Novel non-covalent thrombin inhibitors incorporating P1 4,5,6,7-tetrahydrobenzothiazole arginine side chain mimetics

Article abstract of DOI:10.1016/j.ejmech.2003.12.006

The design, synthesis and biological activity of a series of novel non-covalent D-Phe-Pro-Arg motif-based thrombin inhibitors incorporating 4,5,6,7-tetrahydrobenzothiazol-2-amine as a novel arginine surrogate are described. Compound 9, the most potent in

Full text of DOI:10.1016/j.ejmech.2003.12.006

European Journal of Medicinal Chemistry 39 (2004) 257–265
www.elsevier.com/locate/ejmech
Original article
Novel non-covalent thrombin inhibitors incorporating P1
4,5,6,7-tetrahydrobenzothiazole arginine side chain mimetics
Petra Marinko a, Aleš Krbavcˇicˇ a, Gregor Mlinšek b, Tomaž Šolmajer b,c, Alenka
Trampuš Bakija d, Mojca Stegnar d, Jure Stojan e, Danijel Kikelj a,*
a Faculty of Pharmacy, University of Ljubljana, Aškercˇeva 7, 1000 Ljubljana, Slovenia
b National Institute of Chemistry, Hajdrihova 19, 1115 Ljubljana, Slovenia
c Lek Pharmaceuticals, Verovškova 57, 1526 Ljubljana, Slovenia
d Department of Angiology, University Medical Centre, Zaloška 7, 1525 Ljubljana, Slovenia
e Institute of Biochemistry, Medical Faculty, University of Ljubljana, Vrazov Trg 2, 1000 Ljubljana, Slovenia
Received 15 July 2003; received in revised form 19 December 2003; accepted 22 December 2003
Dedicated to Professor Branko Stanovnik on occasion of his 65th birthday
Abstract
The design, synthesis and biological activity of a series of novel non-covalent D-Phe-Pro-Arg motif-based thrombin inhibitors incorporat-
ing 4,5,6,7-tetrahydrobenzothiazol-2-amine as a novel arginine surrogate are described. Compound 9, the most potent in the series of thrombin
inhibitors, exhibited an in vitro Ki of 128 nM and 342-fold selectivity against trypsin. The binding mode of this novel class of thrombin
inhibitors in the enzyme active site, based on the X-ray crystal structure of compound 9 co-crystallized with human a-thrombin, is discussed.
© 2004 Elsevier SAS. All rights reserved.
Keywords: Thrombin inhibitors; Peptidomimetics; Arginine side chain mimetic; 4,5,6,7-Tetrahydrobenzothiazole
1. Introduction
Numerous thrombin inhibitors, such as inogatran [4] and
melagatran [5], have been developed based on the D-Phe-
Pro-Arg sequence [6], which mimics thrombin’s natural sub-
strate, fibrinogen. A general strategy in the design of these
tripeptidomimetic thrombin inhibitors [3b] continues to be
the replacement of individual amino acids in the D-Phe-Pro-
Arg motif by their mimetic compounds, in order to achieve
an optimal combination of potency, selectivity versus similar
serine proteases and oral bioavailability. Among this class of
thrombin inhibitors, most research has been devoted to com-
pounds lacking the C-terminus of the arginine residue, which
were obtained by incorporating various arginine side chain
mimetic groups [7]. In order to overcome the high basicity of
the P1 guanidine, amidine and aliphatic amine moieties
which impair oral bioavailability of thrombin inhibitors,
various amino heterocycles were investigated as the P1 argi-
nine side chain mimetics [3]. Among these, some bicyclic
aromatic (amino)heterocycles, e.g., 1-isoquinolinamine [8],
3-benzisoxazolamine [9], indole [10], benzimidazole [10],
indazole [10,11], imidazo[1,2-a]pyridine [12] and
pyrrolo[3,2-b]pyridine [13] were recently employed as P1
moieties in tripeptidomimetic thrombin inhibitors.
Thrombin, a trypsin-like serine protease, is a key enzyme
in the blood coagulation cascade [1]. Serving as the terminal
enzyme of the blood coagulation pathway, it catalyzes the
conversion of fibrinogen to fibrin and the cross-linking of
fibrin to form the blood clot. Thrombin is also a potent
stimulant of platelet aggregation. Thus, it plays a major role
in development of thromboembolic diseases, which remain
the leading cause of morbidity and mortality in developed
countries. Inhibition of thrombin is therefore an attractive
therapeutic target for intervention in thrombosis [2]. During
the last decade, the search for orally bioavailable, potent and
selective low molecular weight, active site-directed thrombin
inhibitors has emerged as one of the most intensive areas of
investigation in drug discovery [3].
