Discovery of potent and selective 8-fluorotriazolopyridine c-met inhibitors

Article abstract of DOI:10.1021/jm501913a

The overexpression of c-Met and/or hepatocyte growth factor (HGF), the amplification of the MET gene, and mutations in the c-Met kinase domain can activate signaling pathways that contribute to cancer progression by enabling tumor cell proliferation, survival, invasion, and metastasis. Herein, we report the discovery of 8-fluorotriazolopyridines as inhibitors of c-Met activity. Optimization of the 8-fluorotriazolopyridine scaffold through the combination of structure-based drug design, SAR studies, and metabolite identification provided potent (cellular IC₅₀ < 10 nM), selective inhibitors of c-Met with desirable pharmacokinetic properties that demonstrate potent inhibition of HGF-mediated c-Met phosphorylation in a mouse liver pharmacodynamic model.

Full text of DOI:10.1021/jm501913a

Article
pubs.acs.org/jmc
Discovery of Potent and Selective 8‑Fluorotriazolopyridine c‑Met
Inhibitors
Emily A. Peterson,*,† Yohannes Teffera,† Brian K. Albrecht,† David Bauer,† Steven F. Bellon,†
Alessandro Boezio,† Christiane Boezio,† Martin A. Broome,‡ Deborah Choquette,† Katrina W. Copeland,†
Isabelle Dussault,‡ Richard Lewis,† Min-Hwa Jasmine Lin,† Julia Lohman,† Jingzhou Liu,†
Michele Potashman,† Karen Rex,‡ Roman Shimanovich,† Douglas A. Whittington,† Karina R. Vaida,†
and Jean-Christophe Harmange†
†Amgen Inc., 360 Binney Street, Cambridge, Massachusetts 02142, United States
‡Amgen Inc., One Amgen Center Drive, Thousand Oaks, California 91320, United States
S
* Supporting Information
ABSTRACT: The overexpression of c-Met and/or hepato-
cyte growth factor (HGF), the amplification of the MET gene,
and mutations in the c-Met kinase domain can activate
signaling pathways that contribute to cancer progression by
enabling tumor cell proliferation, survival, invasion, and
metastasis. Herein, we report the discovery of 8-fluorotriazo-
lopyridines as inhibitors of c-Met activity. Optimization of the
8-fluorotriazolopyridine scaffold through the combination of
structure-based drug design, SAR studies, and metabolite
identification provided potent (cellular IC50 < 10 nM), selective inhibitors of c-Met with desirable pharmacokinetic properties
that demonstrate potent inhibition of HGF-mediated c-Met phosphorylation in a mouse liver pharmacodynamic model.
triazolopyridine ring (1c, Table 1). It was later discovered that
compounds of class 1a demonstrated time-dependent inhib-
ition (TDI) of cytochrome P450 3A4 (CYP3A4) and that the
oxygen linker bridging the methoxyquinoline and imidazopyr-
idine rings was a site of metabolic O-dealkylation.7 Considering
these liabilities, we decided to investigate c-Met inhibitors
bearing a more robust carbon linkage, such as previously
reported phenol 2.5 Replacing the phenol with a quinoline
bioisostere provided the more potent inhibitor 3a. As with
inhibitor 1c, incorporation of the 8-fluorine substituent to give
8-fluorotriazolopyridine methylquinoline 3b improved the
potency. Finally, installation of two additional fluorine atoms
on the carbon bridging the 8-fluorotriazolopyridine and
quinoline moieties gave inhibitor 4 (Table 1). This substitution
was pursued to eliminate the potential for metabolic oxidation
of the activated benzylic carbon while maintaining potency.8
Following this promising lead, structure−activity relationship
(SAR) studies were initiated with the aim of further improving
the potency on c-Met, establishing desirable pharmacokinetic
properties and eliminating the inhibition of CYP3A4 observed
with the previously reported scaffold 1. The three structural
elements that were investigated starting from 8-fluorotriazolo-
pyridine 4 were the C6 aryl group, the hinge-binder moiety,
and the substitution of the carbon atom bridging the hinge-
binder and the triazolopyridine core.9 Herein we describe the
INTRODUCTION
■
The regulation of intracellular signal transduction pathways is a
key function of receptor tyrosine kinases (RTKs), and thus,
inappropriate activation of RTKs can lead to deleterious
consequences with regard to downstream signaling. As such,
the receptor tyrosine kinase c-Met has been implicated in the
progression of a variety of human cancers.1 In normal cells,
activation of c-Met occurs through the extracellular binding of
its natural ligand, hepatocyte growth factor/scatter factor
(HGF/SF). It has been shown that a variety of human cancers
contain aberrant HGF/SF or c-Met expression or activating c-
Met kinase mutations.2 Consequently, inhibition of c-Met
activity is a potentially impactful approach to the treatment of
cancers where c-Met is activated.3
An ATP-competitive, small-molecule inhibitor of c-Met has
the potential to inhibit both ligand-dependent and ligand-
independent tumors that are driven by c-Met. As a result, there
has been significant interest in the discovery of small molecule
c-Met inhibitors for the treatment of cancer.4 We previously
reported the synthesis and activity of c-Met inhibitors 1a, which
demonstrated nanomolar inhibition of c-Met phosphorylation
and exquisite selectivity over other kinases.5 Concurrent with
the design and testing of triazolopyridazine inhibitors such as
1a, a modification of the central ring system was investigated in
which the triazolopyridazine was replaced by a triazolopyridine
(Table 1).6 The removal of the N5 pyridazine nitrogen (1a →
1b) led to a measurable loss in potency; however, activity could
be restored by the installation of an 8-fluoro substituent on the
Received: December 10, 2014
Published: February 20, 2015
© 2015 American Chemical Society
2417
DOI: 10.1021/jm501913a
J. Med. Chem. 2015, 58, 2417−2430

Journal of Medicinal Chemistry
Article
Table 1. Modification of the Triazolopyridazine Scaffold
a
Compound
X
R
c-Met IC50 (nM)
1a
1b
1c
2
N
H
H
F
9
31
CH
CH
7
2100
210
60
3a
3b
4
H
F
23
a
Inhibition of c-Met kinase, HTRF, n ≥ 2.
a
Scheme 1. Synthesis of Inhibitor 4
a
Reagents and conditions: (i) X-Phos, Pd(OAc)2, K3PO4, PhB(OH)2, dioxane, 100 °C, 18 h, 94%; (ii) NH2NH2, i-PrOH, 70 °C, 86%; (iii)
Pd(PPh3)4, (2-tert-butoxy-2-oxoethyl)zinc(II) bromide, THF, 75 °C, 18 h, 55%; (iv) LiHMDS, (PhSO4)2NF, −78 °C → rt, 4 h, 47%; (v) (a)
trifluoroacetic acid, CH2Cl2, rt, 18 h, 93%; (b) thionyl chloride, DMF, 0 °C, 1 h, then 6, i-Pr2EtN, DMAP, 18 h, 87% from 8; (vi) Cl3CCN, PS-PPh3,
DCE, 100 °C, microwave, 10%.
design, synthesis, and pharmacodynamic activity of 8-
fluorotriazolopyridine c-Met kinase inhibitors such as 4.
core gave improved potency, presumably due to an optimal
configuration for π-stacking with Tyr1230 (Figure 1, vide
infra).5 The replacement of the nitrogen at C5 with CH (1a →
1b) potentially introduces a steric clash that may result in a
deviation from the desired planarity. Reasoning that inhibitors
such as 4 would be subject to the same interaction, installation
of a 2-pyridyl group at the C6 position gave 10a, which
demonstrated an IC50 = 12 nM for c-Met kinase inhibition
(Table 2). Further enhancement in potency could be obtained
by the installation of five-membered ring heterocycles at the C6
position; crystallography revealed that heterocycles such as the
isoxazole in 10b could participate in an advantageous water
bridge in the kinase active site (Figure 1). The heterocycles
exhibiting the best cellular potency were thiazole 10d, isoxazole
10b, and methylpyrazole 10e (Table 2).
RESULTS AND DISCUSSION
■
The synthesis of 4 (Scheme 1) began with Suzuki (or Stille)10
coupling of phenylboronic acid and 5-chloro-2,3-difluoropyr-
idine (5) followed by SNAr substitution with hydrazine to
afford phenylpyridine 6. The difluoroquinoline ester 8 was
prepared by Negishi coupling of 6-bromoquinoline and the zinc
enolate of tert-butyl acetate followed by bis-fluorination. The
union of hydrazine 6 with the acid chloride derived from tert-
butyl ester 8 provided hydrazide 9. Dehydrative cyclization of
hydrazide 9 could be achieved by microwave heating with
trichloroacetonitrile and solid supported triphenylphosphine to
provide inhibitor 4.11
The next modification investigated in the difluoromethylqui-
noline series (10) was substitution of the hinge-binder
quinoline moiety, which engages hinge-residue Met1160
through a hydrogen bond. Several alternative hinge-binder
groups were investigated using the methylpyrazole at the C6
position. To provide for more rapid access to these analogs, the
inhibitors in Table 3 were prepared without the fluorine atoms
on the bridging carbon using the more synthetically viable
By use of the synthesis described in Scheme 1, the effect of
various aryl groups at the C6 position was investigated with the
guidance of protein structure information. Analysis of the
cocrystal of 1a with the c-Met kinase domain revealed that
coplanarity of the C6 phenyl group with the triazolopyridazine
ring was important for potency. As previously described for
inhibitors 1a, it was observed that substituents at the C6
position that could achieve coplanarity with the triazolopyridine
2418
DOI: 10.1021/jm501913a
J. Med. Chem. 2015, 58, 2417−2430

Journal of Medicinal Chemistry
Article
hydrogen-bonding basicity, weakening the key hinge-binding
interaction with Met1160.12
Table 2. Modification of the C6 Position
Combination of the C6 substituents from the most potent
inhibitors in Table 2 with the methoxyquinoline hinge-binder
component provided compounds 12a and 12b, which
demonstrated single-digit nanomolar inhibition of HGF-
mediated c-Met phosphorylation in PC3 cells (Table 4).
These inhibitors, along with the most potent inhibitors from
Table 2, were further evaluated in vitro in liver microsomes and
in pharmacokinetic studies conducted in male Sprague−Dawley
rats. It was found that the intrinsic clearance (CLint) obtained
from rat liver microsomes (data not shown) agreed well with
the observed clearance in rat and that compounds 12
demonstrated lower turnover than compounds 10 in both rat
and human liver microsomes. Furthermore, compounds 10 and
12 were not potent inhibitors of cytochrome P450 enzymes
CYP2D6 and CYP3A4, an improvement over previously
reported scaffold 1. With the exception of 10d, the compounds
in Table 4 had low clearance, small volume of distribution at
steady state, reasonable oral half-life, and good oral
bioavailability.
Metabolic profiling revealed that the major metabolic fate of
the compounds shown in Table 4 was glutathione S-transferase
(GST) mediated displacement of the aromatic fluorine at the
C8 position with glutathione (GSH) (Figure 2).13 Metabolism
by GSH conjugation can prove problematic if a drug is given at
high or repeated doses, because of potential GSH depletion
leading to oxidative stress.14 In addition, if GST-mediated GSH
conjugation is the predominant metabolic pathway for a
compound, metabolic prediction studies could be complicated
by the occurrence of varying GST isoforms within the human
population.13 Considering these potential drawbacks, diminish-
ing the extent of glutathione conjugation became the primary
focus (see Figure 2 for GSH displacement studies). While
removal of the C8 fluorine would eliminate this metabolic
pathway, the presence of this fluorine was important for
maintaining potency. It was hypothesized that the electrophilic
nature of the 8-fluorotriazolopyridine could be attenuated by
removal of inductively electron-withdrawing groups elsewhere
in the molecule to potentially mitigate the extent of glutathione
displacement. Therefore, initial efforts focused on systemati-
cally replacing the benzylic fluorines with methyl groups and/or
hydrogens to give inhibitors 20−24 (Table 5).
a
b
Inhibition of c-Met kinase activity, n ≥ 2. Inhibition of HGF-
mediated c-Met phosphorylation in PC3 cells, n ≥ 2.
Table 3. Modification of the Hinge-Binder
The synthesis of inhibitors 20 and 21 is described in Scheme
2; the synthesis of compounds 22−24 was performed in a
similar manner from the appropriately functionalized quinoline
carboxylic acids. A palladium-catalyzed acylation of the
appropriately substituted quinoline bromide 13 provided
quinoline ester 15, which could be fluorinated to give
fluoromethylquinoline 16. The carboxylic acid 17, derived
from hydrolysis of the tert-butyl ester 15 or 16, was coupled
with the arylhydrazone 18 under standard amide coupling
conditions to provide hydrazide 19. Subsequent dehydrative
cyclization provided inhibitors 20 and 21.15 Compounds
containing stereogenic centers (20, 21, and 24) could be
obtained in >95% enantiomeric excess following chromato-
graphic separation on chiral media.16
It was found that replacing one of the bridging fluorine atoms
with a methyl group (20) was tolerated, and for several
compounds the cellular IC50 was less than 10 nM (Table 5).
The conformation of the bridging carbon stereocenter had a
significant effect on potency, for example, (R)-20a demon-
strated IC50 = 5 nM in the c-Met kinase inhibition assay
a
Inhibition of c-Met kinase, HTRF, n ≥ 2.
methylpyrazole at the C6 position. Substitution meta to the
quinoline nitrogen atom was investigated, where substituents
could project into the solvent front without interfering with the
key hydrogen bonding interaction of the quinoline with
Met1160. Installation of a methoxy group on the quinoline
maintained the potency observed for parent 11a, whereas a
hydroxyl group was detrimental. Other modifications, such as
quinoxaline 11d or aminobenzthiazoles 11e and 11f were
tolerated but did not improve potency (Table 3). The loss in
potency observed for 11d−f is potentially a result of decreased
2419
DOI: 10.1021/jm501913a
J. Med. Chem. 2015, 58, 2417−2430

