Development of Dual Chitinase Inhibitors as Potential New Treatment for Respiratory System Diseases

Article abstract of DOI:10.1021/acs.jmedchem.9b00681

Acidic mammalian chitinase (AMCase) and chitotriosidase-1 (CHIT1) are two enzymatically active proteins produced by mammals capable of cleaving the glycosidic bond in chitin. Based on the clinical findings and animal model studies, involvement of chitinases has been suggested in several respiratory system diseases including asthma, COPD, and idiopathic pulmonary fibrosis. Exploration of structure-activity relationships within the series of 1-(3-amino-1H-1,2,4-triazol-5-yl)-piperidin-4-amines, which was earlier identified as a scaffold of potent AMCase inhibitors, led us to discover highly active dual (i.e., AMCase and CHIT1) inhibitors with very good pharmacokinetic properties. Among them, compound 30 was shown to reduce the total number of cells in bronchoalveolar lavage fluid of mice challenged with house dust mite extract after oral administration (50 mg/kg, qd). In addition, affinity toward the hERG potassium channel of compound 30 was significantly reduced when compared to the earlier reported chitinase inhibitors.

Full text of DOI:10.1021/acs.jmedchem.9b00681

Subscriber access provided by UNIVERSITY OF TOLEDO LIBRARIES
Article
Development of Dual Chitinases Inhibitors as Potential
New Treatment for Respiratory System Diseases
Marzena Mazur, Barbara Dymek, Robert Koralewski, Piotr L. Sklepkiewicz, Sylwia Olejniczak,
Marcin Piotr Mazurkiewicz, Michal Czeslaw Piotrowicz, Magdalena Salamon, Karol J#drzejczak,
Agnieszka Zagozdzon, Wojciech Czestkowski, Krzysztof Matyszewski, Bart#omiej Borek,
Agnieszka Bartoszewicz, El#bieta Pluta, Aleksandra Rymaszewska, Witold Mozga, Filip Stefaniak,
Pawel Dobrza#ski, Karolina Dzwonek, Jakub Golab, Adam Golebiowski, and Jacek Olczak
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b00681 • Publication Date (Web): 10 Jul 2019
Downloaded from pubs.acs.org on July 11, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Development of Dual Chitinases Inhibitors as
Potential New Treatment for Respiratory System
Diseases
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
†
†
†
†
Marzena Mazur, Barbara Dymek, Robert Koralewski, Piotr Sklepkiewicz, Sylwia Olejniczak,
†
†, ○
†
†
Marcin Mazurkiewicz, Michał Piotrowicz, Magdalena Salamon, Karol Jędrzejczak,
†
†
†
†
Agnieszka Zagozdzon, Wojciech Czestkowski, Krzysztof Matyszewski, Bartłomiej Borek,
†
†
†
†
Agnieszka Bartoszewicz, Elżbieta Pluta, Aleksandra Rymaszewska, Witold Mozga, Filip
Stefaniak,†,# Paweł Dobrzański, Karolina Dzwonek, Jakub Golab, Adam Golebiowski, Jacek
†
†
†,§
†
Olczak*,†
†
OncoArendi Therapeutics SA, Żwirki i Wigury 101, 02-089 Warsaw, Poland
#
Laboratory of Bioinformatics and Protein Engineering, International Institute of Molecular and
Cell Biology in Warsaw, Ks. Trojdena 4, 02-109 Warsaw, Poland
§
Department of Immunology, Medical University of Warsaw, Nielubowicza 5, 02-097 Warsaw,
Poland
ABSTRACT Acidic mammalian chitinase (AMCase) and chitotriosidase-1 (CHIT1) are two
enzymatically active proteins produced by mammals and capable of cleaving glycosidic bond in
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
chitin. Based on the clinical findings and animal models studies, involvement of chitinases has
been suggested in several respiratory system diseases including asthma, COPD and idiopathic
pulmonary fibrosis. Exploration of structure-activity relationships within the series of 1-(3-amino-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1H-1,2,4-triazol-5-yl)-piperidin-4-amines, that was earlier identified as a scaffold of potent
AMCase inhibitors, led us to discovery of highly active dual (i.e. AMCase and CHIT1) inhibitors
with very good pharmacokinetic properties. Among them, compound 30 was shown to reduce the
total number of cells in bronchoalveolar lavage fluid (BALf) of mice challenged with house dust
mite (HDM) extract after oral administration (50 mg/kg, qd). In addition, affinity towards hERG
potassium channel of compound 30 was significantly reduced when compared to the earlier
reported chitinases inhibitors.
INTRODUCTION
Asthma is a chronic inflammatory disease that affects over 330 million people worldwide 1.
Despite availability of rapidly acting bronchodilators and anti-inflammatory drugs such as steroids,
asthma remains an unmet medical need due to variable pathogenesis, side-effects of current
therapeutics, resistance to treatment developing in subsets of patients, and poor patients’
2
compliance to inhaled drugs . There is compelling evidence implicating type 2 immune response
3
in development of asthma pathogenesis . A chronic pulmonary inflammation leads to
development of airway hyperresponsiveness, airways remodeling and fibrosis. Remarkably,
increasing number of studies report changes in the expression and activity of chitinases in patients
with asthma 4-7 and other chronic lung diseases 8-15. Chitinases are chitin-degrading enzymes that
2
ACS Paragon Plus Environment

