The impact of binding site waters on the activity/selectivity trade-off of Janus kinase 2 (JAK2) inhibitors

Article abstract of DOI:10.1016/j.bmc.2019.02.029

Structure based optimization of B39, an indazole-based low micromolar JAK2 virtual screening hit is reported. Analysing the effect of certain modifications on the activity and selectivity of the analogues suggested that these parameters are influenced by

Full text of DOI:10.1016/j.bmc.2019.02.029

Accepted Manuscript
The impact of binding site waters on the activity/selectivity trade-off of Janus
kinase 2 (JAK2) inhibitors
Attila Egyed, Dávid Bajusz, György M. Keserű
PII:
S0968-0896(19)30040-9
https://doi.org/10.1016/j.bmc.2019.02.029
BMC 14765
DOI:
Reference:
Bioorganic & Medicinal Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
7 January 2019
13 February 2019
15 February 2019
Please cite this article as: Egyed, A., Bajusz, D., Keserű, G.M., The impact of binding site waters on the activity/
selectivity trade-off of Janus kinase 2 (JAK2) inhibitors, Bioorganic & Medicinal Chemistry (2019), doi: https://
doi.org/10.1016/j.bmc.2019.02.029
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. 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.

The impact of binding site waters on the activity/selectivity trade-off of Janus
kinase 2 (JAK2) inhibitors
Attila Egyed, Dávid Bajusz, György M. Keserű*
Medicinal Chemistry Research Group, Research Centre for Natural Sciences, Hungarian
Academy of Sciences, Magyar tudósok krt. 2, 1117 Budapest, Hungary
Abstract
Structure based optimization of B39, an indazole-based low micromolar JAK2 virtual
screening hit is reported. Analysing the effect of certain modifications on the activity and
selectivity of the analogues suggested that these parameters are influenced by water molecules
available in the binding site. Simulation of water networks in combination with docking
enabled us to identify the key waters and to optimize our primary hit into a low nanomolar
JAK2 lead with promising selectivity over JAK1.
Keywords
indazole; janus kinase JAK2; inhibitor; ATP site; activity; selectivity; water network
Introduction
Janus kinases (JAKs) are non-receptor tyrosine kinases, consisting of four subtypes
JAK1, JAK2, JAK3, TYK2.1 They have important roles in the JAK/STAT signalling pathway
in mammalian cells through gene transcription,2,3 as this pathway regulates inflammatory
processes and immune responses. Dysregulation of JAK/STAT can contribute to many
disease conditions, including myelofibrosis (MF), essential thrombocythaemia (ET), and
polycythaemia vera (PV) that are considered as JAK2 related malignancies.4 Currently three
basically non-selective drugs have been approved by the FDA: ruxolitinib (1), tofacitinib (2),
and oclaticinib (3) (Figure 1).5–7 Ruxolitinib (1) was the first, dual JAK1/JAK2 inhibitor
globally approved drug in 2011 for the treatment of myelofibrosis and more recently of
polycythemia vera. One year later, it was followed by tofacitinib (2), a pan-JAK inhibitor
marketed for the treatment of rheumatoid arthritis8,9 in the US, however, it is not approved in

the EU.10 Oclacitinib, which shows preference inhibiting JAK1-dependent cytokine
signalling, was released in 2013 and is used in veterinary medicine to treat dogs with allergic
skin diseases.6
Selectivity limitations of approved JAK inhibitors can be traced back to the high
sequence homology and structural similarity of the ATP binding sites of the isoforms (and
likewise, of all kinases).11 Selective inhibitors can help mitigate side effects and better
understand the function of JAKs in relevant indications. More recently, however, it became
possible to achieve up to hundred fold selectivity within the family.12,13 Optimization of
JAK1/JAK2 selectivity, however, is still challenging. Indeed, the ATP pocket of these
isoforms differs in only a few residues: of those, only Gln853 in JAK2 (with Arg879 being its
counterpart in JAK1) and Asp939 (with Glu966 as its counterpart) have the right orientations
toward the binding site to enable direct targeting with a small molecule.1,14 Due to the
mentioned residues being located on the more solvent-exposed parts of the JAK2 ATP pocket,
compounds with improved JAK1/JAK2 selectivity are generally larger than non-selective
inhibitors, as increasing the bulk in the hinge region allows influencing contacts with the key
selectivity residues. In fact, the most selective inhibitor disrupts the water-mediated hydrogen
bond between Arg879 and the hinge Pro960 in JAK1 and has a disfavoured charge interaction
with Arg879.13 This finding suggests that in addition to the shape and composition of the
binding site, water molecules and their network might play a role in the isoform selectivity.
The role of binding site waters in kinases is also highlighted in the KLIFS database, an online
resource of experimental kinase structures with detailed structural information.15,16 New
alternatives for JAK2 inhibition (including type II binders and pseudokinase inhibitors) have
been recently reviewed by Leroy and Constantinescu.17 However, the biochemical toolbox
required for a detailed SAR study on these alternatives is not yet fully developed/accessible.
Recently we have identified the indazole-based compound B39 (4) as a promising
candidate for future optimization towards JAK2 activity and JAK2 vs. JAK1 selectivity
(Figure 1), as it already displays a 14-fold selectivity and low micromolar activity.18 In this
work, we aimed to obtain compounds with improved activity and/or selectivity starting from
B39 (4). In addition, we rationalize the effects of various substitutions on the activity and
selectivity of the compounds, taking into account the water molecules in the binding site. For
a group of compounds, a very precise dissection of the thermodynamic effects of certain
substitutions was possible. Thus, besides reporting this lead series of JAK2 inhibitors, we

provide simple and practical guidelines for the interpretation of the SAR space available at the
ATP site.
CN
N
N
N
N
N
CN
O
N
N
N
N
N
H
H
Ruxolitinib (1)
JAK1 IC50 3.3 nM
JAK2 IC50 2.8 nM
Tofacitinib (2)
JAK1 IC50 3.2 nM
JAK2 IC50 4.1 nM
O
S
O
H
N
O
N
HN
N
N
H
N
N
N
N
N
H
N
N
Oclacitinib (3)
B39 (4)
JAK1 IC50 10 nM
JAK2 IC50 18 nM
JAK1 IC50 33000 nM
JAK2 IC50 2380 nM
Figure 1. Structures and in vitro JAK1 and JAK2 inhibitory activities of the three marketed
JAK inhibitors and our hit compound, B39.
Result and discussion
Synthesis of new JAK2 inhibitors with the indazole and 3-aminoindazole cores
We have identified the compound 4 (B39) in our earlier work as a result of a virtual
screening campaign,19 which also involved the prediction of JAK2 vs. JAK1 selectivity.18
This compound can be divided into 3 parts: the hinge binder (green), the linker (black) and the
selectivity-pocket moieties (orange) (see Figure 2). During the optimization of B39 (4), we
have examined the effects of substitution on each of these major parts on the overall activity
and JAK2 vs. JAK1 selectivity.

Figure 2. The hinge binder (green), the linker (black/white) and the selectivity-pocket
moieties (orange) of the hit compound 4 (A) and predicted binding mode inside the JAK2
ATP-pocket (B).
First, we explored the SAR at the selectivity pocket by varying the position of the
tetrazole ring and its methylene analogue. The meta-substituted derivative (5) displayed a 3.5-
fold activity improvement, but the selectivity was completely lost. The methylene linker, both
in the meta (6) and ortho (7) positions led to significant activity and selectivity loss that
suggested to keep the tetrazole in the ortho position to preserve selectivity.
Table 1 JAK kinase inhibitory activity of B39 analogues with different selectivity-pocket moieties
H
R1
N
N
O
N
H
Compound
R1
JAK1 IC50 (μM) JAK2 IC50 (μM) Selectivity
N
N
N
4 (B39)
33
2.38
14
1
N
N
N
N
N
5
6
0.764
0.718
N
N
N
N
20% (20 μM)
15.2
> 1.3

N
N
N
N
7
9.67
3.54
2.7
Next we investigated the role of the linker by replacing the original amide with
heteroaromatic rings as suggested by literature analogs20,21 and docking results (see
Supplementary information). Keeping the tetrazole at the ortho position we synthesized the
corresponding thiazole, isoxazole and triazole analogues (Scheme 1).
N
N
N
N
NH2
R1
S
N
O
N
15a R1: H
15b R1: Cl
N
N
R1
N
N
Br
S
j
N
H
14
N
16a R1: H
16b R1: Cl
N
H
g,h,i
Br
c
N
N
N
N
H
H
8
11
N
CHO
NO2
N
N
N
a,b
d, e, f
12
9
N
N
N
N
N
N
N
N
N
N
N
N
O
N
N
10
13
N
H
N
H
Scheme 1. Reagents: (a) EtOH, CuI, sodium ascorbate, N,N’-dimethyl-ethylenediamin, NaN3 (b) DMSO, CuSO4, K2CO3, L-proline (c) 1.
CuI, PdCl2(PPh3)2, Et3N, THF, ethynyl-trimethylsilane 2. TBAF, THF (d) NH2OH.HCl, NaOH, t-BuOH, CuI, chloramine-T (e) Na2S, EtOH
(f) CH3COOH, NaN3, CH(OCH3)3 (g) CH2Cl2, Et3N, DMAP, Boc2O (h) PdCl2(PPh3)2, toluene, tributyl(1-ethoxyvinyl)tin (j) 1. PyrBr3, THF
2. TFA, DCM (j) EtOH.
Starting from 5-bromindazole (8) the heterocyclic linkers were formed in the
corresponding coupling reaction. The marginal increase in the activity of the thiazole
derivative (16a) was accompanied by a significant reduction of selectivity (see Table 2).