Journal of Medicinal Chemistry
Article
h
Table 4. PK Properties of Selected Inhibitors
a
b
c
Inhibition of c-Met kinase activity, HTRF, n ≥ 2. Inhibition of HGF-mediated c-Met phosphorylation in PC3 cells, n ≥ 2. Rat IV, 0.25 mg/kg,
d
formulation: DMSO. Rat PO, 2.0 mg/kg, formulation: 2% (w/v) hydroxypropylmethylcellulose, 1% (w/v) Tween 80 in water, pH adjusted to 2.2
with HCl. Formulation: 2% (w/v) hydroxypropylmethylcellulose, 1% (w/v) Tween 80 in water, pH adjusted to 2.2 with MSA. Formulation: 10%
(v/v) dimethylacetamide, 2% (w/v) hydroxypropylmethylcellulose, 1% (w/v) Tween 80 in water, pH adjusted to 2.2 with HCl. NC = not
calculated because of lack of slope to characterize the terminal disposition rate constant. ND = not determined.
e
f
g
h
whereas (S)-20a had IC50 = 780 nM. Analysis of the cocrystal
structure of (R)-20c in the kinase domain of c-Met provides an
explanation for these results (Figure 1).17 The inhibitor (R)-
20c adopts a bent conformation to accommodate Met1211,
which projects the methyl group on the bridging carbon back
into a small pocket. The adjacent fluorine occupies a position
relatively close to the protein surface, which would likely not
accommodate a larger methyl group, thus potentially explaining
the loss in potency for (S)-20a, gem-dimethyl-bridged
compound 22, and cyclopropyl compound 23 (Table 5).
Other key interactions in the kinase active site are hydrogen
bonds between the quinoline and hinge residue Met1160, the
imidazopyridine core and Asp1222, as well as a water bridge
between the isoxazole nitrogen and Arg1086. As described
previously,5 the coplanar nature of the C6 aryl substituent and
the 8-fluorotriazolopyridine core is important for maintaining
the π-stacking interaction with Tyr1230 as well as the
bifurcated hydrogen-bonding interaction between the aromatic
hydrogens of the isoxazole, the C7 position of the
triazolopyridine core, and the carbonyl oxygen atom on residue
Arg1208. This cocrystal structure also reveals the positioning of
the 8-fluorine atom in a small cleft within the active site, where
it sits perpendicular to the backbone carbonyl of Asn 1209 (not
pictured), which may explain the increase in potency observed
for the 8-fluorotriazolopyridine core.18 Inhibitors bearing only a
methyl group on the bridging carbon (21) demonstrated
potencies similar to inhibitors 20. Again, conformation of the
bridging carbon stereocenter had a significant effect on potency,
and the absolute stereochemical configurations of compounds
20, 21, and 24 were determined by potency correlation with
(R)-20c and (S)-21f (see Supporting Information).19 The
steric constraint for this position is further confirmed by the
decreased potency of compounds 22−24, which contain more
bulky substituents on the bridging carbon.
As mentioned previously, it was discovered that the
electronic properties of the substituents at the bridging carbon
and the aromatic group at C6 synergistically affected the
potency, metabolic stability, and PK properties of each scaffold.
One of the major considerations with the 8-fluorotriazolopyr-
idine core was GST-mediated GSH conjugation. To determine
the liability of the 8-fluorotriazolopyridine moiety in each
scaffold toward displacement by GSH, several of the more
promising inhibitors from Tables 4 and 5 were incubated in rat
liver microsomes (RLM) in the presence of GSH. After 2 h of
2420
DOI: 10.1021/jm501913a
J. Med. Chem. 2015, 58, 2417−2430

Journal of Medicinal Chemistry
Table 5. SAR with Respect to Modification of the Bridging Carbon Atom ,
Article
f g
a
b
c
Inhibition of c-Met kinase activity, n ≥ 2. Inhibition of HGF-mediated c-Met phosphorylation in PC3 cells, n ≥ 2. Rat IV, 0.25 mg/kg,
d
formulation: DMSO. Rat PO, 2.0 mg/kg, formulation: 2% (w/v) hydroxypropylmethylcellulose, 1% (w/v) Tween 80 in water, pH adjusted to 2.2
e
f
with HCl. Rat IV, 0.25 mg/kg (20% (w/v) hydroxypropyl β-cyclodextrin in water pH adjusted to 3.5 with MSA). For compounds 20
stereochemistry was assigned by correlation to (R)-20c based on in vitro potency. For compounds 21 and 24 stereochemistry was assigned by
correlation to (S)-21f based on in vitro potency (cocrystal structure of c-Met and (S)-21f in table of contents graphic and Supporting
Information).17 ND = not determined.
g
incubation, the ratio of GSH conjugate to remaining parent was
measured by LC−MS.20 The degree of GSH conjugation was
considerably influenced by the electronic nature of the
substituents on the bridging carbon and the heterocycle at
2421
DOI: 10.1021/jm501913a
J. Med. Chem. 2015, 58, 2417−2430

Journal of Medicinal Chemistry
Article
a
Scheme 2. Synthesis of Inhibitors Bearing a Methyl Group on the Bridging Carbon
a
Reagents and conditions: (i) Pd(dba)2, (t-Bu)3P, NaHMDS, toluene, 80 °C, 24 h, 71−76%; (ii) LiHMDS, (PhSO4)2NF, THF, −78 °C → rt, 71−
76%; (iii) HCl, THF, 2 h, rt, 70−99%; (iv) HATU, i-Pr2EtN, DMF, rt, 78−90%; (v) PPh3, DEAD, TMSN3, THF, rt, 30 min, 22−88%.
Figure 1. Cocrystal structure of inhibitor (R)-20c in the c-Met kinase
domain (PDB code 4XMO).
the C6 position. Inhibitors bearing the more inductively
Figure 2. Effect of substituents on the extent of glutathione
withdrawing isoxazole at C6 and geminal fluorine atoms on the
conjugation. Samples were incubated in rat liver microsomes (RLM)
bridging carbon formed the greatest amounts of GSH
in the presence of glutathione (GSH) for 2 h at pH 7.4. The structures
conjugate. The extent of GSH displacement was greatly
of the GSH adducts were confirmed by high resolution accurate MS
and MS/MS.
decreased by removal of the fluorines present on the bridging
carbon (10b → (R)-20c → (S)-21e, Figure 2). The electron
donating capability of the heterocycle attached at the C6
position also significantly modulated the extent of GSH
displacement, with the more electron donating thiazole and
pyrazole groups showing negligible incorporation of GSH.21 As
anticipated, removal of the bridging fluorine atoms significantly
reduced the amount of GSH addition as demonstrated by
methyl-bridged isoxazole (S)-21e (negligible GSH conjuga-
tion) and methyl-bridged C6 pyrazole (S)-21d (no GSH
conjugate detected).
To determine if the in vitro prediction of glutathione
conjugation accurately reflected the in vivo metabolism, two
compounds were chosen for rat bile duct cannulation (BDC)
studies, 10b and (S)-21f. In the experiment, male rats were
dosed intravenously with 2 mg of compound and the collected
bile was evaluated by LC−MS to determine the metabolites. As
expected, the majority of the metabolites (>90% based on MS
intensity) resulting from treatment with compound 10b were
identified as GSH conjugates. For (S)-21f, which showed
negligible GSH-conjugation in the in vitro study, less than 10%
of the detected metabolites (based on MS intensity) consisted
of products resulting from GSH addition, thus demonstrating
that the in vitro study correlated well with the metabolic
outcome in vivo.22
Similar to the difluoro-bridged compounds 10 and 12, the
most promising compounds contained the isoxazole and
methylpyrazole groups at the C6 position of the 8-
fluoropyridine core (Table 5, (R)-20c, (R)-20d, (S)-21d,
(S)-21e, and (S)-21f). Comparison of the compounds in Table
5 reveals that inhibitors (S)-21e and (S)-21f, which contain the
isoxazole heterocycle at the C6 position, demonstrated the best
combination of potency and pharmacokinetic properties.23
Since inhibitors (S)-21e and (S)-21f showed the best overall
profile with respect to potency and pharmacokinetics with
2422
DOI: 10.1021/jm501913a
J. Med. Chem. 2015, 58, 2417−2430

Journal of Medicinal Chemistry
Article
Figure 3. Metabolite profile of (S)-21e in liver microsomes. Selected ion chromatograms of metabolites identified by accurate mass spectrometry
and MS/MS were used to generate the profiles. MsLM = mouse liver microsomes, RLM = rat liver microsomes, DLM = dog liver microsomes, MLM
= monkey liver microsomes, HLM = human liver microsomes.
Figure 4. Metabolite profile for (S)-21f in liver microsomes. Selected ion chromatograms of metabolites identified by accurate mass spectrometry
and MS/MS were used to generate the profiles. MsLM = mouse liver microsomes, RLM = rat liver microsomes, DLM = dog liver microsomes, MLM
= monkey liver microsomes, HLM = human liver microsomes.
negligible amounts of GST-mediated GSH conjugation, their
metabolic fate was further investigated by incubation with
NADPH in liver microsomes. Analysis by LC−MS revealed
that (S)-21e displayed differing metabolite profiles between
dog, rat, monkey, and human microsomes, which could present
complications for further advancement (Figure 3).24 Incubation
of (S)-21e in monkey and human liver microsomes showed
oxidation of the linker methyl group (M4) as the major
pathway of biotransformation; however, this metabolite was
below the detection limit in mouse, dog, and rat liver
2423
DOI: 10.1021/jm501913a
J. Med. Chem. 2015, 58, 2417−2430

