3,4-Dihydro-2H-1,4-benzoxazine derivatives combining thrombin inhibitory and glycoprotein IIb/IIIa receptor antagonistic activity as a novel class of antithrombotic compounds with dual function

Article abstract of DOI:10.1021/jm8003448

3,4-Dihydro-2H-1,4-benzoxazine derivatives possessing both thrombin inhibitory and glycoprotein IIb/IIIa (GPIIb/IIIa) receptor antagonistic activities were obtained by combining mimetics of the D-Phe-Pro-Arg pharmacophore of thrombin inhibitors and the Arg-Gly-Asp pharmacophore of GPIIb/IIIa receptor antagonists in the same low molecular weight peptidomimetic compound. Systematic variation of the position of substituents around the 3,4-dihydro-2H-1,4-benzoxazine nucleus, the distance between the carboxylate and amidine moieties, together with additional substituents to fill the thrombin S₂ and S₃ pockets resulted in compounds displaying submicromolar inhibition constants (K_i) for thrombin and submicromolar IC₅₀ for inhibition of binding of fibrinogen to platelet GPIIb/IIIa receptor. Some of these compounds, such as 17a, 17b, 17d, and 17h possessing a well balanced activity at both targets, are a good starting point for further optimization. Incorporation of anticoagulant and platelet antiaggregatory activity in the same molecule constitutes a promising approach toward novel antithrombotic agents.

Full text of DOI:10.1021/jm8003448

J. Med. Chem. 2008, 51, 5617–5629
5617
3,4-Dihydro-2H-1,4-benzoxazine Derivatives Combining Thrombin Inhibitory and Glycoprotein
IIb/IIIa Receptor Antagonistic Activity as a Novel Class of Antithrombotic Compounds with
Dual Function
†
†
†
‡
,†
ˇ
Janez Ilasˇ, Ziga Jakopin, Tina Borsˇtnar, Mojca Stegnar, and Danijel Kikelj*
Faculty of Pharmacy, UniVersity of Ljubljana, AsˇkercˇeVa 7, SI-1000 Ljubljana, SloVenia, and Department of Vascular Diseases, UniVersity
Medical Centre Ljubljana, Zalosˇka 7, SI-1525 Ljubljana, SloVenia
ReceiVed March 27, 2008
3,4-Dihydro-2H-1,4-benzoxazine derivatives possessing both thrombin inhibitory and glycoprotein IIb/IIIa
(GPIIb/IIIa) receptor antagonistic activities were obtained by combining mimetics of the D-Phe-Pro-Arg
pharmacophore of thrombin inhibitors and the Arg-Gly-Asp pharmacophore of GPIIb/IIIa receptor antagonists
in the same low molecular weight peptidomimetic compound. Systematic variation of the position of
substituents around the 3,4-dihydro-2H-1,4-benzoxazine nucleus, the distance between the carboxylate and
amidine moieties, together with additional substituents to fill the thrombin S2 and S3 pockets resulted in
compounds displaying submicromolar inhibition constants (Ki) for thrombin and submicromolar IC50 for
inhibition of binding of fibrinogen to platelet GPIIb/IIIa receptor. Some of these compounds, such as 17a,
17b, 17d, and 17h possessing a well balanced activity at both targets, are a good starting point for further
optimization. Incorporation of anticoagulant and platelet antiaggregatory activity in the same molecule
constitutes a promising approach toward novel antithrombotic agents.
Introduction
the blood coagulation cascade, are conversion of soluble
fibrinogen into insoluble fibrin, which forms a mechanical matrix
for the developing blood clot, and the activation of platelet
aggregation.8 The crystal structure of human thrombin9 dem-
onstrated the significance of the thrombin Tyr-Pro-Pro-Trp
insertion loop and the S1 binding pocket for binding and
selectivity and provided the basis for structure-based inhibitor
design. This resulted in many low molecular weight thrombin
inhibitors,2 most of them being mimetics of the D-Phe-Pro-Arg
sequence10 in which arginine (e.g., in argatroban11 ((2R,4R)-
1-[(S)-5-guanidino-2-((R)-3-methyl-1,2,3,4-tetrahydroquinoline-
8-sulfonamido)pentanoyl]-4-methylpiperidine-2-carboxylic acid))
or an arginine surrogate12 (e.g., in melagatran13 (ethyl 2-[[(1R)-
1-cyclohexyl-2-[(2S)-2-[[4-(N′-hydroxycarbamimidoyl)phenyl]-
methylcarbamoyl]azetidin-1-yl]-2-oxo-ethyl]amino]acetate) and
dabigatran14 (3-(2-((4-carbamimidoylphenylamino)methyl)-1-
methyl-N-(pyridin-2-yl)-1H-benzo[d]imidazole-5-carboxamido)-
propanoic acid)) plays a crucial role for binding into the
thrombin active site. Platelet aggregation is mediated by ionic
interaction of Arg-Gly-Asp (RGD) sequence of fibrinogen with
the platelet GPIIb/IIIa receptors (integrin RIIbꢀ3, glycoprotein
IIb/IIIa; GPIIb/IIIa, fibrinogen receptor) of the integrin family.
GPIIb/IIIa receptors are exposed on the surface of thrombocytes
upon activation with, for example, ADP, collagen, and
thrombin.5,15 Among numerous mimetics of the RGD sequence
found to be potent inhibitors of platelet aggregation, eptifi-
batide16 (N-6-(aminoiminomethyl)-N-2-(3-mercapto-1-oxopro-
pyl-L-lysylglycyl-L-R-aspartyl-L-tryptophyl-L-prolyl-L-cystein-
amide) and tirofiban17 ((2S)-2-(butylsulfonylamino)-3-[4-[4-(4-
piperidyl)butoxy]phenylpropanoic acid) have been introduced
into therapy. However, several compounds designed as oral
GPIIb/IIIa antagonists have failed in phase III of clinical trial
because of numerous side effects and even increased mortality
rate. Compounds such as sibrafiban ((S)-2-(1-(2-(4-carbamim-
idoylbenzamido)propanoyl)piperidin-4-yloxy)acetic acid), or-
bofiban ((R)-ethyl 3-(3-(1-(4-carbamimidoylphenyl)-2-oxopyr-
rolidin-3-yl)ureido)propanoate), xemilofiban ((R)-ethyl 3-(4-(4-
Cardiovascular diseases, with an estimated 18 million deaths
yearly, are the main cause of death and morbidity globally. If
appropriate action is not taken, by 2015 an estimated 20 million
people will die from cardiovascular disease every year, mainly
from heart attacks and strokes.1 There is, therefore, a growing
need to discover novel antithrombotic agents as alternatives to
existing treatment strategies. After the major progress made in
the past decade in developing novel antithrombotic agents, e.g.,
thrombin inhibitors,2 inhibitors of factor Xa,3 tissue factor/factor
VIIa inhibitors,4 and platelet GPIIb/IIIaa receptor antagonists,5
the search for novel antithrombotic drugs displaying a new
mechanism of action or making use of a new therapeutic concept
still remains a major challenge to medicinal chemistry.6
While hemostasis is a physiological process by which the
body stops blood loss whenever a blood vessel is severed or
ruptured, thrombosis is a pathological process in which hemo-
static mechanisms, i.e., blood coagulation and platelet aggrega-
tion, are activated in the absence of bleeding. Therefore,
inhibiting coagulation and preventing platelet aggregation at
different stages are essential components of most antithrombotic
therapeutic strategies. Of the several drug targets in the blood
coagulation cascade,7 e.g., thrombin, factor VIIa, and factor Xa,
thrombin has provided the most frequently used basis for the
design of novel anticoagulants over the past decade. The
principal physiological roles of thrombin, the final enzyme in
* To whom correspondence should be addressed. Phone: +386 1
4769561;. Fax: +386 1 4258031. E-mail: danijel.kikelj@ffa.uni-lj.si.
†
University of Ljubljana.
University Medical Centre Ljubljana.
‡
a Abbreviations: GPIIb/IIIa, glycoprotein IIb/IIIa; PDB, Protein Data
Bank; ELISA, enzyme-linked immuno sorbent assay; BTEAC, benzyltri-
ethylammonium chloride; TFA, trifluoroacetic acid; MNDO, modified
neglect of differential overlap; LGA, Lamarckian genetic algorithm; rmsd,
root mean square deviation; NIH, National Institutes of Health; BAEE, N-R-
benzoyl-L-arginine ethyl ester; HBSA, Hepes buffered saline with bovine
serum albumine; Tris, tris(hydroxymethyl)aminomethane; BSA, bovine
serum albumine; PRP, platelet-rich plasma; PPP, platelet-poor plasma.
10.1021/jm8003448 CCC: $40.75
2008 American Chemical Society
Published on Web 08/27/2008

5618 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18
Ilasˇ et al.
carbamimidoylphenylamino)-4-oxobutanamido)pent-4-
ynoate), and lotrafiban ((S)-2-(7-(4,4′-bipiperidine-1-carbonyl)-
4-methyl-3-oxo-2,3,4,5-tetrahydro-1H-benzo[e][1,4]diazepin-2-
yl)acetic acid) showed paradoxically procoagulant activity
mainly due to partial agonist activity at subthreshold GPIIb/
IIIa concentrations.18,19 Recently, crystal structures of the platelet
GPIIb/IIIa ectodomain in complex with fibrinogen-mimetic
therapeutics eptifibatide, tirofiban, and L-739758 (2-(S)-[N-(3-
pyridylsulfonyl)amino]-3-[[2- carbonyl-5-[2-(piperidin-4-yl)ethyl]-
thieno[2,3-b]thiopheneyl]amino]propionic acid) have been solved.
Comparison of these structures demonstrated that the drug-
binding pocket in GPIIb/IIIa is rigid, with the contacting residues
adopting the same conformation with or without the drug.20 This
provided the basis for our structure-based design of novel GPIIb/
IIIa receptor antagonists.
In a recent Letter to this journal we introduced a new type of
low molecular weight peptidomimetic antithrombotic drugs
possessing both thrombin inhibitory and GPIIb/IIIa receptor
antagonistic activities in the same molecule.21,22 Simultaneous
direct inhibition of different targets in the hemostatic system
by different substances is known in hematophageous animals,23
in which multiple inhibition of the blood coagulation enzymes
and platelet activation is frequently involved.24 In clinical
practice a combination of anticoagulant agent (e.g., warfarin
(4-hydroxy-3-(3-oxo-1-phenylbutyl)-2H-chromen-2-one), hep-
arin, or thrombin inhibitor) and antiaggregatory drug (e.g.,
acetylsalicylic acid, ticlopidine (3-[(2-chlorophenyl)methyl]-7-
thia-3-azabicyclo[4.3.0]nona-8,10-diene), clopidogrel ((S)-meth-
yl 2-(2-chlorophenyl)-2-(6,7-dihydrothieno[3,2-c]pyridin-5(4H)-
yl)acetate), or GPIIb/IIIa antagonist) is frequently used to
achieve an effective antithrombotic effect in patients with
valvular heart disease, prosthetic heart valves, or acute coronary
syndromes, in patients following myocardial infarction, or in
patients undergoing thrombolysis.25 In these clinical situations
the combination of anticoagulant and antiplatelet therapy has
an additive effect by suppressing both blood coagulation and
platelet function and is thus more effective than treatment
directed against either thrombin or platelets alone. Efficient
combination of anticoagulant and antiplatelet activity in the same
molecule would produce a novel type of antithrombotic drug
featuring substantial advantages over possible combinations of
anticoagulant and antiplatelet agents, including more predictable
and less complex pharmacokinetics, lower incidence of side
effects, less demanding clinical studies, and more straightfor-
ward registration procedure, which together could render them
the antithrombotic drugs of the future. In this paper we report
at length on the design, synthesis, and in vitro biological activity
of novel 3,4-dihydro-2H-1,4-benzoxazine derivatives that act
both as thrombin inhibitors and GPIIb/IIIa receptor antagonists
and possess a well balanced submicromolar potency against both
targets.
Figure 1. Design of 3,4-dihydro-2H-1,4-benzoxazine derivatives with
thrombin inhibitory and GPIIb/IIIa receptor antagonistic activities.
with Asp189 in the thrombin S1 pocket and to act as a cationic
center for binding to the GPIIb/IIIa receptor, (ii) a carboxylate
group providing ionic or dipolar interaction with the GPIIb/
IIIa receptor, (iii) a central scaffold that would interact with
the thrombin YPPW loop and also provide a ∼1.5 nm spacer
between the two charged groups required for binding to the
GPIIb/IIIa receptor,5,15 and (iv) an aromatic ring in the proximity
of the carboxylate group, required for interaction with the
thrombin S3 binding pocket and to provide a hydrophobic
interaction with the nonpolar binding site of the GPIIb/IIIa
receptor.28 A compromise would be required of the central
scaffold since, owing to the architecture of the enzyme active
site, thrombin inhibitors have to be bent between the P1 and P2
moieties, whereas GPIIb/IIIa receptor antagonists are more
effective with a stretched conformation of the linker joining
the charged groups. Additionally, the central scaffold should,
if possible, provide interactions with the key amino acid residues
in the thrombin active site such as Gly216. On the basis of
preliminary docking experiments, 3,4-dihydro-2H-1,4-benzox-
azine was selected as the central scaffold that, because of its
manifold possibilities of functionalization, should allow for
appropriate positioning of side chains in space. Despite its strong
basicity, which usually results in poor bioavailability, benza-
midine was chosen as the arginine mimetic12 for the first
generation of antithrombotic compounds with dual action, since
it is most suitable for forming a salt bridge with Asp189 in the
thrombin S1 pocket. The benzamidine group also reduces plasma
protein binding, thus improving activity. The problem of the
poor bioavailability of compounds containing a benzamidine
group can be overcome in the future by a prodrug approach, as
shown in the case of ximelagatran13 and dabigatran.14
Results and Discussion
Design. The presence of the arginine moiety in the pharma-
cophores D-Phe-Pro-Arg and Arg-Gly-Asp, the occurrence of
the terminal carboxylic group in the RGD sequence, and the
good tolerability of the P3 carboxylic group26 in some thrombin
inhibitors, e.g., melagatran13c and dabigatran,14 suggested to us
the idea of incorporating the mimetics of both amino acid motifs
into a single molecule that would bind with the same moieties
to the thrombin active site or to the platelet GPIIb/IIIa receptor
(Figure 1), thus featuring a designed multiple ligand with highly
integrated pharmacophores.27 Such dual acting antithrombotic
compounds should incorporate (i) an arginine mimetic to interact
Encouraged by the promising dual antithrombotic activity of
compound 1b (Ki(thrombin)) ) 14.9 µM, IC50(GPIIb/IIIa) ) 1.64
µM)21 (Figure 2), we initiated a synthetic program, the purpose
of which was systematic variation of substituents at positions 2

