Design and synthesis of conformationally constrained inhibitors of non-nucleoside reverse transcriptase

Article abstract of DOI:10.1021/jm2010173

Highly active antiretroviral therapy (HAART) significantly reduces human immunodeficiency virus (HIV) viral load and has led to a dramatic decrease in acquired immunodeficiency syndrome (AIDS) related mortality. Despite this success, there remains a critical need for new HIV therapies to address the emergence of drug resistant viral strains. Next generation NNRTIs are sought that are effective against these mutant forms of the HIV virus. The bound conformations of our lead inhibitors, MK-1107 (1) and MK-4965 (2), were divergent about the oxymethylene linker, and each of these conformations was rigidified using two isomeric cyclic constraints. The constraint derived from the bioactive conformation of 2provided novel, highly potent NNRTIs that possess broad spectrum antiviral activity and good pharmacokinetic profiles. Systematic SAR led to the identification of indazole as the optimal conformational constraint to provide MK-6186 (3) and MK-7445 (6). Despite their reduced flexibility, these compounds had potency comparable to that of the corresponding acyclic ethers in both recombinant enzyme and cell based assays against both the wild-type and the clinically relevant mutant strains.

Full text of DOI:10.1021/jm2010173

Article
pubs.acs.org/jmc
Design and Synthesis of Conformationally Constrained Inhibitors of
Non-Nucleoside Reverse Transcriptase
Robert Gomez,*,† Samson J. Jolly,† Theresa Williams,† Joseph P. Vacca,† Maricel Torrent,‡
Georgia McGaughey,‡ Ming-Tain Lai,§ Peter Felock,§ Vandna Munshi,§ Daniel DiStefano,∥ Jessica Flynn,∥
Mike Miller,⊥ Youwei Yan,# John Reid,# Rosa Sanchez,∞ Yuexia Liang,∞ Brenda Paton,∞
Bang-Lin Wan,∞ and Neville Anthony†,●
†
‡
§
Departments of West Point Discovery Chemistry, Chemistry Modeling and Informatics, In Vitro Pharmacology,
∥Vaccine Research, ID Antiviral HIV Discovery, Structural Chemistry, and ∞DMPK Preclinical WP, Merck Research Labs.,
⊥
#
770 Sumneytown Pike, P.O. Box 4, West Point, Pennsylvania 19486-0004, United States
ABSTRACT: Highly active antiretroviral therapy (HAART)
significantly reduces human immunodeficiency virus (HIV)
viral load and has led to a dramatic decrease in acquired immuno-
deficiency syndrome (AIDS) related mortality. Despite this
success, there remains a critical need for new HIV therapies to
address the emergence of drug resistant viral strains. Next
generation NNRTIs are sought that are effective against these
mutant forms of the HIV virus. The bound conformations of
our lead inhibitors, MK-1107 (1) and MK-4965 (2), were divergent about the oxymethylene linker, and each of these
conformations was rigidified using two isomeric cyclic constraints. The constraint derived from the bioactive conformation of 2
provided novel, highly potent NNRTIs that possess broad spectrum antiviral activity and good pharmacokinetic profiles.
Systematic SAR led to the identification of indazole as the optimal conformational constraint to provide MK-6186 (3) and
MK-7445 (6). Despite their reduced flexibility, these compounds had potency comparable to that of the corresponding
acyclic ethers in both recombinant enzyme and cell based assays against both the wild-type and the clinically relevant
mutant strains.
forms of HIV. The most predominant mutants selected by first
generation NNRTI therapy are the K103N and the Y181C
variants.11,12 Next generation NNRTIs are sought that
possess broad spectrum antiviral activity against key mutant
strains (especially K103N and Y181C) and that also present
a high genetic barrier to the selection of new drug resistant
viral strains. Additionally, to facilitate patient compliance, a
next generation NNRTI that is both efficacious with once
daily dosing and amenable to fixed dose combination
HAART regimens is highly desirable. The NNRTI TMC-
12513 has been approved by the FDA for twice daily oral
dosing and has demonstrated activity toward a number of
key clinical mutations. More recently, the more potent
follow-up compound TMC-27814 has been approved by the
FDA to be administered orally once daily.
INTRODUCTION
■
HIV has a high rate of mutation due to its infidelity during
replication. As a result, single drug therapy is not found to
be durable because of the rapid selection of resistant mutant
strains.1−3 HAART significantly reduces HIV viral load and
has led to a dramatic decrease in AIDS related mortality and
a better quality of life for patients.4 HAART typically
combines at least two anti-HIV drugs simultaneously, as
this greatly reduces the probability of the virus developing
drug resistance.5 The emergence of HIV-1 strains resistant
to at least one antiretroviral drug highlights the need for
further development of antiviral agents with improved
efficacy against the mutant strains that are observed
clinically.6,7
Reverse transcriptase (RT) inhibitors block the conversion of
single stranded viral RNA to double stranded proviral DNA, a
prerequisite to integration into host DNA.8 NNRTIs interact
with the RT enzyme at an allosteric site and induce a change in
the substrate binding site which interferes with its DNA
polymerase activity.9 This class of inhibitor has gained an
increasingly important role in HAART therapy because of its
high specificity and lack of mitochondrial toxicity which
characterizes the nucleoside inhibitors (NRTI’s).10 However,
because of their allosteric mechanism of action, a common
consequence of NNRTI use is the emergence of drug resistant
Tucker et al.15 have previously reported a series of diaryl
ethers that feature a pyrazolopyridine, as exemplified by
MK-1107 (1) (Figure 1). This compound has excellent
potency versus wild type (wt) and key mutant viruses. The
X-ray crystal structure of 1 (magenta) bound to the K103N
RT enzyme (PDB entry 3DRS) (Figure 2) shows that the
terminal benzonitrile ring fills a large lipophilic pocket in the
Received: July 30, 2011
Published: October 10, 2011
© 2011 American Chemical Society
7920
dx.doi.org/10.1021/jm2010173|J. Med. Chem. 2011, 54, 7920−7933

Journal of Medicinal Chemistry
Article
and 2 (gray) present two distinct, divergent conformations
about the oxymethylene linker (Figure 2). This prompted the
speculation that each of these conformations might be
reinforced by the introduction of appropriate cyclic constraints.
Conformational rigidification is a commonly used strategy in
drug design,16,17 as it can lead to improved potency and
selectivity while reducing the potential for drug metabolism.
However, the consequence of rigidifying 1 and 2, in terms of
the effect on their antiviral profiles, was unclear, as it has been
argued that conformational flexibility13 is a necessary feature of
broadly active NNRTIs.
Figure 1. Lead structures.15
After considering a variety of potential cyclic constraints of 1
and 2, we envisioned that a fused five membered heteroaryl
ring in which a nitrogen atom replaces the etheral oxygen atom
would be a conservative way to evaluate this approach. The
bound conformation of 2 can be rigidified with constraint A,
while the bound conformation of 1 can be supported by
constraint B (Figure 2). It was hypothesized that the
increased steric hindrance and rigidity intrinsic to the ring
constraints would reduce oxidative dealkylation of the
pyrazolopyridine moiety which is a major metabolic
clearance pathway for 1.15 Additionally, it was proposed
that increasing the chemical diversity of these ring
constraints by the incorporation of additional nitrogen
atoms would lead to changes in physiochemical properties
which would modulate both the potency and pharmacoki-
netic parameters of these molecules.
Triazoles A and B (X = N, Y = N, Figure 2), each derived
from constraints A and B, respectively, were docked into the
cognate site of RT (PDB entries 3DRP and 3DRS).18 The
resulting poses were compared with the bioactive conforma-
tions of 1 and 2 (Figures 3 and 4). The bioactive conformations
Figure 2. Bound conformation of 1 (magenta) superimposed with X-ray
crystal structure of 2 (gray) in wt RT enzyme (PDB entry 3DRP).15 This
view shows select NNRTI binding site residues within 7 Å of ligand. Key
active site residues are highlighted and labeled in black.
allosteric site formed by Y181−Y188−W229. Additionally, the
central phenyl ring fits into a small lipophilic pocket formed
by Y181−V179−V106. Importantly, key backbone hydrogen
bond interactions between amino acid residue K103 and the
pyrazole nitrogens are observed and may account for its persistent
inhibition in the presence of the K103N mutant. A solvent
exposed region can be accessed near this pyrazolopyridine ring,
and water solubilizing groups have been added to the 6-position
to affect the physical properties of compounds in this series which
ultimately led to the 6-aminopyrazolopyridine, MK-4965 (2).
The crystal structure of the corresponding 6-amino analogue
2 bound to wt RT has also been determined (PDB entry
3DRP), and this molecule makes similar contacts with the
enzyme. Interestingly, the cocrystal structures of 1 (magenta)
Figure 3. X-ray crystal structure of 2 (gray) bound to wt RT
superimposed with the docked orientation of structure A (X, Y = N)
(cyan). This view shows select NNRTI binding site residues within 7 Å of
ligand. Key active site residues are highlighted and labeled in black.
of 1 and 2 showed that the oxygen of the oxymethylene linker
is partially sp3 hybridized and these linkers have measured
dihedral angles of 40° and 32°, respectively, relative to the central
phenyl ring. The nitrogen atoms of the modeled triazoles A
7921
dx.doi.org/10.1021/jm2010173|J. Med. Chem. 2011, 54, 7920−7933

Journal of Medicinal Chemistry
Article
to limit additional substitution at this position. Thus,
computational studies were carried out to quantitate the effects
of increased steric bulk at this position. Structure A compounds
where the 2-position (Y) comprised N, C−H, and C−Me were
each docked in the cognate site of RT. The ligand energy of
each of these bioactive conformations was compared to the
energy of the lowest energy conformer for each compound in
vacuum, and these values are summarized in Table 1. The
Table 1. Ligand Energies Illustrating Effect of 2-Substitution
of Ring Constraint
Figure 4. X-ray crystal structure of 1 (magenta) bound to K103N RT
(PDB code 3DRS) superimposed with the docked orientation of
structure B (X, Y = N) (yellow). This view shows select NNRTI
binding site residues within 7 Å of ligand. Key active site residues are
highlighted and labeled in black.
low energy
ligand energy, bound
conformer,
Y
to wt RT (kcal/mol) vacuum (kcal/mol) Δenergy (kcal/mol)
N
78.06
80.83
85.63
69.47
69.91
71.71
8.59
10.92
13.92
and B, however, are sp2 hybridized resulting in linkages
that are coplanar with the central phenyl ring. Despite
these differences, the modeled triazole constraints A and
B orient the terminal pyrazolopyridine and benzonitrile
rings so that they maintain the key interactions with the
enzyme.
The increased steric environment brought about by the ring
constraint A appears to be accommodated in the cognate site
of RT as shown by the docked model of triazole A (X = N,
Y = N) in wt RT (Figure 3). The proximal residues of RT
include L100 and Y318. L100 is 3.8 Å from the constraint,
while the hydroxyl of Tyr318 is 3.5 Å distal and has the
potential to form hydrogen bonds with heteroatoms at the 2-
position of the constraint. Visual inspection of the docked
conformation of structure A as a benzimidazole (X = N, Y =
CH) reveals a potential non-bonding interaction between the
2-position of the benzimidazole and the 4-position of the
pyrazolopyridine (Figure 5). This interaction was anticipated
C−H
C−Me
change in energy (ΔE) between the bioactive conformation
and the low energy conformation in vacuum provides an
estimate of the amount of energy needed to attain the bioactive
conformation. The ΔE for the methylbenzimidazole (Y = C−
Me, 13.9 kcal/mol) is 5.3 kcal/mol higher than the ΔE for the
2-aza compound (Y = N, 8.6 kcal/mol). In addition, the ΔE of
the benzimidazole (Y = C−H, 10.9 kcal/mol) was 2.3 kcal/mol
higher than the 2-aza compound (Y = N, 8.6 kcal/mol),
suggesting that even a hydrogen at this position would
significantly impact the ability of these compounds to access
the bioactive conformation and was predicted to lead to less
potent inhibition of RT.
The increased steric environment brought about by ring
constraint B, on the other hand, appears to be less well
accommodated in the cognate site of RT as shown by the
docked model of triazole B (X = N,Y = N) in wt RT (Figure 4).
The proximal residues of RT to this constraint include K101
and K102. The carbonyl oxygen of K101 is only 2.4 Å from the
constraint, while the α-carbon of K102 is 2.8 Å from the
constraint. These potential nonbonded interactions with the
constraint suggest that success with constraint B is dependent
on the inherent flexibility of the enzyme13 to accommodate this
additional steric requirement. In addition, the docked model of
triazole B shows a 20° tipping of the core ring, resulting in a 1.6 Å
displacement of the chloro group relative to the corresponding
group in the crystal structure of 1. This tipping was required in
order to maintain the important interactions of the protein with
the terminal rings but is expected to be detrimental to binding.
For these reasons, constraint A was considered more viable and
was evaluated in more detail.
RESULTS AND DISCUSSION
■
Data for the compounds from the cyclic constraint A series
is summarized in Table 2. Compounds were evaluated for
Figure 5. Structure A modeled in bioactive conformation of 2 showing
the potential steric interactions with the 2-hydrogen.
7922
dx.doi.org/10.1021/jm2010173|J. Med. Chem. 2011, 54, 7920−7933

