Synthesis of potent and selective 2-azepanone inhibitors of human tryptase

Article abstract of DOI:10.1016/j.bmcl.2003.11.016

The serine protease tryptase has been associated with a broad range of allergic and inflammatory diseases and, in particular, has been implicated as a critical mediator of asthma. The inhibition of tryptase therefore has the potential to be a valuable therapy for asthma. The synthesis, employing solution phase parallel methods, and SAR of a series of novel 2-azepanone tryptase inhibitors are presented. A member of this series, 8t, was identified as a potent inhibitor of human tryptase (IC₅₀=38 nM) with selectivity ≤330-fold versus related serine proteases (trypsin, plasmin, uPA, tPA, APC, alpha-thrombin, and FXa).

Full text of DOI:10.1016/j.bmcl.2003.11.016

Bioorganic & Medicinal Chemistry Letters 14 (2004) 309–312
Synthesis of potent and selective 2-azepanone inhibitors
of human tryptase
Guohua Zhao,* Scott A. Bolton, Chet Kwon, Karen S. Hartl, Steven M. Seiler,
William A. Slusarchyk, James C. Sutton andGregory S. Bisacchi
The Bristol-Myers Squibb Pharmaceutical Research Institute, PO Box 5400, Princeton, NJ08543-5400, USA
Received25 August 2003; revised31 October 2003; accepted11 November 2003
Dedicated to the memory of Steven M. Seiler, PhD (deceased 31 March 2003). Steve’s untimely passing will not diminish the
continuing impact of his drug discovery research or the admiration of his fellow co-workers
Abstract—The serine protease tryptase has been associated with a broad range of allergic and inflammatory diseases and, in
particular, has been implicatedas a critical mediator of asthma. The inhibition of tryptase therefore has the potential to be a
valuable therapy for asthma. The synthesis, employing solution phase parallel methods, and SAR of a series of novel 2-azepanone
tryptase inhibitors are presented. A member of this series, 8t, was identified as a potent inhibitor of human tryptase (IC50=38 nM)
with selectivity 4330-foldversus relatedserine proteases (trypsin, plasmin, uPA, tPA, APC, alpha-thrombin, andFXa).
# 2003 Elsevier Ltd. All rights reserved.
1. Introduction
dine-containing azetidinones were shown to be effica-
Human tryptase is a structurally unique mast cell spe-
cific trypsin-like serine protease.1 Recent biological
studies have implicated tryptase as a mediator in the
pathology of allergic andinflammatory conidtions
including rhinitis, conjunctivitis, and most notably
asthma.2 In asthma, tryptase has been associatedwith
the acute bronchospastic response, the underlying
inflammatory disease, and long-term airway remodel-
ling. Approximately 150 million people suffer from
asthma worldwide and the problem is growing.3 The
currently available therapy for asthma involves a step-
wise approach in which patients are given increasingly
more efficacious medications to achieve control at the
cious at improving lung function andreudcing lung
inflammatory cell infiltration in an ovalbumin-sensitized
guinea pig model. BMS-363131 and its analogues were
designed for inhalation treatment. In a second phase of
our program we sought potent andselective non-
mechanism-basedtryptase inhibitors having potential
for oral administration.
4
risk of increasedside effects. Therefore new andeffec-
tive drugs are urgently needed for the treatment of
asthma. The design and evaluation of tryptase inhibi-
tors for the treatment of asthma andother inflamma-
tory diseases has been the subject of intense
investigation.5 Our previous report described BMS-
363131 (1) as a potent, mechanism-basedhuman tryp-
tase inhibitor (IC50=2 nM) having exceptionally high
selectivity (>1500-fold) versus related serine proteases
including trypsin, thrombin, FXa, tPA and plasmin.6
Intratracheally dosed BMS-363131 and related guani-
Directedscreening in our laboratories resultedin the
discovery of azepanone 2 as a weak tryptase inhibitor
(IC50=4.43 mM) which was utilizedas a structural
* Corresponding author. Tel.:+1-609-818-5553; fax: +1-609-818-
3450; e-mail: guohua.zhao@bms.com
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.11.016