Abbreviations: Chx, b-cyclohexylalanine; BTZ, 4,5,6,7-tetrahydro-
benzothiazole.
* Corresponding author.
E-mail address: danijel.kikelj@ffa.uni-lj.si (D. Kikelj).
© 2004 Elsevier SAS. All rights reserved.
doi:10.1016/j.ejmech.2003.12.006

258
P. Marinko et al. / European Journal of Medicinal Chemistry 39 (2004) 257–265
Fig. 1. Evolution of 4,5,6,7-tetrahydrobenzothiazol-2-amine arginine side chain mimetics and novel thrombin inhibitors which incorporate them.
Our approach to the design of thrombin inhibitors based
the resulting bromoketone 6 with thiourea afforded the annu-
lated thiazole 7 which, on hydrolysis, gave 6-aminomethyl-
4,5,6,7-tetrahydrobenzothiazol-2-amine (1b).
on the D-Phe-Pro-Arg sequence was focused mainly on re-
placing the arginine moiety with less basic, conformationally
restricted, partially saturated heterobicyclic arginine side
chain mimetics I (Fig. 1). These contain a five- or six-
membered N-heterocyclic ring optionally substituted by an
amino group mimicking the guanidino moiety of arginine,
annulated to a cyclohexane ring mimicking the arginine tri-
methylene side chain. We speculated that the bulky and
lipophilic cyclohexane ring could confer selectivity for
thrombin against trypsin, since the S1 pocket in thrombin is
slightly larger and more hydrophobic than the one in trypsin.
Moreover, a conformationally flexible cyclohexane ring in
the P1 part of the inhibitor could allow a better fit of the
inhibitor in the enzyme active site than inhibitors incorporat-
ing rigid and planar P1 aromatic heterobicycles.
Compounds 9–18 and 21 were prepared by coupling argi-
nine surrogates 1a, 1b and benzothiazole-2,6-diamine to the
optionally protected P2-P3 moiety 8, using N-ethyl-N′-(3-
dimethylaminopropyl)-carbodiimide (EDC) and 1-hydroxy-
benzotriazole (HOBT) as amide bond forming reagents. To
obtain inhibitors with a free N-terminus, the corresponding
N-tert-butyloxycarbonyl protected compounds were cleaved
by hydrogen chloride in glacial acetic acid solution (Fig. 4).
Compounds 19 and 20 were prepared similarly by coupling
of N-benzylsulfonyl-D-phenylalanine with P2-P1 moieties 23
and 24. The target inhibitors were purified with column
In this paper, we report the design, synthesis, in vitro
biological activity, and a binding mode of thrombin inhibi-
tors incorporating 4,5,6,7-tetrahydrobenzothiazol-2-amine
as a P1 moiety.
2. Chemistry
Synthetic routes to 4,5,6,7-tetrahydrobenzothiazol-2-
amine arginine side chain mimetics 1a and 1b for use as P1
moieties are outlined in Figs. 2 and 3, respectively. 4,5,6,7-
Tetrahydrobenzothiazole-2,6-diamine (1a) was prepared
from N-(4-oxocyclohexyl)-acetamide (2) [14] by bromina-
tion, subsequent Hantzsch thiazole synthesis using thiourea
as an S–C–N synthon [15] and final acid hydrolysis of the
acetamido group of 4. Similarly, bromination of N-[(4-
oxocyclohexyl)methyl]-acetamide (5) [16] and reaction of
Fig. 3. Reagents and conditions: (a) Br2, CH2Cl2, rt, 1 h, then reflux, 1 h; (b)
thiourea, EtOH, reflux, 2 h; (c) 47% aq. HBr, reflux, 16 h.
Fig. 2. Reagents and conditions: (a) Br2, AcOH, 60 °C, 1 h; (b) thiourea,
AcOH, reflux, 1.5 h; (c) 47% aq. HBr, reflux, 24 h.
Fig. 4. General scheme for synthesis of compounds 9–18 and 21.