Interestingly by introducing a chlorine substituent to position 2 of the phenyl ring in the
selectivity pocket (16b), we achieved selectivity similar to that of B39 (4). Inspired by the
work of Hirst et al.,22 next we modified the hinge binder to the 3-aminoindazol core (17),
however, this modification did not improve the activity as expected. The incorporation of
other heterocycles, such as isoxazole or triazole resulted in a significant drop in selectivity
and/or activity. Summarizing these results, we concluded in spite of promising literature
examples, heterocyclic linkers are less suited for this particular type of JAK2 scaffold.
Table 2 JAK kinase inhibitory activity of B39 analogies with heterocyclic linker
N
N
N
N
R1
HetCyc
R2
N
N
H
Compound
R1
R2
HetCyc
JAK1 IC50 (μM)
JAK2 IC50 (μM)
Selectivity
10
H
H
8.8
9.53
0.9
N
N
N
13
16a
16b
17
H
H
H
H
3.3
2
1.65
5
N
O
S
7.94
21.6
9.29
1.54
1.68
3.25
N
S
H
Cl
H
12.85
2.85
N
S
NH2
N
Having the positive effect of the 2-halogen substitution we examined this option using
the amide linker and both the original indazole, and the 3-aminoindazole cores. The synthesis
of these compounds was started from the appropriate anthranilic acid (18a, 18b) that was first
equipped by the tetrazole ring (19a, 19b). The carboxylic acid moiety was next converted to
acid chloride that was coupled with the corresponding amine (20, 22). This reaction gave

directly B39 (4) and its 2-chloro analogue (21). The 3-aminoindazole derivatives (24a, 24b)
were obtained from the ring closure reaction of the corresponding amides (23a, 23b).
H
N
H
N
R1
O
N
N
N
H2N
H
20
R1
N
N
4 R1: H
21 R1: Cl
N
COOH
NH2
N
COOH
R1
a
N
b
H
N
F
N
R1
O
R1
O
N
N
N
c
N
CN
N
19a R1: H
19b R1: Cl
18a R1: H
18b R1: Cl
H
H
F
NH2
N
N
N
N
N
N
H2N
CN
N
N
22
23a R1: H
23b R1: Cl
24a R1: H
24b R1: Cl
Scheme 2. Reagents: (a) CH3COOH, NaN3, CH(OCH3)3 (b) 1. SOCl2 2. Et3N, DCM (c) N2H4.H2O23,24
Similar to that observed for the thiazole linker, we found that the introduction of the 2-
chloro substituent at the phenyl ring (21) resulted in a three-fold improvement in the activity,
while maintaining the selectivity of B39 (4). Since the binding modes of 4 and 21 are almost
identical (see Figure 2B and the Supplementary material), we hypothesized that the activity
gain could be attributed to the displacement of a binding site water molecule by the 2-chloro
substituent of 21.
Unfortunately, however, changing the indazole core to 3-aminoindazole (24a, 24b)
failed to improve the activity of the original starting point (4).
Table 3 JAK kinase inhibitory activity of substituted B39 analogues
N
N
N
N
R1
H
N
N
R2
O
N
H
Compound
R1
R2
JAK1 IC50 (μM)
JAK2 IC50 (μM)
Selectivity
4 (B39)
21
H
H
Cl
H
33.0
11.5
2.38
0.82
6.34
76.6
14
14
H
24a
NH2
NH2
33.9
5.3
NA
24b
Cl
10 % (10μM)

Figure 3. A) Overlaid structures of JAK2 with fragment-sized inhibitors in the ATP
binding pocket (green: indazole-based fragment in PDB structure 3E6222, magenta: diaryl-
amide fragment in PDB structure 4GFM25). B) Merging the overlaid experimental binding
poses results in a reference binding mode for our compounds with the merged core.
Given the unexpected effect of the 3-amino group at the hinge binding indazole core
(24a and 24b) we decided to analyse the corresponding binding modes in comparison to the
available crystal structures of 3-aminoindazole-based compounds in the binding pocket of
JAK2.22 In particular, with the overlay of PDB structures 3E62 and 4GFM, the two co-
crystallized ligands can be merged to produce the common core structure of our compounds
(Figure 3). We have used the merged structure presented in Figure 3B as a reference binding
mode when evaluating the predicted binding modes of our compounds.
Figure 4. Predicted binding poses of compounds 24a (A) and 24b (B), overlaid on the
reference binding mode based on PDB entries 3E62 and 4GFM (see Figure 3B).

The expected enhancement associated with the 3-amino group on the indazole core is
counteracted by the much greater ligand strain of these compounds (as compared to simpler
analogues 28a-b and 30a-e), particularly with the suboptimal orientation of the planes of the
amide bond (approx. 90° or 180° flipped) and the phenyl ring (perpendicular to the reference
position).
Comparing the interactions of compounds 24a and 24b to the reference binding mode
revealed that the 3-amino group of these compounds is incompatible with a large substituent
on the phenyl ring (Figure 4). We hypothesized that the strained position of the linker and
phenyltetrazole moiety of 24a and 24b counteract the beneficial effect of the amine group
added to the indazole core. Testing this hypothesis we replaced the large tetrazole group with
a smaller ethyl substituent and synthesized compounds 25 and 26. The synthesis of these
molecules was analogous to 24a and 24b using the N-ethylated analogue of 20 (Scheme 2).
Table 4 JAK kinase inhibitory activity of B39 analogues with N-Et moiety
NH
R1
N
H
N
O
N
H
Compound
R1
JAK1 IC50 (μM) JAK2 IC50 (μM) Selectivity
25
26
H
24.1
22
3.96
1.97
6
NH2
11
Although the activity and the selectivity of 25 was slightly inferior to that of 4, the
corresponding 3-aminoindazole (26) showed better activity and selectivity. This result
suggested that by decreasing the size of the substituents in the selectivity pocket we can
realize the beneficial effect of the 3-amino group at the hinge.

R3
H2N
H
N
N
R2
N
N
H
R1
O
20
R3
N
H
28a R1,R2,R3: H
28b R1:Cl, R2,R3: H
a
R2
COOH
R3
R1
R3
27a R1,R2,R3: H
NH2
N
b
H
27b R1:Cl, R2,R3: H
27c R1,R3: H R2: Cl
27d R1,R2: Cl R3: H
27e R1, R3: Cl, R2: H
H2N
CN
F
N
H
N
R2
CN
R2
R1
O
N
H
R1
O
22
F
30a R1,R2,R3: H
29a R1,R2,R3: H
30b R1:Cl, R2,R3: H
30c R1,R3: H R2: Cl
30d R1,R2: Cl R3: H
30e R1, R3: Cl, R2: H
29b R1:Cl, R2,R3: H
29c R1,R3: H R2: Cl
29d R1,R2: Cl R3: H
29e R1, R3: Cl, R2: H
Scheme 3. Reagents: (a) HATU, DIPEA, DCM (b) N2H4.H2O
To investigate the effect of substituents in detail we performed a haloscan around the
phenyl ring of 4. The corresponding halogen derivatives (28a-b and 30a-e) were synthesized
from the appropriate benzoic acids (27a-e) as described in Scheme 3.
Table 5 JAK kinase inhibitory activity the halogenated derivatives of B39
R3
R4
N
H
N
R2
R1
O
N
H
JAK1 IC50
JAK2 IC50
Compound
R1
H
R2
H
R3
H
R4
H
Selectivity
(μM)
(μM)
17 % (20
28a
13.8
μM)
28b
30a
30b
30c
30d
30e
Cl
H
H
H
H
H
H
H
H
Cl
H
8.53
10.5
0.453
2.78
18.8
3.78
18
NH2
NH2
NH2
NH2
NH2
Cl
H
H
1.82
0.101
0.583
0.411
0.023
Cl
Cl
H
0.943
1.510
0.151
1.63
3.67
6.57
Cl
Cl
Analysing the results we found that the tetrazole of B39 can be replaced by a chlorine
(28b) that improved both the activity and the selectivity. Furthermore, similar to the N-ethyl
analogue (26), the 3-aminoindazole core improved the activity of the 2-chloro derivative

significantly (30b). (The binding modes predicted for these compounds are perfectly in line
with the proposed reference binding mode, see Supplementary material.) Interestingly,
moving the chlorine to the meta position (30c) decreased the activity and diminished the
selectivity. Combining substitutions at positions 2 and 3 (30d) some selectivity was regained,
while the 2,5-dichloro analogue (30e) displayed a twice better selectivity and a significantly
increased, low nanomolar activity (23nM).
R2
R3
R2
R2
O
O
NH2
R3
R3
NC
F
O
O
NC
F
R1
c
a
NH2
N
OH
CN
F
N
H
N
H
b
R1
R1
N
33a R1,R2,R3: H
31
32
H
33b R1: Cl R2,R3: H
33c R1,R2: Cl, R1: H
33d R1,R3: Cl, R2: H
34a R1,R2,R3: H
34b R1: Cl R2,R3: H
34c R1,R2: Cl, R1: H
34d R1,R3: Cl, R2: H
35a R1,R2,R3: H
35b R1: Cl R2,R3: H
35c R1,R2: Cl, R1: H
35d R1,R3: Cl, R2: H
Scheme 4. Reagents: (a) NaH2PO4, NaClO2, H2O, tBuOH 26(b) DCM, SOCl2 (c) N2H4.H2O
After the haloscan, compounds with flipped amid linker were synthetized (Scheme 4). The
activity and selectivity of compounds 35a-c were almost the same with the original amides
(30a, 30b, 30e) (Table 6). The predicted binding modes of compounds with the amide and
flipped amide linkers were found identical (see Supplementary information). From this series
we identified compound 35d having low nanomolar activity and 16 fold selectivity.
Table 6 JAK kinase inhibitory activity flipped amine derivatives of B39
R2
R3
O
NH2
N
H
N
R1
N
H
JAK1 IC50
JAK2 IC50
Compound
R1
R2
R3
Selectivity
(μM)
(μM)
35a
35b
35c
35d
H
H
H
H
H
10.8
2.27
3.23
0.126
0.038
0.028
3.78
18
Cl
Cl
Cl
Cl
H
H
0.414
0.460
11
Cl
16

