Fragment-based discovery of indole inhibitors of matrix metalloproteinase-13

Article abstract of DOI:10.1021/jm201129m

Matrix metalloproteases (MMPs) play an important role in cartilage homeostasis under both normal and inflamed disease states and, thus, have become attractive targets for the treatment of arthritic diseases. Herein, we describe the identification of a pot

Full text of DOI:10.1021/jm201129m

Article
pubs.acs.org/jmc
Fragment-Based Discovery of Indole Inhibitors of Matrix
Metalloproteinase-13
Steven J. Taylor,*,† Asitha Abeywardane,† Shuang Liang,† Ingo Muegge,† Anil K. Padyana,†
Zhaoming Xiong,† Melissa Hill-Drzewi,† Bennett Farmer,† Xiang Li,† Brandon Collins,† John Xiang Li,†
Alexander Heim-Riether,† John Proudfoot,† Qiang Zhang,† Daniel Goldberg,‡ Ljiljana Zuvela-Jelaska,†
Hani Zaher,§ Jun Li,† and Neil A. Farrow†
†Boehringer Ingelheim Pharmaceuticals Inc., 900 Ridgebury Road, Ridgefield, Connecticut 06877-0368, United States
‡Karos Pharmaceuticals, 5 Science Park # 2, New Haven, Connecticut 06511, United States
§Eugene Applebaum College of Pharmacy and Health Sciences, 259 Mack Avenue, Detroit, Michigan 48201, United States
S
* Supporting Information
ABSTRACT: Matrix metalloproteases (MMPs) play an important role in cartilage homeostasis under both normal and inflamed
disease states and, thus, have become attractive targets for the treatment of arthritic diseases. Herein, we describe the
identification of a potent, selective MMP-13 inhibitor, developed using fragment-based structure-guided lead identification and
optimization techniques. Virtual screening methods identified a novel, indole-based MMP-13 inhibitor that bound into the S1′
pocket of the protein exhibiting a novel interaction pattern hitherto not observed in MMP-13 inhibitors. X-ray crystallographic
structures were used to guide the elaboration of the fragment, ultimately leading to a potent inhibitor that was >100-fold selective
over nine other MMP isoforms tested.
typically have a higher hit rate than traditional HTS due to the
fact that smaller compounds are less likely to have mismatched,
unfavorable interactions with the target protein binding site.
Fragments typically bind to their target protein with low
affinity (μM to mM range); thus, a number of sensitive
biophysical techniques can be leveraged to help verify and
triage fragment hits prior to X-ray structure determination.
Biophysical methods such as nuclear magnetic resonance
(NMR),4 surface plasmon resonance (SPR),5 and calorimetry6
have all been developed to detect weak interactions between
fragments and their target protein. As of 2010, at least 13
compounds have been advanced into phase 1 and phase 2
clinical trials discovered with the help of fragment-based lead
identification methodologies.7
Our strategy involved fragment virtual screening and FBS as
a supplement to HTS to discover selective inhibitors of matrix
metalloprotease (MMP)-13 since our HTS screening did not
deliver ideal lead options. MMP-13 was viewed as an attractive
target for FBS because the protein could be readily crystallized.
Inhibition of MMP-13. MMPs are zinc- and calcium-
dependent peptidases, involved in the cleavage of collagen,
INTRODUCTION
■
Fragment-Based Screening (FBS). As productivity in the
pharmaceutical industry has decreased over the past 20 years,
many companies have begun to seek out alternative methods
for discovering druglike leads for “nondruggable” proteins to
complement the traditional high-throughput screening (HTS)
screening of corporate libraries. One method that has gained
acceptance over the past decade is screening low molecular
weight, highly soluble “fragments” against a particular target of
interest.1 These fragments are then evolved, with the aid of
crystallographic costructure data, into lead chemical series. In
its most simple form, fragment screening relies on triaging and
prosecuting hits based more on their ligand efficiency (defined
by binding free energy or potency/number of nonhydrogen
atoms) rather than on their intrinsic potency.2 A primary
rationale for screening fragments [molecular weight (MW) ≤
350 atomic mass units (AMUs)], as opposed to larger
molecules, is that they access a broad chemical space (relative
to molecular size), while screening a limited number of
compounds. Additionally, smaller-sized fragment libraries can
be prescreened for favorable physiochemical properties such as
solubility, log P, and lack of aggregation, such that hits from a
fragment screen would inherently have druglike properties
embedded within the starting scaffold.3 Fragment screens
Received: August 23, 2011
Published: October 21, 2011
© 2011 American Chemical Society
8174
dx.doi.org/10.1021/jm201129m|J. Med. Chem. 2011, 54, 8174−8187

Journal of Medicinal Chemistry
Article
gelatin, and other proteins in the extracellular matrix. MMPs
have been implicated in a wide range of disorders ranging from
oncology to infectious diseases, and as such, a number of pan-
MMP inhibitors have entered clinical trials.8 There are
approximately 23 known human MMPs that are grouped into
subtypes based upon their substrates. To date, however, broad
spectrum MMP inhibitors have been compromised by side
effects, primarily muscular skeletal syndrome (MSS), which is
characterized by painful stiffening of the joints. MSS is thought
to occur by inhibition of intrinsic cellular matrix degradation,
likely due to inhibition of MMPs other than MMP-13.9 MMP-
13 is the most efficient enzyme of this class at degrading
collagen and has been implicated in the development of
osteoarthritis and rheumatoid arthritis.10 As such, selective
inhibition of MMP-13 has become an attractive target for small
molecule intervention.
protein. While several interesting hits were identified, none
were more attractive than the 5-carbamoyl-1H-indole-2-
carboxylic acid ethyl ester identified by virtual screening (1,
Figure 1) due to its high ligand efficiency (>0.3) in the high
Figure 1. Indole-based fragment hits found from virtual screening with
micromolar MMP activity.
While the subunits of each class of MMPs differ, the catalytic
domains share the canonical zinc-binding amino acid sequence
HExxHxxGxxH.11 Classical, broad spectrum inhibitors gained
potency via direct interaction with the zinc residue, which
inevitably made designing selectivity into these inhibitors
challenging.12 Outside the active site, MMPs differ specifically
in the S1′ pocket in a profound manner,13 providing an
opportunity to gain selectivity. MMP-13, in particular,
possesses a large S1′ pocket that, when appropriately occupied
by a ligand, opens up a side pocket (S1′*) not observed in
many other MMPs.14 The structural significance of this pocket
has been exploited by a number of companies, resulting in
potent selective MMP-13 inhibitors that do not interact with
the catalytic zinc. These inhibitors share common binding
motifs: An aromatic group π-stacks against histidine 222, direct
or water-mediated hydrogen bonds are made with residues
(threonine 245, threonine 247, and methionine 253) in the
selectivity loop, and an aromatic group is located in the
hydrophobic portion of the S1′* pocket formed by residues
phenylalanine 217 and leucine 218.
concentration MMP-13 catalytic assay and novel binding mode
within the S1′ pocket of MMP-13.
Figure 2a shows the binding mode of 1 as determined by X-
ray crystallography. Although serendipity has played a role in
virtual screening successes in the past,17 it was still a surprise
that fragment 1 was found to bind shallower in the S1′ pocket
than expected. The ethyl-ester moiety binds adjacent to
histidine 222, one of the three histidine residues that chelate
the catalytic zinc. The indole core is positioned deeper in the
S1′ pocket, with the indole nitrogen forming a hydrogen bond
with the backbone carbonyl of phenylalanine 241. This
interaction had not been described before in the literature of
MMP-13 inhibitors. The nitrogen of the amide group engages
the selectivity loop of the protein though hydrogen bonds with
threonine 245 and water-mediated with threonine 247.
Fragment 1 showed some modest activity over MMP-2 and
-14, which was consistent with a partial occupancy of the S1′
selectivity pocket of MMP-13.
It has previously been shown that MMP-13 selectivity may
be achieved when ligands bind further into the S1′ pocket,
occupying the S1′* subpocket (PDB: 1XUD). Figure 2b shows
an overlay of compound 1 and a selective literature inhibitor. It
is apparent from the figure that the indole fragment is not
accessing the region of the protein that has been reported to
confer selectivity and thus providing a roadmap for increasing
potency against MMP-13, while sparing activity against the
other maxtix metalloproteases.
The initial structure−activity relationship (SAR) focused on
determining if the ethyl ester at the 2-position of the indole was
required for activity since this moiety was viewed as a potential
metabolic liability. On the basis of the cocrystal structure, this
functional group participates only in weak, water-mediated
interactions and potential van der Waals interactions with the
lipophilic side chains of the protein. The SAR in this region is
shown in Table 1. Replacement of the ether oxygen with a
neutral atom (−CH2− 2) or a hydrogen bond donor (NH, 3)
resulted in a complete loss of activity (ketone, amine, and ester
are all in the planar confirmation). This finding suggested the
necessity of a hydrogen bond acceptor (HBA), despite the fact
that the crystal structure did not indicate any direct hydrogen
bonding to the ether oxygen. Maintaining the HBA but
constraining the element within a ring resulted in an analogue
that retained some activity against MMP-13 (Table 1, 5).
Oxazolines similar to that in compound 5 are know to be
hydrolytically liable, so further elaboration of the molecule off
the 6-position of the indole was performed with the ethoxy
ester as an anchoring functional group.
As few examples of selective inhibitors15 were known at the
time of our internal uHTS campaign, we initiated a virtual
screening and fragment-based approach to discover high ligand
efficiency leads with the aim of then evolving them into
selective inhibitors using structure-guided methods to comple-
ment our chemistry strategy.
RESULTS AND DISCUSSION
■
Prior to high concentration screening, we preformed a virtual
screen of our entire corporate library to enrich our fragment
screening deck with compounds that had the potential to bind
within the S1′ pocket of MMP-13.16 Our fragment library of
approximately 1000 diverse scaffolds enriched with the hits
from the virtual screen was screened in a MMP-13 molecular
assay at high inhibitor concentration (500 μM). Among the hits
were two related compounds derived from the virtual screening
efforts that exhibited modest MMP-13 inhibition yet ligand
efficiencies as good as those of the most potent MMP-13
inhibitors reported in the literature (1 and 1a). In parallel, the
fragment library was screened in two additional primary assays
[saturation transfer difference (STD) NMR and size exclusion
chromatography mass spectrometry (SEC-MS)] to detect weak
binding interactions. Fragment hits were prioritized based upon
overlapping activity in the three assays as well as novelty, as
defined by similarity to known MMP-13 scaffold classes.
Fragments were then prosecuted in high-throughput crystal-
lization experiments to elucidate their binding mode within the
8175
dx.doi.org/10.1021/jm201129m|J. Med. Chem. 2011, 54, 8174−8187

Journal of Medicinal Chemistry
Article
Figure 2. (a) X-ray crystal structure of 1. The region of the protein referred to as the “selectivity loop” is colored red. The catalytic zinc atom is
depicted as a gray sphere, while stably bound water molecules are shown as red spheres. (b) Overlay of 1 with the MMP-13 selective, nonzinc-
chelating Aventis inhibitor (PDB: 1XUD) that accesses the S1′* pocket.
Table 1. SAR for Substitution at the 2-Position of Indole
Fragment 1
Table 2. SAR for Substitution at the 5-Position of Indole
Fragment 2
a
IC50 values are represented as μM and averaged from a minimum of
two determinations; LE calculated by RT ln(IC50)/no. of heavy atoms.
On the basis of the X-ray costructure of 2 with MMP-13 and
literature compounds (PDB: 1XUD), we sought to grow the
fragment in the direction of the S1′* pocket to gain potency
and selectivity over all other MMP isoforms. In addition to
testing activity against MMP-13, compounds were tested for
their ability to inhibit MMP-2 and MMP-14, two isoforms that
predict selectivity against a wider panel of metalloproteases.18
Initial attempts at increasing potency and selectivity were
focused on replacing the 6-carboxamide with an aromatic ring
mimicking known selective literature inhibitors.
a
IC50 values are represented as μM and averaged from a minimum of
two determinations; LE calculated by RT ln(IC50)/no. of heavy atoms.
Simple replacement of the amide with a phenyl analogue
resulted in complete loss of enzyme inhibition. Adding a
heteroatom to the ring resulted in increased potency with the 3-
and 4-pyridyl analogues (Table 2, entries 4 and 5), presumably
due to a hydrogen bond interaction with the side chain of
threonine 247, which is not available when the HBA is α to the
attachment indole ring (Table 2, entry 3). Substitution on the
distal aryl ring α to the biraryl connection was envisioned to
Replacement of the carboxamide with an aryl ring was
envisioned to provide a more rigid scaffold that could be further
elaborated to access deep within the S1′* pocket to gain
selectivity. Carboxyamide replacements are shown in Table 2.
8176
dx.doi.org/10.1021/jm201129m|J. Med. Chem. 2011, 54, 8174−8187

