Novel phenylalanine derived diamides as Factor XIa inhibitors

Article abstract of DOI:10.1016/j.bmcl.2015.11.089

The synthesis, structural activity relationships (SAR), and selectivity profile of a potent series of phenylalanine diamide FXIa inhibitors will be discussed. Exploration of P1 prime and P2 prime groups led to the discovery of compounds with high FXIa affinity, good potency in our clotting assay (aPPT), and high selectivity against a panel of relevant serine proteases as exemplified by compound 21. Compound 21 demonstrated good in vivo efficacy (EC₅₀ = 2.8 μM) in the rabbit electrically induced carotid arterial thrombosis model (ECAT).

Full text of DOI:10.1016/j.bmcl.2015.11.089

Bioorganic & Medicinal Chemistry Letters 26 (2016) 472–478
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Novel phenylalanine derived diamides as Factor XIa inhibitors
⇑
Leon M. Smith II , Michael J. Orwat, Zilun Hu, Wei Han, Cailan Wang, Karen A. Rossi, Paul J. Gilligan,
Kumar B. Pabbisetty, Honey Osuna, James R. Corte, Alan R. Rendina, Joseph M. Luettgen, Pancras C. Wong,
Ranga Narayanan, Timothy W. Harper, Jeffrey M. Bozarth, Earl J. Crain, Anzhi Wei,
Vidhyashankar Ramamurthy, Paul E. Morin, Baomin Xin, Joanna Zheng, Dietmar A. Seiffert, Mimi L. Quan,
Patrick Y. S. Lam, Ruth R. Wexler, Donald J. P. Pinto
Research and Development, Bristol-Myers Squibb Company, P.O. Box 5400, Princeton, NJ 08543, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis, structural activity relationships (SAR), and selectivity profile of a potent series of
phenylalanine diamide FXIa inhibitors will be discussed. Exploration of P1 prime and P2 prime groups
led to the discovery of compounds with high FXIa affinity, good potency in our clotting assay (aPPT),
and high selectivity against a panel of relevant serine proteases as exemplified by compound 21.
Received 13 October 2015
Revised 20 November 2015
Accepted 24 November 2015
Available online 26 November 2015
Compound 21 demonstrated good in vivo efficacy (EC50 = 2.8 lM) in the rabbit electrically induced
carotid arterial thrombosis model (ECAT).
Keywords:
Factor XIa
FXIa
Ó 2015 Elsevier Ltd. All rights reserved.
Anticoagulant
Thrombosis
Serine protease
aPTT
Hemostasis is an adaptive process that maintains blood in a fluid
state and preserves vasculature integrity.1 Disruption of hemostatic
balance, if untreated, can lead to thromboembolic complications
such as deep vein thrombosis, pulmonary embolism, myocardial
infarction, and stroke. Recently, new anticoagulants have been
developed that target specific enzymes in the coagulation cascade,
such as the thrombin inhibitor, dabigatran (PradaxaÒ)2 and Factor
Xa (FXa) inhibitors, rivaroxaban (XareltoÒ),3 apixaban (EliquisÒ),4
and edoxaban (SavaysaÒ).5 In part, these agents address the limita-
tions seen with the narrow therapeutic index associated with the
drug warfarin.6 More recently, an intensive effort has been under-
taken to identify agents that target FXIa as potential alternative
therapies to warfarin prophylaxis.
a mild bleeding phenotype.7 Subjects with high Factor XI plasma
levels are at increased thrombosis risk, whereas subjects with sev-
ere FXI deficiency have a reduced incidence of ischemic stroke.8
Direct, active-site inhibitors of FXIa show robust antithrombotic
efficacy with minimum bleeding in rabbit models of thrombosis
and hemostasis.9a,b A recent clinical study (VTE prevention after
knee replacement surgery) demonstrated that administration of a
FXI-directed
antisense
oligonucleotide
(ASO)
prevented
thrombosis and appeared safe with respect to bleeding.10 Together,
the evidence suggests that FXIa inhibition could block pathologic
thrombus formation while preserving normal hemostasis.
Recently, our labs disclosed a novel series of phenylimidia-
zoles11a,b possessing neutral P1 groups including 1 that exhibited
good FXIa potency and modest oral exposure in dogs.11c In this
report, we describe the utility of these neutral P1 groups in a
phenylalanine diamide chemotype 2. Optimization of the
chemotype’s P2 prime (P20) and P1 prime (P10) regions will also
be discussed.