Computational analysis of the binding affinities and the role of binding site waters
Interpreting the activity and selectivity results, we have carried out a detailed binding
thermodynamic analysis of the most important compounds in the series. As a first step, we
have converted the experimentally determined IC50 values, first to Ki values (using the Cheng-
Prusoff formula27), then to binding Gibbs free energies, and determined the margin of error
for these values to be 0.65 kcal/mol. Table 7 summarizes the binding free energy values with
JAK1 and JAK2, while the details of the calculations can be found in the Supplementary
material.
Table 7. Free energies of binding on JAK1 and JAK2
R3
R2
R1
R4
L
R5
N
N
H
ΔGJAK1
ΔGJAK2
Compound
R1
R2
R3
R4
R5
L
Selectivity
(kcal/mol) (kcal/mol)
28a
28b
30a
30b
30c
30d
30e
25
H
H
Cl
H
H
H
H
H
Cl
Cl
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Cl
H
H
H
H
H
H
H
H
Cl
H
H
H
H
H
H
H
H
-CO-NH-
-CO-NH-
-CO-NH-
-CO-NH-
-CO-NH-
-CO-NH-
-CO-NH-
-CO-NH-
-CO-NH-
-CO-NH-
-CO-NH-
-CO-NH-
-CO-NH-
-NH-CO-
-NH-CO-
-NH-CO-
-NH-CO-
> -6.79
-7.29
-7.17
-8.21
-8.59
-8.32
-9.67
-6.68
-6.74
-6.50
-7.12
-6.48
>-7.20
-7.15
-8.08
-9.08
-9.02
-7.01
-9.03
-7.96
-9.91
-8.88
-9.08
-10.78
-7.75
-8.16
-8.05
-8.68
-7.47
-6.00
-7.87
-9.78
-10.49
-10.67
N/A
18.8
3.78
18
H
NH2
NH2
NH2
NH2
NH2
H
Cl
H
1.63
3.67
6.57
8
Cl
Cl
H
N-ethyl
N-ethyl
tetrazole
tetrazole
tetrazole
tetrazole
H
26
NH2
H
H
11
4
H
14
21
H
Cl
H
14
24a
24b
35a
35b
35c
35d
NH2
NH2
NH2
NH2
NH2
NH2
5.3
N/A
3.78
18
Cl
H
Cl
Cl
Cl
H
H
11
Cl
16

Looking at the JAK2 binding free energies in Table 7, one can notice the first four
compounds (28a, 28b, 30a and 30b), whose free energy values can be decomposed linearly,
as follows: addition of an amine group to the indazole core confers a free energy gain of
approx. 0.9 kcal/mol (28a to 30a, and 28b to 30b), while addition of a chlorine to the ortho
position of the phenyl ring results in ~ 2.0 kcal/mol free energy gain (28a to 28b, and 30a to
30b). Rationalising these effects in the light of the proposed contribution of binding site water
molecules, we have predicted the positions of water molecules in the JAK2 binding site
(PDB: 3E64) with WaterFlap (see Experimental section). This analysis highlighted two
important water molecules: a thermodynamically favourable (“happy”) water close to the
hinge region and an unfavourable (“unhappy”) water close to β strands I and II, their
predicted binding free energies being -0.65 and +3.34 kcal/mol, respectively (see Figure 5 and
the Supplementary material). Based on the overlay with the predicted binding modes, the
former is replaced by the 3-amino group of the indazole core and the latter is replaced by the
chloro substituent on the phenyl group.
Figure 5. Binding mode of compound 30b in the JAK2 pocket, predicted with Glide based on
the reference binding mode in Figure 3B. Two important binding site waters are overlaid,
based on the water network prediction by WaterFlap (with the ΔG values indicated).

The gain in the free energy associated to the displacement of these waters can be fully
explained if we also account for the other aspects of binding. The introduction of the amine
group to the indazole core results in an additional H-bond with the hinge backbone. Literature
sources evaluate the free energy gain associated with NH···O=C hydrogen bonds typically in
the range of 1-1.5 kcal/mol.28,29 This is partially counterbalanced by removing an
energetically favourable water molecule from the binding site (0.65 kcal/mol). The difference
of the two contributions gives the experimentally observed 0.9 kcal/mol difference within the
margin of error.
For the chloro substituent on the phenyl ring, the predicted free energy gain associated
with displacing the energetically unfavourable water molecule is 3.34 kcal/mol. In addition,
the van der Waals interaction between the chlorine and the binding pocket should account for
further free energy gain. This is compensated by the slightly higher desolvation energy of 28b
compared to 28a (0.8 kcal/mol difference, as evaluated by QM calculations with Jaguar).30
The difference of these effects (> 2.54 kcal/mol) is closely within, or around the margin of
error from the experimental difference of 2.0 kcal/mol. Compounds 35a and 35b (the flipped-
amide analogues of 30a and 30b) display the same trends, with almost identical binding free
energies on the two enzymes (within the margin of error).
If we move on to the compounds with bigger substituents (N-ethyl, tetrazole), the
addition of the amine group becomes less favourable, due to the increased ligand strain in the
binding poses. While for the N-ethyl bearing 26, the free energy contribution of the 3-amino
group is reduced to 0.4 kcal/mol (compared to 25), for the tetrazole-bearing 4 and 21, the
contribution becomes mildly (+0.6 kcal/mol) and strongly (+2.7 kcal/mol) unfavourable.
Similarly, addition of the chlorine atom to 4 confers only a 0.6kcal/mol free energy gain, and
deteriorates the activity in case of 24a. The same is observed for JAK1, where the addition of
the amine group does not result in a gain of free energy, except for compounds 30a and 30b
with roughly 1.0 kcal/mol.
Compounds 30c and 30d underline the importance of the position of the chloro
substituent with respect to the binding site water molecule: neither of them presents as large
an enhancement over 30a as 30b does. By comparison, the three molecules (30b, c and d)
display almost identical binding free energies on JAK1, which lacks the unfavourable water
molecule in the same position. On the other hand, the addition of a second chlorine atom to

the opposite side of the phenyl ring confers a free energy gain of 1-1.5 kcal/mol on both
enzymes, as demonstrated by compounds 30e, 35c and 35d. As such, we propose that this
effect is independent of the water network of the binding site, and is rather caused by the free
energy gain associated with the burial of a more hydrophobic compound in the binding cavity.
Conclusion
In our previous work, we identified compound 4 (B39), a JAK2 inhibitor by virtual
screening.19 This compound showed 14-fold selectivity over JAK1 and a low micromolar
activity against JAK2. In this study, we investigated the preliminary SAR around 4 that
revealed the importance of structural waters and their networks that influence both the activity
and selectivity of B39 derivatives. Carefully optimizing the substituents at the hinge binding
region and in the selectivity pocket we identified several compounds with improved affinity
and/or JAK2/JAK1 selectivity. Three of them are low nanomolar inhibitors of JAK2, with
compound 35d displaying reasonable selectivity, as well. The analysis of water networks
suggests that binding site waters play a crucial role in the activity-selectivity trade-off of this
set of JAK2 inhibitors. In fact, we were able to dissect the thermodynamic effects of some
substituents, accounting for various terms, including the displacement of binding site water
molecules.
Acknowledgement
This work was supported by the National Research, Development and Innovation
Office of Hungary (OTKA, grant no. K 116904).
Experimental
NMR spectra were recorded in DMSO-d6 or CDCl3 solution at room temperature, on a
1
13
Varian Unity Inova 500 spectrometer (500 MHz for H NMR and 125 MHz for C spectra)
or Varian Unity Inova 300 spectrometer (300 MHz for 1H NMR and 75 MHz for 13C spectra),
with the deuterium signal of the solvent as the lock and TMS as the internal standard.
Chemical shifts (δ) and coupling constants (J) are given in ppm and Hz, respectively.

HPLC-MS measurements were performed using a Shimadzu LCMS-2020 device
equipped with a Reprospher 100 C18 (5 µm; 100x3mm) column and positive-negative double
ion source (DUIS) with a quadrupole MS analyser in a range of 50-1000 m/z. Sample was
eluted with gradient elution using eluent A (0.1 % formic acid in water) and eluent B (0.1 %
formic acid in acetonitrile). Flow rate was set to 1 ml/min. The initial condition was 0% B
eluent, followed by a linear gradient to 100% B eluent by 2 min, from 2 to 3.75 min 100% B
eluent was retained; and from 3.75 to 4.5 min back to initial condition and retained to 5 min.
The column temperature was kept at 30°C and the injection volume was 1 µl.
Biochemical assay
Compounds were tested for JAK inhibition in a Z’-LYTE kinase inhibition assay (Life
Technologies). The Z´-LYTE biochemical assay employs a fluorescence-based, coupled-
enzyme format and is based on the differential sensitivity of phosphorylated and non-
phosphorylated peptides to proteolytic cleavage. The peptide substrate (Z'-LYTE Kinase
Assay Kit - Tyrosine 06 Peptide for JAK1 and JAK2) is labelled with two fluorophores –
coumarin and fluorescein, one at each end – that make up a FRET pair. After incubating the
kinase + peptide + test compound mixture for an hour, a development reaction is carried out,
during which any peptide that was not phosphorylated by the kinase is cleaved, disrupting the
resonance energy transfer between the FRET pair. Based on the ratio of the detected emission
at 445 nm (coumarin) and 520 nm (fluorescein), the ratio of cleaved vs. intact peptide (and
thus, the reaction progress) can be quantified.
The compounds were screened at 20 M concentration in 1% DMSO (final). Each data
point was acquired twice and the results were averaged. For the IC50 measurements, 10-point
curves were obtained with 3-fold serial dilutions of the compounds, starting from 100 M in
1% DMSO. (Duplicate data points were acquired for each concentration.)
The following assay protocols were applied:
JAK1: 5 μL of JAK1 / Substrate mixture is prepared and added to 2.5 μL of Test
Compound solution, to which 2.5 μL of ATP solution is added. The final 10 μL Kinase
Reaction consists of 21.2 - 91.5 ng JAK1 and 2 μM Substrate in 50 mM HEPES pH 7.0,
0.01% BRIJ-35, 10 mM MgCl2, 1 mM EGTA, 0.01% NaN3. After the 1 hour Kinase Reaction
incubation, 5 μL of a 1:128 dilution of Development Reagent A is added. After another hour
of incubation, the fluorescence is read out and the data is analysed.