Journal of Medicinal Chemistry
Article
increase potency by (A) twisting the aromatic ring out of plane
to give a dihedral angle of 46°, the more preferred 40°
mirroring the angle observed in the advanced literature
inhibitors, and (B) positioning a small lipophilic group in the
direction of a small lipophilic notch generated by proline 255
not seen in other MMPs. Methyl pyridyl 10 and N-methyl
imidazole 11 were both tolerated and showed increased
molecular potency. In addition, selectivity against MMP-2
and -14 was >10-fold. To confirm this increase in potency and
selectivity, a crystal structure of imidazole 11 was obtained
(Figure 3). The crystal structure revealed that the gain in
relative to the rest of the scaffold. All of the compounds showed
selectivity over MMP-2 and -14, validating the strategy of
occupying the S1′* pocket as a means to achieve selectivity with
little shift in the presence of exogenous protein highlighting
their potential.
To confirm the binding interactions leveraged by the
elaborated molecules in Table 3 and to further optimize
these analogues, a costructure of compound 15 was obtained
(Figure 4). The structure revealed that the terminal benzoic
acid of the compound accesses the S1′* pocket. The terminal
acid moiety makes a direct hydrogen bond with the Nz
nitrogen of lysine 140, and water mediated interactions with the
backbone carbonyl of serine 209 and the hydroxyl group of
tyrosine 246. Additionally, the inhibitor's position in the S1′
pocket is reinforced by two water-mediated hydrogen bond
interactions with the backbone amide nitrogen of Met253 and a
direct hydrogen bond interaction with the backbone amide
nitrogen of threonine 274.
On the basis of its excellent potency and selectivity profile,
compound 15 was further profiled in more physiologically
relevant metalloprotease assays as well as advanced into
pharmacokinetic profiling. Compound 15 was potent in a
full-length MMP-13 collagen degradation assay (11 nM) and
was able to inhibit degradation of bovine nasal cartilage with an
IC50 of 31 nM. In addition 15 was >1000× selective over nine
other MMPs tested (Figure 5). Passive permeability was
predicated to be high, based upon parallel artificial membrane
permeability assay (PAMPA) data (data not show); however,
significant efflux was observed in human CACO-2 cells, which
is consistent with compounds with comparable polar surface
areas (PSAs) to 15. The in vitro profile was acceptable in h/
rLM, even with the perceived ester as a liability, and as such,
this analogue was moved into rat pharmacokinetics. When
dosed orally at 10 mpk (iv 1 mpk), compound 15 reached
micromolar plasma levels, displayed modest clearance, and
showed acceptable bioavailability (39%). The Vss of the
compound was quite low at 0.26, in line with other carboxylic
containing compounds. To better understand the metabolic
fate of 1, a metabolic ID study upon incubation with rat liver
microsomes was performed. The major metabolite was the
result of ester hydrolysis to give 18 (Table 4, entry 2), which
accounted for >80% of all metabolites formed.
A number of analogues were generated to follow up on the
observed primary metabolite of 15 (Table 4). The diacid
metabolite (18) was inactive against MMP-13, reiterating the
importance of this ester functional group. Strikingly, complete
removal of this functional group, representing removal of only
four atoms in the molecules, resulted in a complete loss of
MMP-13 inhibition. Replacement of the ester, with an ethyl
ether that contained the acceptor but not the carboxyl group,
partially recovers some the activity lost upon remove of the
ester (Table 4, 20, entry 4). These experiments highlight the
critical contribution of this moiety to overall activity, despite
the limited direct interactions observed between the ester and
the protein.
Figure 3. (a) Costructure of 11 bound to MMP-13 and (b) the
imidazole methyl group is directed into a subpocket of S1′, toward
residue Pro255.
selectivity is likely due to the positioning of the methyl group in
an MMP-13-specific pocket formed by the presence of proline
255.
Guided by the costructure of compound 11, we further
elaborated the core scaffold to determine if additional potency
and selectivity could be obtained by fully occupying the S1′*
pocket of the enzyme (Table 3). To access the S1′* pocket
necessitated substitution of the 3-position of the imidazole;
thus, we exchanged the imidazole core for a pyrazole. This
modification had the additional benefit of allowing for another
interaction between the pyrazole nitrogen and a tightly bound
water within the active site. Substitution off of the aryl ring with
a carboxyl group was tolerated (12), which, based on modeling,
provided a synthetic handle for further elaboration into the S1′*
subpocket. In addition, this modification maintained some
selectivity against MMP-2 and -14. Following SAR from
previous inhibitors, the ester was replaced with a benzyl amide
containing various distal HBAs at the 4-postion of the aromatic
ring. The amide modification was hypothesized to rigidify the
core structure, reinforcing the water-mediated hydrogen bond
to the pyrazole and define the proper trajectory of the aryl ring
and placement of the HBA in proximity to interact with
Lys140. The simple 4-pyridyl amide (Table 3, 13, entry 2)
provided the first submicromolar MMP-13 inhibitor and
displayed strong selectivity over MMP-2 and -14 (>5000-
fold). A similar increase in potency was observed with the
pyridine core (compound 17), which had comparable potency
to the pyrazole core but with slightly lower ligand efficiency.
Excising the HBA out of the ring by replacing the pyridine ring
with a pyridone moiety (14) resulted in an analogue that
maintained potency and some selectivity over MMP-12 and -14
(Table 3, entry 3). Replacement of the hydrogen bond donor
(HBD) with a carboxylic acid further improved the potency of
the inhibitors approximately 10-fold, presumably by interacting
via a salt bridge with Lys140 (Table 3, 15, entry 4). Extending
the carboxylate moiety one methylene carbon away from the
aromatic ring (16) resulted in a slight decrease in potency,
highlighting the importance of placement of the carboxylate
Chemistry. Compounds 119 and 520 were synthesized as
described previously, and compound 4 was purchased directly
from commercial vendors. The synthesis of compound 2 is
illustrated in Scheme 1. Weinreb's amide formation from acid
21 generated amide 22, which was treated with propylmagne-
siumbromide, giving the intermediate ketone 23. Palladium-
catalyzed aminocarbonalytion reaction from the corresponding
keto bromide 23 provided compound 2.
8177
dx.doi.org/10.1021/jm201129m|J. Med. Chem. 2011, 54, 8174−8187

Journal of Medicinal Chemistry
Article
Table 3. SAR of Fully Elaborated Analogs of Indole Fragment 1
a
b
IC50 values are represented as μM and averaged from a minimum of two determinations; LE calculated by RT ln(IC50)/no. of heavy atoms. IC50 in
MMP-13 molecular assay with the addition of 1.25% purified human serum albumin.
acid via a palladium-mediated cross-coupling reaction. Alter-
natively, compounds 7−11 were generated by converting 5-
bromo-1H-indole-2-carboxylic acid ethyl ester (25) to the
corresponding boronic ester 26 and then coupling directly with
various aryl bromides.
The synthesis of compound 12 is illustrated in Scheme 4.
Methylhydrazine was condensed with the but-2-ynedioic acid
dimethyl ester, and the intermediate was heated to evolve water
generating hydroxypyrazole 27. This alcohol was converted to
the pyrazole bromide 28, which was then coupled with
intermediate 26 to give the corresponding indole 12.
The synthesis of compounds 13−16 is illustrated in Scheme
Figure 4. Costructure of compound 15 showing the network of
5. Ester 28 was hydrolyzed to the corresponding acid, which
was then coupled with various amines to give substituted
benzylamides. The esters 32 and 34 were hydrolyzed under
basic conditions. The aryl bromides were coupled with 26 to
generating the corresponding elaborated indole analogues.
The synthesis of compound 17 is shown in Scheme 6. Under
palladium-catalyzed conditions, chloropyridine 36 was coupled
with indole boronic ester 6 and then treated with 2-(1H-7-
azabenzotriazol-1-yl)-1,1,3,3-tetramethyl uronium hexafluoro-
hydrogen bond interactions formed between the inhibitor and the
residues in the S1′* pocket of the protein.
The synthesis of compound 3 is illustrated in Scheme 2.
Bromide 21 was converted to the amide 24 and then treated
under aminocarbonylation conditions generating compound 3.
The synthesis of compounds 6−11 is illustrated in Scheme 3.
5-Aryl indole 6 was synthesized from 5-bromo-1H-indole-2-
carboxylic acid ethylester and the corresponding aryl boronic
8178
dx.doi.org/10.1021/jm201129m|J. Med. Chem. 2011, 54, 8174−8187

Journal of Medicinal Chemistry
Article
Figure 5. Extended profiling of 15 and rat pharmacokinetic profile.
The synthesis used to generate ether analogue 20 is
illustrated in Scheme 9. Commercially available indole 25 was
protected with an ethylene trimethylsiloxymethyl (SEM) group,
and the ester was reduced to the primary alcohol 39. The
alcohol was alkylated (40), and the bromide was converted to
the boronic acid (41). The boronic acid was coupled with
intermediate 33, and the SEM group was removed under
fluoride conditions to provide compound 20.
Table 4. Ester SAR on Elaborated Molecules
CONCLUSIONS
■
As compared with the starting indole fragment 1, compound 15
represents a 39000-fold increase in potency while maintaining
the original high ligand efficiency (>0.3). Using costructures of
intermediates and known literature SAR, the indole fragment
was rapidly optimized to a picomolar inhibitor that displayed a
promising pharmacokinetic profile. The described study
demonstrates how a novel chemical series can be obtained
via combined virtual screening and fragment-based drug
discovery strategies guided by high-resolution crystal structures
and subsequent optimization. Fragment 1 was submitted as a
synthetic intermediate for a BIPI legacy program and qualified
for our fragment library due to its intrinsic solubility and low
MW. The study herein highlights how every compound and
intermediate that a chemist generates should be submitted for
testing, regardless of pedigree or perceived limited value due to
molecular size.
a
IC50 values are represented as μM and averaged from a minimum of
two determinations; LE calculated by RT ln(IC50)/ no. of heavy
atoms. IC50 in MMP-13 molecular assay with the addition of 1.25%
purified human serum albumin. In vitro assessment of metabolic
stability of compounds in the presence of isolated human liver
microsome preparations. Data are reported as a % relative to human
hepatic blood flow.
b
c
phosphate methanaminium (HATU) and the methyleneami-
nopyridine to generate 17.
The synthesis of compound 18 is illustrated in Scheme 7.
Compound 15 was saponified to 18 under aqueous basic
conditions.
The synthesis used to generate pyridine analogue 19 is
shown in Scheme 8. Intermediate bromide 33 was coupled with
the 5-indole boronic acid under typical conditions to provide
the elaborated molecule 19.
EXPERIMENTAL SECTION
■
Chemistry General Remarks. Starting materials were obtained
from commercial suppliers and used without further purification unless
otherwise stated. 1H NMR spectra were recorded on a Bruker
UltraShield 400 MHz spectrometer operating at 400 MHz in solvents,
as noted. Proton coupling constants (J values) are rounded to the
nearest Hz. All coupling constants are reported in hertz (Hz), and
multiplicities are labeled s (singlet), bs, (broad singlet), d (doublet), t
(triplet), q (quartet), dd (doublet of doublets), dt (doublet oftriplets),
and m (multiplet). All NMR spectra were referenced to
8179
dx.doi.org/10.1021/jm201129m|J. Med. Chem. 2011, 54, 8174−8187

