Design and synthesis of orally active pyrrolidin-2-one-based factor Xa inhibitors

Article abstract of DOI:10.1016/j.bmcl.2006.04.053

A series of novel, non-basic 3-(6-chloronaphth-2-ylsulfonyl)aminopyrrolidin-2-one-based factor Xa (fXa) inhibitors, incorporating an alanylamide P4 group, was designed and synthesised. Within this series, the N-2-(morpholin-4-yl)-2-oxoethyl derivative 24

Full text of DOI:10.1016/j.bmcl.2006.04.053

Bioorganic & Medicinal Chemistry Letters 16 (2006) 3784–3788
Design and synthesis of orally active pyrrolidin-2-one-based
factor Xa inhibitors
Nigel S. Watson,a,* David Brown,a Matthew Campbell,a Chuen Chan,a Laiq Chaudry,a
a
a
a
a
´
Maire A. Convery, Rebecca Fenwick, J. Nicole Hamblin, Claudine Haslam,
Henry A. Kelly,a N. Paul King,a Cynthia L. Kurtis,b Andrew R. Leach,a
Gary R. Manchee,a Andrew M. Mason,a Charlotte Mitchell,a Champa Patel,a
Vipulkumar K. Patel,a Stefan Senger,a Gita P. Shah,a Helen E. Weston,a
Caroline Whitwortha and Robert J. Younga
aGlaxoSmithKline, Medicines Research Centre, Gunnels Wood Road, Stevenage, Hertfordshire SG1 2NY, UK
bGlaxoSmithKline, 709 Swedeland Road, King of Prussia, PA 19406, USA
Received 27 February 2006; revised 12 April 2006; accepted 15 April 2006
Available online 11 May 2006
Abstract—A series of novel, non-basic 3-(6-chloronaphth-2-ylsulfonyl)aminopyrrolidin-2-one-based factor Xa (fXa) inhibitors,
incorporating an alanylamide P4 group, was designed and synthesised. Within this series, the N-2-(morpholin-4-yl)-2-oxoethyl
derivative 24 was shown to be a potent, selective fXa inhibitor with good anticoagulant activity. Moreover, 24 possessed highly
encouraging rat and dog pharmacokinetic profiles with excellent oral bioavailabilities in both species.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Factor Xa (fXa) plays a central role in the coagulation
cascade where it combines with factor Va and calcium
ions on membrane surfaces to form the prothrombinase
complex that activates prothrombin to thrombin. A key
action of thrombin is to convert fibrinogen into fibrin
leading to clot formation.1 In the search for orally active
anticoagulants with improved efficacy/safety profiles, di-
rect thrombin inhibitors have been investigated by our-
selves2 and others.3 More recently, fXa has been
recognised as an alternative attractive target for devel-
opment of new antithrombotic agents.4
series of non-amidine-based inhibitors with profiles suit-
able for chronic oral therapy.
fXa inhibitors incorporating a 3-amino-2-pyrrolidinone
scaffold and bearing benzamidine-based P1 motifs 1
have been disclosed5 and our in-house programme cen-
tred on identifying alternative N-substituted pyrrolidin-
2-one motifs that retained attractive in vitro potency
and promoted good in vivo profiles.
H
N
N
S
R
O
fXa is a trypsin-like serine protease and thus exhibits a
preference for basic groups to bind in its primary (S1)
specificity pocket. First-generation fXa inhibitors incor-
porated highly basic P1 moieties (e.g., benzamidine) and
whilst such compounds provide high affinity, they gener-
ally suffer from poor pharmacokinetic (PK) profiles and
low oral bioavailabilities. Our strategy has centred on
eliminating this developability risk by identifying novel
O
O
HN
NH2
1
We report here the identification of a non-basic series of
amides (Table 1) and studies to optimise their profiles.
Scheme 1 illustrates the generalised synthetic approach.
The pyrrolidinone 8 was prepared from CBZ-Met-OH
and the tert-butyl ester of the appropriate amino acid
adapting a previously reported approach from our lab-
oratories.7 Removal of the CBZ group followed by sulf-
onylation of the resultant amine provided a late stage
Keyword: Factor Xa.
