond acceptor for
the catalytic lysine (K745) in the “αC-helix out” inactive state. Selective N-methylation of the H-
bond accepting nitrogen ablates inhibitor potency, confirming the role of the K745 H-bond in
potent, non-covalent inhibition of C797S variant. Insights from these studies offer new strategies
for developing next generation inhibitors targeting EGFR in non-small cell lung cancer.
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 34
1
2
3
4
Introduction
5
6
7
8
Activating mutations within the kinase domain of the epidermal growth factor receptor
(EGFR), including L858R and in-frame exon-19 deletions (del), drive tumor carcinogenesis in
non-small cell lung cancer (NSCLC) and are prognostic indicators for the clinical effectiveness
for treatment with tyrosine kinase inhibitors (TKIs), such as gefitinib and erlotinib.1-2 Although
patients initially respond well to TKIs, drug resistance occurs due to the acquisition of the
T790M “gatekeeper” mutation rendering TKIs ineffective due to increased ATP affinity within
the kinase domain.3-4 Development of a novel irreversible binding anilinopyrimidine inhibitor
(WZ4002), containing a C797S-targeting acrylamide Michael acceptor, exhibited exquisite
potency against T790M mutant while minimally inhibited WT EGFR.5 This tool compound
provided proof of principle for mutant-selective irreversible inhibition of T790M-mutant EGFR,
and subsequently the third generation EGFR TKI osimertinib was approved for treatment of
NCSLC patients harboring T790M positive tumors6-7 and more recently as a front-line therapy in
untreated EGFR mutant NSCLC.8 However, in the context of T790M-positive NSCLC, patients
acquire resistance to osimertinib through several mechanisms, including around 30-40% of
patients acquiring an additional C797S mutation that renders third-generation TKIs incapable of
forming their potency-enabling covalent bond.9-10 Therefore, efforts to develop inhibitors
targeting both EGFR(T790M) and EGFR(C797S) mutations represent a critical unmet need in
the treatment of NSCLC.
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
2
ACS Paragon Plus Environment

Page 3 of 34
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 1. Trisubstituted imidazole inhibitor chemical structures.
To overcome resistance to third-generation inhibitors via the acquisition of the C797S
mutation, various drug discovery efforts have been dedicated to discovery of novel scaffolds that
selectively target this mutation.11 While screening for highly effective p38 MAP kinase
inhibitors, trisubstituted imidazole compounds were discovered to exhibit off-target inhibition of
EGFR.12 Medicinal chemistry optimization and molecular docking approaches produced a series
of reversible and irreversible trisubstituted imidazole-based compounds that effectively inhibited
mutant EGFR (Figure 1, 1-5).13 Surprisingly, these molecules exhibited low nanomolar potency
against EGFR(L858R/T790M/C797S) irrespective of whether they contained the covalent
warhead.14-15 Subsequently, several examples of reversible ATP-competitive inhibitors targeting
EGFR(del/T790/C797S) and/or EGFR(L858R/T790M/C797S) variants have been described,
including
Brigatinib,16
2-aryl-4-aminoquinazolines,17
pyrimidopyrimidinones,18-19
pyrrolopyrimidines,20 9-heterocyclyl-purines,21 and rationally designed macrocycles based on
aminobenzimidazoles.22 Despite these successes in developing trisubstituted imidazole
EGFR(T790M/C797S) targeting reversible inhibitors, the structural basis for their potency for
this drug resistant oncogenic EGFR variant remains unclear.
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 34
1
2
3
4
5
6
7
8
9
In this study, we structurally characterize the binding modes of reversible and irreversible
binding trisubstituted imidazole EGFR inhibitors with X-ray crystallography to understand their
targeting of EGFR drug resistant mutant C797S. Crystal structures of inhibitors bound to both
inactive and active EGFR kinases reveal that the imidazole moiety functions as a H-bond
acceptor for the -ammonium group of the catalytic lysine (K745) in the “αC-helix out” inactive
state and blocking this interaction via the synthesis of selective imidazole N-methylation
abolishes potency for EGFR(L858R/T790M/C797S). This H-bonding interaction with the EGFR
kinase, which is not possible with WZ4002 or osimertinib but is observed in examples of other
C797S inhibitors, is key for enabling strong, reversible binding rendering the trisubstituted
imidazole molecules as effective 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
Results
We determined crystal structures of reversible (1-2) and irreversible binding (3-5)
imidazole-based inhibitors (Figure 1) soaked into crystals of EGFR(T790M/V948R) (Figure 2,
S1-2, and Table S1).13-15 The inclusion of the V948R variant enables crystallization of EGFR in
the inactive state by precluding the formation of the activating asymmetric dimer. The structural
hallmark of the inactive EGFR kinase is the outward arrangement of the αC-helix (“αC-helix
out”), which is rotated inward due in the active “αC-helix in” asymmetric kinase dimer.23
Expectedly, reversible imidazole inhibitors (1 and 2) bind to the ATP pocket anchored by H-
bonds to kinase hinge via the pyrrolopyridine (Figure 2A-B). C797 at the periphery of the ATP
binding site is distant from the inhibitor (~ 5.0 Å) indicating that the binding modes of 1 and 2 is
likely agnostic with respect to the C797S mutation. The imidazole group extends away from the
hinge toward K745, the catalytic lysine, and the fluorophenyl group extends into the
hydrophobic pocket positioned between the T790M gatekeeper mutation and K745. The
4
ACS Paragon Plus Environment

Page 5 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
imidazole N3 forms an H-bond with K745, which is also H-bonded to D855 of the DFG motif.
The flexible functional groups extending from the imidazole C2 adopt subtly different
conformations in the four copies of the EGFR kinase domain in the asymmetric unit of the
crystal structures of 1 and 2. As representative examples, the propanol group of 1 engages in H-
bonding with D855 and N842, while the propylether moiety of 2 is incapable of such interactions
and forms an intramolecular H-bond to the imidazole N1. Generally, this experimentally
determined binding mode of the inhibitor is in good agreement with the previously reported
poses generated with in silico docking approaches.13-15 Additionally, in both cases electron
density difference maps indicate the presence of an ion bound underneath the EGFR kinase P-
loop directly interacting with K745 in all cases and R841 in the case of 2 (Figure 2B, Figure S2-
3). While we have not experimentally determined the identity of this ion, we have modeled this
density as a chloride ion since it is closely associated with positive side chains and backbone
amides and a water molecule was not sufficiently electron-dense to model it. We also note the
high concentration of chloride present in the purification and crystallization buffer.
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
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 34
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 2. Structures and binding modes of 1 (A, PDB 6V5N), 2 (B, PDB 6V5P), 3 (C, PDB
6V6O), 4 (D, PDB 6V6K), and 5 (E, PDB 6V66) in complex with EGFR(T790M/V948R). The
V948R mutation enables the kinase domain to crystallize in the inactive state.
6
ACS Paragon Plus Environment

Page 7 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
We also determined structures of inactive “αC-helix out” EGFR(T790M/V948R) crystals
soaked with irreversible compounds 3-5, which contain a 2-aminopyridine phenylacrylamide
Michael acceptor to target C797S (Figure 2C-E). Indeed, crystal structure electron density maps
indicate that 3-5 form their expected covalent bonds with C797 (Figure S2). The
phenylacrylamide group is found to engage in intramolecular H-bonding with the imidazole N1
as predicted by molecular modeling.13 The formation of the covalent bond to C797 does not
significantly alter the inhibitor binding mode in comparison to the reversibly binding compounds
(1 and 2), indicating that both covalent and non-covalent imidazole inhibitors bind EGFR
involving conserved H-bonds to K745.
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
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 34
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. Structures and binding modes of 3 (A, PDB 6VH4), 4 (B, PDB 6VHN), and 5 (C, PDB
6VHN) bound to WT EGFR kinase domain in the active state.
8
ACS Paragon Plus Environment

Page 9 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Active “αC-helix in” kinase structures were determined though compound soaking with
crystals of WT EGFR kinase domain. We were successful in obtaining crystal structures of the
irreversible inhibitors (3-5), but failed to obtain structures in complex with reversibly binding
compounds (1 and 2). As expected, these imidazole-based inhibitors bind to the kinase hinge in
essentially the same manner observed in the inactive-state complexes (Figure 3, Figure S4, and
Table S2). Due to the inward rotation of that αC-helix (Figure 3, red) in the kinase active state,
the E762 carboxylate side chain is now positioned toward K745, establishing a salt bridge
competing the K745 ammonium group away from the imidazole. The crystal structures of 3 and
5 show a covalent bond formed between the inhibitor and C797, while this is not the case for 4
(Figure 3). The absence of a covalent bond with 4 is unclear, but may be due to differences in
compound reactivity and potency (Tables 1 and S4). A structural overlay of the “αC-helix out”
inactive (green) and “αC-helix in” active (blue) kinase crystal structures in complex with 3
illustrates how K745 can toggle between functions as an H-bond donor to the imidazole and
E762 carboxylate as dependent on the orientation of the kinase αC-helix (Figure 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
Figure 4. Structural overlay of 3 crystalized in complex with EGFR(T790M/V948R) (green,
PDB 6V6O) in the inactive state and WT EGFR (blue, PDB 6VH4) in the active state.
