From off-to on-target: New BRAF-inhibitor-template-derived compounds selectively targeting mitogen activated protein kinase kinase 4 (MKK4)

Article abstract of DOI:10.1016/j.ejmech.2020.112963

The mitogen-activated protein kinase (MAP) kinase 4 (MKK4) was found to be a major regulator of liver regeneration and could be a valuable drug target addressing liver related diseases by restoring its intrinsic regenerative capacity. We report on the synthesis and optimization of novel MKK4 inhibitors following a target-hopping strategy from the FDA-approved BRAF^V600E inhibitor PLX4032 (8). Applying an iterative multi-parameter optimization process we carved out essential structural features yielding in compounds with a low nanomolar affinity for MKK4 and excellent selectivity profiles against the main off-targets MKK7 and JNK1, which, upon relevant inhibition, would totally abrogate the pro-regenerative effect of MKK4 inhibition, as well as against the off-targets MAP4K5, ZAK and BRAF with selectivity factors ranging from 40 to 430 for our best-balanced compounds 70 and 73.

Full text of DOI:10.1016/j.ejmech.2020.112963

Journal Pre-proof
From off-to on-target: New BRAF-inhibitor-template-derived compounds selectively
targeting mitogen activated protein kinase kinase 4 (MKK4)
Philip Klövekorn, Bent Pfaffenrot, Michael Juchum, Roland Selig, Wolfgang Albrecht,
Lars Zender, Stefan A. Laufer
PII:
S0223-5234(20)30935-1
DOI:
https://doi.org/10.1016/j.ejmech.2020.112963
EJMECH 112963
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 24 July 2020
Revised Date: 19 October 2020
Accepted Date: 20 October 2020
Please cite this article as: P. Klövekorn, B. Pfaffenrot, M. Juchum, R. Selig, W. Albrecht, L. Zender, S.A.
Laufer, From off-to on-target: New BRAF-inhibitor-template-derived compounds selectively targeting
mitogen activated protein kinase kinase 4 (MKK4), European Journal of Medicinal Chemistry, https://
doi.org/10.1016/j.ejmech.2020.112963.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
©
2020 Published by Elsevier Masson SAS.

GRAPHICAL ABSTRACT.

From off- to on-target: New BRAF-inhibitor-
template-derived compounds selectively targeting
mitogen activated protein kinase kinase 4 (MKK4)
†
†
†
┴
┴
Philip Klövekorn, Bent Pfaffenrot, Michael Juchum, Roland Selig, Wolfgang Albrecht,
‡
X 3
,†X34
Lars Zender
and Stefan A. Laufer*
†
Department of Pharmaceutical/Medicinal Chemistry, Eberhard Karls Universität Tübingen,
Auf der Morgenstelle 8, 72076 Tübingen, DE, Germany
‡
Department of Medical Oncology and Pneumology (Internal Medicine VIII), University
Hospital Tuebingen, 72076 Tuebingen, DE, Germany
┴
HepaRegenix GmbH, Alfred-Mendler-Weg 25/1, 89075 Ulm
X Cluster of Excellence ‘Image Guided and Functionally Instructed Tumor Therapies’ (iFIT),
Eberhard Karls University of Tübingen, 72076 Tübingen, Germany.
3
German Consortium for Translational Cancer Research (DKTK), Partner Site Tübingen,
German Cancer Research Center (DKFZ), 69120 Heidelberg, Germany.
4
Tuebingen Center for Academic Drug Discovery, Auf der Morgenstelle 8, 72076 Tübingen,
DE, Germany

KEYWORDS.
Acute and chronic liver failure, liver failure, NAFLD, NASH, liver regeneration, MEK4,
MKK4, inhibitors, Vemurafenib
ABSTRACT.
The mitogen-activated protein kinase (MAP) kinase 4 (MKK4) was found to be a major
regulator of liver regeneration and could be a valuable drug target addressing liver related
diseases by restoring its intrinsic regenerative capacity. We report on the synthesis and
optimization of novel MKK4 inhibitors following a target-hopping strategy from the FDA-
V600E
approved BRAF
inhibitor PLX4032 (8). Applying an iterative multi-parameter
optimization process we carved out essential structural features yielding in compounds with a
low nanomolar affinity for MKK4 and excellent selectivity profiles against the main off-
targets MKK7 and JNK1, which, upon relevant inhibition, would totally abrogate the pro-
regenerative effect of MKK4 inhibition, as well as against the off-targets MAP4K5, ZAK and
BRAF with selectivity factors ranging from 40 – 430 for our best-balanced compounds 70 and
7
3.
INTRODUCTION.
The liver harbors an enormous regenerative capacity in response to injury, but after chronic
liver damage this regenerative ability can collapse with a mostly fatal outcome for the patient.
Driven by obesity and metabolic syndrome, non-alcoholic fatty liver disease (NAFLD)
including its progressive form, non-alcoholic steatohepatitis (NASH), belong to the major
health issues in industrialized countries affecting nearly 2 billion people worldwide [1,2].

While viral infection can be cured with antiviral drugs, the treatment of NASH is limited to
drastic lifestyle changes and prevention [3]. Based on the medical need for treatment of
NASH or its prevention, several therapeutic concepts are currently under clinical
investigation. Most of these candidates address established metabolic disease targets such as
subtypes of peroxisome proliferator-activated receptors (elafibranor / α/δ, lanifibranor /α/γ/δ,
or saroglitazar / α/γ), the farnesoid x receptor (obeticholic acid, cilofexor or tropifexor),
acetyl-CoA carboxylase (firsocostat) and others.
Wüstefeld et al. identified mitogen-activated protein (MAP) kinase kinase 4 (MKK4) as a
key promotor for liver regeneration with beneficial effects on hepatocyte regeneration,
robustness, fibrosis and Fas-mediated apoptosis [4]. On the molecular level, silencing of
MKK4 via shRNA results in increased signaling through apoptosis signal-regulating kinase 1
(ASK1) and mitogen activated protein (MAP) kinas kinase 7 (MKK7) and thus yielding in
higher phosphorylation of c-Jun-N-terminal kinase 1 (JNK1). Thereby, phosphorylation of
ETS transcription factor (ELK1) and activating transcription factor 2 (ATF2) is intensified,
which leads to a reinforced hepatocyte proliferation. Due to the involved signaling network
MKK7 and JNK1 have to be considered as major off-targets (see Figure 1).
Figure 1. Schematic visualization of signal transduction upon MKK4 silencing modified from
Wüstefeld et al. [4].

