Design, structure-activity relationship, and pharmacokinetic profile of pyrazole-based indoline factor Xa inhibitors

Article abstract of DOI:10.1016/j.bmcl.2007.09.091

A new series of pyrazole-based factor Xa inhibitors have been identified as part of our ongoing efforts to optimize previously reported clinical candidate razaxaban. Concern over the possible formation of primary aniline metabolites via amide hydrolysis led to the replacement of the primary amide linker between the pyrazole and phenyl moieties with secondary amides. This was accomplished by replacing the aniline with a variety of heterobicycles, of which indolines were the most potent. The indoline series demonstrated subnanomolar factor Xa binding K_is, modest to high selectivity versus other serine proteases, and good in vitro clotting activity. A small number of indoline fXa inhibitors were profiled in a dog pharmacokinetic model, one of which demonstrated pharmacokinetic parameters similar to that of clinical candidate razaxaban.

Full text of DOI:10.1016/j.bmcl.2007.09.091

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 17 (2007) 6481–6488
Design, structure–activity relationship, and pharmacokinetic
profile of pyrazole-based indoline factor Xa inhibitors
Jeffrey G. Varnes, Dean A. Wacker,* Irina C. Jacobson, Mimi L. Quan,
Christopher D. Ellis, Karen A. Rossi, Ming Y. He, Joseph M. Luettgen, Robert M. Knabb,
Steven Bai, Kan He, Patrick Y. S. Lam and Ruth R. Wexler
Bristol-Myers Squibb Pharmaceutical Research Institute, PO Box 5400 Princeton, NJ 08542-5400, USA
Received 17 August 2007; revised 26 September 2007; accepted 27 September 2007
Available online 1 October 2007
Abstract—A new series of pyrazole-based factor Xa inhibitors have been identified as part of our ongoing efforts to optimize pre-
viously reported clinical candidate razaxaban. Concern over the possible formation of primary aniline metabolites via amide hydro-
lysis led to the replacement of the primary amide linker between the pyrazole and phenyl moieties with secondary amides. This was
accomplished by replacing the aniline with a variety of heterobicycles, of which indolines were the most potent. The indoline series
demonstrated subnanomolar factor Xa binding Kis, modest to high selectivity versus other serine proteases, and good in vitro clot-
ting activity. A small number of indoline fXa inhibitors were profiled in a dog pharmacokinetic model, one of which demonstrated
pharmacokinetic parameters similar to that of clinical candidate razaxaban.
ꢀ 2007 Elsevier Ltd. All rights reserved.
The trypsin-like serine protease fXa is an attractive tar-
get for the development of new anticoagulant drugs.1 As
a key enzyme in the coagulation cascade, fXa combines
with fVa and calcium to form the prothrombinase com-
plex responsible for the conversion of prothrombin to
thrombin. Thrombin plays a critical role in clot forma-
tion by catalyzing the conversion of fibrinogen to fibrin
and by activating platelets.2 Due to the nature of the
coagulation cascade, small quantities of fXa can pro-
duce large amounts of thrombin. Therefore, fXa inhibi-
tion is expected to be more efficient than direct
inhibition of thrombin, and in vivo studies support a
lower risk of bleeding with fXa inhibitors.3,4
phenyl moieties. While the biarylaniline P4 moiety of
razaxaban tested negative in the Ames assay and was
not observed in vivo,6 we considered it advantageous to
eliminate possible formation of a primary aniline alto-
gether. In practice, the carboxamido linker was converted
to a heterobicyclic moiety. Hydrolysis, if it should occur,
would then yield a secondary aniline, which is less likely to
be mutagenic. It was hoped that this modification would
provide anticoagulants with both potency and selectivity
for factor Xa and acceptable pharmacokinetic profiles for
in vivo testing.
F3C
R1
F
N
H
N
N
R2
N
N
We have described several fXa clinical candidates with
excellent potency, selectivity, and oral efficacy in vivo.5,6
Herein we describe a new series of factor Xa inhibitors
developed as part of our overall optimization strategy.