Page 3 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
belong to the 18 glycosyl hydrolase family, which consists of two enzymatically active molecules
(chitotriosidase [CHIT1] and acidic mammalian chitinase [AMCase]) as well as several chitinase-
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
like proteins, that lack enzymatic activity 4, 7, 12, 16-19. Chitinases were initially hypothesized to play
a significant role in immune response modulation in asthma, owing to the observation that
predominant aeroallergen sources, such as house dust mites, cockroaches and fungi, are chitin-
containing organisms 20. However, several reports have clearly shown their critical role in
facilitating immune response and tissue repair processes in chitin-independent manner. For
example, both enzymes were shown to actively participate in interleukin-13-driven development
of type 2 immune response in asthma 7, 16, 21 or tissue repair in pulmonary fibrosis 15 in animal
models induced by non-chitin containing insults, such as ovalbumin and bleomycin.
AMCase is produced by lung epithelial cells and macrophages at sites of Th2 inflammation in
animal models of asthma, induced by HDM, Aspergillus sp., ovalbumin (OVA), and IL-13 7, 20, 22
.
1
6
Moreover, this enzyme is believed to be the dominant true chitinase in mouse lungs . AMCase
expression and activity was mostly described to modulate type 2 immune response in mouse
models of asthma 7, 16, 17, 20-23 and was linked to human asthma pathogenesis 5, 7. However, the
increase in overall chitinolytic activity observed in the lungs of patients with chronic and
interstitial lung diseases has recently been mostly attributed to CHIT1 enzymatic activity 6, 8, 11-15,
24.
Studies in animal models revealed that CHIT1 deficiency results in decreased TGFβ and IL-13-
1
5
driven fibrotic responses in the lungs . Also other observations support the central role of CHIT1
in human lungs and in expanding spectrum of human lungs disorders ranging from asthma
throughout granulomatous diseases such as sarcoidosis 11-13, 18, tuberculosis 25, 26 to interstitial lung
fibrosis 15. All these observations suggests that dual chitinase inhibition could potentially be
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
considered as the therapeutic strategy not only for asthma patients, where AMCase is considered
to be a crucial modulator of Th2 inflammatory response, but also for chronic interstitial lung
diseases dominated by chronic inflammation and ongoing pulmonary fibrosis.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Our group has recently described compound 3b 21 that displays AMCase vs CHIT1 selectivity
(150-times and 15-times in murine and human enzymes, respectively), has good pharmacokinetic
profile in mice and shows significant anti-inflammatory efficacy in HDM-induced asthma model
after oral administration. Starting from compound 3b we sought to identify dual specificity
inhibitor targeting both chitinases simultaneously and capable of ameliorating both pulmonary
inflammation and bronchial fibrosis.
RESULTS AND DISSCUSSION
We have previously reported 21 that shifting the position of the central nitrogen atom from
piperazine ring to the 4-aminopiperidine system in a moderately potent dual chitinase inhibitor
2
7
(
compound 1), discovered by the Wyeth group (Wyeth 1, Figure 1) , led to a significant, over
one order of magnitude increase in the inhibitory potency against both chitinases (compound 2, R
= Me). Initial exploration of SAR revealed however that increasing the size of the hydrophobic
substituent on the central nitrogen atom exemplified by compounds 3a and 3b-d, leads to a further
increase in potency against hAMCase, but not hCHIT1.
4
ACS Paragon Plus Environment

Page 5 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. Early optimization of Wyeth 1 – moderately potent dual chitinase inhibitor (replacement
of bromine by chlorine atom in the phenyl ring has no major impact on inhibitory potency) 21.
Examination of the published X-ray structures of complexes of hCHIT1 and hAMCase with
their small-molecule inhibitors 27, 28 indicated that the active sites of both enzymes are very similar
and differ by only two amino acids: Met300 and Arg269 in hCHIT1 are replaced by Ile300 and
His269 in hAMCase. Our previously reported compounds bind to the acidic binding site of the
enzymes through the network of the hydrogen bonds between 3-aminotriazole moiety and Tyr27,
2
1
Asp138 and Glu140 residues . Critical for the high activity of those inhibitors is the salt bridge
length of 2.8 Å) between the central basic nitrogen of 4-aminopiperidine fragment and carboxylic
(
group of Asp213. The hydrophobic interactions of para-halogenated phenethyl group and
hydrophobic pocket created by Tyr267, Glu297 and Met300 (hCHIT1) or Ile300 (hAMCase)
residues appear to be well optimized. As shown in Figure 1, with enlargement of the size of the
hydrophobic substituent on the central nitrogen atom from methyl to isobutyl or (ring-
substituted)benzyl, the inhibitory activity against hAMCase is increased by two orders of
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
magnitude while no analogous effect is observed for hCHIT1. We hypothesized that the key factor
determining this high activity against hAMCase are the hydrophobic interactions of the aliphatic
substituent with Ile300 and His269 in the enzyme binding pocket. On the other hand, such a bulky
N-substituent causes the steric clash with the positively charged, guanidine-containing Arg269
side chain in hCHIT1, which is reflected in no activity gain.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Another difference between the analyzed structures was the orientation of Trp99 side-chain.
While in the hAMCase-compound 1 complex (PDB ID: 3RM4) the indole ring of this amino acid
residue is directed slightly outside of the binding pocket (Figure 2A), it is directed towards the
pocket in hCHIT1 complexed with compound 3a (PDB ID: 5NRA; Figure 2B). As this residue
plays an important role in binding of both the oligosaccharide substrate and small molecule
inhibitors, we performed analysis of the experimentally solved structures of hAMCase (see: Figure
5
in Supporting Information). In six structures, both apo and with ligands, the Trp99 side-chain is
placed in the same position as in the hAMCase-compound 1 complex. In one complex (PDB ID:
RM9), Trp99 can adopt a double conformation and only in one case (PDB ID: 3RM8) the Trp99
3
is in the same position as in hCHIT1 complexed with compound 3b. It may be concluded that
although for hAMCase, the Trp99 side-chain may adopt both conformations, the preferred state is
towards outside of the binding pocket. It is also not fully clear what structural features of the ligand
induce the flip of this residue.
6
ACS Paragon Plus Environment