Journal of Medicinal Chemistry
Article
a
Scheme 1.
a
Reagents and conditions: (a) TBTU, N,N-methoxy methylamine, DIEA, DMF. (b) nPrMgCl, THF/DMF. (c) Mo(CO)6, DBU, Herrmann's
Palladacycle, DIEA, HCl dioxane water.
a
Scheme 2.
a
Reagents and conditions: (a) TBTU, ethylamine, THF/DMF. (b) Mo(CO)6, DBU, Hermann's Palladcycle, DIEA, HCl 1,4-dioxane, water.
a
Scheme 3.
a
Reagents and conditions: (a) PhB(OH)2, bistriphenylphosphine palladium(II) chloride, DMF, 2 M aqueous Na2CO3. (b) Bis(dipinacolyl)borane
Pd(dppf)Cl, 1,4-dioxane. (c) Ar-X, bis(triphenylphosphine)palladium(II) chloride, DMF, 2 M aqueous Na2CO3.
a
out under an atmosphere of Ar at room temperature, unless otherwise
Scheme 4.
noted. Mass spectroscopy data were obtained using the Micromass
Platform LCZ (flow injection). In addition to mass spectrometry and
NMR, purity (confirming ≥95% purity) was evaluated by the
following systems.
System 1. Analytical HPLC using a Varian Dynamax SD-200 pump
coupled to a Varian Dynamax UV-1 detector: The solvents were the
following: (A) water + 0.05% TFA and (B) acetonitrile + 0.05% TFA;
flow, 1.2 mL/min. Column Vydac RP-18, 5 m, 250 mm × 4.6 mm,
photodiode array detector at 220 nm; from 95 to 20% solvent (A) over
25 min confirming ≥95% purity.
System 2. HP 1110 Agilent LCMS using a Quaternary G1311A
pump coupled to a Micromass Platform LCZ detector: The solvents
a
Reagents and conditions: (a) Diethyl ether, 0 °C, filter, 100 °C neat.
(b) POBr3, 80 °C. (c) Compound 26, Pd(PPh3)2 Cl2, DMF, 2 M
aqueous Na2CO3.
were as follows: (A) water + 0.1% formic acid and (B) acetonitrile +
0.1% formic acid; flow, 1.5 mL/min. Photodiode array detector at 190
or 400 nm (a) Agilent Zorbax Eclipse XDB-C8 5 μm, 4.6 mm × 150
mm column, 4.6 mm × 30 mm, 3.5 μm, from 99 to 5% solvent (A)
over 10 min; or (b) Column Agilent Zorbax C18 SB 3.5 μm, 4.6 mm
× 30 mm cartridge, from 95 to 5% solvent (A) over 2.5 min
confirming ≥95% purity.
Chemstry. 2-Butyryl-1H-indole-5-carboxylic Acid Amide (2). A
solution of 5-bromo-1H-indole-2-carboxylic acid 21 (500 mg, 2.1
mmol), N,N-diisopropylethylamine (DIEA) (960 mL, 5.2 mmol), and
O-(benzotriazol-1-yl)-N,N,N′,N′-tetramethyluronium tetrafluoroborate
tetramethylsilane (TMS δH 0, δC 0). All solvents were HPLC grade or
higher. The reactions were followed by TLC on precoated Uniplate
silica gel plates purchased from Analtech. The developed plates were
visualized using 254 nm UV illumination or by PMA stain. Flash
column chromatography on silica gel was performed on Redi Sep
prepacked disposable silica gel columns using an Isco Combiflash,
Biotage SP1, or on traditional gravity columns. Reactions were carried
8180
dx.doi.org/10.1021/jm201129m|J. Med. Chem. 2011, 54, 8174−8187

Journal of Medicinal Chemistry
Article
a
Scheme 5.
a
Reagents and conditions: (a) LiOH, THF 1,4-dioxane, water. (b) HATU, DIEA, amine, DMF. (c) LiOH, water, THF 1,4-dioxane, methanol. (d)
Compound 26, Pd(PPh3)2, Cl2, DMF, 2 M aqueous Na2CO3.
a
Scheme 6.
a
Reagents and conditions: (a) Compound 26, Pd(4Me2NPh, PBu)2BF4, DMF, 2 M aqueous Na2CO3. (b) HATU, DIEA, amine, DMF.
a
Scheme 7.
a
Reagents and conditions: (a) LiOH, THF 1,4-dioxane, water.
a
Scheme 8.
a
Reagents and conditions: (a) Pd(PPh)4, 5-indoleboronic acid, THF, 2 M aqueous Na2CO3.
(TBTU) (740 mg, 2.2 mmol) was charged with methoxy methylamine
hydrochloride (220 mg, 2.2 mmol), and the reaction was diluted with
dimethylformamide (DMF) (10 mL) and stirred for 16 h at room
temperature. The reaction was quenched with water (25 mL), and the
product was extracted into EtOAc. The organic layer was dried over
sodium sulfate and reduced to a minimum volume by rotary
evaporation. The residue was purified on silica (heptanes/EtOAc) to
give 5-bromo-1H-indole-2-carboxylic acid methoxy-methyl-amide 22
δ 11.98 (s, 1H), 7.96 (s, 1H), 7.44−7.45 (m, 2H), 7.39 (d, J = 2.0 Hz,
1H), 3.00 (t, J = 7.2 Hz, 2H), 1.77 (m, 2H), 1.00 (t, J = 7.6 Hz, 3H).
A solution of 5-bromo-2-(1-methylene-butyl)-1H-indole 23 (103
mg, 0.4 mmol), molybdenum hexacarbonyl (52 mg, 0.2 mmol), and
Hermann's palladcycle (37 mg, 0.04 mmol) in THF (2 mL) was added
1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) (120 μL, 0.8 mmol), and
the reaction vessel was sealed and heated in a microwave reactor at 150
°C for 20 min. The reactions vessel was cooled, filtered, and diluted
with DMSO HCl/water and purified on reverse-phase LC to give the
title compound as a white solid. (0.01 g, 10%). LCMS (ES+) m/z
found, 231; retention time, 0.60 min. C13H14N2O2 requires 231.
1
(420 mg, 71%). H NMR (400 MHz, DMSO-d6): δ 11.78 (s, 1H),
7.88 (d, J = 1.6 Hz, 1H), 7.43 (d, J = 8.8 Hz, 1H), 7.33 (dd, J = 8.8 Hz,
2.0 Hz, 1H), 7.14 (d, J = 2.0 Hz, 1H), 3.80 (s, 3H), 3.34 (s, 3H).
A mixture of 5-bromo-1H-indole-2-carboxylic acid methoxy-methyl-
amide 22 (200 mg, 0.7 mmol) was dissolved in 10 mL of
tetrahydrofuran (THF) and cooled to 0 °C. Propylmagnesiumbromide
(2.8 mL 5.7 mmol) was added, and the reaction was warmed to room
temperature overnight. The reaction was quenched with 0.5 M HCl,
and the product was extracted into EtOAc. The organic layer was dried
over sodium sulfate and reduced to minimum volume by rotary
evaporation. The residue was added to a 20 g prepacked silica column
and eluted with EtOAc in heptanes to give 5-bromo-2-(1-methylene-
butyl)-1H-indole 23 (103 mg, 53%). 1H NMR (400 MHz, DMSO-d6):
1
HPLC retention time, 7.74 min. H NMR (400 MHz, DMSO-d6): δ
11.92 (s, 1H), 8.29 (s, 1H), 7.93 (bs, 1H), 7.82 (dd, J = 8.8 Hz, 1.6 Hz,
1H), 7.47 (d, J = 1.6 Hz, 1H), 7.44 (d, J = 8.8 Hz, 1H), 7.20 (bs, 1H),
2.97 (t, J = 7.2 Hz, 2H), 1.70 (m, 2H), 0.96 (t, J = 7.6 Hz, 3H).
1H-Indole-2,5-dicarboxylic Acid 5-Amide 2-Ethylamide (3). A
solution of 5-bromo-1H-indole-2-carboxylic acid 21 (500 mg, 2.1
mmol), DIEA (960 mL, 5.2 mmol), and TBTU (740 mg, 2.2 mmol)
was charged with ethyl amine (1.1 mL of 1 M in THF, 2.2 mmol), and
the reaction was diluted with DMF (10 mL) and stirred for 16 h at
room temperature. The reaction was quenched with water (25 mL),
8181
dx.doi.org/10.1021/jm201129m|J. Med. Chem. 2011, 54, 8174−8187