This series evolved from the P1-amino-benzisoxazole pyr-
azole-based scaffold of the previously reported clinical
candidate razaxaban through modification of the carbox-
amido linker that connects the pyrazole and proximal
N
N
N
N
O
O
NH2
NH2
O
O
N
N
razaxaban (BMS-561389, DPC906)
FXa Ki = 0.19 nM (PT IC2x = 1.3 μM)
Thrombin Ki = 540 nM
Trypsin Ki >10,000 nM
Initial leads for the heterobicyclic series are disclosed in
Table 1. The most potent heterobicycles were indolines 2
and 3, which were 16- and 30-fold less active, respec-
Keywords: Factor Xa; Coagulation.
*
Corresponding author. Tel.: +1 609 818 5381; e-mail: dean.wacker@
bms.com
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.09.091

6482
J. G. Varnes et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6481–6488
Table 1. Factor Xa, thrombin, and trypsin binding affinities for heterobicyclic scaffoldsa
F3C
N
R
N
N
O
NH2
O
N
Compound
Scaffold
R
fXa Ki, nM
Thrombin Ki, nM
Trypsin Ki, nM
PT IC2x lM
F
H
N
1
SO2CH3
0.09
6300
>5200
—
N
N
2
3
4
SO2CH3
SO2NH2
SO2CH3
1.5
2.7
>21,000
>21,000
>6300
>2500
>2500
>4200
26
28
—
2000
N
N
O
5
SO2CH3
500
>21,000
>4200
—
OH
6b
SO2CH3
SO2CH3
SO2NH2
1400
>6300
>21,000
>6300
>4200
—
—
—
N
N
O
7
71
—
8
3400
>4200
N
a Kis were obtained from purified human enzymes. Kis and PT values were measured according to Ref. 5a–c.
b Racemic.
tively, than 1, the P4-phenylmethylsulfone analog of
razaxaban. Both compounds also had relatively weak
in vitro clotting activity in the prothrombin time (PT)
assay. Modest to high selectivity for factor Xa over
thrombin and trypsin was observed for all compounds.
Increases in ring size (4, 8), oxidation (5, 6), or incorpo-
ration of a heteroatom (7) decreased fXa binding affinity
by 50-fold or more compared to 2 and 3.
ment of the 3-trifluoromethyl group of the pyrazole func-
tionality with a C-linked primary amide increased
binding affinity for fXa (Table 2). Compounds that also
contained P4 benzylic amines were very potent in the PT
assay, likely due to decreased protein binding as seen
previously with P4 alkyl amines and 3-amidopyrazole
analogs by our group.5c,7,8 For example, replacing the
3-trifluoro-methylpyrazole of 2 with a 3-amidopyrazole
(9) afforded a modest increase in fXa binding but de-
creased in vitro clotting activity. Making the same
change to N,N-dimethylbenzylamine 10 to give 11 re-
sulted in a 19-fold increase in binding affinity. Compared
to sulfone 9, benzylamine 11 was approximately 14-fold
To improve fXa binding affinity and in vitro clotting
activity of the indoline series, the structure–activity rela-
tionships of the pyrazole substitutent (Table 2) and the
indoline P4 group (Tables 2–4) were explored. Replace-

J. G. Varnes et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6481–6488
Table 2. Amide and trifluoromethyl pyrazole substitutentsa
6483
R1
N
R2
N
N
O
NH2
O
N
Compound
R1
R2
fXa Ki, nM
Thrombin Ki, nM
Trypsin Ki, nM
PT IC2x lM
SO2CH3
SO2CH3
2
CF3
1.5
>21,000
>2500
28
41
9
CONH2
0.82
>20,000
19,200
—
—
N
N
10
CF3
19
—
3.0
—
11
12
CONH2
1.0
4.9
—
>20,000
>40,000
N
N
CF3
15,000
13
CONH2
0.63
—
—
2.1
a Kis were obtained from purified human enzymes. Kis and PT values were measured according to Ref. 5a–c.
more efficacious in the clotting assay. Similarly, pyrroli-
dine 13 was 8-fold more potent than 12 and showed
activity in the PT assay comparable to that of 11.
pared and is thought to be due to a preferred orientation
of the P4 group when bound in the aryl S4 binding
pocket of the fXa enzyme.