Page 7 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
9
Figure 2. (A) Docking of compound 2 to the binding pocket of hAMCase (PDB ID: 3RM4 )
and (B) structure of hCHIT1-compound 2 complex (PDB ID: 5NR8). For ligands, carbon atoms
are colored in green, nitrogen atoms in blue, oxygen atoms in red, sulfur atoms in yellow and
bromine atoms in dark red.
In order to design the equally active inhibitor for these two enzymes, during optimization of the
structure of compound 2 we explored only these fragments of the binding pockets, which are
conserved among binding sites of both proteins. Therefore, our initial design strategy to improve
interactions of the inhibitor molecule with the hCHIT1 active site (while maintaining high
hAMCase affinity) was based on the introduction of the small N-substituents containing polar
functional groups that might be capable of creating the additional binding interactions with the
guanidine moiety of Arg269 in hCHIT1. Such an approach was also driven by the fact that our
first generation inhibitors displayed considerable activity against hERG potassium channel and
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
1
dopamine and serotonin transporters . While we realized that the presence of the central basic
nitrogen atom and phenethylamine fragment are the possible structural features of our inhibitors
that trigger those two off-targets activities, we also assumed that we could not afford losing the
strong interaction between the central amine and the Asp213 residue and therefore decided that
this structural motif would be kept intact. On the other hand, it is known from the previous studies30
that implementation of a polar fragment into the peripheral part of the inhibitor resulting in the
overall decrease of the lipophilicity of the final molecule, oftentimes attenuates interaction with
the hERG potassium channel. In a similar manner, introduction of the alkoxy/hydroxyl group at
the -position with respect to the amino group, causing reduction of its basicity could provide yet
another measure of destabilizing interactions with hERG. We also hoped that the presence of the
polar moiety in the inhibitor molecule would help to reduce the monoamine transporters off-target
activity.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
FIRST APPROACH – COMPOUNDS WITH SMALL N-SUBSTITUENTS CONTAINING
POLAR GROUPS
As seen in Table 1 introduction of carboxylic group (compound 4) led to a significant decrease
in activity against both human chitinases in comparison with the parent compound 2. Amidation
of the carboxylic group (compound 5) partially restored the inhibitory potency against hAMCase,
but the activity vs hCHIT1 remained at over 1 micromolar level. Reversal of the sequence of the
amide fragment (compound 6) and especially replacement of the acetyl group with
methanesulfonyl group (compound 7) led to a further increase in activity vs hAMCase with only
marginal improvement of hCHIT1 inhibitory potency. Sulfonamide 8 exhibited analogous profile
as its parent counterpart 7, but introduction of N-methylcarbamoyl moiety effected in significantly
8
ACS Paragon Plus Environment