Since MKK4 plays a pivotal role in cell growth, differentiation and inflammation processes
this kinase could represent a valuable drug target for different diseases [5-7]. However, only a
handful of inhibitors are published until now (see Figure 2). Bayer published MKK4 and
MKK7 inhibitors based on 9H-pyrimido[4,5-b]indol-6-ol scaffold (1 and 2) with IC values
5
0
below 1 µM [8,9]. Also, natural products like isoflavones (THIF 3) show inhibitory potency
towards MKK4 but with lacking selectivity [10]. Recently, Deibler et al. published
arylindazoles showing a great selectivity profile within the MAP2-kinase family and low
nanomolar IC50 range (exemplary 7) [11,12]. These structures were deduced from known
MKK4 inhibitors like pazopanib (4), PLX4720 (5) and LY333531 (6).
Figure 2. Published inhibitors for MKK4. Top: published MKK4/MKK7 dual-inhibitors (1
and 2) and THIF (3) by Kim et al.; bottom: inhibitors with off-target-inhibition for MKK4
used by Deibler et al. for in silico design of novel small molecules.
In this work, we describe the optimization process of 8, an approved v-Raf murine sarcoma
V600E
viral oncogene homolog B1 (BRAF
) kinase inhibitor for the treatment of malignant
melanoma, towards a high affine and selective MKK4 inhibitor. Initially we tested 8 and
compounds 1 and 2 on a panel of important off-targets, namely BRAFwt, MEK kinase kinase

5
(MAP4K5), sterile alpha motif and leucine zipper containing kinase (ZAK), MKK7 and
JNK1 regarding their binding affinities. Among other MAP4 and MAP3 kinases upstream of
JNK both MAP4K5 and ZAK are known activators of JNK. Therefore, to achieve a beneficial
pharmacological effect as obtained with shRNA-mediated MKK4-suppression, the small
molecule MKK4 inhibitor should exhibit sufficient selectivity against MAP4K5 and ZAK
[
13]. Therefore, hit optimization included the determination of the affinity to all known
potential off-targets employing the KINOMEscan technology by Eurofins DiscoverX.
Binding affinities are given as percentage of control (POC) values. In the assay, the
compounds compete with a proprietary immobilized multikinase inhibitor for the binding site
of the DNA-tagged kinase. The readout is via quantitative PCR and the POC is calculated
from the DMSO control. So, a POC of 0 is equal to a complete displacement of the ligand
from the kinase (complete binding) and a POC of 100 is equal to no binding of the compound
to the kinase [14].
Table 1. Initial testing of 1, 2 and 8 as potential starting points for further optimization.
No.
BRAF
MAP4K5
MKK4
MKK7
JNK1
ZAK
a
%
Ctrl at 100 nM
1
2
8
91
82
16
95
92
48
100
100
100
100
88
80
94
81
84
48
14
5.3
a
Binding affinities from KINOMEscan assay at a compound concentration of 100 nM.
Values are shown as POC determined in duplicate.
RESULTS AND DISCUSSION.

Biological evaluation. In the following all structural features of 8 (sulfonamide residue,
substitution of the moiety in 5 position of the azaindole and substitution pattern of the difluoro
phenyl part) will be analyzed successively taking affinity and selectivity towards all before
mentioned off-targets, as well as MKK4 into account, while maintaining the azaindole
scaffold.
Modification of the sulfonamide. At first, we evaluated the role of the sulfonamide residue
of 8 with regards to affinity and selectivity towards MKK4. The substitution of the
sulfonamide by an amide eliminates the affinity towards MKK4 (entry 9, Table 2).
Table 2. Investigations on the sulfonamide structural elements.
No.
8
R
BRAF
16
MAP4K5
MKK4
MKK7
JNK1
94
ZAK
5.3
a
%
Ctrl at 100 nM
48
91
14
100
94
9
82
100
61
78
a
Binding affinities from KINOMEscan assay at a compound concentration of 100 nM.
Values are shown as POC determined in duplicate.
The propyl sulfonamide chain in 8 was optimized for a unique pocket in the BRAF family
[
15]. We speculated that modifications of this propyl side chain would increase selectivity,
especially over BRAF. Therefore, chain lengths, aromatic modifications and incorporation of
polar heteroatoms were examined on their binding effects to MKK4. We started by varying
the length of the alkyl chains from methyl to hexyl. Especially for BRAF, we could observe
an effect on the structure-activity-relationship. With altering the propyl chain, the binding
affinity drops dramatically from POC 16 to at least 89 on BRAF (entries 10 – 14, Table 3).

Regarding MKK4, the propyl chain exhibits the highest affinity and modifications thereof are
more tolerated as for BRAF, while the hexyl moiety is the least tolerated larger side chain.
Fluorination of the propyl sidechain gave varying results. MKK4 affinity of the
monofluorinated compound 15 is comparable to 8, but selectivity against BRAF and ZAK is
reduced. The trifluoro compound 16 in contrast, showed a highly increased affinity for
MKK4, while the affinity for BRAF dropped by factor 2. Polar modifications such as the
methoxyethyl chain (entry 17, Table 3) were not well tolerated by MKK4. To install a bulkier
residue, we attached a phenyl ring to the sulfonamide with variation in spacer length (n = 0 –
2
, entries 18 – 20, Table 3). Compared to 8 the affinity towards MKK4 increased by a factor
of 2 for the benzyl group (19) and 4 for the phenyl group (18), respectively. The ethylene
linker (20) led to a loss in affinity to MKK4. Compound 19 showed improved affinity
MKK4
(POC
= 5.9) and the best selectivity profile among these modifications. Whether
substitution of the benzyl moiety has a further impact on selectivity, this modification was
investigated more detailed with compounds 21 – 25. Fluorine was chosen as substituent to
further increase affinity according to 15 and 16, but no striking effect could be observed. Even
though 21, 24 and 25 showed a higher selectivity against ZAK, the affinity for MKK4 was
lower compared to the unsubstituted compound 19. Just 23 has a comparable profile to 19.
Table 3. Scope of sulfonamide substitution patterns.
No.
R
BRAF
93
MAP4K5
63
MKK4
MKK7
JNK1
100
ZAK
2.6
a
%
Ctrl at 100 nM
10
32
87

1
1
93
16
82
48
43
14
23
29
45
10
2.8
47
3.3
5.9
24
15
13
6
89
100
100
89
100
94
8.3
5.3
29
55
73
1.4
3.5
58
4.9
30
53
61
36
23
87
64
8
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
3
4
5
6
7
8
9
0
1
2
3
4
5
89
75
100
100
100
95
96
98
96
99
81
6.6
32
80
100
100
75
54
100
100
100
100
100
100
97
91
100
83
22
83
100
99
94
100
97
100
n.d.
n.d.
n.d.
n.d.
n.d.
95
100
99
100
100
100
100
97
100
99
19
16
90
100
100
a
Binding affinities from KINOMEscan assay at a compound concentration of 100 nM.
Values are shown as POC determined in duplicate; n.d.= not determinded.