Journal of Medicinal Chemistry
Article
microsomes. The major metabolite in mouse liver microsomes
was demethylation of the methoxyquinoline (M1). This
metabolite was also present in rat and dog liver microsomes
in addition to metabolites resulting from oxidation of the
methylisoxazole (M2) and the triazolopyridine ring (M3). For
comparison, inhibitor (S)-21f, bearing the ethoxymethoxy
solubilizing group on the quinoline, was also incubated with
NADPH in the presence of the liver microsome panel. In this
case the metabolic profile was relatively consistent across all
species.25 The introduction of the ethoxymethoxy solubilizing
chain provided an alternative site of metabolism that was the
major site of metabolism in all species. The primary metabolites
formed after incubation of (S)-21f with mouse, rat, dog,
monkey, and human liver microsomes were demethylation of
the ethoxymethoxy side chain (M6) and subsequent oxidation
to acid M7 (Figure 4).
CONCLUSIONS
■
By utilization of a combination of structure-based drug design
as well as the structure−activity relationships between potency,
pharmacokinetics, and the optimization of metabolite profiles,
new c-Met inhibitors containing an 8-fluorotriazolopyridine
core were discovered. Throughout these studies it was observed
that metabolic liabilities could be significantly improved by
changes in the substituents at positions remote from the site of
reactivity and that the extent of glutathione conjugation could
be mitigated by the removal of inductively electron-with-
drawing groups from sites peripheral to the 8-fluorine
substituent. Optimized molecules from this new scaffold
demonstrated up to 99% inhibition of HGF-stimulated c-Met
phosphorylation in our murine liver pharmacodynamic model
at a 10 mg/kg PO dose as exemplified by inhibitor (S)-21f.
The more consistent metabolic profile for (S)-21f led to its
selection for additional in vivo studies. Before assessing the in
vivo pharmacological activity of (S)-21f, it was evaluated
against a broader panel of kinases to understand its selectivity
profile. This effort revealed that (S)-21f demonstrated exquisite
selectivity for c-Met with no significant inhibition of the 100
other kinases in the panel.26 To demonstrate that compound
(S)-21f inhibits c-Met activity in vivo, a pharmacodynamic
assay measuring changes in HGF-induced phosphorylation of c-
Met in the mouse liver was used. Mice were administered a
single oral dose of compound (S)-21f at 10 mg/kg. Human
recombinant HGF was injected intravenously at 1, 3, 6, 9, 12,
and 24 h postdose. Liver and blood were harvested 5 min after
administration of HGF. Treatment with compound (S)-21f
resulted in inhibition of c-Met phosphorylation in the liver that
was related to unbound plasma concentration (Figure 5).27 The
in vivo efficacy of (S)-21f, which was sustained above 50%
inhibition for up to 9 h, correlated well with the observed in
vitro cellular IC50 of 5 nM.28
EXPERIMENTAL SECTION
■
Chemistry. Unless otherwise noted, all materials were obtained
from commercial suppliers and used as obtained. Anhydrous organic
solvents were purchased from Aldrich packaged under nitrogen in
Sure/Seal bottles and used directly. Microwave reactions were
performed using a Biotage Initiator microwave. Reactions were
monitored using Agilent 1100 series LCMS with UV detection at
254 and 215 nm and a low resonance electrospray mode (ESI).
Medium pressure liquid chromatography (MPLC) was performed on a
CombiFlash Companion (Teledyne Isco) with Redisep normal-phase
silica gel (35−60 μm) columns and UV detection at 254 nm.
Preparative reverse-phase HPLC was performed on a Gilson (GX-281
liquid handler), 150 mm × 30 mm i.d. column, 5 μm particle size,
eluting with a binary solvent system A and B using a gradient elusion
(A, water with 0.1% TFA; B, MeCN with 0.1%TFA) with UV
detection at 254 nm. Purity was measured using Agilent 1100 series
high performance liquid chromatography (HPLC) with UV detection
at 254, 215, and 280 nm (15 min; 1.5 mL/min flow rate; eluting with a
binary solvent system A and B using a gradient elusion (A, water with
0.1% TFA; B, MeCN with 0.1%TFA). Unless otherwise noted, the
purity of all compounds was ≥95%. Enantiomeric excess for
compounds bearing stereogenic centers was determined using
analytical superfluid chromatography (SFC) on a Chiralpak AS-H
1
column. H NMR spectra were recorded on a Bruker AV-400 (400
MHz) spectrometer at ambient temperature. Chemical shifts are
reported in ppm from the solvent resonance (DMSO-d6 2.50 ppm).
Data are reported as follows: chemical shift, multiplicity (s = singlet, d
= doublet, t = triplet, q = quartet, dd = doublet of doublets, br = broad,
m = multiplet), coupling constants, and number of protons.
3-Fluoro-2-hydrazinyl-5-phenylpyridine (6). (i) A sealable vial
was charged with 5-chloro-2,3-difluoropyridine (1.01 g, 6.76 mmol),
X-Phos (0.32 g, 0.68 mmol), palladium(II) acetate (76 mg, 0.34
mmol), potassium phosphate (4.31 g, 20.3 mmol), and phenylboronic
acid (1.00 g, 8.11 mmol). The vial was sealed, and dioxane (27 mL)
was added. The mixture was degassed by bubbling with N2 and then
heated at 100 °C for 30 min. The mixture was concentrated and
purified by MPLC using a gradient of 0−40% EtOAc in hexanes to
afford 2,3-difluoro-5-phenylpyridine (1.21 g, 6.33 mmol, 94% yield) as
a white solid.
(ii) A sealable pressure vessel was charged with 2,3-difluoro-5-
phenylpyridine (1.20 g, 6.28 mmol) and isopropanol (7 mL) followed
by hydrazine (1.20 mL, 38.2 mmol). The vessel was sealed and heated
at 70 °C for 1.5 h until a gray precipitate formed. The precipitate was
collected by vacuum filtration, rinsed with isopropanol (20 mL), and
dried under positive air flow to yield 3-fluoro-2-hydrazinyl-5-
phenylpyridine (6) (1.34 g, 6.59 mmol, 86% yield) as a light gray
solid. 1H NMR (400 MHz, DMSO-d6) δ ppm 8.26 (dd, J = 1.81, 1.12
Hz, 1 H), 7.94 (s, 1 H), 7.73 (dd, J = 12.91, 1.96 Hz, 1 H), 7.60−7.67
(m, 2 H), 7.38−7.47 (m, 2 H), 7.32 (d, J = 7.34 Hz, 1 H), 4.21 (br s, 2
H).
Figure 5. A single dose of compound (S)-21f was administered by oral
gavage (formulation: 2% (w/v) hydroxypropylmethylcellulose, 1% (w/
v) Tween 80 in water, pH adjusted to 2.2 with MSA) at 1, 3, 6, 9, 12,
or 24 h prior to sacrifice. c-Met phosphorylation was induced in the
livers of female Balb/c mice by injection of 12 μg of human
recombinant HGF IV 5 min prior to sacrifice. Levels of c-Met
phosphorylation were determined by an electrochemiluminescent
immunoassay. Data represent the mean standard deviation (n = 3).
Statistical significance was determined by ANOVA followed by
Bonferroni/Dunn post hoc test: (∗) p < 0.0001 compared to HGF
control. Black circles represent the mean terminal concentration of
compound (S)-21f in the plasma standard deviation (n = 3).
2424
DOI: 10.1021/jm501913a
J. Med. Chem. 2015, 58, 2417−2430

