Development of potent and selective Cathepsin C inhibitors free of aortic binding liability by application of a conformational restriction strategy

Article abstract of DOI:10.1016/j.bmcl.2021.128202

Cathepsin C plays a key role in the activation of several degradative enzymes linked to tissue destruction in chronic inflammatory and autoimmune diseases. Therefore, Cathepsin C inhibitors could potentially be effective therapeutics for the treatment of diseases such as chronic obstructive pulmonary disease (COPD) or acute respiratory distress syndrome (ARDS). In our efforts towards the development of a novel series of Cathepsin C inhibitors, we started working around AZD5248 (1), an α-amino acid based scaffold having potential liability of aortic binding. A novel series of amidoacetonitrile based Cathepsin C inhibitors were developed by the application of a conformational restriction strategy on 1. In particular, this work led to the development of a potent and selective Cathepsin C inhibitor 3p, free of aortic binding liability.

Full text of DOI:10.1016/j.bmcl.2021.128202

Bioorg. Med. Chem. Lett. 47 (2021) 128202
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Development of potent and selective Cathepsin C inhibitors free of aortic
binding liability by application of a conformational restriction strategy
Abhisek Banerjeea, Ranganadh Velagaletia, Sandip Patila, Mahesh Pawara, Pravin Yadava,
Pradip Kadama, Mohammad Mohsin Qadria, Samitabh Chakrabortib, Jagmohan S. Sainic,
,*
,
b
,
c
,
Dayanidhi B. Beherad, Keya Karanjaia, Pravin S. Iyera
d, Laxmikant A. Gharata, Sanjib Dasa
a Medicinal Chemistry Division, Glenmark Research Centre, A-607, TTC Industrial Area, MIDC Mahape, Navi Mumbai 400 709, India.
b Pharmacology Division, Glenmark Research Centre, A-607, TTC Industrial Area, MIDC Mahape, Navi Mumbai 400 709, India.
c Computational Chemistry, Glenmark Research Centre, A-607, TTC Industrial Area, MIDC Mahape, Navi Mumbai 400 709, India.
d Drug Metabolism and Pharmacokinetics, Glenmark Research Centre, A-607, TTC Industrial Area, MIDC Mahape, Navi Mumbai 400 709, India
A R T I C L E I N F O
A B S T R A C T
Keyword:
Cathepsin C plays a key role in the activation of several degradative enzymes linked to tissue destruction in
chronic inflammatory and autoimmune diseases. Therefore, Cathepsin C inhibitors could potentially be effective
therapeutics for the treatment of diseases such as chronic obstructive pulmonary disease (COPD) or acute res-
piratory distress syndrome (ARDS). In our efforts towards the development of a novel series of Cathepsin C
Chronic Obstructive Pulmonary Disease
(COPD)
α
-Amino acid based scaffold
Amidoacetonitrile-based Cathepsin C inhibitors
Coronavirus SARS-CoV-2
inhibitors, we started working around AZD5248 (1), an α-amino acid based scaffold having potential liability of
aortic binding. A novel series of amidoacetonitrile based Cathepsin C inhibitors were developed by the appli-
cation of a conformational restriction strategy on 1. In particular, this work led to the development of a potent
and selective Cathepsin C inhibitor 3p, free of aortic binding liability.