NoVel Antithrombotic Compounds with Dual Function
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18 5619
O-alkylated with 4-cyanobenzyl bromide under phase transfer
conditions in the presence of potassium carbonate, which gave
better yields than the previously published method.21 The nitro
derivative 3a, possessing a 4-cyanobenzyloxy group that is easily
removable under catalytic hydrogenation conditions, was con-
verted to amine 4a using sodium borohydride and copper(II)
acetate as catalyst32 (Scheme 1). Aminolysis of diethyl 2-phe-
nylmalonate with amine 4a under microwave irradiation gave
carboxamide 6a in a short time and excellent yield. This was
converted to amidine 7a using Pinner reaction33 (Scheme 2).
The carboxylic acid 4b, obtained by alkaline hydrolysis of 3b,
was transformed to the acyl chloride, which was reacted with
aniline to give the amide 5b. This was subsequently alkylated
with ethyl bromoacetate to give the nitrile 6b, which was finally
converted to the amidine 7b. Coupling of carboxylic acid 4b
with ethyl 3-aminopropionate gave the amide 5d, which was
converted in two steps, by Pinner reaction and subsequent
alkaline hydrolysis of the ethyl ester, to the zwitterionic
compound 8d (Scheme 3).
Figure 2
Scheme 1a
The target compounds 17 and 18 with the 3,4-dihydro-2H-
1,4-benzoxazine scaffold and 2-oxymethylene spacer were
synthesized as shown in Schemes 4–7. Ethyl 2,4-dimethyl-6/
7-nitro-3-oxo-3,4-dihydro-2H-1,4-benzoxazine-2-carboxylates
9a and 9b were prepared from 2-amino-4-nitrophenol and
2-amino-5-nitrophenol according to a published procedure.34
Reduction of esters 9a and 9b with borane dimethyl sulfide
complex proceeded with concomitant reduction of the lactam
carbonyl group and retention of the nitro group, affording
alcohols 10a and 10b. Alcohol 10a was activated for nucleo-
philic substitution as tosylate 11a, which was transformed to
the ether 12a with sodium 4-cyanophenolate. Alternatively,
alcohols 10a and 10b were directly transformed to ethers 12a
and 12b with 4-cyanophenol under Mitsunobu reaction condi-
tions.35 The overall yields were comparable; however, a one-
step Mitsunobu reaction was more favorable because it de-
manded less workup. Reduction of the nitro group in 12a and
12b, using catalytic hydrogenation, afforded aromatic amines
13a and 13b that were N-alkylated, using benzaldehyde or
2-pyridinecarboxaldehyde and sodium triacetoxyborohydride as
reducing agent,36 to afford secondary amines 14a-c or acylated
with benzoyl chloride to give benzamide 15a. The amines 14a
and 14b were alkylated with ethyl bromoacetate to afford tertiary
amines 16a and 16b, while alkylation with ethyl 3-bromopro-
pionate did not give the desired product. Acetylation of amines
14a-c with various acyl chlorides gave amides 16c-g.
Similarly, alkylation of benzamide 15a with ethyl bromoacetate
gave the N-disubstituted benzamide 16h, while reaction with
ethyl 3-bromopropionate was again unsuccessful. Target com-
pounds 17a-h were prepared from nitriles 16a-h using the
Pinner reaction.33 Amidines 17c-e were converted to the
zwitterionic compounds 18c-e by alkaline hydrolysis. Loss of
acetate group was observed during attempted alkaline hydrolysis
in the case of N-benzylglycine derivatives 17a and 17b; hence,
the desired products 18a and 18b were not obtained. (Scheme
4). Aromatic amines 14a and 14b were also acylated with
monoalkyl 2-benzylmalonates using the EDC/HOBt method to
give carboxamides 16i and 16j (Scheme 5). Unsubstituted
carboxamide 15a was also converted to amidine 17k in order
to explore the effect of N-alkylation on biological activity
(Scheme 6). N-(ω-Carboxypropanoyl) derivative 18l was ob-
tained by heating 13a with succinanhydride in THF to give 16l
and subsequent Pinner reaction followed by hydrolysis affording
zwitterionic compound 17l (Scheme 7).
a (a) 4-(Bromomethyl)benzonitrile, BTEAC, K2CO3, CH3CN, 60 °C, 8 h;
(b) NaBH4, Cu(OAc)2, MeOH, room temp, 2 h; (c) 1.5 M NaOH, H2O,
EtOH, room temp, 6 h.
Scheme 2a
a (a) Diethyl 2-phenylmalonate, microwave, 120 °C, 15 min; (b) HCl(g)
EtOH, 0 °C, 30 min, then NH4OAc, EtOH, room temp, 24 h.
,
and 7, with the aim of achieving a well balanced potency for
both targets. All compounds were modeled using HyperChem29
to determine the distances between the cationic and anionic
centers in the minimized conformations and docked to the
thrombin active site (PDB code 1KTS)14 and GPIIb/IIIa receptor
(PDB code 1TY5)20 with AutoDock30 in order to predict their
propensity for binding to both targets. Of the envisaged
compounds with a distance of 1.5 nm between the carbon atoms
of the carboxylate and amidino groups,5,15 modeling predicted
appropriate binding also for 6-substituted 1,4-benzoxazine
derivatives bearing an arginine mimetic moiety at position 2.
Our synthetic efforts were limited, therefore, to compounds
possessing 2,6- and 2,7 substitution patterns.
Chemistry. Compounds containing the 2H-1,4-benzoxazin-
3(4H)-one scaffold were synthesized as depicted in Schemes
1–3. 2-Hydroxy-4-methyl-2H-1,4-benzoxazin-3(4H)-one deriva-
tives 2a and 2b were prepared from 2-aminophenol derivatives
in a five-step synthesis involving N-acylation, cyclization,
N-methylation, bromination, and bromine substitution, with an
overall yield of 83%.31 Compounds 2a and 2b were then

5620 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18
Ilasˇ et al.
Scheme 3a
a (a) Ethyl 3-aminopropanoate, EDC, HOBt, DMF, room temp, overnight; (b) SOCl2, CH2Cl2, reflux, 4 h; (c) aniline, Et3N, DMF, overnight; (d)
BrCH2COOEt, BTEAC, K2CO3, CH3CN, 60 °C, 24 h; (e) HCl(g), EtOH, 0 °C, 30 min, room temp, 48 h, then NH4OAc, EtOH, room temp, 48 h; (f) 1.5 M
NaOH, H2O, EtOH, room temp, 6 h.
Biological Activity. In a previous communication21 we
demonstrated basic structure-activity relationships in balancing
GPIIb/IIIa receptor antagonist and thrombin inhibitory activities
in potential dual antithrombotic compounds based on the central
2H-1,4-benzoxazin-3(4H)-one core. First, we continued our
work on optimizing the P3 moiety of compounds 1a and 1b
(Table 1). Although the inverse amide derivative 7a did not
improve thrombin inhibition, shortening the alkyl chain by one
carbon atom had a positive effect on GPIIb/IIIa receptor
antagonistic activity (IC50 ) 2.44 µM). The more rigid analogue
7b showed no improvement of thrombin inhibitory activity over
1a; however, the acid 8b (R′ ) H) was a 2-fold better thrombin
inhibitor than the respective ester analogue 7b (R′ ) Et). This
trend was the reverse of that of the thrombin inhibitory activities
of 1a and 1b, where the carboxylic acid derivative 1b had a
4-fold lower activity than the ester analogue 1a. This suggests
that in 1a the ethyl moiety forms additional hydrophobic
interactions with the protein surface, while in 7b and 8b the
carboxylate moiety points to the water environment and the
carboxylic acid group is therefore more favorable. Compound
7c was prepared to examine the influence of the carboxylic acid
moiety on biological activity. It had weaker thrombin inhibitory
activity than 7b and 8b, showing that the introduction of a
carboxylic acid moiety improves the thrombin inhibitory activity.
Surprisingly, 7c was not devoid of GPIIb/IIIa receptor antago-
nistic activity but actually showed a 2-fold better IC50 value
than 7b. This might be due to a favorable interaction of the
aniline moiety of 7c with the nonpolar binding site of the GPIIb/
IIIa receptor,28 whereas the sterically more restricted N-[(1,4-
benzoxazin-7-yl)carbonyl]-N-phenylglycine moiety of 7b does
not allow the optimal positioning of the carboxylate group, the
benzoxazinone core, and the phenyl ring in this special case.
Compound 8d, which lacks the P3 aromatic moiety, showed,
as expected, very weak thrombin inhibitory activity and 9-fold
lower GPIIb/IIIa receptor antagonistic activity than 1b, indicat-
ing the importance of an aromatic moiety in the vicinity of the
carboxylic group for binding to GPIIb/IIIa receptor. Surprisingly,
the carboxylic acid derivative 8b showed no inhibition in a
modified ELISA assay but, as expected, had greater activity
than ester 7b in an aggregation inhibition assay. These
unexpected results, i.e., inactivity of 8b on GPIIb/IIIa receptor,
stronger GPIIb/IIIa receptor antagonistic activity of 7c than that
of 7b, generally low thrombin inhibitory activity, and low
selectivity toward factor Xa and trypsin, led to the decision to
abandon the 2-hydroxy-4-methyl-2H-1,4-benzoxazin-3(4H)-one
scaffold. Compounds with a 2-amino-4-methyl-2H-1,4-benzox-
azin-3(4H)-one scaffold and a 2-methylamino linker containing
an O,N-acetal moiety were also considered, but they were found
to be unstable under alkaline and acidic conditions which
hindered their successful preparation.31
Molecular modeling showed that 3,4-dihydro-2H-1,4-ben-
zoxazines with a flexible 2-oxymethylene spacer have the
potential to be suitably accommodated by both targets. Crystal
structures of the GPIIb/IIIa receptor-tirofiban18 and
thrombin-ximelagatran14 complexes show that tirofiban binds
in an extended conformation and ximelagatran in a bent
conformation because of the different geometries of the two
target binding sites. Docking experiments indicated that in order
to achieve the needed flexibility, a spacer of two atoms is
required between the benzamidine moiety and the heterocyclic
core. Compounds with the shorter and less flexible methylene
spacer were also prepared but were devoid of GPIIb/IIIa receptor
antagonistic activity, despite being very potent and selective
thrombin inhibitors.37 Modification of the P3 part containing
the aromatic and carboxylic acid moieties, which are supposed
to be crucial for dual activity, was the main focus of our
optimization strategy of compounds with a 2-oxymethylene
spacer. Molecular modeling studies predicted that introduction
of the P3 moiety at position 6 of a 1,4-benzoxazine scaffold
would lead to a more pronounced GPIIb/IIIa receptor antago-