Several indolines with P4 substituents at either the R2 or
R3 position are shown in Table 3. Comparing the fXa
binding affinities of 9 and 14 shows a marked preference
(>1700-fold) for the P4 group to be in the R2 position.
While there is also a decrease in binding affinity in going
from glycinamide 15 to 16, the effect is significantly less
pronounced, possibly because the P4 moiety of 16 has a
greater degree of flexibility than that of 14. Comparing
the regioisomers of the moderately rigid e-lactam P4-
analogs (17 and 18) also indicates a preference for R2
substitution by approximately 70-fold. This trend was
found to hold for every pair of R2 and R3 isomers pre-
The P4 substitutent of the aminobenzisoxazole-3-amido-
pyrazole indoline scaffold was diversified to further im-
prove the efficacy of our anticoagulants (Table 4).
These substituents were selected because they have
historically provided improved in vitro potency, better
protein binding, and favorable pharmacokinetic proper-
ties.5–8 Replacement of the dimethylbenzylamine of 11
with the dimethylamino-methyl imidazole of razaxaban
decreased fXa binding affinity by approximately twofold
(compound 19). This decrease was more pronounced
upon conversion of the dimethylamine to either the
N-methylacetamide (20) or N-methylsulfonamide (21).

6484
J. G. Varnes et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6481–6488
Table 3. Effect of indoline substitution on fXa binding affinitya
lactam 22 was marginally bioavailable, which is partially
attributable to a significantly lower rate of clearance
(0.22 L/h/kg) than that of either 12 or 13. Coupled with
a low volume of distribution, tetrahydropyridone 22 was
found to have a half-life of 1.6 h. The earlier lead, sul-
fonamide 3, which was equipotent with 22 in terms of
fXa binding affinity, exhibited a pharmacokinetic profile
similar to that of razaxaban. Unfortunately, the sulfon-
amide had low activity in the clotting assay, suggesting a
low free fraction in plasma, and was not further evalu-
ated in in vivo models.
O
H2N
R2
N
N
N
O
R3
NH2
O
N
Compound
R2
R3
H
fXa Ki, nM
The general synthetic strategy for preparing compounds
from the indoline series is shown in Scheme 1. Com-
pounds containing 3-trifluoromethylpyrazoles were pre-
pared from acid chloride 245c by treatment with the
sodium salt of the appropriate heterobicycle. Com-
pounds generated from 3-amidopyrazole 25 were simi-
larly prepared upon treatment of indole and acid with
1,3-diisopropylcarbodiimide in 4:1 pyridine/N,N-dimeth-
ylformamide for 24 h. Conversion to the P1-aminoben-
zisoxazole was carried out using acetohydroxamic acid
and potassium carbonate in DMF.5c At this stage, com-
pounds containing P4-thioanisole moieties were oxidized
to the sulfone using m-chloroperoxybenzoic acid in
dichloromethane. Similarly, compounds containing P4-
tert-butyl-sulfonamides were deprotected using trifluoro-
acetic acid in dichloromethane.
SO2CH3
9
0.82
SO2CH3
14
15
16
17
18
H
1400
6.8
N
H
N
O
N
H
H
350
15
N
O
O
H
N
O
1100
N
Amidopyrazole 25 was prepared as shown in Scheme 2
starting with a lithium hexamethyldisilazide-mediated
addition of commercially available ethanedioic acid
bis(1,1-dimethylethyl) ester (26) to 2-acetylfuran (27).
The resulting lithium salt was then reacted with hydra-
zine 285c in situ to afford furan 29. Ester hydrolysis un-
der acidic conditions, conversion to the acid chloride,
and treatment with 0.5 M ammonia in dioxane afforded
the corresponding 3-amidopyrazole that was then oxi-
dized with KMnO4 to provide acid 25.
a Kis were obtained from purified human enzymes and were measured
according to Ref. 5a–c.