Page 9 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
less active inhibitor (compound 9). In contrast to analogs 5 – 8 that were either hAMCase-selective
inhibitors or only moderately active dual inhibitors (compounds 4 and 9), compound 10 containing
hydroxyethyl group on the central nitrogen atom displayed only ca. 3-fold hAMCase/hCHIT1
selectivity and reasonable activity against both enzymes (IC50 111 nM and 384 nM respectively).
Methylation of the hydroxyl group provided compound 11 that approached the desired dual
inhibitor profile even more closely. Introduction of additional methyl groups of the defined
stereochemistry into the hydroxy/methoxyethyl substituent (analogs 12 – 15) resulted in general
in no positive impact on the inhibitory activity with the exception of compound 15 (IC50 103 nM
and 44 nM against hAMCase and hCHIT1, respectively). Worth noting is the low hERG inhibition
of analog 15 that was achieved by quite minor structural change (as compared to the parent
compound 11: IC50 17.6 M vs 2.8 M respectively). However, 20-fold selectivity in murine
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
enzymes and 33% inhibition at 10 M of dopamine transporter appeared to be a serious
shortcomings of compound 15. Adding yet one more methyl group afforded compounds 16 and
17 displaying distinct AMCase-selectivity both in human and murine enzymes. Similar selectivity
was found for oxetane-containing inhibitor 18, while, conversely, the other two compounds (19
and 20) incorporating the cyclic ether moieties (tetrahydrofuran and tetrahydropyran respectively)
showed very promising potency against both human proteins. However, these two analogs suffered
from the undesired mAMCase-selectivity and were quite active against both off-targets
investigated.
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 1. Structure-activity relationship of the series of polar group – containing inhibitors a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Cmpd #
R
hAMCase
IC50 nM)
hCHIT1
mAMCase
(IC50 nM)
mCHIT1
hERGb
DATc
(
(IC50 nM)
(IC50 nM)
(IC M)
%inh. @
0 M
50
1
4
1040
233
1900
395
21000
8.0
NT
NT
NT
4
±
± 424
± 64
± 2830
5
6
7
8
9
690
1800
1150
± 71
48000
2.6
3.3
7.2
16.0
1.8
±
42
± 141
± 1414
163
495 ± 21
35 ± 9.2
8100
± 71
± 141
37 ± 5.7
28 ± 1.4
425 ± 35
1200 ± 141 22 ± 2.1
12500
±
707
870 ± 141
1430 ± 35
21 ± 0.7
130 ± 14
7550
13
±
1343
9100
NT
±
424
10
ACS Paragon Plus Environment

Page 11 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
10
111 ± 28
98 ± 17
384 ± 48
84 ± 11
135 ± 21
55 ± 1.4
66 ± 16
110 ± 28
180 ± 42
53 ± 7.1
73 ± 16
28 ± 7.1
27 ± 2.8
4800
6.3
2.8
6.1
6.4
14
NT
NT
NT
NT
NT
33
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
± 566
11
1700
±
424
1
1
1
2
3
4
185 ± 7.1
135 ± 35
135 ± 64
103 ± 9.9
40 ± 3.5
28 ± 2.1
83 ± 9.8
330 ± 14
530 ± 156
145 ± 21
44 ± 1.4
395 ± 106
505 ± 7.1
180 ± 28
6800
848
±
12500
2121
±
2350
212
±
15
1150
17.6
19
± 212
1
1
1
6
7
8
4700
NT
NT
NT
±
990
3650
353
3.5
4.4
±
5500
1273
±
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
19
20
21 ± 6.4
68 ± 1.4
27 ± 2.8
21 ± 1.4
17 ± 4.2
56 ± 24
1080
15
74
50
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
± 170
750 ± 14 11
a
IC50 data of each compound were determined in 2−3 independent assays, and mean IC50 ± SD
values are shown in the table. Cmpd, abbreviation of compound. SD, standard deviation.
b
data from Predictor™ hERG Fluorescence Polarization assay 31
c
data from DAT Human Dopamine Transporter Binding (Antagonist Radioligand) Assay,
Cerep32
It is worth noting that while we were able to achieve good activity against three out of four
enzymes tested, even best compounds of this series (analogs 11, 19 and 20) only moderately
inhibited murine CHIT1. As no crystal structures of the murine chitinases are known, we compared
their sequences and three-dimensional models of binding pockets in order to explain observed
differences in activities of our inhibitors between human and murine enzymes.
With regards to AMCase, the only differences within binding pockets observed on the structural
level between human and murine proteins are residues 297 (Glu for human, Gln for murine) and
3
00 (Ile and Phe, respectively). Glu and Gln are similar in size, and they may form analogous
interactions with ligand. Ile and Phe are both hydrophobic amino acids, and despite Phe is more
bulky, its rotation allows for accommodation of even relatively big substituents (like
tetrahydrofuran and tetrahydropyran fragments in compounds 19 and 20).
Human and murine CHIT1 binding pockets differ in residues at the same positions as in
AMCase. Residue 297 is Glu for human enzyme and Asp for murine protein, and residue 300 is
Met and Val, respectively. The observed lower activity of our compounds against mCHIT1 may
12
ACS Paragon Plus Environment

