Exploration of nitrogen heterocycle scaffolds for the development of potent human neutrophil elastase inhibitors

Article abstract of DOI:10.1016/j.bmc.2020.115836

Human neutrophil elastase (HNE) is a potent protease that plays an important physiological role in many processes but is also involved in a variety of pathologies that affect the pulmonary system. Thus, compounds able to inhibit HNE proteolytic activity could represent effective therapeutics. We present here a new series of pyrazolopyridine and pyrrolopyridine derivatives as HNE inhibitors designed as modifications of our previously synthesized indazoles and indoles in order to evaluate effects of the change in position of the nitrogen and/or the insertion of an additional nitrogen in the scaffolds on biological activity and chemical stability. We obtained potent HNE inhibitors with IC₅₀ values in the low nanomolar range (10–50 nM), and some compounds exhibited improved chemical stability in phosphate buffer (t_1/2 > 6 h). Molecular modeling studies demonstrated that inhibitory activity was strictly dependent on the formation of a Michaelis complex between the OH group of HNE Ser195 and the carbonyl carbon of the inhibitor. Moreover, in silico ADMET calculations predicted that most of the new compounds would be optimally absorbed, distributed, metabolized, and excreted. Thus, these new and potent HNE inhibitors represent novel leads for future therapeutic development.

Full text of DOI:10.1016/j.bmc.2020.115836

Journal Pre-proofs
Exploration of Nitrogen Heterocycle Scaffolds for the Development of Potent
Human Neutrophil Elastase Inhibitors
Niccolò Cantini, Andrei I. Khlebnikov, Letizia Crocetti, Igor A. Schepetkin,
Giuseppe Floresta, Gabriella Guerrini, Claudia Vergelli, Gianluca Bartolucci,
Mark T. Quinn, Maria Paola Giovannoni
PII:
S0968-0896(20)30666-0
https://doi.org/10.1016/j.bmc.2020.115836
BMC 115836
DOI:
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
11 August 2020
27 October 2020
29 October 2020
Please cite this article as: N. Cantini, A.I. Khlebnikov, L. Crocetti, I.A. Schepetkin, G. Floresta, G. Guerrini, C.
Vergelli, G. Bartolucci, M.T. Quinn, M. Paola Giovannoni, Exploration of Nitrogen Heterocycle Scaffolds for the
Development of Potent Human Neutrophil Elastase Inhibitors, Bioorganic & Medicinal Chemistry (2020), doi:
https://doi.org/10.1016/j.bmc.2020.115836
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 Ltd.

Exploration of Nitrogen Heterocycle Scaffolds for the Development of Potent Human
Neutrophil Elastase Inhibitors
a
b,c
a*
d
e
Niccolò Cantini , Andrei I. Khlebnikov , Letizia Crocetti , Igor A. Schepetkin , Giuseppe Floresta ,
a
a
a
d
a
Gabriella Guerrini , Claudia Vergelli , Gianluca Bartolucci , Mark T. Quinn , Maria Paola Giovannoni
aNEUROFARBA, Pharmaceutical and Nutraceutical Section, University of Florence, Sesto Fiorentino, Italy
bNational Research Tomsk Polytechnic University, Tomsk 634050, Russia
cFaculty of Chemistry, Tomsk State University, Tomsk 634050, Russia
dDepartment of Microbiology and Immunology, Montana State University, Bozeman, MT, USA
eInstitute of Pharmaceutical Science, King’s College London, Stamford Street, London SE1 9NH, UK
*Corresponding Author
Letizia Crocetti
Department of NEUROFARBA
Via Ugo Schiff 6
Sesto Fiorentino, 50019 Firenze
Tel +39-055-4573683
E-mail: letizia.crocetti@unifi.it
Keywords: human neutrophil elastase, inhibitors, nitrogen heterocycle, stability, ADMET, molecular
docking
Abstract
1

Human neutrophil elastase (HNE) is a potent protease that plays an important physiological role
in many processes but is also involved in a variety of pathologies that affect the pulmonary system.
Thus, compounds able to inhibit HNE proteolytic activity could represent effective therapeutics. We
present here a new series of pyrazolopyridine and pyrrolopyridine derivatives as HNE inhibitors
designed as modifications of our previously synthesized indazoles and indoles in order to evaluate
effects of the change in position of the nitrogen and/or the insertion of an additional nitrogen in the
scaffolds on biological activity and chemical stability. We obtained potent HNE inhibitors with IC50
values in the low nanomolar range (10-50 nM), and some compounds exhibited improved chemical
stability in phosphate buffer (t1/2 > 6 h). Molecular modeling studies demonstrated that inhibitory
activity was strictly dependent on the formation of a Michaelis complex between the OH group of
HNE Ser195 and the carbonyl carbon of the inhibitor. Moreover, in silico ADMET calculations
predicted that most of the new compounds would be optimally absorbed, distributed, metabolized,
and excreted. Thus, these new and potent HNE inhibitors represent novel leads for future therapeutic
development.
1. Introduction
2

Heterocyclic nuclei are considered privileged scaffolds in medicinal chemistry, and a wide variety
of combinations of nitrogen, sulfur, and oxygen atoms in five- and six-membered rings have been
reported. Notably, nitrogen heterocycles play a central role in drug discovery and are represented in
~
85% of all biologically active chemical molecules. Furthermore, the number and the position of the
nitrogen atoms in the ring influence some pharmacokinetic properties, such as solubility,
1
hydrophilic/lipophilic characteristics, and the ability to form hydrogen bonds. We have identified a
number of heterocyclic scaffolds that were modified based on the desired protein target in order to
increase the number or the strength of the interactions with a catalytic site or receptor binding
domains.2-5 For example, the identification and modification of nitrogen heterocycles has been
essential in the discovery of new potent inhibitors of human neutrophil elastase (HNE), a proteolytic
enzyme belonging to the serine protease family. This enzyme is involved in a variety of chronic
6
diseases, including cardiopulmonary pathologies, and its regulation represents a promising
therapeutic approach, as evident by the interest of many pharmaceutical companies. In fact,
research in this field is quite active, and a number of HNE inhibitors are currently in various stages
7
8
of clinical trials, such as Alvelestat developed by AstraZeneca in various formulations , BAY85-8501
9
in phase II trials for bronchiectasis (Bayer HealthCare), and CHF6333, a novel potent inhaled HNE
inhibitor by Chiesi Farmaceutici, in phase I trials for cystic fibrosis.10,11 Currently, only two HNE
®
inhibitors are on the market. The first one, Prolastin (purified α1-AT), is a peptide drug used in the
12
®
treatment of α1-AT deficiency. The second is Sivelestat (ONO-5046, Elaspol 100), a non-peptide
inhibitor approved for intravenous use in Japan and South Korea for the treatment of ALI and ARDS
associated with systemic inflammatory response syndrome.13
We have designed and synthesized a number of potent HNE inhibitors with various scaffolds.14-
18
Among these compounds, the indazole and 7-azaindole nuclei resulted the best nitrogen bicycles,
including HNE inhibitors active in the low nanomolar range.19-22 Based on these previous studies, we
report here new HNE inhibitors that were developed through modification of the indazole and 7-
azaindole (1H-pyrrolo[2,3-b]pyridine) nuclei in order to determine if the introduction of another
nitrogen atom and/or moving nitrogen from its original position could affect biological activity (Figure
1). The profile of selected compounds was further examined in depth with stability and molecular
3

modeling studies. Moreover, in silico pharmacokinetic (ADMET) profiles of the most potent HNE
inhibitors obtained (IC50 < 1 μM) were evaluated in order to identify candidates for future in vivo
studies.
2. Results and Discussion
Two different strategies were pursued in our studies: 1) insertion of an additional nitrogen atom
at position 5 or 7 of the indazole nucleus, resulting in 5- and 7-azaindazole derivatives (1H-
pyrazolo[3,4-b]pyridine and 1H-pyrazolo[4,3-c]pyridine) (Figure 1A) and insertion at position 4 of the
7-azaindole nucleus, resulting in 5H-pyrrolo[2,3-b]pyrazines (Figure 1B); and 2) moving the nitrogen
in the pyrrolopyridine scaffold from position 7 to positions 4, 5, and 6 of the pyridine ring (Figure 1B).
Finally, we synthesized some pyrazolopyrimidines containing an exocyclic amide group instead of
the endocyclic amide function present in all of our previous compounds. In all new compounds, we
inserted those fragments and substituents that lead to the optimal biological results based on our
previous studies and those reported in the literature.
CN
4
5
B)
N
A)
6
CN
N
R
R
R1
5
O
N
N
7
N
N
N
CN
N
N
R1
N
N
X
X
O
R1
Indazole derivatives19,20
R1
N
Pyrrolo[2,3-b]pyridine derivatives21,22
R1
O
Figure 1. Indazole and pyrrolo[2,3-b]pyridine derivative modification.
2.1. Chemistry
All compounds were synthesized as reported in Schemes 1-5, and the structures were confirmed
on the basis of analytical and spectral data. To obtain the 3-substituted derivatives of the 5- and 7-
azaindazoles, we followed the procedures shown in Scheme 1 and 2. Starting from the previously
described compounds of type 1 (1a,b,23 1c,d,24 1e,f,25 1g26 and 1h ), we synthesized the final
compounds 2a-t (Scheme 1) following the indicated reaction conditions. Treatment with m-toluoyl
27
4