310
G. Zhao et al. / Bioorg. Med. Chem. Lett. 14 (2004) 309–312
Scheme 1. (a) LiHMDS, BrCH2CO2Bn, THF, 83%; (b) TFA, CH2Cl2, 99%; (c) biphenyl-4-carboxylic acid, EDC, HOBT, iPr2EtN, CH2Cl2, 85%;
(d) H2, Pd/C, THF, 100%; (e) R-NH2 amines, polystyrene-EDC, iPr2NEt, DMF–ClCH2CH2Cl (1:2), polystyrene-trisamine; (f) 20% TFA in
ClCH2CH2Cl for Boc protectedamines; reverse phase preparative HPLC; 73–94% for two steps.
starting point in an attempt to improve tryptase inhibi-
tory potency. In this letter, we report the solution phase
parallel synthesis of a series of racemic 2-azepanone
analogues of 2 exploring replacements to the aminocy-
clohexylmethyl andbiphenyl substituents leading to the
identification of analogue 8t as a potent andselective
non-mechanism-basedinhibitor of human tryptase.
diamine 7g was synthesizedfrom 7-nitro-tetralone via
oxime mediated hydrogenation followed by Boc protec-
tion of the primary amine.
1-[(3-Aminomethyl-phenyl-carbamoyl)-methyl]-2-oxo-
azepan-3-yl amides 12a–k were synthesizedfrom ester 9,
7
which was preparedusing Skiles’ procedure
(Scheme
3). The hydrolysis of ester 9 with TFA followedby the
reaction with 3-N-Boc aminomethyl aniline (7a) gave
acid 10. Removal of the Cbz group via hydrogenolysis
andcoupling with a variety of acids ACO 2H generated
amides 11 after purification with cation exchange resin.
Subsequently, Boc deprotection provided amides 12a–k.
2. Chemistry
The preparation of biphenyl-4-carboxylic acid(1-car-
bamoylmethyl-2-oxo-azepan-3-yl)-amides 8a–t listedin
Table 1 is outlinedin Scheme 1. Alkylation of 2-azapa-
none 3 with bromoacetic acidbenzyl ester followedby
Boc deprotection gave amine 4. Acylation of 4 with
biphenyl-4-carboxylic acidfollowedby reductive clea-
vage of the benzyl ester gave acid 5. Acid 5 was then
coupled under standard conditions with a variety of
monoamines anddiamines. Where requiredfor selective
coupling, mono-Boc protected diamines were used. Boc-
protected penultimate intermediates were deprotected
using TFA in methylene chloride to afford the final
products 8a–e, 8h–o and 8r–t. Coupling with unpro-
3. Results and discussion
To broadly screen for replacements of the amino-cyclo-
hexylmethyl moiety in 2 over a hundred corresponding
amide analogues of 2 were initially synthesizedusing
commercially available amines. Although neutral R
(Scheme 1) groups were included, an emphasis was
placedon R groups incorporating basic amines since
our working hypothesis was that the aminocyclohexyl
substituent of leadinhibitor 2 may be occupying the S1
pocket of tryptase with the primary amine forming a
salt bridge to Asp 189.1 A subset of these analogues is
shown in Table 1 (compounds 8a–l). Alkylamine analo-
gues of various chain lengths (8a–c) as well as the ami-
nomethylcyclohexyl (8d) or piperidinylmethyl (8e)
analogues were poorly potent for tryptase, as were the
aniline derivitives 8f and 8g. More interesting results
tectedaminoalkyl anilines affordedproducts
8f and 8g
directly, while coupling with m-diethylaminomethyl ani-
line afforded product 8p. Additionally, m-amino-
methylbenzamide was coupled with 5 to affordproduct 8q.
The Boc protecteddiamines 7a–b, 7d–f, anddiamine 7c
usedfor the synthesis of 8l, 8o, 8r–t, and 8p, respectively,
were prepared from the corresponding carboxamido
anilines 6a–f by reduction with lithium aluminum
hydride in THF followed by Boc protection of the pri-
mary amine (Scheme 2). Additionally, the Boc protected
were obtainedfrom aminoalkylphenyl
R groups.
Whereas the p-aminoethylphenyl, o-aminomethylphe-
nyl, and m-aminomethylphenylmethyl analogues (8h–j,
respectively) were poorly potent, the p-aminomethyl-
phenyl analogue 8k showedactivity (IC 50=3.64 mM) at
least as goodas leadcompound 2. However, the corre-
sponding m-aminomethyl analogue, 8l hadan IC
of
50
0.25 mM, a 10-foldimprovement in potency compared
to 2 and 8k.8
Next we decided to use 8l as the starting point for a
study of the replacement of the biphenylamide sub-
stitutent. Over 50 amide analogues 12 (Scheme 3) were
prepared, and a subset of these is shown in Table 2. In
this series, substitution on the terminal phenyl ring (12a–
c) or extending the rings by insertion of a methylene
group (12d,e) ledto less potent compounds, suggesting a
binding pocket of limited size. Replacement of the termi-
nal phenyl ring by an aliphatic ring or chain (12f,g) also
decreases the activity. Several other lipophilic mono- or
bicyclic aromatic groups at various linker lengths also
Scheme 2. (a) LAH, THF; (b) Boc2O, THF: 90–95%; (c) 2 N HCl in
Et2O, 74–96%; (e) HONH2, pyridine, 93%; (f) H2, PtO2, EtOH–
CHCl3 (10:1), 95%.