Page 13 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
indicate importance of interactions of Met300 of human protein with haloaromatic ring of
inhibitors. Also, while formation of the hydrogen bonds between ligands and Glu297 of human
enzyme could be predicted by molecular modeling, it appears that the carboxylic group of Asp297
in mCHIT1 is too distant to form this type of interactions with inhibitors.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
SECOND APPROACH – CONFORMATIONALLY CONSTRAINED COMPOUNDS WITH
OXYGEN ATOM CONTAINING FUNCTIONALITY IN BETA-POSITION WITH RESPECT
TO THE CENTRAL AMINE GROUP
Analysis of the crystallographic structures of complexes of hCHIT1 with inhibitors 2 and 3a
supported by molecular modeling studies indicated that two linear substituents present on the
central nitrogen atom in all of the analogs we synthesized so far could be joined one to another to
form new four, five or six-membered ring (Figure 3). We predicted, that for both human chitinases,
the binding mode of the resulting compounds would be similar to their noncyclic analogs. The
calculated binding score values suggested that the benzylic position as the site of cyclization
resulting in considerably constrained 3-aryl-substituted nitrogen-containing heterocycle is slightly
preferred (Figure 3A). However, the alternative mode of cyclization at the homo-benzylic position
(Figure 3B), while providing less effective conformational constraint, allowed for better alignment
of the benzyl fragment to the hydrophobic pockets of the enzymes.
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 3. Two possible modes of cyclization of two linear N-substituents leading to (A) more and
(B) less rigidified (4-chlorophenyl)-N-heterocycles.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3-(4-chlorophenyl)pyrrolidine and 3-(4-chlorophenyl)piperidine required for the synthesis of the
more conformationally constrained inhibitors were commercially available, what prompted us to
synthesize the corresponding analogs in their racemic forms. On the other hand, the 2-(4-
chlorobenzyl)pyrrolidine and 2-(4-chlorobenzyl)piperidine required for the synthesis of more
flexible inhibitors had to be synthesized. Careful examination of the experimentally solved
structure of hCHIT1 complexed with compound 2 (PDB ID: 5NR8) clearly indicated that only one
of the enantiomers can fulfill the constraints of the binding site (Figure 4) and thus we decided to
prepare these cyclic amines in the optically pure form.
Figure 4. Binding site of hCHIT1 complexed with compound 2 (PDB ID: 5NR8); homobenzylic
hydrogens marked as H
homobenzylic position with the N-methyl group (the resulting 2-(4-chlorobenzyl)pyrrolidine of
the R configuration shown). On the other hand, H can not be displaced in the same way without
A
and H
B
. H
A
can be substituted by the incoming alkyl linker joining the
B
disrupting conformation of the p-halophenetyl substituent.
As seen in Table 2, comparison of the biochemical potencies of inhibitors containing 3-(4-
chlorophenyl)-substituted N-heterocycles (compounds 21 and 22) with more conformationally
14
ACS Paragon Plus Environment

Page 15 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
flexible analogues 23 and 24 clearly indicated the favored mode of cyclization. Introduction of 3-
(4-chlorophenyl)pyrrolidine fragment resulted in poorly active compound 21 (IC50 6 M and 19.5
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
M vs hAMCase and hCHIT1, respectively), and the 3-(4-chlorophenyl)piperidine moiety
provided only modest improvement of activity against both human chitinases. On the other hand,
analogue 23, and especially 24, approached the targeted in vitro pharmacological profile very
closely, being potent, non-selective chitinases inhibitors against both human and murine enzymes.
However, the single-digit micromolar level of hERG inhibition displayed by those two compounds
was perceived as a serious disadvantage and the introduction of the oxygen-containing
functionality placed in the -position with respect to the amino functionality was attempted as a
countermeasure to this undesired side-activity. The first attempt was not very successful, as the 2-
(4-chlorobenzyl)-3-methoxyazetidine containing compound 25 was not only the most potent
hERG inhibitor we synthesized, but, moreover, its inhibitory activities against chitinases were
greatly diminished. Conversely, introduction of oxygen-containing functionality into position 4 of
pyrrolidine ring provided analogs 26 – 29, whose inhibition ability of chitinases was (regardless
the configuration of the additional chiral center) retained by larger part, while their hERG activity
was significantly attenuated. The in vitro pharmacological profile of the best compound of this
series 28 was further improved two-fold, by replacement of 2-(4-chlorobenzyl)-4-
hydroxypyrrolidine fragment with 2-(4-chlorobenzyl)morpholine resulting in inhibitor 30 that
displayed only two- and four-fold selectivity AMCase vs CHIT1 in murine and human enzymes,
respectively. The so obtained analog 30 represented the optimum combination of structural
features to achieve the desired dual inhibition both in human and murine enzymes while
maintaining acceptable hERG channel affinity level of 39 M, and therefore meeting the selected
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
criteria of the primary pharmacological parameters and justifying its further in vitro and in vivo
profiling.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Since lowering of lipophilicity and basicity of the compound is considered to be the method of
controlling the inhibition of the hERG channel, the pK values were calculated and logD7.4 values
a
were measured for the most interesting compounds (23 – 30). Unfortunately, based on this
valuations no simple conclusions could be drawn in relation to how do these parameters affect the
affinity of our inhibitors for hERG. Two compounds of comparable (and low – pK of ca. 7.5)
a
basicity of the central amine, 25 and 30, were the most potent (IC50 0.34 M) and second least
potent (IC50 39 M) hERG inhibitors, respectively, spanning over two orders of magnitude of
activity. Also a pair of stereoisomers, compounds 26 and 28, sharing the same calculated value of
pK , showed significant, 4-fold, difference in hERG activity (IC50 13.2 M and 55 M,
a
respectively). Lipophilicity of the inhibitors within this series of analogues spanned only 1.3 unit
of magnitude (logD7.4 = 0.68 for 28 and logD7.4 = 1.96 for 30; solubility of compound 30 in water
was measured to be 0.31 mg/mL), so the significant variability of impact of this parameter on the
hERG inhibition should not be expected. Nevertheless, it was these two compounds that proved to
be the best and the second best analogs, respectively, in terms of avoiding the potassium channel
binding.
In summary, it can be concluded that the structural factors rather than the physicochemical
properties govern the activity against hERG in this series of the conformationally constrained
chitinases inhibitors.
16
ACS Paragon Plus Environment