Journal of Medicinal Chemistry
Article
Table 2. Enzyme Inhibition and Antiviral Activity of 1 and 2 and Constraint Analogues 3−11
a
b
inhibition of RT, Ki (nM)
antiviral activity in cell culture, CIC95 (nM)
compd
X
Y
Z
wt
K103N
Y181C
0.23 ± 0.02
0.39 ± 0.02
0.60 ± 0.04
0.44 ± 0.02
4.2
wt
10 ± 0.6
4.9 ± 0.3
11 ± 1.1
4.3 ± 0.6
57
K103N
15 ± 1.3
12 ± 1.0
15 ± 1.6
3.8 ± 0.8
232 ± 23
9.7 ± 1.6
5.8
Y181C
22 ± 2.1
27 ± 1.6
58 ± 7
18 ± 6
418 ± 104
68
wt 50% NHS
52 ± 4.2
26 ± 1.5
64 ± 6.1
23 ± 2.0
281 ± 59
43 ± 5
c
1
H
0.12 ± 0.03
0.20 ± 0.01
0.34 ± 0.05
0.24 ± 0.02
2.2
0.22 ± 0.04
0.39 ± 0.01
0.58 ± 0.04
0.43 ± 0.04
4.3
c
2
NH2
H
3
CH
N
N
N
N
N
4
N
H
5
N
H
6
CH
N
NH2
NH2
H
0.61 ± 0.06
1.1 ± 0.6
2.0 ± 0.8
>300
1.7 ± 0.2
0.94
1.9 ± 0.2
1.2
4.1 ± 0.3
3
7
37 ± 2
320 ± 57
ND
36 ± 2
8
N
CH
5.3
3.3
23
66 ± 17
>833
390 ± 225
>833
9
NH
N
CO
CMe
N
H
>600
>300
>833
10
11
H
>300
>600
>300
>833
>833
ND
>833
CCl
H
46
28
42
>500
>2000
>2000
>5000
a
Compounds were evaluated in a standard SPA assay. Values are the geometric mean of at least two determinations. All individual values are within
2-fold of the mean. The standard error of the mean is included for compounds with ≥3 determinations. ND = not determined. Assay protocols are
b
detailed in ref 19. CIC95 (cell culture inhibitory concentration) is defined as the concentration at which the replication of virus is inhibited by >95%
in MT-4 human T-lymphoid cells maintained in RPMI 1640 medium containing either 10% FBS or 50% NHS. Values are the geometric mean of at
least two determinations. All individual values are within 2-fold of the mean. The standard error of the mean is included for compounds with ≥3
determinations. Details of the assay protocol are provided in ref 20. ND = not determined. No cytotoxicity was observed for any of the compounds
c
up to the upper limit of the assay (8.3 μM). Oxyether linker. See Figure 1 for structures.
intrinsic enzyme inhibitory potency19 versus wt RT as well as
the K103N and Y181C mutant variants. Inhibitors were also
evaluated for antiviral potency20 and measured as the cell
inhibitory concentration that blocks 95% of cell replication
(CIC95) against wt and the key mutant viruses in the
presence of 10% fetal bovine serum (FBS) and with 50%
normal human serum (NHS) to evaluate the effects of
protein binding.
3T19).21 Comparison of the bound conformation of 5 from
this X-ray crystal structure with the crystal structure of 2 in
wt RT (PDB entry 3DRP) shows high similarity between the
structures (Figure 6) and is consistent with the previously
Incorporation of a nitrogen atom at the 2-position of the
constraint to minimize the potential steric repulsion with the
C−H bond at the 4-position of pyrazolopyridine led to the
indazole 3. This ring constraint resulted in a 2- to 3-fold
reduction in intrinsic enzyme inhibition of both wt and the
K103N and Y181C mutants relative to the lead oxyether 1.
Indazole 3 was, however, equipotent to 1 in cell culture against
both wt and K103N mutant viruses but was 3-fold less active
against the Y181C mutant. Incorporation of an additional
nitrogen atom at the 3-position of the ring constraint provided
the benzotriazole 4. This compound was slightly more potent
than the indazole 3 in the enzyme assays and was significantly
(3- to 4-fold) more potent in the cell based assays.
Unfortunately, attempts to obtain X-ray structural informa-
tion of either indazole 3 or benzotriazole 4 bound to wt RT
were not successful, presumably because of their limited
aqueous solubility. The crystal structure of 2 bound to RT
(PDB entry 3DRP) reveals that the nitrile group occupies a
solvent exposed region of the enzyme. Reduction of this nitrile
group on a chloro analogue of 4 provided the aminomethyl
derivative 5. The improved physical properties of 5 facilitated
ligand exchange and provided an X-ray crystal structure of 5
complexed with wt RT at a resolution of 2.6 Å (PDB entry
Figure 6. Bound conformation of 521 (aqua) overlaid with crystal
structure of 2 (gray) in wt RT.15 This view shows select NNRTI
binding site residues within 7 Å of ligand. Key active site residues are
highlighted and labeled in black.
described docking studies (Figure 3).18 The Y181 side chain
orientation in this structure is analogous to the orientation of
7923
dx.doi.org/10.1021/jm2010173|J. Med. Chem. 2011, 54, 7920−7933

Journal of Medicinal Chemistry
Article
Figure 7. Antiviral potency of key compounds versus panel of isolated clinical viruses. Compounds were analyzed in a Monogram Bioscience
Phenoscreen assay in the presence of FBS. The IC50 is defined as the concentration of compound in cell culture required to block 50% of viral
replication. Values are derived from the average of two determinations, and all individual values are within 30% of the mean. Details are provided in
ref 24. MDRC4 refers to a multidrug resistant virus. CNDO refers to a wild type virus.
this side chain observed for the ethers 1 and 2 (Figure 2). This
conformation of Y181 has been observed previously15,22 and
provides an explanation for the maintainance of good potency
in the presence of the Y181C mutation. Despite requisite
realignment in the core ring, the key interactions with the
protein residues that comprise the NNRTI binding site are
preserved including the backbone hydrogen bond interactions
between K103 and the pyrazole nitrogen atoms of the
pyrazolopyridine group. Interestingly, NNRTIs containing an
indazole central ring that is unsubstituted on the N1 nitrogen23
are observed to bind in the opposite orientation compared to
the compounds described here. Hydrogen bonding of the N1
hydrogen atom to the K101 backbone carbonyl is responsible
for this alternative binding mode.
The pyrazolopyridine of 5 occupies the solvent exposed
region outlined by P236 and suggests that the corresponding
6-amino group would be tolerated in accord with the findings
with oxyether 2. The 6-amino group was incorporated onto
indazole and benzotriazole scaffolds to yield the more water-
soluble derivatives 6 and 7. As anticipated, 6 and 7 had potent
antiviral activity in cell culture similar to 3 and 4 against both
wt and the K103N and Y181C mutants.
Incorporation of a basic heterocycle into the cyclic constraint
was evaluated in the context of the more basic and polar
benzimidazole 8. Consistent with the computational studies,
introduction of even a hydrogen atom at the 2 position of the
constraining scaffold leads to diminished potency in both
enzyme inhibition (3- to 10-fold) and antiviral activity in cell
culture (2- to 4-fold) relative to the indazole 3. Importantly, the
increased polarity of benzimidazole 8 leads to an increase in
protein binding, as indicated by a larger shift in antiviral activity
in the presence of NHS. As anticipated by our computational
studies, substitution at the 2 position of the benzimidazole 8 to
give the benzimidazol-2-one 9 and the 2-methylbenzimidazole
10 results in a complete loss of enzyme inhibition and antiviral
activity. This demonstrates that the higher energetic cost
associated with each ligand’s ability to access the bioactive
conformation correlates with the observed loss of inhibitory
potency.
The impact of introducing a substituent at the 3 position of
the cyclic constraint was also evaluated. Examination of the
docked analogues of structure A in wt RT (Figure 3) suggested
that there is ample room in the binding site to accommodate
substitution at this position. Surprisingly, the 3-chloro
substituted indazole 11 is 10-fold less active than the parent
indazole 3, perhaps because of a sterically induced reorientation
of the benzonitrile ring.
The conformationally constrained inhibitors 3, 4, and 6 were
evaluated for their antiviral potency against a more extensive
panel of viruses from clinical isolates.24 The activity profiles of
these constrained inhibitors compared to the more flexible
oxymethylene 2 and the preferred first generation NNRTI
inhibitor efavirenz (EFV) are shown in Figure 7. Inhibitors 2, 3,
4, and 6 had markedly improved profiles relative to efavirenz
(magenta). All four inhibitors possess broad spectrum activities
against the mutant viruses tested. The notable exceptions to
this activity include the rare single mutation Y188L (occurs in
0.6% of clinical isolates).25 This is presumably due to disruption
of the π stacking interactions between the benzonitrile ring,
which is common to each of these compounds, and the aryl
ring of the Y188 residue. Each of these compounds was also
ineffective against the triple mutation, V106A/G190A/F227L,
presumably in part because of the loss of the hydrophobic
interaction between the V106 side chain and the central phenyl
ring of the ligands. The indazole 6 (yellow) showed an overall
activity profile comparable to that of 2 (blue) but was 2-fold
less potent against the virus containing the K103N/Y181C
double mutant. Indazole 3 (purple) showed a profile
comparable to that of 2 (blue) against the viruses that
contained single and double mutations and improved activities
against the viruses that contained the triple mutations. The
benzotriazole 4 (brown) had the best profile and showed
marked improvement in potency against many of the viruses
with double and triple mutations. These results demonstrate
that despite their more rigid nature, these conformationally
constrained inhibitors retain activity against a broad panel of
mutant viruses and that the observed activity profiles are
comparable to those of their more flexible oxyether
counterparts.
7924
dx.doi.org/10.1021/jm2010173|J. Med. Chem. 2011, 54, 7920−7933

Journal of Medicinal Chemistry
Article
Our attention then turned to evaluation of the alternative
constraint B. Within this context, the constraint B analogues
benzotriazole 12 and the benzimidazole 13 were prepared, and
the enzyme inhibitory and the antiviral activity potencies are
shown in Table 3. Both 12 and 13 were significantly less active
Table 4. Pharmacokinetics of the Oxymethylene Leads 1 and
2 and Key Cyclically Constrained Leads 3, 4, 6, 7, and 8
a
Clp,
mL min−1
kg−1
nAUC (po)
μM·h
Vd, L/kg
t1/2, h
F, %
compd rat
dog
rat dog rat dog rat dog rat
dog
Table 3. Enzyme Inhibition and Antiviral Activity of Con-
straint B Analogues 12−14
b
1
2
3
4
6
7
8
5.3
2.0
2.6
1.7
28
9
8.9 4.1 11.8 7.5
5.8 3.5 3.9 52 47 2.3
6.1 7.6 12.8 76 48 1.6
3
0.1
0.14
3.1
cd
,
10.5 6.0
18
14
4.7
2.3
5.5
2.5
4.4
3.3
1.3
2.0
1.6 4.8 11.3 7.3 9.5 0.18 0.45
7.6 6.3 2.8 39 32 3.64 1.33
e
f
0.65 11.9 2.2 4.7 12.8 <1 nd 0.02 nd
0.37 25.6 2.2 2.4 12.8 36 nd 0.56 nd
a
Average of at least two Sprague−Dawley rats dosed at 10 mpk po
(Methocel suspension) and 2 mpk iv (DMSO) unless otherwise
noted. All values are within 25% of the mean. Average of at least two
beagle dogs dosed at 2 mpk po (Methocel suspension) and 1mpk iv
antiviral CIC95
a
b
inhibition of RT, Ki (nM)
(nM)
b
compd
Y
wt
K103N
Y181C
wt
K103N
(DMSO) unless otherwise noted. nd = not determined. Average of
two beagle dogs dosed at 1 mpk po (Methocel suspension) and
12
13
N
>300
114
2.3
>300
>100
3.8
>900
>100
3.3
>833
693
24
>833
732
40
c
0.25 mpk iv (DMSO). Average of four Spraque−Dawley rats dosed at
CH
d
10 mpk po (Methocel suspension) and 3 mpk iv (DMSO). Average
c
14
of three beagle dogs dosed at 10 mpk po (Methocel suspension) and
a
e
Compounds were evaluated in a standard SPA assay. All values for 12
and 13 are from at least two determinations. The nonqualified value
for 13 is the geometric mean of two determinations. Assay protocols
are detailed in ref 19. CIC95 (cell culture inhibitory concentration) is
defined as the concentration at which the spread of virus is inhibited
by >95% in MT-4 human T-lymphoid cells maintained in RPMI 1640
medium containing either 10% FBS or 50% NHS. All values for 12 and
13 are from at least two determinations. The nonqualified values for
13 are the geometric mean of two determinations. Details of the assay
protocol are provided in ref 20. No cytotoxicity was observed for any
2 mpk iv (DMSO). Average of two beagle dogs dosed at 2 mpk po
f
(Methocel suspension) and 0.5 mpk iv (DMSO). Average of three
beagle dosed dosed at 0.125 mpk iv (DMSO) as part of a cassette
dosing protocol.
b
dosing in rat and dog. The corresponding oxymethylene ether
1 had similarly limited water solubility (<0.1 μg/mL at pH 2)
but poor oral bioavailability. Incorporation of the solubilizing
6-amino group onto the pyrazolopyridine moiety in 1 to give 2
was shown15 to improved water solubility (10 μg/mL at pH 2)
and increase bioavailability following oral dosing. The 6-amino
analogue 6 was likewise more water-soluble (40 μg/mL) than 3
(<0.1 μg/mL) at pH 2 but was similarly poorly soluble (<0.1
μg/mL) at pH 7. Despite being more soluble at acidic pH, 6
had similar bioavailability and exposure with respect to 3 in
both rat and dog. This suggests that low solubility does not
significantly limit the absorption of indazole 3 and is consistent
with absorption of both compounds occurring in the intestinal
tract at sites of higher pH. It is interesting to note that the effect
of constraining the oxymethylene ether 1 to the indazole 3
leads to a >10-fold increase in exposure whereas this increased
exposure was not realized in the context of the corresponding
the 6-amino derivatives 2 and 6.
The benzotriazole 4 has both lower volumes of distribution
and lower clearances relative to indazole 3, which results in
comparable half-lives. However, despite the presence of an
additional heteroatom, benzotriazole 4 retains poor water
solubility (<0.1 μg/mL at pH 2). Contrary to indazole 3, the
permeability of benzotriazole 4 was presumably insufficient to
overcome its poor aqueous solubility and resulted in low
exposures. The 6-aminobenzotriazole 7 had an iv pharmaco-
kinetic profile very similar to that of the parent 4 and, despite
the solubilizing amino group, had negligible bioavailability.
The effect of introducing a more basic heterocycle into the
constraining scaffold was examined for its effects on solubility
and pharmacokinetic behavior. The more basic benzimidazole
8 (pKa = 3.7 26) is likely to be protonated in the stomach, which
results in increased water solubility (10 μg/mL at pH 2).
However, this compound had a smaller volume of distribution
and a higher plasma clearance in rat than the corresponding
c
of the compounds up to the upper limit of the assay (8.3 μM). The
values for 14 are from single determinations.
(50- to >1000-fold) than the corresponding alternatively
constrained analogues 4 and 8. Clearly the enzyme lacks the
flexibility required to accommodate these heterocycles.
Interestingly and in support of this interpretation, the ring
opened intermediate 14, while lacking the cyclic constraint, is
considerably more potent. This improved activity may result
from a realignment of the core benzene ring to produce a new
hydrogen bonding interaction between the 4-amino group and
the backbone carbonyl of the K101 residue.
The pharmacokinetic parameters of key compounds were
determined in rats and dogs and are tabulated in Table 3.
Sprague−Dawley rats received an intravenous (iv) dose as a
solution in DMSO via an in-dwelling canula implanted in the
jugular vein or an oral dose as a suspension in aqueous methyl-
cellulose. Similar procedures were used to evaluate pharmaco-
kinetic parameters in dogs with the animals receiving iv and oral
doses in a crossover fashion. Plasma concentrations of the indicated
compounds were determined by liquid chromatography−
mass spectrometry analysis, and the absolute oral bioavail-
ability was determined by comparing the oral dose normalized
area under the plasma concentration vs time curve (nAUC)
with the iv nAUC.
The pharmacokinetics of the indazole 3 are characterized by
a large volume of distribution (Vd of 10.5 and 6.0 L/kg), a
moderate plasma clearance (Clp of 18 and 6.1 mL min−1 kg−1),
and long half-lives (t1/2 of 7.6 and 12.8 h) in rats and dogs
compared to the unconstrained oxymethylene ether 1 (Table 4).
Despite its limited water solubility (<0.1 μg/mL at pH 2), 3
had good bioavailability (F of 76% and 48%) following oral
7925
dx.doi.org/10.1021/jm2010173|J. Med. Chem. 2011, 54, 7920−7933