Journal of Medicinal Chemistry
Article
a
Scheme 9.
a
Reagents and conditions: (a) SEM-Cl, THF, NAH. (b) LiAlH4, THF. (c) Ag2O, EtI. (d) Bis(dipinacolyl)borane Pd(dppf)Cl, 1,4-dioxane. (e)
Compound 33, Pd(PPh3)2 Cl2, DMF, 2 M aqueous Na2CO3. (f) TBAF, ethylenediamine, THF.
and the product was extracted into EtOAc. The organic layer was dried
over sodium sulfate and reduced to minimum volume by rotary
evaporation. The residue was purified on silica (heptanes/EtOAc) to
give 5-bromo-1H-indole-2-carboxylic acid ethylamide 24 as a yellow
oil (500 mg, 89%). 1H NMR (400 MHz, DMSO-d6): δ 11.76 (s, 1H),
8.55 (t, J = 5.2 Hz, 1H), 7.83 (s, 1H), 7.38 (d, J = 8.8 Hz, 1H), 7.28
(dd, J = 8.8 Hz, 2.0 Hz, 1H), 7.08 (d, J = 2.4 Hz, 1H), 3.30 (q, J = 7.2
Hz, 2H), 7.14 (t, J = 7.2 Hz, 3H).
To a solution of 5-bromo-1H-indole-2-carboxylic acid ethylamide
24 (213 mg, 0.8 mmol), molybdenum octacarbonyl (105 mg, 0.4
mmol), and Hermann's palladcycle (37 mg, 0.04 mmol) in THF (2
mL) was added DBU (120 μL, 0.8 mmol), and the reaction vessel was
sealed and heated in a microwave reactor at 150 °C for 20 min. The
reaction vessel was cooled, filtered, and diluted with DMSO and HCl/
water and purified on reverse-phase HPLC to give the title compound
as a white solid (0.025 g, 14%). LCMS (ES+) m/z found, 232;
retention time, 0.39 min. C12H13N3O2 requires 232. HPLC retention
time, 5.44 min. 1H NMR (400 MHz, DMSO-d6): δ 11.78 (s, 1H), 8.54
(t, J= 5.6 Hz, 1H), 8.22 (s, 1H), 7.87 (brd, 1H), 7.73 (dd, J = 8.4 Hz,
1.6 Hz, 1H), 7.42 (d, J = 8.4 Hz, 1H), 7.20, (d, J = 1.6 Hz, 1H), 7.14
(brd, 1H), 3.33 (m, 2H), 1.16 (t, J = 3.6 Hz).
was sealed and heated in a microwave reactor at 100 °C for 20 min.
The reaction vessel was cooled and filtered though Celite, and the
Celite was washed well with EtOAc. The combined filtrates were
washed with water, dried (Na2SO4), filtered, and evaporated in vacuo
to give the crude product that was purified on reverse-phase HPLC to
give the title compound 7 as a white solid (0.020 g, 24%). LCMS (ES
+) m/z found, 267; retention time, 0.63 min. C16H14N2O2 requires
1
267. HPLC retention time, 7.42 min. H NMR (400 MHz, DMSO-
d6): δ 12.14 (s, 1H), 8.71 (bs, 1H), 8.41 (s, 1H), 8.10 (bs, 2H), 8.00
(dd, J = 8.4 Hz, 1.6 Hz, 1H), 7.58 (d, J = 8.8 Hz, 1H), 7.51 (bs, 1H),
7.28 (d, J = 1.6 Hz, 1H), 4.37 (q, J = 3.2 Hz, 2H), 1.36 (t, J = 3.2 Hz,
3H).
5-Pyridin-3-yl-1H-indole-2-carboxylic Acid Ethyl Ester (8). The
title compound was prepared according to the general procedure from
indole 7 using 5-(4,4,5,5-tetramethyl-[1,3,2]dioxaborolan-2-yl)-1H-
indole-2-carboxylic acid ethyl ester 26 (100 mg, 0.31 mmol), 3-
bromopyridine (0.03 mL, 0.31 mmol), and bistriphenylphosphine
palladium(II) chloride (25 mg, 0.03 mmol) to give the crude product
that was purified on reverse-phase HPLC to give the title compound as
a white solid (20 mg, 24%). LCMS (ES+) m/z found, 267; retention
time, 0.58 min. C16H14N2O2 requires 267. HPLC retention time, 7.56
min. 1H NMR (400 MHz, DMSO-d6): δ 12.12 (s, 1H), 9.11 (bs, 1H),
8.71 (bs, 1H), 8.47 (d, J = 7.2 Hz, 1H), 8.10 (s, 1H), 7.60−7.80 (m,
3H), 7.24 (s, 1H), 4.37 (q, J = 3.2 Hz, 2H), 1.36 (t, J = 3.2 Hz, 3H).
5-Pyridin-4-yl-1H-indole-2-carboxylic Acid Ethyl Ester (9). The
title compound was prepared according to the general procedure from
indole 7 using 5-(4,4,5,5-tetramethyl-[1,3,2]dioxaborolan-2-yl)-1H-
indole-2-carboxylic acid ethyl ester 26 (100 mg, 0.31 mmol), 4-
bromopyridine hydrochloride (61 mg, 0.31 mmol), and bistriphenyl-
phosphine palladium(II) chloride (25 mg, 0.03 mmol) to give the
crude product that was purified on reverse-phase HPLC to give the
title compound as a white solid (8 mg, 10%). LCMS (ES+) m/z
found, 267; retention time, 0.49 min. C16H14N2O2 requires 267.
5-Phenyl-1H-indole-2-carboxylic Acid Ethyl Ester (6). A 5 mL
microwave vial was charged with phenylboronic acid (45 mg, 0.37
mmol), 5-bromo-1H-indole-2-carboxylic acid ethyl ester 25 (100 mg,
0.37 mmol), and bis(triphenylphosphine)palladium(II) chloride (25
mg, 0.03 mmol). The solids were suspended in DMF (2 mL), and 2 M
sodium carbonate solution (2 mL) was added to the reaction vessel.
The vial was sealed and heated in a microwave reactor at 100 °C for 20
min. The reaction vessel was cooled and filtered though Celite, and the
Celite was washed well with EtOAc. The combined extracts were
washed with water, dried (Na2SO4), filtered, and evaporated in vacuo
to give the crude product that was purified on reverse-phase HPLC to
give the title compound 6 as a white solid (0.026 g, 26%). LCMS (ES
+) m/z found, 266; retention time, 1.11 min. C17H15NO2 requires 266.
1
HPLC retention time, 7.44 min. H NMR (400 MHz, DMSO-d6): δ
12.25 (s, 1H), 8.81, (d, J = 6.4 Hz, 2H), 8.37 (s, 1H), 8.20 (d, J = 6.4
Hz, 2H), 7.82 (dd, J = 8.4 Hz, 1.6 Hz, 1H), 7.63 (d, J = 8.4 Hz, 1H),
7.29 (d, J = 1.6 Hz, 1H), 4.37 (q, J = 3.2 Hz, 2H), 1.36 (t, J = 3.2 Hz,
3H).
5-(3-Methyl-pyridin-4-yl)-1H-indole-2-carboxylic Acid Ethyl Ester
(10). The title compound was prepared according to the general
procedure from indole 7 using 5-(4,4,5,5-tetramethyl-[1,3,2]-
dioxaborolan-2-yl)-1H-indole-2-carboxylic acid ethyl ester 26 (434
mg, 1.4 mmol), 4-chloro-3-methyl-pyridine hydrochloride (188 mg,
1.4 mmol), bis(di-tert-butyl(4-dimethylaminophenyl)phosphine)-
dichloropalladium(II) (40 mg, 0.06 mmol), potassium carbonate
(965 mg, 6.9 mmol), toluene (4 mL), and water (0.5 mL) to give the
crude product that was purified on reverse-phase HPLC to give the
1
HPLC retention time, 13.25 min. C17H15NO2 requires 266. H NMR
(400 MHz, DMSO-d6): δ 11.95 (s, 1H), 7.93 (s, 1H), 7.67 (d, J= 7.2
Hz, 2H), 7.58 (dd, J = 8.8 Hz, 1.6 Hz, 1H), 7.53 (d, J = 8.8 Hz, 1H),
7.45 (t (dd), J = 7.2 Hz, 2H), 7.32 (t, J = 7.2 Hz, 1H), 7.21 (d, J = 2.0
Hz, 1H), 4.35 (q, J = 7.2 Hz, 2H), 1.35 (t, J = 7.2 Hz, 3H).
5-Pyridin-2-yl-1H-indole-2-carboxylic Acid Ethyl Ester (7). A 5
mL microwave vial was charged with 5-(4,4,5,5-tetramethyl-[1,3,2]-
dioxaborolan-2-yl)-1H-indole-2-carboxylic acid ethyl ester 25 (100 mg,
0.31 mmol), 2-bromopyridine (0.03 mL, 0.31 mmol), and
bistriphenylphosphine palladium(II) chloride (25 mg, 0.03 mmol).
The solids were suspended in DMF (2 mL), and 2 M sodium
carbonate solution (2 mL) was added to the reaction vessel. The vial
8182
dx.doi.org/10.1021/jm201129m|J. Med. Chem. 2011, 54, 8174−8187

Journal of Medicinal Chemistry
Article
title compound as a white solid (50 mg, 15%). LCMS (ES+) m/z
layers were dried, filtered, and evaporated in vacuo to give 5-bromo-1-
found, 281; retention time, 0.54 min. C17H16N2O2 requires 281.
HPLC retention time, 7.96 min. C17H16N2O2 requires 281. H NMR
methyl-1H-pyrazole-3-carboxylic acid 29 (7.1 g, 34 mmol 75%) as a
white solid that was used without further purification. H NMR (400
1
1
(400 MHz, DMSO-d6): δ 12.19 (s, 1H), 8.78 (s, 1H), 8.71 (d, J = 5.6
Hz, 1H), 7.86 (s, 1H), 7.78 (d, J = 5.6 Hz, 1H), 7.61 (d, J = 8.8 Hz,
1H), 7.42 (dd, J = 8.8 Hz, 1.6 Hz, 1H), 7.26 (d, J = 1.6 Hz, 1H), 4.37
(q, J = 3.2 Hz, 2H), 2.43 (s, 3H), 1.36 (t, J = 3.2 Hz, 3H).
MHz, DMSO-d6): δ 6.87 (s, 1H), 3.89 (s, 3H).
A flask was charged with 5-bromo-1-methyl-1H-pyrazole-3-carbox-
ylic acid 29 (500 mg, 2.4 mmol) and HATU (1.1 g, 2.7 mmol). The
solids were dissolved in DMF (10 mL), and the reaction was stirred
for 1 h at room temperature after which time pyridin-4-yl-methylamine
(0.27 mL, 2.7 mmol) and diisopropylethylamine (1.1 mL, 6.1 mmol)
were added. The reaction was stirred at room temperature for 24 h and
then poured into water, and the product was extracted with EtOAc.
The combined organic layers were washed with water and then brine,
dried, filtered, and evaporated in vacuo to give 5-bromo-1-methyl-1H-
pyrazole-3-carboxylic acid (pyridin-4-ylmethyl)-amide 30 (456 mg, 1.5
mmol, 63%) as a solid that was used without further purification.
1H-Pyrazole-3-carboxylic acid (pyridin-4-ylmethyl)-amide 30 (100
mg, 0.34 mmol), 5-(4,4,5,5-tetramethyl-[1,3,2]dioxaborolan-2-yl)-1H-
indole-2-carboxylic acid ethyl ester 26 (113 mg, 0.36 mmol), and
bis(di-tert-butyl(4-dimethylaminophenyl)phosphine)-
dichloropalladium(II) (40 mg, 0.06 mmol) were placed into a
microwave vial and suspended in DMF (2 mL), and a 2 M aqueous
solution of sodium carbonate (2 mL) was added. The reaction vessel
was sealed and heated in a microwave reactor for 20 min at 100 °C.
The solution was diluted with EtOAc and filtered though a pad of
Celite. The pad was washed with DCM/methanol (1:1), and the
combined extracts were washed with water. The organic layer was
dried, filtered, and evaporated in vacuo to give an oil that was purified
by reverse-phase HPLC to give the title compound 13 (15 mg, 11%).
LCMS (ES+) m/z found, 404; retention time, 0.56 min. C17H17N3O4
requires 404. HPLC retention time, 8.44 min. C17H17N3O4 requires
404. 1H NMR (400 MHz, DMSO-d6): δ 12.12 (s, 1H), 9.01 (t, J = 6.0
Hz, 1H), 8.70 (d, J = 5.6 Hz, 2H), 7.87 (s, 1H), 7.67 (d, J = 4.8 Hz,
2H), 7.57 (d, J = 8.4 Hz, 1H), 7.44 (d, J = 8.4 Hz, 1H), 7.23 (s, 1H),
6.77 (s, 1H), 4.59 (d, J = 7.0 Hz, 2H), 4.36 (q, J = 3.2 Hz, 2H), 3.93
(s, 3H), 1.35, (t, J = 3.2 Hz, 3H).
5-Pyridin-4-yl-1H-indole-2-carboxylic Acid Ethyl Ester (11). The
title compound was prepared according to the general procedure from
indole 7 using 5-(4,4,5,5-tetramethyl-[1,3,2]dioxaborolan-2-yl)-1H-
indole-2-carboxylic acid ethyl ester 26 (100 mg, 0.31 mmol), 5-
bromo-1-methyl-1H-imidazole (51 mg, 0.31 mmol), and bistriphenyl-
phosphine palladium(II) chloride (25 mg, 0.03 mmol) to give the
crude product that was purified on reverse-phase HPLC to give the
title compound as a white solid (5 mg, 6%). LCMS (ES+) m/z found,
270; retention time, 0.46 min. C15H15N3O2 requires 270. HPLC
1
retention time, 7.39 min. C15H15N3O2 requires 270. H NMR (400
MHz, DMSO-d6): δ 12.20 (s, 1H), 9.11 (s, 1H), 7.91 (s, 1H), 7.80 (s,
1H), 7.61 (d, J = 8.8 Hz, 1H), 7.45 (d, J = 8.8 Hz, 1H), 7.25 (s, 1H),
7.36 (q, J = 3.2 Hz, 2H), 3.82 (s, 3H), 1.35 (t, J = 3.2 Hz, 3H).
5-(2-Ethoxycarbonyl-1H-inden-5-yl)-1-methyl-1H-pyrazole-3-car-
boxylic Acid Methyl Ester (12). But-2-ynedioic acid dimethyl ester
(21.6 g, 151 mmol) was added to diethylether (200 mL), and the
resulting solution was cooled to −5 °C in a methanol/ice bath.
Methylhydrazine (8.1 mL, 151 mmol) in diethyl ether (10 mL) was
added dropwise to the solution at a rate such that the internal
temperature did not rise above 0 °C. The mixture was stirred for 1 h at
−5 °C, and the resulting yellow precipitate was filtered off, washed
with ether, and dried in a 40 °C oven. A flask was then charged with
the solid, and the flask was immersed in an oil bath heated to 100 °C
for 10 min. The solid turned bright orange, and vapor was emitted
from the flask. The resulting solid was cooled to room temperature
and recrystallized from methanol to give 5-hydroxy-1-methyl-1H-
pyrazole-3-carboxylic acid methyl ester 27 as a white solid (15 g, 96
mmol).
A sealed tube reaction flask was charged with 5-hydroxy-1-methyl-
1H-pyrazole-3-carboxylic acid methyl ester (10 g, 64 mmol),
phosphorus oxybromide (91 g, 320 mmol), and acetonitrile (200
mL). The flask was sealed and heated to 80 °C for 15 h. Upon
completion, the reaction vessel was cooled, and the reaction solution
was poured into a precooled solution of saturated sodium bicarbonate.
Additional solid sodium bicarbonate was added to the stirred mixture
until the solution turned basic. The product was extracted into EtOAc,
and the combined organic layers were dried (Na2SO4), filtered, and
evaporated in vacuo to give 5-bromo-1-methyl-1H-pyrazole-3-carbox-
5-{2-Methyl-5-[(6-oxo-1,6-dihydro-pyridin-3-ylmethyl)-carbamo-
yl]-2H-pyrazol-3-yl}-1H-indole-2-carboxylic Acid Ethyl Ester (14).
The title compound was prepared according to the procedure
described for 13 from 5-bromo-1-methyl-1H-pyrazole-3-carboxylic
acid 29 (200 mg, 0.97 mmol), HATU (371 mg, 0.97 mmol), 5-
aminoethyl-1H-pyridin-2-one (121 mg, 0.97 mmol), DMF (1 mL),
and diisopropyethylamine (0.18 mL, 0.97 mmol) to give the amide 14
(200 mg, 66%). LCMS (ES+) m/z found, 420; retention time, 0.66
1
min. C22H21N5O4 requires 420. HPLC retention time, 9.07 min. H
NMR (400 MHz, DMSO-d6): δ 12.06 (brd, 1H), 11.42 (brd, 1H),
8.56 (t, J = 6.4 Hz, 1H), 7.84 (s, 1H), 7.56 (d, J = 8.8 Hz, 1H), 7.44
(dd, J = 9.2 Hz, 2.4 Hz, 1H), 7.41 (dd, J = 8.4 Hz, 1.6 Hz, 1H), 7.24
(d, J = 2.4 Hz, 1H), 7.21 (s, 1H), 6.71 (s, 1H), 6.29 (d, J = 9.2 Hz,
1H), 4.35 (q, J = 3.2 Hz, 2H), 4.14 (d, J = 7.0 Hz, 2H), 3.88 (s, 1H),
1.34 (s, J = 3.2 Hz, 3H).
5-[5-(4-Carboxy-benzylcarbamoyl)-2-methyl-2H-pyrazol-3-yl]-
1H-indole-2-carboxylic Acid Ethyl Ester (15). A flask is charged with
5-bromo-1-methyl-1H-pyrazole-3-carboxylic acid 29 (7.25 g 35
mmol), TBTU (14.7 g, 38 mmol), and DMF (50 mL), and the
reaction was stirred for 30 min. 4-Aminomethyl-benzoic acid methyl
ester hydrochloride (7.84 g, 38 mmol) was added followed by
Hunnings base (16.2 mL, 88 mmol), and the reaction was sealed and
stirred overnight for24 h. The reaction was poured into water and
extracted into EtOAc. The organic layer was washed with water,
saturated sodium bicarbonate, and brine, dried, filtered, and
evaporated in vacuo to give 4-{[(5-bromo-1-methyl-1H-pyrazole-3-
carbonyl)-amino]-methyl}-benzoic acid methyl ester 32, which was
used directly without further purification.
1
ylic acid methyl ester 28 as a tan solid (6.8 g, 31 mmol, 48%). H
NMR (400 MHz, DMSO-d6): δ 6.96 (s, 1H), 3.90 (s, 3H), 3.80 (s,
3H).
The title compound 12 was prepared according to the general
procedure from indole 7 using 5-(4,4,5,5-tetramethyl-[1,3,2]-
dioxaborolan-2-yl)-1H-indole-2-carboxylic acid ethyl ester 26 (113
mg, 0.36 mmol), 5-bromo-1-methyl-1H-pyrazole-3-carboxylic acid
methyl ester 28 (80 mg, 0.37 mmol), and bistriphenylphosphine
palladium(II) chloride (41 mg, 0.06 mmol) to give the crude product
that was purified on reverse-phase HPLC to give the title compound
12 as a white solid (14 mg, 13%). LCMS (ES+) m/z found, 328;
retention time, 0.85 min. C17H17N3O4 requires 328. HPLC retention
1
time, 10.53 min. H NMR (400 MHz, DMSO-d6): δ 12.08 (s, 1H),
7.86 (s, 1H), 7.56 (d, J = 8.8 Hz, 1H), 7.42 (dd, J = 8.4 Hz, 1.6 Hz,
1H), 7.21 (d, J = 1.6 Hz), 6.84 (s, 1H), 4.35 (q, J = 3.2 Hz, 2H), 3.91
(s, 3H), 3.80 (s, 3H), 1.34 (t, J = 3.1 Hz, 3H).
5-{2-Methyl-5-[(pyridin-4-ylmethyl)-carbamoyl]-2H-pyrazol-3-yl}-
1H-indole-2-carboxylic Acid Ethyl Ester (13). A flask was charged
with 5-bromo-1-methyl-1H-pyrazole-3-carboxylic acid methyl ester 28
(10 g, 46 mmol) and dioxane (100 mL), and a 4 N solution of lithium
hydroxide (57 mL) was added. The reaction was stirred at room
temperature for 1 h, then additional water was added, and the organic
layer evaporated in vacuo. The water layer was washed twice with
dichloromethane and then acidified to pH 2.0 with concentrated HCl.
The product was extracted into EtOAc, and the combined organic
The above ester 32 (2.0 g, 5.6 mmol) was dissolved in 1,4-dioxane
(30 mL), and 4 N warm lithium hydroxide (5.67 mL, 22 mmol) was
added to the solution. The reaction was stirred at room temperature
for 1 h, then water was added, and then, the organic layer was
evaporated in vacuo. The water layer was washed with dichloro-
methane and acidified to pH 2 with concentrated HCl. The reaction
was evaporated to 50% volume, and the resulting solid was collected,
8183
dx.doi.org/10.1021/jm201129m|J. Med. Chem. 2011, 54, 8174−8187