Page 17 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 2. Structure-activity relationship of the conformationally constrained analogs a
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
hERGb
DATc
%inh.
pK
a
d
logD7.4e
Cmpd
#
R
hAMCase hCHIT1
mAMCase
(IC50 nM)
mCHIT1
(
IC50 nM)
(IC50 nM)
(IC50 nM)
(IC50 M)
@
10 M
21
5950
19500
2600
37000
NT
NT
NT
NT
NT
NT
±
212
± 2120
± 141
± 4240
2
2
2
3
1350
± 71
1850
335 ± 49
19000
NT
NT
± 354
± 2828
73 ± 5.1
23 ± 2.1
59 ± 1.4
41 ± 8.5
120 ± 0
81 ± 12
430 ± 42
87 ± 7.8
4.0
8.3
NT
NT
9.86
9.74
1.15
1.41
24
2
2
2
5
6
7
395
580
485
5550
0.34
13.2
19
NT
NT
59
7.49
9.08
11.73
1.55
0.81
1.60
± 64
± 85
± 78
± 919
39 ± 7.1
19 ± 0.7
82 ± 3.5
47 ± 5.7
170 ± 28
65 ± 7.1
520 ± 70
345 ± 64
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
8
50 ± 9.2
87 ± 7.1
68 ± 9.9
135 ± 7.1
55
NT
9.08
0.68
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
3
9
0
26 ± 6.4
22 ± 2.1
44 ± 1.4
48 ± 8.2
21 ± 1.4
255 ± 49
74 ± 13.4
15.8
39
NT
95
11.73
7.58
1.56
1.96
30 ± 12.7
a
IC50 data of each compound were determined in 2−3 independent assays, and mean IC50 ± SD
values are shown in the table. Cmpd, abbreviation of compound. SD, standard deviation.
b
data from Predictor™ hERG Fluorescence Polarization assay 31
c
data from DAT Human Dopamine Transporter Binding (Antagonist Radioligand) Assay,
Cerep32
d
pK
a
values were calculated online using chemicalize.com freeware by ChemAxon
e
measured values
PHARMACOKINETIC AND ANTI-INFLAMMATORY EFFECTS OF COMPOUND 30.
Good oral bioavailability of compound 30 and low clearance provided sufficient exposure for
efficacy examination following once daily oral administration in mice (Table 3 and Figure 5).
Considerably higher clearance value in rats (accompanied by still very good bioavailability,
though) suggests that more frequent dosing of inhibitor 30 may be required for in vivo studies in
higher species.
18
ACS Paragon Plus Environment

Page 19 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 3. In vivo pharmacokinetic profile of compound 30 in mice and rats after 3 mg/kg IV and
1
0 mg/kg PO administration
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Mouse IV
4.76
Mouse PO
11.1
Rat IV
4.98
Rat PO
8.01
AUC0-inf
(mg*h/L)
C
max (mg/L)
1.59
10.5
2.39
n/a
16.5
33.7
1.56
n/a
CL
(mL/min/kg)
t
1/2 (h)
2.50
n/a
2.68
1.0
70
7.95
n/a
2.14
1.5
48
t
max (h)
F (%)
n/a
n/a
Figure 5. Plasma total concentration – time course for compound 30 after 3 mg/kg IV bolus and
after 10 mg/kg oral administration (PO) to female BALB/c mice.
Compound 30 passed the test for potential genotoxicity (AMES test 33 was negative) and was
evaluated in the CEREP Diversity Panel 34 in order to identify the possible off-target activities.
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
hERG potassium channel was found to be inhibited with IC50 of 12 M in the automated patch-
clamp assay 35 and also considerable dopamine transporter (DAT) inhibition was revealed showing
IC50 of 7.9 M and 0.3 M in DAT cell-based assay 36 and DAT radioligand binding assay 32,
respectively. This result prompted us to study the plasma and brain concentration of inhibitor 30
following oral administration in mice in the context of possible occurrence of the CNS-related
side-effects. By comparison of AUCs in plasma and brain it was found that approximately 6.8%
of 30 crosses the blood-brain barrier approaching the DAT IC50 radioligand assay value, during
first 2 hours after administration (Figure 6).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. Plasma and brain total concentrations of 30 after 10 mg/kg oral administration in female
BALB/c mice compared to the observed off-target activities.
Common, clinically relevant aeroallergen – house dust mite (HDM) extract was used to induce
allergic airway inflammation in C57BL/6 mice which was represented by the increased CD45-
positive leukocytes influx to the lungs and chitinolytic activity in BALf of mice challenged with
20
ACS Paragon Plus Environment