chloride or cyclopropanecarbonyl chloride, triethylamine in anhydrous dichloromethane at room
temperature resulted in synthesis of compounds 2a,b,l-o,r-t, while derivatives 2e-I,p,q are obtained
using sodium hydride in dry tetrahydrofuran at room temperature. Compound 2c was obtained
starting from intermediate 1b using benzyl bromide and potassium carbonate in dry acetonitrile at
reflux, whereas 2d was obtained using benzoic acid, activating reagents (OHBt and DCC) and
triethylamine in dry tetrahydrofuran.
aSCHEME 1
R3
R3
5
5
a
N
7
N
N
N
N
N
7
H
X
1a-h
R1
2
a-t
1
a
b
c
d
e
f
R
CF
CN
COOEt
COOiPr
CF
CN
COOEt
CH
3
N
7
7
7
7
5
5
5
5
2
a
b
c
d
e
f
g
h
i
N
R
CF
CF
CN
CN
CN
CN
CN
CN
CN
3
X
R
1
3
-Ph
3
7
7
7
7
7
7
7
7
7
7
7
5
5
5
5
5
5
5
3
CO m-CH
CO cC
CH
CO
CO m-CH
CO p-CH -Ph
3
CO
CO
CO
3
3
H
5
2
Ph
Ph
3
3
-Ph
g
h
cC
cC
3
H
H
5
3
5
9
cC
6
H
11
l
COOEt CO m-CH
COOiPr CO m-CH
CF
CF
CN
CN
COOEt CO m-CH
COOEt CO cC
CH CO m-CH
3
3
3
-Ph
-Ph
-Ph
m
n
o
p
q
r
3
CO m-CH
CO cC
CO m-CH
CO cC
3
3
H
5
3
-Ph
3
H
5
3
-Ph
s
t
3
H
5
3
3
-Ph
aReagents and Conditions: a) for 2a,b,l-o,r-t: R-COCl, Et
c: Benzyl bromide, anhydrous CH CN, K CO , reflux, 4 h; for 2d: Ph-COOH, dry THF, HOBt, Et
C, 30’, then r.t., 48 h; for 2e-i,p,q: dry THF, NaH, 0 °C, 30’, then R-COCl, r.t., o/n.
, 0 °C, 2 h, then r.t., 2 h; for
N, DCC, 0
3
2 2
N, anhydrous CH Cl
2
°
3
2
3
3
Scheme 2 shows synthesis of the pyrazolo[3,4-b]pyridine derivatives containing a nitro (4), an
amino (6), or an amide group (7-9) at position 3. The 3-nitro derivative 4 was obtained through an
5

acylation reaction starting from compound 328 and following the procedure previously reported in
Scheme 1. Reduction of the nitro group in compound 3 with tin(II) chloride in acid medium, led to
28
intermediate 5 in high yield, which was then treated with 3-methylbenzoyl chloride, triethylamine in
1
,4-dioxane and DMF. According to the results reported in the literature for pyrazolo-quinolines,29
acylation selectively occurred on the amino group at position 3 under these conditions, resulting in
the final compound 8. Compound 8 was further treated with acetyl chloride under the conditions
shown in Scheme 1 to obtain the final derivative 9. Different results were observed when
intermediate 5 was reacted with cyclopropanecarbonyl chloride, since a mixture of mono-acyl 6 and
bi-acyl derivative 7 was obtained.
aSCHEME 2
NO2
NO2
a
N
N
N
N
N
N
H
4
3
O
b
O
NH2
N
NH2
N
NH
N
c
N
N
N
N
N
N
H
5
6
7
O
O
c
O
O
HN
HN
d
N
N
N
N
N
H
N
8
9
O
6

aReagents and Conditions: a) dry THF, NaH, 0 °C, 30’, then cC
H
-COCl, r.t., o/n; b) HCl 37%, SnCl
3
5
2
2
2H O,
r.t., 30’; c) Et
3
N, DMF/1,4-Dioxane, R-COCl, 50°C, 4 h; d) CH
3
COCl, Et N, anhydrous CH Cl , 0 °C, 2 h, then
3
2
2
r.t., 2 h.
Scheme 3 shows the synthetic pathway used to obtain the pyrrolopyridines 13a-f. The starting
30
31
material 10a-c (10a,b and 10c ) was treated with hydroxylamine hydrochloride and NaHCO at
3
32
high temperature, leading to the corresponding oximes 11a-c (11c ). Dehydration with POCl
3
32
resulted in the 3-CN derivatives 12a-c (12c ), which were finally acylated at position 1 with the
appropriate acyl(aroyl) chloride and triethylamine in anhydrous dichloromethane to obtain the
azaindoles 13a-f.
aSCHEME 3
OH
O
H
N
CN
CN
4
4
4
4
5
6
5
6
5
6
5
6
a
c
b
N
N
N
N
N
N
N
H
N
H
H
R1
10a-c
11a-c
12a-c
13a-f
O
1
0-12
N
4
5
6
1
a
3
N
4
4
5
5
6
6
R
1
a
b
c
m-CH
cC
m-CH
cC
m-CH
cC
3
-Ph
H
b
c
d
e
f
3
5
3
-Ph
3
H
5
3
-Ph
3
H
5
aReagents and Conditions: a) NH
OH·HCl, H
2
c) R-COCl, Et N, anhydrous CH Cl , 0 °C, 2 h, then r.t., 2 h.
2
2 3 3
O, 60 °C, 30 min; NaHCO , reflux, 4 h; b) POCl , reflux, 2 h;
3
2
The synthesis of compounds with a 5H-pyrrolo[2,3-b]pyrazine scaffold (15a,b and 19a,b) is
33
shown in Scheme 4. The final compounds 15a,b were obtained starting from intermediate 14 and
following the acylation procedures described above. To obtain compounds 19a,b, the CN group was
inserted at position 3 as follows: formylation of compound 14 with HTMA and glacial acetic acid
33
(
16 ), transformation into the corresponding oxime (17), and dehydration to compound 18, which
7

was subjected to acylation. Unfortunately, compounds 15a and 19a, which contained a m-toluoyl
fragment at position 1, were not chemically stable and quickly turn into the precursors 14 and 18 at
room temperature, respectively.
aSCHEME 4
N
N
N
a
1
5, 19
a
b
R
N
N
3
m-CH -Ph
cC H
3 5
N
H
14
R
1
5a,b
O
b
OH
O
H
N
CN
CN
N
N
N
N
N
N
c
d
a
N
N
N
N
N
N
H
H
H
R
16
18
19a,b
17
O
aReagents and Conditions: a) R-COCl, Et
CH COOH, reflux, 2 h; c) NH OH·HCl, H O, 60 °C, 30 min; NaHCO
, 0 °C, 2 h, then r.t., 2 h; b) HTMA, glacial
, reflux, 4 h; d) POCl , reflux, 2 h.
3
2 2
N, anhydrous CH Cl
3
2
2
3
3
Scheme 5 shows the steps leading to synthesis of the pyrazolopyrimidine derivatives 22a-h.
Part a of Scheme 5 shows the synthesis of pyrazolopyrimidines containing the amide function at
position 6, and Part b shows synthesis of the final compounds 22g,h displaying the amide group at
34
35
position 3 of the bicyclic scaffold. Starting with compounds 20a,b and 20d , which were previously
described, or compound 20c, which was obtained from 20b through chlorination with N-
chlorosuccinimide (NCS) and benzoyl peroxide in tetrachloromethane at reflux, the 6-carbethoxy
(20a-c) and 3-carbethoxy (20d) derivatives were hydrolyzed using 6N NaOH at reflux, resulting in
the corresponding acids 21a-d (21a,b,d36-38). These compounds were treated with
8