Variation of the phenyl ring in position 5 of the azaindole core. The impact of the
moiety in position 5 of the azaindole of 8 regarding binding affinity and selectivity was
further examined. Initially, the 4-chlorophenyl moiety of 8, was replaced with a phenyl ring
(
26) and the corresponding isosteric alkyne (27). Furthermore, 2-furyl- (28) and 2-thiophenyl
29) compounds were prepared (see Table 4).
(
Table 4. Isosteric replacement of the 4-chlorophenyl part of 8.
No.
8
R
BRAF
MAP4K5
MKK4
MKK7
JNK1
ZAK
a
%
Ctrl at 100 nM
16
21
8.2
9
48
29
89
78
48
14
2.9
11
100
100
100
95
94
81
5.3
0.5
1.4
2.1
2.9
2
2
2
2
6
7
8
9
84
7.3
4.5
94
13
98
100
a
Binding affinities from KINOMEscan assay at a compound concentration of 100 nM.
Values are shown as POC determined in duplicate.
The unsubstituted isosteric derivatives 26 – 29 are showing higher affinity values for
MKK4 compared to 8. Therefore, we further investigated the substitution pattern of the
phenyl moiety in the 5 position of the azaindole (Table 5). First, we elaborated the effect of
monosubstituted derivatives and proceeded with di- and trisubstitution to combine beneficial
moieties of each optimization cycle. Altering substituents in the para-position of the 5-phenyl

moiety increases the affinity towards MKK4 when chlorine is substituted by fluorine, methyl
or a methoxy group (entries 32, 35 and 38, Table 5). Among the regioisomeric mono-
substitution patterns the meta-position (30, 33, 36, Table 5) has no relevant effect, except for
the m-methoxy derivative (entry 39, Table 5) that has a higher affinity to MKK4, but
unfortunately also to BRAF and MAP4K5 compared to 8. The ortho-regioisomers are
showing a positive effect on the selectivity towards BRAF and slightly on ZAK. Substitution
patterns containing o-chloro, o-fluoro, o-methyl and o-methoxy (entries 31, 34, 37 and 40,
Table 5) led to decrease in binding affinity for BRAF compared to the meta- and para-
isomers. This ‘ortho-effect’ was further investigated by the combination of 32 and 38 with
compounds 31, 34 and 37 (entries 41 – 46, Table 5). Entries 45 and 46 represent the most
superior derivatives of these combinations in terms of selectivity and affinity towards MKK4.
The m- and o-methoxy derivatives were also combined with p-methoxy residues (entries 47
and 48, Table 5). Herein, the meta substituted compound (47) had a higher selectivity towards
ZAK but low selectivity towards BRAF, while selectivity to ZAK and BRAF was opposite
for the ortho-methoxy compound (48). Bridging the 3,4-dimethoxy groups with a methylene
and ethylene linker (entries 49 and 51, Table 5), increased the selectivity to BRAF but also
increased the affinity to ZAK. Ongoing modifications with o-chloro substituents led to a
further increase of selectivity (50 and 52, Table 5). These combinations afforded highly affine
inhibitors of MKK4 sparing BRAF, MAP4K5, MKK7, JNK1 and ZAK moderately. In this
example, the ortho chlorine was identified as an outstanding selectivity driver. We also
focused on heteroaromatic moieties like pyridine, pyridazine and pyrimidine (entries 53 – 57,
Table 5). Entries 56 and 57 were inspired by 38 and PLX8394, a further derivative of 8,
which inhibits ERK-signaling without undesired JNK-activation and is currently under
clinical investigations [16-18]. Compared to the substituted phenyl derivatives, the
heteroaromatic compounds have a higher affinity towards MKK4, but also with a higher
binding to the off-targets as well. Among this series, compound 57 showed an improvement

of ZAK-selectivity compared to 38 and 56. We assume, that the attached cyclopropyl moiety
could be in tight proximity to the extended p-loop of ZAK and thus clash with the protein
[
19].
Table 5. Investigation on the 5-position of the azaindole scaffold.
No.
8
R
BRAF
MAP4K5 MKK4
MKK7
JNK1
ZAK
a
%
Ctrl at 100 nM
16
47
65
27
31
47
11
29
72
6.7
48
n.d.
79
14
24
6.2
4.3
9.9
4.1
5
100
85
94
100
93
96
98
69
76
92
88
89
5.3
2.2
7.6
0.9
0.8
2.9
0.85
0.7
6.9
0
3
3
3
3
3
3
3
3
3
0
1
2
3
4
5
6
7
8
100
100
90
67
45
n.d.
28
86
89
28
13
10
6.3
97
67
100
70
38

3
4
4
4
4
4
4
4
4
4
4
9
0
1
2
3
4
5
6
7
8
9
10
64
18
43
5.4
2.8
8.8
4.2
18
100
100
100
93
77
80
23
16
69
100
n.d.
100
n.d.
n.d.
100
14
97
14
46
87
5.9
19
85
100
97
100
100
85
75
7.7
2.4
2.7
1.3
4.5
3.1
13
29
85
6
70
100
84
97
18
0.55
31
65
18
84
87
76
1.6
0.45
13
76
90
94
5
5
5
0
1
2
85
5.7
72
99
84
98
11
2.9
5.3
100
100
97
100
96
26
2.2
20
99
5
5
3
4
1.4
8.7
80
92
1.7
11
81
90
65
83
3.5
39