Page 21 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
HDM as compared to PBS challenged mice (control group) (Figure 7). Oral administration of
compound 30 (50 mg/kg, qd) starting from day 7 after the beginning of the HDM challenge led to
a significant reduction of total CD45-positive leukocyte number (37%) as compared to the HDM
vehicle group.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 7. Dual AMCase/CHIT1 inhibition with compound 30 results in significant anti-
inflammatory effects in HDM-induced asthma mouse model. Flow cytometry analysis of total
infiltrating leukocytes numbers in BALf recovered from naïve mice challenged with PBS or HDM
with or without compound 30 (50 mg/kg; PO; qd) or dexamethasone (10 mg/kg, IP, qd) treatment.
Molecular docking of 30 to binding pockets of hAMCase and hCHIT1 (Figure 8 A, C and B, D,
respectively) revealed that, indeed - morpholine ring is too distant to interact with residues at
positions 269 and 300. The interaction network formed by the 2-aminotriazole ring of 30 in both
2
1
proteins is analogous to that described earlier : direct hydrogen bonds are formed with Tyr27,
Asp138 and the water-mediated hydrogen bond is possibly formed with Glu140. Heterocyclic ring
is further stabilized by the Met - π interaction formed by Met210. Similarly to the earlier described
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
structures of hCHIT1 complexes with compounds 2 and 3a, the nitrogen of the central 4-
aminopiperidine ring may form hydrogen bond with Asp213, however in the hCHIT1-30 and
hAMCase-30 complexes the distances between carboxylic group of aspartic acid residue and
amino group of the ligand appear to be longer (3.7 Å and 3.5 Å respectively). 4-Chlorobenzyl
fragment is located within hydrophobic pocket formed by Leu362, Ala302, Leu301, Tyr267 (in
both enzymes), Ile300 (hAMCase), and Met300 (hCHIT1) residues. This last residue may further
stabilize the phenyl ring by Met - π interaction in hCHIT1 binding pocket. Oxygen atom in the
morpholine ring of 30 may also form a weak hydrogen bond with Trp99 of hCHIT1 and also in
hAMCase provided the side-chain of this residue is directed toward the binding pocket (Figure
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
C).
22
ACS Paragon Plus Environment

Page 23 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 8. Compound 30 docked to the hAMCase (PDB ID: 3RM4, panes A and C) and hCHIT1
(PDB ID: 5NR8, panes B and D) binding pockets. Upper panes (A, B) show key residues forming
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
the binding pockets of both proteins. Bottom panes (C, D) show a sphere representation of binding
pockets of proteins; distinct residues (269 and 300) and Trp99 are shown as sticks. For ligands,
carbon atoms are coloured in green, nitrogen atoms in blue, oxygen atoms in red and sulphur atoms
in yellow.
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
SYNTHESIS
Scheme 1. General Synthesis of Compounds 4 – 20 a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
aReagents and conditions: (a)
chlorophenylacetaldehyde, NaBH(OAc)
S,S’-dimethyl N-cyanodithioiminocarbonate, K
5-55% over 3 steps; (f) NaOH, H O, reflux, 10%.
R
-NH
,
NaBH(OAc)
,
DCE, rt; then (b) 4-
1
2
3
3
, DCE, rt, 45-88% over 2 steps (c) HCl, AcOEt, rt; (d)
o
o
2
CO , MeCN, 82 C; (e) hydrazine, MeCN, 82 C,
3
1
2
Synthesis of analogs 4 – 20 was accomplished in a similar fashion as described previously 21 for
compounds 2, 3a and 3b (Scheme 1). N-Boc-4-piperidone 31, was reductively aminated with a set
of commercially available primary amines and the resulting intermediate amines were in situ
24
ACS Paragon Plus Environment

Page 25 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
subjected to the second reductive alkylation with 4-chlorophenylacetaldehyde to provide
compounds 32a - q. Acidic deprotection of the piperidine followed by a two-step 3-aminotriazole
ring formation completed synthesis of the target compounds. An additional step of the hydrolysis
of the methyl ester was required for the synthesis of compound 4.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Analogues 21 and 22 were prepared in a straightforward manner from commercially available
racemic 3-(4-chlorophenyl)pyrrolidine and 3-(4-chlorophenyl)piperidine respectively, by the
reductive amination of N-Boc-4-piperidone that yielded intermediates 34 and 35. N-deprotection
and 3-aminotriazole ring installation afforded the final analogues.
Scheme 2. Synthesis of analogues 21 and 22 a
aReagents and conditions: (a) 3-(4-chlorophenyl)pyrrolidine (for 34) or 3-(4-
chlorophenyl)piperidine (for 35), NaBH(OAc)
3
, AcOH, DCE, rt, 45% and 23% for 34 and 35,
CO , MeCN,
respectively; (b) HCl, AcOEt, rt; (c) S,S’-dimethyl N-cyanodithioiminocarbonate, K
2
3
o
o
8
2 C; (d) hydrazine, MeCN, 82 C, 55% and 14% over 3 steps for 21 and 22, respectively.
The syntheses of compounds 23 – 30 required the prerequisite cyclic secondary amines that
would be incorporated into the molecules of the final inhibitors to be prepared. Accordingly, Boc-
D-proline 36 was coupled with N,O-dimethylhydroxylamine and the resulting Weinreb amide 37
was reacted with 4-chlorophenyl magnesium bromide affording ketone 38 that was reduced with
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
concomitant removal of the Boc- group by treatment with AlCl
3
/ Et SiH system to give the desired
3
(S)-2-(4-chlorobenzyl)pyrrolidine 39.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 3. Synthesis of (R)-2-(4-chlorobenzyl)pyrrolidine 39 a
a
Reagents and conditions: (a) HN(Me)OMe, CDI, DCM, rt, 52%; (b) 4-chlorophenymagnesium
o
o
bromide, Et
2
O, -70 C to rt, 69%; (c) AlCl
3
, Et SiH, DCM, 0 C to rt, 99%; (d) N-Boc-piperid-4-
3
one, NaBH(OAc)
cyanodithioiminocarbonate, K
3
, AcOH, DCE, rt, 45%; (e) HCl, AcOEt, rt; (f) S,S’-dimethyl N-
o
o
2
CO , MeCN, 82 C; (g) hydrazine, MeCN, 82 C, 39% over 3 steps
3
(S)-2-methyl-2-propanesulfinamide 40 was condensed with 4-chlorophenylacetaldehyde
providing sulfinimine 41, to which allyl Grignard reagent was added yielding sulfinamide 42.
Nitrogen atom was deprotonated with sodium hydride followed by allylation with allyl bromide
and the resulting diene 43 was subjected to the ring-closing metathesis reaction using Grubbs
catalyst (1st generation). Hydrogenation of the double bond in the resulting (R)-6-(4-
chlorobenzyl)-3,4-dehydropiperidine 44 followed by the removal of the tert-butyl sulfonamide
group in acidic conditions from intermediate 45 completed the synthesis of (R)-2-(4-
chlorobenzyl)piperidine 46.
26
ACS Paragon Plus Environment