NoVel Antithrombotic Compounds with Dual Function
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18 5621
Scheme 4a
a (a) Me2S·BH3, THF, reflux, overnight; (b) TsCl, pyridine, room temp, 24 h; (c) NaOC6H4-4-CN, DMF, 80 °C, 24 h; (d) 4-cyanophenol, PPh3, DIAD,
THF, reflux, 48 h; (e) H2, Pd/C, THF, room temp, 2 h; (f) benzaldehyde or 2-pyridinecarbaldehyde, NaBH(AcO)3, 1,2-dichloroethane, room temp, 6 h; (g)
benzoyl chloride, Et3N, THF, room temp, overnight; (h) BrCH2COOEt, BTEAC, K2CO3, CH3CN, 60 °C, 24 h; (i) acyl chloride, Et3N, CH2Cl2, room temp,
overnight; (j) HCl(g), EtOH, 0 °C, 30 min, then NH4OAc, EtOH, room temp, 24 h; (k) 1.5 M NaOH, H2O, EtOH, room temp, 6 h.
nistic activity, while introducing it at position 7 would result in
a more pronounced thrombin inhibitory activity. We therefore
prepared both 2,7- and 2,6-disubstituted-2,4-dimethyl-3,4-di-
hydro-2H-1,4-benzoxazines with a 2-oxymethylene spacer in
order to allow the dual activity of the compounds to be tuned
by simple variation of the substitution pattern.
The results of biological testing of synthesized compounds
are presented in Table 2. In the resulting series of 7-benzylamino
derivatives (compound 17a) and 7-acylamino derivatives (com-
pounds 17c and 17h), compound 17a bearing the P3 N-
(ethoxycarbonylmetyl)benzylamino moiety was the least active
(Ki(thr) ) 0.38 µM) as a thrombin inhibitor. Conformational
restriction, effected by introducing an oxo functionality on the
benzylic CH2 moiety, increased thrombin inhibitory potency by
2-fold (17h, Ki(thr) ) 0.156 µM), whereas the oxo substituent
attached to the acetate CH2 moiety afforded an even more potent
thrombin inhibitor (17c, Ki(thr) ) 0.11 µM). All three compounds
possess low micromolar IC50 values for GPIIb/IIIa receptor
antagonistic activity, thus confirming that the 2,7-disubstitution
pattern favors thrombin inhibition. Extending the N-acyl chain
in compound 17e increased thrombin inhibitory activity but
reduced the GPIIb/IIIa receptor antagonistic activity. Substitution
of benzyl moiety in 17e with pyridin-2-ylmethyl moiety to give
17g improved thrombin inhibitory activity 2-fold (Ki(thr) ) 25
nM) and also improved GPIIb/IIIa receptor antagonistic activity.
Compounds 17b and 17d, bearing a P3 moiety at position 6,
had improved GPIIb/IIIa receptor antagonistic activity but with
concomitantly reduced thrombin inhibitory activity. Compound
17d possessed the most potent and the best balanced dual
antithrombotic activity (Ki(thr) ) 0.32 µM; IC50(GPIIb/IIIa) ) 1.2
µM).
Compounds 17i, 18i, and 17j possessing more distant P3
aromatic moiety showed reduced thrombin inhibitory activity,
proving the importance of optimal position of P3 aromatic
moiety for good thrombin inhibition. Compound 18i showed
good platelet aggregation inhibitory activity (IC50 ) 24.9 µM).
Comparison of activities of compounds 17h with that of 17k
shows that the introduction of a carboxylic acid moiety improved
the thrombin inhibitory activity by an order of magnitude while
surprisingly only slightly improving the GPIIb/IIIa receptor

5622 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18
Ilasˇ et al.
Scheme 5a
and 18e unexpectedly displayed much weaker GPIIb/IIIa
receptor antagonistic activity in a modified ELISA assay than
their ester analogues 17c, 17d, and 17e and showed good activity
in an aggregation assay. Similar behavior had already been
observed with compound 8b and could suggest that these
compounds probably affect the modified ELISA test in an
unknown way. The fact that both carboxylic acids 18 and esters
17 display GPIIb/IIIa receptor antagonistic activity could be
explained by their ion-dipole interaction with the cationic site
of the GP IIb/IIIa receptor. 6-Isomers 17b and 17d possess better
GPIIb/IIIa receptor antagonistic activity than analogous 7-iso-
mers 17a and 17c, which is opposite the trend observed with
thrombin inhibitory activity. The oxo substituent attached to
the acetate CH2 moiety slightly reduced the GPIIb/IIIa receptor
antagonistic activity (1.4-fold in the case of 7-isomer 17c and
1.6-fold in the case of 6-isomer 17d). Introduction of an oxo
functionality to the benzylic CH2 moiety (17h) did not influence
the activity compared with less rigid compound 17a. With the
prolongation of acyl chain, the GPIIb/IIIa receptor antagonistic
activity is reduced (17e and 18e) as expected from molecular
modeling.
a (a) 2-Benzyl-3-alkyloxy-3-oxopropanoic acid, EDC, HOBt, DMF, room
temp, overnight; (b) HCl(g), EtOH, 0 °C, 30 min, then NH4OAc, EtOH,
room temp, 24 h; (c) 1.5 M NaOH, H2O, EtOH, room temp, 6 h.
Scheme 6a
The biological results were explained by docking compounds
7, 8, 17, and 18 to the thrombin active site14 and to the GPIIb/
IIIa receptor binding site.18 The (2R)-isomers displayed better
docking allignments than the (2S)-isomers. The conformations
of compounds 7, 8, 17, 18 docked to both targets are similar
with respect to the bend between benzamidine and benzoxazine
moieties. (Figures 3 and 4 docking results for 18c and 18d are
shown). Compound (R)-18c binds to the former in a similar
manner to ximelagatran. The distance between the C-atoms of
the Asp189 carboxylate and the amidine group in (R)-18c was
4.0 Å, and two hydrogen bonds linked the amidine hydrogens
and carboxylate oxygen atoms. An additional hydrogen bond
was formed between benzamidine and the Gly219 backbone.
The 3,4-dihydro-2H-1,4-benzoxazine scaffold was located in the
S2 binding pocket and the benzyl group in the lipophilic S3
binding pocket, with the P3 carboxylate stretching outward from
the thrombin surface. The oxygen of the 1,4-oxazine ring formed
a hydrogen bond (2.72 Å) with the Gly216 backbone, which is
often regarded as a key interaction in the thrombin active site.38
The oxymethylene spacer between benzamidine and the 3,4-
dihydro-2H-1,4-benzoxazine moiety allows for flexibility, which
enables the compound to assume a bent conformation in the
thrombin binding site. Compound (R)-18d is also positioned
similarly in the thrombin active site; however, the proximity of
1,4-benzoxazine aromatic ring and P3 phenyl ring does not allow
optimum positioning of the P3 aromatic moiety of (R)-18d,
which results in 2-fold lower potency of the 6-isomer. In the
cases of 17e and 17f, where 17e is a more optimized thrombin
inhibitor (Ki ) 60 nM), the difference in the potency of 7-isomer
and 6-isomer is bigger (11-fold). Compound 17i, possessisng a
2-benzylmalonyl moiety, binds also in similar fashion, the only
difference being that the more distant benzyl group is not
optimally bound to the S3 pocket, which accounts for the much
lower thrombin inhibitory potency of this compound.
a (a) HCl(g), EtOH, 0 °C, 30 min, then NH4OAc, EtOH, room temp,
24 h.
Scheme 7a
a (a) Succinanhydride, THF, 60 °C, 60 min; (b) HCl(g), EtOH, 0 °C, 30
min, then NH4OAc, EtOH, room temp, 24 h; (c) 1.5 M NaOH, H2O, EtOH,
room temp, 6 h.
antagonistic activity. Similar to the activity of 7c, the activity
of 17k might be due to a favorable interaction of the benzoyl
moiety of 17k with the nonpolar binding site of the GPIIb/IIIa
receptor.28 Comparison of compounds 17e (R′ ) Et) and 18e
(R′ ) H) with 17l (R′ ) Et) and 18l (R′ ) H) shows that
omitting the aromatic P3 moiety reduced the thrombin inhibitory
activity by an order of magnitude while it improved the GPIIb/
IIIa receptor antagonistic activity. While carboxylic acid 18l
displayed stronger GPIIb/IIIa receptor antagonistic activity in
a modified ELISA assay and better activity in an aggregation
assay than its ester analogue 17l, carboxylic acids 18c, 18d,
Analysis of the binding mode of (R)-18d in the binding site
of GPIIb/IIIa receptor shows that it takes an extended conforma-
tion with a break that is forced by the conformational restriction
on position 2 of 1,4-benzoxazine. The benzamidine moiety
hydrogen-bonds with Asp224 and Phe160 from the RIIb subunit
of the GPIIb/IIIa receptor and makes hydrophobic interactions
with Tyr190, Phe231, and Phe191 of the RIIb subunit. The
oxygen atom of the 1,4-oxazine ring forms a hydrogen bond
with Ala218. The carboxylic acid moiety makes electrostatic