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 34
1
2
3
4
5
6
7
F
F
F
F
a
N
N
N
b
c
O
O
O
N
H
9
CH3
H
OCH3
OCH3
MOM
10
8
8
9
F
F
F
F
N
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
N
N
d
e
N
N
f
N
N
g
N
H
OCH3
(Bu)3Sn
OCH3
MOM
I
N
N
OCH3
MOM
OCH3
MOM
HN
11
N
12
MOM
2
13
Scheme 1. Reagents and conditions: (a) SeO2, 1,4-dioxane, H2O, 60°C to reflux (73%); (b) 4-
methoxybutanal, NH4OAc, MeOH, rt (16%); (c) NaH, MOM-Cl, THF, 0°C to rt (77%); (d) N-
iodosuccinimide, MeOH, 0°C (94%); (e) iPrMgCl*LiCl, Sn(Bu)3Cl, THF (quant.), (f) 4-chloro-
1-(methoxymethyl)-2-phenyl-1H-pyrrolo[2,3-b]pyridine, Pd(OAc)2, XPhos, 1,4-dioxane, reflux
(52%); (g) HCl(aq), MeOH, reflux (58%).
To assess the functional relevance of the imidazole-K745 H-bond for inhibitor potency, we
prepared both regioisomeric N-methylimidazole derivatives based on compound 4. To this end,
we have synthesized compounds 6 and 7 according to our previously reported procedures.13 For
the N3-regioselective introduction of the methyl group, an N1-SEM protected precursor was
methylated with methyl trifluoromethanesulfonate according to Scheme 2. Subsequent acidic
deprotection of the intermediate imidazolium trifluoromethanesulfonate yielded the desired
methylated precursor in one-pot. The identity of the desired N-methyl regioisomer was
determined by X-ray crystallography (Table S3 and Figure S4). Subsequent Suzuki cross-
coupling reaction, followed by a deprotection step and a Buchwald-Hartwig aryl amination
yielded the final compound analogous to our previously described route.13
10
ACS Paragon Plus Environment

Page 11 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
F
F
F
F
F
CH3
c
d
b
N
O
N
N
a
N
S
S
S
S
CH3
CH3
N
CH3
CH3
NH
N
N
Br
Br
Br
SEM
SEM
26
23
24
25
F
F
F
CH3
CH3
CH3
N
N
N
S
S
S
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
CH3
O
f
N
e
CH3
g
CH3
N
N
N
H
N
N
N
CH3
HN
NH2
HN
O
28
27
O
7
F
F
N
N
g
S
S
CH3
N
N
CH3
CH3
CH3
N
N
H
N
NH2
HN
O
O
29
6
Scheme 2. Reagents and conditions: (a) S-methylisothiourea, NaHCO3, THF/water, reflux
(50%); (b) NaH, SEM-Cl, THF, 0°C to rt (91%); (c) NBS, THF, -30°C (85%); (d) MeOtf, DCM,
TFA, rt (70%); (e) N-(4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)pyridin-2-yl)acetamide,
P(t-bu)3 Pd G3, K3PO4, 1,4-dioxane/water, 50°C (39%); (f) NaOH, MeOH/water, 50°C (88%);
(g) N-(3-bromo-4-methoxyphenyl)acrylamide, Brettphos Pd G3, Cs2CO3, t-BuOH/1,4-dioxane,
rf (67%).
F
F
F
F
F
H
N
d
O
O
b
N
c
N
a
S
S
S
N
OH
NH
15
NH
16
Br
NH3+Cl-
SEM
17
OH
14
F
F
F
F
N
N
N
N
f
g
e
h
N
N
S
S
S
S
N
N
SEM
19
Br
Br
OH
OCH3
OCH3
SEM
18
SEM
21
SEM
OCH3
N
N
HN
NH2
CH3
F
F
20
O
N
N
N
S
i
j
S
N
H
OCH3
SEM
N
OCH3
H
N
N
H
N
HN
HN
O
O
O
O
22
5
Scheme 3. Reagents and conditions: (a) urotropine, chloroform, 50°C then EtOH, HCl(aq), rf
(97%); (b) KSCN, HOAc, rf (71%); (c) 2-bromoethan-1-ol, K2CO3, MeOH, rt (quant.); (d) NaH,
SEM-Cl, THF, 0°C (67%); (e) NBS, MeCN, -30°C (quant.);(f) NaH, MeI,THF, -15°C (75%); (g)
N-(4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)pyridin-2-yl)acetamide, P(t-bu)3 Pd G3,
K3PO4, 1,4-dioxane/water, 50°C (56%%); (h) NaOH, THF/water, 50°C (59%); (i) N-(3-bromo-
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 34
1
2
3
4
5
4-methoxyphenyl)acrylamide, Brettphos Pd G3, Cs2CO3, t-BuOH/1,4-dioxane, rf (44%); (j) HCl
in EtOH 2.5 M, rt (70%).
6
7
Table 1. Impact of N-methylation of 4 on biochemical IC50.
8
9
R1
R2
R1
R2
CH3
N
R1
R2
N
N
N
N
7
S
S
S
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
N
H
CH3
CH3
CH3
CH3
6
4
IC50 (nM)
L858R/T790M
< 1
90
111
> 5000
1
L858R/T790M/C797S
> 5000
Biochemical EGFR inhibitory capacities at 50% activity (IC50) measurements were
performed with 4 and selectively N-methylated derivatives 6 and 7 (Table 1 and S4).
Methylation of N1 is found to significantly impair the potency of the inhibitor against
EGFR(L858R/T790M) indicating that the intramolecular H-bond formed between the imidazole
and the acrylamide is structurally required for inhibitor binding. Correspondingly, methylation of
N3, the nitrogen involved in H-bonding to K745 in the inactive “αC-helix out” kinase, does not
significantly alter inhibition of EGFR(L858R/T790M), showing that the potency of 4 does not
depend on the K745 H-bond and more likely driven by the formation of the C797-mediated
covalent bond. It should be noted that methylation at N3 produces an imidazole derivative that
would not form an identical intramolecular H-bond with the acrylamide as seen in the crystal
structures of 4. The impact of this alternative isomer is likely a minor effect, which is likely since
7 exhibits low nanomolar potency against EGFR(L858R/T790M) (Table 1). In the case of
EGFR(L858R/T790M/C797S), where 4 inhibits entirely through reversible binding, it is found
that N-methylation at either N1 or N3 completely abrogates compound potency demonstrating
the critical role for both imidazole nitrogens. Our inactive “αC-helix out” crystal structures
(Figure 2) demonstrate that the N3 imidazole is engaged in H-bonding with K745 whereas this
interaction is not present in the active kinase state (Figure 3-4). Additionally, there is a
12
ACS Paragon Plus Environment

Page 13 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
possibility for N-methylation of these inhibitors to have diminished binding affinity due to steric
hinderance with the EGFR protein. The binding mode of 4 (Figures 1D and 2B) indicates that N-
methylation at N1 will not interfere with EGFR, as this nitrogen is pointed away from EGFR and
out into solvent, while N-methylation at N3 could sterically clash with K745 in the inactive “αC-
helix out” state. Since the H-bond to K745 is a conserved feature in all of our inactive “αC-helix
out” structures, we expect any resulting steric hinderance from the N3 methylation to be minimal
and the domain component of inhibitor reversible binding is the H-bond. Therefore, these
collective findings demonstrate that reversible binding trisubstituted imidazole inhibitors acquire
stronger affinity to the EGFR kinase through an H-bond to K745 of the inactive kinase state,
which enables inhibitor potency against EGFR(L858R/T790M/C797S).
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
Discussion.
While osimertinib is a highly effective therapeutic for NSCLC driven by sensitive EGFR
mutants, including the treatment-acquired T790M resistance mutation, the C797S mutation
abolishes its efficacy as well as that of other third-generation EGFR TKIs. Inhibitors based on a
trisubstituted imidazole scaffold, initially optimized to target EGFR(L858R/T790M), were found
to unexpectedly exhibit low nanomolar activity against EGFR(L858R/T790M/C797S).13-15 To
determine the structural basis for their unique ability to target this drug resistant EGFR mutation
through a non-covalent mechanism, we determined the binding modes of several trisubstituted
imidazole inhibitors in complex with the EGFR kinase domain using X-ray crystallography.
While all inhibitors bind to the EGFR kinase through conserved H-bonding to the kinase hinge,
as expected, marked differences with respect to interactions with the imidazole are observed
depending on the state of the kinase. In the inactive “αC-helix out” state, the imidazole inhibitors
all are directly involved in H-bonding to the catalytic lysine (K745) through the imidazole N3
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 34
1
2
3
4
5
6
7
8
9
nitrogen whereas the inward rotation of the αC-helix in the active state results in a H-bond with
E862, which apparently outcompetes K745 from the inhibitor. Overall, these structures are
consistent with previous crystal structures of related pyridinylimidazole inhibitors in complex
with p3824-25 and JNK3.26 To assess the dependence of inhibitor potency on the ability of K745
to form an H-bond with the imidazole, we synthesized a variant of 4 with selective N-
methylation at either of the two imidazole nitrogens. Biochemical potencies with selective N-
methylation at the imidazole N3 nitrogen of 4 was found to minimally impact inhibitor binding
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
to
EGFR(L858R/T790M),
while
entirely
ablating
inhibitor
efficacy
for
EGFR(L858R/T790M/C797S). Collectively, our findings suggest that the imidazole N3-K745
H-bond, present only in the inactive “αC-helix out” kinase, is responsible for driving reversible
binding and inhibition of EGFR(L858R/T790M/C797S).
Our structures demonstrate the key molecular features of the imidazole compounds in
directing reversible inhibition of the EGFR(L858R/T790M/C797S), which resists inhibition by
third-generation EGFR TKIs. Third-generation EGFR TKIs such as WZ4002 and osimertinib
lack the ability to make this H-bond and reversibly bind EGFR more weakly compared to
imidazole-based inhibitors rendering them ineffective against the C797S mutation (Figure 5A).