Journal of Medicinal Chemistry
Article
tert-Butyl 2,2-Difluoro-2-(quinolin-6-yl)acetate (8). (iii) Zn
dust (26.5 g, 0.42 mol) was suspended in 300 mL of anhydrous THF
under argon. Trimethylsilyl chloride (1.4 mL, 12.0 mmol) was added
dropwise via syringe, and the solution was refluxed until a precipitate
was observed. The mixture was allowed to cool to rt, and tert-butyl
bromoacetate (56 mL, 0.37 mol) was added dropwise via syringe. The
mixture was cooled to 0 °C and a solution of 6-bromoquinoline (25.0
g, 0.12 mol) in 25 mL of THF was added dropwise via syringe
followed by addition of tetrakis(triphenylphosphine)palladium (1.4 g,
1.2 mmol) as a solid in a single portion under positive argon flow. The
mixture was heated to reflux for 3 h. The reaction mixture was cooled
to rt, and saturated aqueous ammonium chloride was added (100 mL).
The mixture was filtered through Celite, partitioned between water
(50 mL) and ethyl acetate (300 mL), and the layers were separated.
The aqueous layer was extracted with ethyl acetate (2 × 100 mL). The
combined organic layers were dried (Na2SO4) and concentrated in
vacuo. The crude residue was purified by MPLC, eluting with 10%
EtOAc in hexanes to obtain tert-butyl 2-(quinolin-6-yl)acetate (16.0 g,
trile (0.54 mL, 5.43 mmol) and diisopropylethylamine (0.95 mL, 5.43
mmol). The mixture was heated with microwave irradiation at 100 °C
for 20 min. The mixture was filtered to remove PS-PPh3, washing with
MeOH (10 mL) and CH2Cl2 (20 mL). The collected mother liquor
was concentrated and purified by reverse-phase HPLC (10−90%
CH3CN in H2O (0.1% TFA additive)) to provide 6-(difluoro(8-
fluoro-6-phenyl[1,2,4]triazolo[4,3-a]pyridin-3-yl)methyl)quinoline
2,2,2-trifluoroacetate. The product was free-based using an SCX-2 ion
exchange column (Biotage), eluting with 7 N ammonia in MeOH (100
mL). The eluent was concentrated to give 6-(difluoro(8-fluoro-6-
phenyl[1,2,4]triazolo[4,3-a]pyridin-3-yl)methyl)quinoline (103 mg,
1
0.26 mmol, 10 % yield) as an off-white solid. H NMR (400 MHz,
DMSO-d6) δ ppm 9.06 (dd, J = 4.21, 1.76 Hz, 1 H), 8.56−8.59 (m, 2
H), 8.50 (d, J = 0.98 Hz, 1 H), 8.23 (d, J = 8.90 Hz, 1 H), 8.00−8.09
(m, 2 H), 7.92 (s, 1 H), 7.75−7.81 (m, 2 H), 7.65−7.71 (m, 1 H),
7.48−7.58 (m, 2 H). LRMS (ESI): m/z (M + H) 391.2.
6-(Difluoro(8-fluoro-6-(pyridin-2-yl)[1,2,4]triazolo[4,3-a]-
pyridin-3-yl)methyl)quinoline (10a). Prepared from 1-(3-fluoro-5-
(pyridin-2-yl)pyridin-2-yl)hydrazine (prepared as described for 6) and
2,2-difluoro-2-(quinolin-6-yl)acetic acid in 29% yield (over 2 steps)
1
65.7 mmol, 55% yield) as an orange waxy solid. H NMR (400 MHz,
CD3OD), δ ppm 8.88−8.67 (m, 1H), 8.34−8.32 (d, 1H, J = 8.0 Hz),
7.99−7.96 (d, 1H, J = 8.8 Hz), 7.83 (s, 1H), 7.67−7.64 (m, 1H),
7.54−7.50 (m, 1H), 3.78 (s, 2H), 1.41 (s, 9H).
1
according to the general procedure described for 4. Data for 10a: H
NMR (400 MHz, DMSO-d6) δ ppm 9.11 (d, J = 0.59 Hz, 1 H), 9.06
(dd, J = 4.21, 1.76 Hz, 1 H), 8.73 (dt, J = 3.86, 0.86 Hz, 1 H), 8.55−
8.62 (m, 1 H), 8.49 (s, 1 H), 8.27−8.35 (m, 1 H), 8.24 (d, J = 8.80 Hz,
1 H), 8.17 (d, J = 8.02 Hz, 1 H), 8.06 (dd, J = 8.90, 2.15 Hz, 1 H), 7.99
(td, J = 7.80, 1.81 Hz, 1 H), 7.68 (dd, J = 8.31, 4.21 Hz, 1 H), 7.49
(ddd, J = 7.51, 4.77, 0.83 Hz, 1 H). LCMS (ESI): m/z (M + H) =
392.4.
5-(3-(Difluoro(quinolin-6-yl)methyl)-8-fluoro[1,2,4]triazolo-
[4,3-a]pyridin-6-yl)-3-methylisoxazole (10b). Prepared from 1-(3-
fluoro-5-(3-methylisoxazol-5-yl)pyridin-2-yl)hydrazine (18a) (pre-
pared via Stille coupling route) and 2,2-difluoro-2-(quinolin-6-yl)acetic
acid in 27% yield (over 2 steps) according to the general procedure
described for 4. Data for 10b: 1H NMR (400 MHz, DMSO-d6) δ ppm
9.05 (dd, J = 4.21, 1.56 Hz, 1 H), 8.75 (s, 1 H), 8.57 (d, J = 0.69 Hz, 1
H), 8.47 (s, 1 H), 8.22 (d, J = 8.80 Hz, 1 H), 7.96−8.10 (m, 2 H), 7.66
(dd, J = 8.31, 4.21 Hz, 1 H), 7.20 (s, 1 H), 2.31 (s, 3 H). LCMS (ESI):
m/z (M + H) = 396.3.
(iv) A solution of tert-butyl 2-(quinolin-6-yl)acetate (1.63 g, 6.70
mmol) and N-fluorobenzenesulfonimide (6.34 g, 20.1 mmol) in THF
(34 mL) was cooled to −78 °C. A THF solution of LiHMDS (23.5
mL, 23.4 mmol) was added, and the solution was maintained at −78
°C for 2 h, then allowed to warm to rt. Saturated aqueous NH4Cl (10
mL) was added, and the mixture was partitioned between EtOAc (30
mL) and additional saturated aqueous NH4Cl (10 mL). The layers
were separated and the aqueous layer was extracted with EtOAc (1 ×
25 mL). The combined organic extracts were dried (Na2SO4),
concentrated in vacuo, and purified by MPLC using a gradient of 2−
90% EtOAc in hexanes to afford tert-butyl 2,2-difluoro-2-(quinolin-6-
yl)acetate (8) (0.87 g, 3.13 mmol, 47% yield) as a light-yellow oil,
which solidified upon standing. 1H NMR (400 MHz, CD3OD), δ ppm
8.97 (d, J = 4.11 Hz, 1 H), 8.51 (d, J = 8.22 Hz, 1 H), 7.88−7.96 (m, 1
H), 8.12−8.29 (m, 2 H), 7.64 (dd, J = 8.31, 4.30 Hz, 1 H), 1.47 (s, 9
H).
5-(3-(Difluoro(quinolin-6-yl)methyl)-8-fluoro[1,2,4]triazolo-
[4,3-a]pyridin-6-yl)-2-methylthiazole (10c). Prepared from 1-(3-
fluoro-5-(2-methylthiazol-5-yl)pyridin-2-yl)hydrazine (prepared by
Stille method described for intermediate 18a) and 2,2-difluoro-2-
(quinolin-6-yl)acetic acid in 3% yield (over 2 steps) according to the
general procedure described for 4. Data for 10c: 1H NMR (400 MHz,
DMSO-d6) δ ppm 9.07 (dd, J = 4.21, 1.66 Hz, 1 H), 8.55−8.65 (m, 1
H), 8.49 (s, 1 H), 8.45 (d, J = 0.98 Hz, 1 H), 8.29 (s, 1 H), 8.23 (d, J =
8.80 Hz, 1 H), 7.99−8.08 (m, 2 H), 7.69 (dd, J = 8.27, 4.25 Hz, 1 H),
2.71 (s, 3 H). LCMS (ESI): m/z (M + H) = 412.2.
4-(3-(Difluoro(quinolin-6-yl)methyl)-8-fluoro[1,2,4]triazolo-
[4,3-a]pyridin-6-yl)thiazole (10d). Prepared from 1-(3-fluoro-5-
(thiazol-4-yl)pyridin-2-yl)hydrazine (prepared by Stille method
described for intermediate 18a) and 2,2-difluoro-2-(quinolin-6-yl)-
acetic acid in 20% yield (over 2 steps) according to the general
procedure described for 4. Data for 10d: 1H NMR (400 MHz, DMSO-
d6) δ ppm 9.24−9.36 (m, 2 H), 8.93−9.07 (m, 2 H), 8.70 (s, 1 H),
8.57 (d, J = 1.37 Hz, 1 H), 8.45 (d, J = 8.90 Hz, 1 H), 8.20−8.34 (m, 2
H), 7.96 (dd, J = 8.22, 4.69 Hz, 1 H). LCMS (ESI): m/z (M + H) =
398.1.
6-(Difluoro(8-fluoro-6-(1-methyl-1H-pyrazol-4-yl)[1,2,4]-
triazolo[4,3-a]pyridin-3-yl)methyl)quinoline (10e). Prepared
from 1-(3-fluoro-5-(1-methyl-1H-pyrazol-4-yl)pyridin-2-yl)hydrazine
(prepared as described for 6) and 2,2-difluoro-2-(quinolin-6-yl)acetic
acid in 10% yield (over 2 steps) according to the general procedure
described for 4. Data for 10e: 1H NMR (400 MHz, DMSO-d6) δ ppm
9.06 (dd, J = 4.21, 1.66 Hz, 1 H), 8.55−8.62 (m, 2 H), 8.43−8.52 (m,
2 H), 8.23 (d, J = 8.80 Hz, 1 H), 8.12 (s, 1 H), 8.04 (dd, J = 8.85, 2.10
Hz, 1 H), 7.92 (dd, J = 12.03, 0.88 Hz, 1 H), 7.68 (dd, J = 8.31, 4.30
Hz, 1 H), 3.89 (s, 3 H). LCMS (ESI): m/z (M + H) = 395.4.
6-((8-Fluoro-6-(1-methyl-1H-pyrazol-4-yl)[1,2,4]triazolo[4,3-
a]pyridin-3-yl)methyl)quinoline (11a). Prepared from 2-(quinolin-
2,2-Difluoro-N′-(3-fluoro-5-phenylpyridin-2-yl)-2-(quinolin-
6-yl)acetohydrazide (9). (v) To a solution of tert-butyl 2,2-difluoro-
2-(quinolin-6-yl)acetate (8) (0.87 g, 3.13 mmol) in CH2Cl2 (6.0 mL)
was added trifluoroacetic acid (1.2 mL, 15.6 mmol). The solution was
maintained at rt for 48 h. The solution concentrated and dried under
reduced pressure to provide 2,2-difluoro-2-(quinolin-6-yl)acetic acid
(65 mg, 93% yield) as a sticky brown solid, which was used without
further purification.
(vi) To a solution of 2,2-difluoro-2-(quinolin-6-yl)acetic acid (65
mg, 2.91 mmol) in DMF (6.3 mL) at 0 °C was added thionyl chloride
(0.5 mL, 6.3 mmol) dropwise via syringe. The solution was maintained
at rt for 1 h, and then triethylamine (0.5 mL, 3.14 mmol) was added
followed by 3-fluoro-2-hydrazinyl-5-phenylpyridine (0.64 mg, 3.14
mmol) as a solid in a single portion, then dimethylaminopyridine (38
mg, 0.31 mmol). The solution was allowed to warm to rt and was
maintained at rt for 18 h. Saturated aqueous NaHCO3 was added
carefully to the solution over 20 min. The mixture was extracted with
EtOAc (3 × 20 mL). The combined organic layers were washed
vigorously with 1 N NaOH (15 mL). The organic layer was dried
(Na2SO4), concentrated in vacuo, and purified by MPLC using a
gradient of 2−10% MeOH in CH2Cl2 to afford 2,2-difluoro-N′-(3-
fluoro-5-phenylpyridin-2-yl)-2-(quinolin-6-yl)acetohydrazide (9) (1.11
g, 2.72 mmol, 87% yield) as a brown solid.
General Procedure for the Preparation of Compounds 3b, 4,
10, and 12. See Scheme 1, steps v−vi, exemplified by the synthesis of
4.
6-(Difluoro(8-fluoro-6-phenyl[1,2,4]triazolo[4,3-a]pyridin-3-
yl)methyl)quinoline (4). (vi) A microwave vial was charged with 2,2-
difluoro-N′-(3-fluoro-5-phenylpyridin-2-yl)-2-(quinolin-6-yl)-
acetohydrazide (1.11 g, 2.72 mmol) and PS-triphenylphosphine (PS-
PPh3) (3.04 g, 6.79 mmol). The vial was sealed with a septum cap, and
dichloroethane (5 mL) was added followed by 2,2,2-trichloroacetoni-
2425
DOI: 10.1021/jm501913a
J. Med. Chem. 2015, 58, 2417−2430