Factor XI, a serine protease of the intrinsic blood coagulation
pathway, is activated to FXIa by thrombin, Factor XIIa, or FXIa
itself. Factor XIa amplifies thrombin generation leading to stable
fibrin clot formation. Targeting FXIa inhibition is supported by
the observation that human FXI deficiency (hemophilia C) exhibits
⇑
Corresponding author. Tel.: +1 609 466 5078; fax: +1 609 818 3450.
E-mail address: leon.smith@bms.com (L.M. Smith).
http://dx.doi.org/10.1016/j.bmcl.2015.11.089
0960-894X/Ó 2015 Elsevier Ltd. All rights reserved.

L. M. Smith II et al. / Bioorg. Med. Chem. Lett. 26 (2016) 472–478
473
Table 1
N N
Structure–activity relationships of the P20 region
N
N
O
N
N
N
N
NHCO Me
N
H
2
N
O
H
N
HN
N
H
R
1
Cl
Cl
O
FXIa Ki = 2.7 nM
aPTT EC1.5x = 5.3 µM
Cl
Compound
R
FXIa Kia
(nM)
PK Kia
(nM)
aPTTb EC1.5x
(lM)
P1`
N
N
N
3
4
40.5
6.4
120
>40.0
5.75
NHCO2Me
N
O
H
N
H
N
N
P1
N
H
P2`
24
28
O
NH
2
2
Cl
OH
5
11.2
16.5
N
H
O
In designing the phenylalanine diamide chemotype, molecular
modeling suggested binding in the FXIa active site would be simi-
lar to phenylimidazole analog 1. It was also postulated that subtle
differences could emerge at the P20.
6
7
8
2
140
210
130
4.2
COOH
O
17
17.5
>40.0
18.8
With this in mind, we began our SAR effort by exploring the P20
group (Table 1). Our initial strategy involved the synthesis of sev-
eral key analogues possessing optimal P20 moieties previously
identified in the phenylimidazole series (compounds 3–5).11,11b
The 4-methoxycarbonylaminophenyl analog 3 exhibited modest
OH
CONH2
9
47
170
NTc
FXIa affinity (FXIa Ki = 40.5 nM) and greater than 40 lM activity
CN
in the aPTT (clotting) assay9c,b compared to the significantly
NHCO2Me
more potent phenylimidazole compound 1 (FXIa Ki = 2.7 nM;
aPTT EC1.5x = 2.7 lM). This result supported our prediction that
10
11
173
350
270
NTc
NTc
COOH
the P20 group SAR would probably deviate between the two
series. While aminoindazole 4 and 4-hydroxyquinolin-2(1H)-one
5 analogs performed better than 3, significant room for further
improvements existed. X-ray crystal structure of 1 showed a
strong hydrogen bond interaction of the carbamate NH with the
carbonyl of His40 and the carbamate carbonyl group with Ile151
45.6
a
Ki values were obtained from purified human enzyme and were averaged from
multiple determinations (n = 2), as described in Ref. 9.
aPTT (activated partial thromboplastin time) in vitro clotting assay was per-
formed in human plasma as described in Ref. 9c.
b
c
NT = not tested.
interacting through
a
highly conserved structural water
molecule.11b However, modeling of 3 suggested that the methyl
carbamate could not efficiently make these same interactions
due to the change in torsional angle when compared to the P20
group of 1. Gratifyingly, 4-benzoic acid analog 6 resulted in a
20-fold increase in FXIa binding affinity (FXIa Ki = 2 nM) and a
>10-fold improvement in aPTT potency (EC1.5x = 4.2 lM)
compared to 3. The carboxylic acid functionality is within
hydrogen bond distance to Tyr143. This modification also led to
improved selectivity over plasma kallikrein (PK Ki = 140 nM) with
a 70:1 selectivity ratio towards FXIa. The 4-phenylacetic acid
analog 7 resulted in a >8-fold decrease in FXIa binding affinity
(FXIa Ki = 17 nM). Meta-substitution as demonstrated by analogs
10 and 11 were relatively less potent.