NoVel Antithrombotic Compounds with Dual Function
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18 5623
Table 1. Biological Activity of 1,4-Benzoxazin-3(4H)-one Derivatives 7 and 8: Inhibition of Serine Proteases Thrombin, Factor Xa, and Trypsin,
Inhibition of Fibrinogen Binding to GP IIb/IIIa Receptors, and Inhibition of ADP Induced Platelet Aggregation
a Selectivity for thrombin vs trypsin. b Selectivity for thrombin vs FXa. c Percent of inhibition of ADP induced platelet aggregation at 50 µM inhibitor
concentration.
interactions with the Mg2+ atom and hydrogen-bonds to the
TLC was performed on silica gel Merck 60 F254 plates (0.25 mm),
using visualization with ultraviolet light and ninhydrin. Column
chromatography was carried out on silica gel 60 (particle size
240-400 mesh). Microwave assisted reactions were performed
using a CEM Discover microwave reactor (CEM Corp.). Melting
points were determined on a Reichert hot stage microscope and
are uncorrected. 1H NMR and 13C NMR spectra were recorded at
300 and 75 MHz, respectively, on a Bruker AVANCE DPX300
spectrometer in CDCl3 or DMSO-d6 solution with TMS as the
internal standard. Spectra were assigned using gradient COSY,
HSQC, and HMBC experiments. IR spectra were recorded on a
Perkin-Elmer 1600 FT-IR spectrometer. Microanalyses were per-
formed on a Perkin-Elmer C, H, N analyzer 240 C. Analyses
indicated by the symbols of the elements were within (0.4% of
the theoretical values. Mass spectra were obtained using a VG-
Analytical Autospec Q mass spectrometer. HPLC analyses were
performed on an Agilent Technologies HP 1100 instrument with
G1365B UV-vis detector (254 nm), using a Luna C18 column
(4.6 mm × 250 mm) at flow rate 1 mL/min. The eluant was a
mixture of 0.1% TFA in water (A) and acetonitrile (B). Gradient
was from 10% B to 80% B in 30 min.
General Procedure for Preparing Derivatives 3a and 3b. A
stirred suspension of 4-cyanobenzyl bromide (0.77 g, 3.94 mmol),
2-hydroxy-4-methyl-2H-1,4-benzoxazin-3(4H)-one derivative 2a or
2b (3.94 mmol), benzyltriethylammonium chloride (0.90 g, 3.94
mmol), and potassium carbonate (1.36 g, 9.85 mmol) in acetonitrile
(70 mL) was heated at 60 °C for 8 h. The suspension was filtered
and the filtrate evaporated under reduced pressure. The residue was
dissolved in ethyl acetate (150 mL) and washed successively with
10% citric acid (3 × 50 mL), saturated solution of NaHCO3 (2 ×
amide proton of Asn215 of the ꢀ3 subunit of the GPIIb/IIIa
receptor. Compound (R)-18c adopts a similar conformation, the
main difference being in the positioning of the 1,4-benzoxazine
moiety, which is more twisted toward the water environment.
Conclusion
In conclusion, we have described the design, synthesis, and
dual activity of several new 3,4-dihydro-2H-1,4-benzoxazine
compounds capable of acting as both thrombin inhibitors and
GPIIb/IIIa receptor antagonists and analyzed the structure-activity
relationship of combining anticoagulant and antiaggregatory
activity into one molecule. We optimized flexibility between
benzamidine moiety and central 1,4-benzoxazine scaffold to
ensure desired activity on both targets. To improve balanced
dual activity on thrombin and platelet GPIIb/IIIa receptor, a
compromise concerning flexibility and bulkiness in P3 part
containing aromatic and carboxylic acid moieties was sought.
Compounds 17a, 17b, 17d, and 17h, which have the most potent
and well balanced dual antithrombotic activity, close to the
nanomolar range, can serve as a lead compounds for the next
generation of dual antithrombotic agents, making use of
established binding modes and structure-activity relationships.
Experimental Section
General. Chemicals were obtained from Acros, Aldrich Chemi-
cal Co., and Fluka and used without further purification. THF was
kept over sodium and distilled immediately prior to use. Analytical

5624 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18
Ilasˇ et al.
Table 2. Biological Activity of 3,4-Dihidro-1,4-benzoxazine Derivatives 17 and 18: Inhibition of Serine Proteases Thrombin, Factor Xa, and Trypsin,
Inhibition of Fibrinogen Binding to GP IIb/IIIa Receptors, and Inhibition of ADP Induced Platelet Aggregatione
a Selectivity for thrombin vs trypsin. b Selectivity for thrombin vs FXa. c Percent of inhibition of ADP induced platelet aggregation at 50 µM inhibitor
concentration or IC50. d Compound did not show any inhibition at 50 µM. e nd ) not determined.
50 mL), and saturated solution of NaCl (1 × 50 mL). The organic
phase was dried with Na2SO4 and the solvent evaporated under
reduced pressure to obtain a solid product. If necessary, the product
was recrystallized from ethyl acetate/petroleum ether.
4-[(4-Methyl-7-nitro-3-oxo-3,4-dihydro-2H-1,4-benzoxazin-2-
yloxy)methyl]benzonitrile (3a). Yellow crystals; yield, 1.26 g (94%);
mp 135-136 °C. 1H NMR (CDCl3): δ ) 3.32 (s, 3H, N-CH3),
3
4.95 (s, 2H, OCH2), 5.88 (s, 1H, 2-H), 7.43 (d, 2H, J ) 8.3 Hz,

NoVel Antithrombotic Compounds with Dual Function
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18 5625
microwave reactor for 15 min at 120 °C. Dichloromethane (5 mL)
was added and the product precipitated with 70 mL of petroleum
ether. The crude product was redissolved in 3 mL of dichlo-
romethane and precipitated again with 70 mL of petroleum ether
to give yellow crystals. Yield, 0.475 g (95.2%); mp 71-73 °C. 1H
NMR (DMSO-d6: δ ) 1.15 (t, 3H, 3J ) 7.1 Hz, CH2-CH3), 3.37
(s, 3H, N-CH3), 4.08-4.21 (m, 2H, CH2-CH3), 4.90 (4.91) (s, 2H,
3
4
OCH2), 5.64 (dd, 1H, J ) 6.8 Hz, J ) 1.5 Hz, Ph-CH), 5.75
(5.76) (s, 1H, 2-H), 7.33 (d, 1H, 3J ) 8.6 Hz, Ar-H5), 7.36-7.54
3
(m, 5H, Ph), 7.49 (d, 2H, J ) 8.0 Hz, Ar-H2′,H6′), 7.64 (d, 1H,
4J ) 1.5 Hz, Ar-H8), 7.74-7.80 (m, 3H, Ar-H6, Ar-H3′,H5′)
ppm. 13C NMR (DMSO-d6: δ ) 170.39 (170.40) (COOEt), 156.03
(156.07) (NHCO), 159.66 (159.67) (C-3), 142.26 (142.28) (C-1′),
140.17 (C-8a), 135.97 (136.03) (C-1′′), 132.15 (132.17) (C-3′, C-5′),
131.08 (131.10) (C-4a), 128.74 (128.75) (C-7), 128.43 (C-2′, C-6′),
128.34, 128.15 (Ph), 123.11 (C-6), 118.54 (CN), 116.82 (116.85)
(C-8), 114.91 (C-5), 110.49 (110.50) (C-4′), 94.79 (94.82) (C-2),
69.00 (69.07) (OCH2-Ar), 60.81 (CH2-CH3), 56.96 (57.00)
(PhCH), 28.18 (N-CH3), 13.86 (CH2-CH3) ppm. MS (ESI): m/z
(%) ) 500.2 (MH+, 100), 454.1 (10). IR (KBr): ν ) 3436, 2229,
1707, 1616, 1498, 1364, 1222, 1179, 1092, 1056 cm-1. Anal.
(C28H25N3O6) C, H, N.
Figure 3. Compounds (R)-18c and (R)-18d docked in the active site
of thrombin.
7-[(2-Ethoxy-2-oxoethyl)(phenyl)carbamoyl]-2-(4-cyanobenzy-
loxy)-4-methyl-2H-1,4-benzoxazin-3(4H)-one (6b). 6b was synthe-
sized from 5b (2.23 g, 5.38 mmol) according to the general
procedure for alkylation under phase-transfer catalysis. The crude
product was purified by column chromatography using dichlo-
romethane/methanol (50:1) as eluant to give yellow crystals. Yield,
1
2.31 g (86.0%); mp 63-64 °C. H NMR (CDCl3): δ ) 1.32 (t,
3
3
3H, J ) 7.1 Hz, CH2CH3), 3.36 (s, 3H, N-CH3), 4.26 (q, 2H, J
2
) 7.1 Hz, CH2CH3), 4.60 (s, 2H, N-CH2COO), 4.73 (d, 1H, J )
12.9 Hz, OCH2-Ar), 4.67 (d, 1H, 2J ) 12.9 Hz, OCH2-Ar), 5.41
(s, 1H, 2-H), 6.86 (d, 1H, 3J ) 8.4 Hz, Ar-H5), 7.04 (d, 1H, 4J )
1.9 Hz, Ar-H8), 7.11-7.24 (m, 5H, Ph, Ar-H2′,H6′), 7.34 (d, 2H,
3J ) 8.3 Hz, Ar-H3′,H5′), 7.62-7.65 (m, 2H, Ph) ppm. MS (EI):
m/z (%) ) 517.2 (MH+, 100), 421.2 (15), 198.2 (45), 196.2 (55),
141.0 (95). IR (KBr): ν ) 2980, 2228, 1746, 1697, 1615, 1380,
1323, 1201, 1088, 1032 cm-1. Anal. (C28H25N3O6) C, H, N.
Ethyl 3-[2-(4-Carbamimidoylbenzyloxy)-4-methyl-3-oxo-3,4-di-
hydro-2H-1,4-benzoxazin-7-ylamino]-3-oxo-2-phenylpropanoate (7a).
7a was synthesized from 6a according to the general procedure
for preparation of amidines from nitriles (Pinner reaction). Brown
Figure 4. Compounds (R)-18c and (R)-18d docked in the binding site
of the GPIIb/IIIa receptor.
Ar-H2′,H6′), 7.47 (d, 1H, 3J ) 9.0 Hz, Ar-H5), 7.72 (d, 1H, 4J )
2.6 Hz, Ar-H8), 7.78 (d, 2H, 3J ) 8.3 Hz, Ar-H3′,H5′), 8.04 (dd,
3
4
1H, J ) 9.0 Hz, J ) 2.6 Hz, Ar-H6) ppm. 13C NMR (CDCl3):
δ ) 159.51 (C-3), 142.63 (C-1′), 142.23 (C-7), 140.51 (C-8a),
134.43 (C-4a), 132.15 (C-3′, C-5′), 128.43 (C-2′, C-6′), 119.15
(CN), 118.50 (C-6), 115.71 (C-5), 112.73 (C-8), 110.59 (C-4′),
94.81 (C-2), 69.56 (OCH2-Ar), 28.52 (N-CH3) ppm. MS (EI): m/z
(%) ) 339 (M+, 10), 195 (100), 149 (49). IR (KBr): ν ) 2224,
1692, 1601, 1525, 1384, 1340, 1223, 1083, 1029, 890 cm-1. Anal.
(C17H13N3O5) C, H, N.
1
oil; yield, 157 mg (27%). H NMR (DMSO-d6): δ ) 1.18 (t, 3H,
3
3J ) 7.1 Hz, CH2-CH3), 3.30 (s, 3H, N-CH3), 4.15 (q, 2H, J )
7.1 Hz, CH2-CH3), 4.89 (s, 2H, OCH2), 5.08 (s, 1H, Ph-CH),
3
5.66 (s, 1H, 2-H), 7.18 (d, 1H, J ) 8.9 Hz, Ar-H5), 7.28-7.41
(m, 4H, Ph-H2′′,H6′′, Ph-H4′′, Ar-H6), 7.42-7.53 (m, 5H, Ar-H8,
Ph-H3′,H5′, Ar-H3′′,H5′′), 7.78 (dd, 2H, 3J ) 8.0 Hz, 4J ) 2.3 Hz,
Ar-H2′,H6′), 9.40 (br s, 4H, amidino H), 10.89 (br s, 1H, CONH)
ppm. 13C NMR (DMSO-d6): δ ) 168.40 (COOEt), 156.75
(NHCO), 165.35 (C(dNH2)NH2), 159.24 (C-3), 142.66 (C-1′),
140.59 (C-8a), 134.95 (C-7), 134.17 (C-1′′), 129.21 (C-3′′, C-5′′),
128.04, 127.81, 127.49 (Ph, C-2′, C-6′, C-2′′, C-6′′), 124.16 (C-
4′), 115.31 (C-5), 113.71 (C-6), 108.42 (C-8), 95.03 (95.05) (C-2),
69.09 (OCH2-Ar), 60.88 (60.90) (CH2-CH3), 57.18 (PhCH), 27.93
(N-CH3), 13.89 (CH2-CH3) ppm. MS (ESI): m/z (%) ) 517.2
(MH+, 100), 421 (15), 196 (57), 141 (100). IR (KBr): ν ) 3061,
1676, 1514, 1399, 1302, 1186, 1088, 1029 cm-1. HPLC: 96.3%,
tR ) 18.59 min. Anal. (C28H28N4O6 ·2.5H2O) C, H, N.
Synthesis of 4-[(7-Amino-4-methyl-3-oxo-3,4-dihydro-2H-1,4-
benzoxazin-2-yloxy)methyl]benzonitrile (4a). Nitro derivative 3a
(1.20 g, 3.54 mmol) and Cu(OAc)2 (0.708 g, 3.54 mmol) were
dissolved in 250 mL of methanol. NaBH4 was added portionwise
over a period of 1 h and the solution stirred for an additional hour.
The solution was filtered through a Celite pad to remove the black
precipitate. Solvent was removed under reduced pressure. The
residue was dissolved in ethyl acetate (200 mL) and washed
successively with 10% citric acid (3 × 50 mL), saturated solution
of NaHCO3 (2 × 50 mL), and saturated solution of NaCl (1 × 50
mL). The organic phase was dried with Na2SO4 and the solvent
evaporated under reduced pressure to obtain a solid product. Yellow
crystals; yield, 0.856 g (78.3%); mp 65-66 °C. 1H NMR (DMSO-
d6): δ ) 3.24 (s, 3H, N-CH3), 4.85 (s, 2H, OCH2), 5.06 (s, 2H,
2-[2-(4-Carbamimidoylbenzyloxy)-4-methyl-3-oxo-N-phenyl-3,4-
dihydro-2H-1,4-benzoxazine-7-carbonylamino]acetic acid (8b). 8b
was synthesized from 7b according to the general procedure for
alkaline hydrolysis of alkyl esters. Brown crystals; yield, 91 mg
(88.2%); mp 110-111 °C. 1H NMR (DMSO-d6): δ ) 3.20 (s, 3H,
3
4
NH2), 5.55 (s, 1H, 2-H), 6.30 (dd, 1H, J ) 8.7 Hz, J ) 2.4 Hz,
Ar-H6), 6.35 (d, 1H, 4J ) 2.4 Hz, Ar-H8), 6.90 (d, 1H, 3J ) 8.7
Hz, Ar-H5), 7.44 (d, 2H, 3J ) 8.4 Hz, Ar-H2′,H6′), 7.80 (d, 2H,
3J ) 8.3 Hz, Ar-H3′,H5′) ppm. MS (EI): m/z (%) ) 309 (55), 194
(80) 165 (100). IR (KBr): ν ) 3436, 2926, 2228, 1676, 1516, 1401,
1304, 1187, 1146, 1088, 1030 cm-1. Anal. (C17H15N3O3) C, H, N.
Ethyl 3-[2-(4-Cyanobenzyloxy)-4-methyl-3-oxo-3,4-dihydro-2H-
1,4-benzoxazin-7-ylamino]-3-oxo-2-phenylpropanoate (6a). Amine
4a (0.309 g, 1.00 mmol) and diethyl 2-phenylmalonate (2.36 g,
10.0 mmol) were sealed in a 10 mL process vial and heated in a
2
N-CH3), 4.42 (s, 2H, N-CH2COOH), 4.60 (d, 1H, J ) 12.2 Hz,
2
OCH2-Ar), 4.71 (d, 1H, J ) 12.2 Hz, OCH2-Ar), 5.54 (s, 1H,
4
2-H), 6.79 (d, 1H, J ) 2.1 Hz, Ar-H8), 7.01-7.24 (m, 7H, Ph,
Ar-H5, Ar-H6), 7.32 (d, 2H, 3J ) 7.9 Hz, Ar-H2′,H6′), 7.75 (d,
3
2H, J ) 7.9 Hz, Ar-H3′,H5′), 9.07 (br s, 2H, amidino H), 9.34
(br s, 2H, amidino H) ppm. MS (ESI): m/z (%) ) 489.2 (MH+,
35), 321.1 (35), 198.2 (75), 196.2 (100). IR (KBr): ν ) 3420, 1576,