trichloroacetonitrile, triphenylphosphine and m-anisidine or 3-methoxybenzilamine to obtain the final
compounds 22a-h.
aSCHEME 5
a)
2
0,21
a
b
c
R
2
R
3
H
H
Cl
CH
t-Bu
t-Bu Cl
3
N
EtO
N
N
O
CH3
20c
a
for 20b
b
R3
R3
R3
N
N
N
b
c
R2
R2
H
R2
N
HO
N
N
N
EtO
N
N
N
R
O
CH3
O
CH3
O
CH3
21a-c
22a-f
20a,b
b)
O
O
O
OEt
NH
OH
R
N
N
N
b
c
H C
3
N
H C
3
N
H C
3
N
N
N
N
O
CH3
0d
O
CH3
22g,h
O
CH3
2
21d
22
R
2
R
3
R
a
CH
t-Bu
3
H
H
m-(OCH
m-(OCH
3
3
3
)-Ph
)-Ph
)-Ph
b
c
d
e
f
t-Bu Cl m-(OCH
CH
t-Bu
3
H
H
m-(OCH
m-(OCH
3
3
3
)-Bz
)-Bz
)-Bz
t-Bu Cl m-(OCH
g
h
-
-
-
-
m-(OCH
m-(OCH
3
)-Ph
)-Bz
3
aReagents and Conditions: a) NCS, CCl
dry DCM, r.t., 4 h then appropriate amine (m-anisidine or 3-methoxybenzylamine), Et
, (PhCOO)
, reflux, 12 h; b) NaOH 6N, reflux, 2 h; c) CCl
N, r.t., 12 h.
4
2
3 3
CN, PPh ,
3
9

2.2 . Biological evaluation and structure-activity relationship (SAR) analysis
All compounds were evaluated for their ability to inhibit HNE, and the results are presented in
Tables 1−3, together with data from representative reference compounds from the previous series
20
21
21
of N-benzoylindazoles (A ), 7-azaindoles (B and C ), and the drug Sivelestat. Keeping the cyano
group that is present in the reference indazole A at position 3 of the 7-azaindazole compounds, we
introduced various substituents at position 1 (Table 1). The introduction of a phenyl (2d) or
cyclopentyl (2h) ring resulted in retention of HNE inhibitory activity (IC50 = 140 and 160 nM,
respectively), whereas introduction of a cyclohexyl ring resulted in loss of activity for compound 2i.
The 7-aza analogue of compound A displaying a m-toluoyl fragment at 1-N (2e), was 47 fold less
potent than A (2e: IC = 330 nM versus A: IC = 7 nM). Isomer 2f obtained by the shift of the methyl
50
50
group from the meta to the para position was even less active (IC = 1.5 µM). The best compound
50
of the 3-CN derivatives was 2g, which contains a cyclopropyl group at the amide function (IC = 34
50
nM). Elimination of the amide function led to complete loss of HNE inhibitory activity (compound 2c),
confirming the importance of this group in the catalytic mechanism. Replacement of the cyano group
at position 3 with various groups or functions selected from our previous publications or from the
literature also gave interesting results. Ester (2l,m), amine (6), amide (7, 9) derivatives were only
moderately active, with HNE inhibition in the micromolar range. In contrast, introduction of the
trifluoromethyl (2a,b) or the nitro group (4) led to potent HNE inhibitors, with IC values comparable
50
to or better than that of Sivelestat (2b, IC50 = 50 nM; 4, IC50 = 21 nM). Thus, the most active
compounds in the 7-azaindazole series were 2b, 2g and 4, which contain common elements,
including a group cyclopropyl at the amide function and the presence of a strong electron
withdrawing group at position 3 (i.e., CN, CF , or NO ).
3
2
Analysis of the 5-azaindazole derivatives (2n-t) revealed a slightly different trend. First, the
substitutions with CF or CN groups (compound 2n-q), as well as the weaker electron withdrawing
3
ester group (2r,s), were tolerated, resulting in potent HNE inhibitors (IC = 10-87 nM). On the other
50
hand, substitution with a methyl group (2t) resulted in significant loss of activity (IC = 760 nM). In
50
addition, replacement with an m-toluoyl fragment resulted in compounds with the highest activity, as
is evident when comparing these pairs: 2n/2o, 2p/2q, and 2r/2s, which was different from the trend
1
0

observed with the 7-azaindazoles. Compound 2p, which is the 5-aza analogue of A, is the most
potent compound of this series, had comparable activity with A (IC = 10 nM).
50
Table 1. HNE inhibitory activity of compounds 2a-t, 4, 6, 7, and 9.
R3
5
N
7
N
N
X
R1
2
a-t
4
, 6, 7, 9
a
Compound
N
7
7
7
7
7
7
7
7
7
7
7
5
5
5
5
5
5
5
7
7
7
7
R3
CF3
CF3
CN
CN
CN
CN
CN
CN
X
R1
IC (μM)
50
2a
2b
2c
2d
2e
CO
CO
CH2
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
m-CH -Ph
0.42 ± 0.12
0.050 ± 0.01
3
cC H
3
5
b
Ph
Ph
N.A.
0.16 ± 0.05
0.33 ± 0.03
1.5 ± 0.051
0.034 ± 0.012
0.14 ± 0.04
1.1 ± 0.22
m-CH -Ph
3
2f
p-CH -Ph
3
2
2
g
h
cC H
3
5
cC H
5
9
2i
CN
cC H
6
11
2l
COOEt
COOiPr
CF3
CF3
CN
m-CH -Ph
2.0 ± 0.44
3
2m
m-CH -Ph
0.98 ± 0.31
0.033 ± 0.011
0.087 ± 0.021
0.010 ± 0.003
0.079 ± 0.023
0.016 ± 0.005
0.069 ± 0.026
0.760 ± 0.14
0.021 ± 0.002
9.9 ± 1.3
3
2
2
2
2
n
o
p
q
m-CH -Ph
3
cC H
3
5
m-CH -Ph
3
CN
cC H
3
5
2r
COOEt
COOEt
CH3
NO2
NH2
m-CH -Ph
3
2s
cC H
3
5
2t
m-CH -Ph
3
4
6
7
9
cC H
3
5
5
5
cC H
3
NHCO-cC H
NHCO-m-CH -Ph
cC H
1.5 ± 0.14
25.2 ± 1.3
3
5
3
CH3
3
Sivelestat
0.050 ± 0.020
0.007 ± 0.0015
20
A
-
CN
CO
m-CH -Ph
3
b
aIC50 values are presented as the mean ± SD of three independent experiments. N.A.: no inhibitory activity
was found at the highest concentration of compound tested (50 μM).
1
1

Modification of the azaindoles 13a-f, which are isomers of the 7-azaindoles B and C and the
pyrrolopyrazines of type 15 and 19, led to the formation of highly active derivatives (Table 2).
Specifically, movement of the nitrogen in the pyridine ring from position 7 of B and C to position 4 or
5
resulted in compounds with IC values of 14-97 nM, with 13c being the best term of this series. In
50
contrast, shift of the nitrogen to position 6 (compounds 13e and 13f) led to reduced activity compared
to the reference compounds B and C, indicating that the nitrogen in this position does not interact
well with the target. The pyrrolopyrazine 19a, which differs from the reference compound C by the
introduction of a second nitrogen at position 4, retained high HNE inhibitory activity (IC = 56 nM),
50
whereas elimination of the CN at position 3 (compound 15) resulted in reduced activity by two orders
of magnitude.
Table 2. HNE inhibitory activity of compounds 13a-f, 15b, and 19b.
CN
R1
4
N
N
5
6
N
N
N
7
R1
O
O
13a-f
15b, 19b
a
Compound
N
R1
IC (μM)
50
1
1
1
1
1
3a
3b
3c
3d
3e
4
4
5
5
6
6
-
m-CH -Ph
0.089 ± 0.034
0.023 ± 0.004
0.014 ± 0.004
0.097 ± 0.01
0.194 ± 0.053
0.183 ± 0.044
7.9 ± 2.5
0.056 ± 0.018
0.050 ± 0.020
0.015 ± 0.004
0.087 ± 0.021
3
cC H
3
5
m-CH -Ph
3
cC H
3
5
m-CH -Ph
3
1
3f
cC H
3
5
1
1
5b
9b
H
CN
-
Sivelestat
21
B
C
7
7
m-CH -Ph
3
21
cC H
3
5
aIC50 values are presented as the mean ± SD of three independent experiments.
1
2