Journal of Medicinal Chemistry
Article
washed with 1 N HCl, and dried in vacuo to give 4-{[(5-bromo-1-
methyl-1H-pyrazole-3-carbonyl)-amino]-methyl}-benzoic acid 33 (1.9
g, 93%). 1H NMR (400 MHz, DMSO-d6): δ 12.78 (bs, 1H), 8.83 (t, J
= 6.4 Hz, 1H), 7.87 (d, J = 8.4 Hz, 2H), 7.37 (d, J = 8.0 Hz, 2H), 6.82
(s, 1H), 4.47 (d, J = 6.4 Hz, 2H), 3.88 (s, 3H).
was charged with 5-(4,4,5,5-tetramethyl-[1,3,2]dioxaborolan-2-yl)-1H-
indole-2-carboxylic acid ethyl ester 26 (510 mg, 1.6 mmol), 4-Chloro-
5-methyl-pyridine-2-carboxylic acid (300 mg, 1.61 mmol), bis(di-t-
butylphosphino (4-dimethylaminophenyl)) palladium(II)dichloride
(230 mg, 0.32 mmol), and the solids suspended in DMF (6 mL)
and 2N sodium carbonate (2 mL). The reaction vessel was sealed and
heated to 100 °C for 20 min in a microwave reactor. The reaction was
cooled to room temperature, filtered through Celite, and washed with
EtOAc and ammonium hydroxide. The aqueous layer was separated,
and acidified with concentrated HCl. The solvent was removed in
vacuo and the crude material 37 used without further purification.
5-(2-Carboxy-5-methyl-pyridin-4-yl)-1H-indole-2-carboxylic acid
ethyl ester 37 (15 mg, 0.05 mmol) was dissolved in DMF (1 mL),
and HATU (26 mg, 0.07 mmol) and DIEA (0.03 mL 0.07 mmol) were
added. The reaction was stirred for 30 min, then C-pyridin-4-yl-
methylamine (8 μL, 0.07 mmol) was added to the reaction, and the
resulting mixture was stirred at room temperature overnight. The
solvents were evaporated in vacuo, and the crude product was purified
directly by reverse-phase HPLC to give the title compound 17 as a
white solid (10 mg, 52%) LCMS (ES+) m/z found, 415; retention
time, 0.67 min. C24H22N4O3 requires 415. HPLC retention time, 9.20
min. 1H NMR (400 MHz, DMSO-d6): δ 12.08 (s, 1H), 9.50 (t, J = 6.4
Hz, 1H), 8.61 (s, 1H), 8.50 (d, J = 4.8 Hz, 2H), 7.90 (s, 1H), 7.77 (s,
1H), 7.57 (d, J = 8.4 Hz, 1H), 7.36 (d, J = 8.4 Hz, 1H), 7.31 (d, J = 4.8
Hz, 2H), 7.23 (s, 1H), 4.52 (s, 2H), 4.36 (q, J = 3.2 Hz, 2H), 2.38 (s,
3H), 1.35 (t, J = 3.2 Hz, 3H).
A reaction flask was charged with 4-{[(5-bromo-1-methyl-1H-
pyrazole-3-carbonyl)-amino]-methyl}-benzoic acid 33 (7.0 g, 20
mmol), 5-(4,4,5,5-tetramethyl-[1,3,2]dioxaborolan-2-yl)-1H-indole-2-
carboxylic acid ethyl ester 26 (6.5 g, 20 mmol), tetrakistriphenyphos-
phine palladium(0) (2.9 g, 2.5 mmol) THF (600 mL), and 2 M
aqueous solution of sodium bicarbaonate (60 mL, 120 mmol). The
mixture was heated to reflux for 24 h and cooled to room temperature,
and the solvent was evaporated in vacuo. The crude material was
partitioned between water and EtOAc, and the aqueous layer was
isolated and acidified to pH 3 with concentrated HCl. The resulting
precipitate was collected by filtration and purified on silica (methanol/
dichloromethane) to give the title compound 15 as a white solid.
LCMS (ES+) m/z found, 447; retention time, 0.80 min. C24H22N4O5
1
requires 447. HPLC retention time, 10.12 min. H NMR (400 MHz,
DMSO-d6): δ 12.12 (s, 1H), 8.80 (t, J = 6.4 Hz, 1H), 7.89 (s, 1H),
7.87 (s, 2H), 7.57 (d, J = 8.4 Hz, 1H), 7.43 (dd, J = 8.4 Hz, 1.6 Hz,
1H), 7.36 (d, J = 8.4 Hz, 2H), 7.22 (d, J = 1.6 Hz, 1H), 6.75 (s, 1H),
4.48 (d, J = 6.0 Hz, 2H), 4.35 (q, J = 3.2 Hz, 2H), 3.91 (s, 3H), 1.35 (t,
J = 3.2 Hz, 3H).
5-[5-(4-Carboxymethyl-benzylcarbamoyl)-2-methyl-2H-pyrazol-
3-yl]-1H-indole-2-carboxylic Acid Ethyl Ester (16). A flask was
charged with 5-bromo-1-methyl-1H-pyrazole-3-carboxylic acid 29
(250 mg, 1.2 mmol), HATU (463 mg, 1.2 mmol), and DMF (2
mL), and the reaction was stirred for 30 min. (4-Aminomethyl-
phenyl)-acetic acid methyl ester hydrochloride (260 mg, 1.2 mmol)
was added followed by DIEA (0.45 mL, 2.4 mmol), and the reaction
was sealed and stirred overnight for 24 h. The solvents were
evaporated in vacuo, and the crude material was partitioned between
EtOAc and water. The organic layer was washed thrice with water,
dried over magnesium sulfate, filtered, and evaporated in vacuo to give
an oil that was purified on silica (EtOAc/heptanes) to give (4-{[(5-
bromo-1-methyl-1H-pyrazole-3-carbonyl)-amino]-methyl}-phenyl)-
acetic acid methyl ester 34 (330 mg, 74%).
The above ester 34 (150 mg, 0.4 mmol) was dissolved in THF (2.5
mL) and water (1.0 mL), and lithium hydroxide monohydrate (50 mg,
2.0 mmol) was added to the solution. The reaction was stirred at room
temperature for 24 h and then quenched with concentrated HCl to a
pH of 2.0. The solvent was evaporated in vacuo, and the solids were
diluted with water and EtOAc. The product was extracted into EtOAc,
and the organic layer was dried (MgSO4), filtered, and evaporated in
vacuo to give a white solid 25 that was used with out further
purification (140 mg, 97%). 1H NMR (400 MHz, DMSO-d6): δ 12.23
(s, 1H), 8.69 (t, J = 6.4 Hz, 1H), 7.19 (q, J = 8.0 Hz, 4H), 6.80 (s,
1H), 4.36 (d, J = 6.4 Hz, 2H), 3.87 (s, 3H), 3.50 (s, 2H).
A reaction flask was charged with (4-{[(5-bromo-1-methyl-1H-
pyrazole-3-carbonyl)-amino]-methyl}-phenyl)-acetic acid 35 (100 mg,
0.28 mmol), 5-(4,4,5,5-tetramethyl-[1,3,2]dioxaborolan-2-yl)-1H-in-
dole-2-carboxylic acid ethyl ester 26 (90 mg, 0.28 mmol),
tetrakistrkistriphenyphosine palladium(0) (10 mg, 0.03 mmol), THF
(5 mL), and 2 M aqueous sodium bicarbonate (0.75 mL). The mixture
was heated to reflux for 24 h and cooled to room temperature, and the
solvent was evaporated in vacuo. The crude material was partitioned
between water and EtOAc, and the aqueous layer was isolated and
acidified to pH 3 with concentrated HCl. The resulting precipitate was
collected by filtration and purified by reverse-phase HPLC to give a
white solid (14 mg, 11%). LCMS (ES+) m/z found, 461; retention
time, 0.80 min. C25H24N4O5 requires 461. HPLC retention time, 10.15
min. 1H NMR (400 MHz, DMSO-d6): δ 12.29 (s, 1H), 12,11 (s, 1H),
8.69 (t, J = 6.0 Hz, 1H), 7.86 (s, 1H), 7.56 (d, J = 8.8 Hz, 1H), 7.43
(dd, J = 8.8 Hz, 1.6 Hz, 1H), 7.24 (d, J = 8 Hz, 2H), 7.22 (d, J = 1.6
Hz, 1H), 7.19 (d, J = 8.4 Hz, 2H),6.73 (s, 1H), 4.40 (d, J = 6.0 Hz,
2H), 4.35 (q, J = 3.2 Hz, 2H), 3.90 (s, 3H), 3.52 (s, 2H), 1.35 (t, J =
3.2 Hz, 3H).
5-[5-(4-Carboxy-benzylcarbamoyl)-2-methyl-2H-pyrazol-3-yl]-
1H-indole-2-carboxylic (18). A flask was charged with 5-[5-(4-
carboxy-benzylcarbamoyl)-2-methyl-2H-pyrazol-3-yl]-1H-indole-2-car-
boxylic acid ethyl ester 15 (35 mg, 0.08 mmol) and taken up in 1,4-
dioxane (1 mL), and lithium hydroxide in water (4 N, 0.1 mL) was
added. The reaction was stirred for 3 h and then acidified to pH 2 with
concentrated HCl. The reaction flask was cooled to 0 °C, and the
resulting precipitate was collected, washed with ice cooled water, and
dried in vacuo to give the title compound (15 mg, 45%). LCMS (ES+)
m/z found, 419; retention time, 0.61 min. C21H18N4O5 requires 419.
1
HPLC retention time, 8.29 min. H NMR (400 MHz, DMSO-d6): δ
13.06 (bs, 1H), 12.85 (b, 1H), 11.97 (s, 1H), 8.81 (t, J = 6.4 Hz, 1H),
7.89 (d, J = 8.0 Hz, 2H), 7.84 (s, 1H), 7.54 (d, J = 8.4 Hz, 1H), 7.42−
7.39 (m, 3H), 7.15 (d, J = 1.6 Hz, 1H), 6.73 (s, 1H), 4.49 (d, J = 6.0
Hz, 2H), 3.91 (s, 3H).
4-({[5-(1H-indol-5-yl)-1-methyl-1H-pyrazole-3-carbonyl]-amino}-
methyl)-bezoic Acid (19). A reaction flask was charged with 4-{[(5-
bromo-1-methyl-1H-pyrazole-3-carbonyl)-amino]-methyl}-benzoic
acid 33 (75 m g, 0.22 mmol, from example 14), 5-indoleboronic acid
(40 mg, 0.24 mmol), and tetrakistriphenyphosphine palladium(0) (9
mg, 0.003 mmol), and the solids were taken up in THF (4 mL). The
reaction was heated to reflux for 24 h, cooled to room temperature,
and diluted with EtOAc and water. The aqueous layer was separated
and acidified to pH 2 with concentrated HCl, and the resulting solid
was collected and purified by reverse-phase HPLC to give the title
compound 19 (20 mg, 24%). LCMS (ES+) m/z found, 375; retention
time, 0.70 min. C21H18N4O3 requires 375. HPLC retention time, 9.20
1
min. C21H18N4O3 requires 375. H NMR (400 MHz, DMSO-d6): δ
12.86 (s, 1H), 11.31 (s, 1H), 8.82 (t, J = 6.4 Hz, 1H), 7.89 (d, J = 8.4
Hz, 2H), 7.72 (s, 1H), 7.51 (d, J = 8.4 Hz, 1H), 7.44 (t, J = 2.8 Hz,
1H), 7.41 (d, J = 8.0 Hz, 2H), 7.23 (dd, J = 8.4 Hz, 1.6 Hz, 1H), 6.70
(s, 1H), 6.51 (m, 1H), 4.49, (d, J = 6.4 Hz, 2H), 3.91 (s, 3H).
4-({[5-(20-Ethoxymethyl-1H-indol-5-yl)-1-methyl-1H-pyrazole-3-
carbonyl]-amino}-methyl)-bezoic Acid (20). A flask was charged
with 5-bromo-1H-indole-2-carboxylic acid ethyl ester 25 (4.5 g, 16
mmol) and THF (50 mL), and the reaction cooled to 0 °C in an ice
bath. Sodium hydride (60% in mineral oil, 740 mg, 18 mmol) was
added portion wise to the reaction over 10 min, and the resulting
slurry was stirred for 15 min, and then, 2-(trimethylsilyl)ethoxymethyl
chloride (SEM-Cl) (3.2 mL, 18 mmol) was added to the reaction. The
reaction was stirred at 0 °C for 4 h and then quenched by the addition
of water, and the product was extracted into diethyl ether. The organic
layers were combined, dried over sodium sulfate, filtered, and
evaporated in vacuo to give an oil that was purified on silica
5-{5-Methyl-2-[(pyridin-4-ylmethyl)-carbamoyl]-pyridin-4-yl}-1H-
indole-2-carboxylic Acid Ethyl Ester (17). A microwave reaction flask
8184
dx.doi.org/10.1021/jm201129m|J. Med. Chem. 2011, 54, 8174−8187