P. Marinko et al. / European Journal of Medicinal Chemistry 39 (2004) 257–265
259
chromatography and obtained as diastereomeric mixtures
with respect to C-6 of the 4,5,6,7-tetrahydrobenzothiazole
ring.
freedom of the P1 part, was not beneficial for inhibitory
activity against thrombin. Thus, compound 21 exhibited
threefold less inhibitory potency and threefold lower selec-
tivity for thrombin over trypsin than inhibitor 9. This is
probably due to the increased length of the molecule, which
is responsible for a shift of the P2 and P3 groups from the sites
of optimal interaction with the active site residues. Com-
pounds 9 and 21 were almost equipotent in all clotting as-
says. Interestingly, the potency of compound 9, with
D-cyclohexylalanine in the P3 part, exceeded the potency of
12, with D-phenylalanine, in all the assays by a factor of 5–6.
Based on the calculated log P values (4.20 0.86 for com-
pound 9 and 3.57 0.89 for compound 12) [18], it is antici-
pated that the potency of 9 is affected to a greater extent than
that of 12 by non-specific binding to plasma proteins.
The replacement of D-phenylalanine in compound 12 by
D-3,4-dichlorophenylalanine resulted in a twofold improve-
ment of inhibition constant (10; Ki = 352 nM). The potency
of compound 16, lacking a P3 benzylsulfonyl group, was
10-fold lower (Ki = 3.392 µM) than that of 10, indicating an
important contribution of the benzylsulfonyl group to the
inhibitory activity of compound 10. Only slightly higher
inhibitory potencies against human thrombin were observed
in 14 (Ki = 1.100 µM) and 15 (Ki = 2.360 µM) featuring a
D-diphenyl moiety at the P3 part of the molecule.
3. Pharmacology
The ability of the new thrombin inhibitors to inhibit the
enzymatic activity of thrombin, trypsin and factor Xa was
measured by amidolytic assay and expressed as inhibitory
constants (Ki values), determined according to Cheng and
Prusoff based on IC50 [17a], or from a relation between
reaction velocity in the absence and presence of inhibitor
using the relevant Km [17b]. The selectivity for thrombin over
trypsin was expressed as the ratio Ki(trypsin)/Ki(thrombin). In-
hibitors were also tested on normal pool plasma with stan-
dard clotting assays, including thrombin time (TT), activated
partial thromboplastin time (APTT) and prothrombin time
(PT) assays, which were used as qualitative in vitro indica-
tors of potential antithrombotic activity.
4. Results and discussion
The results are summarized in Table 1. Several of these
compounds showed good in vitro potency against thrombin,
with desirable selectivity towards trypsin. Inhibitor 9, pos-
sessing 4,5,6,7-tetrahydrobenzothiazol-2,6-diamine to fill
the selectivity pocket, proline in the central part and a ben-
zylsulfonyl group attached to D-cyclohexylalanine as the P3
part of the molecule, was the most potent with a Ki value of
128 nM and a 342-fold selectivity for thrombin over trypsin.
It doubled TT, APTT and PT at concentrations of 16, 59 and
114 µM, respectively. Contrary to our expectation, a methyl-
ene linker between the cyclohexane ring of a heterocycle and
the amino group, which might allow a substantial rotational
Compounds 11 and 13, which can be regarded as
N-carboxymethyl analogs of 9, displayed submicromolar
inhibition constants for thrombin (11: Ki = 560 nM; 13:
Ki = 800 nM), proving that the N-benzylsulfonyl group of 9
undergoes a more favorable interaction with the enzyme
active site than the N-tert-butyloxycarbonyl-methyl and
N-carboxymethyl groups of 11 and 13, respectively. Inhibitor
11 displayed the highest, 810-fold, selectivity for thrombin
over trypsin of all the compounds studied.
In an attempt to evaluate the influence of planarity of the
heterobicyclic P1 moiety, benzothiazole-2,6-diamine was ex-
Table 1
Inhibitory activities of compounds 9-21: P1, P2 and P3 modifications
a Due to precipitation measurements were not possible.

260
P. Marinko et al. / European Journal of Medicinal Chemistry 39 (2004) 257–265
replacing proline in 19 by glycine can be explained by dimin-
ished rigidity of inhibitor 20 in the P2 part, allowing a better
overall fit of the molecule in the thrombin active site. In
contrast, substitution of proline by glycine or alanine in a P1
4,5,6,7-tetrahydrobenzothiazol-2-amine containing inhibitor
12, resulting in compounds 17 or 18, was detrimental to
anti-thrombin activity (Ki > 35.1 µM).
All the thrombin inhibitors listed in Table 1 were also
selective with respect to inhibition of factor Xa. Inhibition of
the latter appears to be in the same range for all the com-
pounds tested, and is hardly affected by any of the replace-
ments studied.