The pyrazolopyrimidine derivatives 22a-h containing an exocyclic amide group instead of an
endocyclic amide function were all completely inactive, indicating that this modification was not
tolerated (Table 3).
Table 3. HNE inhibitory activity of compounds 22a-h.
O
NH
R3
R
N
N
R2
H
N
H C
3
N
N
N
N
R
O
CH3
O
CH3
22a-f
22g,h
a
Compound
R
R2
CH3
t-Bu
t-Bu
CH3
t-Bu
t-Bu
-
R3
H
H
Cl
H
H
Cl
-
IC (μM)
50
b
22a
22b
22c
22d
22e
m-(OCH )-Ph
m-(OCH )-Ph
m-(OCH )-Ph
m-(OCH )-Bn
m-(OCH )-Bn
m-(OCH )-Bn
m-(OCH )-Ph
m-(OCH )-Bn
N.A.
3
b
N.A.
3
b
N.A.
3
b
N.A.
3
b
N.A.
3
b
22f
N.A.
3
b
2
2
2g
2h
N.A.
3
b
-
-
N.A.
3
Sivelestat
0.050 ± 0.020
aIC50 values are presented as the mean ± SD of three independent experiments. N.A.: no inhibitory activity
b
was found at the highest concentration of compound tested (50 μM).
The 12 most potent HNE inhibitors were evaluated for their chemical stability in aqueous buffer.39
Spontaneous hydrolysis rates and reaction orders of the inhibitors were measured in phosphate
buffer at pH 7.4 and 20 °C using spectrophotometry to detect compound hydrolysis. As examples,
Figure 2 shows that the absorbance maxima at 305 nm decreased over time for compound 4, while
a new peak appeared at 255 nm.
1
3

0
0
0
0
0
0
0
0
0
.45
.40
.35
.30
.25
.20
.15
.10
.05
0
15
min
Compound 4
3
0
45
6
9
1
0
0
20
1
80
2
40
00
3
3
60
0
min
2
50
275
300
325
350
Wavelength (nm)
Figure 2. Analysis of changes in compound absorbance resulting from spontaneous hydrolysis of 4. The
changes in absorbance spectra of the compounds (20 μM in 0.05 M phosphate buffer, pH 7.4, 20 °C) were
monitored over time in solution.
As shown in Table 4, compounds 2b, 2o, 2s, and 13b were the most stable with t1/2 > 4 h. We
found that compounds containing a cyclopropyl group at position 1 (compounds 2s and 13b) were
more stable when compared to the analogs with a m-toluoyl substituent (compounds 2r and 13a,
respectively). Conversely, the azaindazole derivatives (series 2 with 3 nitrogen atoms in the
heterocycle) were more stable than the azaindoles and indazoles (2 nitrogen atoms in the
heterocycle). Moreover, the most stable compound 2b had a CF at position 3 (compared to
3
compounds 2q and 2s with CN and COOEt groups, respectively). Noticeably, acceptor substituents
at position 3 exhibiting mesomeric effects with respect to the heterocyclic moiety (CN and NO ) were
2
present in mostly unstable or moderately unstable compounds (e.g., 2p, 2q, 4, 13a, 13c, 19b). This
observation is in agreement with amide hydrolysis according to the well-known acyl substitution
40
mechanism, which includes a nucleophilic attack on the carbonyl carbon atom in the rate-limiting
1
4

step. In addition, it should be mentioned that N-acyl derivatives of azoles are highly reactive amides,
as compared to other compounds containing an amide function.41,42
Table 4. Half-life (t ) for the spontaneous hydrolysis of selected derivatives. LUMO energies of the
1/2
compounds calculated by DFT B3LYP/6-31+G(d,p).
Compound
t (min)
E(LUMO), eV
-2.376
-2.547
-2.386
-2.819
-2.743
-2.536
-2.418
-3.374
-2.414
-2.242
-2.484
-2.604
1/2
2b
2n
2o
2p
2q
880.0 ± 77.5
106.6 ± 2.4
386.1 ± 43.2
12.2 ± 4.0
79.0 ± 15.6
147.0 ± 12.3
436.8 ± 50.4
93.8 ± 7.6
2r
2s
4
13a
13b
13c
19b
76.6 ± 1.8
268.3 ± 29.3
163.3 ± 8.1
100.6 ± 5.2
In general, it was difficult to find a clear relationship between hydrolytic activity and electronic or
steric effects of the heterocyclic moieties and substituents within the series of the investigated
compounds. However, we found that the observed hydrolytic instability could be explained by the
energies of the lowest unoccupied molecular orbital (LUMO) calculated for the optimized geometries
of compounds 2b, 2n, 2o, 2p, 2q, 2r, 2s, 4, 13a, 13b, 13c, and 19 using density functional theory
(DFT) (Table 4). The optimized structures of cyclopropyl-containing compounds, except for 13b,
have C symmetry with heterocyclic and cyclopropane groups lying in mutually perpendicular planes.
s
For toluoyl-substituted derivatives, the angle between the planes of the heterocycle and m-tolyl group
varied between 30 and 50˚. Relatively stable compounds 2b, 2o, 2s, and 13b can be distinguished
from hydrolytically unstable derivatives according to their LUMO energies E(LUMO) (Figure 1S, see
Supplementary Material). The unstable compounds had systematically lower values of E(LUMO)
and, hence, were more prone to reactions with nucleophiles. This conclusion agrees with the above-
mentioned mechanism of the base-catalyzed hydrolysis involving an attack of the hydroxide anion
on the carbonyl group. It should be noted that LUMO has a significant localization at the carbonyl
carbon atom in both stable and unstable compounds (see examples in Figure 3) and, thus, its energy
1
5

can strongly affect electrophilicity of the carbonyl group. The relationship between E(LUMO) and
hydrolytic stability is not quantitative, (i.e., LUMO energy is not tightly correlated with t1/2 or rate
constant k‘). Clearly, reactivity is determined not just by the electronic structure of a starting
compound but is also highly dependent on properties and stability of the tetrahedral intermediate
formed on the addition of the hydroxide ion to the carbonyl group in the rate-limiting step.
Classification of stable and unstable compounds according to other calculated characteristics,
such as energy of the highest occupied molecular orbital (HOMO), LUMO-HOMO energy gap, or
some local indices of reactivity (e.g., atomic charges, Wiberg index for the amide bond) did not lead
to a separation similar to the result presented in Figure 1S for LUMO energies.
Figure 3. Surfaces of LUMO (isovalue 0.02) for compounds 2b (A) and 2r (B) calculated by the DFT method
at B3LYP/6-31+G(d,p) level.
2.3.
Molecular Modeling
We performed molecular modeling of selected compounds in the active site of HNE (1HNE
entry of Protein Data Bank) using previously reported methods.20,21 According to the literature,50
inhibitors interact with HNE via a nucleophilic attack of the Ser195 hydroxyl oxygen atom onto a
carbonyl carbon atom of a ligand, which leads to the formation of a Michaelis complex. This
interaction is accompanied by proton transfer from Ser195 to Asp102. For the docking poses
obtained, we have evaluated the geometric parameters d and α, which are important for Michaelis
1
1
6

complex formation (Figure 4). In addition, length of the proton transfer channel within the catalytic
triad from Ser195 to Asp102 through His57 was calculated using d and d distances between the
2
3
key atoms (see Methods).
X
Y
d2
O  d1
HO
O
O
H
d3
N
Asp102
N
d2
Ser195
His57
Figure 4. Interaction of a carbonyl-containing ligand XC(O)Y with the HNE catalytic triad (Ser195, His57, and
Asp102). Distance d and angle α determine conditions for the Michaelis complex formation. The value of α
was measured as an angle between C(carbonyl)···O(Ser195) axis and C=O bond. Distances d and d
influence on the proton transfer from Ser195 to Asp102. Length of the proton transfer channel was calculated
as L = min(d ) + d
1
2
3
2
3
.
Compounds 2e (IC = 330 nM) and 2p (IC = 10 nM) contain an additional nitrogen atom at
50
50
20
different positions of the fused aromatic ring compared to compound A. The docking poses of these
molecules are shown in Figure 5. Although these three molecules are isosteric, the presence of an
additional nitrogen atom in the heterocycle of compounds 2e and 2p favors arrangement of their
bicyclic fragments among polar amino acid residues versus compound A with an indazole
heterocycle. Molecule 2e forms a H-bond with Val216 via participation of the nitrogen atom at
position 7 of the heterocycle, while compound 2p is H-bonded to Ser195 with two nitrogen atoms of
the five-membered ring. In addition, strong H-bonds are formed between the carbonyl oxygen of 2p
and Gly193 and Ser195. Note that the nitrogen atom at position 5 of the heterocycle in 2p is located
near an electropositive (blue) region of the binding site surface (Figure 5D), which should, along
with the H-bonds, stabilize the molecule in the obtained docking pose.
1
7