This is consistent as the imidazole inhibitors are found to inhibit EGFR with enhanced Ki ~ 15-
fold compared to osimertinib.14 While reversible inhibitors that bind EGFR more tightly may be
able to overcome the drug-resistant C797S mutant, it is more so a liability with respect to
developing a targeted drug since strong interactions with conserved amino acids will diminish
the ability for the inhibitor to preferentially bind to the mutant compared to the wild-type kinase.
Since this is the case for the trisubstitued imidazole inhibitors in this study, further optimization
is required to achieve mutant selectivity.
14
ACS Paragon Plus Environment

Page 15 of 34
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 5. A) Structural overlay of EGFR crystals structures in complex with 1 (green, 6V5N)
and WZ4002 (magenta, 3IKA). B) EGFR crystal structures of JN3229 (5ZT0) and C)
aminobenzimidazole compound 1 (6S9B) from ref 22.
In addition to the trisubstituted imidazoles, several small molecules have been
subsequently reported to inhibit C797S-containing EGFR mutants operating through non-
covalent mechanisms.11 Analysis of all structurally-characterized inhibitor binding modes reveals
that many of these compounds function as a H-bond acceptor for K745. The inhibitor Brigatinib,
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 34
1
2
3
4
5
6
7
8
9
originally developed as an ALK inhibitor, was shown to exhibit off-target inhibition of other
kinases, including EGFR(del/T790M/C797S), which is maximally effective if dosed in
combination with an anti-EGFR antibody cetuximab.16 While no structure of Brigatinib bound to
EGFR is reported to our knowledge, the crystal structure of Brigatinib in complex with ALK in
the active state shows that the inhibitor phosphine oxide moiety is pointing toward the catalytic
lysine (K1150), potentially confirming enhanced potency in a similar mechanism as the
trisubstituted imidazoles.27 Related compounds containing a sulfonamide in the position of the
phosphine oxide effectively inhibit ALK and EGFR and, based on docking models, are expected
to form H-bonds with the respective EGFR/ALK catalytic lysine in the active state.28
Additionally, pyrimidopyrimidinone inhibitors were developed to target EGFR(C797S) and co-
crystal structures indicate that the inhibitors ketone is positioned as a H-bond acceptor with K745
through a bridging water molecule in the active kinase (Figure 5B).18-19 It is likely that in the
inactive “αC-helix out” kinase state such molecules form a stronger, more direct H-bond with
K745 operating through a similar mechanism to the trisubstituted imidazole inhibitors (1-2). A
more distinct example is the recent report of a macrocyclic aminobenzimidazole inhibitor (BI-
4020) that selectively inhibits T790M/C797S-containing EGFR due to the compounds
macrocyclic structure with marked efficacy for regressing EGFR(del/T790M/C797S) tumor
xenografts in vivo.22 While the best performing macrocycle inhibitor has not been structurally
characterized, crystal structures of several precursor molecules indicate that the mutant-selective
aminobenzimidazole obtains their strong potency through direct H-bonds with K745 and T854 in
the active kinase state and is potentially made mutant selective due to limited steric interference
with T790M (Figure 5C).22 Compared to our structures here, the lack of mutant-selectivity in the
imidazole inhibitors maybe due to steric clashing between T790M and the fluorophenyl group on
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
16
ACS Paragon Plus Environment

Page 17 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
C4 of the imidazole inhibitor requiring the methionine to adopt a forced conformation (Figure
5A and C).22 In addition, a distinct strategy has also emerged, which involves designing EGFR
drug molecules that purposefully form strong H-bonds with the mutant C797S serine alcohol.
Inhibitors consisting of 2-aryl-4-aminoquinazoline17 and 9-heterocyclyl substituted 9H-purine21
scaffolds have been reported to exhibit selectivity for EGFR(T790M/C797S) mutations. It is
clear that an EGFR inhibitor operating through a non-covalent mechanism can overcome the
C797S mutation through engineering of additional intermolecular interactions with the kinase,
although with the caveat that targeting conserved residues, such as K745, will negatively impact
the inhibitor mutant selectivity as well as overall kinase selectivity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As an alternative to ATP-competitive TKIs, a new class of highly mutant-selective EGFR
allosteric inhibitors (EAIs) has been developed to bind to an allosteric pocket of the inactive
EGFR kinase made accessible by the oncogenic EGFR(L858R/T790M) mutations.29-31
Importantly, the most effective single agent of the EAIs (JBJ-04-125-02) exhibits synergy in vivo
with the third-generation TKI osimertinib, potentially in a co-operative mechanism involving
simultaneous inhibitor binding the EGFR kinase domain.30 Efforts to completely span these two
sites have resulted in a compound consisting of a chimeric fusion of an EAI scaffold (EAI045)
with
a
TKI (vandetinib) that exhibited low nanomolar potency against
EGFR(L858R/T790M/C797S), although at similar potencies for the WT kinase.32 Although this
is a single example, it is probable that compounds consisting of elements of effective TKIs and
mutant-selective allosteric inhibitors may represent a productive avenue to produce a C797S-
targeting reversible inhibitor with enhanced mutant selectivity.
Conclusions
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 34
1
2
3
4
5
6
7
8
9
We have employed a combination of structural and biochemical studies to define the
molecular basis for non-covalent inhibition of EGFR(L858R/T790M/C797S) by trisubstituted
imidazole inhibitors. The reversible binding imidazole inhibitors overcome the C797S mutation
through stabilization of the inactive kinase state anchored by an imidazole-dependent H-bond
with K745. Insights from this work demonstrate the diversity of avenues available for the
development of next generation therapies in EGFR mutant NSCLC. Future directions of these
inhibitors are dependent on further optimization for compounds with selectivity for the
oncogenic mutants and sparingly limited potency against the WT EGFR kinase.
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
General Information.
Reagents and solvents were purchased from Sigma Aldrich, Acros, CombiBlocks or Fluorochem.
Commercially available reagents and solvents were used as received without any further
purification or drying procedures unless otherwise noted. All NMR spectra were measured with a
Bruker Avance 200, with a Bruker Avance 400 or with a Bruker Avance 600 MHz (research
group Prof. Dr. Pichler, Dep. of Preclinical Imaging and Radio pharmacy, Tübingen). Solvents
for NMR are noted in the experimental procedures for each compound. Residual solvent peaks
were used to calibrate the chemical shifts. Chemical shifts (δ) are reported in parts per million
(ppm). LRMS were determined either by TLC-MS (Advion expressions) (ESI) or FAB-MS at
the Mass Spectrometry Department, Institute of Organic Chemistry, Eberhard Karls Universität
Tübingen. ESI-HRMS was obtained from the research group of Prof. Dr. Lämmerhofer
(Pharmaceutical (Bio)-Analyses), University of Tübingen. TLC analyses were performed on
fluorescent silica gel 60 F254 plates (Merck). Compound spots were identified either via UV
illumination at 254 nm and 366 nm or developed with a suitable stain. For Column
chromatography either Davisil LC60A 20-45 micron silica from Grace Davison or Geduran Si60
63-200 micron silica from Merck was used as stationary phase. VWR LaFlash automated flash
chromatography system or standard glass columns were used for flash chromatography. The
purity of tested compounds was determined via reverse phase high performance liquid
chromatography (HPLC) on Agilent 1100 LC equipped with a UV diode array detector (DAD,
detection at 230 nm and 254 nm) and in all cases was determined to be > 95%. The
chromatographic separation was performed on a Phenomenex Luna 5u C8 column (150 mm x
4.6 mm, 5 μm) at 35 °C oven temperature. The injection volume was 5 μL and the gradient of the
used method was (Flow: 1.5 mL/min): 0.01 M KH2PO4, pH 2.3 (Solvent A), methanol (Solvent
B): 40 % B to 85 % B in 8 min, 85 % B for 5 min, 85% to 40 % B in 1 min, 40 % B for 2 min,
stop time 16 min.
18
ACS Paragon Plus Environment

Page 19 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
Synthesis
7
8
9
2-(4-fluorophenyl)-2-oxoacetaldehyde (8):
Selenium dioxide (14.89 g, 134.2 mmol) in 200 ml 1,4-dioxane and 6 ml water was heated to
60°C until a clear solution formed. Then 4’-Fluoroacetophenone (16.85 g, 14.8ml, 122.0 mmol)
was added and the solution was refluxed for 16 hours. After cooling to room temperature, the
solid was filtered. The filtrate was evaporated in vacuum to yield a yellow oil that was distilled at
85°C at 1 mbar to give the pure product in 73% yield (13.52 g, 88.9 mmol). 1H-NMR (200 MHz,
CDCl3) δ 9.62 (s, 1H), 8.47 – 7.84 (m, 2H), 7.19 (t, J = 8.7 Hz, 2H). 13C-NMR (50 MHz,
CDCl3) δ 189.6, 186.0, 167.1 (d, J = 258.8 Hz), 133.6 (d, J = 9.8 Hz), 128.3 (d, J = 3.0 Hz),
116.5 (d, J = 22.0 Hz). FAB: Calculated: 152.03 for C13H10N2. Found: 153.0 [M+H]+.
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-(4-fluorophenyl)-2-(3-methoxypropyl)-1H-imidazole (9):
A solution of 8 (3.09 g, 20.3 mmol) in methanol was added dropwise over 30 minutes to a stirred
solution of 4-methoxybutanal (2.07 g, 20.3 mmol) and NH4OAc (7.51 g, 97.4 mmol) in
methanol. The mixture was stirred overnight. The reaction was quenched by the addition of a
saturated aqueous NaHCO3 solution and stirred for 30 minutes. Methanol was removed by rotary
evaporation and the residual aqueous phase extracted three times with ethyl acetate. The
combined organic layers were dried over Na2SO4, filtered and the solvents removed in vacuum.