The X-ray crystal structure of 6 bound in the active site of FXIa
at 2.5 Å resolution is depicted in Figure 1.12 The chlorophenyl
tetrazole moiety occupies the S1 pocket with the Cl atom forming
an interaction with Tyr228 at the bottom of the S1 pocket, while
the tetrazole N-3 atom is optimally oriented within H-bonding dis-
tance to the e-amino group of Lys192 (3.1 Å). The tetrazole N-2
nitrogen is near the Lys192 backbone NH (3.3 Å) and the disulfide
bridge Cys219–Cys192 (3.1–4.0 Å). The acidic tetrazole CH proton
appears to be in close proximity to the carbonyl of Gly216
(3.3 Å). The cinnamide linker carbonyl is positioned nicely within
the oxyanion hole by forming key hydrogen bonds with Ser195,
Asp194, and Gly193 residues. Overall, the P1 of 6 is similarly
Figure 1. X-ray crystal structure of 6 complexed with FXIa.
oriented to that observed for compound 1. The benzyl side chain,
which occupies the S10 pocket, forms a hydrophobic edge-on con-
tact with the disulfide bridge (Cys42–Cys58). Lastly, the P20 amide

474
L. M. Smith II et al. / Bioorg. Med. Chem. Lett. 26 (2016) 472–478
Table 2
NH makes a water mediated hydrogen bond with Leu41 while the
carboxylic acid makes a water mediated hydrogen bond with His40
and a direct hydrogen bond with Tyr143 (2.7 Å) in the enzyme
backbone. Overall, the ligand binds to FXIa in a highly efficient
manner with each atom on the ligand productively interacting
with the enzyme active site residues. The difference in SAR trends
between phenylalanine diamides and the phenylimidazole series
can be explained by the overlay of co-crystal structures of 6 with
1 as depicted in Figure 2.13 The phenyl group of the 4-benzoic acid
P20 moiety of 6 does not go as deep into the S20 pocket and is
rotated approximately 60° when compared to the 4-methoxycar-
bonylaminophenyl group in 1. Each motif takes advantage of key
hydrogen bond interactions with a complex network of conserved
water molecules or via direct interactions with the FXIa enzyme
backbone residues.
Carboxylic acid replacement with biosteres
N
N
N
N
O
H
N
N
H
O
R
Cl
Compound
R
FXIa Kia (nM) PK Kia (nM) aPTTb EC1.5x
(lM)
OH
6
2
140
23
4.2
4.6
O
N
H
N
12
1.4
N
N
Next, we decided to expand the SAR of the P20 group by replac-
ing the carboxylic acid moiety in 6 with bioisosteres (Table 2). The
C-substituted tetrazole acid bioisostere analog 12 was equipotent
while the corresponding N-substituted tetrazole 13 was substan-
tially weaker. Interestingly, 12 led to a decrease in selectivity over
plasma kallikrein (16:1). Imidazole 14 and 1H-1,2,4-triazole 15
also exhibited satisfactory in vitro results. However, carboxylic
acid surrogates 1,2,4-oxadiazol-5(2H)-one 16 and 1,3,4-oxadia-
zol-2(3H)-one 17 had good FXIa affinity but were 2–3-fold weaker
in the clotting assay which is probably attributable to high protein
binding.
13
14
43
240
150
>40.0
7.9
N
5.5
HN
N
N
15
16
14
2
120
36
8.6
9.6
NH
N
O
HN
O
We next explored the effect of substitution on the P10 phenyl
group (Table 3). It should be noted that introduction of a fluorine
in the 2-position of the P1 phenyl 18 resulted in a modest but sig-
nificant improvement in aPTT potency when compared to 6. Sub-
stitution of the 4-position of the phenyl group with non-polar
groups as well as basic groups is well tolerated as exemplified by
the FXIa affinities of compounds 19–24. Propyl analog 19 and
biphenyl analog 20 did not offer an advantage when compared to
N
NH
O
17
5.8
20
12.7
O
a
Ki values were obtained from purified human enzyme and were averaged from
multiple determinations (n = 2), as described in Ref. 9.
aPTT (activated partial thromboplastin time) in vitro clotting assay was per-
formed in human plasma as described in Ref. 9c.
b
6
resulting in a >4-fold loss in aPTT potency. However,
methylpiperazine urea analog 21 gave improved clotting activity
(submicromolar potency) while maintaining good FXIa affinity.
Interestingly, inserting fluorine into the P1 phenyl did not enhance
APTT as seen in 18. Installation of a cycloproylamide linked
hydrophobic P10 analog 23 led to sub-nanomolar FXIa affinity.
The boost in affinity may be attributed to the cyclopropyl group
having a favorable interaction with the hydrophobic region near
Leu41 as depicted in the molecular model of 23 (Fig. 3).