5626 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18
Ilasˇ et al.
1421, 1090, 1033, 836 cm-1. HPLC: 98.7%, tR ) 15.19 min. Anal.
(C26H24N4O6 ·2H2O) C, H, N.
3), 38.55 (N-CH3), 21.46 (2-CH3) ppm. MS (EI): m/z (%) ) 340
(MH+, 56), 154 (100), 136 (82), 55 (87). IR (KBr): ν ) 3468,
2219, 1604, 1534, 1468, 1254, 1174, 1031, 838, 744 cm-1. Anal.
(C18H17N3O4) C, H, N.
Reduction of Ethyl 2,4-Dimethyl-6/7-nitro-3-oxo-3,4-dihydro-
2H-1,4-benzoxazine-2-carboxylates 9a and 9b. Compound 9a or
9b (10.66 g, 36.2 mmol) was dissolved in anhydrous THF (100
mL). Borane dimethyl sulfide complex (5.51 g, 72.5 mmol) was
added, and the solution was refluxed overnight. Hydrochloric acid
(4 M, 20 mL) was added, and the solution was refluxed for a further
30 min. After neutralization with 1 M NaOH, the solvent was
evaporated under reduced pressure and the precipitated product
recrystallized from methanol/water.
From Tosylate 11a (Method B). A suspension of tosylate 11a
(1.18 g, 3.0 mmol) and sodium 4-cyanophenolate (0.445 g, 3.15
mmol) in anhydrous N,N-dimethylformamide (40 mL) was heated
for 24 h at 80 °C. The solvent was removed under reduced pressure,
and the residue was dissolved in ethyl acetate (400 mL) and washed
successively with 10% citric acid (3 × 75 mL), 1 M NaOH (2 ×
75 mL) and saturated NaCl solution (1 × 75 mL). The organic
phase was dried over Na2SO4, filtered, and the solvent was
evaporated under reduced pressure. The residue was recrystallized
from methanol to give 0.85 g (84% yield) of 4a as orange crystals.
The product was identical to that obtained by method A.
(2,4-Dimethyl-7-nitro-3,4-dihydro-2H-1,4-benzoxazin-2-yl)meth-
anol (10a). Red crystals; yield, 7.57 g (88%); mp 125-127 °C. 1H
3
NMR (CDCl3): δ ) 1.32 (s, 3H, 2-CH3), 1.98 (dd, 1H, J ) 5.9
Hz, 3J ) 6.9 Hz, OH), 3.09 (s, 3H, N-CH3), 3.11 (d, 1H, 2J ) 12.1
2
2
Ethyl 2-(Benzyl{2-[(4-cyanophenoxy)methyl]-2,4-dimethyl-3,4-
dihydro-2H-1,4-benzoxazin-7-yl}amino)acetate (16a). 16a was syn-
thesized from 14a (1.14 g, 2.87 mmol) according to the general
procedure for alkylation under phase-transfer catalysis. Red oil;
yield, 0.69 g (49%). 1H NMR (CDCl3): δ ) 1.27 (t, 3H, 3J ) 7.2
Hz, CH2-CH3), 1.48 (s, 3H, 2-CH3), 2.79 (s, 3H, N-CH3), 2.91 (d,
1H, 2J ) 11.4 Hz, 3-H), 3.18 (d, 1H, 2J ) 11.7 Hz, 3-H), 3.93 (d,
Hz, 3-H), 3.52 (d, 1H, J ) 12.1 Hz, 3-H), 3.63 (dd, 1H, J )
3
2
3
11.6 Hz, J ) 6.9 Hz, CH2-OH), 3.72 (dd, 1H, J ) 11.6 Hz, J
) 5.9 Hz, CH2-OH), 6.62 (d, 1H, 3J ) 9.0 Hz, Ar-H5), 7.69 (d,
1H, 4J ) 2.6 Hz, Ar-H8), 7.84 (dd, 1H, 3J ) 9.0 Hz, 4J ) 2.6 Hz,
Ar-H6) ppm. 13C NMR (CDCl3): δ ) 141.44, 141.21 (C-4a, C-8a),
138.76 (C-7), 119.36 (C-6), 112.61 (C-8), 109.83 (C-5), 75.31 (C-
2), 67.05 (-CH2-OH), 53.98 (C-3), 39.00 (N-CH3), 20.77 (2-CH3)
ppm. MS (FAB): m/z (%) ) 239 (MH+, 82), 154 (84), 136 (71),
81 (46), 69 (100), 55 (95). IR (KBr): ν ) 3515, 2928, 1605, 1534,
1495, 1310, 1220, 1061, 970, 804, 743 cm-1. Anal. (C11H14N2O4)
C, H, N.
2
1H, J ) 9.0 Hz, CH2O), 3.99 (s, 2H. N-CH2-COO), 4.14-4.24
(m, CH2O, 3-H, CH2-CH3), 4.58 (s, 2H, Ph-CH2), 6.28-6.31 (m,
2H, Ar-H6, Ar-H8), 6.61 (d, 1H, 3J ) 8.1 Hz, Ar-H5), 6.97 (d,
3
2H, J ) 9.0 Hz, Ar-H2′,H6′), 7.24-7.34 (m, 5H, Ph), 7.58 (d,
2H, 3J ) 9.0 Hz, Ar-H3′,H5′) ppm. MS (ESI): m/z (%) ) 486
(MH+, 45), 485 (M+, 100). IR (KBr): ν ) 2979, 2359, 2340, 2223,
1743, 1628, 1604, 1572, 1519, 1452, 1371, 1301, 1256, 1172, 1113,
1092, 1029 cm-1. Anal. (C29H31N3O4) C, H, N.
(2,4-Dimethyl-7-nitro-3,4-dihydro-2H-1,4-benzoxazin-2-yl)me-
thyl 4-Methylbenzenesulfonate (11a). A solution of alcohol 10a
(3.28 g, 13.8 mmol) and tosyl chloride (2.89 g, 15.2 mmol) in
anhydrous pyridine (40 mL) was stirred for 24 h at room
temperature. Solvent was removed under reduced pressure. The
residue was dissolved in ethyl acetate (600 mL) and washed
successively with 2 M hydrochloric acid (3 × 100 mL), 1 M sodium
hydroxide (2 × 100 mL), and brine (1 × 100 mL). The organic
phase was dried over Na2SO4 and filtered, and the solvent was
4-{2-[(4-Cyanophenoxy)methyl]-2,4-dimethyl-3,4-dihydro-2H-
1,4-benzoxazin-7-ylamino}-4-oxobutanoic Acid (16l). A solution of
amine 13a (0.309 g, 1.00 mmol) and succinic anhydride (1.00 g,
10.0 mmol) in anhydrous tetrahydrofuran was heated for 2 h at 60
°C. The solvent was removed under reduced pressure to give 4.09 g
(100%) of a yellow oil. 1H NMR (DMSO-d6: δ ) 1.38 (s, 3H,
2-CH3), 2.50 (4H, COCH2CH2CONH, signal overlapped with
1
evaporated under reduced pressure to /4 of the starting volume.
The residual solution was stored at 4 °C, and the precipitated
crystals were filtered off and washed with cold ethyl acetate. Yield,
4.60 g (85%); yellow crystals; mp 157-159 °C. 1H NMR (CDCl3):
δ ) 1.37 (s, 3H, 2-CH3), 2.98 (s, 3H, Ar′-CH3), 3.02 (s, 3H,
2
DMSO), 2.80 (s, 3H, N-CH3), 2.99 (d, 1H, J ) 11.6 Hz, 3-H),
3.22 (d, 1H, 2J ) 11.6 Hz, 3-H), 4.07 (d, 1H, 3J ) 10.2 Hz, CH2O),
4.15 (d, 1H, 3J ) 10.2 Hz, CH2O), 6.68 (d, 1H, 3J ) 8.7 Hz,
3
4
Ar-H5), 6.97 (dd, 1H, J ) 8.7 Hz, J ) 2.3 Hz, Ar-H6), 7.07
2
2
N-CH3), 3.16 (d, 1H, J ) 12.3 Hz, 3-H), 3.35 (d, 1H, J ) 12.3
(d, 1H, 3J ) 2.3 Hz, Ar-H8), 7.16 (d, 2H, 3J ) 8.9 Hz,
2
2
Hz, 3-H), 3.90 (d, 1H, J ) 10.0 Hz, CH2O), 4.05 (d, 1H, J )
Ar-H2′,H6′), 7.75 (d, 2H, J ) 8.9 Hz, Ar-H3′,H5′), 9.16 (s, 1H,
3
10.0 Hz, CH2O), 6.58 (d, 1H, J ) 9.0 Hz, Ar-H5), 7.35 (d, 2H,
3
3J ) 8.4 Hz, Ar-H3′,H5′), 7.51 (d, 1H, 4J ) 2.6 Hz, Ar-H8), 7.76
CONH), 12.07 (br s, 1H, COOH) ppm. MS (EI): m/z (%) ) 409
(M+, 7), 323 (6), 309 (100). IR (NaCl): ν ) 3328, 2225, 1712,
1605, 1510, 1257, 1173, 835 cm-1. Anal. (C22H23N3O5) C, H, N.
HRMS.
(d, 2H, J ) 8.4 Hz, Ar-H2′,H6′), 7.81 (dd, 1H, J ) 9.0 Hz, J
) 2.6 Hz, Ar-H6) ppm. MS (FAB): m/z (%) ) 393 (MH+, 58),
154 (100), 137 (82). IR (KBr): ν ) 3442, 1603, 1528, 1500, 1364,
1315, 1226, 1172, 1074, 999, 886, 843, 747, 673 cm-1. Anal.
(C18H20N2O6S) C, H, N.
3
3
4
Ethyl 4-(Benzyl{2-[(4-carbamimidoylphenoxy)methyl]-2,4-di-
methyl-3,4-dihydro-2H-1,4-benzoxazin-7-yl}amino)-4-oxobu-
tanoate Acetate (17e). 17e was synthesized from 16e according to
the general procedure for preparation of amidines from nitriles
(Pinner reaction). Brown powder; yield, 276 mg (37%); mp
125-127 °C. 1H NMR (DMSO-d6): δ ) 1.17 (t, 3H, 3J ) 7.1 Hz,
CH2-CH3), 1.37 (s, 3H, 2-CH3), 1.89 (s, 3H, CH3COOH), 2.32 (t,
2H, 2J ) 6.1 Hz, COCH2CONH), 2.50 (COCH2CH2CONH,
Synthesis of Compounds 12a and 12b by Mitsunobu Reaction
(Method A). A solution of alcohol 10a or 10b (6.12 g, 25.7 mmol),
4-cyanophenol (3.07 g, 25.7 mmol), and triphenylphosphine (8.76
g, 33.4 mmol) in anhydrous tetrahydrofuran (100 mL) was cooled
on an ice bath under argon. Diisopropyl azodicarboxylate (6.75 g,
33.4 mmol) dissolved in 30 mL of anhydrous tetrahydrofuran was
added, and the solution was stirred for 30 min at 0 °C and then
refluxed for 48 h. The solvent was removed under reduced pressure,
and the residue was triturated with boiling petroleum ether (150
mL) and finally recrystallized from petroleum ether/ethyl acetate.
4-[(2,4-Dimethyl-7-nitro-3,4-dihydro-2H-1,4-benzoxazin-2-yl)-
methoxy]benzonitrile (12a). Red crystals; yield, 7.31 g (83.9%),
mp 155-156 °C. 1H NMR (CDCl3): δ ) 1.51 (s, 3H, 2-CH3), 3.08
2
overlapped with DMSO-d5), 2.85 (s, 3H, N-CH3), 3.08 (d, 1H, J
) 11.8 Hz, 3-H), 3.29 (d, 1H, 2J ) 11.8 Hz, 3-H), 4.03 (q, 2H, 3J
) 7.1 Hz, CH2-CH3), 4.08 (d, 1H, 2J ) 10.2 Hz, CH2O), 4.16 (d,
2
4
1H, J ) 10.2 Hz, CH2O), 4.76 (s, 2H, Ph-CH2), 6.56 (d, 1H, J
) 2.3 Hz, Ar-H8), 6.60 (dd, 1H, 3J ) 8.4 Hz, 4J ) 2.3 Hz,
Ar-H6), 6.70 (d, 1H, 3J ) 8.4 Hz, Ar-H5), 7.18 (d, 2H, 3J ) 8.7
Hz, Ar-H2′,H6′), 7.20-7.29 (m, 5H, Ph), 7.84 (d, 2H, J ) 8.7
3
Hz, Ar-H3′,H5′), 9.15 (br s, 4H, amidino-H) ppm. MS (EI): m/z
2
2
(s, 3H, N-CH3), 3.27 (d, 1H, J ) 12.1 Hz, 3-H), 3.52 (d, 1H, J
) 12.1 Hz, 3-H), 3.96 (d, 1H, 2J ) 9.1 Hz, CH2O), 4.05 (d, 1H, 2J
) 9.1 Hz, CH2O), 6.65 (d, 1H, 3J ) 9.0 Hz, Ar-H5), 6.98 (d, 2H,
3J ) 9.0 Hz, Ar-H2′,H6′), 7.62 (d, 2H, 3J ) 9.0 Hz, Ar-H3′,H5′),
(%) ) 544 (M+, 15), 527 (100), 399 (51), 309 (35), 91 (42). IR
(KBr): ν ) 3063, 1731, 1661, 1608, 1518, 1265, 1178, 1036 cm-1
.
HPLC: 100.0%, tR ) 21.44 min. Anal. (C31H36N4O5 · CH3CO-
OH·H2O) C, H, N.
4
3
4
7.72 (d, 1H, J ) 2.6 Hz, Ar-H8), 7.86 (dd, 1H, J ) 9.0 Hz, J
) 2.6 Hz, Ar-H6) ppm. 13C NMR (CDCl3): δ ) 161.50 (C-1′),
140.90, 140.60 (C-4a, C-8a), 138.49 (C-7), 134.03 (C-3′, C-5′),
118.90 (C-6), 118.84 (CN), 115.31 (C-2′, C-6′), 112.16 (C-8),
109.78 (C-5), 104.87 (C-4′), 73.86 (C-2), 70.50 (2-CH2), 53.76 (C-
2-(Benzyl{2-[(4-carbamimidoylphenoxy)methyl]-2,4-dimethyl-
3,4-dihydro-2H-1,4-benzoxazin-7-yl}amino)-2-oxoacetic Acid (18c).
18c was synthesized from 17c according to the general procedure
for alkaline hydrolysis of alkyl esters. Green crystals; yield, 75 mg