Journal of Medicinal Chemistry
Article
a
Scheme 1
a
Reagents and conditions: (a) K2CO3, NMP, 140 °C, 18 h, 65%; (b) LDA, THF, −78 °C, then DMF, 72%; (c) hydrazine hydrate, EtOH, 95°C,
4 days, 64%; (d) Zn(CN)2, Pd(PPh3)4, DMF, 90 °C, 1 h, 89%; (e) Cs2CO3, DMF, then TFA, 36% (+20% 2-isomer); (f) KOtBu, PMB-Br, DMF,
68% (+22% isomer); (g) (i) PMB-NH2, NMP, 80 °C, 3 h; (ii) LAH, THF, 0 °C to room temp, 84% (two steps); (h) thionyl chloride, CHCl3, 94%;
(i) (i) LiOtBu, DMF, 0 °C to room temp; (ii) TFA reflux; (iii) HCl, CHCl3/MeOH, 50% (+40% 2-isomer).
indazole 3, which resulted in lower exposure following oral
administration.
Synthesis of compounds 4 and 7−10 are outlined in Scheme 2.
The key transformation of this sequence was a vicarious
amination27 of the nitrobenzene 25 to give a 1:1 mixture of
regioisomeric aniline products that were readily purified to give
a 36% yield of 26 and 37% yield of 27. The nitroaniline 27 was
reduced to the diaminobenzene 28 with tin(II) chloride in
refluxing ethanol. Cyanide displacement of this bromide was
unsuccessful under a variety of conditions, so it became necessary
to form the heterocycles first and cyanate subsequently. All
four heterocycles were prepared from this versatile diamine
intermediate. Treatment of an acetic acid solution of the
bisaniline 28 with sodium nitrite gave the benzotriazole 29. The
benzimidazole 30 was formed by heating 28 to 100 °C in
formic acid. Treatment of 28 with triphosgene with pyridine as
base provided the benzimidazole-2-one 31. The 2-methyl-
benzimidazole 32 was synthesized by treating diamine 28 with
acetic acid at 100 °C. Heteroaryls 29−32 were each cyanated
using zinc cyanide in the presence of catalytic palladium
tetrakistriphenylphosphine to give 33−36. Heterocycles 33−36
were then each alkylated with bromide 20 using Cs2CO3 in
DMF to form a regioisomeric mixture of products which
were deprotected with TFA and purified to form the
benzotriazole 4, the benzimidazole 8, the benzimidazole-2-
one 9, and the 2-methylbenzimidazole 10. Additionally, the
lithium salt of benzotriazole 33 was alkylated with the bis PMB
protected chloride 24. The resulting regioisomeric mixture was
purified by chromatography, and the resulting bis PMB
intermediate was deprotected in refluxing TFA to give triazole
analogue 7.
The two constrained analogues that had the best overall
combination of activity and pharmacokinetic profiles were the
indazoles 3 and 6. Both of these inhibitors have pharmacoki-
netics in rat and dog that suggests a potential for once daily
dosing in humans.
CHEMISTRY
■
The synthesis of the indazole 3 was carried out as shown in
Scheme 1. SNAr reaction of 1-bromo-3-chloro-5-fluorobenzene
with 2-chloro-5-fluorophenol at 140 °C in NMP gave the
requisite biaryl ether 16. The key step in this synthesis was a
fluoride directed deprotonation of 16 with LDA which,
followed by quenching with DMF, gave the aldehyde 17
regioselectively. This aldehyde was reacted with excess
hydrazine to afford the core indazole 18. Palladium catalyzed
cyanation afforded the key indazole 19. This intermediate was
alkylated with the bromide 20 in DMF using Cs2CO3 as base to
provide a mixture of 1- and 2-substituted indazoles. However,
this reaction was complicated by a transfer of the Boc
protecting group from 20 to provide Boc protected 19 along
with bis and tris alkylated byproducts. Fortunately, this mixture
of products could be deprotected with TFA and purified
by chromatography to give a 36% yield of the desired N1
derivative 3 along with 20% of the undesired N2 isomer.
Compound 6 was similarly prepared by alkylation of the
lithium anion of indazole 19 with the bis PMB protected
chloride 24. This mixture was purified by chromatography and
deprotected in refluxing TFA to give 6.
The syntheses of constraint B compounds 12 and 13 are
outlined in Scheme 3. The synthesis takes advantage of the
7926
dx.doi.org/10.1021/jm2010173|J. Med. Chem. 2011, 54, 7920−7933

Journal of Medicinal Chemistry
Article
a
Scheme 2
a
+ −
Reagents and conditions: (a) K2CO3, NMP, 120 °C, 2 h, 85.4%; (b) KOtBu, NH2NMe3 I , cat. CuI, DMF, 0−20 °C, 37% (+36% isomer);
(c) SnCl2, EtOH, 75 °C, 95%; (d) NaNO2, AcOH, 95%; (e) HCO2H, 100 °C, 1 h, 100%; (f) triphosgene, pyridine, DCM, 0 °C, 20 min, 67%;
(g) AcOH, 100 °C, 6 h, 99%; (h) Zn(CN)2, Pd(PPh3)4, DMF, 90 °C, 69−80%; (i) Cs2CO3, DMF, then TFA, 7−43%; (g) (i) LiOtBu, DMF, 0 °C
to room temp; (ii) TFA reflux, 35%.
a
Scheme 3
a
Reagents and conditions: (a) PMB-NH2, toluene, reflux, 99%; (b) K2CO3, NMP, 80 °C, 1 h, 74%; (c) TFA, 75 °C,100%; (d) Cs2CO3, DMF, then
TFA, 32%; (e) Zn(CN)2, Pd(PPh3)4, DMF, 95 °C, 2 h, 80%; (f) SnCl2 H2O, EtOH, reflux; (g) NaNO2, HOAc; (h) HCO2H, 50 °C, 2 h.
chemoselective displacement of an activated nitro group over a
similarly activated chloro group. Specifically, reaction of the
symmetrical 1,2-dinitro-4,5-dichlorobenzene with PMB amine
gave only the nitro displaced adduct 37 in 99% yield. SNAr
displacement of the activated para chloride atom with 3-bromo-
5-chlorophenol gave 38 in 74% yield. This PMB protected
amine was treated with TFA to afford the nitroaniline 39. This
compound was alkylated with 20 to give a mixture of starting
aniline 39, desired alkylation product 40, and some of the
corresponding bis alkylation product. This mixture was purified
by chromatography to give a 32% yield of 40. The nitro group
was reduced with tin(II) chloride in refluxing ethanol to give
the differentially alkylated diamine 14. This common
intermediate was treated with sodium nitrite to form the
benzotriazole 12 and with formic acid at 50 °C to form the
benzimidazole 13.
DISCUSSION AND CONCLUSIONS
■
The bound conformations of the acyclic oxyether NNRTIs 1
and 2 in RT were reinforced by two types of cyclic confor-
mational constraints. Constraint B, which constrains the
molecule to a conformation seen in oxymethylene ether 1
bound to the K103N mutant form of RT, gave only modest
inhibition of reverse transcriptase presumably because of steric
interactions with the residues surrounding the NNRTI binding
pocket of the enzyme. In contrast, constraint A, which was
7927
dx.doi.org/10.1021/jm2010173|J. Med. Chem. 2011, 54, 7920−7933