JAK2: 5 μL of JAK2 / Substrate mixture is prepared and added to 2.5 μL of Test
Compound solution, to which 2.5 μL of ATP solution is added. The final 10 μL Kinase
Reaction consists of 0.05 - 0.42 ng JAK2 and 2 μM Substrate in 50 mM HEPES pH 7.5,
0.01% BRIJ-35, 10 mM MgCl2, 1 mM EGTA. After the 1 hour Kinase Reaction incubation, 5
μL of a 1:128 dilution of Development Reagent A is added. After another hour of incubation,
the fluorescence is read out and the data is analysed.
Chemistry
1. Method A
To the solution of carboxylic acid (1 eq.) in thionyl chloride were added one drop
DMF. The mixture was heated to reflux temperature and stirred for 3 hours. The solvent was
removed and the crude material dissolved in DCM. To this solution were added the amine (1
eq.) and triethylamine (1.5 eq). The mixture was stirred for 3 hours at room temperature under
argon. Afterwards the DCM was removed. The crude material was dissolved in ethyl acetate
and washed with water and brine. The resulting organic layer was dried over MgSO4, and the
solvent was removed by evaporation. The obtained residue was purified by flash
chromatography (CH2Cl2/MeOH).
N-(1H-indazol-5-yl)-2-(1H-tetrazol-1-yl)benzamide (4)
2-(1H-tetrazol-1-yl)benzoic acid (19a) (0.62 mg, 0.33 mmol), thionyl chloride (235
mg, 1.98 mmol) in toluene (4 ml). 1H-indazol-5-amine (22) (44 mg, 0.33 mmol),
triethylamine (50 mg, 0.5 mmol) in DCM (4 ml). 88 mg (87 %) white solid was obtained. 1H
NMR (500 MHz, DMSO) δ 13.02 (s, 1H), 10.54 (s, 1H), 9.85 (s, 1H), 8.04 (d, J = 8.7 Hz,
2H), 7.92 – 7.89 (m, 1H), 7.80 (s, 3H), 7.49 (d, J = 8.8 Hz, 1H), 7.40 (d, J = 8.8 Hz, 1H). 13C
NMR (125 MHz, DMSO) δ 163.74, 137.11, 133.47, 133.08, 131.51, 131.20, 131.08, 130.59,
129.22, 126.15, 122.54, 120.64, 110.62, 110.09 HRMS (ESI) (M+H)+ calcd for C15H12N7O+,
306.1103; found 306.1104
N-(1H-indazol-5-yl)-3-(1H-tetrazol-1-yl)benzamide (5)
3-(1H-tetrazol-1-yl)benzoic acid (370 mg, 1.95 mmol) in thionyl chloride (4 ml, 55
mmol). 1H-indazol-5-amine (22) (260 mg, 1.95 mmol), triethylamine (295 mg, 2.92 mmol) in
DCM (20 ml) 300 mg (50 %) white solid was obtained. 1H NMR (500 MHz, DMSO) δ 13.01
(s, 1H), 10.45 (s, 1H), 10.18 (s, 1H), 8.49 (s, 1H), 8.24 (s, 1H), 8.17 – 8.09 (m, 2H), 8.06 (s,

1H), 7.82 (t, J = 7.9 Hz, 1H), 7.63 (d, J = 9.1 Hz, 1H), 7.54 (d, J = 8.8 Hz, 1H). 13C NMR (125
MHz, DMSO) δ 164.41, 142.92, 137.72, 137.39, 134.29, 134.00, 132.14, 130.74, 129.09,
124.33, 123.14, 121.86, 120.82, 111.89, 110.55. HRMS (ESI) (M+H)+ calcd for C15H12N7O+,
306.1097; found 330.1098
2-chloro-N-(1H-indazol-5-yl)-6-(1H-tetrazol-1-yl)benzamide (21)
2-chloro-6-(1H-tetrazol-1-yl)benzoic acid (19b) (310 mg, 1.68 mmol) in thionyl
chloride (4 ml, 55 mmol), 1H-indazol-5-amine (22) (186 mg, 1.68 mmol), triethylamine (210
1
mg, 2.1 mmol) in DCM (20 ml) 398 mg (85 %) white solid was obtained. H NMR (300
MHz, DMSO) δ 13.06 (s, 1H), 10.75 (s, 1H), 9.84 (s, 1H), 8.08 (s, 1H), 8.02 (s, 1H), 7.93 (d,
13
J = 1.3 Hz, 1H), 7.89 – 7.77 (m, 2H), 7.53 (d, J = 8.9 Hz, 1H), 7.34 (d, J = 8.9 Hz, 1H). C
NMR (75 MHz, DMSO) δ 161.43, 145.30, 138.00, 134.29, 133.93, 132.64, 132.42, 132.33,
132.15, 131.64, 125.51, 123.32, 121.15, 111.45, 111.07. HRMS (ESI) (M+H)+ calcd for
C15H11N7OCl+, 340.0708; found 340.0718
N-(3-cyano-4-fluorophenyl)-2-(1H-tetrazol-1-yl)benzamide (23a)
2-(1H-tetrazol-1-yl)benzoic acid (95 mg, 0.5 mmol), in thionyl chloride (2 ml, 27.5 mmol), 5-
amino-2-fluorobenzonitrile (22) (70 mg, 0.5 mmol), triethylamine (75 mg, 0.74 mmol) in
1
DCM (5 ml) 100 mg (65 %) white solid was obtained. H NMR (500 MHz, DMSO) δ 10.90
(s, 1H), 9.85 (s, 1H), 8.01 (dd, J = 5.8, 2.6 Hz, 1H), 7.89 (d, J = 6.8 Hz, 1H), 7.86 – 7.75 (m,
4H), 7.50 (t, J = 9.1 Hz, 1H). 13C NMR (125 MHz, DMSO) δ 168.70, 162.07, 159.98, 142.53,
138.65, 136.09, 133.12, 132.72, 132.24, 132.19, 127.60, 125.99, 125.36, 117.33, 93.89. LC-
MS (DUIS) m/z 309 (M+H)
N-(1H-indazol-5-yl)benzamide (28a)
In this case, no benzoyl chloride was prepared, which was purchased. benzoyl chloride
(211 mg, 1.5 mmol), 1H-indazol-5-amine (20) (200 mg, 1.5 mmol), triethylamine (151 mg,
1
1.5 mmol) in DCM (5 ml) 315 mg (88 %) white solid was obtained. H NMR (500 MHz,
DMSO) δ 12.98 (s, 1H), 10.23 (s, 1H), 8.24 (s, 1H), 8.05 (s, 1H), 7.98 (d, J = 7.1 Hz, 2H),
13
7.64 (d, J = 8.7 Hz, 1H), 7.58 (d, J = 7.1 Hz, 1H), 7.53 (dd, J = 12.5, 5.8 Hz, 3H). C NMR
(125 MHz, DMSO) δ 165.86, 137.58, 135.59, 133.90, 132.51, 131.79, 128.78, 128.01,
123.13, 121.92, 111.63, 110.39 HRMS (ESI) (M+H)+ calcd for C14H12N3O+, 238.0974; found
238.0973
N-(3-cyano-4-fluorophenyl)benzamide (29a)

In this case, the benzoyl chloride wasn’t prepared, which was purchased. benzoyl
chloride (211 mg, 1.5 mmol), 5-amino-2-fluorobenzonitrile (20) (204 mg, 1.5 mmol),
triethylamine (366 mg, 1.5 mmol) in DCM (5 ml) 200 mg (55 %) white solid was obtained.
1H NMR (500 MHz, DMSO) δ 10.56 (s, 1H), 8.26 (s, 1H), 8.06 (s, 1H), 7.94 (s, 2H), 7.55 (d,
13
J = 33.0 Hz, 4H). C NMR (125 MHz, DMSO) δ 166.24, 159.79, 157.78, 136.76, 134.57,
132.41, 128.91, 128.14, 128.01, 124.61, 117.45, 117.28, 114.40. HRMS (ESI) (M+H)+ calcd
for C14H11N3OCl+, 272.0585; found 272.0591
3-cyano-4-fluoro-N-phenylbenzamide (34a)
3-cyano-4-fluorobenzoic acid (32) (400 mg, 2.42 mmol) in thionyl chloride (4 ml, 55
mmol) aniline (33a) (400 mg, 4.3 mmol), triethylamine (295 mg, 3.63 mmol) in DCM (20 ml)
1
387 mg (67 %) white solid was obtained. H NMR (500 MHz, DMSO) δ 10.26 (s, 1H) 7.46
(s, 1H), 7.44 (s, 1H), 7.43 (d, J = 1.5 Hz, 1H), 7.34 (d, J = 1.0 Hz, 1H), 7.33 (d, J = 1.7 Hz,
13
1H), 7.32 (t, J = 1.6 Hz, 2H), 7.30 (s, 1H). C NMR (125 MHz, DMSO) δ 166.86, 162.09,
136.68, 132.32, 132.24, 129.03, 128.15, 124.91, 122.49, 117.42, 116.64, 96.46. LC-MS
(DUIS) m/z 241 (M+H)
N-(2-chlorophenyl)-3-cyano-4-fluorobenzamide (34b)
3-cyano-4-fluorobenzoic acid (32) (400 mg, 2.42 mmol) in thionyl chloride (4 ml, 55
mmol) 2-chloroaniline (33b) (546 mg, 4.3 mmol), triethylamine (295 mg, 3.63 mmol) in
DCM (20 ml) 266 mg (40 %) white solid was obtained. 1H NMR (500 MHz, DMSO) δ 10.28
(s, 1H), 8.49 (dd, J = 6.2, 2.3 Hz, 1H), 8.38 – 8.29 (m, 1H), 7.70 (t, J = 9.0 Hz, 1H), 7.55 (dd,
J = 8.0, 1.5 Hz, 2H), 7.38 (dd, J = 7.8, 1.5 Hz, 1H), 7.32 (dd, J = 7.8, 1.7 Hz, 1H). 13C NMR
(125 MHz, DMSO) δ 206.86, 136.19, 136.11, 134.96, 134.04, 130.09, 129.00, 128.34,
128.01, 117.55, 117.39, 113.92. LC-MS (DUIS) m/z 275 (M+H)
3-cyano-N-(2,5-dichlorophenyl)-4-fluorobenzamide (34c)
3-cyano-4-fluorobenzoic acid (32) (400 mg, 2.42 mmol) in thionyl chloride (4 ml, 55
mmol) 2,5-dichloroaniline (33c) (700 mg, 4.3 mmol), triethylamine (295 mg, 3.63 mmol) in
DCM (20 ml) 388 mg (52 %) white solid was obtained. 1H NMR (500 MHz, DMSO) δ 10.36
(s, 1H), 8.49 (dd, J = 6.2, 2.2 Hz, 1H), 8.32 (dd, J = 4.4, 1.9 Hz, 1H), 7.74 – 7.66 (m, 2H),
7.59 (d, J = 8.6 Hz, 1H), 7.38 (dd, J = 8.7, 2.5 Hz, 1H). 13C NMR (125 MHz, DMSO) δ