NoVel Antithrombotic Compounds with Dual Function
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18 5627
1
(24%); mp 227-228 °C. H NMR (DMSO-d6): δ ) 1.34 (s, 3H,
chromogenic substrate then added. Absorbance at 405 nm at 25
°C was measured every 10 s. Measurements were carried out in
triplicate with three concentrations of the inhibitor and two
concentrations of the substrate. For every combination of concentra-
tions Ki was calculated from the change of absorbance in the initial,
linear part of the curve according to the method of Cheng and
Prusoff39 and the final result given as their average value.
Argatroban (thrombin, Ki ) 12 ( 2 nM) was used as control.
Inhibition of in Vitro Binding of Fibrinogen to Isolated GPIIb/
IIIa Receptor. Binding affinities to GPIIb/IIIa receptor were
measured by a solid-phase competitive displacement assay.40
Human fibrinogen (100 mg) was dissolved in aqueous NaCl (0.3
M, 5 mL) at 30 °C and then diluted with 0.1 M NaHCO3 in water
to a final concentration of 1 mg/mL. Biotin N-hydroxysuccinimide
ester (2 mg) was dissolved in DMF (2 mL) and added to 6 mL of
fibrinogen solution. The reaction mixture was incubated for 90 min
at 30 °C and dialyzed for 3 h at room temperature against buffer 1
(3 L, 20 mM Tris, 150 mM NaCl, pH 7.4). After dialysis, the
solution was centrifuged for 5 min at 5400 rpm and Tween-20
(0.005%) added to give the stock solution. Human integrin (10 µL
of GPIIb/IIIa receptor solution (Calbiochem)) was diluted in 10.2
mL of buffer 2 (20 mM Tris, 150 mM NaCl, 1 mM CaCl2, 1 mM
MgCl2, 1 mM MnCl2, pH 7.4) and adsorbed to 96-well (100 µL/
well) high-binding microtiter plates (Greiner, Lumitrac 600)
overnight at 4 °C. Nonspecific receptor-binding sites were blocked
with 1% BSA in buffer 2 (200 µL/well). Following incubation for
1 h at room temperature, the plates were washed twice with buffer
3 (buffer 2 containing 0.1% Tween-20). The potential antagonists
were serially diluted with buffer and test solutions added (50 µL/
well) together with biotinylated fibrinogen (50 µL/well, 1:10 dilution
of stock solution in buffer 2) to each well. The plates were incubated
for 2 h at room temperature, then washed twice with buffer 3.
Peroxidase-conjugated antibiotin goat antibody (100 µL/well of a
1:1000 dilution of purchased solution (Calbiochem) in buffer 3
containing 0.1% of BSA) was added to each well and incubated
for another hour. The microtiter plates were washed three times
with buffer 3. Finally, chemiluminescent substrate (POD, Roche
Diagnostics, Boehringer Mannheim) (50 µL/well) was added and
the luminescence measured with a GENios (Tecan Group AG)
multimode research reader three times over 10 min. Positive controls
received no inhibitors, while negative controls received no ligands.
RGDS peptide with IC50 of 1.3 ( 0.06 µM was used as the internal
standard. Assays were performed in triplicate at least. The mean
experimental data were fitted to the sigmoid model and IC50 values
calculated from the dose-response curve (OriginPro, OriginLab,
version 7.5).
Inhibition of in Vitro Human Platelet Aggregation in Platelet
Rich Plasma. Healthy male donors who had not taken any
antiplatelet drugs for at least 2 weeks were fasted for 8 h prior to
withdrawing of blood; then 10 mL of whole blood was collected
using a butterfly needle and a 10 mL plastic syringe containing 1
mL of 0.129 M buffered sodium citrate (3.8%). Platelet-rich plasma
(PRP) was prepared by centrifugation at 1000g for 15 min at room
temperature, allowing the centrifuge to coast to a stop without
braking. Platelet poor plasma (PPP) was prepared by centrifuging
the remaining blood at 2000g for 10 min at room temperature,
allowing the centrifuge to coast to a stop without braking. The PRP
was adjusted with PPP to a count of (250 ( 25) × 106 mL. Then
15 µL of tested compounds was added to 135 µL of PRP. After 5
min of incubation of 15 µL of adenosine 5′-diphosphate (ADP)
(11 µM final concentration) was added to the cuvettes and
aggregation was monitored by measuring optical density for 10 min
on the Behring coagulation timer (BCT, Dade Behring, Marburg,
Germany). The entire procedure was run within 2 h, since this is
the maximal viability of the platelets. Saline, in place of test
compound, was used to determine the maximal aggregation.
2-CH3), 2.82 (s, 3H, N-CH3), 3.06 (d, 1H, 2J ) 11.8 Hz, 3-H),
3.27 (d, 1H, 2J ) 11.8 Hz, 3-H), 4.08 (d, 1H, 2J ) 10.2 Hz, CH2O),
4.13 (d, 1H, 2J ) 10.2 Hz, CH2O), 4.81 (s, 2H, Ph-CH2), 6.55 (d,
1H, 4J ) 2.3 Hz, Ar-H8), 6.57 (dd, 1H, 3J ) 8.7 Hz, 4J ) 2.3 Hz,
Ar-H6), 6.64 (d, 1H, 3J ) 8.7 Hz, Ar-H5), 7.17 (d, 2H, 3J ) 8.9
Hz, Ar-H2′,H6′), 7.18-7.31 (m, 5H. Ph), 7.81 (d, 2H, J ) 8.9
3
Hz, Ar-H3′,H5′), 8.84 (br s, 2H, amidino-H), 9.16 (br s, 2H,
amidino-H) ppm. MS (ESI): m/z (%) ) 489 (MH+, 70), 145.0 (95),
104.0 (100). IR (KBr): ν ) 3425, 3000, 1630, 10608, 1521, 1487,
1457, 1264, 1179, 1179, 1043 cm-1. HPLC: 100%, tR ) 18.43
min. Anal. (C27H29N4O5) C, H, N. HRMS (MH+).
Docking Studies. Autodock 3.0530 was used to predict the ligand-
binding mode in the thrombin active site and GPIIb/IIIa receptor
binding site. The energetically most favorable conformation was
found for each stereoisomer with HyperChem, using the semiem-
pirical method MNDO.29 All amide bonds had trans configuration
and were marked as nonrotatable. Compounds were protonated on
the basic center (guanidine, amidino, or amino group) and were
assigned with a negative charge on the acidic center (ꢀ carboxylic
group of aspartate or its bioisoster). For docking calculations, the
crystal structures of thrombin, with a resolution of 2.4 Å (PDB
code 1KTS),14 and RIIbꢀ3 receptor, with a resolution of 2.9 Å (PDB
code 1TY5),20 were used. The PDB crystal structure was prepared
for docking by removing water molecules and ligands. In the case
of 1TY5, the magnesium ion in the active site was retained and
manually assigned charge of +2 and solvation parameters. The
ligands were docked into a restricted box centered in the active
site (22 Å × 22 Å × 22 Å). For the global search using the LGA
(Lamarckian genetic algorithm), the size of the initial random
population was 250 individuals, the maximal number of energy
evaluations 1.25 × 106, the maximal number of generations 500,
the number of top individuals that survived into the next generation,
the elitism, 1, the probability that a gene would undergo a random
change 0.02, and the crossover probability 0.80, and the average
of the worst energy was calculated over a window of 10 generations.
For a purely local search, the pseudo Solis and Wets method was
used, whereas the Solis and Wets method was used for the LGA
part of the local search. The parameters used for local search in
both cases were a maximum of 1000 iterations per local search,
the probability of performing a local search on an individual was
1.0, the maximal number of consecutive successes or failures before
doubling or halving the step size of the local search was 4, and the
lower bound on the step size, 0.01, was the termination criterion
for the local search. A total of 250 dockings were performed, and
ligands with rmsd less than 1 were joined in clusters. The ligand
with the lowest docked energy was chosen for interpreting the
docking results.
Enzyme Assay for Inhibition of Serine Proteases. The enzyme
amidolytic method for determining inhibition was based on the
spectrophotometric determination of absorbance of the product
formed after amide bond cleavage of a chromogenic substrate in
the presence of the enzyme. Ki, which is a quantitative measure of
inhibitor potency, was determined from the kinetics of substrate
hydrolysis with and without the addition of the inhibitor.39
Measurements (spectrophotometer, Tecan, Sunrise) were performed
in microtiter plates with a final volume of 200 µL. Thrombin was
tested at a final concentration of 0.5 NIH E/mL with the substrate
S-2238 (Chromogenix) at 20 and 40 µM final concentration, and
factor Xa at final concentration of 1 mBAEE E/mL with the
substrate S-2222 (Chromogenix) at 100 and 200 µM final concen-
trations. Trypsin was assayed at a final concentration of 0.5 nkat/
mL using the substrate S-2222 (Chromogenix) at 50 and 100 µM
final concentrations. Inhibitors were dissolved in DMSO (concen-
tration of stock solutions, 10 mmol/L) and diluted with distilled
water to concentrations from 0.5 to 100 µM. Reaction rates were
measured in the presence and absence of the inhibitor. Then 50
µM HBSA buffer, 50 µM solution of each inhibitor concentration
(or of HBSA buffer in case of measurement without inhibitor), and
50 µM of enzyme solution were pipetted into the microtiter wells.
The plate was incubated for 15 min at 25 °C and 50 µL of
Acknowledgment. This work was supported financially by
the Slovenian Research Agency (Grant No. P1-0208). The
authors thank Professor Roger Pain for critical reading of the
manuscript.