Journal of Medicinal Chemistry
Article
then quenched with water (1000 mL), and the aqueous layer was
extracted with EtOAc (3 × 500 mL). The combined extracts were
washed with water (500 mL), dried over MgSO4 and the solvent
removed in vacuo. This residue was purified on a silica gel column
(0−40% CH2Cl2/hexanes) to give 17 (77.51 g, 71.5%) as a clear
designed to preorganize the molecule in a manner that mimics
the bound conformation of oxymethylene ether 2, provided
potent inhibitors with activity profiles that are similar to their
more flexible acyclic counterparts. The indazole 3 and the
benzotriazole 4 derived cores demonstrated the most potent
enzyme inhibition and broadest spectrum of antiviral activity.
The more water-soluble 6-aminopyrazolopyridine analogues,
6 and 7, were prepared in analogy to the oxymethylene series
and were found to be equipotent to 3 and 4. The indazoles 3
and 6 had good pharmacokinetic profiles in rats and dogs
and offer potential for once daily dosing in humans. These
compounds provide a combination of excellent pharmacoki-
netic profiles, potent enzyme inhibition, and broad spectrum
antiviral activities. As such, 3 and 6 were selected as preclinical
development candidates and designated MK-6186 and MK-
7445, respectively. Further details of the clinical development of
these compounds including formulation, pharmacokinetic
characterization, and process chemistry optimization will be
presented in subsequent publications from our laboratories.
1
liquid. H NMR (CDCl3): δ 10.22 (s, 1H), 7.72 (dd, J = 5.6 and 9.0
Hz, 1H), 7.24 (m, 1H), 7.15 (dd, J = 9 Hz, 1H), 6.88 (t, J = 1.7 Hz,
1H), and 6.77 (t, J = 1.7 Hz, 1H) ppm. LRMS (M − 18 + 1) = 346.7
4-(3-Bromo-5-chlorophenoxy)-5-chloro-1H-indazole (18).
To a suspension of 17 (77.5 g, 213 mmol) in EtOH (120 mL) was
added hydrazine hydrate (31 mL, 639 mmol). The resulting mixture
was then heated at reflux for 4 days, after which time the mixture was
cooled to room temperature, diluted with water (100 mL) and the
resulting solid was filtered. The solid was preabsorbed onto silica gel
(300 g) and purified on a silica gel (1500 g) column (0−40% EtOAc/
CH2Cl2) to give 18 (40.0 g, 52.3%) as a white solid. 1H NMR
(CDCl3) δ 10.15 (br s, 1H), 7.80 (s, 1H), 7.48 (d, J = 8.8 Hz, 1H),
7.38 (d, J = 8.8 Hz, 1H), 7.24 (m, 1H), 6.98 (t, J = 1.7 Hz, 1H) and
6.86 (t, J = 1.7 Hz, 1H) ppm. LRMS (M + 1) = 358.8.
3-Chloro-5-[(5-chloro-1H-indazol-4-yl)oxy]benzonitrile
(19). To a degassed suspension of 18 (40 g, 112 mmol) and Zn(CN)2
(15.74 g, 134 mmol) in DMF (400 mL) was added palladium
tetrakistriphenylphosphine (38.7 g, 33.5 mmol), and the mixture was
heated to 90 °C for 1 h. After this time, the mixture was cooled to
room temperature and partitioned between saturated aqueous
NH4Cl (500 mL), water (500 mL), and EtOAc (2 × 1000 mL).
The combined extracts were dried over MgSO4, absorbed onto silica
gel (200 g), and the solvent was removed in vacuo. This solid was
purified on a silica gel (1000 g) column (0−10% EtOAc/CH2Cl2) to
EXPERIMENTAL SECTION
■
Silica gel flash chromatography was performed on an ISCO
Combiflash Companion unit using prepacked ISCO RediSep silica
gel cartridges. Reverse phase chromatographic separations were
performed on a Gilson automated HPLC system using Luna C18,
10 μm, 100 Å columns. 1H NMR spectra were routinely obtained on a
400 MHz Varian FT NMR spectrometer and are referenced versus
TMS. NMR solvents used were CDCl3, CD3OD, or DMSO-d6 as
indicated. Chemical shifts are reported in δ units relative to the internal
standard TMS. LC−MS was performed using a Waters Alliance
2695/Micromass ZQ LC/MS system with a YMC Pro C18 5 μm/120 Å,
3 mm × 50 mm cartridge and a 95:5 A/B gradient (A = 92% water/8%
ACN/0.05% TFA; B = 100% ACN/0.0425% TFA) over 4 min at a
flow rate of 2 mL/min, with UV (215 or 254 nm) and mass detection.
Retention times and purity assessments for all final compounds are
reported from a second HPLC method using a Thermo Separations
system with a YMS Pro C18 S-5 120 Å, 4.6 mm × 33 mm cartridge
and a 95:5 gradient (A = water/0.1% TFA, B = ACN/0.085% TFA)
over 5 min at a flow rate of 1 mL/min with UV (215 and 254 nm)
detection. All compounds were assessed at >95% purity by both
analytical methods. Electrospray (ES) and atmospheric pressure
chemical ionization (APSI) accurate mass spectral data were acquired
by use of a Bruker Daltonics 7 T Fourier transform ion cyclotron
resonance (FTICR) spectrometer. External calibration was accom-
plished with oligomers of polypropylene glycol. Aldrich Sure-Seal
solvents were used when anhydrous solvents were necessary. All
reactions were carried out under inert atmosphere (argon or nitrogen).
3-Bromo-5-chlorophenyl 2-chloro-5-fluorophenyl ether
(16). To a solution of 2-chloro-5-fluorophenol (82.3 g, 562 mmol)
and 1-bromo-3-chloro-5-fluorobenzene (124 g, 590 mmol) in NMP
(200 mL) was added K2CO3 (155 g, 1.123 mol). The resulting mixture
was then heated to 140 °C for 30 h, after which the mixture was
poured into water (1500 mL) and the aqueous layer extracted with
EtOAc (2500 + 1500 mL). The combined extracts were washed with
water and brine. The solvent was removed in vacuo and the resulting
residue was distilled under high vacuum at 135−190 °C to give 16
(121.24 g, 64.3%) as a clear, colorless liquid. 1H NMR (CDCl3):
δ 7.45 (dd, 1H, J = 9.0 and 4.5 Hz), 7.28 (t, J = 1.7 Hz, 1H), 7.26 (s,
1H), 7.00 (t, 1 h, J = 1.95 Hz), 6.92 (ddd, J = 10.5, 7.6, and 2.7 Hz,
1H), 6.89 (t, J = 1.95 Hz, 1H) ppm.
1
give 19 (30.18 g, 89%) as a white solid. H NMR (CDCl3) δ 10.25
(br s, 1H), 7.82 (s, 1H), 7.49 (d, J = 8.8 Hz, 1H), 7.42 (d, J = 8.8 Hz,
1H), 7.36 (t, J = 1.5 Hz, 1H), 7.18 (t, J = 1.9 Hz, 1H), 7.02 (dd, J = 1.3
and 2.4 Hz, 1H) ppm. LRMS (M + 1) = 303.9.
3-Chloro-5-{[5-chloro-1-(1H-pyrazolo[3,4-b]pyridin-3-
ylmethyl)-1H-indazol-4-yl]oxy}benzonitrile (3). To a suspension
of 19 (24 g, 79 mmol) and tert-butyl 3-(bromomethyl)-1H-
pyrazolo[3,4-b]pyridine-1-carboxylate (20) (24.63 g, 79 mmol) in
DMF (150 mL) at 0 °C was added Cs2CO3 (51.4 g, 158 mmol). The
resulting mixture was allowed to warm to room temperature and then
stirred for 2 h. LC−MS shows 47% Boc protected 3, 28.8% Boc
protected 2-isomer. This mixture was partitioned between water (1000
mL) and EtOAc (1000 mL). The organic extract was washed with
water (1000 mL) and then with brine (200 mL), dried over MgSO4
and the solvent removed in vacuo. This residue was purified on a silica
gel (500 g) column (0−10% ACN/CH2Cl2) to give Boc protected 3
and 3 (from deprotection on column). The Boc protected 3 was
dissolved in TFA (50 mL) and allowed to stand at room temperature
for 10 min, after which the solvent was removed in vacuo. The
resulting solid and the previously isolated 3 were then preabsorbed
onto silica gel (50 g) and purified on a silica gel (330 g) column (10−
1
30% ACN/CH2Cl2) to give 3 (12.2 g, 35.6%) as a white solid. H
NMR (DMSO-d6): δ 13.65 (s, 1H), 8.49 (d, J = 4.6 Hz, 1H), 7.91 (d,
J = 8.3 Hz, 1H), 7.89 (s, 1H), 7.82 (d, J = 8.8 Hz, 1H), 7.62 (d, J = 8.6
Hz, 1H), 7.47 (m, 1H), 7.39 (m, 1H), 7.13 (dd, J = 7.7 and 4.6 Hz,
1H), and 6.05 (s, 2H) ppm. HRMS: measured m/z, [M + H]+
=
435.0514 (theoretical, 435.0522). LC purity 99.99%; tR = 3.754 min.
2-(3-Bromo-5-chlorophenoxy)-1-chloro-3-nitrobenzene
(25). 2,3-Dichloronitrobenzene (34.2 g, 178 mmol), 3-bromo-5-
chlorophenol (46.2 g, 223 mmol), and K2CO3 (61.6 g, 445 mmol)
were suspended in NMP (200 mL) and placed in an oil bath at 120 °C
for 2 h. The mixture was allowed to cool to room temperature. Water
(1000 mL) was added, and the aqueous layer was extracted with
EtOAc (2 × 1000 mL). The combined organic extracts were washed
with water (2 × 500 mL), dried over MgSO4, filtered, and the solvent
was evaporated in vacuo to a solid. This solid was recrystallized from
EtOAc (200 mL) to give 25 (51.23 g, 63.2%) as a crystalline solid. The
filtrate was further purified by chromatography on silica gel (330 g),
eluting with 0−25% CH2Cl2/hexanes to yield additional 25 (18 g,
22.2%). 1H NMR (CD3CN) δ 8.02 (dd, J = 1.5 and 8.3 Hz, 1H), 7.88
2-(3-Bromo-5-chlorophenoxy)-3-chloro-6-fluorobenzal-
dehyde (17). To a solution of 16 (100 g, 298 mmol) in THF (300
mL) cooled to −78 °C over a dry ice/acetone bath was added 1.8 M
lithium diisopropylamide in hexanes/THF/ethylbenzene (174 mL,
313 mmol) over 10 min. The resulting mixture was stirred for 20 min
and then treated with DMF (46.1 mL, 595 mmol). This mixture was
then removed from the cooling bath and allowed to warm to room
temperature and kept at room temperature for 1 h. The reaction was
7928
dx.doi.org/10.1021/jm2010173|J. Med. Chem. 2011, 54, 7920−7933

Journal of Medicinal Chemistry
Article
(dd, J = 1.6 and 8.2 Hz, 1H), 7.52 (dd, J = 8.2 and 8.3 Hz, 1H), 7.38
(m, 1H), 7.07 (m, 1H), and 6.97 (m, 1H) ppm.
raphy, eluting with 25−50% EtOAc/hexanes to afford the title
compound (1.10 g, 31%) as a white solid. 1H NMR (CDCl3): δ 8.72−
8.70 (m, 1H), 8.15−8.12 (m, 1H), 7.96 (d, J = 9.0 Hz, 1H), 7.83 (d,
J = 9.0 Hz, 1H), 7.82−7.80 (m, 1H), 7.58−7.54 (m, 2H), 7.45−7.41
(m, 1H), 6.47 (s, 2H), and 1.62 (s, 9H) ppm. LRMS (M + 1) = 435.7.
3-Chloro-5-{[5-chloro-1-(1H-pyrazolo[3,4-b]pyridin-3-
ylmethyl)-1H-1,2,3-benzotriazol-4-yl]oxy}benzonitrile (4). tert-
Butyl 3-{[5-chloro-4-(3-chloro-5-cyanophenoxy)-1H-1,2,3-benzotria-
zol-1-yl]methyl}-1H-pyrazolo[3,4-b]pyridine-1-carboxylate (1.10 g,
2.05 mmol) was dissolved in TFA (10 mL) at room temperature.
After 5 min, the solvent was evaporated in vacuo to a solid. This was
purified by silica gel chromatography, eluting with 0−5% MeOH/
CH2Cl2 to afford 4 (752 mg, 84%) as a white solid. 1H NMR (CDCl3):
δ 8.52−8.46 (m, 1H), 8.35 (d, J = 4.15 Hz, 1H), 7.59 (d, J = 8.79 Hz,
1H), 7.57 (d, J = 8.78 Hz, 1H), 7.36 (s, 1H), 7.33−7.27 (m, 1H), 6.98
(s, 1H), and 6.25 (s, 2H) ppm. HRMS: measured m/z, [M + H]+ =
436.0467 (theoretical, 436.0475). LC purity 99.7%; tR = 3.572 min.
1-{3-Chloro-5-[2-chloro-5-(1H-pyrazolo[3,4-b]pyridin-3-
ylmethoxy)phenoxy]phenyl}methenamine (5). A solution of
2,5-dichloro-3-{[5-chlro-1-(1H-pyrozolo[3,4-b]pyridine-3-ylmethyl)-
1H-1,2,3-benzotriazol-4-yl]oxy}benzonitrile (prepared as in 4) (283 mg,
0.601 mmol) in THF (30 mL) was cooled to 0 °C over an ice bath,
and to it was added 1 M LAH in THF (1.202 mL, 1.202 mmol). Then
the mixture was removed from the cooling bath and stirred at room
temperature for 1 h. After this time, the mixture was recooled over
an ice bath and then quenched with EtOAc (50 mL) and allowed to
warm to room temperature. This mixture was treated with saturated
aqueous Na2SO4 and stirred for 10 min to give a white suspension.
This suspension was treated with solid Na2SO4 until dry, and this
suspension was filtered and the solvent removed in vacuo. This residue
was purified by reverse phase preparative HPLC. The desired fractions
were combined, and the solvent was removed in vacuo. This residue
was partitioned between saturated aqueous Na2CO3 and CH2Cl2.
The organic extract was dried over MgSO4, filtered and the solvent
removed in vacuo. This residue was redissolved in 4 M HCl in dioxane
and the solvent was removed in vacuo to give 5 (82 mg, 27%)
3-(3-Bromo-5-chlorophenoxy)-4-chloro-2-nitroaniline (27).
CuCl (5.45 mg, 0.055 mmol) and K2CO3 (247 mg, 2.204 mmol)
were suspended in DMF (3 mL) at 0 °C and treated with 1,1,
1-trimethylhydrazinium iodide (139 mg, 0.689 mmol). This mixture
was allowed to stir over an ice bath for 5 min and then cooled to
−40 °C. This mixture was treated dropwise with a solution of 25
(200 mg, 0.551 mmol) in DMF (3 mL). After 15 min, this mixture was
quenched cold with aqueous saturated NH4Cl (10 mL) and water
(10 mL) and then allowed to warm to 25 °C. The aqueous layer was
extracted with EtOAc (2 × 50 mL), and the combined extracts were
washed with water (3 × 50 mL). The resulting organic layer was dried
over MgSO4, filtered and the solvent was removed in vacuo. This solid
was adsorbed onto silica gel (500 mg) and purified by silica gel
chromatography (0−25% EtOAc/CH2Cl2) to afford 26 (75.5 mg,
1
36%) and 27 (76.5 mg, 37%). 26: H NMR (CD3CN) δ 7.98 (d, J =
8.5 Hz, 1H), 7.33 (dd, J = 2 Hz, 1H), 7.04 (dd, J = 2 Hz, 1H), 6.94
(dd, J = 2 Hz, 1H), 6.83 (d, J = 9.3 Hz, 1H), and 5.67 (br s, 2H) ppm.
LRMS (M + 1) = 378.7. 27: 1H NMR (CD3CN) δ 7.41 (d, J = 9.3 Hz,
1H), 7.35−7.32 (m, 1H), 7.07−7.05 (m, 1H), 6.98−6.95 (m, 1H),
6.90 (d, J = 9.3 Hz, 1H) and 5.87 (br s, 2H) ppm. LRMS (M + 1) =
378.7.
3-(3-Bromo-5-chlorophenoxy)-4-chlorobenzene-1, 2-diamine
(28). Tin(II) chloride dihydrate (17.0 g, 75 mmol) and 27 (5.7 g, 15.1
mmol) were suspended in MeOH (100 mL) and heated to 75 °C.
After 11 h, the mixture was allowed to cool to room temperature and
the solvent was removed in vacuo. The resulting residue was diluted
with EtOAc (150 mL), and 10% aqueous Na2CO3 (250 mL) was
added with vigorous stirring until the pH was 10. The resulting white
paste was filtered through diatomaceous earth and the pad rinsed with
additional EtOAc (200 mL). The biphasic filtrate was separated and
the aqueous layer extracted again with EtOAc (200 mL). The
combined organic extracts were dried (MgSO4), filtered and the
1
solvent was removed in vacuo to yield 28 (4.97 g, 95%) as a solid. H
NMR (CD3CN): δ 7.30−7.28 (m, 1H), 6.97−6.94 (m, 1H), 6.87−
6.84 (m, 1H), 6.68 (d, J = 8.5 Hz, 1H), 6.59 (d, J = 8.5 Hz, 1H) ppm.
LRMS (M + 1) = 348.8.
1
as a white solid. H NMR (CDCl3): δ 8.53 (d, J = 4.5 Hz, 1H),
8.12 (d, J = 8 Hz, 1H), 7.95 (br s, 3H), 7.89 (d, J = 9 Hz, 1H),
7.79 (d, J = 9 Hz, 1H), 7.48 (s, 1H), 7.21 (dd, 5 and 8 Hz, 1H), 6.92
(s, 1H), 6.39 (s, 2H), and 3.57 (s, 2H) ppm. HRMS: measured m/z
4-(3-Bromo-5-chlorophenoxy)-5-chloro-1H-benzotriazole
(29). To a solution 28 (4.97 g, 14.28 mmol) in AcOH (25 mL) at
15 °C was added aqueous NaNO2 (0.4M, 39.3 mL). The mixture was
allowed to warm to room temperature. After 1.5 h, the mixture was
diluted with EtOAc (300 mL), and the organic layer was separated and
washed with water (3 × 100 mL). This extract was dried over MgSO4,
filtered and the solvent removed in vacuo to afford 29 (5.01 g, 98%) as
a solid. 1H NMR (DMSO-d6) δ 8.14 −8. 00 (m, 1H), 7.86−7.54 (m, 2H),
7.50−7.46 (m, 1H), 7.20−7.14 (m, 1H), 7.12−7.04 (m, 1H) ppm.
3-Chloro-5-[(5-chloro-1H-1,2,3-benzotriazol-4-yl)oxy]-
benzonitrile (33). A degassed suspension of 29 (5.00 g, 13.93
mmol), Zn(CN)2 (1.96 g, 16.71 mmol), and palladium tetrakis-
triphenylphosphine (4.83 g, 4.18 mmol) in dry DMF (50 mL) was
heated to 90 °C for 2.5 h. After this time, the mixture was allowed to
cool to room temperature, diluted with EtOAc (300 mL), washed with
water (4 × 100 mL), dried over MgSO4, and filtered. Silica gel (20 g)
was added to the filtrate, and the solvent was removed in vacuo. This
dry powder was purified using silica gel chromatography, eluting with
[M + H]+ = 474.0392 (theoretical, 474.0398). LC purity 97%; tR
=
2.372 min.
tert-Butyl 6-Fluoro-1-(4-methoxybenzyl)-1H-pyrazolo[3,4-
b]pyridine-3-carboxylate (22) and tert-Butyl 6-Fluoro-2-(4-
methoxybenzyl)-2H-pyrazolo[3,4-b]pyridine-3-carboxylate. To
a solution of 21 (29.55 g, 125 mmol) in THF (300 mL) cooled in an
ice bath was added KOtBu (13.98 g, 125 mmol) at such a rate as to
maintain the temperature between 5 and 10 °C. This mixture was
treated with 4-methoxybenzyl bromide (18.2 mL, 125 mmol) and
stirred over an ice bath for 1 h and then at 25 °C for 18 h. The
resulting suspension was quenched with saturated aqueous NH4Cl
(200 mL) and then the aqueous layer extracted with EtOAc (2 × 300
mL). The combined organic extracts were washed with brine, dried
over MgSO4, and the solvent was removed in vacuo. The resulting
residue was purified by chromatography using silica gel (1000 g),
eluting with 0−30% EtOAc/hexanes to give tert-butyl 6-fluoro-2-(4-
methoxybenzyl)-2H-pyrazolo[3,4-b]pyridine-3-carboxylate (9.75 g,
21.9%) and 22 (30.3 g, 68.2%) as white solids. 22: 1H NMR
(CDCl3): δ 8.44 (t, J = 8 Hz, 1H), 7.40 (d, J = 8.8 Hz, 2H), 6.90 (d,
J = 8.8 Hz, 1H), 6.83 (d, J = 8.5 Hz, 2H), 5.61 (s, 2H), and 3.76 (s, 3H)
ppm. LRMS (M + 1) = 380.1. tert-Butyl 6-fluoro-2-(4-methoxybenzyl)-
1
0−5% MeOH/CH2Cl2 to give 33 (3.83 g, 90%) as a solid. H NMR
(DMSO-d6) δ 7.90−7.80 (m, 2H), 7.75−7.60 (m, 1H), 7.60−7.45 (m,
2H) ppm.
tert-Butyl 3-{[5-Chloro-4-(3-chloro-5-cyanophenoxy)-1H-
1,2,3-benzotriazol-1-yl]methyl}-1H-pyrazolo[3,4-b]pyridine-1-
carboxylate. To a suspension of 33 (2.00 g, 6.55 mmol) and
Cs2CO3 (2.14 g, 6.55 mmol) in DMF (15 mL) at room temperature
was added 20 (1.84 g, 5.90 mmol) as a solution in DMF (5 mL). After
the mixture was stirred for 1 h at room temperature, the reaction was
quenched with aqueous NH4Cl (20 mL). The mixture was diluted
with water (10 mL), and the aqueous layer was extracted with EtOAc
(3 × 75 mL). The combined organic extracts were washed with water
(3 × 75 mL), dried over MgSO4, filtered and the solvent removed in
vacuo. The resulting residue was purified using silica gel chromatog-
1
2H-pyrazolo[3,4-b]pyridine-3-carboxylate: H NMR (CDCl3): δ 8.35
(t, J = 8.5 Hz, 1H), 7.43 (d, J = 8.5 Hz, 2H), 6.88 (d, J = 8.5 Hz, 1H),
6.82 (d, J = 8.5 Hz, 2H), 6.01 (s, 2H), and 3.76 (s, 3H) ppm. LRMS
(M + 1) = 380.1.
{1-(4-Methoxybenzyl)-6-[(4-methoxbenzyl)amino]-1H-
pyrazolo[3,4-b]pyridin-3-yl}methanol (23). To a solution of 22
(2.55 g, 7.14 mmol) in NMP (30 mL) was added 4-methoxbenzyl-
amine (2.80 mL, 21.41 mmol), and the mixture was heated to 80 °C
for 2 h. This mixture was allowed to cool to room temperature, diluted
7929
dx.doi.org/10.1021/jm2010173|J. Med. Chem. 2011, 54, 7920−7933