Journal of Medicinal Chemistry
Article
6-yl)acetic acid hydrochloride and 1-(3-fluoro-5-(1-methyl-1H-pyr-
azol-4-yl)pyridin-2-yl)hydrazine (prepared as described for 6) in 63%
yield (over 2 steps) according to the general procedure described for
4-(3-(Difluoro(3-methoxyquinolin-6-yl)methyl)-8-fluoro-
[1,2,4]triazolo[4,3-a]pyridin-6-yl)thiazole (12b). Prepared from
1-(3-fluoro-5-(thiazol-4-yl)pyridin-2-yl)hydrazine (prepared via Stille
coupling route) and 2,2-difluoro-2-(3-methoxyquinolin-6-yl)acetic acid
in 7% yield (over 2 steps) according to the general procedure
described for 4. Data for 12b: 1H NMR (400 MHz, DMSO-d6) δ ppm
9.29 (s, 1H), 8.94 (s, 1 H), 8.79 (s, 1 H), 8.53 (s, 1 H), 8.22−8.35 (m,
2 H), 8.17 (d, J = 7.6, 1 H), 7.96 (s, 1H), 7.87 (d, J = 7.2, 1 H), 3.94
(s, 3 H). LCMS (ESI): m/z (M + H) = 428.1.
6-(Difluoro(8-fluoro-6-(1-methyl-1H-pyrazol-4-yl)[1,2,4]-
triazolo[4,3-a]pyridin-3-yl)methyl)-3-methoxyquinoline (12c).
Prepared from 1-(3-fluoro-5-(1-methyl-1H-pyrazol-4-yl)pyridin-2-yl)-
hydrazine (prepared as described for 6) and 2,2-difluoro-2-(3-
methoxyquinolin-6-yl)acetic acid in 15% yield (over 2 steps) according
to the general procedure described for 4. Data for 12c: 1H NMR (400
MHz, DMSO-d6) δ ppm 8.78 (d, J = 2.93 Hz, 1 H), 8.57 (d, J = 0.78
Hz, 1 H), 8.45 (s, 1 H), 8.32 (d, J = 1.47 Hz, 1 H), 8.15 (d, J = 8.71
Hz, 1 H), 8.12 (d, J = 0.68 Hz, 1 H), 7.96 (d, J = 2.74 Hz, 1 H), 7.92
(dd, J = 12.03, 0.98 Hz, 1 H), 7.87 (dd, J = 8.80, 2.15 Hz, 1 H), 3.94
(s, 3 H), 3.89 (s, 3 H). LCMS (ESI): m/z (M + H) = 425.4.
General Procedure for the Preparation of Compounds 11,
20−24. See Scheme 2, steps i−v, exemplified by the synthesis of 21e
(R = isoxazole, R1 = OMe, R2 = H).
1
(S)-21e. Data for 11a: H NMR (400 MHz, DMSO-d6) δ ppm 8.86
(dd, J = 4.21, 1.76 Hz, 1 H), 8.62 (d, J = 1.17 Hz, 1 H), 8.31 (ddd, J =
8.46, 1.61, 0.59 Hz, 1 H), 8.25 (d, J = 0.39 Hz, 1 H), 7.96−8.03 (m, 2
H), 7.90 (d, J = 1.66 Hz, 1 H), 7.77 (dd, J = 8.71, 2.05 Hz, 1H), 7.64
(dd, J = 12.28, 1.12 Hz, 1 H), 7.50 (dd, J = 8.31, 4.21 Hz, 1 H), 4.80
(s, 2 H), 3.82 (s, 3H). LCMS (ESI): m/z (M + H) = 359.4.
6-((8-Fluoro-6-(1-methyl-1H-pyrazol-4-yl)[1,2,4]triazolo[4,3-
a]pyridin-3-yl)methyl)-3-methoxyquinoline (11b). Prepared
from 2-(3-methoxyquinolin-6-yl)acetic acid and 1-(3-fluoro-5-(1-
methyl-1H-pyrazol-4-yl)pyridin-2-yl)hydrazine (prepared as described
for 6) in 41% yield (over 2 steps) according to the general procedure
described for (S)-21e. Data for 11b: 1H NMR (400 MHz, DMSO-d6)
δ ppm 8.55−8.66 (m, 2 H), 8.25 (s, 1 H), 8.00 (d, J = 0.68 Hz, 1 H),
7.93 (d, J = 8.61 Hz, 1 H), 7.68−7.75 (m, 2 H), 7.58−7.67 (m, 2 H),
4.79 (s, 2 H), 3.89 (s, 3 H), 3.87 (s, 3 H). LCMS (ESI): m/z (M + H)
= 389.0.
6-((8-Fluoro-6-(1-methyl-1H-pyrazol-4-yl)[1,2,4]triazolo[4,3-
a]pyridin-3-yl)methyl)quinolin-3-ol (11c). Prepared as a by-
1
product in the synthesis of 11b. Data for 11c: H NMR (400 MHz,
DMSO-d6) d ppm 10.59 (br s, 1 H), 8.57−8.66 (m, 2 H), 8.27 (s, 1
H), 8.01 (s, 1 H), 7.91 (d, J = 8.80 Hz, 1 H), 7.80 (s, 1 H), 7.65 (d, J =
12.32 Hz, 1 H), 7.61 (s, 1 H), 7.52−7.59 (m, 1 H), 4.76 (s, 2 H), 3.88
(s, 3 H). LCMS (ESI): m/z (M + H) = 375.1.
6-((8-Fluoro-6-(1-methyl-1H-pyrazol-4-yl)[1,2,4]triazolo[4,3-
a]pyridin-3-yl)methyl)quinoxaline (11d). Prepared from 1-(3-
fluoro-5-(1-methyl-1H-pyrazol-4-yl)pyridin-2-yl)hydrazine (prepared
as described for 6) and 2-(quinoxalin-6-yl)acetic acid hydrochloride
in 28% yield (over 2 steps) according to the general procedure
described for 4. Data for 11d: 1H NMR (400 MHz, DMSO-d6) δ ppm
8.92 (q, J = 1.76 Hz, 2 H), 8.70 (s, 1 H), 8.27 (s, 1 H), 8.05−8.13 (m,
2 H), 8.02 (s, 1 H), 7.87 (dd, J = 8.66, 1.91 Hz, 1 H), 7.65 (d, J =
12.23 Hz, 1 H), 4.88 (s, 2 H), 3.88 (s, 3 H). LCMS (ESI): m/z (M +
H) = 360.3.
N-(6-((8-Fluoro-6-(1-methyl-1H-pyrazol-4-yl)[1,2,4]triazolo-
[4,3-a]pyridin-3-yl)methyl)benzo[d]thiazol-2-yl)acetamide
(11e). Prepared from 2-(2-acetamidobenzo[d]thiazol-6-yl)acetic acid
hydrochloride and 1-(3-fluoro-5-(1-methyl-1H-pyrazol-4-yl)pyridin-2-
yl)hydrazine (prepared as described for 6) in 7% yield (over 2 steps)
according to the general procedure described for (S)-21e. Data for
11e: 1H NMR (400 MHz, DMSO-d6) δ ppm 12.29 (s, 1 H), 8.58 (d, J
= 1.08 Hz, 1 H), 8.26 (s, 1 H), 8.00 (d, J = 0.68 Hz, 1 H), 7.95 (d, J =
1.27 Hz, 1 H), 7.68 (d, J = 8.31 Hz, 1 H), 7.63 (dd, J = 12.28, 1.03 Hz,
1 H), 7.42 (dd, J = 8.36, 1.81 Hz, 1 H), 4.68 (s, 2 H), 3.88 (s, 3 H),
2.18 (s, 3 H). LCMS (ESI): m/z (M + H) = 422.0.
tert-Butyl 2-(3-Methoxyquinolin-6-yl)propanoate (15, R1
=
OMe).29 A sealable vial was charged with bis(benzylideneacetone)-
palladium (60.4 mg, 105 μmol), tri-tert-butylphosphonium tetrafluor-
oborate (30 mg, 105 μmol), sodium bis(trimethylsilyl)amide (1.73 g,
9.45 mmol), 6-bromo-3-methoxyquinoline (13, R1 = OMe) (0.50 g,
2.10 mmol). The vial was sealed with a septum-cap, flushed with N2,
and then tert-butyl propionate (14) (0.73 mL, 4.9 mmol) and toluene
(6 mL) were added. The mixture was sparged with N2 for 5 min and
then maintained at rt for 24 h. The mixture was poured into saturated
aqueous NH4Cl (15 mL) and extracted with EtOAc (2 × 10 mL). The
combined organic layers were dried (Na2SO4), concentrated, and
purified by MPLC using a gradient of 2−60% EtOAc in hexanes to
afford tert-butyl 2-(3-methoxyquinolin-6-yl)propanoate 15 (0.46 g,
76% yield) as a colorless oil. 1H NMR (400 MHz, CDCl3) δ ppm 8.62
(d, J = 2.84 Hz, 1 H), 7.94−8.02 (m, 1 H), 7.61 (d, J = 1.76 Hz, 1 H),
7.49 (dd, J = 8.71, 1.96 Hz, 1 H), 7.33 (d, J = 2.64 Hz, 1 H), 3.91 (s, 3
H), 3.73−3.85 (m, 1 H), 1.53 (d, J = 7.14 Hz, 3 H), 1.40 (s, 9 H).
LCMS (ESI): m/z (M + H) = 288.3.
tert-Butyl 2-Fluoro-2-(3-methoxyquinolin-6-yl)propanoate
(16, R1 = OMe). To a solution of tert-butyl 2-(3-methoxyquinolin-
6-yl)propanoate (2.50 g, 8.7 mmol) in THF (8.7 mL, 8.7 mmol) at
−78 °C was added a THF solution of LiHMDS (11.0 mL, 1 M). The
solution was warmed to room temperature over 1 h, then recooled to
−78 °C, and a THF solution of N-fluorobenzenesulfonimide (3.6 g, 11
mmol, 1 M) was added. The solution was allowed to warm to −10 °C
over 1 h. Upon completion, the solution was filtered through a plug of
Celite, washing with EtOAc (30 mL). The filtrate was concentrated,
then redissolved in EtOAc (50 mL) and washed with saturated
aqueous NH4Cl (20 mL). The layers were separated, and the aqueous
layer was extracted with EtOAc (2 × 10 mL). The combined organic
layers were dried (MgSO4), concentrated, and the residue was purified
by MPLC, eluting with 40% EtOAc:hexanes to provide tert-butyl 2-
fluoro-2-(3-methoxyquinolin-6-yl)propanoate (16) (2.03 g, 76% yield)
6-((8-Fluoro-6-(1-methyl-1H-pyrazol-4-yl)[1,2,4]triazolo[4,3-
a]pyridin-3-yl)methyl)benzo[d]thiazol-2-amine (11f). A solution
of sodium hydroxide (2.8 mL, 1 N) was added to 11e (40 mg, 0.09
mmol). The resulting mixture was heated at 60 °C for 24 h. The
mixture was cooled to rt and the precipitated solid was collected by
vacuum filtration and dried under high vacuum to yield 6-((8-fluoro-6-
(1-methyl-1H-pyrazol-4-yl)[1,2,4]triazolo[4,3-a]pyridin-3-yl)methyl)-
1
benzo[d]thiazol-2-amine (11f) (10 mg, 0.03 mmol, 33% yield). H
NMR (400 MHz, DMSO-d6) δ ppm 8.53 (s, 1 H), 8.26 (s, 1 H), 7.99
(s, 1 H), 7.58−7.68 (m, 2 H), 7.41 (s, 2 H), 7.24−7.32 (m, 1 H),
7.17−7.23 (m, 1 H), 4.58 (s, 2 H), 3.88 (s, 3 H). LCMS (ESI): m/z
(M + H) = 380.2.
1
as a yellow oil. H NMR (400 MHz, DMSO-d6) δ ppm 8.68 (d, J =
2.93 Hz, 1H), 7.97−8.05 (m, 2H), 7.89 (d, J = 2.93 Hz, 1H), 7.62 (dd,
J = 2.10, 8.75 Hz, 1H), 3.93 (s, 5H), 3.33 (s, 3H), 1.38 (s, 9H). LCMS
(ESI): m/z (M + H) = 306.4.
5-(3-(Difluoro(3-methoxyquinolin-6-yl)methyl)-8-fluoro-
[1,2,4]triazolo[4,3-a]pyridin-6-yl)-3-methylisoxazole (12a). Pre-
pared from 1-(3-fluoro-5-(3-methylisoxazol-5-yl)pyridin-2-yl)-
hydrazine (prepared via Stille coupling route as described for 18a)
and 2,2-difluoro-2-(3-methoxyquinolin-6-yl)acetic acid in 20% yield
(over 2 steps) according to the general procedure described for 4.
2-(3-Methoxyquinolin-6-yl)propanoic Acid (17, R1 = OMe, R2
= H). HCl gas was bubbled through an EtOAc solution of tert-butyl 2-
(3-methoxyquinolin-6-yl)propanoate (6.20 g, 21.6 mmol) for 5 min to
produce a white precipitate. The precipitate was collected by vacuum
filtration to give 2-(3-methoxyquinolin-6-yl)propanoic acid hydro-
1
Data for 12a: H NMR (400 MHz, CDCl3) δ ppm 8.75−8.84 (m, 2
1
chloride (5.48 g, 95% yield) as a colorless solid. H NMR (400 MHz,
H), 8.21 (d, J = 8.71 Hz, 1 H), 8.17 (d, J = 1.08 Hz, 1 H), 7.83 (dd, J =
8.80, 2.05 Hz, 1 H), 7.47 (d, J = 2.84 Hz, 1 H), 7.39 (dd, J = 9.78, 1.17
Hz, 1 H), 6.54 (s, 1 H), 3.97 (s, 3 H), 2.41 (s, 3 H). LCMS (ESI): m/z
(M + H) = 426.0.
DMSO-d6) δ ppm 8.77 (d, J = 2.25 Hz, 1H), 7.96−8.05 (m, 2H), 7.87
(s, 1H), 7.62 (dd, J = 1.56, 8.61 Hz, 1H), 3.94 (s, 3H), 1.47 (d, J =
7.14 Hz, 3H). LCMS (ESI): m/z (M + H) = 232.2.
2426
DOI: 10.1021/jm501913a
J. Med. Chem. 2015, 58, 2417−2430