163.43, 136.31, 136.21, 134.16, 131.94, 131.45, 128.45, 128.24, 127.98, 117.58, 117.42,
113.87. LC-MS (DUIS) m/z 310 (M+H)
3-cyano-N-(2,6-dichlorophenyl)-4-fluorobenzamide (34d)
3-cyano-4-fluorobenzoic acid (32) (400 mg, 2.42 mmol) in thionyl chloride (4 ml, 55
mmol) 2,6-dichloroaniline (33d) (700 mg, 4.3 mmol), triethylamine (295 mg, 3.63 mmol) in
DCM (20 ml) 374 mg (50 %) white solid was obtained. 1H NMR (500 MHz, DMSO) δ 10.55
(s, 1H), 8.48 (dd, J = 6.2, 2.2 Hz, 2H), 8.36 (ddd, J = 8.5, 5.2, 2.3 Hz, 2H), 7.67 (t, J = 9.0 Hz,
2H), 7.54 (d, J = 8.1 Hz, 4H), 7.33 (t, J = 8.1 Hz, 2H) 13C NMR (125 MHz, DMSO) δ 165.59,
163.51, 162.81, 136.10, 136.02, 133.94, 129.24, 128.91, 117.50, 117.34, 113.92 LC-MS
(DUIS) m/z 310 (M+H)
2. Method B
To the solution of carboxylic acid (1 eq.) in DCM were added HATU (1,05 eq.),
amine (1,1 eq.) and DIPEA (2 eq.). The mixture was stirred for 16 hour at room temperature.
Afterwards, the solution was washed with water and brine. The resulting organic layer was
dried over MgSO4, and the solvent was removed by evaporation. The crude material was
purified by flash chromatography (CH2Cl2:MeOH).
2-(ethylamino)-N-(1H-indazol-5-yl)benzamide (25)
2-(ethylamino)benzoic acid (100 mg, 0.6 mmol) in 5 ml DCM, 1H-indazol-5-amine (20) (87
mg, 0.65 mmol), HATU (240 mg 0.63 mmol), DIPEA (210 µl, 1.2 mmol). 30 mg (17 %)
1
white solid was obtained. H NMR (500 MHz, DMSO) δ 12.98 (s, 1H), 10.07 (s, 1H), 8.14
(s, 1H), 8.04 (s, 1H), 7.72 (d, J = 7.7 Hz, 1H), 7.57 (d, J = 8.9 Hz, 1H), 7.50 (d, J = 8.9 Hz,
1H), 7.38 (t, J = 4.5 Hz, 1H), 7.33 (t, J = 7.7 Hz, 1H), 6.72 (d, J = 8.4 Hz, 1H), 6.64 (t, J = 7.4
Hz, 1H), 3.18 – 3.11 (m, 2H), 1.21 (t, J = 7.1 Hz, 3H) 13C NMR (125 MHz, DMSO) δ 168.89,
150.85, 137.43, 135.39, 133.20, 131.24, 128.45, 127.13, 122.57, 122.43, 117.91, 117.81,
115.54, 110.33, 36.91, 13.72. HRMS (ESI) (M+H)+ calcd for C16H16N4O+, 280.1324; found
280.1326
2-chloro-N-(1H-indazol-5-yl)benzamide (28b)
2-chlorobenzoic acid (27b) (213 mg, 1.36 mmol) in 5 ml DCM, 1H-indazol-5-amine
(20) (200 mg, 1.5 mmol), HATU (517 mg 1.36 mmol), DIPEA (470 µl, 2.72 mmol). 272 mg

1
(73 %) white solid was obtained. H NMR (500 MHz, DMSO) δ 12.99 (s, 1H), 10.46 (s, 1H),
8.25 (s, 1H), 8.06 (s, 1H), 7.55 (ddd, J = 40.2, 21.7, 15.0 Hz, 6H). 13C NMR (125 MHz,
DMSO) δ 165.21, 137.65, 133.96, 132.38, 131.38, 130.43, 130.07, 129.38, 127.67, 123.12,
120.96, 110.73, 110.62, 47.49. LC-MS (DUIS) m/z 272, 274 (M+H)
2-chloro-N-(3-cyano-4-fluorophenyl)benzamide (29b)
2-chlorobenzoic acid (27b) (213 mg, 1.35 mmol) in 5 ml DCM, 5-amino-2-
fluorbenzonitrile (22) (205 mg, 1.5 mmol), HATU (517 mg, 1.575 mmol), DIPEA (470 µl, 3
1
mmol). 210 mg (57 %) white solid was obtained. H NMR (300 MHz, DMSO) δ 10.25 (s,
1H), 8.15 (d, J = 7.7 Hz, 1H), 7.95 (d, J = 7.5 Hz, 1H), 7.76 (t, J = 7,7 Hz, 1H), 7.48 (t, J =
6,5 Hz, 1H), 7.35-7,25 (m, 2H), 7.2-7,15 (m, 1H). 13C NMR (75 MHz, DMSO) δ 166.64,
161.03, 136.08, 134.15, 134.05, 132.22, 131.68, 130.77, 127.35, 127.02, 125.37, 117.44,
117.42, 94.01. LC-MS (DUIS) m/z 276, 278 (M+H)
3-chloro-N-(3-cyano-4-fluorophenyl)benzamide (29c)
3-chlorobenzoic acid (27c) (157 mg, 1 mmol) in 5 ml DCM, 5-amino-2-fluorbenzonitrile (22)
(150 mg, 1.1 mmol), HATU (400 mg, 1.05 mmol), DIPEA (350 µl, 2 mmol). 164 mg (60 %)
white solid was obtained. 1H NMR (500 MHz, DMSO) δ 10.55 (s, 1H), 8.35 (s, 1H), 8.15 (d,
J = 7.2 Hz, 1H), 8.02 (d, J = 6.9 Hz, 1H), 7.75 – 7.63 (m, 3H), 7.47 (dd, J = 6.9, 2.3 Hz, 1H).
13C NMR (125 MHz, DMSO) δ 166.86, 161.03, 136.08, 135.05, 133.67, 132.22, 131.14,
129.69, 128.52, 127.43, 125.39, 117.42, 117.40, 93.78. LC-MS (DUIS) m/z 276, 278 (M+H)
2,3-dichloro-N-(3-cyano-4-fluorophenyl)benzamide (29d)
2,3-dichlorobenzoic acid (27d) (191 mg, 1mmol) in 5 ml DCM, 5-amino-2-fluorobenzonitrile
(22) (150 mg, 1.1 mmol), HATU (400 mg, 1.05 mmol), DIPEA (350 µl, 2mmol). 216 mg (70
1
%) white solid was obtained. H NMR (500 MHz, DMSO) δ 10.36 (s, 1H), 8.53 (d, J = 6.2
Hz, 1H), 8.38 (d, J = 7.8 Hz, 1H), 7.80 (t, J = 9.0 Hz, 1H), 7.65 (dd, J = 7.7, 1.7 Hz, 1H), 7.57
13
(dd, J = 7.8, 1,5 Hz, 1H), 7.32 (d, J = 7.8 Hz, 1H). C NMR (125 MHz, DMSO) δ 166.62,
161.05, 136.10, 135.35, 134.15, 134.05, 132.94, 132.25, 128.60, 126.58, 125.25, 117.44,
117.42, 94.08. LC-MS (DUIS) m/z 310 (M+H)
2,5-dichloro-N-(3-cyano-4-fluorophenyl)benzamide (29e)
2,5 dichlorobenzoic acid (27e) (191 mg, 1 mmol), in 5 ml DCM, 5-amino-2-
fluorobenzonitrile (22) (150 mg, 1.1 mmol) HATU (400 mg, 1.05 mmol), DIPEA (350 µl,
1
2mmol). 108 mg (35 %) white solid was obtained. H NMR (500 MHz, DMSO) δ 10.43 (s,