Table 3
Structure–activity relationships in the P1 prime region
R
N
N
N
N
O
H
N
N
H
O
OH
R'
Cl
O
Compound
R
R0
FXIa/PK Kia (nM)
aPTTb EC1.5x
4.2
(lM)
6
H
H
F
2.0/140
18
H
1.7/46
1.2
19
20
H
H
7.1/804
18.5
38.9
11.5/317
O
O
O
21
22
H
F
2.0/59
1.8/55
0.36/7.5
0.5
0.7
1.5
N
H
N
N
N
H
N
N
23c
F
N
H
a
Ki values were obtained from purified human enzyme and were averaged from
multiple determinations (n = 2), as described in Ref. 9.
b
aPTT (activated partial thromboplastin time) in vitro clotting assay was per-
formed in human plasma as described in Ref. 9c.
Figure 2. Superposition of crystal structures of diamide 6 (cyan) and phenylimi-
dazole 1 (orange; PDB ID 4Y8X) in active site of FXIa.
c
Diastereomeric mixture.

L. M. Smith II et al. / Bioorg. Med. Chem. Lett. 26 (2016) 472–478
475
Table 4
To further our understanding on how these molecules bound to
Structure–activity relationship of reversed diamides
FXIa, we also evaluated compounds wherein the P1 and P20 groups
were reversed (Table 4). To enable the p-phenyltetrazole P1 to bind
to the S1 region, the reversed diamides required the incorporation
of an ethylene linker on the P1 phenyl group. The SAR for com-
pounds 24–27 was the exact opposite for that observed for the pre-
vious diamides (Tables 1–3). For example, compounds 24 and 25
were >1000-fold less potent when compared to compounds 6
and 12. It was interesting to note that the SAR in the reversed
diamides series tracked closely to that observed in the phenylimi-
dazole series (e.g., compounds 26 and 27 contain the P20 groups
which showed most potency in phenylimidazole series).
Superposition of the crystal structures of phenylimidazole 1 and
reversed diamide 27 (Fig. 4) reveals the P20 phenyl groups of both
chemotypes overlay in a highly efficient manner.4a The primary
difference in their binding modes is that P1 linker carbonyl of 27
approaches the oxyanion hole from a higher position but all
critical interactions are maintained.
N
N
N
Ph
N
O
H
N
N
H
R
O
Cl
Compound
R
FXIa
PK Kia
(nM)
aPTTb EC1.5x
(lM)
Kia(nM)
24
25
26
2688
1656
3.4
>6000
>6000
340
NTc
NTc
4.2
COOH
H
N
N
N
N
NHCO2Me
OH
The compounds in Tables 1–3 were prepared according to gen-
eral synthetic approach outlined in Scheme 1. Boc L-phenylalanine
27
1.6
110
1.0
28 was coupled with P20 anilines 29 in pyridine using phosphoryl
oxychloride to give 30 in good yields. This method proved
to be particularly useful when electron deficient anilines were
employed. Likewise, standard peptide coupling methods can be
utilized with electron rich inputs. Intermediates 32a and 32b were
obtained from commercially available 4-chloro-2-iodoaniline and
2-amino-5-chlorobenzaldehyde using known literature meth-
ods.13a The resulting iodo compound 33a was converted to
N
H
O
a
Ki values were obtained from purified human enzyme and were averaged from
multiple determinations (n = 2), as described in Ref. 9.
b
aPTT (activated partial thromboplastin time) in vitro clotting assay was per-
formed in human plasma as described in Ref. 9c.
c
NT = not tested.
cinnamate ester 32a via
a
Heck cross coupling with ethyl
acrylate. Alternatively,
a Horner Emmons Wadsworth
condensation of benzaldehyde 33b and methyl diethyl
phosphonoacetate provided the corresponding methyl ester
32b.13b Saponification products 31a or 31b and the free amine of
30 were coupled to afford compounds 3–20.