Journal of Medicinal Chemistry
Article
with water (200 mL), and the aqueous layer was extracted with EtOAc
(2 × 200 mL). The combined extracts were washed with water
(100 mL), dried over MgSO4, and filtered. Silica gel (17 g) was added,
and the solvent was removed in vacuo. This solid was purified by
chromatography using silica gel (80 g), eluting with 0−100% EtOAc/CH2Cl2
to give tert-butyl 1-(4-methoxybenzyl)-6-[(4-methoxybenzyl)amino]-
and placed in an oil bath at 75 °C for 2 h. This mixture was allowed to
cool to room temperature, and the solvent was removed in vacuo. The
resulting residue was suspended in CHCl3/EtOAc (4:1, 1200 mL),
and aqueous NaHCO3 (500 mL) was added and vigorously stirred
for 10 min. The layers were partitioned and the aqueous portion was
back-extracted with CHCl3/EtOAc (4:1, 1200 mL) and then EtOAc
(2 × 1000 mL). The extracts were each washed with water (500 mL)
and enough MeOH to keep them clarified. The combined extracts
were dried over MgSO4, filtered, and the solvent was removed in
vacuo. The resulting residue was purified by chromatography using
silica gel (330 g), eluting with 0−10% MeOH/CHCl3 to give 6 (7.08
1
1H-pyrazolo[3,4-b]pyridine-3-carboxylate. H NMR (CDCl3): δ 7.98
(d, J = 8.8 Hz, 1H), 7.33 (d, J = 8.8 Hz, 2H), 7.29 (d, J = 8.6 Hz, 2H),
6.86 (d, J = 8.6 Hz, 2H), 6.77 (d, J = 8.8 Hz, 2H), 6.36 (d, J = 8.8 Hz,
1H), 5.52 (s, 2H), 5.00 (t, J = 5.5 Hz, 1H), 4.61 (d, J = 5.5 Hz, 2H),
3.81 (s, 6H), and 1.54 (s, 9H) ppm. LRMS (M + 1) = 474.0.
1
To a solution of tert-butyl 1-(4-methoxybenzyl)-6-[(4-
methoxybenzyl)amino]-1H-pyrazolo[3,4-b]pyridine-3-carboxylate
(38.4 g, 81 mmol) in THF (300 mL) cooled over an ice bath was
added 2 M LAH in THF (50 mL, 100 mmol) at such a rate to keep the
temperature below 10 °C. The resulting mixture was stirred over an ice
bath for 20 min and then allowed to warm to 25 °C for 4 h. This
mixture was then recooled over an ice bath and treated sequentially
with water (3.8 mL), 15% aqueous NaOH (3.8 mL), and water (11.4 mL)
and then stirred for 30 min. The resulting suspension was filtered
through diatomaceous earth, and the resulting cake was rinsed with
more THF (300 mL). The combined filtrates were treated with silica
gel (100 g), and the solvent was removed in vacuo. This preabsorbed
solid was purified by chromatography using silica gel (500 g), eluting
with 0−20% MeOH/CH2Cl2 to give 23 (23.6 g, 83% for two steps) as
a white solid. 1H NMR (DMSO-d6): δ 7.73 (d, J = 9 Hz, 1H), 7.26 (d,
J = 8 Hz, 2H), 7.21 (d, J = 8 Hz, 2H), 6.87 (d, J = 8 Hz, 4H), 6.41 (d,
J = 9 Hz, 1H), 5.49 (t, 1H), 5.34 (s, 2H), 4.74 (d, J = 5.4 Hz, 2H), 4.45
(d, J = 5.4 Hz, 2H), and 3.72 (s, 6H) ppm. LRMS (M + 1) = 404.9.
3-(Chloromethyl)-N,1-bis(methoxybenzyl)-1H-pyrazolo[3,4-
b]pyridin-6-amine (24). To a slurry of 23 (1.49 g, 3.68 mmol) in
CHCl3 (15 mL) cooled over an ice bath was added thionyl chloride
(538 μL, 7.37 mmol). The resulting mixture was then allowed to warm
to room temperature and left at room temperature for 30 min. The
reaction was determined by LC/MS monitoring not to be complete, so
the mixture was recooled over an ice bath and treated with additional
thionyl chloride (100 μL) and then stirred at room temperature for
30 min. This mixture was poured into aqueous NaHCO3 (50 mL),
extraction was with CHCl3 (100 mL), and the solvent was removed in
vacuo. This residue was purified by chromatography using silica gel
(120 g), eluting with 0−60% EtOAc/CHCl3 to give 24 (1.46 g, 94%)
g, 95%) as a white solid. H NMR (DMSO-d6): δ 12.60 (s, 1H), 7.86
(s, 1H), 7.81−7.78 (m, 1H), 7.73 (d, J = 9.0 Hz, 1H), 7.58 (d, J = 9.0
Hz, 1H), 7.49−7.47 (m, 1H), 7.38−7.34 (m, 2H), 6.30 (s, 2H), 6.21
(d, J = 8.8 Hz, 1H), and 5.8882 (s, 2H) ppm. LRMS (M + 1) = 449.7.
3-{[5-Chloro-4-(3-chloro-5-cyanophenoxy)-1H-indazol-1-yl]-
methyl}-1H-pyrazolo[3,4-b]pyridin-6-aminium chloride
(6·HCl). To a warmed (40 °C) solution of 6 (7.08 g, 15.72 mmol)
in 20% CH3OH/CHCl3 (1200 mL) was added 1 N HCl (15.72 mL,
15.72 mmol), and the resulting mixture was then concentrated in
vacuo. This solid was azeotroped from ACN (3 × 200 mL) and dried
under high vacuum overnight to give 6·HCl (7.19 g, 94%) as a white
solid. 1H NMR (DMSO-d6 with puff NH3 vapor): δ 7.86 (s, 1H), 7.81
(t, J = 1.5 Hz, 1H), 7.73 (d, J = 9 Hz, 1H), 7.58 (d, J = 9 Hz, 1), 7.49
(dd, J = 1.5 Hz, 1H), 7.37 (d, J = 6 Hz, 1H), 7.36 (s, 1H), 6.30 (br s,
2H), 6.23 (d, J = 8.8 Hz, 1H), and 5.82 (s, 2H) ppm. HRMS:
measured m/z, [M + H]+ = 450.0623 (theoretical, 450.0631). LC
purity 98.4%; tR = 2.888 min.
3-Chloro-5-{[5-chloro-1-({2-(4-methoxybenzyl)-6-[(4-
methoxybenzyl)amino]-2H-pyrazolo[3,4-b]pyridin-3-yl}-
methyl)-1H-1,2,3-benzotriazol-4-yl]oxy}benzonitrile. The re-
gioisomer of 24, 3-(chloromethyl)-N,2-bis(4-methoxybenzyl)-2H-
pyrzolo[3,4-b]pyridin-6-amine, was prepared analogously to 24
starting from tert-butyl 6-fluoro-2-(4-methoxybenzyl)-2H-pyrazolo-
[3,4-b]pyridine-3 carboxylate. LiOtBu (37 mg, 0.459 mmol) and 33
(140 mg, 0.459 mmol) were dissolved in DMF (1 mL), and the
mixture was allowed to stir for 20 min. The mixture was then cooled
over an ice bath, and a solution of 3-(chloromethyl)-N,2-bis(4-
methoxybenzyl)-2H-pyrazolo[3,4-b]pyridine-6-amine (198 mg, 0.468
mmol) in DMF (1 mL) was added dropwise. After 1 h, the reaction
was quenched with saturated aqueous NH4Cl (3 mL). The mixture
was diluted with water (3 mL), and the aqueous layer was extracted
with EtOAc (2 × 25 mL). The combined organic extracts were washed
with brine (4 × 10 mL), dried over MgSO4, filtered, and the solvent
was removed in vacuo. This residue was purified by chromatography
using silica gel (12 g), eluting with 0−100% EtOAc/hexanes to give
the title compound (135 mg, 42.5%) and the 2-isomer 3-chloro-5-{[5-
chloro-2-({2-(4-methoxybenzyl)-6-[40methoxybenzyl)amino]-2H-
pyrazolo[3,4-b]pyridine-3-yl}methyl)-2H-1,2,3-benzotriazol-4-yl]-
oxy}benzonitrile. 1H NMR (CD3CN): δ 7.58 (d, J = 11 Hz, 1H), 7.53
(t, J = 1.5 Hz, 1H), 7.46 (d, J = 8.9 Hz, 1H), 7.32−7.25 (m, 4H), 7.17
(d, J = 8.9 Hz, 1H), 6.91−6.83 (m, 4H), 6.67 (d, J = 8.7 Hz, 2H), 6.45
(d, J = 9 Hz, 1H), 6.12 (s, 2H), 5.96 (br t, 1H), 5.44 (s, 2H), 4.53 (d, J =
6 Hz, 2H), 3.74 (s, 3H), and 3.70 (s, 3H) ppm. LRMS (M + 1) = 690.5.
3-({1-[(6-Amino-1H-pyrazolo[3,4-b]pyridin-3-yl)methyl]-5-
chloro-1H-1,2,3-benzotriazol-4-yl}oxy)-5-chlorobenzonitrile
(7). A solution of 3-chloro-5-{[5-chloro-1-({2-(4-methoxybenzyl)-6-
[(4-methoxybenzyl)amino]-2H-pyrazolo[3,4-b]pyridin-3-yl}methyl)-
1H-1,2,3-benzotriazol-4-yl]oxy}benzonitrile (135 mg, 0.195 mmol) in
TFA (10 mL) was heated to 75 °C for 2 h. This mixture was allowed
to cool to room temperature, and the solvent was removed in vacuo.
The resulting residue was diluted with CHCl3/EtOAc (4:1, 50 mL),
and CH3OH was added until a clear solution was achieved. This solution
was washed with 50% aqueous Na2CO3 (10 mL), and the aqueous
portion was back-extracted with CHCl3 (25 mL). The combined
extracts were dried (MgSO4), filtered, and the solvent was removed in
vacuo. This residue was preabsorbed onto silica gel (500 mg) and
purified by chromatography using silica gel (12 g), eluting with 0−5%
MeOH/CH2Cl2 to afford 7 (51 mg, 58%) as a white solid. 1H
NMR (DMSO-d6) δ 7.82 (d, J = 9.0 Hz, 1H), 7.81 (m, 1H), 7.75
1
as a white solid. H NMR (CDCl3): δ 7.76 (d, J = 8.8 Hz, 1H), 7.30
(d, J = 8.8 Hz, 2H), 7.26 (d, J = 8.8 Hz, 2H), 6.86 (d, J = 8.5 Hz, 2H),
6.79 (d, J = 8.8 Hz, 2H), 6.29 (d, J = 8.5 Hz, 1H), 5.41 (s, 2H), 5.00 (t,
J = 5.8 Hz, 1H), 4.80 (s, 2H), 4.61 (d, J = 5.8 Hz, 2H), 3.80 (s, 3H),
and 3.76 (s, 3H) ppm. LRMS (M + 1) = 422.9.
3-Chloro-5-{[5-chloro-1-({1-(4-methoxybenzyl)-6-[(4-
methoxybenzyl)amino]-1H-pyrazolo[3,4-b]pyridin-3-yl}-
methyl)-1H-indazol-4-yl]oxy}benzonitrile. A solution of 19 (9.0 g,
29.6 mmol) in DMF (75 mL) was treated with LiOtBu (2.37 g, 29.6
mmol) and stirred for 5 min. This mixture was cooled over an ice
bath and treated with a solution of 24 (12.77 g, 30.2 mmol) in DMF
(75 mL). This mixture was stirred at 25 °C for 18 h. After this time,
aqueous NH4Cl (200 mL) was added and extracted with EtOAc (2 ×
500 mL). The combined extracts were dried over MgSO4, filtered, and
the solvent was removed in vacuo. This residue was purified by
chromatography using silica gel (330 g), eluting with 0−10% EtOAc/
CH2Cl2 to give the title compound (11.49 g, 56.2%) and 3-chloro-5-
{[5-chloro-2-({1-(4-methoxybenzyl)-6-[(4-methoxybenzyl)amino]-
1H-pyrazolo[3,4-b]pyridin-3-yl}methyl)-2H -indazol-4-yl]oxy}-
benzonitrile (8.23 g, 40.3%) as white solids. 1H NMR (CDCl3): δ 7.72
(s, 1H), 7.36−7.23 (m, 7H), 7.14 (m, 1H), 6.94 (m, 1H), 6.83 (d, J =
9.5 Hz, 2H), 6.81 (d, J = 8.8 Hz, 2H), 6.13 (d, J = 8.8 Hz, 1H), 5.78 (s,
2H), 5.44 (s, 2H), 4.94 (t, J = 5.6 Hz, 1H), 4.55 (d, J = 5.6 Hz, 2H),
and 3.78 (s, 6H) ppm. LRMS (M + 1) = 691.8.
3-({1-[(6-Amino-1H-pyrazolo[3,4-b]pyridin-3-yl)methyl]-5-
chloro-1H-indazol-4-yl}oxy)-5-chlorobenzonitrile (6). 3-Chloro-
5-{[5-chloro-1-({1-(4-methoxybenzyl)-6-[(4-methoxybenzyl)amino]-
1H-pyrazolo[3,4-b]pyridin-3-yl}methyl)-1H-indazol-4-yl]oxy}-
benzonitrile (11.49 g, 16.64 mmol) was dissolved in TFA (100 mL)
7930
dx.doi.org/10.1021/jm2010173|J. Med. Chem. 2011, 54, 7920−7933