Page 27 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 4. Synthesis of (R)-2-(4-chlorobenzyl)piperidine 46 a
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Reagents and conditions: (a) 4-chlorophenylacetaldehyde, Ti(OEt)
4
, DCM, reflux, then rt; (b)
o
allylmagnesium bromide, -20 C to rt, 23% over two steps; (c) allyl bromide, NaH, DMF, rt, 71%;
d) Grubbs catalyst 1 generation, DCM, reflux, 94%; (e) H
st
(
2
, Pd/C (cat.), MeOH, rt, 77%; (f) HCl,
MeOH, rt, 72%.
Boc-L-(4-chloro)phenylalanine 47 was transformed into the corresponding -diazoketone 48 via
reaction of the preformed mixed anhydride with an excess of diazomethane. Rhodium acetate
catalyzed cyclization afforded S-2-(4-chlorophenyl)azetidinone 49 that was stereoselectively
reduced to the secondary alcohol 50 (stereoselectivity > 10:1; minor isomer was rejected by
crystallization from ethyl acetate/hexanes mixture, ca. 1 : 5 ratio; the configuration of the newly
created chiral center was confirmed by single-crystal X-ray diffraction – see Supporting
Information). Williamson etherification of the hydroxyl group and removal of the N-Boc
protection from intermediate 51 afforded the desired azetidine 52.
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Scheme 5. Synthesis of (2S,3S)-2-(4-chlorobenzyl)-3-methoxyazetidine 52 a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
o
Reagents and conditions: (a) methyl chloroformate, Et
3
N, THF, -15 C, then CH
2
2
N solution
o
o
in Et
NaBH
2
O, -30 C to rt, 96%; (b) Rh
2
(OAc)
4
(0.5% mol), Et
3
N (1% mol), DCM, 0 C to rt, 46%; (c)
, MeOH, 0 C to rt, 86%; (d) NaH, MeI, DMF, rt, 100%; (e) HCl, 1,4-dioxane, rt, 77%.
o
4
Fully protected Weinreb amides 55 and 61 were prepared in few straightforward synthetic steps
starting from commercially available trans- and cis- D-4-hydroxyproline derivatives 53 and 58,
respectively, and reacted with 4-chlorophenyl magnesium bromide yielding ketones 56 and 62.
Reduction of a carbonyl groups in a strongly acidic conditions accompanied by concomitant
removal of protecting groups afforded the desired (2S,4S)-2-(4-chlorobenzyl)-4-
hydroxypyrrolidine 57 and (2S,4R)-2-(4-chlorobenzyl)-4-hydroxypyrrolidine 63.
28
ACS Paragon Plus Environment

Page 29 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 6. Synthesis of (2S,4S)-2-(4-chlorobenzyl)-4-hydroxypyrrolidine 57 and (2S,4R)-2-(4-
chlorobenzyl)-4-hydroxypyrrolidine 63 a
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Reagents and conditions: (a) HN(Me)OMe, CDI, DCM, rt, 49% for 54, 67% for 61; (b) TBSCl,
o
imidazole, DMF, rt, 92%; (c) 4-chlorophenymagnesium bromide, Et
0% for 62; (d) AlCl
DCM, reflux, 68%; (f) LiOH, THF/MeOH/H
2
O, -70 C to rt, 86% for 56,
o
4
3
, Et
3
SiH, DCM, 0 C to rt, 81% for 57, 98% for 63; (e) Boc
O, rt, 94%.
2
O, Mg(ClO
4
2
) ,
2
(S)-2-amino-(4-chlorophenyl)propan-1-ol 64 was reacted with chloroacetyl chloride and the
resulting amide 65 was cyclized in basic conditions. The so obtained S-5-(4-
chlorophenyl)morpholin-3-one 66 was reduced with borane dimethyl sulfide complex affording
morpholine 67.
29
ACS Paragon Plus Environment