Advanced intermediates for P20 phenylmethylcarbamate 3 and
4-hydroxyquinolin-2(1H)-one
5 compounds in Table 1 were
Figure 4. Superposition of crystal structures of reversed diamide 27 (green) and
phenylimidazole 1 (magenta) complexed with FXIa.
prepared as shown in Scheme 2 starting from Pd/C catalyzed
hydrogenation of 34. Compound 36 was accessed by treatment
of aniline 35 with methyl chloroformate. (S)-tert-Butyl (1-((4-ami-
nophenyl)amino)-1-oxo-3-phenylpropan-2-yl)carbamate 35 also
served as a useful intermediate in the preparation of 38 by
coupling with malonic acid mono tert-butyl ester followed by
cyclization under PPA conditions11b and in situ re-protection of
the Boc group in good yields. Aminoindazole compound 4 was
prepared from appropriate starting materials utilizing known
procedures.11
Compound 21 (Table 4) was accessed via nitro reduction of 39
and subsequent urea formation by treatment with triphosgene
and N-methylpiperazine, respectively, (Scheme 3) and standard
attachment of the P1 group. In a similar manner, the cyclopropyl
Figure 3. Molecular model of 23 in active site of FXIa.

476
L. M. Smith II et al. / Bioorg. Med. Chem. Lett. 26 (2016) 472–478
NO2
NH2
R
R
NH2
O
O
O
O
H
N
a
H
N
H
N
OH
+
R'
O
N
H
O
N
H
a
O
N
H
O
N
H
R'
O
30
O
28
O
39
O
O
40
O
29
O
O
N
N
N
N
N
N
N
H
R
N
N
O
N
O
N
O
H
N
b,c, d
O
OH
O
30
N
or
b, e
+
H
N
R'
H
b, c
O
Y
O
N
Y
3 - 20
H
Cl 31a:Y = H
O
41
O
Cl
31b: Y = F
Scheme 3. Reagents and conditions: (a) 10% Pd/C, MeOH, H2, quantitative; (b)
trisphosgene, DIPEA; (c) N-methylpiperazine, 54%.
i
N
N
N
N N
NH2
N
X
N
O
OR"
Y
N
X = I,
g
f
N
N
N
N N
X
N
N
N
N
Y
N
O
N
O
N
X = CHO,
h
Cl
Y
NH2
OH
NH2
b,c
a
d
X = I, CHO
Y = H or F
31a
Cl
Cl
33a: X = I
33b:
32a: R" = Et
32b: R" = Me
43
42
44
X = CHO
Cl
Cl
Cl
Scheme 1. Reagents and conditions: (a) POCl3, pyridine, DCM, ꢀ15–0 °C, 18–93%;
(b) TFA, DCM or HCl, dioxane, 51–95%; (c) 1-hydroxypyrrolidine-2,5-dione, DIC,
DMF/THF, 64–100%; (d) DIEA, DMF, 37–73%; (e) EDC, HOBt, DIPEA, DMF, 8–48%; (f)
NaN3, CH(OMe)3, AcOH, rt, 24 h, 90%; (g) ethyl acrylate, TEA, Pd(OAc)2, CH3CN 85 °C,
24 h, 43%; (h) NaH, methyl diethylphosphonoacetate, THF, rt, 24 h, 57%; (i) NaOH,
MeOH, water, rt, 94%.
Scheme 4. Reagents and conditions: (a) PtO2, H2, MeOH, 83%; (b) ClCO2Et, Et3N,
THF, ꢀ78 °C, (c) NH3, MeOH; (d) PhI(O2CCF3)2, pyridine, MeCN, H2O, 95% over three
steps.
N
N
N
N
N
N
Ph
Ph
carboxamide 23 was prepared by coupling of 40 with cyclopropyl
carboxylic acid under EDC conditions.
N
R
COOH
N
O
H
N
H
N
47
a, b
NH2
44
c
N
H
R
The synthesis of reversed linear diamides in Table 4 is described
in Schemes 4 and 5. The double bond of cinnamic acid 31a was
hydrogenated under PtO2 conditions to give carboxylic acid 42
which was easily converted into primary amide 43 by treatment
of the mixed anhydride with ammonia. Compound 43 was sub-
jected to Hofmann degradation to afford primary amine 44 in good
yields. The phenyltetrazole phenethyl amine 44 was coupled with
O
O
45
24 - 27
Cl
Cl
Scheme 5. Reagents and conditions: (a) Boc
L
-phenylalanine 28, EDC, HOBt, DIPEA,
DMF (b) HCl, dioxane, 85% over two steps; (c) EDC, HOBt, DIPEA, DMF, 21–89%.
The pharmacokinetics of representative compounds 6, 21, and
27 were evaluated following intravenous administration in dog
(Table 5).4a Generally, all compounds exhibited short t1/2 lives
Boc L-phenylalanine 28 followed by TFA deprotection and coupling
with appropriate carboxylic acids under EDC/HOBt conditions to
give compounds 24–27 (Scheme 5).