1H), 8.91 (d, J = 4.1 Hz, 1H), 8.76 (d, J = 8.4 Hz, 1H), 7.84 – 7.74 (m, 3H), 7.69 (dd, J = 8.4,
4.5 Hz, 1H). 13C NMR (125 MHz, DMSO) δ 166.61, 161.03, 136.08, 134.96, 134.53, 133.60,
132.26, 132.22, 130.74, 127.63, 125.39, 117.48, 117.46, 93.72. LC-MS (DUIS) m/z 310
(M+H)
2-chloro-N-(3-cyano-4-fluorophenyl)-6-(1H-tetrazol-1-yl)benzamide (23b)
2-chloro-6-(1H-tetrazol-1-yl)benzoic acid (19b) (130 mg, 0.56 mmol) in 5 ml DCM, 5-amino-
2-fluorobnzonitrile (22) (84 mg, 0.62 mmol), HATU (280 mg, 0.6 mmol), DIPEA (195 µl,
1.12 mmol). 107 mg (56 %) white solid was obtained. 1H NMR (500 MHz, dmso) δ 10.30 (s,
1H), 9.49 (s, 1H), 8.05 (d, J = 7.3, 1H), 7.85 (t, J = 9.0 Hz, 1H), 7.55 (d, J = 6.9, 1H), 7.40
(dd, J = 7.8, 1.5 Hz, 1H), 7.32 - 7.15 (m, 2H). 13C NMR (125 MHz, DMSO) δ 170.66, 161.13,
142.53, 139.00, 136.08, 135.26, 133.83, 132.95, 132.22, 125.39, 124.32, 122.32, 117.69,
117.62, 93.68. LC-MS (DUIS) m/z 344, 346 (M+H)
3. Method C
The 2-fluorobenzonitrile derivatives was suspended in hydrazine hydrate. The mixture was
heated to reflux temperature at 2 hour. In the next step the solvent was evaporated. The crude
material was washed with water.
N-(3-amino-1H-indazol-5-yl)-2-(1H-tetrazol-1-yl)benzamide (24a)
N-(3-cyano-4-fluorophenyl)-2-(1H-tetrazol-1-yl)benzamide (23a) (90 mg, 0.3 mmol) in 1 ml
1
hydrazine hydrate. 65 mg (67 %) white solid was obtained. H NMR (300 MHz, DMSO) δ
13.06 (s, 1H), 10.75 (s, 1H), 9.84 (s, 1H), 8.08 (s, 1H), 8.02 (s, 1H), 7.93 (d, J = 1.3 Hz, 1H),
13
7.89 – 7.77 (m, 2H), 7.53 (d, J = 8.9 Hz, 1H), 7.34 (d, J = 8.9 Hz, 1H) C NMR (75 MHz,
DMSO) δ 161.43, 145.30, 138.00, 134.29, 133.93, 132.64, 132.42, 132.33, 132.15, 131.64,
125.51, 123.32, 121.15, 111.45, 111.07. . HRMS (ESI) (M+H)+ calcd for C15H12N8O+,
321.1206; found 321.1213
N-(3-amino-1H-indazol-5-yl)-2-chloro-6-(1H-tetrazol-1-yl)benzamide (24b)
2-chloro-N-(3-cyano-4-fluorophenyl)-6-(1H-tetrazol-1-yl)benzamide (23b) (100 mg, 0.29
1
mmol) in 1 ml hydrazine hydrate. 74 mg (72 %) white solid was obtained. H NMR (500
MHz, DMSO) δ 11.29 (s, 1H), 8.47 (s, 1H), 8.01 (s, 1H), 7.49 (t, J = 8.0 Hz, 1H), 7.33 (d, J =
13
8.5 Hz, 3H), 7.20 (d, J = 8.8 Hz, 1H), 7.14 (d, J = 7.7 Hz, 1H), 5.27 (s, 2H). C NMR (125

MHz, DMSO) δ 170.52, 145.61, 142.53, 139.00, 135.52, 135.26, 133.83, 132.95, 129.99,
124.32, 122.32, 120.88, 111.51, 111.49, 108.36., HRMS (ESI) (M+H)+ calcd for
C15H12N8OCl+, 355.0817 found 355.0811
N-(3-amino-1H-indazol-5-yl)benzamide (30a)
N-(3-cyano-4-fluorophenyl)benzamide (29a) (200 mg, 0.83 mmol) in 2 ml hydrazine hydrate.
1
86 mg (41 %) white solid was obtained. H NMR (300 MHz, DMSO) δ 11.37 (s, 1H), 10.19
(s, 1H), 8.15 (s, 1H), 8.02 (d, J = 6.9 Hz, 2H), 7.65 – 7.51 (m, 3H), 7.47 (d, J = 8.7 Hz, 1H),
7.25 (d, J = 8.8 Hz, 1H), 5.31 (s, 2H). 13C NMR (75 MHz, DMSO) δ 165.88, 149.99, 139.74,
135.89, 132.00, 130.09, 129.07, 128.24, 122.87, 114.40, 113.27, 109.94. HRMS (ESI)
(M+H)+ calcd for C14H13N4O+, 253.1083; found 253.1087
N-(3-amino-1H-indazol-5-yl)-2-chlorobenzamide (30b)
2-chloro-N-(3-cyano-4-fluorophenyl)benzamide (29b) (300 mg, 1.1 mmol) in 2 ml hydrazine
1
hydrate. 112 mg (35 %) white solid was obtained. H NMR (300 MHz, DMSO) δ 11.37 (s,
1H), 10.37 (s, 1H), 8.19 (s, 1H), 7.65 – 7.57 (m, 2H), 7.57 – 7.44 (m, 2H), 7.40 (dd, J = 8.9,
13
1.3 Hz, 1H), 7.24 (d, J = 8.8 Hz, 1H), 5.31 (s, 2H). C NMR (75 MHz, DMSO) δ 165.26,
150.03, 139.66, 138.09, 131.56, 130.73, 130.34, 130.06, 129.63, 127.94, 121.76, 114.33,
112.08, 110.13. HRMS (ESI) (M+H)+ calcd for C14H12N4OCl+, 287.0694; found 287.0703
N-(3-amino-1H-indazol-5-yl)-2,5-dichlorobenzamide (30e)
2,5-dichloro-N-(3-cyano-4-fluorophenyl)benzamide (29e) (150 mg, 0.48 mmol) in 2 ml
1
hydrazine hydrate. 149 mg (96 %) white solid was obtained. H NMR (300 MHz, DMSO) δ
11.36 (s, 1H), 10.43 (s, 1H), 8.14 (s, 1H), 7.71 (s, 1H), 7.59 (s, 2H), 7.37 (d, J = 8.5 Hz, 1H),
7.22 (d, J = 8.6 Hz, 1H), 5.30 (s, 2H). 13C NMR (75 MHz, DMSO) δ 163.76, 150.05, 139.72,
139.47, 132.53, 132.11, 131.34, 129.83, 129.63, 129.32, 121.71, 114.30, 112.13, 110.20.
+
HRMS (ESI) (M+H)+ calcd for C14H11N4OCl2 , 321.0304; found 321.0315
N-(3-amino-1H-indazol-5-yl)-2,3-dichlorobenzamide (30d)
2,3-dichloro-N-(3-cyano-4-fluorophenyl)benzamide (29d) (100 mg, 0.32 mmol) in 2 mil
1
hydrazine hydrate. 85 mg (83 %) white solid was obtained. H NMR (300 MHz, DMSO) δ
11.36 (s, 1H), 10.43 (s, 1H), 8.15 (s, 1H), 7.75 (d, J = 7.7 Hz, 1H), 7.56 (d, J = 6.6 Hz, 1H),
13
7.48 (t, J = 7.7 Hz, 1H), 7.35 (d, J = 8.7 Hz, 1H), 7.22 (d, J = 8.8 Hz, 1H), 5.30 (s, 2H). C
NMR (75 MHz, DMSO) δ 164.47, 150.05, 140.34, 139.70, 132.77, 131.82, 129.83, 129.38,

128.95, 128.09, 121.65, 114.32, 112.11, 110.22. HRMS (ESI) (M+H)+ calcd for
+
C14H11N4OCl2 , 321.0304 found 321.0307
N-(3-amino-1H-indazol-5-yl)-3-chlorobenzamide (30c)
3-chloro-N-(3-cyano-4-fluorophenyl)benzamide (29c) (150 mg, 0,55 mmol) in 2 ml hydrazine
1
hydrate. 72 mg (46 %) white solid was obtained. H NMR (300 MHz, DMSO) δ 11.38 (s,
1H), 10.29 (s, 1H), 8.13 (s, 1H), 8.06 (s, 1H), 7.97 (d, J = 7.6 Hz, 1H), 7.69 (d, J = 7.9 Hz,
1H), 7.60 (t, J = 7.8 Hz, 1H), 7.46 (d, J = 8.7 Hz, 1H), 7.25 (d, J = 8.8 Hz, 1H), 5.31 (s, 2H).
13C NMR (75 MHz, DMSO) δ 164.35, 150.01, 139.78, 137.85, 133.92, 131.85, 131.10,
129.78, 128.01, 127.05, 122.72, 114.35, 113.31, 109.99. HRMS (ESI) (M+H)+ calcd for
C14H12N4OCl+, 287.0694; found 287.0697
3-amino-N-phenyl-1H-indazole-5-carboxamide (35a)
3-cyano-4-fluoro-N-phenylbenzamide (34a) (300 mg, 1.25 mmol) in 2 ml hydrazine
1
hydrate. 195 mg (62 %) white solid was obtained. H NMR (500 MHz, DMSO) δ 10.14 (s,
1H), 8.50 (s, 1H), 7.84 (d, J = 8.7 Hz, 1H), 7.79 (d, J = 8.0 Hz, 2H), 7.34 – 7.19 (m, 4H), 7.03
13
(t, J = 7.3 Hz, 1H), 5.51 (s, 2H). C NMR (125 MHz, DMSO) δ 166.50, 150.78, 142.81,
140.12, 128.92, 126.03, 124.76, 123.57, 122.43, 120.63, 114.13, 109.28. HRMS (ESI)
(M+H)+ calcd for C14H13N4O+, 253.1089; found 253.1076
3-amino-N-(2-chlorophenyl)-1H-indazole-5-carboxamide (35b)
N-(2-chlorophenyl)-3-cyano-4-fluorobenzamide (34b) (200 mg, 0.72 mmol) in 2 ml
hydrazine hydrate. 161 mg (78 %) white solid was obtained. 1H NMR (500 MHz, DMSO) δ
9.75 (s, 1H), 8.44 (s, 1H), 7.85 (dd, J = 8.7, 1.2 Hz, 1H), 7.67 (d, J = 7.9 Hz, 1H), 7.53 (d, J =
8.2 Hz, 1H), 7.36 (t, J = 7.8 Hz, 1H), 7.29 (d, J = 8.7 Hz, 1H), 7.24 (t, J = 7.6 Hz, 2H), 5.55
(s, 2H). 13C NMR (125 MHz, DMSO) δ 166.25, 142.94, 136.04, 129.90, 129.20, 128.07,
127.85, 127.79, 127.22, 125.86, 123.91, 122.52, 114.25, 109.42. HRMS (ESI) (M+H)+ calcd
for C14H12N4OCl+, 287.0694; found 287.0688
3-amino-N-(2,5-dichlorophenyl)-1H-indazole-5-carboxamide (35c)
3-cyano-N-(2,5-dichlorophenyl)-4-fluorobenzamide (34c) (150 mg, 0.5 mmol) in 2 ml
1
hydrazine hydrate. 104 mg (65 %) white solid was obtained. H NMR (500 MHz, DMSO) δ
9.82 (s, 1H), 8.44 (s, J = 0.9 Hz, 1H), 7.85 – 7.80 (m, 2H), 7.56 (t, J = 8.6 Hz, 2H), 7.33 –
13
7.27 (m, 2H), 5.57 (s, 2H). C NMR (125 MHz, DMSO) δ 166.30, 137.34, 134.09, 131.87,