5
5
5
5
6
7
2.2
0.75
1.1
75
86
2.5
0.3
0.2
83
84
87
74
73
11
8.9
30
n.d.
100
a
Binding affinities from KINOMEscan assay at a compound concentration of 100 nM.
Values are shown as POC determined in duplicate; n.d.= not determinded.
Alteration of the difluoro pattern. In the ongoing course of our studies, we examined the
impact of the difluorophenyl moiety of 8. This part is considered to have an impact on the
hydrophobic interaction, since the fluorine atoms are embedded in a hydrophobic pocket in
BRAF [15]. To identify the relevant substitution, we initially tested the unsubstituted
compound 58. Removing the fluorine atoms led to a significant drop in affinity towards
BRAF but also to a loss in affinity to the whole panel of kinases, including MKK4.
Concluding the necessity of the fluorine from these results, mono substitution patterns (59 –
6
4, Table 6) were investigated, which revealed that the position 2 needs to be substituted with
fluorine (8, 59) in order to maintain affinity towards MKK4. Substituting fluorine in 2-
position with chlorine or methyl (63, 64) led to a loss in affinity for MKK4. Compound 59 (2-
fluorophenyl) has a comparable affinity to MKK4 as 8, but with improved selectivity against
BRAF by factor 2.
Table 6. Altered substitution pattern at the difluorophenyl part of 8.
No.
R
BRAF
MAP4K5
MKK4
MKK7
JNK1
ZAK

a
%
Ctrl at 100 nM
8
2,6-F
-H
16
100
31
94
91
82
31
94
48
44
51
76
99
60
79
40
14
53
16
32
100
45
78
92
100
94
98
5.3
n.d.
5
5
8
9
0
1
2
3
4
100
92
5
6
6
6
6
6
2-F
4-F
5-F
6-F
2-Cl
97
93
95
37
93
100
100
100
n.d.
n.d.
16
92
100
91
68
2-CH
n.d.
3
a
Binding affinities from KINOMEscan assay at a compound concentration of 100 nM.
Values are shown as POC determined in duplicate; n.d.= not determinded.
Combinatorial approach. From previous results, we combined structural elements which
improved affinity on MKK4 and/or led to a markedly better selectivity profile. By
combination of the 2-fluorophenyl moiety of 59 and the benzyl sulfonamide group of 19
(
yielding entry 66, Table 7) the affinity towards MKK4 was increased by the factor of 2 while
the selectivity ratio to all other kinases improved by a factor range of 7 – 11, compared to 8.
In comparison to the difluorophenyl derivative 19, the monofluorophenyl compound 66
showed an improved selectivity against ZAK with comparable affinity towards MKK4. The
combination of the benzyl sulfonamide and 2-fluoro phenyl moiety yielded an excellent
selectivity profile and was thus used for the final modifications.
Table 7. Monofluoro substituted compounds 59, 65 and 66 in comparison to their difluoro
counterparts.

No.
8
X
Y
BRAF
MAP4K5 MKK4
MKK7
JNK1
ZAK
a
%
Ctrl at 100 nM
-F
-H
-F
-propyl
-propyl
-methyl
-benzyl
-methyl
-benzyl
16
31
48
51
63
94
78
72
14
16
100
92
94
5.3
5
5
1
1
6
6
9
0
9
5
6
97
93
32
87
100
100
100
100
2.6
30
27
66
-F
100
91
5.9
28
100
81
-H
-H
100
8.9
97
a
Binding affinities from KINOMEscan assay at a compound concentration of 100 nM.
Values are shown as POC determined in duplicate.
Final modifications. As last stage of this optimization study, we combined the selectivity
and affinity increasing residues from 66 (benzylsulfonamide and 2-fluorophenyl moiety) with
aryl residues of the 5 position of the azaindole from Table 5, that were found to have an
already high affinity to MKK4 but with still lacking selectivity, especially towards BRAF and
ZAK. We selected the residues from 41, 46, 50 and 52 of the 5-position of the azaindole to
combine with the 2-fluorophenyl linker and the benzylsulfonamide from 66 (Entries 67 – 70,
Table 8). Within this selection entries 68 and 70 are excellent compounds showing high
affinity to MKK4 and a good selectivity profile within the kinase panel. Furthermore, the 5-
position of the azaindole was substituted with pyridines (entries 71 and 72, Table 8) and the
residues from 56 and 57 were chosen for this combination (entries 73 and 74, Table 8).
Pyridines 71 and 72 showed a high affinity for MKK4, but also a strong affinity towards
ZAK. Furthermore, MAP4K5 arises as an undesired off-target affinity for these compounds.

The combined 4-methoxy and 4-cyclopropyl pyrimidine compounds 73 and 74 have
outstandingly high affinity to MKK4, while the critical off-targets were discriminated by the
factor >190, making these compounds the best compounds from all compounds prepared,
concerning affinity and selectivity for MKK4.

Table 8. Combinational compounds derived from 66.
O
O
R
S
O
N
F
H
N
N
H
No.
8
R
BRAF
MAP4K5 MKK4
MKK7
JNK1
ZAK
a
%
Ctrl at 100 nM
-
16
100
96
48
72
79
97
66
98
26
20
86
82
14
8.9
2.4
3.3
2.9
2.2
0
100
97
94
100
100
100
97
5.3
66
6
6
6
6
7
7
7
7
7
6
7
8
9
0
1
2
3
4
100
98
87
100
98
90
100
100
95
80
99
97
94
86
75
0.95
1.7
38
91
0
85
55
O
N
84
0.2
0.35
70
86
N
83
86
92
80
a
Binding affinities from KINOMEscan assay at a compound concentration of 100 nM.
Values are shown as POC determined in duplicate.
For a further biological characterization of the most potent compounds, we determined
Kd-values for MKK4 for 70, 73 and 74 as well as for 8 for comprehension (see Table 9).

Table 9. Comparison of Kd-values (MKK4) for the most potent compounds.
a
No.
8
Structure
Kd-Value (nM)
8.2
7
7
7
0
3
4
3.5
0.85
1.1
a
Kd-values for MKK4 from KINOMEscan, determined in duplicates.
The Kd-values for MKK4 were within in the low one-digit nanomalor range for all tested
compounds. Our newly developed compounds outperformed 8 with a gain in affinity towards
MKK4 up to factor 10 (73), combined with a superior selectivity profile. Compounds 73 and
7
4, showing Kd-values of 0.85 and 1.1 respectively, can be seen as equal in affinity towards
MKK4. From the previous testing (see Table 8), 74 was the compound with the best
selectivity and also one with the highest affinity for MKK4. We determined the kinome
selectivity for 74 using the scanEDGE assay provided by DiscoverX at a concentration of 1
µM (1000x Kd-value). The panel consists of 97 kinases, from which 90 are non-mutant
kinases. In this panel, a POC lower than 35 was measured only for 5 kinases (1 mutant-kinase,
see SI), whereas a POC lower than 10 was not observed. This leads to a selectivity score of
(S35) = 0.044 at 1 µM.
CHEMISTRY.