Aortic binding
Catalytic cysteine residue (Cys234)
Cathepsin C (Cat C), also known as dipeptidyl peptidase I (DPP I), is a
constitutively expressed lysosomal cysteine protease which sequentially
removes dipeptides with broad specificity from the free N-terminal of
proteins and peptides.1 Many members of the Cathepsin family are
important for intracellular protein degradation and cellular homeosta-
sis. While Cathepsin C plays a crucial role in these processes, it also
activates a number of granule-associated serine proteases with pro-
inflammatory and immune functions. Some of the downstream pro-
teases matured by Cathepsin C are neutrophil elastase (NE), cathepsin G
(Cat G) and proteinase-3 (PR- 3), which are present in neutrophils as
well as in macrophages. Additionally, mast cell proteases and gran-
zymes, present in cytotoxic T lymphocytes (CTLs) and natural killer (NK)
cells, are also matured by Cathepsin C.2 Maturation of these downstream
proteases result in degradation of extracellular matrix (ECM) compo-
nents, cleavage of inflammatory mediators, cytokine as well as chemo-
kine induction, increased chemotaxis, increased airway secretions and
maintenance of inflammation.3
believed to play a role in the pathology of several respiratory diseases,
such as Chronic Obstructive Pulmonary Disease (COPD), Cystic Fibrosis
(CF), non-Cystic Fibrosis (non-CF), bronchiectasis and also ARDS asso-
ciated with COVID-19 infections.4
Cat C knockout mice are resistant to lung airspace enlargement and
inflammatory cell infiltration in both cigarette smoke and ozone expo-
sure models of COPD.5,6 The pharmacological inhibition of Cat C activity
via therapeutic administration of the covalent, irreversible compound,
MOTP, results in protection from smoke-induced pulmonary inflam-
mation, and alveolar destruction comparable to the levels of protection
exhibited by Cat C-/- mice. The results suggest Cat C as a potential target
for new therapies addressing unmet needs in chronic respiratory dis-
eases and emphysema.7 Brensocatib (AZD7986, 2 in Fig. 1), a Cat C
inhibitor from Astrazeneca/Insmed, has been reported to achieve its
primary end-point in a Phase II clinical trial with non-cystic fibrosis
bronchiectasis patients. Moreover, the company announced very
recently that brensocatib will be evaluated in the STOP-COVID19 (Su-
periority Trial of Protease Inhibition in COVID-19, EudraCT no.
2020–001643-13) trial in up to 300 hospitalized patients with COVID-
19. The eventualities in COVID-19 often result from acute respiratory
A number of respiratory diseases are associated with an over-
abundant accumulation of neutrophils and the presence of increased
levels of at least some neutrophil serine proteases. These enzymes are
* Corresponding author.
E-mail address: Sanjib.Das@glenmarkpharma.com (S. Das).
https://doi.org/10.1016/j.bmcl.2021.128202
Received 25 February 2021; Received in revised form 5 June 2021; Accepted 10 June 2021
Available online 15 June 2021
0960-894X/© 2021 Elsevier Ltd. All rights reserved.

A. Banerjee et al.
Bioorganic & Medicinal Chemistry Letters 47 (2021) 128202
In our quest, we envisioned that
α-amino acids with appropriate
conformational constrain around the reactive amino group on 1 (Fig. 1),
can overcome this aortic binding liability. To test the hypothesis, a series
of novel Cat C inhibitors with variety of
α -amino acids, with different
degrees of conformational constrain, were synthesized (3a-q) and
screened (Tables 1–3). During this exercise, ([1,1′-biphenyl]-4-yl)pro-
panenitrile portion of 1 (Fig. 1) was kept constant. Together with 1 and
2, selected compounds were screened against propionaldehyde based in
vitro assay (Table 4).
The preparations of compounds 3a-3q were achieved using the
general synthetic route illustrated in Scheme 2. As previously reported,
amine 72 was synthesized from compound 4 via sequential N-Boc pro-
tection, amidation, Suzuki coupling with (4-cyanophenyl)boronic acid
and N-Boc deprotection. The obtained amine 7 was then coupled with
Fig. 1. Cat C inhibitors developed by AstraZeneca.
distress syndrome (ARDS). Accumulated evidence indicates that lung
injury in such cases result due to release of proteases from activated
neutrophils.8 As seen previously in chronic respiratory syndromes, Cat C
might be a major driver of lung damage and hence, inhibition of this
upstream protease can fulfil an important medical need in this pandemic
situation. Given such findings, Cat C has been recognized as a potential
therapeutic target for the treatment of a broad range of diseases due to
the important role that it plays in biological systems. Thus, an object of
the present invention is to provide compounds which are inhibitors of
Cathepsin C.
various known and novel N-Boc-protected cyclic α-amino acids 8a-q
Table 1
Cat C inhibitors with 6-membered α-amino acids.