The potency and selectivity of inhibitor 9 prompted us to
determine its precise binding conformation in the thrombin
active site by X-ray crystallography. Fig. 5 shows the overall
binding mode of 9 in the thrombin active site with the
partially saturated heterobicyclic arginine side chain mimetic
occupying the S1 selectivity pocket, the proline ring nestling
under the YPPW loop, and the cyclohexane ring occupying
the aryl binding site. Altogether, inhibitor 9 forms six hydro-
gen bonds to the surrounding residues. As shown in Fig. 6,
the 2-amino-4,5,6,7-tetrahydrobenzothiazole NH2 group
forms a hydrogen bond with Od1 Asp189 (2.71 Å) and O
Ala190 (3.35 Å), and is additionally hydrogen bonded
through a water mediated hydrogen bond to the backbone
carbonyl oxygens of Phe227 and Trp215. The crystals of
thrombin with the inhibitor 9 bound in the active site were
prepared using a diastereomeric mixture of 9, with respect to
the stereogenic center at C6 of the 4,5,6,7-
tetrahydrobenzothiazole ring, whereas the model of 9 that fits
best into the final 2Fo-Fc density has the S configuration at
C6 and equatorial orientation of the chain attached to C6.
Consequently, the sulfur atom of the benzothiazole ring is
directed towards Val213. The S2 site is occupied by proline,
with its carbonyl group forming a hydrogen bond to both the
Oc Ser195 (2.80 Å) and Ne2 His57 (3.06 Å) of the catalytic
triad and to Gly193 via a water molecule. The P3 cyclohexyl
Fig. 5. Connolly surface map of the X-ray structure of the human
a-thrombin—inhibitor 9 complex at 1.9 Å resolution. Inhibitor 9 is shown as
sticks. Colors green, blue, red and yellow identify carbon, nitrogen, oxygen
and sulfur, respectively. Contact with Asp 189 is shown.
plored as an arginine surrogate instead of 4,5,6,7-
tetrahydrobenzothiazol-2-amine. The resulting compound
19 was devoid of inhibitory activity against thrombin
(Ki > 35.1 µM) and trypsin (Ki > 68.3 µM). A great drop in
potency of 19 (calcd. pKa of N3 = 5.13 0.30) [18] with
regard to the corresponding tetrahydro derivative 12 (calcd.
pKa of N3 = 6.14 0.40) [18] cannot be interpreted only by
difference in basicity of P1 moiety of both compounds but
also in terms of higher flexibility of compound 12. Thus, the
activity of 19 was improved by replacing proline by glycine
in the P2 part of the inhibitor to give 20 with an inhibition
constant Ki = 5.10 µM for thrombin. The beneficial effect of
Fig. 6. Schematic representation of inhibitor 9 bound in the active site of thrombin. Thin lines indicate hydrogen bonds. Water molecules are represented as red
spheres. Distances are given in Ångstroms.

P. Marinko et al. / European Journal of Medicinal Chemistry 39 (2004) 257–265
261
ring lies in the lipophilic aryl binding site. The electron
6.1.2. N-(Benzylsulfonyl)-D-phenylalanyl-L-alanine;
density of the benzyl group was contoured at 0.8r. It points
loosely towards the solvent, which implies that it is not stably
bound. Thus, the contribution to affinity of the ligand is not
favorable. Crystallographic analysis of 9 bound in the throm-
bin active site indicated that one of the N-terminal sulfona-
mide oxygens participates in a hydrogen-bonding interaction
with the NH of Gly219 (3.29 Å) [19].
general procedure for the synthesis of P3–P2 fragments 8
6.1.2.1. N-Benzylsulfonyl-D-phenylalanine. A solution of
benzylsulfonyl chloride (5.91 g, 31 mmol) in dioxane
(100 ml) was added dropwise to a stirred mixture of
D-phenylalanine hydrochloride (5.44 g, 27 mmol) and
NaOH (1 M, 27 ml) in dioxane (100 ml) at 0 °C. After the
addition was complete, the mixture was stirred overnight at
room temperature. The solvent was removed under reduced
pressure, the residue was dissolved in water (50 ml), acidified
to pH 2 with saturated aqueous KHSO4 and extracted with
ethyl acetate (3 × 50 ml). The organic phase was washed with
brine, dried over Na2SO4 and evaporated in vacuum to give a
yellow oil; yield: 3.88 g (45%); IR (NaCl-Film): m 3262,
3064, 2931, 1731, 1496, 1456, 1374, 1322, 1264, 1153,
1127, 1043, 956, 846, 782, 698 cm–1; 1H-NMR (300 MHz,
DMSO-d6): 2.78–3.06 (ABX-system, 2H, JAB = 13.66 Hz,
JAX = 8.85 Hz, JBX = 5.65 Hz, CH2Ph), 3.92–4.09 (m, 3H,
PhCH2SO2 + CH), 7.17–7.37 (2 × m, 10H, 2 × Ph), 7.65 (d,
1H, J = 8.67 Hz, –SO2NHCH–), 12.5 (s br, 1H, COOH) ppm.