5628 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18
Ilasˇ et al.
platelet aggregation based upon the Arg-Gly-Asp sequence of fibrino-
gen (aminobenzamidino)succinyl (ABAS) series of orally active
fibrinogen receptor antagonists. J. Med. Chem. 1995, 38, 2378–2394.
Supporting Information Available: Synthetic procedures and
compound characterization, microanalysis data of synthesized
compounds, HRMS results, and double-reciprocal plot. This mate-
rial is available free of charge via the Internet at http://pubs.acs.org.
(16) Curran, M. P.; Keating, G. M. Eptifibatide: a review of its use in
patients with acute coronary syndromes and/or undergoing percuta-
neous coronary intervention. Drugs 2005, 65, 2009–2035.
References
(17) (a) Cook, J. J.; Bednar, B.; Lynch, J. J.; Gould, R. J.; Egbertson, M. S.;
Halczenko, W.; Duggan, M. E.; Hartman, G. D.; Lo, M. W.; Murphy,
G. M.; Deckelbaum, L. I.; Sax, F. L.; Barr, E. Tirofiban (Aggrastat).
CardioVasc. Drug ReV. 1999, 17, 199–224. (b) Mukherjee, D.; Roffi,
M. Current strategies with high-dose tirofiban. Expert Opin. Drug
Metab. Toxicol. 2007, 3, 275–280.
(18) (a) Bhatt, D. L.; Topol, E. J. Scientific and therapeutic advances in
antiplatelet therapy. Nat. ReV. Drug DiscoVery 2003, 2, 15–28. (b)
McNicol, N.; Israels, S. J. Platelets and anti-platelet therapy. J. Phar-
macol. Sci. 2003, 93, 381–396.
(19) Hantgan, R. R.; Stahle, M. C.; Connor, J. H.; Connor, R. F.; Mousa,
S. A. RIIbꢀ3 priming and clustering by orally active and intravenous
integrin antagonists. J. Thromb. Hemostasis 2007, 5, 542–550.
(20) Xiao, T.; Takagi, J.; Coller, B. S.; Wang, J.-H.; Springer, T. A.
Structural basis for allostery in integrins and binding to fibrinogen-
mimetic therapeutics. Nature 2004, 432, 59–67.
(1) (a) Lopez, A. D.; Mathers, C. D.; Ezzati, M.; Jamison, D. T.; Murray,
C. J. L. Global and regional burden of disease and risk factors, 2001:
systematic analysis of population health data. Lancet 2006, 367, 1747–
1757. (b) http://www.who.int/topics/cardiovascular_diseases/en/.
(2) (a) Schwienhorst, A. Direct thrombin inhibitors, a survey of recent
developments. Cell. Mol. Life Sci. 2006, 63, 2773–2791. (b) Srivastava,
S.; Goswami, L. N.; Dikshit, D. K. Progress in the design of low
molecular weight thrombin inhibitors. Med. Res. ReV. 2005, 25, 66–
92. (c) Steinmetzer, T.; Stu¨rzebecher, J. Progress in the development
of synthetic thrombin inhibitors as new orally active anticoagulants.
Curr. Med. Chem. 2004, 11, 2297–2321.
(3) (a) Kubitza, D.; Haas, S. Novel factor Xa inhibitors for prevention
and treatment of thromboembolic diseases. Expert Opin. InVest. Drugs
2006, 15, 843–855. (b) Casimiro-Garcia, A.; Dudley, D. A.; Heemstra,
R. J.; Filipski, K. J.; Bigge, C. F.; Edmunds, J. J. Progress in the
discovery of factor Xa inhibitors. Expert Opin. Ther. Pat. 2006, 16,
119–145. (c) Quan, M. L.; Smallheer, J. M. The race to an orally
active factor Xa inhibitor. Recent advances. Curr. Opin. Drug
DiscoVery 2004, 7, 460–469.
(4) (a) Lazarus, R. A.; Olivero, A. G.; Eigenbrot, C.; Kirchhofer, D.
Inhibitors of tissue factor/factor VIIa for anticoagulant therapy. Curr.
Med. Chem. 2004, 11, 2275–2290. (b) Kranjc, A.; Kikelj, D.; Peterlin-
Masˇicˇ, L. Recent advances in the discovery of tissue factor/factor VIIa
inhibitors and dual inhibitors of factor VIIa/factor Xa. Curr. Pharm.
Des. 2005, 11, 4207–4227.
(5) Andronati, S. A.; Karaseva, T. L.; Krysko, A. A. Peptidomimetics.
Antagonists of the fibrinogen receptors: molecular design, structures,
properties and therapeutic applications. Curr. Med. Chem. 2004, 11,
1183–1211.
ˇ
(21) Stefanicˇ Anderluh, P.; Anderluh, M.; Ilasˇ, J.; Mravljak, J.; Sollner
Dolenc, M.; Stegnar, M.; Kikelj, D. Toward a novel class of
antithrombotic compounds with dual function. Discovery of 1,4-
benzoxazin-3(4H)-one derivatives possessing thrombin inhibitory and
fibrinogen receptor antagonistic activity. J. Med. Chem. 2005, 48,
3110–3113.
(22) For simplicity, the term GPIIb/IIIa receptor antagonist is used in this
paper to denote compounds that inhibit binding of fibrinogen to the
GPIIb/IIIa receptor (platelet fibrinogen receptor) irrespective of
resulting signaling and possible conformational change of the integrin
that might account for their paradoxical procoagulant activity in lower
doses.18,19
(23) (a) Markwardt, F. Coagulation inhibitors from blood-sucking animals.
A new line of developing antithrombotic drugs. Pharmazie 1994, 49,
313–316. (b) Nar, H.; Bauer, M.; Schmid, A.; Stassen, J.-M.; Wienen,
W.; Priepke, H. W. M.; Kauffmann, I. K.; Ries, U. J.; Hauel, N. H.
Structural basis for inhibition promiscuity of dual specific thrombin
and factor Xa blood coagulation inhibitors. Structure 2001, 9, 29–37.
(24) (a) Rigbi, M.; Orevi, M.; Eldor, A. Platelet aggregation and coagulation
inhibitors in leech saliva and their roles in leech therapy. Semin.
Thromb. Hemostasis 1996, 22, 273–278. (b) Waxman, L.; Connolly,
T. M. Isolation of an inhibitor selective for collagen-stimulated platelet
aggregation from the soft tick Ornithodorus moubata. J. Biol. Chem.
1993, 268, 5445–5449. (c) Nikonov, G. I.; Titova, E. A.; Seleznev,
K. G. A stable prostacyclin-like substance produced by the medicinal
leech Hirudo medicinalis. Prostaglandins Other Lipid Mediators 1999,
58, 1–7.
(25) (a) Bussey, H. I. An overview of anticoagulants, antiplatelet agents,
and the combination in patients with mechanical heart valves. J. Heart
ValVe Dis. 2004, 13, 319–324. (b) Bonow, R. O.; Carabello, B. A.;
Chatterjee, K.; de Leon, A. C.; Faxon, D. P.; Freed, M. D.; Gaasch,
W. H.; Lytle, B. W.; Nishimura, R. A.; O’Gara, P. T.; O’Rourke,
R. A.; Otto, C. M.; Shah, P. M.; Shanewise, J. S.; Smith, S. C.; Jacobs,
A. K.; Adams, C. D.; Anderson, J. L.; Antman, E. M.; Faxon, D. P.;
Fuster, V.; Halperin, J. L.; Hiratzka, L. F.; Hunt, S. A.; Lytle, B. W.;
Nishimura, R.; Page, R. L.; Riegel, B. ACC/AHA 2006 guidelines
for the management of patients with valvular heart disease. Circulation
2006, 114, E84–E231. (c) Sabatine, M. S.; Tu, T. M.; Jang, I. K.
Combination of a direct thrombin inhibitor and a platelet glycoprotein
IIb/IIIa blocking peptide facilitates and maintains reperfusion of
platelet-rich thrombus with alteplase. J. Thromb. Thrombolysis 2000,
10, 189–196. (d) Walenga, J. M.; Jeske, W. P.; Wallis, D. E.; Bakhos,
M.; Lewis, B. E.; Leya, F.; Fareed, J. Clinical experience with
combined treatment of thrombin inhibitors and GPIIb IIIa inhibitors
in patients with HIT. Semin. Thromb. Hemostasis 1999, 25 (Suppl.
1), 77–81. (e) Lincoff, A. M.; Bittl, J. A.; Harrington, R. A.; Feit, F.;
Kleiman, N. S.; Jackman, J. D.; Sarembock, I. J.; Cohen, D. J.; Spriggs,
D.; Ebrahimi, R.; Keren, G.; Carr, J.; Cohen, E. A.; Betriu, A.; Desmet,
W.; Kereiakes, D. J.; Rutsch, W.; Wilcox, R. G.; de Feyter, P. J.;
Vahanian, A.; Topol, E. J. Bivalirudin and provisional glycoprotein
IIb/IIIa blockade compared with heparin and planned glycoprotein IIb/
IIIa blockade during percutaneous coronary intervention. REPLACE-2
Randomized Trial. JAMA, J. Am. Med. Assoc. 2003, 289, 853–863.
(f) Watson, T.; Lip, G. Y. H. Combining antiplatelet drugs and oral
anticoagulant therapy in atrial fibrillation: acute coronary syndromes
and/or percutaneous coronary intervention/stenting revisited. Stroke
2007, 38, E107–E108. (g) Eikelboom, J. W.; Hirsh, J. Combined
antiplatelet and anticoagulant therapy: clinical benefits and risks. J.
Thromb. Hemostasis 2007, 5 (Suppl. 1), 255–263.
(6) (a) Strong, S. H.; Halperin, J. L. Confronting atrial fibrillation in the
elderly. Stroke risk stratification and emerging antithrombotic therapies.
Geriatrics 2007, 62, 22–27. (b) Ruef, J.; Katus, H. A. New antithrom-
botic drugs on the horizon. Expert Opin. InVest. Drugs 2003, 12, 781–
797.
ˇ
(7) Prezˇelj, A.; Anderuh, P.S.; Peternel, L.; Urleb, U. Recent advances in
serine protease inhibitors as anticoagulant agents. Curr. Pharm. Des.
2007, 13, 287–312.
(8) (a) Bode, W. The structure of thrombin: a janus-headed proteinase.
Semin. Thromb. Hemostasis 2006, 32 (Suppl. 1), 16–31. (b) Bode,
W. The structure of thrombin, a chameleon-like proteinase. J. Thromb.
Hemostasis 2005, 3, 2379–2388.
(9) Bode, W.; Mayr, I.; Baumann, U.; Huber, R.; Stone, S. R.; Hofsteenge,
J. The refined 1.9 Å crystal structure of human R-: interaction with
D-Phe-Pro-Arg chloromethyl ketone and significance of the Tyr-Pro-
Pro-Trp insertion segment. EMBO J. 1989, 8, 3467–3475.
(10) Bajusz, S.; Szell, E.; Bagdy, D.; Barabas, E.; Horvath, G.; Dioszegi,
M.; Fittler, Z.; Szabo, G.; Juhasz, A.; Tomori, E.; Szilagyi, G. Highly
active and selective anticoagulants: D-Phe-Pro-Arg-H, a free tripeptide
aldehyde prone to spontaneous inactivation, and its stable N-methyl
derivative, D-MePhe-Pro-Arg-H. J. Med. Chem. 1990, 33, 1729–1735.
(11) (a) Hijikata Okunomiya, A.; Okamoti, S. A strategy for a rational
approach to designing synthetic selective inhibitors. Semin. Thromb.
Hemostasis 1992, 18, 135–149. (b) Rewinkel, J. B. M.; Adang, A. E. P.
Strategies and progress towards the ideal orally active thrombin
inhibitor. Curr. Pharm. Des. 1999, 5, 1043–1075.
(12) Peterlin-Masˇicˇ, L.; Kikelj, D. Arginine mimetics. Tetrahedron 2001,
57, 7073–7105.
(13) (a) Gustafsson, D.; Bylund, R.; Antonsson, T.; Nilsson, I.; Nystrom,
J. E.; Eriksson, U.; Bredberg, U.; Teger-Nilsson, A. C. A new oral
anticoagulant: the 50-year challenge. Nat. ReV. Drug DiscoVery 2004,
3, 649–659. (b) Crowther, M. A.; Weitz, J. I. Ximelagatran: the first
oral direct thrombin inhibitor. Expert Opin. InVest. Drugs 2004, 13,
403–413. (c) Evans, H. C.; Perry, C. M.; Faulds, D. Ximelagatran/
melagatran. A review of its use in the prevention of venous
thromboembolism in orthopaedic surgery. Drugs 2004, 64, 649–678.
(14) Hauel, N. H.; Nar, H.; Priepke, H.; Ries, U.; Stassen, J. M.; Wienen,
W. Structure-based design of novel potent nonpeptide thrombin
inhibitors. J. Med. Chem. 2002, 45, 1757–1766.
(15) Zablocki, A. J.; Rico, J. G.; Garland, R. B.; Rogers, T. E.; Williams,
K.; Schretzman, L. A.; Rao, S. A.; Bovy, P. R.; Tjoeng, F. S.;
Lindmark, R. J.; Toth, M. V.; Zupec, M. E.; Mcmackins, D. E.; Adams,
S. P.; Miyano, M.; Markos, C. S.; Milton, M. N.; Paulson, S.; Herin,
M.; Jacqmin, P.; Nicholson, N. S.; Panzerknodle, S. G.; Haas, N. F.;
Page, J. D.; Szalony, J. A.; Taite, B. B.; Salyers, A. K.; King, L. W.;
Campion, J. G.; Feigen, L. P. Potent in vitro and in vivo inhibitors of