Many organizations like, GlaxoSmithKline, AstraZeneca, Boehringer
Ingelheim, Prozymex and Janssen have been extensively working on Cat
C as a therapeutic target over the years. Around thirty-five international
patent publications disclose variety of scaffolds as Cat C inhibitors.9,10
Many known Cat C inhibitors form an irreversible covalent bond with
Cys-234 of the enzyme’s active-site. Typically, these inhibitors are di-
peptides carrying reactive functional groups such as diazomethyl ke-
tones,11,12 vinyl sulfones,13,14 acyloxymethyl and fluoromethyl
ketones,13 and O-acyl hydroxamic acids.15 In addition, naturally
occurring epoxysuccinyl derivative E64 was also shown to inhibit Cat C
at high concentrations.16 Known reversible inhibitors of Cat C are
dipeptide semicarbazides,17 arginine-based peptides,18 phosphinic tri-
peptides19 and dipeptide nitriles.20–23
Compound
R1
% Cat C
Inhibition
(1 µM)
% Cat C
Inhibition
(10 µM)
Cat C
Cat C
Cellular
IC50
Enzyme
IC50
(nM)a
(nM)a
1
99.2
99.9
1
5
3a
3b
75.54
2.2
96.31
14.2
308.1
NDb
9724
NDb
3c
3.29
2.19
NDb
NDb
Most recently, AstraZeneca reported potent and selective dipeptide
nitrile based Cat C inhibitors 1 (AZD5248) and 2 (AZD7986) (Fig. 1). In
particular, AstraZeneca’s first generation Cat C inhibitor 1 showed
aortic binding in a rat quantitative whole body autoradiography. As a
result, its development was terminated before human dosing due to
3d
3e
68.26
1.2
94.99
0
394
NDb
NDb
NDb
safety concerns. The chemical reactivity of α-amino acid structure of 1
with an aldehyde of elastin peptide chain (Scheme 1) from aortic tissue
is responsible for the toxicity.2,24
Specifically, the chemical reactivity of 4-aminotetrahydro-2H-pyran-
4-carboxamide of 1 with the aldehyde of elastin occurs through for-
mation of a stable five-membered imidazolidin-4-one (Scheme 1).
AstraZeneca overcame this aortic binding liability by employing
β-amino acids and ultimately developed second generation Cat C in-
hibitor 2 (Fig. 1), which was in Phase II clinical trials for treatment of
COPD.24 In our hands, we could reproduce the reactivity profiles of 1
(fast; 15% remaining after 1 h) and 2 (slow; 95% remaining after 15 h) in
propionaldehyde based in vitro chemical reactivity assay,25 developed to
support lead optimization process by eliminating the possibility of aortic
binding. (Table 4).
3f
4.63
1.11
22.75
1.11
NDb
NDb
NDb
NDb
3g
a
Values are mean of two independent experiments in duplicate.
b
ND refers to ‘not determined’.
Scheme 1. Imidazolidin-4-one formation of 1 with elastin.
2

A. Banerjee et al.
Bioorganic & Medicinal Chemistry Letters 47 (2021) 128202
cyclization yielded benzyl ethers 15i-l.32 O-Debenzylation by hydro-
genolysis provided corresponding amino alcohols 16i-l, which upon
oxidation provided N-Boc amino acids 8i-l.
Table 2
N-Boc amino acids 8m,33 8o,34 8p,35 and 8q36 are known and syn-
thesized as previously reported. Novel N-Boc amino acid 8n was syn-
thesized as illustrated in Scheme 4.37 Reductive amination of keto ester
Cat C inhibitors with 5-membered spiro
α
-amino acids.
Compound
R1
% Cat C
Inhibition
(1 µM)
% Cat C
inhibition
(10 µM)
Cat C
Enzyme
IC50
Cat C
Cellular
IC50
17 with (R)-(+)-
α
-methylbenzylamine provide amino ester 1838 which
upon further reduction provided amino alcohol 19. Subsequent N-
alkylation of 19 with ethyl bromoacetate provided amino alcohol 20,
which upon mesylation and intramolecular c-alkylation provided bicy-
clic amino ester 21. N-Dealkylation and in situ Boc-protection was fol-
lowed by ester hydrolysis to provide N-Boc amino acid 8n.