Journal of Medicinal Chemistry
Article
1-(3-Fluoro-5-(3-methylisoxazol-5-yl)pyridin-2-yl)hydrazine
(18a, R = methylisoxazole). To a sealable pressure vessel charged
with 5-(5,6-difluoropyridin-3-yl)-3-methylisoxazole (26) (5.32 g, 27.1
mmol) and isopropanol (30 mL) was added anhydrous hydrazine (3.0
mL, 94.9 mmol). The vessel was sealed and heated at 60 °C for 18 h
until a white precipitate formed. The mixture was cooled to rt and
concentrated to dryness. The resulting solids were suspended in
saturated aqueous NaHCO3, then collected by vacuum filtration and
dried under reduced pressure to yield 1-(3-fluoro-5-(3-methylisoxazol-
5-yl)pyridin-2-yl)hydrazine (18a) (1.47 g, 77.8% yield) as a tan
6-(1-Fluoro-1-(8-fluoro-6-(1-methyl-1H-pyrazol-4-yl)[1,2,4]-
triazolo[4,3-a]pyridin-3-yl)ethyl)-3-methoxyquinoline (20d).
Prepared from 1-(3-fluoro-5-(1-methyl-1H-pyrazol-4-yl)pyridin-2-yl)-
hydrazine (prepared as described for 6) and 2-fluoro-2-(3-methox-
yquinolin-6-yl)propanoic acid hydrochloride in 57% yield (over 2
steps) according to the general procedure described for 21e.
Separation of enantiomers provided 22% yield of (R)-20d (>99%
ee) and 26% yield of (S)-20d (>99%ee). Data for (R)-20d: 1H NMR
(400 MHz, DMSO-d6) δ ppm 8.66 (d, J = 2.93 Hz, 1H), 8.16 (s, 1H),
8.02 (d, J = 8.80 Hz, 1H), 7.91 (s, 1H), 7.88 (d, J = 1.86 Hz, 1H), 7.82
(d, J = 2.84 Hz, 1H), 7.78 (d, J = 0.68 Hz, 1H), 7.72 (d, J = 0.88 Hz,
1H), 7.62 (dd, J = 2.10, 8.85 Hz, 1H), 3.88 (s, 3H), 3.80 (s, 3H),
2.35−2.45 (m, 3H). LCMS (ESI): m/z (M + H) = 421.2.
1
amorphous solid. H NMR (400 MHz, DMSO-d6) δ ppm 8.42 (br s,
1H), 8.35−8.38 (m, 1H), 7.77 (d, J = 1.86 Hz, 1H), 7.74 (s, 1H), 6.70
(s, 1H), 4.33 (br s, 2H), 2.25 (s, 3H).
N′-(3-Fluoro-5-(3-methylisoxazol-5-yl)pyridin-2-yl)-2-(3-me-
thoxyquinolin-6-yl)propanehydrazide (19, R = Methylisoxa-
zole, R1 = OMe, R2 = H). To a flask charged with 1-(3-fluoro-5-(3-
methylisoxazol-5-yl)pyridin-2-yl)hydrazine (18a) (2.63 g, 12.6 mmol),
2-(3-methoxyquinolin-6-yl)propanoic acid hydrochloride (17) (3.25 g,
12.1 mmol), O-(7-azabenzotriazol-1-yl)-N,N,N′,N′-tetramethyluro-
nium hexafluorophosphate (HATU) (7.20 g, 18.9 mmol), and
diisopropylethylamine (6.6 mL, 37.9 mmol) was added DMF (45
mL). The resulting solution was maintained at rt for 2 h, then
concentrated to 50% volume, which resulted in the formation of a
precipitate. The solid was collected by vacuum filtration and washed
with EtOAc to give N′-(3-fluoro-5-(3-methylisoxazol-5-yl)pyridin-2-
yl)-2-(3-methoxyquinolin-6-yl)propanehydrazide (19) (4.71 g, 89%
5-(8-Fluoro-3-(1-(quinolin-6-yl)ethyl)[1,2,4]triazolo[4,3-a]-
pyridin-6-yl)-3-methylisothiazole (21a). Prepared from 1-(3-
fluoro-5-(3-methylisothiazol-5-yl)pyridin-2-yl)hydrazine (prepared by
Stille method as described for 18a) and 2-(quinolin-6-yl)propanoic
acid hydrochloride in 52% yield (over 2 steps) according to the general
procedure described for 21e. Separation of enantiomers provided 19%
yield of (S)-21a (>99% ee) and 19% yield of (R)-21a (>99% ee). Data
1
for (S)-21a: H NMR (400 MHz, DMSO-d6) δ ppm 8.86 (dd, J =
1.71, 4.16 Hz, 1H), 8.64 (d, J = 1.17 Hz, 1H), 8.32 (d, J = 7.43 Hz,
1H), 8.00 (d, J = 8.61 Hz, 1H), 7.94 (d, J = 1.96 Hz, 1H), 7.81 (dd, J =
2.01, 8.75 Hz, 1H), 7.69 (dd, J = 1.17, 11.74 Hz, 1H), 7.67 (s, 1H),
7.51 (dd, J = 4.16, 8.26 Hz, 1H), 5.24 (d, J = 7.14 Hz, 1H), 2.44 (s,
3H), 1.90 (d, J = 7.04 Hz, 3H). LCMS (ESI): m/z (M + H) = 390.0.
5-(8-Fluoro-3-(1-(3-methoxyquinolin-6-yl)ethyl)[1,2,4]-
triazolo[4,3-a]pyridin-6-yl)-3-methylisothiazole (21b). Prepared
from 1-(3-fluoro-5-(3-methylisothiazol-5-yl)pyridin-2-yl)hydrazine
(prepared by Stille method as described for 18a) and 2-(3-
methoxyquinolin-6-yl)propanoic acid hydrochloride in 35% yield
(over 2 steps) according to the general procedure described for 21e.
Separation of enantiomers provided 10% yield of (S)-21b (>99% ee)
and 8% yield of (R)-21b (>99%ee). Data for (S)-21b: 1H NMR (400
MHz, DMSO-d6) δ ppm 8.55−8.61 (m, 2H), 7.93 (d, J = 8.61 Hz,
1H), 7.75 (d, J = 1.76 Hz, 1H), 7.66−7.70 (m, 2H), 7.62−7.66 (m,
2H), 5.21 (d, J = 7.04 Hz, 1H), 3.89 (s, 3H), 2.44 (s, 3H), 1.90 (d, J =
7.14 Hz, 3H). LCMS (ESI): m/z (M + H) = 420.2.
1
yield) as a gray-white solid. H NMR (400 MHz, DMSO-d6) δ ppm
10.17 (d, J = 1.66 Hz, 1H), 9.16 (s, 1H), 8.60 (d, J = 2.93 Hz, 1H),
8.35 (d, J = 1.56 Hz, 1H), 7.89−7.95 (m, 2H), 7.86 (d, J = 1.86 Hz,
1H), 7.74 (d, J = 2.35 Hz, 1H), 7.65 (dd, J = 2.05, 8.71 Hz, 1H), 6.78
(s, 1H), 3.94−4.01 (m, 1H), 3.93 (s, 3H), 2.26 (s, 3H), 1.50 (d, J =
7.14 Hz, 3H). LCMS (ESI): m/z (M + H) = 422.0.
5-(8-Fluoro-3-(1-fluoro-1-(quinolin-6-yl)ethyl)[1,2,4]triazolo-
[4,3-a]pyridin-6-yl)-3-methylisothiazole (20a). Prepared from 1-
(3-fluoro-5-(3-methylisothiazol-5-yl)pyridin-2-yl)hydrazine (prepared
by Stille method as described for 18a) and 2-fluoro-2-(quinolin-6-
yl)propanoic acid hydrochloride in 59% yield (over 2 steps) according
to the general procedure described for 21e. Separation of enantiomers
provided 15% yield of (R)-20a (>99% ee) and 11% yield of (S)-20a
(>99%ee). Data for (R)-20a: 1H NMR (400 MHz, DMSO-d6) δ ppm
8.95 (dd, J = 1.76, 4.21 Hz, 1H), 8.45 (td, J = 0.90, 7.58 Hz, 1H),
8.05−8.16 (m, 2H), 8.02 (d, J = 0.88 Hz, 1H), 7.71−7.85 (m, 2H),
7.52−7.65 (m, 2H), 2.41−2.48 (m, 3H), 2.40 (s, 3H). LCMS (ESI):
6-(1-(8-Fluoro-6-(1-methyl-1H-imidazol-4-yl)[1,2,4]triazolo-
[4,3-a]pyridin-3-yl)ethyl)-3-methoxyquinoline (21c). Prepared
from 1-(3-fluoro-5-(1-methyl-1H-imidazol-4-yl)pyridin-2-yl)hydrazine
and 2-(3-methoxyquinolin-6-yl)propanoic acid hydrochloride in 34%
yield (over 2 steps) according to the general procedure described for
21e. Separation of enantiomers provided 10% yield of (S)-21c (>99%
1
m/z (M + H) = 408.0. As anticipated, the H NMR and LCMS data
1
for (S)-20a were identical to those obtained for (R)-20a.
ee) and 8% yield of (R)-21c (>99%ee). Data for (S)-21c: H NMR
5-(8-Fluoro-3-(1-fluoro-1-(3-methoxyquinolin-6-yl)ethyl)-
[1,2,4]triazolo[4,3-a]pyridin-6-yl)-3-methylisothiazole (20b).
Prepared from 1-(3-fluoro-5-(3-methylisothiazol-5-yl)pyridin-2-yl)-
hydrazine (prepared by Stille method as described for 18a) and 2-
fluoro-2-(3-methoxyquinolin-6-yl)propanoic acid hydrochloride in
70% yield (over 2 steps) according to the general procedure described
for 21e. Separation of enantiomers provided 32% yield of (R)-20b
(>99% ee) and 39% yield of (S)-20b (>99%ee). Data for (R)-20b: 1H
NMR (400 MHz, DMSO-d6) δ ppm 8.67 (d, J = 2.93 Hz, 1H), 8.03
(d, J = 8.80 Hz, 1H), 7.98 (s, 1H), 7.91 (d, J = 1.96 Hz, 1H), 7.83 (d, J
= 2.64 Hz, 1H), 7.80 (dd, J = 1.17, 11.44 Hz, 1H), 7.60 (dd, J = 2.15,
8.80 Hz, 1H), 7.56 (s, 1H), 3.89 (s, 3H), 2.37−2.47 (m, 3H), 2.39 (s,
3H). LCMS (ESI): m/z (M + H) = 438.3.
5-(8-Fluoro-3-(1-fluoro-1-(3-methoxyquinolin-6-yl)ethyl)-
[1,2,4]triazolo[4,3-a]pyridin-6-yl)-3-methylisoxazole (20c). Pre-
pared from 1-(3-fluoro-5-(3-methylisoxazol-5-yl)pyridin-2-yl)-
hydrazine (18a) and 2-fluoro-2-(3-methoxyquinolin-6-yl)propanoic
acid hydrochloride in 25% yield (over 2 steps) according to the general
procedure described for 21e. Separation of enantiomers provided 10%
yield of (R)-20c (>99% ee) and 8% yield of (S)-20c (>99%ee). Data
for (R)-20c: 1H NMR (400 MHz, DMSO-d6) δ ppm 8.67 (d, J = 2.93
Hz, 1H), 8.12 (s, 1H), 8.04 (d, J = 8.80 Hz, 1H), 7.92 (d, J = 11.44 Hz,
1H), 7.85 (d, J = 1.96 Hz, 1H), 7.82 (d, J = 2.84 Hz, 1H), 7.61 (dd, J =
2.10, 8.85 Hz, 1H), 7.02 (s, 1H), 3.88 (s, 3H), 2.35−2.46 (m, 3H),
2.23 (s, 3H). LCMS (ESI): m/z (M + H) = 422.2.
(400 MHz, DMSO-d6) δ ppm 8.57 (d, J = 2.93 Hz, 1H), 8.23 (d, J =
0.98 Hz, 1H), 7.93 (d, J = 8.61 Hz, 1H), 7.67−7.70 (m, 2H), 7.61−
7.66 (m, 3H), 7.58 (dd, J = 2.05, 8.61 Hz, 1H), 5.09 (d, J = 7.14 Hz,
1H), 3.87 (s, 3H), 3.65 (s, 3H), 1.89 (d, J = 7.04 Hz, 3H). LCMS
(ESI): m/z (M + H) = 403.2.
6-(1-(8-Fluoro-6-(1-methyl-1H-pyrazol-4-yl)[1,2,4]triazolo-
[4,3-a]pyridin-3-yl)ethyl)-3-methoxyquinoline (21d). Prepared
from 1-(3-fluoro-5-(1-methyl-1H-pyrazol-4-yl)pyridin-2-yl)hydrazine
(prepared as described for 6) and 2-(3-methoxyquinolin-6-yl)-
propanoic acid hydrochloride in 40% yield (over 2 steps) according
to the general procedure described for 21e. Separation of enantiomers
provided 18% yield of (S)-21d (>99% ee) and 21% yield of (R)-21d
(>99% ee). Data for (S)-21d: 1H NMR (400 MHz, DMSO-d6) δ ppm
8.57 (d, J = 2.84 Hz, 1H), 8.39 (d, J = 1.08 Hz, 1H), 8.20 (s, 1H),
7.88−7.97 (m, 2H), 7.74 (d, J = 1.76 Hz, 1H), 7.69 (d, J = 2.74 Hz,
1H), 7.65 (dd, J = 2.05, 8.71 Hz, 1H), 7.61 (dd, J = 1.03, 12.28 Hz,
1H), 5.00−5.18 (m, 1H), 3.88 (s, 3H), 3.85 (s, 3H), 1.90 (d, J = 7.04
Hz, 3H). LCMS (ESI): m/z (M + H) = 403.2.
5-(8-Fluoro-3-(1-(3-methoxyquinolin-6-yl)ethyl)[1,2,4]-
triazolo[4,3-a]pyridin-6-yl)-3-methylisoxazole (21e). To a THF
solution (120 mL) of N′-(3-fluoro-5-(3-methylisoxazol-5-yl)pyridin-2-
yl)-2-(3-methoxyquinolin-6-yl)propanehydrazide (19) (4.71 g, 11.2
mmol), and triphenylphosphine (4.40 g, 16.8 mmol) was added
trimethylsilylazide (2.2 mL, 16.8 mmol) via syringe followed by
dropwise addition of diethyl azodicarboxylate (DEAD) (2.6 mL, 16.8
2427
DOI: 10.1021/jm501913a
J. Med. Chem. 2015, 58, 2417−2430