Journal of Medicinal Chemistry
Article
(d, J = 9.0 Hz, 1H), 7.58 (m, 1H), 7.55 (d, J = 8.8 Hz, 1H), 7.52 (m,
1H), 6.36 (br s, 2H), 6.30 (d, J = 8.8 Hz, 1H), and 6.16 (s, 2H) ppm.
LRMS (M + 1) = 450.6.
desired monoalkylated product were combined and lyophilized. This
residue was dissolved in TFA (3 mL) for 5 min and the solvent was
removed in vacuo. This residue was again purified using reverse phase
1
3-{[5-Chloro-4-(3-chloro-5-cyanophenoxy)-1H-1,2,3-benzo-
triazol-1-yl]methyl}-1H-pyrazolo[3,4-b]pyridin-6-aminium
chloride (7·HCl). A suspension of 7 (48 mg, 0.106 mmol) was
chromatography to give 9 (8.3 mg, 7%) as a white solid. H NMR
(DMSO-d6): δ 13.6 (s, 1H), 11.58 (s, 1H), 8.51 (dd, J = 1.5 and
3 Hz, 1H), 8.19 (d, J = 8 Hz, 1H), 7.76 (m, 1H), 7.37 (m, 1H), 7.33
(m, 1H), 7.22 (d, J = 8.5 Hz, 1H), 7.20 (dd, J = 4.4 and 8 Hz, 1H),
7.13 (d, J = 9 Hz, 1H), and 5.34 (s, 2H) ppm. HRMS: measured
m/z, [M + H]+ = 451.0466 (theoretical, 451.0474). LC purity 100%;
tR = 3.242 min.
1
treated as in 6·HCl to give 7·HCl (52 mg, 100%) as a white solid. H
NMR (DMSO-d6 with puff of NH3 vapor): δ 7.84−7.78 (m, 2H), 7.75
(d, J = 9 Hz, 1H), 7.58 (m, 1H), 7.56 (d, J = 9 Hz, 1H), 7.52 (m, 1H),
6.37 (br s, 2H), 6.30 (d, J = 8.8 Hz, 1H), and 6.16 (s, 2H) ppm.
HRMS: measured m/z, [M + H]+ = 451.0577 (theoretical, 451.0584).
LC purity 99.7%; tR = 2.835 min.
4-(3-Bromo-5-chlorophenoxy)-5-chloro-2-methyl-1H-benzi-
midazole (32). A solution of 28 (204 mg, 0.586 mmol) in AcOH
(3.4 mL) was heated to 100 °C for 6 h. The solvent was removed in
vacuo, and the residue was partitioned between saturated aqueous
NaHCO3 (10 mL) and CH2Cl2 (20 mL). The organic extract was
dried over MgSO4, filtered, and the solvent was removed in vacuo.
This residue was purified by chromatography using silica gel (12 g),
eluting with 0−10% MeOH/CH2Cl2 to give 32 (217 mg, 99.5%) as a
4-(3-Bromo-5-chlorophenoxy)-5-chloro-1H-benzimidazole
(30). A solution of 28 (400 mg, 1.49 mmol) in 90% formic acid
(5 mL) was heated to 100 °C for 1 h. This mixture was concentrated
in vacuo and then partitioned between EtOAc (40 mL) and saturated
aqueous NaHCO3 (40 mL). The organic extract was dried over
MgSO4, filtered and the solvent removed in vacuo to give 30 (382 mg,
1
1
93%) as a white solid. H NMR (CDCl3): δ 8.02 (s, 1H), 7.5 (br s,
white solid. H NMR (CDCl3): δ 7.4 (br s, 1H), 7.33−7.28 (m, 2H),
1H), 7.40 (d, J = 8.5 Hz, 1H), 7.26 (m, 1H), 6.95 (m, 1H), and 6.83
(m, 1H) ppm. LRMS (M + 1) = 358.7.
7.15 (s, 1H), 6.85 (s, 1H), 6.74 (s, 1H), and 2.56 (s, 3H) ppm. LRMS
(M + 1) = 372.7.
3-Chloro-5-[(5-chloro-1H-benzimidazol-4-yl)oxy]-
benzonitrile (34). 34 was prepared as for 33 using 30 (430 mg,
3-Chloro-5-[(5-chloro-2-methyl-1H-benzimidazol-4-yl)oxy]-
benzonitrile (36). 36 was prepared as for 33 using 32 (217 mg,
1
1.201 mmol) to give 34 (278 mg, 76%) as a white solid. H NMR
1
0.583 mmol) to give 36 (129 mg, 69%) as a white solid. H NMR
(CD3OD): δ 8.22 (s, 1H), 7.65−7.50 (m, 2H), 7.49 (br s, 1H), 7.45 (d,
J = 8.7 Hz, 1H), and 7.15−7.10 (m, 2H) ppm. LRMS (M + 1) = 303.8.
3-Chloro-5-{[5-chloro-1-(1H-pyrazolo[3,4-b]pyridin-3-
ylmethyl)-1H-benzimidazol-4-yl]oxy}benzonitrile Dihydro-
chloride (8). 34 was treated with 20 as in the synthesis of 3 to
give the free base of 8. The solid was purified by chromatography
using silica gel, eluting with 0−10% MeOH/CH2Cl2. The pure
fractions were combined, treated with 1 N HCl to pH 3 and the
solvent was removed in vacuo to provide 8 (46 mg, 29%) as a white
solid. 1H NMR (DMSO-d6): δ 13.7 (s, 1H), 8.58 (s, 1H), 8.52 (d, J =
4.4 Hz, 1H), 8.16 (d, 8.0 Hz, 1H), 7.74 (dd, J = 1.5 Hz, 1H), 7.63 (d,
J = 8.8 Hz, 1H), 7.47 (d, J = 8.5 Hz, 1H), 7.39 (dd, J = 1.5 Hz, 1H),
7.29 (dd, J = 2 Hz, 1H), 7.19 (dd, J = 4.5 and 8.1 Hz, 1H), and 5.92 (s,
2H) ppm. HRMS: measured m/z, [M + H]+ = 435.0512 (theoretical,
435.0522). LC purity 96.5%; tR = 3.090 min.
4-(3-Bromo-5-chlorophenoxy)-5-chloro-1,3-dihydro-2H-
benzimidazol-2-one (31). To a solution of 28 (204 mg, 0.586
mmol) in DMF (3.5 mL) was added pyridine (119 μL, 1.465 mmol),
and the mixture was cooled over an ice bath. This solution was treated
with triphosgene (69.6 mg, 0.234 mmol) and stirred at 0 °C for 20 min.
The reaction was quenched with water (10 mL) and the mixture
acidified with 1 N HCl to pH 3. Extraction was with CH2Cl2 (30 mL)
with MeOH (at 1 mL to solubilize). The aqueous layer was further
extracted with CH2Cl2 (20 mL). The combined extracts were washed
with water (10 mL), dried over MgSO4, filtered, and the solvent was
removed in vacuo. This residue was preabsorbed onto silica gel (1.5 g)
and purified by chromatography using silica gel (12 g), eluting with
0−10% MeOH/CH2Cl2 to give 31 (147 mg, 67%) as a white solid. 1H
NMR (CDCl3): δ 7.40 (m, 2H), 7.22 (t, J = 1.7 Hz, 1H), 7.14 (d, J =
8.3 Hz, 1H), 6.93 (d, J = 8.3 Hz, 1H), 6.93 (t, J = 1.7 Hz, 1H), and
6.82 (t, J = 1.9 Hz, 1H) ppm. LRMS (M + 1) = 374.7.
3-Chloro-5-[(5-chloro-2-oxo-2,3-dihydro-1H-benzimidazol-
4-yl)oxy]benzonitrile (35). 35 was prepared as for 33 using 31 (147
mg, 0.393 mmol) to give 35 (905 mg, 72%) as a white solid. 1H NMR
(DMSO-d6): δ 11.15 (s, 1H), 10.0 (s, 1H), 7.76 (t, J = 1.5 Hz, 1H), 7.37
(dd, J = 1.2 and 2.4 Hz, 1H), 7.31 (t, J = 2 Hz, 1H), 7.14 (d, J = 8.3
Hz, 1H), and 6.91 (d, J = 8.3 Hz, 1H) ppm. LRMS (M + 1) = 339.0.
3-Chloro-5-{[5-chloro-2-oxo-1-(1H-pyrazolo[3,4-b]pyridin-3-
ylmethyl)-2,3-dihydro-1H-benzimidazol-4-yl]oxy}benzonitrile
(9). To a solution of 35 (83.9 mg, 0.262 mmol) and 20 (54.0 mg,
0.173 mmol) in DMF (2 mL) at 0 °C was added Cs2CO3 (85 mg,
0.262 mmol), and the mixture was stirred for 10 min over an ice bath.
Then the mixture was removed from the cooling bath and stirred for
an additional 1 h. LC−MS indicates mostly 35 and bis alkylated
product with only minor amounts of the 2 monoalkylated products.
The mixture was filtered through a Gelman Acrodisc and purified
using reverse phase chromatography. The fractions containing the
(DMSO-d6): δ 12.66 (s, 1H), 7.73 (t, J = 1.5 Hz, 1H), 7.43 (d, J = 8.5
Hz, 1H), 7.34 (m, 1H), 7.33 (d, J = 8.5 Hz, 1H), 7.22 (t, J = 1.8 Hz,
1H), and 2.5 (s, 3H) ppm. LRMS (M + 1) = 317.8.
3-Chloro-5-{[5-chloro-2-methyl-1-(1H-pyrazolo[3,4-b]-
pyridin-3-ylmethyl)-1H-benzimidazol-4-yl]oxy}benzonitrile tri-
fluoroacetate (10). A solution of 36 (104 mg, 0.327 mmol) and
20 (102 mg, 0.327 mmol) in DMF (3 mL) was stirred over an ice bath
for 5 min and then treated with Cs2CO3 (107 mg, 0.327 mmol) and
then stirred at room temperature for 1 h. The mixture was filtered
through Gelman Acrodisc and purified using reverse phase
chromatography. The desired fractions were combined and lyophi-
lized. This solid was deprotected in TFA (3 mL) for 5 min, and the
solvent was removed in vacuo. This residue was again purified using
reverse phase chromatography to give 10 (79.2 mg, 43%) as a white
solid. 1H NMR (DMSO-d6): δ 13.75 (s, 1H), 8.52 (dd, J = 1.5 and 4.5
Hz, 1H), 8.02 (d, J = 7.8 Hz, 1H), 7.74 (t, J = 1.5 Hz, 1H), 7.69 (d, J =
8.8 Hz, 1H), 7.42 (d, J = 8.5 Hz, 1H), 7.37 (dd, J = 1.2 and 2.5 Hz,
1H), 7.25 (t, J = 4.5 and 8.2 Hz, 1H), 5.87 (s, 2H), and 2.59 (s, 3H)
ppm. HRMS: measured m/z, [M + H]+ = 449.0671 (theoretical,
449.0679). LC purity 100%; tR = 2.801 min.
3-Chloro-5-[3,5-dichloro-1H-indazol-4-yl)oxy]benzonitrile. To
a solution of 19 (100 mg, 0.329 mmol) in DMF (1 mL) were added
1.0 M KOtBu in THF (329 μL, 0.329 mmol) and then NCS (44 mg,
0.329 mmol). The resulting mixture was stirred for 10 min and then
purified by reverse phase chromatography to give the title compound
(42 mg, 37.8%) as a white solid. 1H NMR (DMSO-d6): δ 10.1 (s, 1H),
7.54 (d, J = 9 Hz, 1H), 7.40 (d, J = 9 Hz, 1H), 7.35 (m, 1H), 7.15 (m,
1H), and 6.95 (m, 1H) ppm. LRMS (M + 1) = 337.9.
3-Chloro-5-{[3,5-dichloro-1-(1H-pyrazolo[3,4-b]pyridin-3-
ylmethyl)-1H-indazol-4-yl]oxy}benzonitrile (11). 11 was pre-
pared as for 8 using 3-chloro-5-[3,5-dichloro-1H-indazol-4-yl)oxy]-
benzonitrile (42 mg, 0.124 mmol) to give 11 (44 mg, 76%) as a white
1
solid. H NMR (CDCl3): δ 8.71 (dd, J = 1.7 and 4.6 Hz, 1H), 8.01
(dd, J = 1.7 and 7.9 Hz, 1H), 7.53 (d, J = 9.0 Hz, 1H), 7.48 (d, J = 9.0
Hz, 1H), 7.33 (m, 1H), 7.25 (m, 1H), 7.10 (m, 1H), 6.91 (m, 1H),
and 5.89 (s, 2H) ppm. HRMS: measured m/z, [M + H]+ = 469.0122
(theoretical, 469.0133). LC purity 99.5%; tR = 4.111 min.
4,5-Dichloro-N-(methoxybenzyl)-2-nitroaniline (37). 1,2-Dichloro-
4,5-dinitrobenzene (1,86 g, 7.85 mmol) and 4-methoxybenzylamine (2.051
mL, 15.7 mmol) were heated in toluene (10 mL) to
90 °C for 1 h. This mixture was diluted with water (100 mL), and the
aqueous layer was extracted with EtOAc (200 mL). The extract was
washed with water (75 mL), dried over MgSO4, filtered, treated with silica
gel (10 g), and the solvent was removed in vacuo. This preabsorbed solid
was purified by chromatography using silica gel (40 g), eluting with
40% CH2Cl2/hexanes to provide 37 (2.545 g, 99%) as a red crystalline
7931
dx.doi.org/10.1021/jm2010173|J. Med. Chem. 2011, 54, 7920−7933