O
H
N
O
O
N
H
O
O
N
H
O
36
b
O
H
N
H
N
a
c
O
N
H
O
N
H
O
34
O
NH2
NO2
35
O
H
N
O
OH
d, e
OtBu
H
N
O
N
O
O
O
N
H
H
O
37
O
N
N
H
O
H
38
Scheme 2. Reagents and conditions: (a) 10% Pd/C, MeOH, H2, quantitative; (b) ClCO2Me, pyridine, DCM, 0 °C, 87–100%; (c) malonic acid mono tert-butyl ester, BOP reagent,
Et3N, THF, 92%; (d) PPA, 120 °C; (e) Boc2O, NaOH, 68% for two steps.

L. M. Smith II et al. / Bioorg. Med. Chem. Lett. 26 (2016) 472–478
477
Table 5
Dog PK summarya
Compound
Cl (mL/min/kg)
Vdss (L/kg)
t1/2 (h)
AUCtot (nM * h)
6
21
27
16
14
32
0.5
0.7
1.3
0.6
0.9
1.0
396
314
206
a
Compounds were dosed at 0.4 mg/kg, i.v. in a cassette format.
Table 6
Enzyme selectivity profiles of compounds 6, 21, and 27
Human enzyme (Ki)a
6 (nM)
21 (nM)
27 (nM)
Figure 6. Compound 21 ex vivo aPTT vs. PT.
Factor XIa
2.0
2.0
1.6
Factor VIIa
Factor IXa
Factor Xa Ki
Factor XIIa
Trypsin
>13,000
>73,000
5200
>28,000
>6200
>13,000
122
>13,000
>24,000
>9000
>3700
>6200
>13,000
59
>13,000
>73,000
>9000
>28,000
>6200
>13,000
111
prolonged ex vivo aPTT by 1.8-fold but not PT, as expected with the
XIa mechanism (Fig. 6).
In conclusion, phenylalanine derived diamides have been suc-
cessfully established as potent and selective FXIa inhibitors. SAR
and X-ray crystallography revealed that P2 prime groups in the
diamides are orientated in a slightly different manner than the
phenylimidazole series. Furthermore, we were able to identify
molecules with subnanamolar FXIa potency by installing
hydrophobic groups in the P10 region of the template. We also dis-
covered that substituents on the core amide bond can be reversed
and maintain good FXIa potency. Ultimately, our studies led to the
identification of compound 21 which demonstrated good efficacy
Thrombin
Plasma kallikrein
Chymotrypsin
Activated protein C
Urokinase
7649
280
1650
>64,000
>28,000
>46,000
>30,000
>60,000
>15,000
>8,600
>30,000
>64,000
>28,000
>46,000
>30,000
tPA
Tissue kallikrein
a
All Ki values were obtained using human purified enzymes and were averaged
from multiple determinations (n = 2).
(EC50 = 2.8 lM) in the rabbit electrically induced carotid arterial
thrombosis model (ECAT). Further exploration of phenylalanine
derived diamides will be disclosed in due course.
Acknowledgments
The authors would like to thank the BMS Department of
Discovery Synthesis and BMS Biocon Research Center for the
preparation of synthetic intermediates and Dr. Steven Sheriff for
depositing the coordinates and data for the complexes of FXIa with
compounds 6 and 27, respectively.
References and notes
1. Schumacher, W. A.; Luettgen, J. M.; Quan, M. L.; Seiffert, D. A. Arterioscler.
Thromb. Vasc. Biol. 2010, 30, 388.
2. (a) Hauel, N. H.; Nar, H.; Priepke, H.; Ries, U.; Stassen, J. M.; Wienen J. Med.
Chem. 2002, 45, 1757; (b) Ieko, M. Curr. Opin. Investig. Drugs 2007, 8, 758; (c)
Van Ryn, J.; Stangier, J.; Haertter, S.; Liesenfeld, K. H.; Wienen, W.; Feuring, M.
Thromb. Haemost. 2010, 103, 1116; (d) Hankey, G. J.; Eikelboom Circulation
2011, 123, 1436.
3. (a) Perzborn, E.; Roehring, S.; Straub, A.; Kubitza, D.; Misselwitz, F. Nat. Rev.
Drug Discovery 2011, 10, 61; (b) Straub, A.; Roehrig, S.; Hillisch, A. Curr. Top.