Journal of Medicinal Chemistry
Article
mmol). The solution was maintained at rt for 30 min. The solution
was concentrated and the residue was purified by MPLC, eluting with
50−80% EtOAc in CH2Cl2 to afford 5-(8-fluoro-3-(1-(3-methoxyqui-
nolin-6-yl)ethyl)[1,2,4]triazolo[4,3-a]pyridin-6-yl)-3-methylisoxazole
(3.21 g, 7.95 mmol, 71% yield) as a racemic white solid. The racemic
product 21e was further purified by superfluid chromatography (SFC)
by repeated 1 mL injections of a 25 mg/mL solution of 21e onto a
Chiralpak AS-H, 2 cm × 25 cm (i.d. × length) column, eluting with
20% methanol/0.2% diethylamine/80% CO2 to provide enantioen-
riched products (R)-21e (1.31, 29% yield, 97% ee) and (S)-21e (1.39
g, 31% yield, >99% ee). Data for (R)-21e: As anticipated, the 1H NMR
and LCMS data for (R)-21e were identical to those obtained for (S)-
noic acid hydrochloride in 25% yield (over 2 steps) according to the
general procedure described for (21e).
5-(5,6-Difluoropyridin-3-yl)-3-methylisoxazole (26). A seal-
able vessel was charged with 5-chloro-2,3-difluoropyridine (5) (0.35
mL, 3.34 mmol), X-Phos (0.22 g, 0.47 mmol), and palladium(II)
acetate (53 mg, 0.23 mmol). Dioxane (16 mL) was added followed by
3-methyl-5-(tributylstannyl)isoxazole (27) (1.87 g, 5.02 mmol). The
mixture was purged with Ar, sealed and heated at 100 °C for 3 h. The
solution was concentrated in vacuo and the residue purified by MPLC,
eluting with 25% EtOAc in hexanes to provide the desired product,
which was still contaminated with some residual tin salts. The product
was triturated with hexanes (3 × 20 mL) to provide 5-(5,6-
difluoropyridin-3-yl)-3-methylisoxazole (26) (0.348 g, 53.1% yield)
1
21e. Data for (S)-21e: H NMR (400 MHz, DMSO-d6) δ ppm 8.65
1
as a white solid. H NMR (400 MHz, DMSO-d6) δ ppm d 8.42−8.69
(d, J = 1.08 Hz, 1H), 8.58 (d, J = 2.93 Hz, 1H), 7.93 (d, J = 8.61 Hz,
1H), 7.77 (dd, J = 1.12, 11.59 Hz, 1H), 7.72 (d, J = 1.86 Hz, 1H), 7.69
(d, J = 2.35 Hz, 1H), 7.63 (dd, J = 2.05, 8.71 Hz, 1H), 7.00 (s, 1H),
5.23 (d, J = 7.04 Hz, 1H), 3.88 (s, 3H), 2.27 (s, 3H), 1.89 (d, J = 7.04
Hz, 3H). LCMS (ESI): m/z (M + H) = 404.2.
(m, 2H), 7.08 (s, 1H), 2.31 (s, 3H).
3-Methyl-5-(tributylstannyl)isoxazole (27). To a solution of
tributyl(ethynyl)tin (55.0 g, 174.6 mmol) in benzene (200 mL) was
added nitroethane (13.8 mL, 192.0 mmol), phenyl isocyanate (38.0
mL, 349.1 mmol), and triethylamine (1.4 mL, 9.78 mmol). A reflux
condenser was attached, and the solution was heated at 50 °C for 18 h.
This resulted in a yellow solution containing a white precipitate. The
mixture was diluted with hexanes (200 mL) and then filtered to
remove the solid. The filter cake was washed with additional hexanes
(500 mL), and these rinses were combined with the mother liquor.
The combined solution was washed with saturated aqueous K2CO3 (3
× 100 mL) and brine (1 × 100 mL), then dried (Na2SO4) and
concentrated. The residue was dissolved in hexanes and passed
through a plug of silica, eluting with hexanes (500 mL), then 5%
EtOAc in hexanes (∼400 mL). The resulting solution was then
concentrated and subjected to high vacuum for 18 h to provide 3-
methyl-5-(tributylstannyl)isoxazole (27) (64.32 g, 99.02% yield) as a
yellow oil. 1H NMR (400 MHz, CDCl3) δ ppm 6.20 (s, 1H), 2.32 (s,
3H), 1.45−1.70 (m, 6H), 1.34 (t, J = 7.53 Hz, 6H), 1.15 (d, J = 8.22
Hz, 6H), 0.89 (t, J = 7.29 Hz, 9H).
Biology. Kinase Assay. IC50 measurements of inhibitor activity
against the recombinant c-Met kinase domain were determined using
homogeneous time-resolved fluorescence using a gastrin peptide as
substrate. Each reaction consists of 10 μL of an 8 nM phosphorylated
c-Met kinase domain (WT or mutant), increasing concentrations of
inhibitor in a volume of 1.6 and 48 μL of buffer (60 mM HEPES, pH
7.4, 50 mM NaCl, 20 mM MgCl2, 5 mM MnCl2, 2 mM DTT, 0.1 mM
Na3VO4, 0.05% BSA) for 30 min at room temperature. An amount of
20 μL of ATP and gastrin (final concentrations are 4 μM for ATP (2/3
of Km) and 1 μM for the biotinylated gastrin) in the same buffer is
then added to the reaction in a final volume of 80 μL and incubated at
room temperature for 60 min. An amount of 5 μL of the above
reaction is then added to a reaction mixture containing 11 nM
streptavidin−allophycocyanin (S-APC) and 0.1 nM europium-labeled
anti-phosphotyrosine antibody (Eu-PT66) in a final volume of 85 μL
for 30 min at room temperature before data capture using a
fluorescence plate reader.
Cell-Based Assay. IC50 measurements of inhibitor activity on
HGF-mediated c-Met autophosphorylation were determined in serum-
starved PC-3 cells using a quantitative electrochemiluminescence
immunoassay. PC-3 cells were plated in high glucose DMEM with
10% FBS at a density of 20 000 cells/well in 96-well plates. The next
day, cells were starved in low glucose DMEM containing 0.1% BSA for
16 h. Cells in the starvation media were then treated with a 10-point
serial dilution of the inhibitor for 1 h at 37 °C followed by stimulation
with 200 ng/mL of recombinant human HGF for 10 min at 37 °C.
Cells were washed once with PBS and lysed (1% Triton X-100, 50 mM
Tris, pH 8.0, 100 mM NaCl, Na3VO4, and protease inhibitors). Cell
lysates were then used to measure the levels of c-Met phosphorylation
using a quantitative assay as follows: A biotinylated antibody against c-
Met (R&D Systems no. BAF358) was preincubated with streptavidin
beads (IGEN no. 110029) for 30 min at rt with rotation. Cell lysates
(25 μL) were then added to 25 μL of biotin labeled anti-c-Met
antibodies for 1 h at room temperature with shaking. An
antiphosphotyrosine antibody 4G10 (Upstate no. 05-321) (12.5 μL)
was then added and allowed to incubate for 1 h at rt followed by the
5-(8-Fluoro-3-(1-(3-(2-methoxyethoxy)quinolin-6-yl)ethyl)-
[1,2,4]triazolo[4,3-a]pyridin-6-yl)-3-methylisoxazole (21f). Pre-
pared from 1-(3-fluoro-5-(3-methylisoxazol-5-yl)pyridin-2-yl)-
hydrazine (18a) and 2-(3-(2-methoxyethoxy)quinolin-6-yl)propanoic
acid hydrochloride in 65% yield (over 2 steps) according to the general
procedure described for 21e. Separation of enantiomers provided 25%
yield of (S)-21f (>99% ee) and 35% yield of (R)-21f (>99%ee). Data
for (S)-21f: 1H NMR (400 MHz, DMSO-d6) δ ppm 8.65 (d, J = 0.98
Hz, 1H), 8.58 (d, J = 2.84 Hz, 1H), 7.92 (d, J = 8.61 Hz, 1H), 7.75
(dd, J = 0.93, 11.49 Hz, 1H), 7.71 (d, J = 1.76 Hz, 1H), 7.70 (d, J =
2.84 Hz, 1H), 7.63 (dd, J = 2.01, 8.66 Hz, 1H), 6.98 (s, 1H), 5.22 (q, J
= 7.08 Hz, 1H), 4.12−4.35 (m, 2H), 3.64−3.82 (m, 2H), 3.31 (s, 3H),
2.27 (s, 3H), 1.89 (d, J = 7.04 Hz, 3H). LCMS (ESI): m/z (M + H) =
448.2.
6-(2-(8-Fluoro-6-(3-methylisoxazol-5-yl)[1,2,4]triazolo[4,3-
a]pyridin-3-yl)propan-2-yl)quinolin-3-ol (22). To a flask charged
with 25 (51 mg, 0.10 mmol) was added trifluoroacetic acid (1 mL).
The solution was heated at 65 °C for 18 h. The solution was cooled to
rt and concentrated. The crude residue was purified by MPLC using a
gradient of 0−10% MeOH in CH2Cl2 to afford 30 mg (75% yield) of
22. 1H NMR (400 MHz, DMSO-d6) δ ppm 10.38 (s, 1H), 8.55 (d, J =
2.74 Hz, 1H), 7.87 (d, J = 2.05 Hz, 1H), 7.82 (d, J = 8.80 Hz, 1H),
7.72 (dd, J = 0.88, 11.44 Hz, 1H), 7.58 (d, J = 1.08 Hz, 1H), 7.55 (d, J
= 2.35 Hz, 1H), 7.20 (dd, J = 2.15, 8.80 Hz, 1H), 6.83 (s, 1H), 2.17 (s,
3H), 1.98 (s, 6H). LCMS (ESI): m/z (M + H) = 404.4.
5-(8-Fluoro-3-(1-(3-methoxyquinolin-6-yl)cyclopropyl)-
[1,2,4]triazolo[4,3-a]pyridin-6-yl)-3-methylisoxazole (23). Pre-
pared from 1-(3-fluoro-5-(3-methylisoxazol-5-yl)pyridin-2-yl)-
hydrazine (18a) and 1-(3-methoxyquinolin-6-yl)cyclopropane-
carboxylic acid hydrochloride in 22% yield (over 2 steps) according
1
to the general procedure described for 21e. H NMR (400 MHz,
DMSO-d6) δ ppm 8.56 (d, J = 2.93 Hz, 1H), 8.31 (d, J = 1.17 Hz, 1H),
7.89 (d, J = 8.80 Hz, 1H), 7.84 (dd, J = 1.12, 11.59 Hz, 1H), 7.63 (d, J
= 2.64 Hz, 1H), 7.50 (d, J = 2.15 Hz, 1H), 7.33 (dd, J = 2.20, 8.85 Hz,
1H), 7.05 (s, 1H), 3.84 (s, 3H), 2.25 (s, 3H), 1.69−1.85 (m, 4H).
LCMS (ESI): m/z (M + H) = 416.4.
5-(8-Fluoro-3-(1-(3-methoxyquinolin-6-yl)propyl)[1,2,4]-
triazolo[4,3-a]pyridin-6-yl)-3-methylisoxazole (24). Prepared
from 1-(3-fluoro-5-(3-methylisoxazol-5-yl)pyridin-2-yl)hydrazine
(18a) and 2-(3-methoxyquinolin-6-yl)butanoic acid hydrochloride in
47% yield (over 2 steps) according to the general procedure described
for (21e). Separation of enantiomers provided 17% yield of 24 (>99%
ee, S-enantiomer). 1H NMR (400 MHz, DMSO-d6) δ ppm 8.65 (d, J =
1.08 Hz, 1H), 8.58 (d, J = 2.93 Hz, 1H), 7.93 (d, J = 8.61 Hz, 1H),
7.77 (dd, J = 1.12, 11.59 Hz, 1H), 7.72 (d, J = 1.86 Hz, 1H), 7.69 (d, J
= 2.35 Hz, 1H), 7.63 (dd, J = 2.05, 8.71 Hz, 1H), 7.00 (s, 1H), 5.23 (q,
J = 7.07 Hz, 1H), 3.88 (s, 3H), 2.52 (d, J = 1.96 Hz, 2H), 2.27 (s, 3H),
1.89 (d, J = 7.04 Hz, 3H). LCMS (ESI): m/z (M + H) = 418.2.
5-(3-(2-(3-(Benzyloxy)quinolin-6-yl)propan-2-yl)-8-fluoro-
[1,2,4]triazolo[4,3-a]pyridin-6-yl)-3-methylisoxazole (25). Pre-
pared from 1-(3-fluoro-5-(3-methylisoxazol-5-yl)pyridin-2-yl)-
hydrazine (18a) and 2-(3-(benzyloxy)quinolin-6-yl)-2-methylpropa-
2428
DOI: 10.1021/jm501913a
J. Med. Chem. 2015, 58, 2417−2430