*
Corresponding author. Tel.: +44 0 1438 763375; fax: +44 0 1438
763620; e-mail: Nigel.S.Watson@gsk.com
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.04.053

N. S. Watson et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3784–3788
3785
Table 1. fXa inhibitory activities6a for compounds 2–6
Table 2. fXa inhibitory activities6a for compounds 13 and 14
R1
R1
R2
H
N
R2
N
O
N
N
O
O
S
O
S
O
O
O
N
O
N
Cl
Cl
n
Compound
R1
R2
fXa Ki (nM)
Compound
R1
R2
n
fXa Ki (nM)
13a
13b
14
Et
Et
H
320
1020
H
Me
2
3
4
5
6
Me
Me
H
H
2
1
1
1
1
60
72
63
1
Me
10,600
H
H
H
Me
Me
Me
2
N3
N3
X
N3
CO2H
O
X = Br, I
EtO
O
O
N
NH
O
R1
R2
R1
R2
H
H
H
N
I
S+
N
CBZ
a, b
c
d,e
S
N
CBZ
CBZ
NH2
R1
O
O
O
R2
O
O
N
O
O
HN
HO
15
16
O
R2
a, b
c
d,e
O
R2
NH2
O
S
O
H
O
H
N
O
O
N
R2
S
O
O
7
8
O
Cl
O
O
Cl
f, g
N
N
O
S
R1
R2
O
R1
R2
H
H
N
O
O
N
S
O
O
N
O
Cl
O
O
Cl
f, g
N
N
17
18
O
R2
O
R2
Scheme 2. Reagents and conditions: (a) Et3N, MeCN, 50 ꢁC; (b)
LiOH, THF–H2O, rt; (c) (PhO)2P(O)N3, Et3N, DMF, rt; (d) H2, Pd–
i
C, EtOH, rt; (e) RSO2Cl, Pr2NEt, DCM, rt; (f) TFA, DCM, rt; (g)
N
R3
R4
O
9
10
h
piperidine, EDC, HOBT, Et3N, DCM, rt.
O
O
S
N
N
S
to give the racemic pyrrolidinone 16. Catalytic hydroge-
nation of the azido group followed by sulfonylation of
the derived amine gave intermediates 17 which were con-
verted into the desired products 18 using procedures
analogous to those in Scheme 1.
O
O
Cl
O
O
Cl
f, g
O
N
N
R2
O
R2
N
R3
R4
O
11
12
Focusing on the glycyl linker in 4, a series of branched-
chain analogues (Tables 1 and 2) showed that the
a-methyl-(alanyl) group conferred potent fXa inhibitory
activity in both the secondary and tertiary sulfonamide
series (5 and 6, respectively); related a-ethyl- and a,a-
di-methyl-analogues (13 and 14, respectively) showed
significantly reduced potency.
Scheme 1. Reagents and conditions: (a) EDC, 1-hydroxybenzotriazole
(HOBT), Et3N, DMF, rt; (b) MeI, acetone, rt; (c) DOWEX [OH]À,
MeCN; (d) H2, Pd–C, EtOH, rt; (e) 6-chloronaphth-2-yl-SO2Cl
(RSO2Cl), pyridine, MeCN, rt; (f) TFA, DCM; (g) R3R4NH, EDC,
HOBT, Et3N, DCM, rt; (h) MeOTs, [(Me)3Si]2NLi, THF, À78 ꢁC to
rt.
intermediate 9; deprotection of the tert-butyl ester fol-
lowed by coupling with a range of amines provided
the final products 10. Alternatively, N-alkylation of
intermediate 9 to provide 11 followed by deprotection
and amide coupling as described above provided a series
of related tertiary sulfonamides 12.
Retaining the alanyl linker, a wide range of terminal
amides was assessed and a representative subset is
shown in Table 3. The piperidine amide 5 was more
potent than the related pyrrolidine analogue 19 as were
the bridged analogues, 20 and 21.10 In contrast, the
piperazine 22 and N-methylpiperazine 23 derivatives10
showed significantly reduced activities. A range of sub-
stituents at C2 and C3 on the piperidine ring was also
well tolerated.11a All analogues tested from these series
possessed poor rat PK profiles characterised by moder-
ate to high plasma clearance and short half-lives.
Analogues 138 and 14 (Table 2) were prepared as race-
mates by an alternative route (Scheme 2).