G. Zhao et al. / Bioorg. Med. Chem. Lett. 14 (2004) 309–312
Table 1. Tryptase inhibition for compounds 8a–t Table 2. Tryptase inhibition for compounds 12a–k
311
CompdR
Tryptase IC
50 (nM)
>33,000
CompdA
Tryptase IC
50 (nM)
8a
12a
1300
8b
8c
12,000
12b
12c
12d
12e
12f
12g
12h
12i
2900
2100
5200
980
>33,000
8d
8e
12,000
>33,000
8f
>33,000
>33,000
11,000
>33,000
13,000
3600
980
8g
8h
8i
4600
2900
4600
15,000
200
8j
12j
8k
8l
12k
250
8m
8n
8o
8p
8q
8r
8s
8t
>33,000
>33,000
>33,000
>33,000
>33,000
890
afforded less potent compounds (12h–j). Interestingly,
the branched 3,3-diphenylpropionamide analogue 12k
emergedas essentially equipotent to 8l. However, in this
limitedstuyd no clear improvement in potency was
achievedby replacement of the biphenylamide.
At this point, we returnedto 8l to explore the effect of
conformational constraints or substitution on the m-
aminomethylphenyl substitutent. Constrainedanalo-
gues 8m and 8n were inactive while small alkyl
substitution on the terminal amino groups also resulted
in loss of activity (8o,p). The corresponding neutral
amide analogue (8q) of 8l was also poorly potent, fur-
ther suggesting the preference for a basic nitrogen on
this part of the molecule. Small substitutents para posi-
tion to the methyl amine group were tolerated. Methoxy
substitutedanalogue 8r (IC50=0.89 mM) was somewhat
less potent than the unsubstitutedparent 8l. However,
methyl andchloro substitution affordeda substantial
boost in potency comparedto 8l. The methyl analogue 8s
hadan IC 50 of 74 nM andthe chloro analogue 8t hadan
IC50 of 38 nM, approximately a 10-foldimprovement in
potency comparedto the unsubstitutedparent 8l, and
100-foldimprovement comparedwith the original lead
compound 2.
74
38