B
A
C
D
Figure 5. Docking poses of 2e, 2p and reference compound A. Panel A. Docking poses of 2e (violet), 2p (light
blue), and reference compound A (green). Co-crystallized methoxysuccinyl-Ala-Ala-Pro-Ala chloromethyl
ketone (MSACK) is shown in yellow sticks. Residues within 4 Å from the co-crystallized ligand are visible.
Panel B. Docking pose of compound 2e. Panel C. Docking pose of compound 2p. For Panels B and C,
residues within 4 Å of the pose are visible. Panel D. Docking pose of compound 2p. The semi-transparent
surface of HNE is shown.
Compound 4 containing nitro and cyclopropyl groups is anchored in the HNE binding site due to
H-bonds with Val216 (with the participation of the nitro group), Gly193, and Ser195 (with the
participation of the amide oxygen atom) (Figure 6A). Compound 2r has two potential centers for
nucleophilic attack by Ser195 (i.e., carbonyl oxygen atoms of the amide and ester groups). H-bonds
with Gly193 formed with participation of the amide oxygen atom and one of the nitrogen atoms in
the five-membered ring play a role in anchoring 2r inside the binding site (Figure 6B). The d1
distances O=C···O(Ser195) in the pose obtained for compound 2r are 2.946 and 3.817 Å for the
carbonyl carbon atoms of the ester and amide groups, respectively (Table 5). From this point of
1
8

view, nucleophilic attack on the ester group is more likely. In addition, the angle α in this case is
6.7˚, being in the range of 80-120˚, which is favorable for formation of the Michaelis complex,44-46
8
which is in contrast to α = 132.9˚ obtained for the amide C=O bond.
A
B
Figure 6. Docking pose of compound 4 (Panel A) and 2r (Panel B). For both panels, residues within 4 Å of
the pose are visible.
A comparative analysis of molecules containing one nitrogen atom in the five-membered ring
21
(
13a, 13c, 13e, and B ) shows that they have a somewhat similar orientation within the binding site
Figure 7A). Both heterocyclic and m-methylbenzoyl groups were located approximately in the same
(
corresponding regions of space inside the enzyme. Some features of their interactions with the HNE
amino acids should be mentioned and are illustrated in Figure 7. Compounds of this series form H-
bonds with Gly193 involving participation of the amide oxygen atom (13e and 13c), or with Ser195
involving participation of the oxygen and nitrogen atoms of the amide function (13e and 13a). In
addition, compound 13a is H-bonded to Val216 via the nitrogen atom at position 4 of the heterocycle.
A comparison of the docking poses of compounds 13c and 2p, of which the latter has an additional
nitrogen atom in the five-membered ring instead of the CH group, showed a slight difference in
orientation of the molecules within the binding site (not shown). Although positions of cyano and
1
9

carbonyl groups of these compounds are close in space, the heterocyclic and m-toluoyl fragments
do not mutually coincide. Obviously, the difference in the arrangement of 2p and 13c docking poses
is due to peculiarities of the 2p H-bonding pattern (Figure 7, panel C).
A
B
D
C
Figure 7. Docking poses of 13e, 13a, 13c, and reference compound B. Panel A. Docking poses of compounds
B (magenta), 13e (dark green), 13a (light blue), and 13c (orange). Co-crystallized MSACK is shown in yellow
sticks. Residues within 4 Å of the co-crystallized ligand are visible. Panel B. Docking pose of compound 13e.
Panel C. Docking pose of compound 13a. Panel D. Docking pose of compound 13c. For Panels B-D residues
within 4 Å of the pose are visible.
2
0

Table 5 shows the geometric parameters (distance d and angle α) for the docking poses, as
1
well as the distances d and d between the key atoms of the catalytic triad of Ser195, His57, and
2
3
Asp102 (Figure 4), which determine the length L of the proton transfer channel. Values of α for most
o
of the compounds evaluated were between 80 to 120 and are optimal for formation of the Michaelis
complex. The exceptions were the less active compounds 2e and 13e. The pose of another
o
compound with reduced activity (13a) is characterized by an angle α close to 80 ; however, the
proton transfer channel L for this molecule has a rather high value due to mutual displacements of
the residues of the catalytic triad upon binding of 13c to HNE.
Table 5. HNE inhibitory activity of selected compounds and geometric parameters of the enzyme–
inhibitor complexes predicted by molecular modelinga
Compound
IC (nM)
휶 (˚)
d1
d2
d3
L
5
0
Å
2
e
330
21
76.3
85.2
89.8
86.7
64.9
79.7
140.0
3.620
3.298
3.001
2.946b
3.622
3.256
4.324
2.196, 3.606
2.349, 3.883
1.790, 3.334
1.769, 3.340
1.857, 3.396
4.852, 5.345
1.790, 3.334
3.291
3.294
2.826
2.856
2.872
2.478
2.815
5.487
5.643
4.616
4.625
4.729
7.330
4.605
4
2p
10
2
r
16
13e
13a
13c
194
89
14
aGeometric parameters correspond to the formation of Michaelis complex with the ester carbonyl group.
bThe indicated d
value corresponds to the carbonyl carbon atom of the ester function of compound 2r. For
amide carbon atom of this compound, the d distance equals 3.817 Å.
1
1
2.4. ADMET assessment
To facilitate selection of lead compounds and to further support the molecular modelling and in
vitro evaluation, we also performed an in silico absorption, distribution, metabolism, and excretion-
toxicity (ADMET) pharmacokinetics evaluation. The in silico assessment was generated through the
evaluation of pharmacokinetic profiles and possible adverse side effects for compounds 2a-t, 4, 6,
7, 9, 13a-f, 15, and 19. ADMET molecular studies were conducted using SwissADME
2
1

(
http://swissadme.ch)47 and pkCSM (http://biosig.unimelb.edu.au/pkcsm/)48 (see Supplemental
Tables 1S-10S). Most of the compounds were predicted to be orally available, with high
gastrointestinal absorption and high water solubility (Table 2S, 3S). Compounds with predicted
-5
moderate solubility (10 mol/L range) were 2a, 2i, 2l, 2m, 2n, and 2r. Only compounds 2o, 6 and 7
were predicted to be P-glycoprotein substrates, whereas most of the compounds were predicted to
be inhibitors of the CYP1A2. Some compounds were also predicted to inhibit CYP2C19/CYP2C9,
but no inhibition was predicted for CYP3A4 and CYP2D6. Otherwise, most of the compounds were
predicted to be CYP3A4 substrates but should not be metabolized by CYP2D6 (Table 7S).
Interestingly, none of the compounds violated the Lipinski rule of 5, nor did they violate the other
drug-likeness rules (Ghose, Egan, Veber, and Muegee).49-53
The absorption and distribution calculated parameters are shown in an Egan–Egg model (Figure
8
) (Brain or IntestinaL EstimateD, BOILED-Egg). The Egan–Egg model highlights that most of the
compounds were predicted to passively permeate the blood-brain barrier, whereas compounds 2g,
q, 2s, 4, 6, 7, and 19 are only able to be passively absorbed by the gastrointestinal tract pkCSM
calculated absorption properties showed > 95 % intestinal absorption (aside from molecule 6 with
7.3%), due to the calculated optimal (> 0.90) Caco-2 cell permeability (Table 5S). Moreover, most
2
7
of the compounds are predicted to be adsorbed by the skin (exploitable for transdermal drug delivery)
as shown by the Log Kp < -2.5 (see skin permeability, Table 5S).
The calculated values of steady state volume of distribution are relatively low for some
compounds (Log VDss < -0.15, Table 6S), but most of the compounds were predicted to have a
significant unbound fraction in the plasma, thus becoming available for interaction with the
pharmacological target (Table 6S). The calculated values of the total clearance (Table 8S) indicate
that most of the compounds were predicted to have good renal elimination (0.20–0.84 mL/min/kg),
and no compounds were predicted to be substrates of the renal organic cation transporter 2. pkCSM
calculated toxicity properties indicated concerns about the AMES test, the T. pyriformis toxicity and
hepatoxicity. On the other hand, the compounds did not have skin sensitization properties and were
not predicted to be inhibitors of hERG I and II (Table 9S). Due to the toxicology concerns indicated
by
pkCSM,
toxicology
was
also
evaluated
with
Datawarrior
(v.
5.2.1
2
2

http://www.openmolecules.org/datawarrior/). In this case, all compounds were predicted to be non-
mutagenic, non-tumorigenic, and not irritants. A low risk for reproductive system effects was
associated with molecules 2l-n and 2r,s (Table 10S). In silico toxicity evaluations were also
performed
with
preadmet
(https://preadmet.bmdrc.kr/toxicity/)
and
admetsar1
(http://lmmd.ecust.edu.cn/admetsar1). The results for preadmet are presented in Table S11, while
the results for admetsar1 are presented in Tables 12S-41S. Preadmet considered most of the
compounds mutagenic in the same test, whereas admetsar1 considered most of the compounds as
nontoxic in the same test. Despite this result, most of the compounds were considered non-
carcinogenic for mouse or rat, with a medium to low risk for their hERG inhibition in both in silico
evaluations. Further biological evaluation is needed due to the somewhat contrasting test results,
but the ADMET general profile was good for our set of compounds, particularly the most potent
compounds 2p, 2r, 4, 13b, and 13c, which had excellent ADMET profiles and shouldn't have any
particular toxicity issues. The IC50 measured potency, together with the ADMET profile will be of
fundamental importance for the future selection of compounds for in vivo evaluation.
Figure 8. BOILED-Egg plot. Points located in the BOILED-Egg’s yellow are the compounds predicted to
permeate the BBB passively, differently the ones in the white are the molecules predicted to be only passively
absorbed by the gastrointestinal tract. Blue dots indicate molecules expected to be refluxed from the central
nervous system (CNS) by the P-glycoprotein, whereas the red ones are not transported by the P-glycoprotein.
3. Experimental section
2
3