(nM)a
(nM)a
3h
3i
91.47
92.36
99.07
99.49
91.51
NDb
77.56
NDb
A quick structure–activity relationship (SAR) of 6-membered
α-amino acids (3a – 3g, Table 1) was established around compound 1
(AZD5248), which showed potent human Cat C inhibition with enzyme
IC50 of 1 nM and cellular IC50 of 5 nM. Compound 3a showed moderate
Cat C enzyme inhibition with IC50 308.1 nM, whereas its diastereomer
3b showed poor Cat C inhibition (2.2% and 14.2% inhibition at 1 µM
and 10 µM concentration, respectively). 3-Hydroxyadamantanyl-
3j
97.19
98.19
27.83
150.3
3k
3l
99.04
84.3
99.67
99.32
10.17
227.9
68.98
NDb
α-amino acid derivative 3c also appeared inactive in our assay. Except
for 3d (Cat C IC50 = 394 nM), remaining conformational restricted 6-
membered α-amino acids (3e-3g, Table 1) showed poor Cat C inhibi-
tion at both 1 µM and 10 µM concentrations.
At this stage, we turned our attention to 5-membered ring α-amino
acids (3h – 3l, Table 2). Compound 3h with a simple 5-membered
α
-amino acid (proline) showed Cat C enzyme inhibition with IC50 =
a
b
Values are mean of two independent experiments in duplicate.
91.51 nM. However, as anticipated, it indicated fast aortic binding (8%
remaining after 2 h; half-life of 0.56 h) when tested in the propio-
naldehyde reactivity assay (Table 4). To get rid of the aortic binding
liability, spiro analogs of various ring sizes (3–6) by systematic homol-
ogation on proline were pursued (3i-3l). In comparison to 3h, increase
in Cat C enzymatic potency was observed from 3-membered ring to 5-
ND refers to ‘not determined’.
(Scheme 3 and 4) to provide amides 9a-q. Subsequent dehydration
provided nitriles 10a-q which upon N-Boc deprotection provided com-
pounds 3a-q.
In particular, N-Boc amino acids 8a-8c and 8h were commercially
available, whereas 8d-8g were synthesized as previously reported.26
Spiro N-Boc aminoacids 8i-8l are also commercially available, however,
they were synthesized using a modified synthetic strategy as illustrated
in Scheme 3.27 O-Benzylation of commercially available chiral alcohol
1128 followed by epoxide opening with cyclic carbonitriles 12i-l, pro-
vided nitriles 13i-l.29–31 Subsequent nitrile reduction and N-Boc pro-
tection provided amino alcohols 14i-l which upon intramolecular
membered ring (3i: IC50 = 77.56 nM; 3j: IC50 = 27.83 nM; 3k: IC50
10.17 nM), whereas, 6-membered ring (3l) was less potent (IC50
=
=
227.9 nM). In addition, increase in cellular potency was also observed
from 3j (IC50 = 150.3 nM) to 3k (IC50 = 68.98 nM) with increase in ring
size. At this stage, the most potent compound from this series 3k was
advanced for aldehyde binding studies; however, this analog too showed
tendency of fast aortic binding (0% remaining after 15 h; half-life of
0.56 h; Table 4).
Table 3
Cat C inhibitors with fused 5-membered α-amino acids.
Compound
R1
% Cat C inhibition (1 µM)
% Cat C inhibition (10 µM)
Cat C Enzyme IC50 (nM)a
222.6
Cat C CellularIC50 (nM)a
NDb
3m
83.89
97.87
3n
3o
3p
6.55
85.61
99.18
50.15
98.23
99.96
NDb
NDb
77.73
9.27
79.07
699.2
3q
88.72
99.03
153.6
1716
a
b
Values are mean of two independent experiments in duplicate.
ND refers to ‘not determined’.
3

A. Banerjee et al.
Bioorganic & Medicinal Chemistry Letters 47 (2021) 128202
Table 4
Cathepsin aldehyde adduct assay.