Journal of Medicinal Chemistry
Article
(hexanes/EtOAc) to give 5-bromo-1-(2-trimethylsilanyl-ethoxymeth-
yl)-1H-indole-2-carboxylic acid ethyl ester 38 as a clear oil. H NMR
time, 0.78 min. C24H24N4O4 requires 433. HPLC retention time, 9.82
min. H NMR (400 MHz, DMSO-d6): δ 12.85 (brd, 1H), 11.37 (s,
1
1
(400 MHz, DMSO-d6): δ 7.94 (d, J = 2.0 Hz, 1H), 7.68 (d, J = 8.8 Hz,
1H), 7.49 (dd, J = 8.8 Hz, 2.0 Hz, 1H), 7.32 (s, 1H), 5.94 (s, 2H), 4.32
(q, J = 7.2 Hz, 2H), 3.41 (t, J = 8.0 Hz, 2H), 1.32 (t, J = 7.2 Hz, 3H),
0.75 (t, J = 8.0 Hz, 2H), 0.14 (s, 9H).
The above ester 38 (5.1 g, 12 mmol) was placed into a flask and
dissolved in THF (40 mL) and ether (40 mL). The flask was cooled to
0 °C, and LiAlH4 (1.0 g, 25 mmol) was added portion wise to the
reaction (Caution: hydrogen gas evolution) under argon. The resulting
slurry was stirred for 1.5 h at 0 °C and then quenched by the addition
of 1 mL of water, 1 mL of 15% aqueous sodium hydroxide, and 3 mL
of water to give a suspension that is filtered, and the filtrate was
evaporated in vacuo to give [5-bromo-1-(2-trimethylsilanyl-ethox-
ymethyl)-1H-indol-2-yl]-methanol 39 as an oil that was used without
further purification.
A flask was charged with the above indole 39 (500 mg, 1.4 mmol),
silver(II) oxide (1 g, 4.2 mmol), ethyl iodide (1.6 mL, 14 mmol), and
acetonitrile (5 mL). The reaction was stirred at room temperature for
4 h and then heated to 50 °C for 16 h. After it was cooled, the reaction
was passed though Celite and evaporated in vacuo to give an oil that
was purified on silica (hexanes/EtOAc) to give 5-bromo-2-
ethoxymethyl-1-(2-trimethylsilanyl-ethoxymethyl)-1H-indole 40 as a
clear oil (238 mg, 50%). 1H NMR (400 MHz, DMSO-d6): δ 7.72 (d, J
= 2.0 Hz, 1H), 7.53 (d, J = 8.4 Hz, 1H), 7.28 (dd, J = 8.8 Hz, 2.0 Hz,
1H), 6.50 (s, 1H), 5.55 (s, 2H), 4.64 (s, 2H), 3.49 (q, J = 7.2 Hz, 2H),
3.46 (t, J = 8.0 Hz, 2H), 1.13 (t, J = 7.2 Hz, 3H), 0.79 (t, J = 8.0 Hz,
2H), 0.10 (s, 9H).
1H), 8.81 (t, J = 6.4 Hz, 1H), 7.89 (d, J = 8.4 Hz, 2H), 7.65 (s, 1H),
7.44 (d, J = 8.4 Hz, 1H), 7.40 (d, J = 8.4 Hz, 2H), 7.20 (dd, J = 8.4 Hz,
1.6 Hz, 1H), 6.67 (s, 1H), 6.43 (s, 1H), 4.58 (s, 2H), 4.49 (d, J = 6.4
Hz, 2H), 3.89 (s, 3H), 3.49 (q, J = 6.8 Hz, 2H), 1.14 (t, J = 6.8 Hz,
3H).
Crystallography. A fragment of human MMP-13 corresponding
to residues 104−274 was expressed using the pET29a vector system
(Invitrogen) and E. coli BL21(DE3) host strain. The protein was
expressed in insoluble form. Cell pellets were lysed at 4 °C using five,
1 min cycles of sonication in a buffer containing 50 mM MES, 75 mM
NaCl, and 0.5% Triton at pH 6.5. Following sonication, the pellet was
recovered by centrifugation, and the supernatant was decanted. This
sonication process was repeated three times, with the final pellet,
containing the inclusion bodies, being frozen at −80 °C. The pellet
containing the inclusion bodies was thawed and resolubilized at room
temperature in a buffer containing 6 M urea and 150 mM MES, pH
6.5. Inhibition of protease activity was achieved through the addition
of Complete protease inhibitor cocktails (Roche). The resuspended
material was clarified using centrifugation, and the supernatant was
recovered. The supernatant was dialyzed against a buffer containing 6
M urea and 50 mM MES, pH 6.5. The dialyzed material was then
purified using an SP Sepharose Fast Flow resin (GE LifeSciences).
Separation was achieved using a buffer containing 6 M urea and 50
mM MES, pH 6.5, and a gradient of 0−250 mM NaCl. Fractions
containing the MMP-13 were pooled and stored at −80 °C.
The protein was refolded using sequential dialysis of the protein (at
a concentration of 0.25 mg/mL) against a buffer containing 50 mM
MES, pH 6.5, 500 mM NaCl, 10 mM CaCl2, 1.0 mM ZnCl2, and urea.
The concentration of urea was reduced in the following steps: 4, 2, 1,
0, and 0 M, over the course of approximately 2 days. The dialyzed
protein was then concentrated to ∼13 mg/mL, and purification and
buffer exchange was achieved via SEC on a Superdex75 resin (GE
LifeSciences) with a running buffer of 50 mM Tris, pH 8.0, 150 mM
NaCl, and 5 mM CaCl2. The fractions corresponding to MMP-13
were pooled and concentrated to 7 mg/mL.
A 2-fold excess of compound 1 was added to the protein, and
crystals were obtained using a hanging drop procedure with a
precipitant containing 10% w/v PEG4000, 1 M ammonium formate,
and 100 mM Tris, pH 8.0. Crystals with compounds 11 and 15 were
obtained by soaking out of crystals containing compound 1.
X-ray diffraction data were collected using either a Rigaku FR-E
SuperBright generator and Saturn92 CCD detector or the PX1
beamline at the SwissLight Source. Data reduction was achieved using
HKL2000. An initial structure was obtained via molecular replacement
using published coordinates of MMP-13 (PDB: 1XUD) as a starting
model. Multiple rounds of refinement were performed using Phenix.
The costructure of MMP-13 with compound 1 had a resolution of
1.85 Å and R/Rfree statistics of 0.17/0.23. The costructure of MMP-13
with compound 15 had a resolution of 2.03 Å and R/Rfree statistics of
0.16/0.19. The costructure of MMP-13 with compound 11 had a
resolution of 1.79 Å and R/Rfree statistics of 0.24/0.28.
A flask was charged with 5-bromo-2-ethoxymethyl-1-(2-trimethylsi-
lanyl-ethoxymethyl)-1H-indole 40 (500 mg, 1.3 mmol), potassium
carbonate (380 mg, 3.9 mmol), ditetramethylenediborane (360 mg,
1.4 mmol), and bis-diphenylferrocenylpalladium(II) dichloride (210
mg, 0.3 mmol), and the solids were suspended in 1,4-dioxane (10 mL)
and DMF (7 mL). The reaction was sealed and heated to 100 °C for 1
h and then cooled to room temperature. The solution was diluted with
EtOAc and washed with water. The organic layer was dried filtered,
evaporated in vacuo, and purified on silica (heptanes/EtOAc) to give
2-ethoxymethyl-5-(4,4,5,5-tetramethyl-[1,3,2]dioxaborolan-2-yl)-1-(2-
trimethylsilanyl-thoxymethyl)-1H-indole 41 as a white solid (350 mg,
1
62%). H NMR (400 MHz, DMSO-d6): δ 7.91 (s, 1H), 7.54 (d, J =
8.4 Hz, 1H), 7.46 (d, J = 8.0 Hz, 1H), 6.54 (s, 1H), 5.56 (s, 2H), 4.64
(s, 2H), 3.50 (q, J = 7.2 Hz, 2H), 3.45 (t, J = 8.4 Hz, 2H), 1.29 (s,
12H), 1.14 (t, J = 7.2 Hz, 3H), 0.80 (t, J = 8.0 Hz, 2H), 0.10 (s, 9H).
A reaction flask was charged with the boronic acid 41 from above
(185 mg, 0.4 mmol), 4-{[(5-bromo-1-methyl-1H-pyrazole-3-carbon-
yl)-amino]-methyl}-benzoic acid 33 (120 mg, 0.33 mmol, from
example 14), and tetrakistrkistriphenyphosine palladium(0) (32 mg,
0.03 mmol), the solids were suspended in THF (3 mL), and 2 N
aqueous sodium carbonate (0.75 mL) was added. The reaction was
heated for 6 h at reflux and then cooled to room temperature, and the
solvent evaporated in vacuo. The resulting oil was taken up in EtOAc/
water and acidified to pH 2 with concentrated HCl. The product was
then extracted into copious amounts of ethyl acetate. The combined
organic layers were dried (MgSO4), filtered, and evaporated in vacuo
to give an oil that was purified on silica (MeOH/DCM) to give a solid
42 (150 mg, 75%). 1H NMR (400 MHz, DMSO-d6): δ 8.81 (t, J = 6.4
Hz, 1H), 7.89 (d, J = 8.0 Hz, 2H), 7.72 (d, J = 1.6 Hz, 1H), 7.69 (d, J
= 8.8 Hz, 1H), 7.37 (d, J = 8.0 Hz, 2H), 7.34 (dd, J = 8.4 Hz, 1.6 Hz,
1H), 6.73 (s, 1H), 6.60 (s, 1H), 5.61 (s, 2H), 4.68 (s, 2H), 4.49 (d, J =
6.4 Hz, 2H), 3.91 (s, 3H), 3.46−3.54 (m, 4H), 1.15 (t, J = 7.2 Hz,
3H), 0.82 (t, J = 8.0 Hz, 2H), 0.08 (s, 9H).
A flask was charged with 4-[({5-[2-ethoxymethyl-1-(2-trimethylsi-
lanyl-ethoxymethyl)-1H-indol-5-yl]-1-methyl-1H-pyrazole-3-carbon-
yl}-amino)-methyl]-benzoic acid 42 (25 mg, 0.04 mmol), THF (1
mL), and ethylenediame (10 μL, 0.13 mmol), and tetrabutlyammo-
nium fluoride (0.25 mL, 1 M in THF) was added dropwise to the
reaction. The reaction was heated to 70 °C for 5 h, then cooled to
room temperature, and evaporated in vacuo to give a solid that was
purified directly on reverse-phase HPLC to give the title compound 20
as a white solid (6 mg, 31%). LCMS (ES+) m/z found, 433; retention
Biological Assays. Catalytic domains of human MMP-13 were
expressed and purified from E. coli in house. MMP-13 was assayed in a
20 μL volume containing a 125 pM catalytic domain of human MMP-
13 and 2 μM 520 MMP FRET Substrate XIV [sequence: QXL520-
GABA-Pro-Cha-Abu-Smc-His-Ala-Dab(5-FAM)-Ala_Lys-NH2]
(Anaspec) in a buffer containing 100 mM Tris HCl, pH 7.5, 100 mM
NaCl, 10 mM CaCl2, and 0.05% BRIJ 35, and 1% DMSO for 30 min at
28 °C/80% humidity, and fluoresence was read at 485 nm excitation
and 535 nm emission. All other MMPs were assayed as described for
MMP-13 with enzyme substitutions as follows: 4 nM catalytic domain
of recombinant human MMP-1(Biomol), 2 nM catalytic domain of
recombinant human MMP-2 (Biomol), 0.8 nM catalytic domain of
recombinant human MMP-3 (Biomol), 0.6 nM catalytic domain of
recombinant human MMP-7 (Biomol), 2 nM catalytic domain of
recombinant human MMP-8 (Biomol), 1.5 nM catalytic domain of
recombinant human MMP-9 (Biomol), 1 nM catalytic domain of
recombinant human MMP-10 (Biomol), and 0.125 nM catalytic
8185
dx.doi.org/10.1021/jm201129m|J. Med. Chem. 2011, 54, 8174−8187