Journal of Medicinal Chemistry
Article
addition of 12.5 μL of an ORI-Tag-labeled antimouse IgG (IGEN no.
110087) for 30 min at rt. PBS (175 μL) was then added to each
reaction, and levels of c-Met phosphorylation were measured using an
IGEN instrument (Biomek FX). The IC50 values are calculated using
Xlfit4-parameter equation.
Microsomal Incubations. Rat and mouse liver microsomes were
incubated at 0.25 mg/mL protein, 1.0 μM substrate, and 1 mM
NADPH in pH 7.4 phosphate buffer at 37 °C. Aliquots were removed
from the incubation at 0, 10, 20, 30, and 40 min and added to an equal
volume of acetonitrile. The samples were centrifuged and analyzed by
LC/MS/MS. CLint was calculated from the in vitro half-life (t1/2) of
the disappearance of substrate according to the following equation:
CLint = 0.693/(t1/2 × 0.25 mg protein/mL).30
(3) Comoglio, P. M.; Giordano, S.; Trusolino, L. Drug development
of MET inhibitors: targeting oncogene addition and expedience. Nat.
Rev. Drug Discovery 2008, 7, 504−516.
(4) (a) Cui, J. J. Targeting receptor tyrosine kinase MET in cancer:
small molecule inhibitors and clinical progress. J. Med. Chem. 2014, 57,
4427−4453. (b) Porter, J. Small molecule c-Met kinase inhibitors: a
review of recent patents. Expert Opin. Ther. Pat. 2010, 20, 159−177.
(5) Albrecht, B. A.; Harmange, J.-C.; Bauer, D.; Berry, L.; Bode, C.;
Boezio, A. A.; Chen, A.; Choquette, D.; Dussault, I.; Fridrich, C.; Hirai,
S.; Hoffman, D.; Larrow, J. F.; Kaplan-Lefko, P.; Lin, J.; Lohman, J.;
Long, A. M.; Moriguchi, J.; O’Connor, A.; Potashman, M. H.; Reese,
M.; Rex, K. R.; Siegmund, A.; Shah, K.; Shimanovich, R.; Springer, S.
K.; Teffera, Y.; Yang, Y.; Zhang, Y.; Bellon, S. F. Discovery and
optimization of triazolopyridazines as potent and selective inhibitors of
the c-Met kinase. J. Med. Chem. 2008, 51, 2879−2882.
ASSOCIATED CONTENT
■
(6) All potency values listed as “kinase” or “HTRF” refer to the IC50
measurements of inhibitor activity against the recombinant c-Met
kinase domain as described in the Experimental Section under “Kinase
Assay.” All potency values listed as “cell” are IC50 measurements of
inhibitor activity on HGF-mediated c-Met autophosphorylation and
were determined in serum-starved PC-3 cells using a quantitative
electrochemiluminescence immunoassay, also described in the
Experimental Section under the description of “Cell-Based Assay”.
For both assays, n ≥ 2.
S
* Supporting Information
Crystallographic data collection and refinement statistics for
(R)-20c (PDB code 4XMO) and the cocrystal structure of (S)-
21f (PDB code 4XYF) with c-Met, the kinase selectivity panel
assay data for (S)-21f, and the rat bile-duct cannulation ion
chromatograms (10b and (S)-21f). This material is available
free of charge via the Internet at http://pubs.acs.org.
(7) Boezio, A. A.; Berry, L.; Albrecht, B. K.; Bauer, D.; Bellon, S. F.;
Bode, C.; Chen, A.; Choquette, D.; Dussault, I.; Hirai, S.; Kaplan-
Lefko, P.; Larrow, J. F.; Lin, M.-H. J.; Lohman, J.; Potashman, M. H.;
Rex, K.; Santostefano, M.; Shah, K.; Shimanovich, R.; Springer, S. K.;
Teffera, Y.; Yang, Y.; Zhang, Y.; Harmange, J.-C. Discovery and
optimization of potent and selective triazolopyridazine series of c-Met
inhibitors. Bioorg. Med. Chem. Lett. 2009, 19, 6307−6312.
(8) Silverman, R. B. Drug Metabolism. The Organic Chemistry of Drug
Design and Drug Action; Academic Press: New York, 1992; pp 300−
301.
(9) For a review of kinase inhibitor design, see the following:
Zuccotto, F.; Ardini, E.; Casale, E.; Angiolini, M. Through the
“gatekeeper door”: exploiting the active kinase conformation. J. Med.
Chem. 2010, 53, 2681−2694. The hinge binder residue for c-Met is
Met1160.
AUTHOR INFORMATION
■
Corresponding Author
*Phone: 617-444-5027. Fax: 617-621-3907. E-mail: emily.
peterson@amgen.com.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors are grateful to Michael Zhang and Yajing Yang for
enzyme and cell assay support, Loren Berry for in vitro PKDM
analysis, and Larry Miller and Matt Potter for chiral separation
of products bearing stereogenic centers.
(10) General Stille coupling conditions described for the synthesis of
intermediate 26 are in the Experimental Section.
ABBREVIATIONS USED
■
(11) Wang, Y.; Sarris, K.; Sauer, D. R.; Djuric, S. W. A simple and
efficient automatable one step synthesis of triazolopyridines from
carboxylic acids. Tetrahedron Lett. 2007, 48, 2237−2240.
(12) (a) Laurence, C.; Brameld, K. A.; Graton, J.; Le Questel, J.-Y.;
Renault, E. The pKBHX database: toward a better understanding of
hydrogen-bond basicity for medicinal chemists. J. Med. Chem. 2009,
52, 4703−4086. (b) Laurence, C.; Gal, J.-F. Thermodynamic and
Spectroscopic Scales of Hydrogen-Bond Basicity and Affinity. In Lewis
Basicity and Affinity Scales: Data and Measurement; Wiley: New York,
2010; pp 119−135.
(13) Incubation of inhibitors with RLM and glutathione led to
displacement of the 8-fluorine even in the absence of NADPH. For a
review on GSH conjugation see the following: Hayes, J. D.; Pulford, D.
J. The glutathione S-transferase supergene family: regulation of GST*
and the contribution of the isoenzymes to cancer chemoprotection
and drug resistance. Crit. Rev. Biochem. Mol. Biol. 1995, 30, 445−600.
(14) Hinson, J. A.; Roberts, D. W.; James, L. P. Adverse Drug
Reactions. In Handbook of Experimental Pharmacology; Uetrecht, J.,
Ed.; Springer: New York, 2010; Vol. 196, Part 3, pp 369−405.
(15) Roberge, J. Y.; Yu, G.; Mikkilineni, A.; Wu, X.; Zhu, Y.;
Lawrence, R. M.; Ewing, W. R. Synthesis of triazolopyridines and
triazolopyrimidines using a modified Mitsunobu reaction. ARKIVOC
2007, 12, 132−147.
HGF, hepatocyte growth factor; RTK, receptor tyrosine kinase;
TDI, time-dependent inhibition; CYP3A4, cytochrome P450
3A4; SAR, structure−activity relationship; CLint, intrinsic
clearance; GST, glutathione S-transferase; GSH, glutathione;
CL, clearance; Vss, volume of distribution; RLM, rat liver
microsome; HLM, human liver microsome; BDC, bile duct
cannulation; MsLM, mouse liver microsome; DLM, dog liver
microsome; MLM, monkey liver microsome; X-Phos, dicyclo-
hexylphosphino-2′,4′,6′-triisopropylbiphenyl
REFERENCES
■
(1) Gherardi, E.; Birchmeier, W.; Birchmeier, C.; Vande Woude, G.
Targeting MET in cancer: rationale and progress. Nat. Rev. Cancer
2012, 12, 89−103.
(2) (a) Schmidt, L.; Duh, F. M.; Chen, F.; Kishida, T.; Glenn, G.;
Choyke, P.; Scherer, S. W.; Zhuang, Z.; Lubensky, I.; Dean, M.;
Allikmets, R.; Chidambaram, A.; Bergerheim, U. R.; Feltis, J. T.;
Casadevall, C.; Zamarron, A.; Bernues, M.; Richard, S.; Lips, C. J.;
Walther, M. M.; Tsui, L. C.; Geil, L.; Orcutt, M. L.; Stackhouse, T.;
Lipan, J.; Slife, L.; Brauch, H.; Decker, J.; Niehans, G.; Hughson, M.
D.; Moch, H.; Storkel, S.; Lerman, M. I.; Linehan, W. M.; Zbar, B.
Germline and somatic mutations in the tyrosine kinase domain of the
MET proto-oncogene in papillary renal carcinomas. Nat. Genet. 1997,
16, 68−73. (b) Liu, X.; Yao, W.; Newton, R. C.; Scherle, P. A.
Targeting the c-Met signaling pathway for cancer therapy. Expert Opin.
Invest. Drugs 2008, 17, 997−1011.
(16) The more potent inhibitor when R2 = F (20) is the (R)
enantiomer, and when R2 = H (21), it is the (S) enantiomer. In both
cases the methyl group is oriented as drawn in Scheme 2; the
stereochemical assignment changes due to the priority assignment
rules for fluorine vs hydrogen.
2429
DOI: 10.1021/jm501913a
J. Med. Chem. 2015, 58, 2417−2430

Journal of Medicinal Chemistry
Article
(17) Atomic coordinates and structure factors for the cocrystal
structures of (R)-20c and (S)-21f (pictured in graphical abstract and
in Supporting Information) bound to c-Met have been deposited in
the PDB with accession codes 4XMO and 4XYF, respectively.
(18) Muller, K.; Faeh, C.; Diederich, F. Fluorine in pharmaceuticals:
̈
looking beyond intuition. Science 2007, 317, 1881−1886.
(19) Note the change of (R) vs (S) assignment changes from
compound 20 to compound 21 as the presence of the fluorine atom
alters the priority of substituents for stereochemical assignment.
(20) For a more detailed description of the methods used in these
experiments see the following: Teffera, Y.; Colletti, A. E.; Christophe-
Harmange, J.-C.; Hollis, L. S.; Albrecht, B. K.; Boezio, A. A.; Liu, J.;
Zhao, Z. Chemical reactivity of methoxy 4-O-aryl quinolines:
identification of glutathione displacement products in vitro and in
vivo. Chem. Res. Toxicol. 2008, 21, 2216−2222.
(21) For an analysis of electronic properties of various five-
membered heterocycles, see the following: Katritzky, A. R.;
Pozharskii, A. F. Handbook of Heterocyclic Chemistry, 2nd ed.;
Pergamon: Amsterdam, 2000; p 378.
(22) See Supporting Information.
(23) Similar to compounds 10 and 12, the in vitro predicted
clearance (CLint) for compounds 20 and 21 agreed with the observed
in vivo clearance. The unbound clearance (CLu) values for (S)-21e
and (S)-21f were 14 and 25 L h−1 kg−1, respectively.
(24) The parent compound (S)-21e was also observed in all species
(peak not shown).
(25) The parent compound (S)-21f was also observed in all species
(peak not shown).
(26) Kinase counterscreen performed at Ambit; see Supporting
Information for heat map.
(27) Plasma protein binding (PPB) values for (S)-21f: mouse =
85.4%, human = 94.7%, dog = 85.1%, rat = 97.8%, and monkey =
92.9%.
(28) For a discussion of the relationship between c-Met target
coverage and tumor growth inhibition, see the following: Yamazaki, S.;
Skaptason, J.; Romero, D.; Lee, J. H.; Zou, H. Y.; Christensen, J. G.;
Koup, J. R.; Smith, B. J.; Koudriakova, T. Pharmacokinetic−
pharmacodynamic modeling of biomarker response and tumor growth
inhibition to an orally available cMet kinase inhibitor in human tumor
xenograft mouse models. Drug Metab. Dispos. 2008, 36, 1267−1274.
(29) Procedure adapted from the following: Lee, S.; Beare, N. A.;
Hartwig, J. F. Palladium-catalyzed α-arylation of esters and protected
amino acids. J. Am. Chem. Soc. 2001, 123, 8410−8411.
(30) Obach, R. S. Prediction of human clearance of twenty-nine
drugs from hepatic microsomal intrinsic clearance data: an
examination of in vitro half-life approach and nonspecific binding to
microsomes. Drug Metab. Dispos. 1999, 27, 1350−1359.
2430
DOI: 10.1021/jm501913a
J. Med. Chem. 2015, 58, 2417−2430