Alkylation of the appropriate amino tert-butyl ester
with ethyl azido-4-halobutanoate9 followed by selective
saponification of the ethyl ester gave the acid 15 which
was cyclised in the presence of diphenylphosphoryl azide
A key finding from these studies was that reducing
hydrophobicity resulted in a better correlation between

3786
N. S. Watson et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3784–3788
Table 3. fXa inhibitory activities6a for compounds 5 and 19–23
O
S
H
H
H
N
CBZ
a,b
N
CBZ
c, d
N
H
N
O
O
N
S
O
Cl
O
O
O
O
N
N
N
Cl
O
O
O
R1
O
O
N
N
Compound
R1
fXa Ki (nM)
29
30
24
O
O
R1
O
S
5
1
N
N
25: R1 = Me
19
20
21
50
3
N
N
O
O
Cl
31: R1 = CH2CO2But
e
N
f
O
26: R1 = CH2CO2H
27: R1 = CH2CONH2
28: R1 = CH2COEt
13
N
NH
Me
N
22
23
320
N
N
NH
N
O
Scheme 3. Reagents and conditions: (a) TFA, DCM, rt; (b) morpho-
line, EDC, HOBT, Et3N, DCM, rt; (c) H2, Pd–C, EtOH, rt; (d)
RSO2Cl, pyridine, MeCN, rt; (e) MeOTs or R1-Br, [(Me)3Si]2NLi,
THF, À78 ꢁC to rt; (f) TFA, DCM, rt.
2060
intrinsic and anticoagulant potency as illustrated by the
in vitro profiles for the lipophilic bridged analogue 20
[CHI logD (pH 7.4) 2.94, 1.5· PT 21.7 lM] and the
more hydrophilic bispidine derivative 21 (CHI logD
(pH 7.4) 1.47, 1.5· PT 2.5 lM).6b,11–13 A difference in
the plasma protein binding of hydrophobic and more
hydrophilic analogues is believed to be a contributory
factor to this observation.14a
The preferred absolute stereochemistry in the highly
promising secondary sulfonamide 24 was unambiguous-
ly established utilising the tert-butyl esters of D- or L-al-
anine and D- or L-CBZ-Met-OH in the route shown in
Scheme 1. Evaluation of the four derived stereoisomers
24 and 32–34 demonstrated that the 1S,3S isomer 24
conferred the greatest potency (Table 5); similar findings
were established for the related piperidine analogues
(compare 5 with 35–37, Table 5).
Building upon the benefits of reducing logD, a major
breakthrough was achieved by replacing the piperidine
with a morpholine ring 24 [CHI logD (pH 7.4) 2.2].
Not only did 24 retain good fXa inhibitory activity (Ta-
ble 4), it also showed promising anticoagulant activity
and a rat PK profile characterised by low plasma clear-
ance and good oral bioavailability (85%). A range of ter-
tiary sulfonamides was assessed11b and a subset 25–28 is
shown in Table 4. Whilst these retained good in vitro
profiles, these analogues generally possessed poorer rat
PK profiles (Table 4).
Without a basic P1 motif, a key question was the bind-
ing mode of this series. A combination of approaches14a
shed light on this issue culminating in a high resolution
X-ray crystal structure of 24 bound in the active site of
fXa. The molecule could be unambiguously fitted to the
Fo À Fc electron density map establishing the binding
mode shown in Figure 1.15
Reverse exploitation of the orthogonally protected inter-
mediate 29 provided 24 with high (>98%) diastereomeric
purity and provided a preferred route to this analogue
(Scheme 3); tertiary sulfonamides 25–28 were derived
using standard procedures.
Thus, the chloronaphthyl group binds in S1 where the
chlorine atom makes a key interaction with Tyr228.14a,16
The carbonyl group in the pyrrolidinone template makes
a water-mediated interaction with Ser214. Additional
affinity is achieved with the (S)-methyl group that binds
Table 4. fXa inhibitory activities,6a anticoagulant potency6b and rat PK profilesa for 24 and a series of tertiary sulfonamides 25–28
R1
N
N
O
O
S
O
O
N
Cl
O
Compound
R1
fXa Ki (nM)
1.5· PT (lM)
t1/2 (h)
Clp (mL/min/kg)
Vss (L/kg)
24
25
26
27
28
H
Me
6
10
2
5
0.9
0.5
0.2
0.3
0.5
13
15
37
24
37
0.6
0.4
0.3
0.5
1.1
—
7
CH2CO2H
CH2CONH2
CH2COEt
3
3
1
4
a Compounds were administered either iv as a bolus or po via gavage to male Han Wistar rats at nominal doses of 1 mg/kg iv and 2.5 mg/kg po.