5. Conclusion
In summary, we have prepared a series of thrombin inhibi-
tors incorporating P1 4,5,6,7-tetrahydrobenzothiazol-2-
amine moiety as a novel arginine side chain mimetic. Varia-
tions in the P2 and P3 parts led to thrombin inhibitor 9 which
exhibited an in vitro Ki of 128 nM and 342-fold selectivity
against trypsin. The results of the crystal structure of 9
complexed in the thrombin active site will be used for further
structure-based design of improved D-Phe-Pro-Arg se-
quence based thrombin inhibitors. The studies concerning
oral bioavailability of these thrombin inhibitors will be re-
ported in due course.
6.1.2.2. N-Benzylsulfonyl-D-phenylalanyl-L-alanine methyl
ester. To a solution of N-benzylsulfonyl-D-phenylalanine
(1.59 g, 5 mmol) and L-alanine methyl ester hydrochloride
(0.63 g, 4.5 mmol) in N,N-dimethylformamide (12.5 ml),
1-hydroxybenzotriazole (0.68 g, 5 mmol) and, after adjusting
6. Experimental protocols
the pH of the resulting solution to
8
with
N-methylmorpholine, 1-(3-dimethylaminopropyl)-3-ethyl-
carbodiimide hydrochloride (0.96 g, 5 mmol) were added.
The reaction mixture was stirred overnight at room tempera-
ture, the solvent was evaporated in vacuo and the residue
partitioned between EtOAc (50 ml) and saturated aqueous
NaHCO3 (3 × 50 ml). The organic layer was washed with
brine (2 × 50 ml), dried over MgSO4, filtered and the solvent
was removed in vacuo to give 1.73 g (95%) of a pale yellow
foam; IR (KBr): m 3281, 2952, 1736, 1648, 1543, 1456, 1317,
1221, 1151, 1125, 952, 790, 749, 698, 548 cm–1; MS (FAB):
405 (MH+, 47%), 91 (100%); 1H-NMR (300 MHz, DMSO-
d6): 1.19 (d, 3H, J = 7.16 Hz, CHCH3), 2.86 (ABX-system,
2H, JAB = 13.56 Hz, JAX = 9.04 Hz, JBX = 6.03 Hz, CH2Ph),
3.58 (s, 3H, COOCH3), 4.01 (AB system, 2H,
JAB = 13.56 Hz, PhCH2SO2), 4.15–4.36 (m, 2H, CH–Ala,
CH–Phe), 7.19–7.35 (m, 10H, 2 × Ph), 7.67 (d, 1H,
J = 9.04 Hz, –SO2NHCH–), 8.61 (d, 1H, J = 7.16 Hz, CONH)
ppm.
6.1. Chemistry
6.1.1. Materials and methods
Chemicals were purchased from Aldrich Chemical Co.,
Fluka and Synthetech and used without further purification.
THF was kept over sodium and distilled immediately prior to
use.Analytical TLC was performed on Merck silica gel (60 F
254) plates (0.25 mm). Visualization was effected with ultra-
violet light, ninhydrin and 2,4-dinitrophenylhydrazine. Col-
umn chromatography was carried out on Florisil® (particle
size 100–200 mesh) and silica gel 60 (particle size 240–
400 mesh). Melting points were determined on a Reichert hot
stage microscope and are uncorrected. 1H-NMR spectra were
recorded on a Bruker AVANCE DPX300 spectrometer in
CDCl3 or DMSO solution with TMS as the internal standard.
In proline containing compounds 9–16 and 21, due to
cis/trans-isomerism of the proline peptide bond and due to
the presence of stereogenic center at C-6 the NMR spectra
are very complex and therefore, only 1H-NMR spectrum of 9
is reported [20, 21a]. IR spectra were obtained 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 or functions were
within +0.4% of the theoretical values. Mass spectra were
obtained using a VG-Analytical Autospec Q mass spectrom-
eter. Chemical names were generated usingACD/Name soft-
ware.