3.1.
Chemistry
All melting points were determined on a Büchi apparatus (New Castle, DE) and are uncorrected.
Extracts were dried over Na SO , and the solvents were removed under reduced pressure. Merck
2
4
F-254 commercial plates (Merck, Durham, NC) were used for analytical TLC to follow the course of
reactions. Silica gel 60 (Merck 70-230 mesh, Merck, Durham, NC) was used for column
chromatography. H-NMR and 13C-NMR spectra were recorded on an Avance 400 instrument
1
(Bruker Biospin Version 002 with SGU, Bruker Inc., Billerica, MA). Chemical shifts (δ) are reported
in ppm to the nearest 0.01 ppm using the solvent as an internal standard. Coupling constants (J
values) are given in Hz and were calculated using TopSpin 4.0.8 software (Nicolet Instrument Corp.,
Madison, WI) and are rounded to the nearest 0.1 vHz. High resolution mass spectrometry (HRMS)
analysis was performed with a Thermo Finnigan LTQ Orbitrap mass spectrometer equipped with an
electrospray ionization source (ESI). The solvents used in HRMS were acetonitrile (Chromasolv
grade) purchased from Sigma-Aldrich (Milan, Italy) and mQ water 18 MΩ cm, obtained from
Millipore’s Simplicity system (Milan, Italy). The accurate mass measure was performed by
-1
introducing sample solution (1.0 g mL in mQ water: acetonitrile 50:50) via syringe pump at 10 L
-1
min , and the signal of the positive ions was acquired. The experimental conditions allowed
+
monitoring of the protonated molecules ([M+H] species) with a proper dwell time to achieve 60.000
units of resolution at full width at half maximum (FWHM). Microanalyses indicated by the symbols of
the elements or functions were performed with a Perkin–Elmer 260 elemental analyzer (PerkinElmer,
Inc., Waltham, MA) for C, H, and N, and the results were within ±0.4% of the theoretical values,
unless otherwise stated. Reagents and starting material were commercially available.
3.1.1. General procedure for compounds (2a,b, 2l-o, and 2r-t). To a cooled (0 °C) suspension of
17
18
19
20
21
the appropriate substrate 1a, 1c,d, 1e, 1g, or 1h (0.32 mmol) in anhydrous CH Cl (2 mL),
2
2
0
.64 mmol of Et N and 0.96 mmol of the appropriate acyl/aroyl chloride were added. The mixture
3
was stirred at 0 °C for 2 h and then at room temperature for an additional 2 h. The solvent was
evaporated, cold water was added, and the mixture was neutralized with 0.5 N NaOH. The reaction
mixture was extracted with CH Cl (3 x 15 mL), the solvent was dried over sodium sulfate and
2
2
2
4

evaporated in vacuum. The final compound 2o was purified by crystallization with ethanol while the
compounds 2a,b,l-n,r-t were purified by flash column chromatography using cyclohexane/ethyl
acetate 3:1 (2a,b,n) or 1:1 (2s,t), hexane/ethyl acetate 5:2 (2l,m), dichloromethane/methanol 9:1
(2r) as eluents. Finally, all compounds were crystallized from ethanol, and melting points were
performed after crystallization.
m-Tolyl-(3-(trifluoromethyl)-1H-pyrazolo[3,4-b]pyridin-1-yl)methanone (2a). Yield = 26%; mp =
1
6
7-69 °C (white powder). H-NMR (400 MHz, CDCl ) δ 2.43 (s, 3H, m-CH -Ph), 7.38-7.49 (m, 3H,
3
3
Ar), 7.80 (d, 1H, Ar, J = 7.2 Hz), 7.85 (s, 1H, Ar), 8.26 (d, 1H, Ar, J = 8.0), 8.89 (d, 1H, Ar, J = 3.2).
13
C-NMR (100 MHz, CDCl ) δ 21.3 (CH ), 117.7 (C), 121.0 (CH), 122.8 (C), 128.1 (CH), 128.8 (CH),
3
3
129.7 (CH), 130.4 (C), 131.9 (CH), 134.4 (CH), 138.2 (C), 142.1 (C), 152.1 (CH), 159.1 (C), 165.0
+
(
C). ESI-HRMS (m/z) calculated for [M+H] ion species C H F N O= 306,0849, found 306,0847.
15
11
3
3
Anal. C H F N O (C, H, N).
15
10
3
3
Cyclopropyl(3-(trifluoromethyl)-1H-pyrazolo[3,4-b]pyridin-1-yl)methanone (2b). Yield = 23%;
1
oil. H-NMR (400 MHz, CDCl ) δ 1.16-1.23 (m, 2H, CH cC H ), 1.38-1.44 (m, 2H, CH cC H ), 3.31-
3
2
3
5
2
3
5
13
3
.36 (m, 1H, COCH), 7.43-7.48 (m, 1H, Ar), 8.24 (d, 1H, Ar, J = 8.0), 8.85 (d, 1H, Ar, J = 2.8). C-
NMR (100 MHz, CDCl ) δ 11.9 (CH ), 13.6 (CH), 114.3 (C), 118.8 (C), 120.9 (CH), 121.9 (C), 129.7
3
2
+
(
CH), 151.6 (C), 152.1 (CH), 172.5 (C). ESI-HRMS (m/z) calculated for [M+H] ion species
C H F N O = 256,0692, found 256,0689. Anal. C H F N O (C, H, N).
11
9
3
3
11
8
3
3
Ethyl 1-(3-methylbenzoyl)-1H-pyrazolo[3,4-b]pyridine-3-carboxylate (2l). Yield = 30%; mp =
1
1
85-187°C (white powder). H-NMR (400 MHz, CDCl ) δ 1.47 (t, 3H, OCH CH , J = 6.8), 2.44 (s,
3
2
3
3
H, m-CH -Ph), 4.52 (q, 2H, OCH CH , J = 6.8), 7.40-7.48 (m, 3H, Ar), 7.83 (d, 1H, Ar, J = 7.2), 7.88
3
2
3
13
(
s, 1H, Ar), 8.60 (d, 1H, Ar, J = 8.0), 8.83 (d, 1H, Ar, J = 4.0). C-NMR (100 MHz, CDCl ) δ 14.3
3
(
CH ), 21.3 (CH ), 62.0 (CH ), 108.0 (C), 121.1 (CH), 128.1 (CH), 128.9 (CH), 130.5 (C), 131.8 (CH),
3
3
2
132.1 (CH), 133.0 (C), 134.3 (CH), 136.7 (C), 138.1 (C), 151.4 (CH), 159.1 (C), 165.0 (C). ESI-
+
HRMS (m/z) calculated for [M+H] ion species C H N O = 310,1186, found 310,1183. Anal.
17
16
3
3
C H N O (C, H, N).
17
15
3
3
Isopropyl 1-(3-methylbenzoyl)-1H-pyrazolo[3,4-b]pyridine-3-carboxylate (2m). Yield = 28%; mp
1
=
192-194 °C (white powder). H-NMR (400 MHz, CDCl ) δ 1.46 (d, 6H, CH(CH ) , J = 6.0), 2.44 (s,
3
3 2
2
5