Med. Chem. 2010, 10, 257; (c) Misselwitz, F.; Berkowitz, S. D.; Perzborn, E. Ann.
N. Y. Acad. Sci. 2011, 1222, 64; (d) Lassen, M. R.; Ageno, W.; Borris, L. C.;
Lieberman, J. R.; Rosencher, N.; Bandel, T. J.; Misselwitz, F.; Turpie, A. G. G. N.
Engl. J. Med. 2008, 358, 2776.
4. (a) Pinto, D. J. P.; Orwat, M. J.; Koch, S.; Rossi, K. A.; Alexander, R. S.; Smallwood,
A.; Wong, P. C.; Rendina, A. R.; Luettgen, J. M.; Knabb, R. M.; He, K.; Xin, B.;
Wexler, R. R.; Lam, P. Y. S. J. Med. Chem. 2007, 50, 5339; (b) He, K.; Luettgen, J.
M.; Zhang, D.; He, B.; Grace, J. E., Jr.; Xin, B.; Pinto, D. J. P.; Wong, P. C.; Knabb, R.
M.; Lam, P. Y. S.; Wexler, R. R.; Grossman, S. J. Eur. J. Drug Metab. Pharm. 2011,
36, 129; (c) Pinto, D. J. P.; Smallheer, J. M.; Cheney, D. L.; Knabb, R. M.; Wexler,
R. R. J. Med. Chem. 2010, 53, 6243; (d) Pinto, D. J. P.; Qiao, J. X.; Knabb, R. M.
Expert Opin. Ther. Patents 2012, 22, 645.
5. Furugohri, T.; Isobe, K.; Honda, Y.; Kamisato-Matsumoto, C.; Sugiyama, N.;
Nagahara, T.; Morishima, Y.; Shibano, T. J. Thromb. Haemost. 2008, 6, 1542.
6. (a) Hyers, T. M. Arch. Intern. Med. 2003, 163, 759; (b) Stein, P. D.; Grandison, D.;
Hua, T. A. Postgrad. Med. J. 1994, 70, S72; (c) Hirsh, J.; Poller, L. Arch. Intern. Med.
1994, 154, 282.
Figure 5. Compound 21 in rabbit ECAT model (prevention mode).
(61 h). Compounds 6 and 21 exhibited moderate clearance and
low volume of distribution while compound 27 had high
clearance and volume of distribution.
Table 6 shows the enzyme selectivity comparison of com-
pounds 6, 21, and 27. In general, all three compounds were highly
selective over a panel of relevant serine proteases (>1000-fold)
including good selectivity over plasma kallkrein (>60-fold). Com-
pound 21 showed less selectivity over chymotrypsin than observed
with the other two compounds The compounds exhibited differen-
tiated human liver stability profiles with 6 and 21 (t1/2 = >120 min)
being significantly more stable than 27 (t1/2 = >32 min). Generally,
normal diamides such as compounds (3–22) had exceptional HLM
stability (t1/2 = >100 min) while reversed diamides 24–27 were
poor to moderately stable (t1/2 = 2–32 min).
Compound 21 had the best balance of good potency, selectivity,
and pharmacokinetic profile and was chosen for evaluation. In the
rabbit electrically induced carotid arterial thrombosis model
(ECAT).14 As depicted by Figure 5 and 21 exhibited dose-dependent
7. (a) Seligsohn, U. J. Thromb. Haemost. 2009, 7, 84; (b) Renné, T.; Oschatz, C.;
Seifert, S.; Müller, F.; Antovic, J.; karlman, M.; Benz, P. M. J. Thromb. Haemost.
2009, 7, 79.
8. (a) Salomon, O.; Steinberg, D. M.; Koren-Morag, N.; Tanne, D.; Seligsohn, U.
Blood 2008, 111, 4113; (b) Salomon, O.; Steinberg, D. M.; Dardik, R.; Rosenberg,
reduction in thrombosis with an EC50 = 2.8 lM. At the top dose, 21

478
L. M. Smith II et al. / Bioorg. Med. Chem. Lett. 26 (2016) 472–478
N.; Zivelin, A.; Tamarin, I.; Ravid, B.; Berliner, S.; Seligsohn, U. J. Thromb.
Haemost. 2003, 1, 658.