Of the P4 N-linked moieties examined, lactam 17 (Table 3)
was the least active, possibly due to the ring size exceed-
ing the capacity of the S4 binding pocket. However,
incorporation of the lactam P4 moiety found in apix-
aban (BMS-562247) afforded lactam 22 with binding
affinity for fXa on par with that of imidazole 19 and a
strong response in the PT assay. The corresponding aro-
matic analog of 22, pyridone 23, gratifyingly afforded an
increase in fXa binding by approximately 10-fold com-
pared to 22 and threefold compared to 13. Pyridone
23 was also very active in the clotting assay, proving
to be the most potent indoline prepared to date with
activity comparable to that of razaxaban in vitro.
The heterobicycles were generated according to Schemes
3–7.5c,b,c,6 Representative chemistry for preparing the
indoline moieties of compounds 2–3, 9–14, 19–21, and
23 is shown in Scheme 3. Reduction and Boc-protection
of commercially available 5-iodoindole (6-iodoindole in
the case of 14) afforded indoline 31, which could then be
reacted with boronic acid 32a–c under Suzuki conditions
to yield biaryls 33a–c. Reductive amination of aldehyde
33c with either pyrrolidine or dimethylamine afforded
benzyl amines 33d and 33e, respectively. In an analo-
gous manner, iodoindoline 31 was reacted under Ullman
conditions with previously reported imidazoles 35a–b5c
to afford indolines 36a–b. Treatment of 36b with acetyl
chloride or methanesulfonyl chloride under basic condi-
tions resulted in amides 36c–d. The indoline moiety of
23 was prepared by substituting 2-hydroxypyridine for
imidazole 35. Removal of the Boc-carbamate for all
compounds of Scheme 3 was accomplished using triflu-
oroacetic acid followed by conversion to the free base
upon workup using sodium bicarbonate.
Poor Caco-2 permeability was observed for the most ac-
tive compounds of Table 4 (11, 13, 22, and 23). This
raised the question of whether bioavailability would be
problematic upon oral dosing. To determine the phar-
macokinetic parameters of compounds from the indo-
line series and to assess how well the in vivo
permeability correlated with that observed in vitro, four
compounds were profiled in dog pharmacokinetic stud-
ies (Table 5). Pyrrolidines 12 and 13 had high rates of
clearance that were partially offset by high to moderate
volumes of distribution to afford modest intravenous
half-lives (t1/2(iv) = 2.2 and 2.3 h, respectively). Regard-
less, neither 12 or 13 was detectable in plasma when
dosed orally, most likely because of both poor absorp-
tion and significant first-pass metabolism. In contrast,
The indoline moieties of compounds 15–18 and 22 were
prepared according to the chemistry in Scheme 4 start-

J. G. Varnes et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6481–6488
6485
Table 4. Data for indoline P4 analogsa
O
H2N
R2
N
N
N
O
NH2
O
N
Compound
R2
fXa Ki, nM
PT IC2x lM
Caco-2 Papp · 10À6 cm/s
N
N
11
1.0
3.0
0.3 0.1
13
0.63
2.1
0.7 0.1
N
19
20
21
2.20
4.80
6.9
—
—
—
—
—
—
N
N
N
N
N
N
O
S
O
N
O
N
O
22
23
2.4
3.30
1.9
0.1
0.1
N
O
0.20
N
a Kis were obtained from purified human enzymes. Kis, PT, and Caco-2 values were measured according to Ref. 5a–c.
Table 5. Indoline pharmacokinetic profilesa
Compound
Caco-2 Papp · 10À6 cm/s
Cl L/h/kg
Vdss L/Kg
t1/2 (po)
F% (po)
Razaxaban
3
5.56
6.8 2.9
—
1.1
3.4
2.0
20
5.3
4.0
nd
nd
1.6
84
66
bql
bql
6
0.63
7.4
12
13
22
0.7 0.1
0.1
3.57
0.22
6.2
0.2
a Compounds were dosed in an N-in-1 format at 0.5 mg/kg iv and 0.2 mg/kg po (n = 1).5a–c
ing from commercially available 5- or 6-nitroindole (38).