The crude product was purified by flash chromatography (ethyl acetate + 2% NH3 in MeOH
(7M)) to give an orange gum in 16% yield. (761 mg, 3.25 mmol). 1H-NMR (200 MHz, CDCl3) δ
9.92 (s, 1H), 7.71-7.55 (m, 2H), 7.13 (s, 1H), 7.10-6.96 (m, 2H), 3.47 (t, J = 5.8 Hz, 2H), 3.37 (s,
13
3H), 2.88 (t, J = 6.9 Hz, 2H), 2.04-1.90 (m, 2H). C-NMR (50 MHz, CDCl3) δ 161.8 (d, J =
245.0 Hz), 149.2, 129.9 (d, J = 2.7 Hz), 126.3 (d, J = 7.9 Hz), 115.6 (d, J = 21.5 Hz), 72.6, 58.8,
28.0, 26.3. TLC-MS (ESI+) (m/z): Calculated: 234,12 for C13H15FN2O. Found: 234,9 [M+H]+.
4-(4-fluorophenyl)-1-(methoxymethyl)-2-(3-methoxypropyl)-1H-imidazole (10):
A cooled solution of 9 (719 mg, 3.06 mmol) in 20 ml dry DMF was treated portion wise with
sodium hydride (60%, 3.52 mmol) and stirred for 30 minutes at 0°C. MOM-Cl (2.5 M in toluene,
3.06 mmol, 1.22 ml) was added dropwise to the stirred solution. The mixture was stirred for 10
minutes at 0°C and for 30 minutes at room temperature. The reaction was quenched by the
addition of a saturated aqueous NH4Cl solution and stirred for 30 minutes. The aqueous mixture
was extracted three times with ethyl acetate. The combined organic layers were dried over
Na2SO4, filtered and the solvents removed in vacuum. The crude product was purified by
gradient flash chromatography with 100% DCM to DCM/MeOH 93:7 to give the pure product as
1
a yellow oil in 77% yield (657 mg, 2.36 mmol). H-NMR (200 MHz, CDCl3) δ 7.77-7.64 (m,
2H), 7.14 (s, 1H), 7.10-6.95 (m, 2H), 5.20 (s, 2H), 3.46 (t, J = 6.1 Hz, 2H), 3.34 (s, 3H), 3.31 (s,
13
3H), 2.91-2.79 (m, 2H), 2.17-1.99 (m, 2H). C-NMR (50 MHz, CDCl3) δ 162.0 (d, J = 244.9
Hz), 149.2, 139.5, 130.5 (d, J = 3.2 Hz), 126.5 (d, J = 7.9 Hz), 115.4 (d, J = 21.5 Hz), 114.9,
76.7, 71.7, 58.6, 56.1, 28.3, 23.5. TLC-MS (ESI+): calculated: 278,14 for C15H19FN2O2.
Found: 278,9 [M+H]+.
4-(4-fluorophenyl)-5-iodo-1-(methoxymethyl)-2-(3-methoxypropyl)-1H-imidazole (11):
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 34
1
2
3
4
5
6
7
8
9
To a cooled solution of 10 (284 mg, 1.09 mmol) in 10 ml MeCN was added dropwise at 0°C N-
Iodosuccinimide (281 mg, 1.25 mmol) dissolved in 5 ml MeCN. The solution was stirred for 10
minutes at 0°C, the ice bath was removed and stirring continued overnight at room temperature.
The solvent was removed in vacuum and the oily residue purified by gradient flash
chromatography (from 100 % DCM to DCM/MeOH 97:3) to give the pure product as an orange
1
oil in 94% yield (412mg, 1.02 mmol). H-NMR (200 MHz, CDCl3) δ 7.90-7.79 (m, 2H), 7.15-
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
7.02 (m, 2H), 5.30 (s, 2H), 3.47 (t, J = 6.0 Hz, 2H), 3.38 (s, 3H), 3.34 (s, 3H), 3.00-2.89 (m, 2H),
2.16-2.00 (m, 2H). 13C-NMR (50 MHz, CDCl3) δ 162.4 (d, J = 246.5 Hz), 152.6, 142.9, 130.1 (d,
J = 3.2 Hz), 129.6 (d, J = 8.1 Hz), 115.2 (d, J = 21.5 Hz), 77.4, 71.6, 67.7, 58.7, 56.2, 28.1, 24.5.
TLC-MS (ESI+) (m/z): Calculated: 404,04 for C15H18FIN2O2. Found: 404,7 [M+H]+.
4-(4-fluorophenyl)-1-(methoxymethyl)-2-(3-methoxypropyl)-5-(tributylstannyl)-1H-
imidazole (12):
A cooled solution of 11 (513mg, 1.27 mmol) in 15 ml dry THF was treated with a solution of i-
PrMgCl*LiCl (1.3M in THF, 1.65 mmol, 1.27 ml) dropwise at 0°C. The mixture was stirred for
30 minutes at 0°C. Next, tributyltin chloride (1.78mmol, 482µl) was added and the reaction
stirred for 10 minutes at 0°C. The ice bath was removed and the solution stirred at room
temperature overnight. The reaction was quenched by the addition of an aqueous NaHCO3
solution. The aqueous mixture was extracted three times with ethyl acetate. The combined
organic layers were dried over Na2SO4, filtered and the solvents removed in vacuum. The crude
product was used in the next step without further purification or characterization of the mixture.
4-(4-(4-fluorophenyl)-1-(methoxymethyl)-2-(3-methoxypropyl)-1H-imidazol-5-yl)-1-
(methoxymethyl)-2-phenyl-1H-pyrrolo[2,3-b]pyridine (13):
A mixture of Pd(OAc)2 (2 mol%) and XPhos (6 mol%) was dissolved in dry and degassed 1,4-
dioxane (5 ml) under an atmosphere of argon and stirred for 30 minutes at room temperature. To
this solution, crude 12 and 4-chloro-1-(methoxymethyl)-2-phenyl-1H-pyrrolo[2,3-b]pyridine
(690 mg, 2.54 mmol) dissolved in 10 ml 1,4-dioxane was added. The mixture was refluxed for
until complete consumption of the azaindol. After cooling to room temperature, a 40% aqueous
KF solution was added and the mixture stirred for 1 hour at room temperature. The aqueous
mixture was extracted three times with ethyl acetate. The combined organic layers were dried
over Na2SO4, filtered and the solvents removed in vacuum. The crude product was purified with
flash chromatography (DCM/MeOH 97:3 + 2 % AcOH) to give the pure product as a yellow
1
gum in 52% yield (170 mg, 0.33 mmol). H-NMR (200 MHz, CDCl3) δ 8.66 (s, 2H), 8.43 (d, J
= 5.0 Hz, 1H), 7.62-7.51 (m, 2H), 7.48-7.32 (m, 5H), 7.16 (d, J = 5.0 Hz, 1H), 6.92-6.78 (m,
2H), 6.18 (s, 1H), 5.63 (dd, J = 26.9, 10.3 Hz, 2H), 5.08 (dd, J = 37.3, 10.6 Hz, 2H), 3.58-3.46
13
(m, 5H), 3.37 (s, 3H), 3.10 (s, 3H), 3.00 (t, J = 7.5 Hz, 2H), 2.25-2.11 (m, 2H). C-NMR (50
MHz, CDCl3) δ 162.0 (d, J = 245.7 Hz), 150.6, 150.2, 143.3, 137.3, 131.4, 130.4, 130.0 (d, J =
2.8 Hz), 129.3, 129.1, 129.0, 128.9, 124.8, 120.7, 119.1, 115.2 (d, J = 21.4 Hz), 100.5, 74.4,
73.0, 71.9, 58.6, 56.7, 55.9, 28.3, 23.8. TLC-MS (ESI+) (m/z): Calculated: 514,24 for
C30H31FN4O3. Found: 514,9 [M+H]+.
4-(4-(4-fluorophenyl)-2-(3-methoxypropyl)-1H-imidazol-5-yl)-2-phenyl-1H-pyrrolo[2,3-
b]pyridine (2):
A solution of 13 (170 mg, 0,33 mmol) in 5ml methanol and 5 ml concentrated hydrochloric acid
was stirred under reflux for 24 hours. After complete conversion, the mixture was cooled to
20
ACS Paragon Plus Environment

Page 21 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
room temperature and neutralized with a saturated aqueous solution of K2CO3. The aqueous
mixture was extracted three times with ethyl acetate. The combined organic layers were dried
over Na2SO4, filtered and the solvents removed in vacuum. The crude product was purified with
gradient flash chromatography (from 100% DCM to DCM/MeOH 90:10) to give the pure
product as a yellow solid in 58% yield (82 mg, 0.19 mmol). H-NMR (400 MHz, DMSO-d6) δ
12.36 (s, 1H), 12.17 (s, 1H), 8.16 (s, 1H), 7.77 (d, J = 7.4 Hz, 2H), 7.49-7.37 (m, 4H), 7.35-7.28
(m, 1H), 7.22-7.08
1
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
(m, 2H), 7.04 (d, J = 4.7 Hz, 1H), 6.55 (s, 1H), 3.45 (t, J = 5.9 Hz, 2H), 3.28 (s, 3H), 2.79 (t, J =
7.4 Hz, 2H), 2.09-1.95 (m, 2H). C-NMR (100 MHz, DMSO-d6) δ 161.1 (d, J = 243.8 Hz),
150.5, 148.8, 142.7, 137.7, 131.5, 129.3, 129.3, 128.8, 127.9, 125.1, 118.3, 115.1 (d, J = 21.4
Hz), 114.9, 97.9, 71.3, 57.8, 27.9, 24.6.. TLC-MS (ESI−) (m/z): Calculated: 426,19 for
C26H23FN4O. Found: 425,0 [M−H]−.