3
H, m-CH -Ph), 5.43-5.48 (m, 1H, CH(CH ) ), 7.40-7.50 (m, 3H, Ar), 7.85 (d, 1H, Ar, J = 7.6), 7.90
3
3
2
1
3
(
(
(
s, 1H, Ar), 8.57 (d, 1H, Ar, J = 8.0), 8.82 (d, 1H, Ar, J = 3.2). C-NMR (100 MHz, CDCl ) δ 21.3
3
CH ), 21.9 (CH ), 70.0 (CH), 116.9 (C), 121.0 (CH), 128.3 (CH), 129.0 (CH), 131.8 (CH), 132.1
3
3
CH), 134.2 (CH), 138.1 (C), 138.9 (C), 151.3 (CH), 153.2 (C), 160.8 (C), 166.7 (C). ESI-HRMS (m/z)
+
calculated for [M+H] ion species C H N O = 324,1343, found 324,1346. Anal. C H N O (C, H,
18
18
3
3
18 17
3
3
N).
m-Tolyl(3-(trifluoromethyl)-1H-pyrazolo[4,3-c]pyridin-1-yl)methanone (2n). Yield = 66%; mp =
1
8
0-81 °C (colorless solid). H-NMR (400 MHz, CDCl ) δ 2.46 (s, 3H, m-CH -Ph), 7.41-7.49 (m, 2H,
3
3
13
Ar), 7.92 (d, 2H, Ar, J = 6.4), 8.42 (d, 1H, Ar, J = 5.6), 8.80 (d, 1H, Ar, J = 6.0), 9.29 (s, 1H, Ar). C-
NMR (100 MHz, CDCl ) δ 21.4 (CH ), 110.4 (CH), 116.6 (C), 118.9 (C), 121.7 (C), 128.2 (CH), 128.8
3
3
(CH), 130.7 (C), 131.9 (CH), 134.5 (CH), 138.3 (C), 144.2 (CH), 145.0 (C), 148.4 (CH), 167.6 (C).
+
ESI-HRMS (m/z) calculated for [M+H] ion species C H F N O = 306,0849, found 306,0846. Anal.
15
11
3
3
C H F N O (C, H, N).
15
10
3
3
Cyclopropyl(3-(trifluoromethyl)-1H-pyrazolo[4,3-c]pyridin-1-yl)methanone (2o). Yield = 30%;
1
mp = 93-94 °C (colorless solid). H-NMR (400 MHz, CDCl ) δ 1.25-1.30 (m, 2H, CH cC H ), 1.39-
3
2
3
5
1
5
1
.44 (m, 2H, CH cC H ), 3.18-3.24 (m, 1H, COCH), 8.37 (d, 1H, Ar, J = 6.0), 8.75 (d, 1H, Ar, J =
2 3 5
13
.6), 9.28 (s, 1H, Ar). C-NMR (100 MHz, CDCl ) δ 12.0 (CH ), 12.9 (CH), 110.1 (CH), 119.0 (C),
3
2
+
21.7 (C), 143.4 (C), 144.1 (CH), 148.3 (CH), 174.3 (C). ESI-HRMS (m/z) calculated for [M+H] ion
species C H F N O = 256,0692, found 256,0693. Anal. C H F N O (C, H, N).
11
9
3
3
11
8
3
3
Ethyl 1-(3-methylbenzoyl)-1H-pyrazolo[4,3-c]pyridine-3-carboxylate (2r). Yield: 49%; mp = 87-
1
8
9 °C (brown solid). H-NMR (400 MHz, CDCl ) δ 1.48 (t, 3H, CH CH , J = 7.2), 2.44 (s, 3H, m-CH
3
3
2
3-
Ph), 4.55 (q, 2H, CH CH , J = 7.2), 7.39-7.46 (m, 2H, Ar), 7.94 (d, 2H, Ar, J = 7.2), 8.36 (d, 1H, Ar,
3
2
13
J = 6.0), 8.74 (d, 1H, Ar, J = 5.6), 9.58 (s, 1H, Ar). C-NMR (100 MHz, CDCl ) δ 14.3 (CH ), 21.4
3
3
(
CH ), 62.3 (CH ), 110.1 (CH), 120.8 (C), 128.3 (CH), 129.0 (CH), 131.1 (C), 132.1 (CH), 134.4 (CH),
3 2
138.3 (C), 141.0 (C), 145.3 (C), 146.3 (CH), 147.8 (CH), 160.8 (C), 168.0 (C). ESI-HRMS (m/z)
+
calculated for [M+H] ion species C H N O = 310,1186, found 310,1182. Anal. C H N O
17
16
3
3
17 15
3
3
(C,H,N).
2
6

Ethyl 1-(cyclopropanecarbonyl)-1H-pyrazolo[4,3-c]pyridine-3-carboxylate (2s). Yield: 44%; mp
1
=
94-96 °C (white powder). H-NMR (400 MHz, CDCl ) δ 1.23 (m, 2H, CH cC H ), 1.36 (m, 2H, CH
3
2
3
5
2
cC H ), 1.50 (t, 3H, CH CH , J = 7.2), 3.25-3.31 (m, 1H, CH cC H ), 4.57 (q, 2H, CH CH , J = 7.2),
3
5
2
3
3
5
2
3
13
8
.25 (d, 1H, Ar, J = 5.6), 8.67 (d, 1H, Ar, J = 5.6), 9.52 (s, 1H, Ar). C-NMR (100 MHz, CDCl ) δ 8.8
3
(
CH), 12.2 (CH ), 14.1 (CH ), 60.9 (CH ), 105.3 (C), 113.7 (CH), 126.2 (C), 136.7 (C), 143.1 (CH),
2 3 2
+
1
48.0 (CH), 160.6 (C), 172.3 (C). ESI-HRMS (m/z) calculated for [M+H] ion species C H N O =
13 14 3 3
2
60,1030, found 260,1030. Anal. C H N O (C,H,N).
13
13
3
3
(
(
(
3-Methyl-1H-pyrazolo[4,3-c]pyridin-1-yl)(m-tolyl)methanone (2t). Yield = 43%; mp = 104-105 °C
1
white powder). H-NMR (400 MHz, CDCl ) δ 2.45 (s, 3H, m-CH -Ph), 2.67 (s, 3H, 3-CH ), 7.39-7.45
3
3
3
m, 2H, Ar), 7.89 (d, 2H, Ar, J = 8.8), 8.38 (d, 1H, Ar, J = 6.6), 8.70 (d, 1H, Ar, J = 5.6), 9.11 (s, 1H,
13
Ar). C-NMR (100 MHz, CDCl ) δ 12.4 (CH ), 21.4 (CH ), 102.9 (C), 110.3 (CH), 122.8 (C), 127.9
3
3
3
(CH), 128.4 (CH), 131.5 (CH), 132.4 (C), 133.5 (CH), 138.9 (C), 144.0 (CH), 144.8 (C), 147.9 (CH),
+
1
48.5 (C), 168.1 (C). ESI-HRMS (m/z) calculated for [M+H] ion species C H N O = 252,1131,
15
14
3
found 252,1129. Anal. C H N O (C, H, N).
15
13
3
3.1.2. 1-Benzyl-1H-pyrazolo[3,4-b]pyridine-3-carbonitrile (2c). To a suspension of the
17
intermediate 1b (0.42 mmol) in anhydrous CH CN (2 mL), 0.50 mmol of K CO and 0.84 mmol of
3
2
3
benzyl bromide were added. The mixture was stirred at reflux for 4 h. After evaporation of the solvent,
ice-cold water (20 mL) was added and the precipitate was recovered by vacuum filtration. The final
compound 2c was purified by crystallization with ethanol. Yield = 40%; mp = 110-112 °C (white
1
powder). H-NMR (400 MHz, CDCl ) δ 5.77 (s, 2H, CH -Ph), 7.42-7.49 (m, 6H, Ar), 8.18 (d, 1H, Ar,
3
2
13
J = 8.0), 8.68 (d, 1H, Ar, J = 2.8). C-NMR (100 MHz, CDCl ) δ 52.1 (CH ), 105.0 (C), 112.9 (C),
3
2
1
2
2
17.1 (C), 119.6 (CH), 128.4 (CH), 128.4 (CH), 128.8 (CH), 135.3 (C), 150.6 (CH), 154.5 (C). IR =
-1
+
241 cm (CN). ESI-HRMS (m/z) calculated for [M+H] ion species C H N = 235,0978, found
14
11
4
35,0981. Anal. C H N (C, H, N).
14
10
4
3.1.3. 1-Benzoyl-1H-pyrazolo[3,4-b]pyridine-3-carbonitrile (2d). To a solution of intermediate
17
1
b (1.04 mmol), 2.08 mmol of Benzoic acid and 1.56 mmol of HOBt in dry THF (5 mL), 2.08 mmol
2
7