Protection with di-tert-butyl dicarbonate and subse-
quent hydrogenation afforded aniline 39. The aniline
could then be converted to the lactam (41a–b) upon
treatment with either 40a or 40b and two equivalents
of potassium t-butoxide. Deprotection using 4 M hydro-
gen chloride in dioxane followed by basification upon
workup gave indolines 42a–b. Conversion of aniline 39

6486
J. G. Varnes et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6481–6488
R1
N
R1
The heterobicycles of compounds 4–6 were prepared
from known 6-bromo-2,3-dihydro-4(1H)-quinolinone
(45).9 Suzuki coupling with 2-thioanisole boronic acid
afforded ketone 46a. To prepare the corresponding race-
mic alcohol, 46a was treated with sodium borohydride
in tetrahydrofuran after the coupling described in
Scheme 1 but before aminobenzisoxazole formation.
Alternatively, 45 was reduced with an aluminum trichlo-
ride/lithium aluminum hydride melt prior to coupling
with 2-thioanisole boronic acid to generate 46b.
H
R2
N
N
N
24: a,c
25: b,c
N
N
P4
+
O
O
CN
NH2
P4
F
O
N
24: R1 = CF3, R2 = Cl
25: R1 = CONH2, R2 = OH
Scheme 1. Reagents and conditions: (a) NaH, CH2Cl2, 60–70%; (b)
DIC, py/DMF, 20–80%; (c) Ac-NHOH, K2CO3, DMF, 50 ꢁC, 20–50%.
The heterobicycle of 7 was prepared from commer-
cially available aniline 47, which was converted to
benzoxazinone 48 upon acylation with acetoxyacetyl-
chloride and cyclization under basic conditions. Suzu-
ki coupling with 2-thioanisoleboronic acid followed
by reduction with lithium aluminum hydride afforded
amine 49.
to glycinamide 43 was carried out via a pyBOP-medi-
ated coupling with N-methylglycine and subsequent
methylation. Deprotection and basification produced
indoline 44.
O
O
O
O
O
a, b
c, d, e
O
O
25
+
N
O
N
NHNH2 HCl
O
26
27
CN
CN
F
F
28
29
Scheme 2. Reagents and conditions: (a) LiHMDS, Et2O, À78 to 25 ꢁC, 83%; (b) 28, HOAc, 50 ꢁC, 93%; (c) TFA, CH2Cl2, 95%; (d) (COCl)2, DMF,
CH2Cl2 then NH3, dioxane/CH2Cl2, 69%; (e) KMnO4, t-BuOH, 5% NaH2PO4, H2O, 60 ꢁC, 71%.
I
N
H
30
R
R
a, b
I
c
f
R
N
N
H
N
Boc
33a: R = SCH3
33b: R = SO2NHtBu
33c: R = CHO
33d: R = CH2-pyrrolidine
33e: R = CH2N(CH3)2
Boc
31
34a: R = SCH3
B(OH)2
32a: R = SCH3
32b: R = SO2NHtBu
32c: R = CHO
34b: R = SO2NHtBu
34c: R = CH2-pyrrolidine
34d: R = CH2N(CH3)2
d
e
R
N
R
N
g
N
N
N
f
N
N
N
R
N
N
H
NH
Boc
36a: R = CH3
36b: R= H
36c: R = COCH3
36d: R = SO2CH3
37a: R = CH3
37b: R = COCH3
37c: R = SO2CH3
35a: R = CH3
35b: R = H
h
i
Scheme 3. Reagents and conditions: (a) HOAc, NaBH3CN; (b) Boc2O, NaHCO3, H2O/THF, 71% (two steps); (c) 32, Pd2(dba)3, PPh3, Na2CO3, tol,
110 ꢁC, 65–80%; (d) pyrrolidine, NaBH(OAc)3, ClCH2CH2Cl, 63%; (e) H2NCH3, NaBH(OAc)3, ClCH2CH2Cl, 90%; (f) TFA, CH2Cl2 then
NaHCO3, 60–80%; (g) 35, CuI, K2CO3, DMSO, 130 ꢁC, 33–60%; (h) AcCl, py, CH2Cl2 99%; (i) CH3SO2Cl, Et3N, CH2Cl2, 99%.