13
2-(4-fluorophenyl)-2-oxoethan-1-aminium chloride (14):
2-Bromo-1-(4-fluorophenyl)ethan-1-one (7.9 g, 36.6 mmol) and Urotropine (5.6 g, 39.9 mmol)
were dissolved in 150 ml dry chloroform and stirred at 50°C for 2 hours. The precipitate was
removed by filtration and washed with chloroform and ethanol. The solid was dissolved in 150
ml abs. ethanol and 20 ml concentrated hydrochloric acid and refluxed for 2 hours. The mixture
was filtered and the filtrate was evaporated to dryness to give the pure product as a white solid in
1
97% yield (6.7 g, 35.2 mmol). H NMR (200 MHz, DMSO-d6) δ 8.60 (s, 3H), 8.22 – 8.05 (m,
2H), 7.47 – 7.38 (m, 2H), 4.58 (s, 2H). 13C NMR (50 MHz, DMSO-d6) δ 191.78, 165.97 (d, J =
253 Hz), 131.75 (d, J = 9.8 Hz), 130.82 (d, J = 3.6 Hz), 116.46 (d, J = 22.2 Hz), 45.08.
4-(4-fluorophenyl)-1,3-dihydro-2H-imidazole-2-thione (15):
A suspension of 14 (2.0 g, 10.7 mmol) and potassium thiocyanate (2.61 g, 26.8 mmol) in 15 ml
glacial acid was refluxed for 1 hour. After cooling to 0°C and the addition of water, the
precipitated solid was collected by filtration. The solid was washed with water and dried in
vacuum overnight to give the product as a pale yellow solid in 71% yield (1.5 g, 7.6 mmol). The
1
obtained product was used without further purification. LRMS (ESI-MS): 193.1 [M-H]-. H
NMR (200 MHz, DMSO-d6) δ 12.55 (s, 1H), 12.16 (s, 1H), 7.79 – 7.65 (m, 2H), 7.37 (s, 1H),
7.23 (t, J = 8.7 Hz, 2H). 13C NMR (50 MHz, DMSO-d6) δ 162.14, 161.63 (d, J = 245 Hz) 128.52,
126.34 (d, J = 8.3 Hz), 125.34 (d, J = 3.0 Hz), 116.12 (d, J = 21.9 Hz), 112.44.
2-((4-(4-fluorophenyl)-1H-imidazol-2-yl)thio)ethan-1-ol (16):
Intermediate 15 (1.0 g, 5.15 mmol) was dissolved in 20 ml dry methanol. Potassium carbonate
(890 mg, 6.44 mmol) was added to the solution at room temperature under vigorous stirring. The
resulting suspension was stirred for 30 minutes at room temperature. 2-Bromoethanol (402 µl,
5.67 mmol) was added and the suspension was stirred at room temperature overnight. After
complete reaction, water and a saturated NH4Cl solution was added. The aqueous mixture was
extracted three times with ethyl acetate. The combined organic layers were dried over Na2SO4,
filtered and the solvents removed in vacuum. The obtained pure product was used without
further purification. After washing with n-pentane and drying overnight, the product was
1
obtained in quantitative yield. LRMS (ESI-MS): 237.0 [M-H]-. H NMR (200 MHz, DMSO-d6)
δ 12.39 (s, 1H), 7.86 – 7.70 (m, 2H), 7.65 (s, 1H), 7.26 – 7.13 (m, 2H), 5.14 (s, 1H), 3.76 – 3.60
13
(m, 2H), 3.16 (t, J = 6.5 Hz, 2H). C NMR (50 MHz, DMSO-d6) δ 161.24 (d, J = 243 Hz),
140.67, 140.41, 131.24, 126.28 (d, J = 8.0 Hz), 115.61 (d, J = 21.1 Hz), 114.85, 61.06, 36.11.
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 34
1
2
3
4
5
6
2-((4-(4-fluorophenyl)-1-((2-(trimethylsilyl)ethoxy)methyl)-1H-imidazol-2-yl)thio)ethan-1-
ol (17):
To a solution of 16 (965 mg, 4.05 mmol) in dry THF (20 ml) was added sodium hydride (60%,
4.45 mmol) at -15°C under vigorous stirring. The cooling bath was removed and the mixture
warmed to room temperature over 30 minutes. The solution was again cooled to -15°C and
SEM-Cl (716 µl, 4.05 mmol) was added dropwise. After complete consumption of the starting
material, brine was added. The aqueous mixture was extracted three times with ethyl acetate. The
combined organic layers were dried over Na2SO4, filtered and the solvents removed in vacuum.
The crude mixture of regioisomers was dissolved in MeCN (2 ml) and treated with a catalytic
amount of SEM-Cl (5 mol%). The solution was stirred at 70°C overnight. After cooling to room
temperature, the solvent was removed in vacuum. The crude product was purified with gradient
flash chromatography (from 100% n-hexane to n-hexane/ethyl acetate 3:2) to give the pure
product as a pale yellow oil in 67% yield (1.03 g, 2.71 mmol). LRMS (ESI-MS): 369.1 [M+H]+.
1H NMR (200 MHz, Chloroform-d) δ 7.88 – 7.59 (m, 2H), 7.30 (s, 1H), 7.18 – 6.98 (m, 2H),
5.49 – 5.27 (m, 3H), 4.14 – 4.04 (m, 2H), 3.68 – 3.53 (m, 2H), 3.33 – 3.23 (m, 2H), 1.04 – 0.93
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
13
(m, 2H), 0.04 (s, 9H). C NMR (50 MHz, Chloroform-d) δ 162.10 (d, J = 246 Hz), 143.50,
140.74, 128.88 (d, J = 3.1 Hz), 126.31 (d, J = 8.3 Hz), 116.21, 115.51 (d, J = 21.5 Hz), 66.64,
63.72, 37.02, 17.64, -1.48.
2-((5-bromo-4-(4-fluorophenyl)-1-((2-(trimethylsilyl)ethoxy)methyl)-1H-imidazol-2-
yl)thio)ethan-1-ol (18):
To a cooled solution of 17 (1.0 g, 2.71 mmol) in 30 ml MeCN was added dropwise at -30°C N-
Bromosuccinimide (507 mg, 2.85 mmol) dissolved in 15 ml MeCN. The solution was stirred for
1 hour at -30°C, the ice bath was removed and stirring continued at room temperature until
complete consumption of the starting material. Saturated NaHCO3 solution and solid sodium
thiosulfate was added and the mixture stirred for 10 minutes. The aqueous mixture was extracted
three times with ethyl acetate. The combined organic layers were washed with saturated
NaHCO3 solution, dried over Na2SO4, filtered and the solvents removed in vacuum. The obtained
pure product was used without further purification. The product was isolated as a yellow oil in
64% yield (771 mg, 0.64 mmol). LRMS (ESI-MS): 447.1/449.2 [M+H]+.1H NMR (200 MHz,
DMSO-d6) δ 7.81 – 7.54 (m, 2H), 7.54 – 7.26 (m, 2H), 5.55 (s, 2H), 3.96 – 3.40 (m, 7H), 0.85 (t,
J = 6.1 Hz, 2H), 0.03 (s, 9H). 13C NMR (50 MHz, DMSO-d6) δ 163.95 (d, J = 252 Hz), 143.32,
138.48, 128.28 (d, J = 8 Hz), 127.88 (d, J = 3 Hz), 116.31 (d, J = 22 Hz), 98.09, 75.29, 66.42,
60.44, 35.33, 18.06, -1.43.
5-bromo-4-(4-fluorophenyl)-2-((2-methoxyethyl)thio)-1-((2-(trimethylsilyl)ethoxy)methyl)-
1H-imidazole (19):
Intermediate 18 (193 mg, 0.43 mmol) was dissolved in 10 ml dry THF and cooled to -15°C. To
the vigorously stirred solution was added sodium hydride (60%, 21 mg, 0.52 mmol) portion
wise. After complete addition the reaction was slowly warmed to room temperature and methyl
iodide (33 µl, 0.52 mmol) was added. The mixture was stirred for 1 hour at room temperature.
After complete reaction, saturated NH4Cl solution was added. The aqueous mixture was
extracted three times with DCM. The combined organic layers were dried over Na2SO4, filtered
and the solvents removed in vacuum. The crude product was purified with gradient flash
chromatography (from 100% n-hexane to n-hexane/ethyl acetate 3:1) to give the pure product as
22
ACS Paragon Plus Environment

Page 23 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
a colorless oil in 75% yield (148 mg, 0.32 mmol). H NMR (200 MHz, Chloroform-d) δ 8.14 –
7.87 (m, 2H), 7.23 – 7.01 (m, 2H), 5.41 (s, 2H), 3.85 – 3.55 (m, 4H), 3.46 – 3.36 (m, 5H), 1.14 –
0.86 (m, 2H), 0.04 (s, 9H). 13C NMR (50 MHz, Chloroform-d) δ 162.06 (d, J = 247 Hz), 143.67,
138.43, 128.91 (d, J = 3.3 Hz), 128.35 (d, J = 8.2 Hz), 115.06 (d, J = 21.5 Hz), 100.06, 73.80,
71.00, 66.53, 58.55, 33.30, 17.76, -1.46.