Journal of Medicinal Chemistry
Article
domain of recombinant human MMP-12 (Biomol). MMP-14 was
assayed as described for MMP-13 with an enzyme substitution of 8.8
nM catalytic domain of recombinant human MMP-14 (Biomol) and
an extended linear incubation time of 60 min at 28 °C/80% humidity.
MMP-13 with 1.25% human serum was assayed as described for
MMP-13 with the addition of 1.25% human serum to the assay buffer.
For the collagen degradation assay (MMP-13 IC50 full length), the
Rapid Collagen Assay kit by Condrex (catalog #3002) was applied to
measure the activity of MMP-13 in degrading FITC labeled collagen
II. Full-length MMP-13 was activated by 1 mM AMPA, and the
activated FL-MMP13 (600 ng/mL) was incubated with FITC-labeled
type II collagen in the kit assay buffer for 1.5 h at 37 °C. The degraded
collagen fragments were extracted into solution and measured by
fluorescent intensity of FITC (490 nm ex/520 nm em) as described in
the assay kit protocol (www.chondrex.com/protocols/Collagenase_
Kit.pdf). IC50 values are determined by nonlinear curve fitting of the
data from a duplicate 10-point concentration−response curve.
The bovine nasal cartilage (BNC) degradation assay measures the
ability of compounds to inhibit full-length MMP-13-induced cartilage
degradation. BNC explants were purchased from Northland
Laboratories (www.northlandlabs.com) as 3 mm explants in 96-well
plates. The BNC explants were washed with PBS and cultured with
activated full-length human MMP-13 (50 μg/mL) (BioMol) in the
assay buffer containing 50 mM Tris, pH 6.5, 250 mM NaCl, 5 mM
CaCl2, 1 μM ZnCl2, and 10% human serum for 2 h at 37 °C. The
cartilage degradation products C-terminal telopeptide of type II
collagen (CTXII) were quantified by an CTXII ELISA (IDS, formerly
Nordic Bioscience, catalog #3CAL4000) using the cartilage culture
supernatants. Test compound EC50 values are determined by
nonlinear curve fitting of the data from duplicate 8-point
concentration−response curve.
ene; SEM-Cl, 2-(trimethylsilyl)ethoxymethyl chloride; TBAF,
tetra-n-butylammonium fluoride; HPLC, high performance
liquid chromatography; BNC, bovine nasal cartilage; h/r LM,
human/rat liver microsome stability
REFERENCES
■
(1) Eitner, K.; Koch, U. From fragment screening to potent binders:
strategies for fragment-to-lead evolution. Mini Rev. Med. Chem. 2009,
9, 956−961. (b) Schulz, M. N.; Hubbard, R. E. Recent progress in
fragment-based lead discovery. Curr. Opin. Pharmacol. 2009, 9, 615−
621. (c) Mayr, L. M.; Bojanic, D. Novel trends in high-throughput
screening. Curr. Opin. Pharmacol. 2009, 9, 580−5888.
(2) Bembenek, S. D.; Tounge, B. A.; Reynolds, C. H. Ligand
efficiency and fragment-based drug discovery. Drug Discovery Today
2009, 14, 278−283.
(3) Jacoby, E.; Davies, J.; Blommers, M. J. Design of small molecule
libraries for NMR screening and other applications in drug discovery.
Curr. Top. Med. Chem. 2003, 3, 11−23.
(4) (a) Shuker, S. B.; Hajduk, P. J.; Meadows, R. P.; Fesik, S. W.
Discovering high-affinity ligands for proteins: SAR by NMR. Science
1996, 274, 1531−1542. (b) Hajduk, P. J.; Sheppard, G.; Nettesheim,
D. G.; Olejniczak, E. T.; Shuker, S. B.; Meadows, R. P.; Steinman, D.
H.; Carrera, G. M. Jr.; Marcotte, P. A; Severin, J.; Walter, K.; Smith,
H.; Gubbins, E.; Simmer, R.; Holzman, T. F.; Morgan, D. W.;
Davidsen, S. K.; Summers, J. B.; Fesik, S. W. Discovery of potent
nonpeptide inhibitors of stromelysinusing SAR by NMR. J. Am. Chem.
Soc. 1997, 119, 5818−5827. (c) Fejzo, J.; Lepre, C. A.; Peng, J. W.;
Bemis, G. W.; Murcko, M. A.; Moore, J. M. The SHAPES strategy: An
NMR-based approach for lead generation in drug discovery. Chem.
Biol. 1999, 6, 755−769.
(5) Neumann, T.; Junker, H. D.; Schmidt, K.; Sekul, R. SPR-based
fragment screening: Advantages and applications. Curr. Top. Med.
Chem. 2007, 7, 1630−1642.
(6) Recht, M. I.; De Bruyker, D.; Bell, A. G.; Wolkin, M. V.; Peeters,
E.; Anderson, G. B.; Kolatkar, A. R.; Bern, M. W.; Kuhn, P.; Bruce, R.
H.; Torres, F. E. Enthalpy array analysis of enzymatic and binding
reactions. Anal. Biochem. 2008, 377, 33−39.
ASSOCIATED CONTENT
■
S
* Supporting Information
In-depth procedures for target independent assays. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(7) (a) Alex, A. A.; Flocco, M. M. Fragment Based Drug Discovery.
Drug Discovery Today 2009, 13−14, 668−675. (b) Chessari, G.;
Woodhead, A. J. From fragment to clinical candidateA historical
perspective. Curr. Top. Med. Chem. 2007, 7, 1544−1567.
AUTHOR INFORMATION
■
Corresponding Author
*Tel: 203-791-6371. E-mail: steven.taylor@boehringer-
ingelheim.com.
́
(8) (a) Morrison, C. J.; Butler, G. S.; Rodriguez, D.; Overall, C. M.
Matrix metalloproteinase proteomics: substrates, targets, and therapy.
Curr. Opin. Cell Biol. 2009, 5, 645−653. (b) Cudic, M.; Fields, G. B.
Extracellular proteases as targets for drug development. Curr. Protein
Pept. Sci. 2009, 4, 297−307. (c) Tu, G.; Xu, W.; Huang, H.; Li, S.
Progress in the development of matrix metalloproteinase inhibitors.
Curr. Med. Chem. 2008, 15, 1388−1395. (d) Fingleton, B. Matrix
metalloproteinases as valid clinical targets. Curr. Pharm. Des. 2007, 13,
333−346. (e) Li, X.; Li, J. Recent advances in the development of
MMPIs and APNIs based on the pyrrolidine platforms. Mini Rev. Med.
Chem. 2010, 9, 794−805. (f) Woessner, J. F., Nagase, H., Eds. Matrix
Metalloproteinases and TIMPs; Oxford University Press: New York,
2000; pp 1−240.
(9) Historically, MMP-1 and -14 were the postulated key players in
causing MSS. However, the nonselective MMP-13 inhibitor CP-
544439, which is inactive against MMP-1, has caused MSS in clinical
trials: Reiter, L. A.; Freeman-Cook, K. D.; Jones, C. S.; Martinelli, G.
J.; Antipas, A. S.; Berliner, M. A.; Datta, K.; Downs, J.; Eskra, J. D.;
Forman, M. D.; Greer, E. M.; Guzman, R.; Hardink, J. R.; Janat, F.;
Keene, N. F.; Laird, E. R.; Liras, J.; Lopresti-Morrow, L. L.; Mitchell, P.
G.; Pandit, J.; Robertson, D.; Sperger, D.; Vaughn-Bowser, M. L.;
Waller, D. M.; Yocum, S. A. Potent, selective pyrimidinetrione-based
inhibitors of MMP-13. Bioorg. Med. Chem. Lett. 2006, 5822−5826.
MMP-14 has KO mice data supporting it being a strong candidate for
MSS Rowan, A. D.; Litherland, G. J.; Hui, W.; Milner, J. M.
Metalloproteases as potential therapeutic targets in arthritis treatment.
Expert Opin. Ther. Targets 2008, 12, 1−18, and references therein..
ACKNOWLEDGMENTS
■
The authors thank Kathleen Haverty, Alistair Baptiste and
Xiang Zhu for expression and purification of the human MMP-
13 catalytic domain.
ABBREVIATIONS USED
■
MMP, matrix metalloproteases; HTS, high-throughput screen-
ing; ALI, alternative lead identification; FBS, fragment-based
screening; MW, molecular weight; AMU, atomic mass units;
clog P, calculated partion coefficient in octanol:water; NMR,
nuclear magnetic resonance; SPR, surface plasmon resonance;
MSS, muscular skeletal syndrome; STD-NMR, saturation
transfer difference nuclear magnetic resonance; SEC-MS, size
exclusion chromatography mass spectrometry; HBA, hydrogen
bond acceptor; HBD, hydrogen bond donor; LE, ligand
efficiency RT ln(IC50)/no. of heavy atoms; PAMPA, parallel
artificial membrane permeability assay; PSA, polar surface area;
DMF, dimethylformamide; HATU, 2-(1H-7-azabenzotriazol-1-
yl)-1,1,3,3-tetramethyl uronium hexafluorophosphate methana-
minium; TBTU, O-(benzotriazol-1-yl)-N,N,N′,N′-tetramethy-
luronium tetrafluoroborate; DIEA, N,N-diisopropylethylamine;
THF, tetrahydrofuran; DBU, 1,8-diazabicyclo[5.4.0]undec-7-
8186
dx.doi.org/10.1021/jm201129m|J. Med. Chem. 2011, 54, 8174−8187