6.1.2.3. N-Benzylsulfonyl-D-phenylalanyl-L-alanine. To a
solution of N-benzylsulfonyl-D-phenylalanyl-alanine me-
thyl ester (1.60 g, 3.96 mmol) in methanol/water (1:1,
100 ml), lithium hydroxide (2.2 M, 2.38 ml) was added
dropwise during 0.5 h to keep pH between 12 and 13. The
mixture was stirred overnight, then pH was adjusted to 7 with
KHSO4 and methanol was evaporated under reduced pres-
sure. The residual aqueous solution was acidified to pH
2 with KHSO4 and extracted with ethylacetate (3 × 50 ml).

262
P. Marinko et al. / European Journal of Medicinal Chemistry 39 (2004) 257–265
The combined organic extracts were washed with brine (2 ×
75 ml), dried Na2SO4, filtered and evaporated in vacuum to
give 896 mg (58%) of a pale yellow foam; IR (KBr): m 3321,
2930, 1734, 1654, 1541, 1497, 1456, 1321, 1152, 1031, 950,
783, 747, 699, 621 cm–1; MS (FAB): 391 (MH+, 100%);
1H-NMR (300 MHz, DMSO-d6): 1.15–1.21 (m, 3H, Ala–
rophenyl)propanoyl]-2-pyrrolidinecarboxamide (10). Us-
ing a general procedure described above, 10 was prepared
from N-(benzylsulfonyl)-D-(3,4-dichlorophenyl)-alanyl-L-
proline[21a] (140 mg, 0.29 mmol) and 4,5,6,7-tetrahydro-
1,3-benzothiazole-2,6-diamine dihydrobromide (86 mg,
0.26 mmol); yield: 25 mg (15%), pale brown foam; IR (KBr):
m 3326, 1654, 1522, 1448, 1317, 1152, 1128, 1031, 697,
544 cm–1. MS (FAB): 636 (MH+, 100%). Anal.
C28H31Cl2N5O4S2 (C, H, N).
CH3), 2.86 (ABX-system, 2H, JAB
= 13.56 Hz,
JAX = 9.04 Hz, JBX = 6.03 Hz, CH2Ph), 4.03 (AB system, 2H,
JAB = 13.56 Hz, PhCH2SO2), 4.17–4.35 (m, 2H, CH–Ala,
CH–Phe), 7.20–7.33 (m, 10H, 2 × Ph), 7.65 (d, 1H,
J = 9.04 Hz, –SO2NHCH–), 8.50 (d, 1H, J = 7.53 Hz,
CONH), 12.62 (s br, 1H, COOH) ppm; Anal. C19H22N2O5S
(C, H, N).
Using the same general procedure, N-(benzylsulfonyl)-D-
(b-cyclohexyl)alanyl-L-proline, N-(benzylsulfonyl)-D-phe-
nylalanyl-glycine, N-(benzylsulfonyl)-D-phenylalanyl-L-pro-
line, and N-(benzylsulfonyl)-D-(3,4-dichloro)phenyl-alanyl-
L-proline were prepared.
6.1.3.2. tert-Butyl 2-[[(1R)-2-((2S)-2-{[(2-amino-4,5,6,7-te-
trahydro-1,3-benzothiazol-6-yl)amino]carbonyl}-pyrrolidi-
nyl)-1-(cyclohexylmethyl)-2-oxoethyl](tert-butoxycarbonyl)
-amino]acetate (11). Using a general procedure described
above, 11 was prepared from (2S) 1-((2R)-2-{(tert-buto-
xycarbonyl)[2-(tert-butoxy)-2-oxoethyl]amino}-3-cyclohe-
xylpropanoyl)-proline [21b] (241 mg, 0.50 mmol) and
4,5,6,7-tetrahydro-1,3-benzothiazole-2,6-diamine dihydro-
bromide (149 mg, 0.45 mmol); yield: 185 mg (65%), white
crystals; mp 99–101 °C. IR (KBr): m 3318, 2978, 2927, 2585,
1750, 1698, 1646, 1529, 1447, 1368, 1228, 1158 cm–1. MS
(FAB): 634 (MH+, 100%). Anal. C32H51N5O6S (C, H, N).