Journal of Medicinal Chemistry
Article
solid. 1H NMR (CDCl3): δ 8.30 (s, 1H), 8.24 (br s, 1H), 7.25 (d, J =
8.6 Hz, 2H), 6.98 (s, 1H), 6.92 (d, J = 8.6 Hz, 2H), 4.43 (d, J = 5 Hz,
2H), and 3.83 (s, 3H) ppm.
treated with saturated aqueous Na2CO3 (20 mL) and stirred for
30 min. The resulting mixture was partitioned, and the aqueous
portion was back-extracted with EtOAc (40 mL). The combined
extracts were washed with water (20 mL), dried over MgSO4, filtered,
and the solvent was removed in vacuo. This residue was purified
by chromatography using silica gel (4 g), eluting with 40−100%
5-(3-Bromo-5-chlorophenoxy)-4-chloro-N-(4-methoxyben-
zyl)-2-nitroaniline (38). To a suspension of 37 (2.50 g, 7.64 mmol)
in NMP (10 mL) was added 3-bromo-5-chlorophenol (3.17 g, 15.28
mmol) and K2CO3 (2.112 g, 15.28 mmol). The mixture was heated to
100 °C for 2 h. This mixture was diluted with water (400 mL), and the
aqueous layer was extracted with EtOAc (2 × 400 mL). The combined
extracts were further washed with water (200 mL) and then brine (100 mL).
This solution was dried over MgSO4, filtered and the solvent removed
in vacuo. This residue was preabsorbed onto silica gel (13 g) and
purified by chromatography using silica gel (80 g), eluting with 10−
1
EtOAc/CH2Cl2 to provide 14 (8.9 mg, 40%) as a solid. H NMR
(CDCl3): δ 8.57 (d, J = 4.6 Hz, 1H), 8.11 (d, J = 8 Hz, 1H), 7.55−7.50
(m, 1H), 7.26 (m, 1H), 7.16 (dd, J = 4.6 and 8 Hz, 1H), 7.08 (t, J = 2
Hz, 1H), 6.96 (br t, 1H), 6.83 (s, 1H), 6.57 (m, 1H), and 4.63 (s, 2H)
ppm. HRMS: measured m/z, [M + H]+ = 425.0679 (theoretical,
425.0679). LC purity 100%; tR = 2.926 min.
3-Chloro-5-{[5-chloro-1-(1H-pyrazolo[3,4-b]pyridine-3-
ylmethyl)-1H-1,2,3-benzotriazol-6-yl]oxy}benzonitrile (12). A
solution of 14 (0.96 mg, 0.0023 mmol) in AcOH (200 μL) was
treated with a solution of NaNO2 (0.23 mg, 0.0034 mmol) in water
(200 μL), and the mixture was stirred for 10 min. The solvent was
removed in vacuo and the residue was purified by reverse phase
chromatography and lyophilized to give 12 (0.95 mg, 96%) as a solid.
1H NMR (CD3OD): δ 8.50 (d, J = 4.5 Hz, 1H), 8.24 (s, 1H), 8.14 (d,
1
100% CH2Cl2/hexanes to give 38 (2.814 g, 74%) as a solid. H NMR
(CDCl3): δ 8.49 (m, 1H), 8.33 (s, 1H), 7.39 (t, J = 1.6 Hz, 1H), 7.07
(d, J = 8.8 Hz, 2H), 7.03 (t, J = 2 Hz, 1H), 6.91 (t, J = 2 Hz, 1H), 6.85
(d, J = 8.8 Hz, 2H), 6.17 (s, 1H), 4.29 (d, J = 5.5 Hz, 2H), and 3.81 (s,
3H) ppm.
5-(3-Bromo-5-chlorophenoxy)-4-chloro-2-nitroaniline (39). A
solution of 38 (2.814 g, 5.63 mmol) in TFA (20 mL) was heated to
reflux for 1 h. The solvent was removed in vacuo, and the residue
was partitioned between CH2Cl2 (2 × 100 mL) and aqueous Na2CO3
(50 mL). This extract was concentrated in vacuo and the residue was
purified by chromatography using silica gel (40 g), eluting with 10−
100% CH2Cl2/hexanes to give 39 (2.125 g, 100%) as a solid. 1H NMR
(CDCl3): δ 8.29 (s, 1H), 7.40 (m, 1H), 7.13 (m, 1H), 7.03 (m, 1H),
6.21 (s, 1H), and 6.12 (br s, 2H) ppm.
J = 8 Hz, 1H), 7.60 (m, 1H), 7.58 (s, 1H), 7.29 (m, 1H), 7.27 (m,
1H), 7.19 (dd, J = 4.5 and 8 Hz, 1H), and 6.27 (s, 2H) ppm. HRMS:
measured m/z, [M + H]+ = 436.0466 (theoretical, 436.0475). LC
purity 100%; tR = 3.474 min.
3-Chloro-5-{[5-chloro-1-(1H-pyrazolo[3,4-b]pyridine-3-
ylmethyl)-1H-benzimidazol-6-yl]oxy}benzonitrile trifluoroace-
tate (13). A solution of 14 (0.96 mg, 0.0023 mmol) in formic acid
(500 μL) was heated to 100 °C for 1 h. This mixture was evaporated
in vacuo and the residue was purified by reverse phase chromatog-
5-(3-Bromo-5-chlorophenoxy)-4-chloro-2-nitro-N-(1H-
pyrazolo[3,4-b]pyridin-3-ylmethyl)aniline (40). To a solution
of 39 (81 mg, 0.214 mmol) and 20 (66.9 mg, 0.214 mmol) in
DMF (500 μL) was added cesium carbonate (69.8 mg, 0.214 mmol),
and the mixture was stirred at room temperature for 4 days. The
resulting mixture was partitioned between EtOAc (2 × 10 mL) and
water (10 mL). The combined extracts were washed with water
(5 mL), dried over MgSO4, filtered, and the solvent was removed in
vacuo. This residue was further dried under high vacuum for 24 h to
remove excess DMF and then redissolved in TFA (4 mL). This
solution was stirred for 30 min, and then the solvent was removed in
vacuo. This residue was partitioned between EtOAc (2 × 10 mL) and
aqueous NaHCO3 (10 mL). The combined extracts were concentrated
in vacuo and then purified by chromatography using silica gel (12 g),
eluting with 0−40% EtOAc/CH2Cl2 to provide recovered 39 (35 mg,
43%), bis alkylated product (21.2 mg, 15.6%), and 40 (35.3 mg,
1
raphy and lyophilized to give 13 (1.05 mg, 85%) as a solid. H NMR
(DMSO-d6): δ 13.68 (s, 1H), 8.74 (s, 1H), 8.51 (d, J = 4,4 Hz, 1H),
8.16 (d, J = 8.2 Hz, 1H), 7.98 (s, 1H), 7.78 (t, J = 1.6 Hz, 1H), 7.74 (s,
1H), 7.42 (m, 1H), 7.33 (t, J = 2.1 Hz, 1H), 7.17 (dd, J = 4.5 and 8.1
Hz, 1H), and 5.89 (s, 2H) ppm. HRMS: measured m/z, [M + H]+ =
435.0513 (theoretical, 435.0522). LC purity 99%; tR = 2.800 min.
AUTHOR INFORMATION
■
Corresponding Author
*Phone: 215-652-4715. Fax: 215-652-3971. E-mail: Robert_
gomez@merck.com.
1
Present Addresses
32.4%) as a solid. H NMR (CDCl3): δ 10.58 (br s, 1H), 8.77 (m,
●Department of Boston Discovery Chemistry, Merck Research
Labs., 33 Avenue Louis Pasteur, Boston, Massachusetts 02115-
5717, United States.
1H), 8.56 (d, J = 3.1 Hz, 1H), 8.33 (s, 1H), 8.08 (d, J = 7.5 Hz, 1H),
7.42 (t, J = 1.7 Hz, 1H), 7.18 (dd, J = 4.6 and 8.0 Hz, 1H), 7.04 (t, J =
2 Hz, 1H), 6.95 (t, J = 2 Hz, 1H), 6.52 (s, 1H), and 4.73 (s, 2H) ppm.
LRMS (M + 1) = 509.3.
3-Chloro-5-{2-chloro-4-nitro-5-[(1H-pyrazolo[3,4-b]pyridin-
3-ylmethyl)amino]phenoxy}benzonitrile trifluoroacetate. To
a solution of 40 (35 mg, 0.069 mmol) in DMF (500 μL) was added
palladium tetrakistriphenylphosphine (23.8 mg, 0.021 mmol) and
Zn(CN)2 (8.07 mg 0.069 mmol). The mixture was heated to 95 °C for
2 h. The reaction was monitored by LC−MS. To the mixture were
added additional palladium tetrakistriphenylphosphine (23.8 mg, 0.021
mmol) and Zn(CN)2 (8.07 mg, 0.069 mmol). The mixture was heated
to 95 °C for 1 h. The mixture was filtered through a Gelman Acrodisc
and purified using reverse phase chromatography to provide recovered
40 (10 mg, 23%) and the title compound (29.6 mg, 76%) as a solid.
1H NMR (CDCl3): δ 8.75 (m, 1H0, 8.53 (d, J = 4.4 Hz, 1H), 8.36 (s,
1H), 8.31 (d, J = 7 Hz, 1H), 7.53 (s, 1H), 7.32 (dd, J = 4.9 and 8.0 Hz,
1H), 7.21 (t, J = 2 Hz, 1H), 7.14 (s, 1H), 6.22 (s, 1H), and 4.80 (d, J =
5.9 Hz, 2H) ppm.
3-{4-Amino-2-chloro-5-[(1H-pyrazolo[3,4-b]pyridin-3-
ylmethyl)amino]phenoxy}-5-chlorobenzonitrile (14). A suspen-
sion of 3-chloro-5-{2-chloro-4-nitro-5-[(1H-pyrazolo[3,4-b]pyridin-3-
ylmethyl)amino]phenoxy}benzonitrile trifluoroacetate (29.6 mg 0.052
mmol) and SnCl2 (105 mg, 0.466 mmol) in MeOH (3 mL) was
heated to reflux for 1 h. The resulting mixture was allowed to cool to
room temperature and diluted with EtOAc (40 mL). This mixture was
ACKNOWLEDGMENTS
■
The authors thank Dr. Steve Young for his guidance and support
and Elizabeth Cauchon, Manuel Chan, Wanda Cromlish, and
Sebastian Guiral for their assistance in acquiring necessary data for
publication.
ABBREVIATIONS USED
■
HAART, highly active antiretroviral therapy; HIV, human
immunodeficiency virus; AIDS, acquired immunodeficiency
syndrome; NNRTI, non-nucleoside reverse transcriptase
inhibitor; wt, wild type; RT, reverse transcriptase; NRTI,
nucleoside reverse transcriptase inhibitor; ΔE, change in
energy; CIC95, cell inhibitory concentration that blocks 95%
of cell replication; FBS, fetal bovine serum; NHS, normal
human serum; EFV, efavirenz; iv, intravenous; nAUC,
normalized area under the plasma concentration vs time
curve; Vd, volume of distribution; Clp, plasma clearance; t1/2
half-life; F, bioavailability
,
7932
dx.doi.org/10.1021/jm2010173|J. Med. Chem. 2011, 54, 7920−7933