N. S. Watson et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3784–3788
Table 5. Influence of stereochemistry on fXa activities6a in morpho-
line- and piperidine-alanylamide series
3787
C3
H
N
N
S
C1
O
O
O
O
N
Cl
X
Compound
X
fXa Ki (nM)
24
32
33
34
5
O
O
O
O
1S,3S
1R,3R
1S,3R
1R,3S
1S,3S
1R,3R
1S,3R
1R,3S
6
82
53
160
1
CH2
CH2
CH2
CH2
35
36
37
40
16
82
Figure 2. X-ray crystal structure of 24 complexed with fXa (alternative
view).
was selected for further evaluation and findings from
these studies will be reported in separate publications.14
References and notes
1. Colman, R. W.; Clowes, A. W.; George, J. N.; Hirsh, J.;
Marder, V. J. In Hemostasis and Thrombosis: Basic
Principles and Clinical Practice; Lippincott, Williams and
Wilkins: Philadelphia, 2001 (Chapter 1).
2. Pass, M.; Abu-Rabie, S.; Baxter, A.; Conroy, R.; Coote, S.
J.; Craven, A. P.; Finch, H.; Hindley, S.; Kelly, H. A.;
Lowdon, A. W.; McDonald, E.; Mitchell, W. L.; Pegg, N.
A.; Procopiou, P. A.; Ramsden, N. G.; Thomas, R.;
Walker, D. A.; Watson, N. S.; Jhoti, H.; Mooney, C. J.;
Tang, C.-M.; Thomas, P. J.; Parry, S.; Patel, C. Bioorg.
Med. Chem. Lett. 1999, 9, 1657, and references therein.
3. (a) Sanderson, P. E. J.; Naylor-Olsen, A. M. Curr. Med.
Chem. 1998, 5, 289; (b) Coburn, C. A. Exp. Opin. Ther.
Patents 2001, 11, 721; (c) Steinmetzer, T.; Sturzebecher, J.
Curr. Med. Chem. 2004, 11, 2297.
Figure 1. X-ray crystal structure of 24 complexed with fXa.
4. (a) Quan, M. L.; Smallheer, J. M. Curr. Opin. Drug
Discovery Dev. 2004, 7, 460; (b) Gould, W. R.; Leadley, R.
J. Curr. Pharm. Des. 2003, 9, 2337.
in a small pocket so providing a rationale for the in-
creased potency seen with alanyl- compared with gly-
cyl-based analogues (see Table 1). The morpholine-
alanylamide group fits into the S4 pocket formed by
Phe174, Tyr99 and Trp215 but makes no specific H-
bonding interactions with the protein. An alternative
view of this X-ray crystal complex (Fig. 2) illustrates
that the sulfonamide group provides the appropriate
geometry for both the chloronaphthyl group to access
S1 and the rest of the inhibitor to be directed into S4.
5. (a) Ewing, W. R.; Becker, M. R.; Pauls, H. W.; Cheney, D.
L.; Mason, J. S.; Spada, A. P.; Choi-Sledeski, Y. M. WO
9640679 A1, 1996; (b) Ewing, W. R.; Becker, M. R.;
Manetta, V. E.; Davis, R. S.; Pauls, H. W.; Mason, H.;
Choi-Sledeski, Y. M.; Green, D.; Cha, D.; Spada, A. P.;
Cheney, D. L.; Mason, J. S.; Maignan, S.; Guilloteau, J.-
P.; Brown, K.; Colussi, D.; Bentley, R.; Bostwick, J.;
Kasiewski, C. J.; Morgan, S. R.; Leadley, R. J.; Dunwid-
die, C. T.; Perrone, M. H.; Chu, V. J. Med. Chem. 1999,
42, 3557.
6. (a) fXa inhibitory activities were determined using N-a-Z-
D-Arg-Gly-Arg-p-nitroanilide as chromogenic substrate;
details are described in Ref. 11; (b) Anticoagulant activ-
ities were determined in the prothrombin time (PT) assay;
details (test 2) are described in Ref. 11.
7. Macdonald, S. J. F.; Clarke, G. D. E.; Dowle, M. D.;
Harrison, L. A.; Hodgson, S. T.; Inglis, G. G. A.;
Johnson, M. R.; Shah, P.; Upton, R. J.; Walls, S. B.
J. Org. Chem. 1999, 64, 5166.
While some residual activity against thrombin (37-fold
selectivity) was apparent, morpholinamide 24 showed
good selectivity (>100-fold) over all other trypsin-like
serine proteases tested (fVIIa, fIXa, fXIIa, aPC, plas-
min, kallikrein, trypsin and elastase). Furthermore, 24
possessed an encouraging PK profile in the dog (t1/2
,
2.1 h; Clp, 2.7 ml/min/kg; Vss, 0.44 L/kg) with excellent
oral bioavailability (100%).17 On the basis of this prom-
ising profile, the 2-(morpholin-4-yl)-2-oxoethyl series
8. Compounds 13a and 13b were prepared as racemates from
the corresponding diastereomeric intermediates 16 which

3788
N. S. Watson et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3784–3788
were separated by chromatography; the relative stereo-
chemistry of the diastereomers 13a and 13b have not been
determined.