NoVel Antithrombotic Compounds with Dual Function
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18 5629
ˇ
(26) A P3 carboxylic group in some thrombin inhibitors possessing a basic
P1 amidine moiety increases oral biovalilability of these compounds
and reduces amidine toxicity.
(34) Kikelj, D.; Suhadolc, E.; Urleb, U.; Zbontar, U. A convenient
synthesis of 3,4-dihydro-2-methyl-3-oxo-2H-1,4-benzoxazine-2-
carboxylic acids and 3,4-dihydro-2-methyl-3-oxo-2H-pyrido[3,2-
b]-1,4-benzoxazine-2-carboxylic acids. J. Heterocycl. Chem. 1993,
30, 597–602.
(27) (a) Morphy, R.; Kay, C.; Rankovic, Z. From magic bullets to designed
multiple ligands. Drug DiscoVery Today 2004, 9, 641–651. (b)
Morphy, J. R.; Rankovic, Z. Designed multiple ligands, an emerging
drug discovery paradigm. J. Med. Chem. 2005, 48, 6523–6543. (c)
Morphy, R.; Rankovic, Z. The physicochemical challenges of design-
ing multiple ligands. J. Med. Chem. 2006, 49, 4961–4970.
(28) (a) Ojima, I.; Chakravarty, S.; Dong, Q. Antithrombotic agents: from
RGD to peptide mimetics. Bioorg. Med. Chem. Lett. 1995, 3, 337–
360. (b) Marinelli, L.; Lavecchia, A.; Gottschalk, K.-E.; Novellino,
E.; Kessler, H. Docking studies on Rvꢀ3 integrin ligands: pharma-
cophore refinement and implications for drug design. J. Med. Chem.
2003, 46, 4393–4404. (c) Egbertson, M. S.; Chang, C. T. C.; Duggan,
M. E.; Gould, R. J.; Halczenko, W.; Hartman, G. D.; Laswell, W. L.;
Lynch, J. J.; Lynch, R. J.; Manno, P. D.; Naylor, A. M.; Prugh, J. D.;
Ramjit, D. R.; Sitko, G. R.; Smith, R. S.; Turchi, L. M.; Zhang, G. X.
Nonpeptide fibrinogen receptor antagonists. 2. Optimization of a
tyrosine template as a mimic for Arg-Gly-Asp. J. Med. Chem. 1994,
37, 2537–2551.
(35) Mitsunobu, O. The use of diethyl azodicarboxylate and triphenylphos-
phine in synthesis and transformation of natural products. Synthesis
1981, 1–28.
(36) (a) Abdel-Magid, A. F.; Carson, K. G.; Harris, B. D.; Maryanoff, C. A.;
Shah, R. D. Reductive amination of aldehydes and ketones with sodium
triacetoxyborohydride. Studies on direct and indirect reductive ami-
nation procedures. J. Org. Chem. 1996, 61, 3849–3862. (b) Anderluh,
ˇ
M.; Cesar, J.; Stefanic, P.; Kikelj, D.; Janesˇ, D.; Murn, J.; Nadrah,
K.; Tominc, M.; Addicks, E.; Giannis, A.; Stegnar, M.; Sollner-Dolenc,
M. Design and synthesis of novel platelet fibrinogen receptor
antagonists with 2H-1,4-benzoxazine-3(4H)-one scaffold. A systematic
study. Eur. J. Med. Chem. 2005, 40, 25–49.
(37) Ilasˇ, J.; Tomasˇic´, T.; Kikelj, D. Novel potent and selective thrombin
inhibitors based on a central 1,4-benzoxazin-3(4H)-one scaffold.
J. Med. Chem. 2008, 51, 2863–2867.
(38) Lee, L.; Kreutter, K. D.; Pan, W.; Crysler, C.; Spurlino, J.; Player,
M. R.; Tomczuk, B.; Lu, T. 2-(2-Chloro-6-fluorophenyl)acetamides
as potent thrombin inhibitors. Bioorg. Med. Chem. Lett. 2007, 17,
6266–6269.
(39) Cheng, Y.; Prusoff, W. H. Relationship between the inhibition constant
(Ki) and the concentration of inhibitor which causes 50% inhibition
(IC50) of an enzymatic reaction. Biochem. Pharmacol. 1973, 22, 3099–
3108.
(40) Addicks, E.; Mazitschek, R.; Giannis, A. Synthesis and biological
investigation of novel tricyclic benzodiazepinedione-based RGD
analogue. ChemBioChem 2002, 3, 1078–1088.
(29) HyperChem, release 5.01 for Windows; Hypercube, Inc.: Gainesville,
FL, 1996.
(30) Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Har,
T. W. E.; Belew, R. K.; Olson, A. J. Auomated docking using a
Lamarckian genetic algorithm and an empirical binding free energy
function. J. Comput. Chem. 1998, 19, 1639–1662.
(31) Ilasˇ, J.; Kikelj, D. Ring opening of 2-benzylamino-2H-1,4-benzoxazin-
3(4H)-ones and 2-bromo-2H-1,4-benzoxazin-3(4H)-ones. HelV. Chim.
Acta 2008, 91, 654–664.
(32) Cowan, J. A. Cu2+/BH4- reduction system. Synthetic utility and mode
of action. Tetrahedron Lett. 1986, 27, 1205–1208.
(33) Roger, R.; Neilson, D. G. The chemistry of imidates. Chem. ReV. 1961,
61, 179–211.
JM8003448