J. G. Varnes et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6481–6488
6487
O
O
c
d
N
N
n
O
N
N
H
n
Br
Boc
Cl
n
41a: n = 1
41b: n = 2
42a: n = 1
42b: n = 2
40a: n = 4
40b: n = 5
a, b
O2N
H2N
N
H
N
Boc
38
39
g
e, f
N
N
N
N
O
N
H
N
O
Boc
44
43
Scheme 4. Reagents and conditions: (a) Boc2O, DMAP, THF, 60%; (b) Pd/C, H2, MeOH, 90%; (c) 40, t-BuOK, THF, 52–67%; (d) 4 M HCl/dioxane
then NaHCO3, 46–88%; (e) N-methylglycine pyBOP, Et3N, DMF, 91%; (f) CH3I, DMF, 55%; (g) TFA, CH2Cl2, then NaHCO3, 85%.
SCH3
In conclusion, we have prepared a series of potent and
O
X
selective indoline fXa inhibitors designed to prevent the
release of a primary aniline upon hydrolysis in vivo.
The more potent compounds of this series demonstrate
subnanomolar binding affinity and in vitro clotting activ-
ity under 5 lM. Conversion from a 3-trifluoromethyl to
a 3-amidopyrazole improved binding affinity by approx-
imately 2- to 19-fold, while substitution at the 5-position
of the indoline was preferred over the 6-position. Incor-
poration of an N-substituted pyridone at the P4 position
gave an indoline fXa inhibitor with potency comparable
to that of razaxaban. Subsequent evaluation of a small
number of indolines in a dog pharmacokinetic model
indicated that phenylsulfonamide 3 (fXa Ki = 2.7 nM)
had a pharmacokinetic profile similar to that of razax-
aban. However, sulfonamide 3 also demonstrated re-
duced in vitro clotting activity. Permeability and oral
bioavailability remain a significant issue for compounds
with improved potency in the clotting assay.
Br
a or
b, a
N
H
N
H
46a: X = O
45
46b: X = H, H
Scheme 5. Reagents and conditions: (a) 2-Thioanisole boronic acid,
Pd(PPh3)4, 2 M NaHCO3, tol, 80 ꢁC, 65%; (b) AlCl3, LAH, THF,
60 ꢁC, 30%.
S
Br
F
O
Br
O
a, b
c, d
NH2
N
H
N
H
O
47
48
49
Scheme 6. Reagents and conditions: (a) ClCOCH2OAc, K2CO3,
CHCl3, 75%; (b) NaH, DMF, 55%; (c) 2-thioanisoleboronic acid,
Pd(PPh3)4, 2 M Na2NaCO3, 80 ꢁC, 72%; (d) LAH, AlCl3, THF, 71%.
References and notes
O
CO2Et
1. Hirsh, J.; O’Donnell, M.; Weitz, J. I. Blood 2005, 105(2),
453.
2. Rosenberg, J. S.; Beeler, D. L.; Rosenberg, R. D. J. Biol.
Chem. 1975, 250, 1607.
Br
Br
CO2H
NH2
Br
a, b, c
d, e
N
H
N
H
51
O
3. (a) Schaffer, J. A.; Davidson, J. T.; Vlasuk, G. P.; Siegel, P.
K. Circulation 1991, 84, 1741; (b) Wong, P. C.; Crain, E. J.;
Watson, C. A.; Zaspel, A. R.; Knabb, R. M. J. Pharmacol.
Exp. Ther. 2002, 303, 993; (c) Wong, P. C.; Pinto, D. J.;
Knabb, R. M. Cardiovasc. Drug Rev. 2002, 202, 137.
