J. Med. Chem. 2007, 50, 1727-1730
1727
in terms of its component actions, we set out to identify and
study selective inhibitors of Cat G and chymase. We now report
on the discovery of a novel series of nonpeptide chymase
inhibitors possessing a â-carboxamido-phosphon(in)ic acid motif
(2). This chemical series has yielded potent, selective inhibitors
that exhibit useful oral bioavailability and promising anti-
inflammatory pharmacology.
Discovery of Potent, Selective, Orally Active,
Nonpeptide Inhibitors of Human Mast Cell
Chymase
Michael N. Greco,*,† Michael J. Hawkins,†
Eugene T. Powell,† Harold R. Almond, Jr.,†
Lawrence de Garavilla,† Jeffrey Hall,† Lisa K. Minor,†
Yuanping Wang,† Thomas W. Corcoran,† Enrico Di Cera,‡
Angelene M. Cantwell,‡,§ Savvas N. Savvides,‡,|
Bruce P. Damiano,† and Bruce E. Maryanoff*,†
Research and Early DeVelopment, Johnson & Johnson
Pharmaceutical Research and DeVelopment, Spring House,
PennsylVania 19477-0776, and Department of Biochemistry and
Molecular Biophysics, Washington UniVersity, School of Medicine,
St. Louis, Missouri 63110
ReceiVed January 16, 2007
Abstract: A series of â-carboxamido-phosphon(in)ic acids (2) was
identified as a new structural motif for obtaining potent inhibitors of
human mast cell chymase. For example, 1-naphthyl derivative 5f had
an IC50 value of 29 nM and (E)-styryl derivative 6g had an IC50 value
of 3.5 nM. An X-ray structure for 5f‚chymase revealed key interactions
within the enzyme active site. Compound 5f was selective for inhibiting
chymase versus eight serine proteases. Compound 6h was orally
bioavailable in rats (F ) 39%), and orally efficacious in a hamster
model of inflammation.
The original lead compound that led us to 1 was â-ketophos-
phonic acid 3a, which was identified via high-throughput
screening for inhibitors of human Cat G.10 In exploring related
structures as potential Cat G inhibitors, we prepared aza-
homologue 3b.11 This compound was devoid of Cat G inhibition
(0% inhibition at 100 µM), but had moderate potency against
human chymase (IC50 ) 190 nM).12 Because there is a close
sequence homology (>50%) between Cat G and chymase, and
close structural homology between their active sites (∼80%),
3b was modeled into the chymase active site in the mode
observed for 3a‚Cat G.10 Thus, the 1-naphthyl and 2-naphthyl
groups were oriented in the S2 and S1 pockets, respectively
(Figure S1, Supporting Information). The vacant, hydrophobic
S3/S4 region was seen as having potential for additional binding
interactions via attachment of appropriate substituents to the
3-position of the 2-naphthyl group, as exemplified in 4.10 While
this approach parallels the one that we used to progress from
3a to 1, with a 100-fold potency enhancement,8 various attempts
to effect this outcome with 3b met with disappointment. For
example, the chymase IC50 value for 4 was >10 µM. Faced
with this outcome, we decided to systematically investigate
structural modifications of molecule 3b.
The 1-naphthyl ring, thought to occupy the S2 subsite of
chymase, was replaced by different bicyclic heterocycles. We
synthesized the target compounds by the route outlined in
Scheme 1, which is exemplified for 5e (Table 1). Triethylphos-
phite and 3-(bromomethyl)benzothiophene were heated at reflux
to furnish diethyl (3-benzothienyl)methylphosphonate (57%
yield),13 which was deprotonated in THF with butyllithium (2.6
M in hexane), reacted with 2-naphthylisocyanate, and de-
esterified with bromotrimethylsilane in pyridine to give 5e (11%
yield).
A variety of cell types, such as neutrophils, mast cells,
monocytes, macrophages, eosinophils, and lymphocytes, can
play a major role in immune-mediated inflammatory diseases.1
However, whereas the recruitment of such cells into injured
tissues is crucial to tissue repair and host defense,2 this
accumulation can also be deleterious. One source of the adverse
actions is secreted proteases.3 During inflammatory events, tissue
damage4 ensues when there is an excess of protease activity
that overwhelms the levels of endogenous protease inhibitors.5
Consequently, the administration of exogenous inhibitors could
rectify this imbalance and provide a therapeutic benefit, such
as in pulmonary inflammatory diseases like asthma and chronic
obstructive pulmonary disease (COPD).
Because asthma and COPD are serious, unmet medical needs
with an expanding incidence worldwide, we developed a keen
interest in the serine proteases cathepsin G (EC 3.4.21.20; Cat
G)4,6 and chymase (EC 3.4.21.39).6a,7 These chymotrypsin-like
enzymes can degrade extracellular matrix (e.g., collagen, elastin,
proteoglycans, and fibronectin), induce leukocyte migration, and
promote tissue remodeling.4,6a,7a,b We recently described 1 (JNJ-
10311795) as a potent dual inhibitor of human Cat G (Ki ) 38
nM) and human chymase (Ki ) 2.3 nM), and established its
molecular interactions in the active site of each enzyme.8 In
investigating the biological properties of 1, we have observed
notable activity in several animal models of inflammation.8,9
To gain a better understanding of the pharmacology, especially
* To whom correspondence should be addressed. Tel.: 215-628-5614
(M.N.G.); 215-628-5530 (B.E.M.). Fax: 215-628-4985 (M.N.G. and
B.E.M.). E-mail: mgreco@prdus.jnj.com (M.N.G.); bmaryano@prdus.jnj.com
(B.E.M.).
Considering the set of phosphonic acid compounds, 3b and
5a-g, potency for inhibition of human chymase varied over a
200-fold range (Table 1). Relative to the initial 1-naphthyl
derivative, 3b, the 3-benzofuranyl (5a) and 3-indolyl (5b)
analogues were less potent, the 3-benzothienyl (5e) and 3-(1-
methylindolyl) (5c) analogues were marginally more potent, but
† Johnson & Johnson Pharmaceutical Research and Development.
‡ Washington University, School of Medicine.
§ Present address: Department of Microbiology and Immunology,
University of Texas Health Science Center, San Antonio, TX 78229.
| Present address: Laboratory for Protein Biochemistry and Protein
Engineering, Ghent University, 9000 Ghent, Belgium.
10.1021/jm0700619 CCC: $37.00 © 2007 American Chemical Society
Published on Web 03/16/2007