6.1.3. (2S)-N-(2-Amino-4,5,6,7-tetrahydro-1,3-benzothia-
zol-6-yl)-1-{(2R)-2-[(benzylsulfonyl)-amino]-3-cyclohexyl-
propanoyl}-2-pyrrolidinecarboxamide (9); general proce-
dure for the synthesis of compounds 9–12, 14, 17, 18
and 21–22 by coupling of P3–P2 fragments 8 with 4,5,6,7-
tetrahydro-1,3-benzothiazole-2,6-diamine (1a), 6-(amino-
methyl)-4,5,6,7-tetrahydro-1,3-benzothiazol-2-amine (1b)
and 1,3-benzothiazole-2,6-diamine
To a solution of N-(benzylsulfonyl)-D-(b-cyclohexyl)-
alanyl-L-proline (250 mg, 0.59 mmol) and 4,5,6,7-
tetrahydro-1,3-benzothiazole-2,6-diamine dihydrobromide
(178 mg, 0.54 mmol) in N,N-dimethylformamide (1.5 ml),
1-hydroxybenzotriazole (80 mg, 0.59 mmol) and, after ad-
justing the pH of the resulting solution to 8 with
N-methylmorpholine, 1-(3-dimethylaminopropyl)-3-ethyl-
carbodiimide hydrochloride (113 mg, 0.59 mmol) were
added. The reaction mixture was stirred overnight at room
temperature, the solvent was evaporated in vacuo and the
residue partitioned between EtOAc (10 ml) and saturated
aqueous NaHCO3 (3 × 10 ml). The organic layer was washed
with brine (2 × 10 ml), dried over MgSO4, filtered and the
solvent was removed in vacuo. The crude product was puri-
6.1.3.3. (2S)-N-(2-Amino-4,5,6,7-tetrahydro-1,3-benzothia-
zol-6-yl)-1-{(2R)-2-[(benzylsulfonyl)amino]-3-phenyl-pro-
panoyl}-2-pyrrolidinecarboxamide (12). Using a general
procedure described above, 12 was prepared from
N-(benzylsulfonyl)-D-phenylalanyl-L-proline (166 mg,
0.40 mmol) and 4,5,6,7-tetrahydro-1,3-benzothiazole-2,6-
diamine dihydrobromide (120 mg, 0.36 mmol); yield:
100 mg (49%), pale yellow solid; mp 126–129 °C. IR (KBr):
m 3362, 3192, 2926, 1646, 1522, 1454, 1315, 1152, 1124,
944, 751, 698, 547 cm–1. MS (FAB): 568 (MH+, 100%).
Anal. C28H33N5O4S2 (C, H, N).
6.1.3.4. 2-{[(1R)-2-((2S)-2-{[(2-Amino-4,5,6,7-tetrahydro-
1,3-benzothiazol-6-yl)amino]carbonyl}-pyrrolidinyl)-1-(cy-
clohexylmethyl)-2-oxoethyl]amino}acetic acid dihydrochlo-
ride (13). Dry HCl gas was bubbled into a solution of 11
(211 mg, 0.33 mmol) in glacial acetic acid (4 ml) for 1 h at
room temperature. The acetic acid was removed in vacuo and
the residue triturated several times with anhydrous ether. The
white precipitate was filtered off and dried in vacuo over
NaOH; yield: 167 mg (92%); mp 180–182 °C. IR (KBr): m
2929, 2288, 1742, 1639, 1548, 1448, 1227, 1116, 880,
710 cm–1. MS (FAB): 478 [(MH-2HCl)+, 100%]. Anal.
C23H35N5O4S × 2HCl × CH3COOH (C, H, N).
fied
by
column
chromatography
(silica
gel,
CHCl3/MeOH = 9:1) to give 9 as a faint brown crystalline
solid; yield: 167 mg (54%); mp 127–130 °C. IR (KBr): m
3357, 2924, 2848, 1638, 1523, 1448, 1311, 1125, 754, 697,
546 cm–1. MS (FAB): 574 (MH+, 100%). 1H-NMR
(300 MHz, DMSO-d6): d = 0.70–0.98 (m, 2H, Chx), 1.00–
1.28 (m, 3H, Chx), 1.30–1.52 (m, 2H, Chx), 1.53–2.10 (m,
9H, 3 × CH2–Chx, bCH–Pro, c-CH2–Pro), 2.20–2.60 (m,
6H, 3 × CH2–BTZ), 2.60–2.74 (m, 1H, CH), 3.38–3.48 (m,
1H, aCH–Chx), 3.68–3.88 (m, 2H, dCH2–Pro), 4.10–4.38
(m, 4H, CH2Ph, aCH–Pro, CH), 6.58–6.68 (m, 2H, NH2),
7.30–7.58 (m, 6H, Ph, NH), 7.82 (7.86, 8.25) (d, 1H,
J = 7.91 Hz, NH). Anal. C28H39N5O4S2 (C, H, N).