312
G. Zhao et al. / Bioorg. Med. Chem. Lett. 14 (2004) 309–312
Scheme 3. (a) TFA, CH2Cl2; (b) 7a, EDC, HOBT, NMM, CH2Cl2, 78% for two steps; (c) H2, Pd/C, THF, 100%; (d) acids, EDC, DMAP, DMF–
ClCH2CH2Cl (1:2); SCX cation exchange column purification; (e) TFA, ClCH2CH2Cl; reverse-phase preparative HPLC if necessary; 78–97% for
two steps.
Table 3. Tryptase inhibition andselectivities for selectedcompounds
selectivity against other serine protease including
trypsin.
Enzyme
Compound
8l
12k
8s
8t
Acknowledgements
Tryptase
IC50 (nM)
250
200
74
38
We acknowledge Dr. Martin Ogletree for providing the
management support that enabledthe progression of
compounds through biology testing.
Selectivity ratio
Trypsin
Plasmin
uPA
tPA
APC
>130
>130
>130
>130
>130
>130
>130
>170
>170
>170
>170
>170
>170
>170
230
>450
>450
>450
>450
>450
>450
360
330
>870
>870
>870
>870
>870
References and notes
Thrombin
FXa
1. Pereira, P. J. B.; Bergner, A.; Macedo-Ribeiro, S.; Huber,
R.; Matschiner, G.; Fritz, H.; Sommerhoff, C.; Bode, W.
Nature 1998, 392, 306.
Compounds 8l, 12k, 8s, and 8t were screenedfor activ-
ity against a panel of relatedserine proteases ( Table 3)
including trypsin, plasmin, urokinase plasminogen acti-
vator (uPA), tissue plasminogen activator (tPA), acti-
vatedprotein C (APC), alpha-thrombin (FIIa), and
FXa. All compounds were selective against this panel.
In particular, 8t, the most potent compoundof this ser-
ies, exhibited >870-foldselectivity versus uPA, tPA,
APC, thrombin, andFXa, andwas 330-foldselective
against plasmin and360-foldselective against trypsin.
Unlike the beta-lactam BMS-363131 andits analogues,
these azepanones are not reactive mechanism-based
inhibitors. Further they do not incorporate a highly
basic guanidine group, and thus may have potential for
oral administration.
2. (a) Elrod, K. C.; Numerof, R. P. Emerg. Ther. Targets
1999, 3, 203. (b) Zhang, M.-Q.; Timmerman, H. Mediat.
Inflamm. 1997, 6, 311. (c) Abraham, W. Am. J. Physiol.
Lung Cell Mol. Physiol. 2002, 282, L193.
3. WHO Fact Sheet No. 206, revisedJanuary 2000. Availible
at http://www.who.int/inf-fs/en/fact206 .html.
4. (a) Greenfeder, S.; Anthes, J. C. Curr. Opin. Chem. Biol.
2002, 6, 526. (b) Bjermer, L. Respir. Med. 2001, 95, 703.
(c) White, M. V.; Sander, N. J. Allergy Clin. Immunol.
1994, 104, S47.
5. (a) Newhouse, B. IDrugs 2002, 5, 682. (b) Burgess, L. E.
Drug News Perspect. 2000, 13, 147.
6. Slusarchyk, W. A.; Bolton, S. A.; Hartl, K. S.; Huang,
M.-H.; Jacobs, G.; Meng, W.; Ogletree, M. L.; Pi, Z.;
Schumacher, W. A.; Seiler, S. M.; Sutton, J. C.; Treuner,
U.; Zahler, R.; Zhao, G.; Bisacchi, G. S. Bioorg. Med.
Chem. Lett. 2002, 12, 3235.
7. Skiles, J. W.; Sorcek, R.; Jacober, S.; Miao, C.; Mui,
P. W.; McNeil, D.; Rosenthal, A. S. Bioorg. Med. Chem.
1993, 3, 773.
8. Tryptase inhibitors containing benzylamine functionality
have recently been disclosed by others; see ref 5a and
references therein.
In summary, we have studied the parallel synthesis
andSAR of a series of N-1 andC-3 2-azepanone
tryptase inhibitors. As the result of the preparation
of three libraries, compound 8t has been identified as
a potent inhibitor of human tryptase with high