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
N-(4-(4-(4-fluorophenyl)-2-((2-methoxyethyl)thio)-1-((2-(trimethylsilyl)ethoxy)methyl)-1H-
imidazol-5-yl)pyridin-2-yl)acetamide (20):
To a solution of 19 (148 mg, 0.32 mmol) and N-(4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-
yl)pyridin-2-yl)acetamide (126 mg, 0.48 mmol) in 4 ml degassed 1,4-dioxane was added a n
aqueous 1M K3PO4 solution (1 ml). The biphasic mixture was degassed 3 times, by evacuating
and backfilling with argon. Under argon, P(t-Bu)3Pd G3 (5 mol%, 9mg) was added, the reaction
flask was sealed and the suspension stirred for 5 hours at 50°C. After complete reaction, the
mixture was cooled to room temperature and brine was added. The aqueous mixture was
extracted three times with DCM. The combined organic layers were dried over Na2SO4, filtered
and the solvents removed in vacuum. The crude product was purified with gradient flash
chromatography (from 100% n-hexane to n-hexane/ethyl acetate 1:1) to give the pure product as
1
a colorless oil in 56% yield (92 mg, 0.17 mmol). LRMS (ESI-MS): 539.6 [M+Na]+. H NMR
(200 MHz, Chloroform-d) δ 9.25 (s, 1H), 8.42 – 8.09 (m, 2H), 7.55 – 7.37 (m, 2H), 7.13 – 6.85
(m, 3H), 5.25 (s, 2H), 3.76 (t, J = 6.2 Hz, 2H), 3.64 – 3.36 (m, 7H), 2.22 (s, 3H), 0.94 (t, J = 8.3
Hz, 2H), -0.00 (s, 9H). C NMR (50 MHz, Chloroform-d) δ 168.76, 162.02 (d, J = 246 Hz),
13
152.25, 147.90, 144.49, 140.87, 139.49, 129.61 (d, J = 3.0 Hz), 129.12 (d, J = 7.9 Hz), 127.42,
121.19, 115.30, 115.05 (d, J = 21.4 Hz), 71.12, 66.32, 58.59, 33.35, 24.77, 24.49, 17.70, -1.53.
4-(4-(4-fluorophenyl)-2-((2-methoxyethyl)thio)-1-((2-
(trimethylsilyl)ethoxy)methyl)-1H-imidazol-5-yl)pyridin-2-amine (21):
The intermediate 20 (92 mg, 0.17 mmol) was dissolved in 5 ml methanol and 1 ml of an aqueous
solution of NaOH (5M) was added under stirring. The suspension was stirred for 3 hours at
50°C. After complete reaction, the mixture was cooled to room temperature and a saturated
aqueous NH4Cl solution was added. The aqueous mixture was extracted three times with ethyl
acetate. The combined organic layers were dried over Na2SO4, filtered and the solvents removed
in vacuum. The crude product was purified with gradient flash chromatography (from 100% n-
hexane to n-hexane/ethyl acetate 1:1) to give the pure product as a white solid in 59% yield (50
mg, 0.11 mmol). LRMS (ESI-MS): 475.5 [M+H]+. 1H NMR (200 MHz, Chloroform-d) δ 8.11 (d,
J = 5.4 Hz, 1H), 7.60 – 7.39 (m, 2H), 6.97 (t, J = 8.6 Hz, 2H), 6.69 (d, J = 5.2 Hz, 1H), 6.59 (s,
1H), 5.19 (s, 2H), 4.71 (s, 2H), 3.77 (t, J = 6.2 Hz, 2H), 3.60 (t, J = 8.2 Hz, 2H), 3.52 – 3.38 (m,
13
5H), 0.95 (t, J = 8.3 Hz, 2H), 0.03 (s, 9H). C NMR (50 MHz, Chloroform-d) δ 161.93 (d, J =
246 Hz), 159.48, 148.32, 144.04, 140.06, 138.76, 129.75, (d, J = 3.1 Hz), 128.86 (d, J = 8.0 Hz),
127.95, 115.35, 115.07 (d, J = 21.5 Hz), 109.65, 72.87, 71.11, 66.46, 58.59, 33.30, 17.89, -1.48.
N-(3-((4-(4-(4-fluorophenyl)-2-((2-methoxyethyl)thio)-1-((2-(trimethylsilyl)ethoxy)methyl)-
1H-imidazol-5-yl)pyridin-2-yl)amino)-4-methoxyphenyl)acrylamide (22):
A solution of 21 (50 mg, 0,11 mmol) and N-(3-bromo-4-methoxyphenyl)acrylamide (32 mg,
0.126 mmol) and cesium carbonate (171 mg, 0.525 mmol) in a mixture of 1,4-dioxane/t-BuOH
(4ml + 1ml) was degassed three times by evacuating and backfilling with argon under stirring.
Under an atmosphere of argon, Brettphos Pd G3 (5 mol%, 5 mg, 0.005 mmol) was added. The
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 34
1
2
3
4
5
6
7
8
9
solution was stirred under reflux for 3 hours. The mixture was cooled to room temperature and a
saturated aqueous NH4Cl solution was added. The aqueous mixture was extracted three times
with ethyl acetate. The combined organic layers were dried over Na2SO4, filtered and the
solvents removed in vacuum. The crude product was purified with gradient flash
chromatography (from 100% n-hexane to 100 % ethyl acetate) to give the pure product as a
white solid in 44% yield (30 mg, 0.046 mmol). LRMS (ESI-MS): 651.0 [M+H]+ . 1H NMR (200
MHz, Chloroform-d) δ 8.28 (d, J = 5.3 Hz, 1H), 8.11 (s, 1H), 7.60 – 7.45 (m, 4H), 7.23 (s, 1H),
7.05 – 6.81 (m, 5H), 6.53 – 6.20 (m, 2H), 5.80 – 5.70 (m, 1H), 5.23 (s, 2H), 3.88 (s, 3H), 3.78 (t,
J = 6.1 Hz, 3H), 3.64 – 3.44 (m, 6H), 0.98 – 0.89 (m, 2H), 0.02 (s, 9H).
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
N-(3-((4-(4-(4-fluorophenyl)-2-((2-methoxyethyl)thio)-1H-imidazol-5-yl)pyridin-2-
yl)amino)-4-methoxyphenyl)acrylamide (23):
The intermediate 22 (30 mg, 0.046 mmol) was dissolved in an ethanolic HCl solution (1.25 M)
and stirred overnight at room temperature. After complete consumption of the starting material,
the solution was evaporated to dryness and the oily residue taken up in ethyl acetate. The organic
solution was washed with a saturated aqueous NaHCO3 solution, dried over Na2SO4, filtered and
the solvents removed in vacuum. The crude product was dissolved in a minimum amount of
ethyl acetate and triturated with n-pentane to give the pure product as a pale yellow solid in 70%
yield (16 mg, 0.031 mmol). 1H NMR (200 MHz, DMSO-d6) δ 12.73 (s, 1H), 10.00 (s, 1H), 8.31
(d, J = 9.7 Hz, 1H), 8.19 – 7.95 (m, 2H), 7.62 – 7.40 (m, 3H), 7.37 – 7.12 (m, 3H), 6.95 (dd, J =
8.9, 3.5 Hz, 1H), 6.73 – 6.18 (m, 3H), 5.80 – 5.61 (m, 1H), 3.81 (s, 3H), 3.64 (t, J = 6.3 Hz, 2H),
13
3.32 – 3.24 (m, 5H). C NMR (50 MHz, DMSO-d6) δ 163.07, 162,20 (d, J = 249 Hz), 156.74,
147.32, 146.03, 143.33, 141.78, 140.92, 135.31, 132.54, 132.27, 130.95 (d, J = 8.3 Hz), 130.51,
129.97, 129.76, 126.45, 116.11 (d, J = 21.6 Hz), 113.22, 112.41, 111.22, 108.32, 71.15, 58.26,
56.23, 32.35. LRMS(ESI-MS): 517.9 [M-H]-.
Synthesis of 4-(4-fluorophenyl)-2-(methylthio)-1H-imidazole (23):
3.00 g (13.82 mmol) 2-Bromo-1-(4- fluorophenyl)ethan-1-one in 15 ml THF was added
dropwise to a refluxing mixture of 1.92 g (13.82 mmol) 2-methyl-2-thiopseudourea hemi sulfate
salt and 3.82 g (27.64 mmol) K2CO3 in THF/H2O (8 ml/2 ml). The mixture was refluxed for an
additional h until TLC indicated complete consumption of the starting material. The mixture was
evaporated to dryness and the residue dissolved in EtOAc and water. The layers were separated.
The organic layer was washed twice with brine, dried over Na2SO4, filtered and evaporated to
dryness. The crude mixture was purified via flash chromatography (n-hexane/EtOAc 1:1) to
obtain the title product as a beige solid in 50 % yield (1.44 g, 6.91 mmol). 1H-NMR (200 MHz,
d6-DMSO) δ 12.61 – 12.10 17 (m, 1H), 7.83 - 7.66 (m, 2H), 7.62 (s, 1H), 7.28 – 7.05 (m, 2H),
2.55 (s, 3H). 13C-NMR (50 MHz, d6-DMSO) δ 161.2(d, J = 242 Hz) 141.5, 140.5, 131.3 (d, J = 3
Hz), 126.3, 126.2 (d, J = 8 Hz), 115.5 (d, J = 21 Hz), 114.6, 15.8. ESI-MS: 209.1 [M+H]+.
4-(4-fluorophenyl)-2-(methylthio)-1-((2-(trimethylsilyl)ethoxy)methyl)-1H-imidazole (24):
710 mg (3.41 mmol) 23 was dissolved in 10 ml dry THF and cooled to −15°C with an
ice/acetone mixture under an argon atmosphere. 150 mg NaH (3.75 mmol, 60 % dispersion in
mineral oil) was added in small portions. The resulting suspension was stirred for 20 min at
−15°C after which, 663 µl (3.75 mmol) SEM-Cl was added dropwise to the mixture. The mixture
was allowed to warm to room temperature slowly. The reaction was quenched by the addition of
saturated aqueous NH4Cl solution and the crude product was extracted three times with DCM.