The synthetic route was mainly adapted from the procedures published in the corresponding
patent [20]. Due to strategic issues, the order of introducing building blocks was modified in
particular cases. Scheme 1 displays the synthetic route to obtain 2,6-difluoro-3-
(propylsulfonamido)benzoic acid (77), which was used in a subsequent Friedel-Crafts
acylation (FC acylation) under standard conditions using AlCl as mediator for acylation. The
3
benzoic acid 77 was synthesized following the procedure of Wenglowsky et al. [21]. The
intermediate 78 was finally modified using the corresponding boronic acids or the pinacol
ester thereof according to the procedure from Buck et al. under microwave irradiation
conditions [22]. This procedure was also applied to synthesize all modified benzoic acids used
for the synthesis of compounds 58 – 64. For compound 63 the reduction step (Scheme 1, b or
c) was performed using a Béchamp-like procedure to avoid hydrodehalogenation.
Scheme 1. Top: Synthesis of substituted carboxylic acids; here, exemplary synthesis of 2,6-
difluoro-3-(propylsulfonamido)benzoic acid; bottom: synthesis and modification of 81 using
a
microwave irradiation.
a
Reagents and conditions: (a) Oxalyl chloride, DMF, DCM, RT, then MeOH (quant.); (b)
Pd/C, H , EtOH (quant.); (c) Fe, 1M HCl ., EtOH (98%); (d) 1-Propanesulfonyl chloride,
2
aq
TEA, DCM, RT (85%); (e) 4N NaOH , THF/MeOH (4:1, v/v), RT (80%); (f) 77, oxalyl
aq.
chloride, DMF, AlCl , DCM, RT (85%); (g) corresponding boronic acid/pinacol ester
3
boronate, Pd(PPh ) , K CO , 1,4-dioxane/H O (4:1), µw, 120°C, 15 – 90 minutes, (25 –
3
4
2
3
2
9
1%).

For varying the sulfonamide substitution pattern, the commercially available 5-bromo-1H-
pyrrolo[2,3-b]pyridine was substituted in the first step with (4-chlorophenyl)boronic acid
applying Suzuki coupling resulting 79 in high yields (Scheme 2). Using the FC acylation
according to the procedure of Zhang et al., the 2,6-difluoro-3-nitrobenzoic acid was
introduced (79 to 80) [18]. Using nitromethane as solvent allowed a workup without the need
for further column purification. After reduction of the nitro group in 80 to the corresponding
aniline 81 with SnCl dihydrate, the desired sulfonyl chlorides were used to assemble the
2
sulfonamides.
a
Scheme 2: Synthesis of the common precursor 81 varying the sulfonamide residue.
F
Cl
Cl
O
Br
h
i
NO2
N
N
F
H
N
N
H
N
N
H
79
80
j
F
Cl
O
k
NH2
17 examples
F
N
N
H
8
1
a
Reagents and conditions: (h) (4-Chlorophenyl)boronic acid, Pd(PPh ) , K CO ,
3
4
2
3
MeCN/H O (4:1, v/v), 60°C (87%); (i) 2,6-difluoro-3-nitrobenzoic acid, oxalyl chloride,
2
DMF, AlCl , MeNO , 60°C (67%); (j) SnCl dihydrate, THF/EtOAc (1:1, v/v), 60°C (89%);
3
2
2
(k) corresponding sulfonyl chloride, pyridine, 60°C (27 – 91%).
The synthetic strategy for compounds 66 – 74 is depicted in Scheme 3. All steps to
intermediate 84 were performed as shown above. Suzuki coupling was again performed under
microwave irradiation to introduce the favored boronic acids within a short reaction time. For
compounds 71 – 74 the route was altered by a protection step of the pyrrole NH introducing

2
,6-dichlorobenzoyl protection group, followed by a Miyaura borylation, yielding the
intermediate 86 in high yields. For final modifications the corresponding hetero aryl bromides
were used, reacting in the presence of Pd(dppf)Cl and the crude product was deprotected
2
with K CO in MeOH.
2
3
Scheme 3: Synthesis of 66 – 74 from commercially available 5-bromo-1H-pyrrolo[2,3-
a
b]pyridine.
a
Reagents and conditions: (l) 2-fluoro-3-nitrobenzoic acid, oxalyl chloride, DMF, AlCl3,
DCM, RT (84%); (m) SnCl2 dihydrate, THF/EtOAc (1:1, v/v), 60°C (76%); (n)
phenylmethanesulfonyl chloride, pyridine, 60°C (64%); (o) corresponding boronic
acid/pinacol ester boronate, Pd(PPh ) , K CO , 1,4-dioxane/H O (4:1), µw, 120°C, 15 – 90
3
4
2
3
2
min, (33 – 64%); (p) 2,6-dichlorobenzoyl chloride, TEA, THF, 4-DMAP, 0°C – RT (93%);
q) bis(pinacolato)diboron, AcOK, Pd(PPh ) Cl , 1,4-dioxane, 80°C (60%); (r) aryl bromides,
(
3
2
2
K CO , Pd(dppf)Cl , 1,4-dioxane/H O (4:1, v/v), 60°C, then K CO , MeOH (29 – 55%, over
2
3
2
2
2
3
two steps).
CONCLUSION.
In this work, we reported on the development of novel inhibitors with a high affinity to the
on-target MKK4 and an outstanding selectivity profile towards the off-targets MKK7, JNK1,
V600E
BRAFwt, MAP4K5 and ZAK, all derived from the BRAF
inhibitor-template 8. Our
synthetic efforts yielded a set of compounds, which points out structural hot spots driving
affinity and selectivity. We performed a multi-parameter optimization process emphasizing a
distinct SAR. With compound 70, 73 and 74 we demonstrated the design of new inhibitors

MKK4
with a high affinity to MKK4 (POC
= 2.2, 0.2 and 0.35 @ 0.1 µM) while sparing the off-
targets in the range of factor 40 to 190. To our knowledge, these compounds are the first-in-
class inhibitors selectively targeting MKK4 for the purpose of hepatocyte proliferation to
restore the liver’s intrinsic regenerative capacity. The findings of our work could be
implemented in further investigations on the valuable drug target MKK4 in terms of using 70,
7
3 and 74 as tool compounds.
EXPERIMENTAL SECTION.
General. All commercially purchased chemicals and solvents were used as received
without further purification. Air- or moisture-sensitive reactions were performed under argon
atmosphere and in anhydrous solvents (Acros Organics, AcroSeal). Organic extracts were
dried over anhydrous sodium sulfate which was filtered off afterwards. For reaction controls,
thin layer chromatography was applied using TLC Silica Gel 60 F254 aluminum sheets
provided by Merck, detection at λ = 254 nm und 366 nm. Flash chromatography was
performed manually or with Interchim PuriFlash F430 systems and Grace Davison Davisil
LC60A 20 – 45 microns silica. Purity determination of all final compounds was measured
with HPLC analysis on Agilent 1100 Series Liquid Chromatograph using a Phenomenex Luna
C8 150x4.6 mm, 5-micron column with gradient elution (MeOH/0.01 M KH PO buffer, pH
2
4
2
.3, flow rate 1.5 mL/minute) and detection at λ = 230 and 254 nm. The minimum purity of
all final compounds is >95%, if not otherwise quoted. Mass spectra (TLC-MS) for
intermediates was performed on Advion DCMS interface (ESI voltage: 3.50 kV, capillary
voltage: 187 V, source voltage: 44 V, capillary temperature: 250 °C, desolvation gas
temperature: 250 °C, gas flowrate: 5L/minute N ), using MeOH as elution solvent. High
2
resolution mass spectra measurements for final compounds were performed by Mass
Spectrometry Department, Eberhard Karls Universitaet Tuebingen on Bruker maXis 4G ESI-