131.25, 127.84, 127.44, 126.96, 126.76, 125.91, 123.48, 122.70, 111.92, 109.49. HRMS (ESI)
+
(M+H)+ calcd for C14H11N4OCl2 , 321.0309; found 321.0302
3-amino-N-(2,6-dichlorophenyl)-1H-indazole-5-carboxamide (35d)
3-cyano-N-(2,6-dichlorophenyl)-4-fluorobenzamide (34d) (150 mg, 0.5 mmol) in 2 ml
1
hydrazine hydrate. 93 mg (58 %) white solid was obtained. H NMR (500 MHz, DMSO) δ
11.70 (s, 1H), 10.07 (s, 1H), 8.43 (s, 1H), 7.87 (d, J = 8.6 Hz, 1H), 7.56 (d, J = 8.1 Hz, 2H),
7.36 (t, J = 8.1 Hz, 1H), 7.29 (d, J = 8.7 Hz, 1H), 5.52 (s, 2H). 13C NMR (125 MHz, DMSO)
δ 166.22, 150.86, 142.95, 134.77, 134.35, 129.51, 128.90, 125.81, 123.46, 122.77, 114.30,
+
109.30. HRMS (ESI) (M+H)+ calcd for C14H11N4OCl2 , 321.0309; found 321.0304
Synthesis of 2-chloro-6-(1H-tetrazol-1-yl)benzoic acid (19b)
To the solution of 2-amino-6-chlorobenzoic acid (18b) (624 mg, 3.6 mmol) in glacial
acetic acid were added NaN3 (710 mg, 10.9 mmol) and trimethyl ortoformate (1.16 g, 10.9
mmol). The mixture was stirred for 2 hours at room temperature. The precipitate was filtered
1
and wash with diethyl ether. 485 mg (60%) white solid was obtained. H NMR (500 MHz,
DMSO) δ 14.06 (s, 1H), 9.89 (s, 1H), 7.83 (d, J = 7.8 Hz, 1H), 7.78 (d, J = 7.8 Hz, 1H), 7.73
13
(t, J = 7.6 Hz, 1H). C NMR (125 MHz, DMSO) δ 164.93, 144.83, 132.25, 132.09, 131.79,
131.30, 131.14, 125.11. LC-MS (ESI) m/z 223 (M-H)
Synthesis of 3-((1H-tetrazol-1-yl)methyl)-N-(1H-indazol-5-yl)benzamide (6)
To the solution of 3-(aminomethyl)benzoic acid (235 mg, 1.6 mmol) in glacial acetic acid
(8 ml) were added NaN3 (300 mg, 4.7 mmol) and trimethyl ortoformate (495 mg, 4.7 mmol).
The mixture was stirred for 2 hours on 80 °C. The precipitate was filtered and wash with
diethyl ether. 110 mg (30 %) white solid was obtained. The crude material was further for the
next step.
Method B was used to synthesize 3-((1H-tetrazol-1-yl)methyl)-N-(1H-indazol-5-
yl)benzamide (6) 3-((1H-tetrazol-1-yl)methyl)benzoic acid (102 mg, 0.5 mmol) in thionyl
chloride (2 ml, 27.5 mmol), 1H-indazol-5-amine (22) (65 mg, 0.5 mmol), triethylamine (75
1
mg, 0.74 mmol) in DCM (10 ml). 92 mg (57 %) white solid obtained. H NMR (500 MHz,
DMSO) δ 12.98 (s, 1H), 10.27 (s, 1H), 9.54 (s, 1H), 8.18 (s, 1H), 8.04 (s, 1H), 7.96 – 7.93 (m,
13
2H), 7.62 – 7.48 (m, 4H), 5.80 (s, 2H). MS (DUIS) C NMR (125 MHz, DMSO) δ 166.86,
144.48, 140.47, 137.43, 137.39, 135.51, 131.11, 129.45, 129.31, 127.17, 127.02, 125.73,

122.60, 117.91, 110.33, 49.99. m/z 320 (M+H), HRMS (ESI) (M+H)+ calcd for C16H14N7O+,
320.1254; found 320.1257
Synthesis of 2-((1H-tetrazol-1-yl)methyl)-N-(1H-indazol-5-yl)benzamide (7)
To the solution of 2-(aminomethyl)benzoic acid (258 mg, 1.7 mmol) in glacial acetic acid
(3 ml) were added NaN3 (331 mg, 5.1 mmol) and trimethyl ortoformate (541 mg, 5.1 mmol).
The mixture was stirred for 2 hours at room temperature. The precipitate was filtered and
wash with diethyl ether. 140 mg (40 %) white solid was obtained. The crude material was
further for the next step.
Method B was used to synthesize 2-((1H-tetrazol-1-yl)methyl)-N-(1H-indazol-5-
yl)benzamide (7). 2-((1H-tetrazol-1-yl)methyl)benzoic acid (139 mg, 0.68 mmol) in thionyl
chloride (2 ml, 27.5 mmol), 1H-indazol-5-amine (22) (91 mg, 0.68 mmol), triethylamine (103
1
mg, 1 mmol) in DCM (10 ml). 82 mg (37 %) white solid was obtained. H NMR (500 MHz,
DMSO) δ 13.00 (s, 1H), 10.44 (s, 1H), 9.36 (s, 1H), 8.23 (s, 1H), 8.07 (s, 1H), 7.73 – 7.67 (m,
1H), 7.54 (dd, J = 11.1, 5.9 Hz, 4H), 7.41 – 7.36 (m, 1H), 5.91 (s, 2H). 13C NMR (125 MHz,
DMSO) δ 166.45, 144.29, 137.15, 136.08, 133.47, 132.71, 131.76, 130.52, 129.99, 128.63,
128.22, 122.60, 121.03, 110.92, 109.99, 48.74, 39.52. MS (DUIS) m/z 320 (M+H), HRMS
(ESI) (M+H)+ calcd for C16H14N7O+, 320.1254; found 320.1259
Synthesis of 1-(2-ethynylphenyl)-1H-tetrazole (9)
To the solution of 2-((trimethylsilyl)ethynyl)aniline (100 mg, 0.53 mmol) in glacial
acetic acid (5 ml) were added NaN3 (120 mg, 1.9 mmol) and trimethyl ortoformate (200 mg,
1.9 mmol). The mixture was stirred for 2 hours at room temperature. The precipitate was
filtered and wash with diethyl ether. On the next step, the protecting group was removed with
TBAF (150 mg, 0.57 mmol) in THF (5 ml). The mixture was stirred for 2 hours at room
temperature. After, the mixture was washed with water (10 ml), and the inorganic phase was
extracted with EtOAc (3X5 ml). The combined organic phase was dried with NaSO4 and
evaporated. The crude material was purified with flash chromatography (hexane:EtOAc 0-
1
20%). 65 mg (92 %) solid was obtained. H NMR (500 MHz, DMSO) δ 9.57 (s, 1H), 7.44
13
(dd, J = 13.3, 7.5 Hz, 2H), 7.12 (d, J = 8.2 Hz, 1H), 7.05 (t, J = 7.6 Hz, 1H), 4.58 (s, 1H). C
NMR (125 MHz, DMSO) δ 133.73, 131.03, 123.24, 115.53, 109.90, 88.22, 31.12. LC-MS
(DUIS) m/z 171 (M+H)
Synthesis of 5-(4-(2-(1H-tetrazol-1-yl)phenyl)-1H-1,2,3-triazol-1-yl)-1H-indazole (10)