15. The fXa complex crystal structure with 24 was obtained
using an adaptation of the method outlined in Maignan,
S.; Guilloteau, J.-P.; Pouzieux, S.; Choi-Sledeski, S.;
Becker, M. R.; Klein, S. I.; Ewing, W. R.; Pauls, H. W.;
Spada, A. P.; Mikol, V. J. Med. Chem. 2000, 43, 3226,
Further details are described in Ref. 14a. The structure
9. (a)2-Azido-4-iodobutanoate was derived from the 4-bro-
mo derivative by reaction with sodium iodide in acetone;
2-azido-4-bromobutanoate was prepared from the 2,4-
dibromo derivative (Ref. 9b) by the method reported for
the corresponding methyl ester. J. Org. Chem. 1982, 47,
1319; (b) Wasserman, H. H.; Lipshutz, B. H.; Tremper, A.
W.; Wu, J. S. J. Org. Chem. 1981, 46, 2991.
10. Compounds 21–22 were derived by deprotection of the
corresponding Boc-protected derivatives by treatment
with TFA in DCM. Compound 23 was prepared as a
1:1 mixture of two diastereomers by the route outlined in
Scheme 2 utilising the tert-butyl ester of L-alanine.
11. (a) Chan, C.; Hamblin, J. N.; Kelly, H. A.; King, N. P.;
Mason, A. M.; Patel, V. K.; Senger, S.; Shah, G. P.;
Watson, N. S.; Weston, H. E.; Whitworth, C.; Young, R.
J. WO 02100886 A1, 2002.; (b) Chan, C.; Hamblin, J. N.;
Kelly, H. A.; King, N. P.; Mason, A. M.; Patel, V. K.;
Senger, S.; Shah, G. P.; Watson, N. S.; Weston, H. E.;
Whitworth, C.; Young, R. J. WO 02100830 A1, 2002.
12. Valko, K.; Du, C. M.; Bevan, C.; Reynolds, D. P.;
Abraham, M. H. Curr. Med. Chem. 2001, 8, 1137.
˚
was refined at 2.1 A (overall Rmerge is 0.077) in Refmac5 to
a final Rfactor of 0.192 and Rfree of 0.242. Coordinates are
deposited in the protein data bank with code 2cji.
16. Similar findings have been reported recently for other
series of fXa inhibitors (a) Choi-Sledeski, Y. M.;
Kearney, R.; Poli, G.; Pauls, H.; Gardner, C.; Gong,
Y.; Becker, M.; Davis, R.; Spada, A.; Liang, G.; Chu,
V.; Brown, K.; Collussi, D.; Leadley, R., Jr.; Rebello,
S.; Moxey, P.; Morgan, S.; Bentley, R.; Kasiewski, C.;
Maignan, S.; Guilloteau, J.-P.; Mikol, V. J. Med. Chem.
2003, 46, 681; (b) Jia, Z. J.; Su, T.; Zuckett, J. F.; Wu,
Y.; Goldman, E. A.; Li, W.; Zhang, P.; Clizbe, L. A.;
Song, Y.; Bauer, S. M.; Huang, W.; Woolfrey, J.; Sinha,
U.; Arfsten, A. E.; Hutchaleelaha, A.; Hollenbach, S. J.;
Lambing, J. L.; Scarborough, R. M.; Zhu, B.-Y. Bioorg.
Med. Chem. Lett. 2004, 14, 2073; (c) Roehrig, S.;
Straub, A.; Pohlmann, J.; Lampe, T.; Pernerstorfer, J.;
Schlemmer, K.-H.; Reinemer, P.; Perzborn, E. J. Med.
Chem. 2005, 48, 5900.
13. Compound 21 shows good selectivity over other trypsin-
like serine proteases tested, including thrombin.
14. (a) Chan, C. et al. J. Med. Chem., manuscript in
preparation.; (b) Senger, S. et al. Bioorg. Med. Chem.
Lett., manuscript in preparation.
17. Compound 24 was administered iv (1 mg/kg) as a slow
bolus over 5 min and po (1 mg/kg) via gavage to female
Beagle dogs. The dosing vehicle was a 5:95% (v/v) mixture
of DMSO and 50:50 PEG-200:sterile water.