TOF (capillary voltage: 4 kV, endplate voltage -500 V, source temperature 200°C, gas flow
rate: 6 L/minute). NMR-spectra were measured on Bruker Avance 200 or 400, Avance III
HDX 600 and 700. The chemical shifts (δ) are denoted by parts per million (ppm) relative to
tetramethylsilane. Herein not listed intermediates and final compounds are described in the
supporting information named R and/or with its given number from the text above.
Suzuki-coupling under microwave irradiation (General procedure A). A microwave
tube was equipped with a magnetic stir bar. The vessel was charged with the corresponding 5-
bromo-1H-pyrrolo[2,3-b]pyridine (78 or 84) (1.0 eq.), the appropriate boronic acid/ or pinacol
ester boronate (1.2 eq.), solid K CO (2.0 eq.) and 1,4-dioxane/water (4:1, v/v, 0.25 M). The
2
3
suspension was purged with argon (5 minutes), the catalyst (10 mol%) was added, and the
resulting mixture was heated to 120°C under microwave irradiation (50 W) for 30 minutes.
When TLC revealed completion of the reaction, the crude was passed through a pad of Celite,
diluted with EtOAc and neutralized with saturated aqueous NH Cl. The organic phase was
4
separated and dried over sodium sulfate, and the solvent was removed under reduced pressure.
All products were isolated applying flash chromatography.
Synthesis of the sulfonamides (General procedure B). To a solution of the corresponding
aniline (81 or 83) (1.0 eq.) and pyridine (1 M), the appropriate sulfonyl chloride was added
dropwise (1.1 eq). The resulting mixture was stirred at 60°C until TLC revealed complete
consumption of the starting material (3-12 hrs). After cooling to RT, the crude was diluted
with EtOAc, and then washed with aqueous 1M HCl solution. The organic layer was dried
over sodium sulfate, and the solvent was evaporated. All products were isolated applying
flash chromatography.
Suzuki-coupling using pinacol ester boronates at azaindole core (General procedure
C). An oven-dried flask was charged with 86 / R40 or R45 (1.0 eq.), K CO (2.0 eq.) and the
2
3
appropriate aryl bromide (1.5 eq.). A mixture of 1,4-dioxane/H O (2:1, 1 M) was added, and
2
the flask was evaporated and backfilled with argon (3x). Pd(dppf)Cl (6 mol%) was added,
2

and the resulting suspension was heated to 60°C for 0.5 – 3h. After TLC revealed completion
of the reaction, the crude was poured onto a pad of Celite, flushed with EtOAc, and the filtrate
was evaporated to dryness. The residue was resolved in MeOH, K CO (5.0 eq.) was added,
2
3
and the reaction was stirred at RT for 3h, and then neutralized with aqueous 1M HCl solution.
The mixture was extracted with EtOAc, the organic layers were dried over sodium sulfate,
and the solvent was removed under reduced pressure. All compounds were isolated via flash
chromatography.
Methyl 2,6-difluoro-3-nitrobenzoate (75). Oxalyl chloride (1.3 g, 10.4 mmol, 1.05 eq.)
was added to a suspension of 2,6-difluoro-3-nitrobenzoic acid (2.0 g, 9.9 mmol, 1.0 eq.) in
dry DCM (0.5 M), and some drops of DMF were added successively. The mixture was stirred
at RT until the gas formation was complete. An excess of MeOH was added, and the solution
was stirred for 5 minutes. All volatiles were removed under reduced pressure, and the product
1
was obtained as a white solid without further purification. Yield: 2.1 g, 9.7 mmol, 98%. H
NMR (DMSO-d 200 MHz, ppm): δ 8.45 (td, J = 9.0, 5.6 Hz, 1H), 7.52 (td, J = 9.4, 1.8 Hz,
6
,
1
3
1
1
H), 3.95 (s, 3H); C NMR (DMSO-d , 50 Hz, ppm): δ 162.1 (dd, J = 263.5, 5.7 Hz), 159.7,
6
53.6 (dd, J = 271.1, 7.6 Hz), 134.4 (dd, J = 7.5, 4.1 Hz), 130.6 (dd, J = 12.0, 1.5 Hz), 113.4
(dd, J = 23.8, 4.4 Hz), 112.2 (dd, J = 20.3, 18.0 Hz), 53.6.
Methyl 3-amino-2,6-difluorobenzoate (76). To a solution of 75 (2.1 g, 9.7 mmol, 1.0 eq.)
in abs. EtOH (2 M), Pd on charcoal (Pd/C 10%, w/w, 514 mg, 485 µmol, 0.05 eq.) was added.
The suspension was charged with H and the reaction was stirred at RT until TLC revealed
2
,
completion of the reduction. The crude was poured over a pad of Celite, flashed with EtOAc,
and the filtrate was evaporated under reduced pressure to complete dryness. Yield: 1.8 g,
9
.2 mmol, 95%.
Alternatively: 75 (1 g, 4.6 mmol, 1.0 eq.) was suspended in abs. Ethanol (0,25 M) and
0
aqueous HCl solution (1 M, 4.6 ml, 1.0 eq.) and heated to 80°C. Fe (282 mg, 5.1 mmol,
1
.1 eq.) was added to the mixture and stirred at 80°C until complete consumption of the