Compound
MW
Propionaldehyde Reactiona
MW of Adductb
% Remainingc
Duration of Incubation (h)
Half-life of Analyte (h)
1
374.4
420.5
360.41
344.4
398.5
384.5
384.5
384.5
Fast
414
15
95
48
8.5
0
1
0.36
> 15
15.75
0.56
0.56e
–
2
Slow
Moderate
Fast
NTd
15
15
2
3a
3h
3k
3n
3o
3p
399 (NEG)
383 (NEG)
437 (NEG)
423 (NEG)
423 (NEG)
NTd
Fast
15
15
15
15
Slow
Fast
100
7
5.3
Slow
97
> 15
a
Stability was evaluated for all the compounds in a single experiment using bulk spiking (n = 1), wherein the compound was spiked into the solution of propio-
naldehyde in 0.1 M phosphate buffer (pH 7.4); samples were aliquoted at intervals and analyzed in HPLC-UV.
b
Mass of the adduct was confirmed as addition of (+) 40 amu to respective molecular ion peak.
c
In-house data based on hour of incubation with propionaldehyde.
d
e
NT- Not tested as compound was not reactive.
Only 2 h data considered for calculation of t1/2 as no area of analyte observed in 15 h.
Scheme 2. Reagents and conditions: (a) (i) di-tert-butyl dicarbonate, Na2CO3, H2O, THF, RT, 48 h, 80% (ii) NH4Cl, HATU, HOBt, DIPEA, DMF, RT, 16 h, 96%; (b) (4-
cyanophenyl)boronic acid, Pd(dppf)Cl2.CH2Cl2, K2CO3, 1,4-dioxane, H2O, 80 ◦C, 18 h, 50%; (c) HCl in diisopropyl ether, THF, 10 ◦C, 4 h, 96%; (d) (i) Acid 8a-q,
HATU, HOBt, DIPEA, DMF, RT, 18 h, 50–60%; (e) Burgess reagent, CH2Cl2, RT, 18 h, 80–90%; (f) Formic acid, THF, 0 ◦C - RT, 18 h, 20–30%.
Then, we shifted our focus to fused bicyclic ring system where
accordance to our docking experiments performed using Schrodinger’s
Glide module.39 The X-ray crystal structure of Cat C (PDB Code: 4CDE) is
used to perform all docking studies (Figure 2).2 Structural information
revealed human Cat C to exist in a tetrameric form containing four
identical subunits.1 Each identical subunit comprises of three chains,
light and heavy chains forming the catalytic domain, and a third chain
constituting the exclusion domain. The active site of the protein harbors
the catalytic cysteine residue (Cys234), which is located at the interface
of these three chains. Cat C has an open S1 pocket located at the entrance
of the active site, and a large S2 pocket which is deeply buried within the
protein. The exclusion domain, a unique characteristic of Cat C, blocks
access of substrates beyond the S2 site and explains the exopeptidase
activity of Cat C. In addition, the side chain of Asp1 located in the
α
-amino acid proline was kept constant (3m-3q, Table 3). Compound
3m (IC50 = 222.6 nM) with a 3-membered fused ring was equipotent to
3l, but was much less active compared to 3h-3k (Table 2). This differ-
ence in potency opened up a possibility of occupying the available space
with right orientation so that the enzymatic potency can be improved. 5-
membered fused ring with a trans- ring junction (3n) and two possible
cis- ring junctions (3o and 3p) were prepared. To our delight, the
enzymatic potency improved significantly from 3n (50% inhibition at
10 μM, Table 3) to 3p (IC50 = 9.27 nM). However, with the introduction
of 6-membered fused ring (3q), there was a 16-fold drop in potency
(IC50 = 153.6 nM).
The difference in potency between compounds 3n, 3o and 3p were in
4

A. Banerjee et al.
Bioorganic & Medicinal Chemistry Letters 47 (2021) 128202
interaction with Gln228. Furthermore, it was seen that cyano (CN)
group in compound 3p makes a stable hydrogen bond with Asn65,
which is missing in compound 3o.40 Lack of this interaction could be
attributed to the reduced potency of 3o. The (2S)-octahydrocyclopenta
[b]pyrrole-ring forms a hydrogen bond with residues Gly277 and Asp1
in both compounds 3p and 3o, whereas in compound 3n, interaction
with Asp1 is missing due to change in chiral center and rotation of the
bicyclic ring. The absence of this critical interaction might be respon-
sible for loss in potency of compound 3n.
Compound 3p proved to be the first Cat C inhibitor from this series
with significantly reduced aortic binding tendency (97% remaining after
15 h; half-life > 15 h) in propionaldehyde studies (Table 4). In contrast,
its diastereomer 3o had higher aortic binding possibility (7% remaining
after 15 h; half-life = 5.3 h). In addition, the less active analog 3n also
showed reduced aortic binding (100% remaining after 15 h) tendency.