To the solution of 5-azido-1H-indazole31,32 (159 mg, 1 mmol) in 10 ml CH3CN were added
sodium ascorbate (99 mg, 0.5 mmol), 1-(2-ethynylphenyl)-1H-tetrazole (9) (114 mg, 1
mmol), CuSO4 solution (16 mg, 0.1 mmol in 1 ml water). The suspension was stirred for 12
hours at room temperature. Afterwards, DMSO (5 ml), L-proline (138 mg, 1.2 mmol), K2CO3
(165 mg, 1.2 mmol) were added to the mixture and heated to 85 °C. The solution was stirred
for 12 hours. Thereafter, EtOAc (15 ml) was added to the solution and washed with water and
brine. The organic phase was dried with MgSO4 and evaporated. The crude material was
purified with flash chromatography (CH2Cl2:MeOH 0-10%) 150 mg (46 %) white solid was
1
obtained. H NMR (500 MHz, DMSO) δ 13.00 (s, 1H), 10.68 (s, 1H), 9.78 (s, 1H), 8.02 (s,
1H), 7.96 (s, 1H), 7.88 (d, J = 8.0 Hz, 1H), 7.79 (dd, J = 19.7, 7.6 Hz, 3H), 7.47 (d, J = 8.8 Hz,
1H), 7.29 (d, J = 8.8 Hz, 1H). 13C NMR (75 MHz, DMSO) δ 145.98, 144.14, 140.13, 135.19,
132.20, 131.48, 130.59, 130.54, 130.25, 128.92, 127.83, 123.29, 122.92, 120.28, 113.15,
+
112.40. HRMS (ESI) (M+H)+ calcd for C16H12N9 , 330.1210; found 330.1216
Synthesis of 3-(2-(1H-tetrazol-1-yl)phenyl)-5-(1H-indazol-5-yl)isoxazole (13)
2-nitrobenzaldehyde (432 mg, 2.86 mmol), hydroxylamine hydrochloride (195 mg, 1.49
mmol) and NaOH (130 mg, 3.25 mmol) were dissolved in the mixture of tert-butanol (16.25
ml) and water (16.25 ml). After 3 hour stirring at room temperature the tert-butyl 5-ethynyl-
1H-indazole-1-carboxylate (360 mg, 1.18 mmol), Chloramine-T (911 mg, 3.23 mmol) and
copper(I)-iodide were added to the solution and stirred for 18 hours at room temperature. The
end of the reaction was check with TLC (hexane:EtOAc 2:1). After that the mixture was
extracted with EtOAc, then the organic phase was washed with solution of ammonium
chloride (5 %) and brine. The organic phase was dried over NaSO4 and distilled under
reduced pressure. The crude produce was purified with flash chromatography (hexane:EtOAc
1
0-15 %). 96 mg (21 %) 5-(1H-indazol-5-yl)-3-(2-nitrophenyl)isoxazole was obtained. H
NMR (500 MHz, DMSO) δ 8.37 (s, 1H), 8.23 (s, 1H), 8.09 (d, J = 8.1 Hz, 1H), 7.88 – 7.83
13
(m, 3H), 7.81 – 7.78 (m, 1H), 7.70 (d, J = 8.7 Hz, 1H), 7.31 (s, 1H) C NMR (125 MHz,
DMSO) δ 170.91, 160.54, 148.95, 135.24, 133.79, 131.86, 131.72, 129.75, 128.03, 124.91,
124.21, 123.20, 119.52, 119.26, 111.71, 99.84. LC-MS (DUIS) m/z 307 (M+H)
5-(1H-indazol-5-yl)-3-(2-nitrophenyl)isoxazole (96 mg, 0.313 mmol) was dissolved in
ethanol and Na2S (48.48 mg, 0.622 mmol) was added to them. The mixture was refluxed at 16

hours and the end of the reaction was followed with TLC (hexane:EtOAc 1:1). After the
removal of ethanol the precipitation was dissolved in EtOAc and extracted with water. The
organic phase was dried over NaSO4 and distilled under high vacuum. The crude material was
further for the next step. 1H NMR (500 MHz, DMSO) δ 8.36 (s, 1H), 8.22 (s, 1H), 8.09 (d, J =
8.4 Hz, 1H), 7.89 – 7.83 (m, 3H), 7.81 – 7.77 (m, 1H), 7.72 (d, J = 8.7 Hz, 1H), 7.30 (s, 1H),
6.27 (s, 2H). LC-MS (DUIS) m/z 277 (M+H)
2-(3-(1H-indazol-5-yl)isoxazol-5-yl)aniline (82 mg 0.294 mmol), sodium azide (57.25
mg, 0.88 mmol) and trimethyl orthoformate (97 μl) were dissolved in 99% acetic acid (330
μl). The mixture was stirred for 10 hours at 80 °C and the end of the reaction was analyzed
with TLC (hexane:EtOAc 1:1). After the removal of acetic acid by distillation, the precipitate
was dissolved in EtOAc and extracted with distilled water. The organic phase was dried over
Na2SO4 and finally was distilled under reduced pressure. 47 mg (48 %) white solid was
obtained. 1H NMR (500 MHz, DMSO) δ 9.78 (s, 1H), 8.26 (s, 1H), 8.21 (s, 1H), 8.03 (d, J =
7.6 Hz, 1H), 7.91 – 7.85 (m, 1H), 7.83 – 7.77 (m, 3H), 7.75 (d, J = 8.8 Hz, 1H), 7.67 (d, J =
13
8.8 Hz, 1H), 7.02 (s, 1H). C NMR (125 MHz, DMSO) δ 166.79, 156.39, 147.36, 142.53,
135.39, 131.74, 130.52, 128.07, 126.45, 126.31, 126.26, 124.91, 124.21, 121.73, 119.41,
114.44, 97.41. HRMS (ESI) (M+H)+ calcd for C17H12N7O+, 330.1097; found 330.1102
Synthesis of 2-(2-(1H-tetrazol-1-yl)phenyl)-4-(1H-indazol-5-yl)thiazole (16a)
The 2-(1H-tetrazol-1-yl)benzoic acid (723 mg, 3.81 mmol) was dissolved in thionyl
chloride and stirred at 90 °C for 4 hours. Afterwards, the solution was cooled to room
temperature and distilled under reduced pressure. The precipitate was dissolved in DCM and
added to the solution of ammonium hydroxide (25 %) in portions with stirring under cooling
(5 °C). The white precipitate was filtered. 515 mg (72 %) 2-(1H-tetrazol-1-yl)benzamide was
obtained. 1H NMR (500 MHz, DMSO) δ 9.67 (s, 1H), 7.99 (s, 1H), 7.69 (ddd, J = 15.9, 11.1,
13
4.4 Hz, 4H), 7.48 (s, 1H) C NMR (125 MHz, DMSO) δ 167.56, 145.03, 133.47, 131.47,
131.42, 131.03, 129.27, 127.01. LC-MS (DUIS) m/z 190 (M+H)
The 2-(1H-tetrazol-1-yl)benzamide (903 mg, 4.78 mmol) was dissolved in absolute THF
(100 ml) and Lawesson’s reagent (965 mg, 2.4 mmol) was added to the solution. The mixture
was stirred for 2 hours at 65 °C under inert atmosphere. The end of the reaction was checked
with TLC (CH3Cl:MeOH 10:1). The product was purified with flash chromatography
(DCM:MeOH 0-10%). 661 mg (67 %) 2-(1H-tetrazol-1-yl)benzothioamide (15a) was
1
obtained. H NMR (500 MHz, DMSO) δ 10.00 (s, 1H), 9.70 (s, 1H), 9.63 (s, 1H), 7.62 (s,

13
3H), 7.55 (s, 1H) C NMR (126 MHz, DMSO) δ 198.48, 145.10, 139.61, 130.92, 130.31,
129.64, 128.16, 126.70 LC-MS (DUIS) m/z 206 (M+H)
2-bromo-1-(1H-indazol-5-yl)ethan-1-one20 (203 mg, 0.85 mmol) and 2-(1H-tetrazol-1-
yl)benzothioamide (131 mg, 0.64 mmol) were dissolved in ethanol (10 ml) during stirring.
After that the mixture was stirred for 14 hours at 80 °C. The reaction was analyzed with TLC
1
(CH3Cl:MeOH 20:1). 73 mg (33 %) produce was obtained. H NMR (300 MHz, DMSO) δ
13.49 – 12.70 (m, 1H), 9.93 (s, 1H), 8.24 (d, J = 7.7 Hz, 1H), 8.20 – 8.10 (m, 3H), 7.95 – 7.72
(m, 5H), 7.59 (d, J = 8.8 Hz, 1H) 13C NMR (75 MHz, DMSO) δ 162.46, 156.48, 146.24,
140.44, 134.72, 132.68, 131.69, 131.23, 131.04, 130.29, 129.63, 126.97, 125.45, 123.79,
118.81, 115.45, 111.14 HRMS (ESI) (M+H)+ calcd for C17H12N7S+, 346.0869; found
346.0875
Synthesis of 2-(2-chloro-6-(1H-tetrazol-1-yl)phenyl)-4-(1H-indazol-5-yl)thiazole (16b)
2-chloro-6(1H-tetrazol-1-yl)benzoic acid (500 mg, 2.23 mmol) was dissolved in thionyl
chloride (25 ml) and stirred for 3 hours at 90 °C. After the removal of thionyl chloride was
dissolved in DCM and was added in portions to THF which was saturated by ammonia. The
mixture was cooled to 0 °C and was stirred for 20 minutes. The end of the reaction was
followed with TLC (CH3Cl:MeOH 10:1) and the crude product was purified with flash
chromatography (DCM:MeOH 0-10%). 382 mg (77 %) 2-chloro-6-(1H-terazol-1-
yl)benzamide (15b) was obtained. LC-MS (DUIS) m/z 225, 227 (M+H)
2-chloro-6- (1H-tetrazol-1-yl)benzamide (183 mg, 0.82 mmol), thiophosphoryl-chloride
(90 μl), and water (16μl) were dissolved in triethylamine (2 ml). The mixture was
microwaved for 15 minutes at 70 °C. The end of the reaction was checked with TLC (SiO2
TLC, CH3Cl:MeOH-10:1). The precipitate was dissolved in DCM and washed with water and
brine. The organic phase was dried over MgSO4 and evaporated. The crude material was
purified with flash chromatography (DCM:MeOH 0-10%). 20 mg (10 %) 2- chloro-6-(1H-
tetrazol-1-yl)benzothioamide was obtained. LC-MS (DUIS) m/z 241, 243 (M+H)
2- chloro-6-(1H-tetrazol-1-yl)benzothioamide (20 mg, 0.090 mmol) and 2-bromo-1-(1H-
indazol-5-yl)ethan-1-one20 (19 mg, 0.079 mmol) were dissolved in ethanol. The solution was
stirred at 80 °C overnight. The end of the reaction was analyzed with TLC (SiO2 TLC,
CH3Cl:MeOH-10:1). The solvent was evaporated and the crude material was purified with
1
preparative TLC (CH3Cl:MeOH 10:1). 5.3 mg (18%) solid was obtained. . H NMR (500
MHz, DMSO) δ 9.68 (s, 1H), 8.20 (s, 1H), 8.16 (s, 1H), 8.09 (s, 1H), 8.01 (dd, J = 7.6 Hz,