starting material. The crude was poured through a pad of Celite and the filtrate dried in vacuo.
The product was obtained as a white solid and was used without further purification steps.
Yield: 843 mg, 4.5 mmol, 98%.
2,6-Difluoro-3-(propylsulfonamido)benzoic acid (77). The aniline 76 (1.7 g, 8.8 mmol,
1
.0 eq.) was dissolved in DCM (0.25 M), treated with TEA (2.0 g, 19.4 mmol, 2.2 eq.) and
cooled to 0°C. Propane-1-sulfonyl chloride (2.8 g, 19.4 mmol, 2.2 eq.) was added slowly to
the solution at 0°C. The ice bath was removed, and the reaction was stirred at RT until TLC
revealed completion of the reaction. Water was added, and the mixture was extracted with
EtOAc. The combined organic phases were dried over Na SO . The solvent was evaporated
2
4
under reduced pressure, and the intermediary formed disulfonamide was purified using flash
chromatography (SiO , nHex/EtOAc 9/1, v/v). This intermediate was suspended in
2
THF/MeOH (4/1, v/v, 1 M), and aqueous 2 M NaOH solution (13 ml, 26 mmol, 3.0 eq.) was
added and stirred at RT overnight. The organic solvents were evaporated, and the residue was
acidified with aqueous 1 M HCl solution. The thus formed precipitate was filtered off, washed
with water, and the white solid was dried in vacuum to dryness. Yield: 1.7 g, 6.0 mmol, 68%
1
(
over two steps). H NMR (DMSO-d 200 MHz, ppm): δ 14.01 (s, 1H), 9.74 (s, 1H), 7.54
6
,
(dd, J = 14.8, 8.7 Hz, 1H), 7.20 (t, J = 9.2 Hz, 1H), 3.15 – 3.02 (m, 2H), 1.85 – 1.63 (m, 2H),
1
3
0
.97 (t, J = 7.4 Hz, 3H); C NMR (DMSO-d , 50 Hz, ppm): δ 161.8, 157.3 (dd, J = 174.8, 6.9
6
Hz), 152.3 (dd, J = 178.1, 6.9 Hz), 129.8 (dd, J = 10.2, 2.2 Hz), 122.0 (dd, J = 13.5, 3.8 Hz),
-
1
12.8 (dd, J = 21.3, 19.3 Hz), 112.3 (dd, J = 22.6, 4.1 Hz), 53.8, 16.9, 12.6. [M-H] : 278,0.
N-(3-(5-bromo-1H-pyrrolo[2,3-b]pyridine-3-carbonyl)-2,4-difluorophenyl)propane-1-
sulfonamide (78). An oven-dried flask was charged with 77 (15.6 g, 55.8 mmol, 1.1 eq),
suspended in MeNO (first half of total volume, 0.2 M) and treated with oxalyl chloride
2
(
7.1 g, 55.8 mmol, 1.1 eq.). Some drops of DMF were added to initiate the activation. A
second oven dried flask was charged with 5-bromo-1H-pyrrolo[2,3-b]pyridine (10.0 g,
0.8 mmol, 1.0 eq), AlCl (40.6 g, 305 mmol, 6.0 eq.) and MeNO (second half of total
5
3
2

volume, 0.2 M). This suspension was stirred for at least 30 minutes. Upon completion of gas
formation of the activation step, the fully activated carboxylic acid solution was added to the
second flask. The resulting mixture was stirred at 50°C overnight. After cooling the crude to
0
°C, MeOH was slowly added. Finally, the suspension was diluted with water, and the
resulting precipitate was filtered off. The solids were suspended in EtOAc, sonicated for 5
minutes and filtered off again to obtain the product which was dried in vacuum. Yield: 19.9 g,
1
4
3.3 mmol, 85%. H NMR (DMSO-d 200 MHz, ppm): δ 13.14 (s, 1H), 9.78 (s, 1H), 8.59 (d,
6
,
J = 1.8 Hz, 1H), 8.51 (d, J = 2.0 Hz, 1H), 8.28 (s, 1H), 7.59 (td, J = 9.0, 6.4 Hz, 1H), 7.28 (t, J
1
3
=
8.8 Hz, 1H), 3.19 – 3.06 (m, 2H), 1.86 – 1.62 (m, 2H), 0.96 (t, J = 7.3 Hz, 3H).; C NMR
(
DMSO-d , 50 Hz, ppm): δ 180.6, 156.6 (dd, J = 184.1, 7.6 Hz), 151.7 (dd, J = 187.1, 7.7
6
Hz), 147.8, 145.3, 139.3, 131.1, 128.9 (dd, J = 10.1, 2.1 Hz), 122.0 (dd, J = 13.6, 3.8 Hz),
1
1
19.0, 117.8 (dd, J = 24.3, 22.1 Hz), 114.9, 114.3, 112.4 (dd, J = 22.8, 3.8 Hz), 53.5, 16.8,
-
2.6. [M-H] : 456,1.
5
-(4-Chlorophenyl)-1H-pyrrolo[2,3-b]pyridine (79). A dry flask was charged with 5-
bromo-1H-pyrrolo[2,3-b]pyridine (7.0 g, 35.5 mmol, 1.0 eq.), K CO (9.8 g, 71.1 mmol,
2
3
2
.0 eq.) and (4-chlorophenyl)boronic acid (6.1 g, 39.1 mmol, 1.1 eq.). The solids were
suspended in MeCN/H O (4/1, v/v, 36 ml), and the vessel was evacuated and backfilled with
2
argon (3x). Pd(PPh ) (410 mg, 355 µmol, 1 mol%) was added, and the resulting mixture was
3
4
heated to reflux overnight. After cooling the suspension to RT, the solids were filtered off and
washed successively with water and cold MeCN. The obtained solids were dried in vacuum.
1
Yield: 7.1 g, 31.0 mmol, 87%. H NMR (DMSO-d 400 MHz, ppm): δ 11.76 (s, 1H), 8.51 (d,
6
,
J = 2.1 Hz, 1H), 8.20 (d, J = 1.9 Hz, 1H), 7.72 (d, J = 8.5 Hz, 2H), 7.57 – 7.43 (m, 3H), 6.50
13
(
dd, J = 3.2, 1.7 Hz, 1H).; C NMR (DMSO-d , 101 Hz, ppm): δ 148.2, 141.4, 138.0, 131.7,
6
-
1
28.9, 128.6, 127.1, 126.9, 126.1, 119.7, 100.2. [M-H] : 227.0.
(5-(4-Chlorophenyl)-1H-pyrrolo[2,3-b]pyridin-3-yl)(2,6-difluoro-3-
nitrophenyl)methanone (80). An oven-dried flask was charged with 2,6-difluoro-3-