C. A.; Wei, Z.; Zheng, J. J.; Seiffert, D. A.; Wexler, R. R.; Quan, M. L. ACS Med.
Chem. Lett. 2015, 6, 590; (c) Pinto, D. J. P.; Smallheer, J. M.; Corte, J. R.; Austin, E.
J. D.; Wang, C.; Fang, T.; Smith, L. M., II; Rossi, K. A.; Rendina, A. R.; Bozarth, J.
M.; Zhang, G.; Wei, A.; Ramamurthy, V.; Sheriff, S.; Myers, J. E., Jr.; Morin, P. E.;
Luettgen, J. M.; Seiffert, D. A.; Quan, M. L.; Wexler, R. R. Bioorg. Med. Chem. Lett.
2015, 25, 1635.
9. (a) Wong, P. C.; Crain, E. J.; Watson, C. A.; Schumacher, W. A. J. Thromb.
Thrombolysis 2011, 32, 129; (b) Wong, P. C.; Quan, M. L.; Watson, C. A.; Crain, E.
J.; Harpel, M. R.; Rendina, A. R.; Luettgen, J. M.; Wexler, R. R.; Schumacher, W.
A.; Seiffert, D. A. J. Thromb. Thrombolysis 2015. Ahead of Print; (c) Quan, M.;
Wong, P. C.; Wang, C.; Woerner, F.; Smallheer, J. M.; Barbera, F. A.; Bozarth, J.
M.; Brown, R. L.; Harpel, M. R.; Luettgen, J. M.; Morin, P. E.; Peterson, T.;
Ramamurthy, V.; Rendina, A. R.; Rossi, K. A.; Watson, C. A.; Wei, A.; Zhang, G.;
Seiffert, D.; Wexler, R. R. J. Med. Chem. 2014, 57, 955.
12. Coordinates and structure amplitudes for the compound 6 (PDB ID 5E2O) and
compound 27 (PDB ID 5E2P), complexes were deposited into the Protein Data
Bank.
13. (a) Young, M. B.; Barrow, J. C.; Glass, K. L.; Lundell, G. F.; Newton, C. L.; Pellicore,
J. M.; Rittle, K. E.; Selnick, H. G.; Stauffer, K. J.; Vacca, J. P.; Williams, P. D.; Bohn,
D.; Clayton, F. C.; Cook, J. J.; Krueger, J. A.; Kuo, L. C.; Lewis, S. D.; Lucas, B. J.;
McMasters, D. R.; Miller-Stein, C.; Pietrak, B. L.; Wallace, A. A.; White, R. B.;
Wong, B.; Yan, Y.; Nantermet, P. G. J. Med. Chem. 2004, 47, 2995; (b) Howard,
N.; Abell, C.; Blakemore, W.; Chessari, G.; Congreve, M.; Howard, S.; Jhoti, H.;
Murray, C. W.; Seavers, L. C. A.; van Montfort, R. L. M. J. Med. Chem. 2006, 49,
1346.
10. Büller, H. R.; Bethune, C.; Bhanot, S.; Gailani, D.; Monia, B. P.; Raskob, G. E.;
Segers, A.; Verhamme, P.; Weitz, J. I. N. Engl. J. Med. 2015, 372, 232.
11. (a) Hangeland, J. J.; Friends, T. J.; Rossi, K. A.; Smallheer, J. M.; Wang, C.; Sun, Z.;
Corte, J. R.; Fang, T.; Wong, P. C.; Rendina, A. R.; Barbera, F. A.; Bozarth, J. M.;
Luettgen, J. M.; Watson, C. A.; Zhang, G.; Wei, A.; Ramamurthy, V.; Morin, P. E.;
Bisacchi, G. S.; Subramaniam, S.; Arunachalam, P.; Mathur, A.; Seiffert, D. A.;
Wexler, R. R.; Quan, M. L. J. Med. Chem. 2014, 57, 9915; (b) Hu, Z.; Wong, P. C.;
Gilligan, P. J.; Han, W.; Pabbisetty, K. B.; Bozarth, J. M.; Crain, E. J.; Harper, T.;
Luettgen, J. M.; Myers, J. E., Jr.; Ramamurthy, V.; Rossi, K. A.; Sheriff, S.; Watson,
14. Wong, P. C.; Crain, E. J.; Knabb, R. M.; Meade, R. P.; Quan, M. L.; Watson, C. A.;
Wexler, R. R.; Wright, M. R.; Slee, A. M. J. Pharmacol. Exp. Ther. 2000, 295, 212.