Overall, compound 3p not only well occupied the S2 pocket with its
fused ring system to provide high potency, but also the conformational
rigidity of the ring junction kept the two nitrogen far enough such that
the imidazolin-4-one formation with aldehyde was not feasible.
Scheme 3. Reagents and conditions: (a) (i) Benzyl bromide, NaH, THF, RT, 5 h,
56% (ii) cyclic carbonitrile 12i-l, diisopropylamine, n-BuLi, THF, ꢀ 78 ◦C, 1 h,
40–55%; (b) cobalt (II) chloride hexahydrate, NaBH4, MeOH, RT, 1 h, di-tert-
butyl dicarbonate, NaOH, H2O, CH2Cl2, RT, 18 h, 80–85%; (c) trimethyl
orthoacetate, BF3 etherate, CH2Cl2, RT, 2 h 40–50%; (d) H2, 10% Pd on carbon
(wet), MeOH, 40 psi, RT, 2 h, RT, 90–95%; (e) 4-methylmorpholine N-oxide,
tetrapropylammonium perruthenate, CH3CN, RT, 2 h, 95–98%
Table 5
SAR and metabolic stability data of 24a-e.
Compound
R
Cat C
IC50
Metabolic stability (% remaining)b
Human
Rat
Mouse
(nM)a
3p
9.27
1 ± 0.5
3 ± 0.6
6 ± 0.5
24a
44.01
14 ± 1.0
16 ± 0.5
25 ± 2.7
0 ± 0.0
32 ± 3.4
12 ± 2.0
Scheme 4. Reagents and conditions: (a) (R)-(+)-α-methylbenzylamine, AcOH,
ethanol, 70–75 ◦C, 4 h, then sodium cyanoborohydride, reflux, 12 h, 24%; (b)
lithium triethylborohydride solution (1 M in THF), RT, 3 h, 80%; (c) ethyl
bromoacetate, NaHCO3, CH3CN, reflux, 14 h, 72%; (d) (i) methanesulfonyl
chloride, triethylamine, CH2Cl2, 0 ◦C -RT, 30 min, 93% (ii) sodium tert-but-
oxide, THF, 70–80 ◦C, 5 h, 33%; (e) (i) di-tert-butyl dicarbonate, palladium on
carbon (10% wet), EtOH, H2, 60 psi, RT, 18 h, 81% (ii) lithium hydroxide, THF,
methanol, RT, 18 h, 87%.
24b
24c
24d
24e
4.62
14.89
6.45
2 ± 0.2
9 ± 0.3
8 ± 0.7
2 ± 0.9
34 ± 5.2
29 ± 3.9
16 ± 6.0
47 ± 5.4
22 ± 1.5
exclusion domain governs substrate recognition by interacting with the
α
-amino group of the substrate peptide.24 Critical residues that form the
1248
hot-spots for ligand–protein interactions are Cys234 (responsible for
covalent bond formation), Gly277, Asn380, Asn 65 and Asp1. From the
docked-ligand pose analysis, it is observed that all three compounds 3p,
3o and 3n make covalent interaction with Cys234 and hydrogen bond
a
values are reported as mean of three replicates (±SD).
b
Percentage remaining at 60 min in liver microsomal preparations.
Fig. 2. Docking of Compounds 3p (A), 3o (B) and 3n (C) in the binding pocket of Cat C (PDB 4CDE). Indicated in the figures are the key hydrogen bonds between the
protein and ligands within a distance of 4 Å. (A) 3p makes interactions with Asp1, Cys234, Gln228, Gly277 and Asn65 (B) 3o makes interactions with Asp1, Cys234,
Gln228, Gly277; Asn65 interaction is missing (decreased potency) (C) 3n does not make critical interactions with Asp1 due to rotation of the bi-cycle; Asp1 is critical
residue reported for activity.