nitrobenzoic acid (977 mg, 4.8 mmol, 1.1 eq), suspended in MeNO (first half of total volume,
2
0
.2 M) and treated with oxalyl chloride (658 mg, 4.8 mmol, 1.1 eq.). Some drops of DMF
were added to initiate the activation. A second oven-dried flask was charged with 81 (1.0 g,
.4 mmol, 1.0 eq.), AlCl (3.5 g, 26.4 mmol, 6.0 eq.) and MeNO (second half of total
4
3
2
volume, 0.2 M). This suspension was stirred for at least 30 minutes. Upon completion of gas
formation of the activation step, the fully activated carboxylic acid solution was added to the
second flask. The resulting mixture was stirred at 50°C overnight. After cooling the crude to
0
°C, MeOH was slowly added. Finally, the suspension was diluted with water, and the
resulting precipitate was filtered off. The solids were suspended in EtOAc, sonicated for 5
minutes and filtered off again to obtain the product which was dried in vacuum. Yield: 1.2 g,
1
2
8
.9 mmol, 67%. H NMR (DMSO-d 400 MHz, ppm): δ 13.13 (s, 1H), 8.75 – 8.68 (m, 2H),
6
,
1
3
.50 – 8.38 (m, 2H), 7.80 (d, J = 8.5 Hz, 2H), 7.59 – 7.49 (m, 3H). C NMR (DMSO-d , 101
6
Hz, ppm): δ 178.8, 161.5 (dd, J = 257.1, 7.5 Hz), 152.7 (dd, J = 264.2, 9.4 Hz), 149.1, 144.1,
1
39.8, 136.9, 134.3 (dd, J = 7.6, 3.6 Hz), 132.5, 130.4, 129.0, 128.9, 127.2, 119.5 (dd, J =
-
2
5.1, 23.4 Hz), 117.4, 115.4, 113.2 (dd, J = 24.1, 3.7 Hz). [M-H] : 411.7.
(
3-Amino-2,6-difluorophenyl)(5-(4-chlorophenyl)-1H-pyrrolo[2,3-b]pyridin-3-
yl)methanone (81). A solution of 80 (1.2 g, 2.8 mmol, 1.0 eq.) in EtOAc (0.1 M) and THF
0.1 M) was heated to 60°C. SnCl dihydrate (2.2 g, 9.8 mmol, 3.5 eq.) was added portion
(
2
wise, and the resulting mixture was stirred at 60°C until TLC revealed completion of the
reaction. The mixture was cooled to RT, and a half-saturated, aqueous sodium bicarbonate
solution was added. The suspension was filtered, and the filtrate was washed with brine and
water. The organic phase was dried over Na SO and the solvent was removed under reduced
2
4,
1
pressure. Yield: 970 mg, 2.5 mmol, 89%. H NMR (DMSO-d 400 MHz, ppm): δ 12.95 (s,
6
,
1
8
1
H), 8.70 (d, J = 1.9 Hz, 1H), 8.64 (s, 1H), 8.12 (s, 1H), 7.78 (d, J = 8.4 Hz, 2H), 7.55 (d, J =
1
3
.4 Hz, 2H), 7.00 – 6.87 (m, 2H), 5.24 (s, 2H). C NMR (DMSO-d , 101 Hz, ppm): δ 182.2,
6
49.2 (dd, J = 235.1, 6.7 Hz), 148.9, 146.0 (dd, J = 241.2, 8.0 Hz), 143.7, 138.0, 137.1, 133.4

(dd, J = 13.1, 2.3 Hz), 132.4, 130.1, 129.0, 128.8, 127.0, 117.5, 116.6 (dd, J = 7.9, 6.2 Hz),
-
1
15.9, 111.3 (dd, J = 22.1, 3.3 Hz). [M-H] : 381.9.
(5-Bromo-1H-pyrrolo[2,3-b]pyridin-3-yl)(2-fluoro-3-nitrophenyl)methanone (82). An
oven-dried flask was charged with 2-fluoro-3-nitrobenzoic acid (3.1 g, 16.8 mmol, 1.1 eq),
suspended in DCM (0.5 M) and treated with oxalyl chloride (2.1 g, 16.8 mmol, 1.1 eq.). Some
drops of DMF were added to initiate the activation. A second oven-dried flask was charged
with 5-bromo-1H-pyrrolo[2,3-b]pyridine (3.0 g, 15.2 mmol, 1.0 eq.), AlCl (10.0 g, 76 mmol,
3
5
.0 eq.) and DCM (0.5 M). This suspension was stirred for at least 30 minutes. Upon
completion of gas formation of the activation step, the fully activated carboxylic acid solution
was added to the second flask. The resulting mixture was stirred at RT overnight. After
cooling the crude to 0°C, MeOH was slowly added. Brine was added, and the mixture was
extracted with EtOAc. The organic phases were dried over sodium sulfate, and the solvent
was evaporated. The residue was suspended in EtOAc, sonicated for 5 minutes, and the solids
1
were filtered off to obtain the product. Yield: 4.7 g, 12.8 mmol, 84%. H NMR (DMSO-d6
,
4
7
1
1
00 MHz, ppm): δ 13.10 (s, 1H), 8.64 (s, 1H), 8.49 (s, 1H), 8.37 – 8.12 (m, 2H), 7.99 (s, 1H),
1
3
.62 – 7.48 (m, 1H). C NMR (DMSO-d , 101 Hz, ppm): δ 183.9, 151.7 (d, J = 264.4 Hz),
6
47.7, 145.1, 139.4, 137.5 (d, J = 8.2 Hz), 135.3 (d, J = 4.2 Hz), 131.3, 130.3 (d, J = 16.2 Hz),
-
27.8 (d, J = 2.0 Hz), 125.2 (d, J = 4.7 Hz), 119.4, 114.1, 114.0. [M-H] : 363.0.
(
3-Amino-2-fluorophenyl)(5-bromo-1H-pyrrolo[2,3-b]pyridin-3-yl)methanone (83). A
solution of 82 (4.7 g, 12.8 mmol, 1.0 eq.) in EtOAc (0.1 M) and THF (0.1 M) was heated to
0°C. SnCl dihydrate (10.1 g, 44.8 mmol, 3.5 eq.) was added portion wise, and the resulting
6
2
mixture was stirred at 60°C until TLC revealed completion of the reaction. The mixture was
cooled to RT, and a half-saturated, aqueous sodium bicarbonate solution was added. The
suspension was filtered, and the filtrate was washed with brine and water. The organic phase
was dried over Na SO and the solvent was removed under reduced pressure. Yield: 3.3 g,
2
4,
1
9
.7 mmol, 76%. H NMR (DMSO-d 200 MHz, ppm): δ 12.90 (s, 1H), 8.58 (d, J = 2.2 Hz,
6,