Journal of Medicinal Chemistry
Article
(16) De Luca, S.; Saviano, M.; Della Moglie, R.; Digilio, G.; Bracco,
C.; Aloj, L.; Tarallo, L.; Pedone, C.; Morelli, G. Conformationally
constrained CCK8 analogues obtained from a rationally designed
peptide library as ligands for cholecystokinin type B receptor.
ChemMedChem. 2006, 1 (9), 997−1006.
(17) Hruby, V. J.; Al-Obeidi, F.; Kazmierski, W. M. Emerging
approaches in the molecular design of receptor selective peptide
ligands: conformational, topographical and dynamic considerations.
Biochem. J. 1990, 268, 249−262.
(18) The low mode/mixed-torsional sampling algorithm as
implemented in software from Schrodinger, LLC, version 2009.1,
was used with a distance dependent dielectric constant of 2 with the
MMFFs force field to determine the lowest energy conformer. The
protein−ligand complex was minimized using the same force field and
parameters.
(19) Full experimental protocols for the HIV-1 RT polymerase SPA
assay are described in the following patent application: Saggar, S. A.;
Sisko, J. T.; Tucker, T. J.; Tynebor, R. M.; Su, D.-S.; Anthony, N. J.
Preparation of Heterocyclic Phenoxy Compounds as HIV Reverse-
Transcriptase Inhibitors. U.S. Patent Appl. US 2007/021442 A1, 2007.
(20) Vacca, J. P.; Dorsey, B. D.; Schleif, W. A.; Levin, R. B.;
McDaniel, S. L.; Darke, P. D.; Zugay, J.; Quintero, J. C.; Blahy, O. M.;
Roth, E.; Sardana, V. V.; Schlabac, A. J.; Graham, P. I.; Condra, J. H.;
Gotlib, L.; Holloway, M. K.; Lin, J.; Chen, I.-W.; Vastag, K.; Ostovic,
D.; Anderson, P. S.; Emini, E. E.; Huff, J. R. L-735,524: an orally
bioavailable human immunodeficiency virus type 1 protease inhibitor.
Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 4096−4100.
REFERENCES
■
(1) Coffin, J. M. Genetic variation in AIDS viruses. Cell 1986, 46 (1),
1−4.
(2) Perrino, F. W.; Preston, B. D.; Sandell, L. L.; Loeb, L. A.
Extension of mismatched 3′ termini of DNA is a major determinant of
the infidelity of human immunodeficiency virus type 1 reverse
transcriptase. Proc. Natl. Acad. Sci. U.S.A. 1989, 86 (21), 8343−8347.
(3) Menendez-Arias, L. Molecular basis of fidelity of DNA synthesis
and nucleotide specificity of retroviral reverse transcriptases. Prog.
Nucleic Acid Res. Mol. Biol. 2002, 71, 91−147.
(4) Gulick, R. M.; Mellors, J. W.; Havlir, D; Eron, J. J.; Meibohm, A.;
Condra, J. H.; Valentine, F. T.; McMahon, D.; Gonzalez, C.; Jonas, L.;
Emini, E. A.; Chodakewitz, J. A.; Isaacs, R.; Richman, D. D. 3-Year
suppression of HIV viremia with indinavir, zidovudine, and
lamivudine. Ann. Intern. Med. 2000, 133 (1), 35−39.
(5) Cane, P. A. New developments in HIV drug resistance. J. Antimicrob.
Chemother. 2009, 64 (Suppl. 1), i37−i40.
(6) Little, S. J.; Holte, S.; Routy, J.-P.; Daar, E. S.; Markowitz, M.;
Collier, A. C.; Koup, R. A.; Mellors, J. W.; Connick, E.; Conway, B.;
Kilby, M.; Wang, L.; Whitcomb, J. M.; Hellmann, N. S.; Richman,
D. D. Antiretroviral-drug resistance among patients recently infected
with HIV. N. Engl. J. Med. 2002, 347 (6), 385−394.
(7) Richman, D. D.; Morton, S. C.; Wrin, T.; Hellmann, N.; Berry,
S.; Shapiro, M. F.; Bozzette, S. A. The prevalence of antiretroviral drug
resistance in the United States. AIDS (London) 2004, 18 (10), 1393−
1401.
(8) de Bethune, M.-P. Non-nucleoside reverse transcriptase
inhibitors (NNRTIs), their discovery, development, and use in the
treatment of HIV-1 infection: a review of the last 20 years (1989−
2009). Antiviral Res. 2010, 85 (1), 75−90.
(9) Sluis-Cremer, N.; Temiz, N. A.; Bahar, I. Conformational
changes in HIV-1 reverse transcriptase induced by nonnucleoside
reverse transcriptase inhibitor binding. Curr. HIV Res. 2004, 2 (4),
323−232.
(10) Fokunang, C. N.; Hitchcock, J.; Spence, F.; Tembe-Fokunang,
E. A.; Burkhardt, J.; Levy, L.; George, C. An overview of mitochondrial
toxicity of nucleoside reverse transcriptase inhibitors associated with
HIV therapy. Int. J. Pharmacol. 2006, 2 (1), 152−162.
(11) Cheung, P. K; Wynhoven, B.; Harrigan, P. R. Which HIV-1
drug resistance mutations are common in clinical practice? AIDS Rev.
2004, 6 (2), 107−116.
(21) The X-ray diffraction data of the wt RT and inhibitor 5 complex
crystal (Figure 4) were obtained at 2.6 Å resolution and refined to an
R-factor of 0.199 and an Rfree of 0.248. The X-ray diffraction data of the
K103N mutant RT and inhibitor 5 complex crystal were obtained at
2.4 Å and refined to an R-factor of 0.207 and an Rfree of 0.250. Both
structures are deposited in the Protein Data Bank: 3T19 (wild type)
and 3T1A (K103N mutant).
(22) Sweeney, Z. H.; Harris, S. F.; Arora, N; Javanbakht, H.; Li, Y.;
Fretland, J.; Davidson, J. P.; Billedeau, J. R.; Gleason, S. F.; Hirschfeld,
D.; Kennedy-Smith, J. J.; Mirzadegan, T.; Roetz, R.; Smith, M.; Sperry,
S.; Suh, J.; Wu, J.; Tsing, S.; Villasenor, A. G.; Paul, A.; Su, G.; Heilek,
̃
G.; Hang, J. Q.; Zhou, A. S.; Jernelius, J. A.; Zhang, F.-J.; Klumpp, K.
Design of annulated pyrazoles as inhibitors of HIV-1 reverse
transcriptase. J. Med. Chem. 2008, 51 (23), 7449−7458.
(23) Jones, L.; Allan., G.; Barba, O.; Burt., C.; Corbau, R.; Dupont,
(12) Tambuyzer, L.; Azijn, H.; Rimsky, L. T.; Vingerhoets, J.;
Lecocq, P.; Kraus, G.; Picchio, G.; de Bethune, M.-P. Compilation and
prevalence of mutations associated with resistance to non-nucleoside
reverse transcriptase inhibitors. Antiviral Ther. 2009, 14 (1), 103−109.
(13) Das, K.; Clark, A. D. Jr.; Lewi, P. J.; Heeres, J.; De Jonge, M. R.;
Koymans, L. M. H.; Vinkers, H. M.; Daeyaert, F.; Ludovici, D. W.;
Kukla, M. J.; De Corte, B.; Kavash, R. W.; Ho, C. Y.; Ye, H.;
Lichtenstein, M. A.; Andries, K.; Pauwels, R.; De Bethune, M.-P.;
Boyer, P. L.; Clark, P.; Hughes, S. H.; Janssen, P. A. J.; Arnold, E. Roles
of conformational and positional adaptability in structure-based design
of TMC125-R165335 (etravirine) and related non-nucleoside reverse
transcriptase inhibitors that are highly potent and effective against
wild-type and drug-resistant HIV-1 variants. J. Med. Chem. 2004, 47
(10), 2550−2560.
(14) Arasteh, K.; Riege, R. A.; Yeni, P.; Pozniak, A.; Boogaerts, G.;
van Heeswijk, R.; de Bethune, M.-P.; Peeters, M.; Woodfall, B. Short-
term randomized proof-of-principle trial of TMC278 in patients with
HIV type-1 who have previously failed antiretroviral therapy. Antiviral
Ther. 2009, 14 (5), 713−722.
(15) Tucker, T. J.; Sisko, J. T.; Tynebor, R. M.; Williams, T. M.;
Felock, P. J.; Flynn, J. A.; Lai, M.-T.; Liang, Y.; McGaughey, G.; Liu,
M.; Miller, M.; Moyer, G.; Munshi, V.; Perlow-Poehnelt, R.; Prasad, S.;
Reid, J. C.; Sanchez, R.; Torrent, M.; Vacca, J. P.; Wan, B.-L.; Yan, Y.
Discovery of 3-{5-[(6-amino-1H-pyrazolo[3,4-b]pyridine-3-yl)-
methoxy]-2-chlorophenoxy}-5-chlorobenzonitrile (MK-4965): a po-
tent, orally bioavailable HIV-1 non-nucleoside reverse transcriptase
inhibitor with improved potency against key mutant viruses. J. Med.
Chem. 2008, 51 (20), 6503−6511.
T.; Knochel.; Irving, S.; Middleton, D. S.; Mowbray, C. E.; Perros, M.;
̈
Ringrose, H.; Swain, N. A.; Webster, R.; Westby, M.; Phillips, C. Novel
indazole non-nucleoside reverse transcriptase inhibitors using
molecular hybridization based on crystallographic overlays. J. Med.
Chem. 2009, 52 (4), 1219−1223.
(24) The compounds were evaluated in a Monogram Bioscience
Phenosense assay versus a panel of clinically derived RT mutants in
the presence of 10% FBS. The IC50 is defined as the concentration of
compound in cell culture required to block 50% of viral replication.
Values are derived from the average of two determinations. Assays
were performed by Monogram Biosciences, South San Francisco, CA.
Details of the assay protocols are available at http://www.
monogramhiv.com/phenosense_introduction.aspx.
(25) Dykes, C.; Demeter, L. M. Y188H, Which Is a Potential
Intermediate between Wild-Type (WT) and Y188L, Has Markedly
Reduced Replication Efficiency. Presented at the 14th Conference on
Retroviruses and Opportunistic Infections, Los Angeles, CA, Feb 25−
28, 2007; Poster 633.
(26) pKa prediction using ACD/pKa DB from Phys/Chem Suite by
Advanced Chemical Development, Inc. The base compound, N-
methylbenzimidazole, has a pKa of 5.40. Substituent effects using the
Dewar-Grisdale method is calculated at −1.733, giving 8 a predicted
pKa of 3.667.
(27) Grzegozek, M. Vicarious nucleophilic amination of nitroquino-
lines by 1,1,1-trimethylhydrazinium iodide. J. Heterocycl. Chem. 2008,
45 (6), 1879−1882.
7933
dx.doi.org/10.1021/jm2010173|J. Med. Chem. 2011, 54, 7920−7933