24
ACS Paragon Plus Environment

Page 25 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The organic layer was dried over Na2SO4, filtered and evaporated to dryness. The resulting
mixture of regioisomeres (60:40) was dissolved in 10 ml dry acetonitrile under an argon
atmosphere. 28 µl (0.171 mmol) SEM-Cl was added and the biphasic mixture was stirred for 4 h
at 80°C to perform the “SEM-switch”. The reaction was quenched by the addition of a few drops
of 10 % aqueous ammonia solution. The mixture was evaporated to dryness and the crude
product was purified by flash chromatography (from n-hexane 100 % to n-hexane/EtOAc 3:1) to
1
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
obtain the title compound as a colorless oil in 91% yield (1050 mg, 3.11 mmol). H-NMR (200
MHz, d6 -DMSO) δ 7.88 – 7.67 (m, 3H), 7.31 – 7.02 (m, 2H), 5.26 (s, 2H), 3.52 (t, J = 7.9 Hz,
2H), 2.58 (s, 3H), 0.81 – 0.90 (t, J = 7.9 Hz, 2H), 0.01 (s, 9H). 13C-NMR (50 MHz, d6 -DMSO) δ
170.6, 161.0 (d, J = 243 Hz), 143.5, 139.9, 130.4 (d, J = 3 Hz), 126.2 (d, J = 8 Hz), 118.5, 18
115.6 (d, J = 22 Hz), 74.9, 65.9, 17.5, 16.2, -1.0. ESI-MS: 339.1 [M+H]+.
5-Bromo-4-(4-fluorophenyl)-2-(methylthio)-1-((2-(trimethylsilyl)ethoxy)methyl)-1H-
imidazole (25):
895 mg (2.64 mmol) 24 was dissolved in 20 ml dry acetonitrile under argon atmosphere. The
solution was cooled to −30°C. 470 mg (2.64 mmol) N-Bromosuccinimide dissolved in 5 ml dry
acetonitrile was added dropwise under vigorous stirring maintaining −30°C. The reaction
mixture was stirred for 1h at −30°C and let allowed to slowly warm to room temperature. TLC
indicated complete consumption of the starting material. The reaction was quenched by the
addition of an aqueous sodium thiosulfate solution (10 %). The product was partitioned between
water and DCM. The organic layer was washed with brine, dried over Na2SO4, filtered and
evaporated to dryness. The crude product was purified via flash chromatography (hexane/EtOAc
5:1) to yield the title product as a colorless oil in 85 % yield (940 mg, 2.25 mmol). 1H-NMR (200
MHz, d6-DMSO) δ 7.97 – 7.85 (m, 2H), 7.34 – 7.19 (m, 2H), 5.30 (s, 2H), 3.58 (t, J = 7.9 Hz,
2H), 2.60 (s, 3H), 0.95 – 0.77 (t, J = 7.9 Hz, 2H), - 0.05 (s, 9H). 13C-NMR (50 MHz, d6 -DMSO)
δ 161.7 (d, J = 244 Hz), 145.3, 137.5, 129.5 (d, J = 3 Hz), 128.4 (d, J = 8 Hz), 115.7 (d, J = 22
Hz), 100.8, 100.8, 74.0, 66.3, 17.5, 15.8, -1.03. ESI-MS: Found: 417/418 [M+H]+.
4-bromo-5-(4-fluorophenyl)-1-methyl-2-(methylthio)-1H-imidazole (26):
25 (675 mg, 1.62 mmol) was dissolved in dry DCM and stirred at room temperature. To this
solution, methyl trifluoromethanesulfonate (230 µl, 2.10 mmol) was added and stirring continued
for 2 hours. 1 ml trifluoroacetic acid was added and the solution stirred for 30 minutes at room
temperature. After complete consumption of the imidazolium intermediate, the reaction was
quenched by the addition of concentrated ammonia. The aqueous mixture was extracted three
times with DCM. The combined organic layers were dried over Na2SO4, filtered and the solvents
removed in vacuum. The crude product was purified with flash chromatography (100% DCM) to
1
give the pure product as colorless crystals in 70% yield (252 mg, 1.13 mmol). H NMR (200
MHz, Chloroform-d) δ 7.37 – 7.26 (m, 2H), 7.17 – 7.07 (m, 2H), 3.53 (s, 3H), 2.65 (s, 3H).
LRMS (ESI-MS): not detected.
N-(4-(5-(4-fluorophenyl)-1-methyl-2-(methylthio)-1H-imidazol-4-yl)pyridin-2-yl)acetamide
(27):
To a solution of 26 (158 mg, 0.52 mmol) and N-(4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-
yl)pyridin-2-yl)acetamide (343 mg, 1.31 mmol) in 6 ml degassed 1,4-dioxane was added a n
aqueous 1M K3PO4 solution (1.5 ml). The biphasic mixture was degassed 3 times, by evacuating
and backfilling with argon. Under argon, P(t-Bu)3Pd G3 (5 mol%, 9 mg) was added, the reaction
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 34
1
2
3
4
5
6
7
8
9
flask was sealed and the suspension stirred for 5 hours at 50°C. After complete reaction, the
mixture was cooled to room temperature and brine was added. The aqueous mixture was
extracted three times with DCM. The combined organic layers were dried over Na2SO4, filtered
and the solvents removed in vacuum. The crude product was purified with gradient flash
chromatography (from 100% n-hexane to n-hexane/ethyl acetate 1:2) to give the pure product as
a colorless solid in 39% yield (72 mg, 0.20 mmol). 1H NMR (200 MHz, Chloroform-d) δ 9.46 (s,
1H), 8.27 (s, 1H), 8.03 (d, J = 5.4 Hz, 1H), 7.36 – 7.07 (m, 5H), 3.33 (s, 3H), 2.71 (s, 3H), 2.09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
(s, 3H). C NMR (50 MHz, Chloroform-d) δ 163.12 (d, J = 250 Hz), 151.90, 146.89, 144.31,
144.27, 135.56, 132.34 (d, J = 8.4 Hz), 132.07, 125.78 (d, J = 3.8 Hz), 116.42 (d, J = 22 Hz),
110.99, 31.15, 24.38, 15.53. LRMS (ESI-MS): not detected.
4-(5-(4-fluorophenyl)-1-methyl-2-(methylthio)-1H-imidazol-4-yl)pyridin-2-amine (7):
The intermediate 27 (70 mg, 0.223 mmol) was dissolved in 5 ml methanol and 1 ml of an
aqueous solution of NaOH (5M) was added under stirring. The suspension was stirred for 3
hours at 60°C. After complete reaction, the mixture was cooled to room temperature and a
saturated aqueous NH4Cl solution was added. The aqueous mixture was extracted three times
with ethyl acetate. The combined organic layers were dried over Na2SO4, filtered and the
solvents removed in vacuum. The crude product was purified with gradient flash
chromatography (from 100% n-hexane to n-hexane/ethyl acetate 1:1) to give the pure product as
1
a pale yellow solid in 88% yield (62 mg, 0.196 mmol). LRMS (ESI-MS): 315.5 [M+H]+. H
NMR (200 MHz, Chloroform-d) δ 7.78 (d, J = 5.5 Hz, 1H), 7.34 – 7.23 (m, 2H), 7.22 – 7.09 (m,
2H), 6.76 (d, J = 1.3 Hz, 1H), 6.51 (dd, J = 5.5, 1.5 Hz, 1H), 4.50 (s, 2H), 3.33 (s, 3H), 2.68 (s,
13
3H). C NMR (50 MHz, Chloroform-d) δ 163.01 (d, J = 249.7 Hz), 158.58, 147.25, 143.85,
143.34, 135.83, 132.39 (d, J = 8.3 Hz), 131.44, 126.15 (d, J = 3.5 Hz), 116.28 (d, J = 21.7 Hz),
111.42, 105.22, 31.22, 15.85.
N-(3-((4-(5-(4-fluorophenyl)-1-methyl-2-(methylthio)-1H-imidazol-4-yl)pyridin-2-
yl)amino)-4-methoxyphenyl)acrylamide (7):
A solution of 7 (63 mg, 0,20 mmol), N-(3-bromo-4-methoxyphenyl)acrylamide (23 mg, 0.24
mmol) and cesium carbonate (326 mg, 1.0 mmol) in a mixture of 1,4-dioxane/t-BuOH (8ml +
2ml) was degassed three times by evacuating and backfilling with argon under stirring. Under an
atmosphere of argon, Brettphos Pd G3 (5 mol%, 9 mg, 0.01 mmol) was added. The solution was
stirred under reflux for 3 hours. The mixture was cooled to room temperature and a saturated
aqueous NH4Cl solution was added. The aqueous mixture was extracted three times with ethyl
acetate. The combined organic layers were dried over Na2SO4, filtered and the solvents removed
in vacuum. The crude product was purified with gradient flash chromatography (from 100% n-
hexane to 100% ethyl acetate) to give the pure product as a white solid in 67% yield (74 mg,
0.15 mmol). LRMS (ESI-MS): 490[M+H]+.1H NMR (200 MHz, DMSO-d6) δ 9.98 (s, 1H), 8.22
(s, 1H), 8.01 (s, 1H), 7.91 (d, J = 5.4 Hz, 1H), 7.59 – 7.26 (m, 5H), 7.22 (s, 1H), 6.94 (d, J = 8.8
Hz, 1H), 6.62 – 6.12 (m, 3H), 5.71 (d, J = 9.7 Hz, 1H), 3.81 (s, 3H), 3.34 (s, 3H), 2.68 (s, 3H).