5

A. Banerjee et al.
Bioorganic & Medicinal Chemistry Letters 47 (2021) 128202
Scheme 5. Reagents and conditions: (a) (i) substituted boronic acid, Pd(dppf)Cl2.CH2Cl2, K2CO3, 1,4-dioxane, H2O, 80 ◦C, 18 h, 50–55% (ii) HCl in diisopropyl
ether, THF, 10 ◦C, 4 h, 95–98%; (b) (i) Acid 8p, HATU, HOBt, DIPEA, DMF, RT, 18 h, 50–60%; (c) (i) Burgess reagent, CH2Cl2, RT, 18 h, 80–90% (ii) Formic acid,
THF, 0 ◦C - RT, 18 h, 25–30%.
development of 3p or a close analog, as an inhaled therapeutic agent will
Table 6
be reported in due course.
Selectivity profile for 3p.
Enzyme
Fold selectivity
% inhibition (1 µM)
% inhibition (10 µM)
Declaration of Competing Interest
Cathepsin K
Cathepsin L
Cathepsin S
Cathepsin B
CYP1A2
93
10,780
4405
2260
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
15.2
69.7
2.2
43.7
88.3
33.8
32.4
38.4
40.1
CYP2D6
CYP3A4
Acknowledgements
CYP2C9
12
CYP2C19
CYP2C8
17.1
11.8
We thank the analytical team (NCE Research) for all the analytical
support related to this publication. We also thank Glenmark Pharma-
ceuticals Limited for providing research facilities and consent for
publication.
Therefore, the difference in reactivity towards aldehyde binding assay
with various diastereomers clearly supported our hypothesis that
α
-amino acids with appropriate conformational rigidity on 1 (Figure 1)
Appendix A. Supplementary data
can overcome this aortic binding liability.
With encouraging data of compound 3p from aldehyde binding
assay, we developed a quick SAR around the 4-cyano group (Table 5).
The compounds 24a-e were synthesized by following the strategy
illustrated in Scheme 5, starting with 4-iodophenyl derivative 5. The
synthetic route followed was the same as described in scheme 2 for the
preparation of compounds 3a-q.
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.bmcl.2021.128202.
References
´
1
2
3
4
5
6
Turk D, Janjic V, Stern I, et al. EMBO J. 2001;20:6570.
Furber M, Tiden A, Gardiner P, et al. J Med Chem. 2014;57:2357.
Korkmaz B, Horwitz MS, Jenne DE, Gauthier F. Pharmacol Rev. 2010;62:726.
Korkmaz B, Caughey GH, Chapple I, et al. Pharmacol Ther. 2018;190:202.
Guay D, Beaulieu C, Percival MD. Curr Top Med Chem. 2010;10:708.
Podolin PL, et al. Inflammation Research. 2008;57(Suppl 2):S104 (Symposium
Abstract).
The replacement of 4-cyano group of 3p with 4-methylsulfone 24a
resulted in a 5-fold drop in potency and marginal improvement in
metabolic stability across all the species. On the other hand, analogs
24b-d were at par with 3p in terms of potency and metabolic stability. 3-
Methylbenzooxazolone substitution in case of 24e, caused>100-fold
7
8
9
Podolin PL, Bolognese BJ, Foley JP, et al. Am J Respir Crit Care Med. 2013;187:
A1500. www.atsjournals.org (Symposium Abstract funded by GlaxoSmithKline).
Korkmaz B, Lesner A, Marchand-Adam S, Moss C, Jenne DE. J Med Chem. 2020;63:
13258.
drop in potency to 1.2 μM. Overall, compounds 24a-e showed very poor
metabolic stability against human microsome and low to moderate
stability against rodent microsomes.
Laine DI, Busch-Petersen J. Expert Opin Ther Pat. 2010;20:497.
3p was found to be selective for Cathepsin C over Cathepsins K, L, S
and B. In addition, it did not show significant inhibition of CYP450
enzymes except CYP2D6 (Table 6). However, this compound was
rapidly metabolized in human, mouse and rat liver microsomes but
stable in plasma as tested in vitro (data not shown).
10 Selected International Patent Publications viz., WO 2010128324, WO 2010142985,
WO 2011025799, WO 2011075631, WO 2011154677, WO 2012119941, WO
2012109415, WO 2012112733, WO 2013041497, WO 2014140081, WO
2014140078, WO 2014140091, WO 2014140075, WO 2014091443, WO
2015032942, WO 2015110826, WO 2016016242, WO 2016038007 and WO
2020001392.