Journal of Medicinal Chemistry
Article
(10) Vincenti, M. P.; Brinckerhoff, C. E. Transcriptional regulation of
collagenase (MMP-1, MMP-13) genes in arthritis: Integration of
complex signaling pathways for the recruitment of gene-specific
transcription factors. Arthritis Res. 2002, 4, 157−164.
(18) (a) Heim-Riether, A.; Taylor, S. J.; Liang, S.; Gao, D. A.; Xiong,
Z.; August, M. E.; Collins, B. K.; Farmer, B. T. 2nd; Haverty, K.; Hill-
Drzewi, M.; Junker, H. D.; Margarit, M. S.; Moss, N.; Neumann, T.;
Proudfoot, J. R.; Keenan, L. S.; Sekul, R.; Zhang, Q.; Li, J.; Farrow, N.
A. Improving potency and selectivity of a new class of non-Zn-
chelating MMP-13 inhibitors. Bioorg. Med. Chem. Lett. 2009, 18,
5321−5324. (b) Gao, D. A.; Xiong, Z.; Heim-Riether, A.; Amodeo, L.;
August, E. M.; Cao, X.; Ciccarelli, L.; Collins, B. K.; Harrington, K.;
Haverty, K.; Hill-Drzewi, M; Li, X.; Liang, S.; Margarit, S. M.; Moss,
N.; Nagaraja, N.; Proudfoot, J.; Roman, R.; Schlyer, S.; Keenan, L. S.;
Taylor, S.; Wellenzohn, B.; Wiedenmayer, D.; Li, J.; Farrow, N. A. SAR
studies of non-zinc-chelating MMP-13 inhibitors: Improving selectivity
and metabolic stability. Bioorg. Med. Chem. Lett. 2010, 17, 5039−5043.
(19) Boger, D. L. Preparation of CBI analogues of the duocarmycins
and CC-1065 as antitumor agents with DNA alkylating activity. PCT
Int. Appl. WO2004101767, 2004; 144.
(11) Tallant, C.; Marrero, A.; Gomis-Ruth, F. X. Matrix metal-
̈
loproteinases: Fold and function of their catalytic domains. Biochim.
Biophys. Acta 2010, 1, 20−28.
(12) (a) Jacobsen, J. A.; Major Jourden, J. L.; Miller, M. T.; Cohen, S.
M. To bind zinc or not to bind zinc: an examination of innovative
approaches to improved metalloproteinase inhibition. Biochim. Biophys.
Acta 2010, 1, 72−94. (b) Nuti, E.; Tuccinardi, T.; Rossello, A. Matrix
metalloproteinase inhibitors: new challenges in the era of post broad-
spectrum inhibitors. Curr. Pharm. Des. 2007, 20, 2087−2100.
(13) Engel, C. K.; Pirard, B.; Schimanski, S.; Kirsch, R.; Habermann,
J.; Klingler, O.; Schlotte, V.; Weithmann, K. U.; Wendt, K. U.
Structural basis for the highly selective inhibition of MMP-13. Chem.
Biol. 2005, 2, 181−189.
(20) Yasuma, T.; Ujikawa, O.; Iwata, H. Preparation of indoles and
related compounds as glucokinase activators. PCT Int. Appl.
WO2006112549, 2006; 379 pp.
(14) Devel, L.; Czarny, B.; Beau, F.; Georgiadis, D.; Stura, E.; Dive,
V. Third generation of matrix metalloprotease inhibitors: Gain in
selectivity by targeting the depth of the S1′ cavity. Biochimie 2010, 11,
1501−1508.
(15) (a) Wu, J.; Rush, T. S. 3rd; Hotchandani, R.; Du, X.; Geck, M.;
Collins, E.; Xu, Z. B.; Skotnicki, J.; Levin, J. I.; Lovering, F. E.
Identification of potent and selective MMP-13 inhibitors. Bioorg. Med.
Chem. Lett. 2005, 18, 4105−4109. (b) Monovich, L. G.; Tommasi, R.
A.; Fujimoto, R. A.; Blancuzzi, V.; Clark, K.; Cornell, W. D.; Doti, R.;
Doughty, J.; Fang, J.; Farley, D.; Fitt, J.; Ganu, V.; Goldberg, R.;
Goldstein, R.; Lavoie, S.; Kulathila, R.; Macchia, W.; Parker, D. T.;
Melton, R.; O'Byrne, E.; Pastor, G.; Pellas, T.; Quadros, E.; Reel, N.;
Roland, D. M.; Sakane, Y.; Singh, H.; Skiles, J.; Somers, J.; Toscano,
K.; Wigg, A.; Zhou, S.; Zhu, L.; Shieh, W. C.; Xue, S.; McQuire, L. W.
Discovery of potent, selective, and orally active carboxylic acid based
inhibitors of matrix metalloproteinase-13. J. Med. Chem. 2009, 11,
3523−3538. (c) Levin, J. I.; Chen, J. M.; Du, M. T.; Nelson, F. C.;
Wehr, T.; DiJoseph, J. F.; Killar, L. M.; Skala, S.; Sung, A.; Sharr, M.
A.; Roth, C. E.; Jin, G.; Cowling, R.; Di, L.; Sherman, M.; Xu, Z. B.;
March, C. J.; Mohler, K. M.; Black, R. A.; Skotnicki, J. S. The discovery
of anthranilic acid-based MMP inhibitors. Bioorg. Med. Chem. Lett.
2001, 22, 2975−2978. (d) Skiles, J. W.; Gonnella, N. C.; Jeng, A. Y.
The design, structure, and clinical update of small molecular weight
matrix metalloproteinase inhibitors. Curr. Med. Chem. 2004, 11, 2911−
2977. (e) Rao, B. G. Recent developments in the design of specific
Matrix Metalloproteinase inhibitors aided by structural and computa-
tional studies. Curr. Pharm. Des. 2005, 11, 295−322. (f) Wasserman, Z.
R. Making a new turn in matrix metalloprotease inhibition. Chem. Biol.
2005, 2, 143−144.
(16) (a) Engel, C. K.; Pirard, B.; Schimanski, S.; Kirsch, R.;
Habermann, J.; Klingler, O.; Schlotte, V.; Weithmann, K. U.; Wendt,
K. U. Structural Basis for the Highly Selective Inhibition of MMP-13.
Chem. Biol. 2005, 12, 181−189. (b) Klingler, O.; Kirsch, R.;
Habermann, J.; Weithmann, K.-U.; Engel, C.; Pirard, B. Preparation
of novel pyrimidine-4,6-dicarboxamides for the selective inhibition of
collagenases. PCT Int. Appl. WO 2004041788, 2004. The crystal
structure of the selective dicarboxamide pyrimidine MMP-13 inhibitor
in complex with MMP-13 revealed a distinctive hydrogen-bonding
network in the catalytic site. A structure-based virtual screening query
was devised requiring hydrogen bonding to T245, T247, A238, and
either to water 7 found in the cocrystal structure (PDB ID: 1XUD) or
directly to M253. In addition, reaching either the S1′ or the S1′*
pocket was required. Our internal compound collection was virtually
screening against MMP-13 using Glide docking with constraints as
described above yielding a hit set of approximately 15000 compounds.
After clustering, visual inspection, and discarding structurally obvious
analogues of known MMP-13 inhibitors, we chose 100 compounds to
be tested for MMP-13 inhibition in a molecular assay.
(17) Salemme, F. R.; Spurlino, J.; Bone, R. Serendipity meets
precision: the integration of structure-based drug design and
combinatorial chemistry for efficient drug discovery. Structure 1997,
3, 319−324.
8187
dx.doi.org/10.1021/jm201129m|J. Med. Chem. 2011, 54, 8174−8187