4. Swaminathan, A.; Frost, C.; Knabb, R.; Bai, S. A.;
Kornhauser, D.; Mosqueda-Garcia, R. Clin. Pharmacol.
Ther. 2004, 75, P6.
5. (a) Pinto, D. J. P.; Orwat, M. J.; Wang, S.; Fevig, J. M.;
Quan, M. L.; Amparo, E.; Cacciola, J.; Rossi, K. A.;
Alexander, R. S.; Smallwood, A. M.; Luettgen, J. M.;
Liang, L.; Aungst, B. J.; Wright, M. R.; Knabb, R. M.;
Wong, P. C.; Wexler, R. R.; Lam, P. Y. S. J. Med. Chem.
2001, 44, 566; (b) Pruitt, J. R.; Pinto, D. J. P.; Galemmo, R.
A.; Alexander, R. S.; Rossi, K. A.; Wells, B. L.; Drum-
mond, S.; Bostrom, L. L.; Burdick, D.; Bruckner, R.; Chen,
H.; Smallwood, A.; Wong, P. C.; Wright, M. R.; Bai, S.;
50
52
Scheme 7. Reagents and conditions: (a) H2SO4, EtOH, reflux, 13%; (b)
ClCO(CH2)2CO2Et, CH2Cl2, 60%; (c) KH, DMF, tol, 80 ꢁC, 62%; (d)
DMSO, H2O, 150 ꢁC, 70%; (e) AlCl3, LAH, THF, 60 ꢁC, 33%.
The heterobicycle of 8 was generated from commer-
cially available aniline 50. Esterification and acylation
followed by base-induced ring closure gave benzaze-
pine-2,5-dione 51. Thermal decarboxylation and deoxy-
genation employing an aluminum trichloride/lithium
aluminum hydride melt provided amine 52. This amine
was coupled as described in Scheme 1 with acid chlo-
ride 24 prior to a Suzuki coupling with boronic acid
32b as described in Scheme 3.

6488
J. G. Varnes et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6481–6488
Luettgen, J. M.; Knabb, R. M.; Lam, P. Y. S.; Wexler, R.
R. J. Med. Chem. 2003, 46, 5298; (c) Quan, M. L.; Lam, P.
Wexler, R. R.; Lam, P. Y. S. Bioorg. Med. Chem. Lett.
2006, 16, 4141.
Y. S.; Han, Q.; Pinto, D. J. P.; He, M. Y.; Li, R.; Ellis, C.
D.; Clark, C. G.; Teleha, C. A.; Sun, J.-H.; Alexander, R.
S.; Bai, S.; Luettgen, J. M.; Knabb, R. M.; Wong, P. C.;
Wexler, R. R. J. Med. Chem. 2005, 48, 1729.
7. Quan, M. L.; Han, Q.; Fevig, J.; Lam, P.; Bai, S.; Luettgen,
J. M.; Knabb, R. M.; Wong, P. C.; Wexler, R. R. Bioorg.
Med. Chem. Lett. 2006, 16, 1795.
8. Pinto, D. J. P.; Orwat, M. J.; Koch, S.; Rossi, K. A.;
Alexander, R. S.; Smallwood, A.; Wong, P. C.; Rendina,
A.; Luettgen, J. M.; Knabb, R. M.; He, K.; Xin, B.; Wexler,
R. R.; Lam, P. Y. S. J. Med. Chem. 2007. doi:10.1021/
jm070245n.
6. Pinto, D. J. P.; Orwat, M. J.; Quan, M. L.; Han, Q.;
Galemmo, R. A.; Amparo, E.; Wells, B.; Ellis, C.; He,
M. Y.; Alexander, R. S.; Rossi, K. A.; Smallwood, A.;
Wong, P. C.; Luettgen, J. M.; Rendina, A. R.; Knabb,
R. M.; Mersinger, L.; Kettner, C.; Bai, S.; He, K.;
9. Hurd, C. D.; Hayao, S. J. Am. Chem. Soc 1954, 76, 5065.