11 Thiele DL, Lipsky PE, McGuire MJ. U.S. Patent. 1997;5602:102.
12 Green GDJ, Shaw E. J Biol Chem. 1923;1981:256.
In conclusion, we have identified 3p, a potent and selective α-amino
acid based Cat- C inhibitor devoid of aortic binding possibilities. In
addition, it showed weak inhibition of the CYP450 enzymes except
CYP2D6 and having reasonably good selectivity over Cathepsins K, L, S
and B. Poor metabolic stability of 3p encouraged us to explore the op-
portunity of developing it for inhaled therapy to minimize the systemic
side effects of Cathepsin C inhibitors. Further efforts towards the
¨
13 Kam C-M, Gotz MG, Koot G, et al. Arch Biochem Biophys. 2004;427:123.
14 Yuan F, Verhelst SHL, Blum G, Coussens LM, Bogyo M. J Am Chem Soc. 2006;128:
5616.
15 Demuth H-U, Heiser U, Niestroj A. WO. 2003;03022871:A2.
16 Barrett AJ, Kembhavi AA, Brown MA, et al. Biochem J. 1982;201:189.
17 Bondebjerg J, Fuglsang H, Valeur KR, et al. Bioorg Med Chem. 2005;13:4408.
18 Horn M, Pavlik M, Doleckova L, Baudys M, Mares M. Eur J Biochem. 2000;267:3330.
19 Mucha A, Pawelczak M, Hurek J, Kafarski P. Bioorg Med Chem Lett. 2004;14:3113.
6

A. Banerjee et al.
Bioorganic & Medicinal Chemistry Letters 47 (2021) 128202
20 Thompson SA, Andrews PR, Hanzlik RP. J Med Chem. 1986;29:104.
21 Bondebjerg, J.; Fuglsang, H.; Naerum, L. WO2004110988, 2004.
22 Bondebjerg J, Fuglsang H, Valeur KR, Pedersen J, Naerum L. Bioorg Med Chem Lett.
2006;16:3614.
31 Takahashi M, Shioura Y, Murakami T, Ogasawara K. Tetrahedron-Asymmetry. 1997;8:
1235–1242.
32 Park H, Kwon Y, Shin JE, et al. Synthesis. 2017;49:2761–2767.
33 Lopez, O. D.; St. Laurent, D. R.; Goodrich, J.; Romine, J. L.; Serrano-Wu, M.; Yang, F.;
Kakarla, R.; Yang, X.; Qiu, Y.; Snyder, L. B. US 20110294819, 2011.
34 Harwood LM, Lilley IA. Tetrahedron Lett. 1993;34:537.
23 Guay D, Beaulieu C, Truchon JF, et al. M. Bioorg Med Chem Lett. 2009;19:5392.
24 Doyle K, Lonn H, Kack H, et al. J Med Chem. 2016;59:9457.
25 Bragg RA, Brocklehurst S, Gustafsson F, et al. Chem Res Toxicol. 1991;2015:28.
26 Falorni M, Porcheddu A. Giacomelli G. Tetrahedron: Asymm. 1999;1996:7.
27 Bryans JS, Horwell DC, Receveur, J-M. WO1999061424A1, 1999.
28 Taylor CA, Zhang Y-A, Snyder S. A. Chem Sci. 2020;11:3036–3041.
29 Taber DF, Kanai K, Jiang Q, Bui G. J Am Chem Soc. 2000;122:6807–6808.
30 Xu F, Murry JA, Simmons B, et al. Org Lett. 2006;8:3885–3888.
35 Kretzschmar, G.; Oehme, J.; Rossen, K. WO 2015189108, 2015.
36 Wang S, Su M, Wang J, et al. Bioorg Med Chem. 2014;22:6684.
37 Chemburkar, S. R.; Reddy, R. E.; Reamer, D. M.; Pavlina, J. T. Ulrey, S. S.; Kotecki, B.
J. WO 2011009021, 2011.
38 LePlae PR, Umezawa N, Lee H-S, Gellman S. H. J Org Chem. 2001;66:5629–5632.
39 Halgren TA, Murphy RB, Friesner RA, et al. J Med Chem. 2004;47:1750.
40 Wang Y, Du Y, Huang N. Future Med Chem. 2018;10:2713–2728.
7