13C NMR (50 MHz, DMSO) δ 163.07, 162.85(d, J = 246.7 Hz), 156.72, 147.25, 146.17, 143.38,
142.94, 135.38, 133.22 (d, J = 8.4 Hz), 132.55, 132.30, 131.94, 130.43, 126.54 (d, J = 3.1 Hz),
126.39, 116.55 (d, J = 21.6 Hz), 113.36, 112.53, 111.82, 111.24, 107.27, 56.20, 31.59, 15.72.
N-(3-((4-(4-(4-fluorophenyl)-1-methyl-2-(methylthio)-1H-imidazol-5-yl)pyridin-2-
yl)amino)-4-methoxyphenyl)acrylamide (6):
26
ACS Paragon Plus Environment

Page 27 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
A solution of 4-(4-(4-fluorophenyl)-1-methyl-2-(methylthio)-1H-imidazol-5-yl)pyridin-2-amine
(60 mg, 0,190 mmol ,synthesized according to WO 2006089798 A1 Aug 31, 2006) and N-(3-
bromo-4-methoxyphenyl)acrylamide (63 mg, 0.247 mmol) and cesium carbonate (311 mg, 0.954
mmol) in a mixture of 1,4-dioxane/t-BuOH (8ml + 2ml) was degassed three times by evacuating
and backfilling with argon under stirring. Under an atmosphere of argon, Brettphos Pd G3 (5
mol%, 9 mg, 0.01 mmol) was added. The solution was stirred under reflux for 3 hours. The
mixture was cooled to room temperature and a saturated aqueous NH4Cl solution was added.
The aqueous mixture was extracted three times with ethyl acetate. The combined organic layers
were dried over Na2SO4, filtered and the solvents removed in vacuum. The crude product was
purified with gradient flash chromatography (from 100% n-hexane to n-hexane/ethyl acetate 1:1)
to give the pure product as a white solid in 81% yield (75 mg, 0.154 mmol). LRMS (ESI-MS):
490.5 [M+H]+. 1H NMR (400 MHz, DMSO-d6) δ 10.01 (s, 1H), 8.46 (d, J = 2.6 Hz, 1H), 8.32 (s,
1H), 8.23 (d, J = 5.2 Hz, 1H), 7.46 (dd, J = 8.7, 5.7 Hz, 2H), 7.42 – 7.37 (m, 1H), 7.16 – 7.08 (m,
2H), 7.04 (s, 1H), 6.96 (d, J = 8.8 Hz, 1H), 6.74 – 6.67 (m, 1H), 6.51 – 6.41 (m, 1H), 6.23 (dd, J
= 17.0, 2.2 Hz, 1H), 5.71 (dd, J = 10.1, 2.2 Hz, 1H), 3.80 (s, 3H), 3.44 (s, 3H), 2.64 (s, 3H). 13C
NMR (101 MHz, DMSO) δ 163.29, 161.51 (d, J = 243.7 Hz), 156.96, 148.52, 146.01, 143.92,
139.53, 137.23, 132.59, 132.35, 131.05, 130.11, 128.77, 128.68 (d, J = 8.0 Hz), 126.55, 115.98,
115.57 (d, J = 21.4 Hz), 113.76, 112.56, 112.41, 111.34, 56.35, 31.98, 15.84.
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
EGFR protein purification. Constructs spanning residues 696–1022 of the human EGFR
(including WT and T790M/V948R mutant sequences) were prepared in a GST-fusion format
using the pTriEX system (Novagen) for expression in Sf9 insect cells as described
previously.3, 33 Briefly, EGFR kinase proteins were purified by Ni-NTA and glutathione-
affinity chromatography in 50 mM Tris-HCl (pH = 8.0), 500 mM NaCl, 5% glycerol, 5 mM
TCEP followed by size-exclusion chromatography after cleavage with Tomato etch virus
(TEV) to remove the GST fusion partner following established procedures.
Structure determination. EGFR(T790M/V948R) pre-incubated with 1 mM AMP-PNP and 10
mM MgCl2 on ice was prepared by hanging-drop vapor diffusion over a reservoir solution
containing 0.1 M Bis-Tris (pH = 5.5), 25% PEG-3350, and 5 mM TCEP (buffer A). WT EGFR
crystals were prepared by prepared by hanging-drop vapor diffusion over a reservoir solution
containing 1.3 M sodium citrate and 0.1 M MES pH = 6.5 (buffer B). Drops containing crystals
in buffer A were exchanged with solutions of buffer A containing ~1.0 mM 1-5 three times for
an hour and then left overnight. WT crystals were soaked with solutions of buffer B containing
10% DMSO and were soaked for 24 - 48 hours prior to harvesting. In both case, crystals were
flash frozen after rapid immersion in a cryoprotectant solution with buffer A or B, for
T790M/V948R or WT, respectively, containing 25% Ethylene glycol. X-ray diffraction data was
collected at 100K the Advanced Light Source a part of the Northeastern Collaborative Access
Team (NE-CAT) on Beamline 24-ID-C or 24-ID-E or at the National Synchrotron Light Source
II 17-ID-1. Diffraction data was processed and merged in Xia2 using aimless and XDS. The
structure was determined by molecular replacement with the program PHASER using the
inactive EGFR kinase from our previous allosteric inhibitor work excluding the EAI001 ligand
(PDB 5D41)29 or WT EGFR (PDB 6D8E)34 Repeated rounds of manual refitting and
crystallographic refinement were performed using COOT and Phenix. The inhibitor was
modeled into the closely fitting positive Fo − Fc electron density and then included in
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 34
1
2
3
4
5
following refinement cycles. Statistics for diffraction data processing and structure refinement
are shown in Table S1.
6
7
Biochemical assays. Inhibition data (IC50) against EGFR-wildtype, EGFR-L858R/T790M and
EGFR-L858R/T790M/C797S were conducted at Reaction Biology corp (Malvern, PA) via
commercial radiolabeled P(ATP) kinase assays. IC50 values were measured in duplicates at 5
8
9
33
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
different concentration with 10-fold dilution steps starting from 5 µM. Staurosporine was tested
as control compound in singlicate at 10 different concentrations with a 4-fold dilution starting at
20 µM. A detailed description of the assay format is given in the literature.35-36
ASSOCIATED CONTENT
Supporting Information
A listing of the contents of each file supplied as Supporting Information should be included. For
instructions on what should be included in the Supporting Information as well as how to prepare
this material for publication, refer to the journal’s Instructions for Authors.
The Supporting Information is available free of charge on the ACS Publications website.
Crystallography and biochemical data (file type, i.e., PDF)
AUTHOR INFORMATION
Corresponding Authors
Michael J. Eck: eck@crystal.harvard.edu
Stefan A. Laufer: stefan.laufer@uni-tuebingen.de
OCRIDs
DEH: 0000-0002-0722-5160
MG: 0000-0002-4921-462X
FW: 0000-0001-9028-2109
SAL: 0000-0001-6952-1486
MJE: 0000-0003-4247-9403
Author Contributions
The manuscript was written through contributions of all authors. All authors have given approval
to the final version of the manuscript.
Funding Sources
This work was supported by the National Institutes of Health grants R01CA201049 and
R01CA116020 (to M.J.E). Funded by the Deutsche Forschungsgemeinschaft (DFG, German
Research Foundation) under Germany's Excellence Strategy - EXC 2180 – 390900677 (to
S.A.L). This work is based upon research as a part of the Cluster of Excellence iFIT (EXC 2180)
28
ACS Paragon Plus Environment

Page 29 of 34
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
"Image-Guided and Functionally Instructed Tumor Therapies", University of Tübingen,
Germany. This work is based upon research conducted at the Northeastern Collaborative Access
Team (NECAT) beamlines, which are funded by the National Institute of General Medical
Sciences from the National Institutes of Health (P30 GM124165). The Pilatus 6M detector on
24-ID-C beam line is funded by a NIH-ORIP HEI grant (S10 RR029205). This research used
resources of the Advanced Photon Source, a U.S. Department of Energy (DOE) Office of
Science User Facility operated for the DOE Office of Science by Argonne National Laboratory
under Contract No. DE-AC02-06CH11357. This research used resources 17-ID FMX and AMX
of the National Synchrotron Light Source II, a U.S. Department of Energy (DOE) Office of
Science User Facility operated for the DOE Office of Science by Brookhaven National
Laboratory under Contract No. DE-SC0012704. The Life Science Biomedical Technology
Research resource is primarily supported by the National Institute of Health, National Institute of
General Medical Sciences (NIGMS) through a Biomedical Technology Research Resource P41
grant (P41GM111244), and by the DOE Office of Biological and Environmental Research
(KP1605010).
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
Notes
Conflicts of Interest
M.J.E. reports receiving a commercial research grant from Novartis Institutes for Biomedical
Research, reports receiving other commercial research support from Takeda, and has been a
consultant for Novartis Institutes for Biomedical Research.
ACKNOWLEDGMENT
We thank Dr. Tyler S. Beyett for technical assistance and helpful discussions.
ABBREVIATIONS
Tyrosine kinase inhibitors (TKIs), Epidermal growth factor receptor (EGFR), non-small cell lung
cancer (NSCLC), H-bond (hydrogen bond).
Associated Content
Supporting Information
The supporting information is available free of charge on the ACS Publications Website at DOI:
X-ray crystallography statistics; supporting structural images; biochemical potencies of
1-3 and 5; additional chemistry for precursor.
SMILES molecular formula strings.
Accession Codes
29
ACS Paragon Plus Environment