of triethylamine was added. The solution was brought to 0 °C, 1.56 mmol of N,N'-dicyclohexyl
carbodiimide (DCC) was added and stirred at 0 °C for about 15’. The reaction mixture was stirred at
room temperature for 48 h. After evaporation of the solvent, the crude compound was diluted in
CH Cl and first washed with 2M Na CO solution and then with brine. The organic phase was
2
2
2
3
recovered, dried over sodium sulfate and evaporated under vacuum. The final compound 2d was
1
purified by crystallization with ethanol. Yield = 16%; mp = 208-210 °C (white powder). H-NMR (400
MHz, CDCl ) δ 7.55-7.65 (m, 3H, Ar), 7.68 (d, 1H, Ar, J = 6.0), 8.03 (d, 2H, Ar, J = 6.4), 8.31 (d, 1H,
3
13
Ar, J = 6.8), 8.93 (s, 1H, Ar). C-NMR (100 MHz, CDCl ) δ 111.6 (C), 118.1 (C), 121.6 (CH), 122.3
3
(C), 128.4 (CH), 129.2 (CH), 131.1 (C), 131.6 (CH), 133.9 (CH), 152.7 (CH), 155.0 (C), 165.7 (C).
-1
+
IR = 2243 cm (CN). ESI-HRMS (m/z) calculated for [M+H] ion species C H N O =249,0771, found
14
9
4
2
49,0769. Anal. C H N O (C, H, N).
14 8 4
3.1.4. General procedure for compounds (2e-i, 2p, 2q). To a suspension of the appropriate
17
19
substrate 1b or 1f (0.86 mmol) in 10 mL of anhydrous THF, 1.72 mmol of sodium hydride and
1.03 mmol of the appropriate acyl/aroyl chloride were added. The mixture was stirred at room
temperature overnight. The solvent was evaporated, cold water was added, and the mixture was
neutralized with 2.5 N NaOH. The precipitate was recovered by vacuum filtration to obtain the final
compounds 2e-i,p,q, which were purified by crystallization with ethanol.
1-(3-Methylbenzoyl)-1H-pyrazolo[3,4-b]pyridine-3-carbonitrile (2e). Yield = 25%; mp = 210-213
1
°
C (white powder). H-NMR (400 MHz, CDCl ) δ 2.46 (s, 3H, m-CH -Ph), 7.41-7.55 (m, 3H, Ar), 7.77-
3
3
1
3
7
2
.82 (m, 2H, Ar), 8.31 (d, 1H, Ar, J = 8.0), 8.92 (d, 1H, Ar, J = 2.8). C-NMR (100 MHz, CDCl ) δ
3
1.7 (CH ), 105.0 (C), 110.5 (C), 114.2 (C), 119.6 (CH), 127.3 (CH), 128.4 (CH), 129.8 (CH), 130.4
3
-1
(
C), 130.7 (CH), 133.0 (C), 134.4 (CH), 138.9 (C), 150.2 (CH), 165.7 (C). IR = 2241 cm (CN). ESI-
+
HRMS (m/z) calculated for [M+H] ion species C H N O = 263,0927, found 263,0926. Anal.
15
11
4
C H N O (C, H, N).
15
10
4
1-(4-Methylbenzoyl)-1H-pyrazolo[3,4-b]pyridine-3-carbonitrile (2f). Yield = 10%; mp = 140-143
1
°
C (white powder). H-NMR (400 MHz, CDCl ) δ 2.48 (s, 3H, p-CH -Ph), 7.37 (d, 2H, Ar, J = 8.0),
3
3
7.50-7.55 (m, 1H, Ar), 7.94 (d, 2H, Ar, J = 8.0), 8.30 (d, 1H, Ar, J = 8.0), 8.92 (d, 1H, Ar, J = 3.6).
13
C-NMR (100 MHz, CDCl ) δ 21.9 (CH ), 105.0 (C), 111.7 (C), 118.0 (C), 121.5 (CH), 122.0 (C),
3
3
2
8

1
27.9 (CH), 129.2 (CH), 132.4 (CH), 145.3 (C), 151.9 (C), 152.6 (CH), 165.6 (C). IR = 2241 cm-1
+
(
CN). ESI-HRMS (m/z) calculated for [M+H] ion species C H N O = 263,0927, found 263,0927.
15
11
4
Anal. C H N O (C, H, N).
15
10
4
1-(Cyclopropanecarbonyl)-1H-pyrazolo[3,4-b]pyridine-3-carbonitrile (2g). Yield = 34%; mp =
1
1
47-149 °C (white powder). H-NMR (400 MHz, CDCl ) δ 1.24-1.29 (m, 2H, CH cC H ), 1.45-1.50
3
2
3
5
(
(
(
m, 2H, CH cC H ), 3.32-3.37 (m, 1H, COCH), 7.48-7.53 (m, 1H, Ar), 8.27 (d, 1H, Ar, J = 8.0), 8.88
2 3 5
13
d, 1H, Ar, J = 4.4). C-NMR (100 MHz, CDCl ) δ 12.3 (CH ), 13.7 (CH), 105.0 (C), 110.4 (C), 114.1
3
2
-1
C), 121.4 (CH), 129.2 (CH), 133.0 (C), 152.5 (CH), 173.6 (C). IR = 2243 cm (CN). ESI-HRMS (m/z)
+
calculated for [M+H] ion species C H N O = 213,0771, found 213,0772. Anal. C H N O (C, H, N).
11
9
4
11
8
4
1-(Cyclopentanecarbonyl)-1H-pyrazolo[3,4-b]pyridine-3-carbonitrile (2h). Yield = 32%; mp =
1
1
23-125 °C (white powder). H-NMR (400 MHz, CDCl ) δ 1.73-1.79 (m, 2H, CH cC H ), 1.80-1.85
3
2
5
9
(
m, 2H, CH cC H ), 1.98-2.08 (m, 2H, CH cC H ), 2.10-2.16 (m, 2H, CH cC H ), 4.12 (quin, 1H,
2 5 9 2 5 9 2 5 9
13
COCH, J = 8.0), 7.47-7.53 (m, 1H, Ar), 8.25 (d, 1H, Ar, J = 8.0), 8.89 (d, 1H, Ar, J = 4.0). C-NMR
(
100 MHz, CDCl ) δ 25.8 (CH ), 26.2 (CH ), 30.0 (CH ), 30.3 (CH ), 44.2 (CH), 111.7 (C), 118.3 (C),
3 2 2 2 2
-1
1
21.4 (CH), 122.2 (C), 129.1 (CH), 151.0 (C), 152.6 (CH), 174.4 (C). IR = 2243 cm (CN). ESI-
+
HRMS (m/z) calculated for [M+H] ion species C H N O = 241,1084, found 241,108. Anal.
13
13
4
C H N O (C, H, N).
13
12
4
1-(Cyclohexanecarbonyl)-1H-pyrazolo[3,4-b]pyridine-3-carbonitrile (2i). Yield = 24%; mp = 124-
1
1
27 °C (white powder). H-NMR (400 MHz, CDCl ) δ 1.30-1.50 (m, 4H, 2 x CH cC H ), 1.73-1.78
3
2
6
11
(
m, 2H, CH cC H ), 1.88 (d, 2H, CH cC H , J = 16.0), 2.05 (d, 2H, CH cC H , J = 16.0), 3.76 (t,
2 6 11 2 6 11 2 6 11
13
1
H, COCH, J = 8.0), 7.47-7.52 (m, 1H, Ar), 8.25 (d, 1H, Ar, J = 8.0), 8.89 (d, 1H, Ar, J = 4.0). C-
NMR (100 MHz, CDCl ) δ 25.3 (CH ), 25.6 (CH ), 29.0 (CH ), 43.9 (CH), 111.6 (C), 118.2 (C), 121.4
3
2
2
2
-1
(
CH), 122.2 (C), 129.1 (CH), 151.0 (C), 152.6 (CH), 171.1 (C). IR = 2243 cm (CN). ESI-HRMS (m/z)
+
calculated for [M+H] ion species C H N O = 255,1240, found 255,1243. Anal. C H N O (C, H,
14
15
4
14 14
4
N).
1-(3-Methylbenzoyl)-1H-pyrazolo[4,3-c]pyridine-3-carbonitrile (2p). Yield = 47%; mp = 192-194
1
°
C (colorless solid). H-NMR (400 MHz, CDCl ) δ 2.48 (s, 3H, m-CH -Ph), 7.46 (t, 1H, Ar, J = 7.6),
3
3
7.52 (d, 1H, Ar, J = 7.2), 7.88 (d, 2H, Ar, J = 6.8), 8.45 (d, 1H, Ar, J = 6.0), 8.84 (d, 1H, Ar, J = 6.0),
2
9