6.1.3.5. tert-Butyl (1R)-2-((2S)-2-{[(2-amino-4,5,6,7-tetra-
hydro-1,3-benzothiazol-6-yl)amino]carbonyl}pyrrolidinyl)-
1-benzhydryl-2-oxoethylcarbamate (14). Using a general
procedure described above, 14 was prepared from N-(tert-
butyloxycarbonyl)-D-3,3-diphenylalanyl-L-proline
[21a]
(219 mg, 0.50 mmol) and 4,5,6,7-tetrahydro-1,3-
benzothiazole-2,6-diamine dihydrobromide (149 mg,
0.45 mmol); yield: 191 mg (72%); mp 218–220 °C. Anal.
C32H39N5O4S × 0.5H2O (C, H, N).
6.1.3.1. (2S)-N-(2-Amino-4,5,6,7-tetrahydro-1,3-benzothia-
zol-6-yl)-1-[(2R)-2-[(benzylsulfonyl)amino]-3-(3,4-dichlo-

P. Marinko et al. / European Journal of Medicinal Chemistry 39 (2004) 257–265
263
6.1.3.6. (2S)-1-[(2R)-2-Amino-3,3-diphenylpropanoyl]-N-
(2-amino-4,5,6,7-tetrahydro-1,3-benzothiazol-6-yl)-2-pyrro-
lidinecarboxamide dihydrochloride (15). Dry HCl gas was
bubbled into a solution of 14 (185 mg, 0.31 mmol) in glacial
acetic acid (10 ml) for 1 h at room temperature, the acetic
acid was removed in vacuo and the residue triturated several
times with anhydrous ether. The white precipitate was fil-
tered off and dried in vacuo over NaOH; yield: 169 mg
(97%); mp 208–210 °C. IR (KBr): m 3390, 2974, 1641, 1449,
1347, 1242, 1112, 913, 753, 704 cm–1. MS (FAB): 490
[(MH-2HCl)+, 73%], 149 (100%). Anal. C27H31N5O2S ×
2HCl × CH3COOH (C, H, N).
6.1.3.9. (2R)-N-{(1S)-2-[(2-Amino-4,5,6,7-tetrahydro-1,3-
benzothiazol-6-yl)amino]-1-methyl-2-oxoethyl}-2-[(benzyl-
sulfonyl)amino]-3-phenylpropanamide (18). Using a gen-
eral procedure described above, 18 was prepared from
N-(benzylsulfonyl)-D-phenylalanyl-alanine
(195
mg,
0.50 mmol) and 4,5,6,7-tetrahydro-1,3-benzothiazole-2,6-
diamine dihydrobromide (149 mg, 0.45 mmol); yield: 90 mg
(37%), white solid; mp 122–125 °C. IR (KBr): m 3314, 3200,
2929, 1654, 1523, 1455, 1371, 1315, 1231, 1152, 1126, 952,
750, 698, 545 cm–1. MS (FAB): 542 (MH+, 100%). 1H-NMR
(300 MHz, DMSO-d6): d = 1.03–1.15 (m, 3H, CH3–Ala),
1.57–1.89 (m, 2H, CH2), 2.20–2.50 (m, 3H, CH2, CH), 2.57–
2.98 (m, 3H, CH2Ph, CH), 3.80–4.33 (3 × m, 5H, PhCH2SO2
+ 3 × CH), 6.62 (s, 2H, NH2), 7.11–7.38 (m, 10H, 2×Ph),
7.58–7.66 (m, 1H, NHSO2), 7.84–7.98 (m, 1H, NH), 8.26–
8.38 (m, 1H, NH–Ala) ppm. Anal. C26H31N5O4S2 × H2O (C,
H, N).
6.1.3.7. (2R)-1-((2S)-2-{[(2-Amino-4,5,6,7-tetrahydro-1,3-
benzothiazol-6-yl)amino]-carbonyl}pyrrolidinyl)-3-(3,4-di-
chlorophenyl)-1-oxo-2-propanamine dihydrochloride (16).
Using a general procedure described above, tert-butyl (1R)-
2-((2S)-2-{[(2-amino-4,5,6,7-tetrahydro-1,3-benzothiazol-
6-yl)amino]carbonyl}pyrrolidinyl)-1-(3,4-dichlorobenzyl)-
2-oxoethylcarbamate (22) was prepared from N-(tert-
butoxycarbonyl)-D-(3,4-dichlorophenyl)-alanyl-L-proline
[21a] (474 mg, 1.1 mmol) and 4,5,6,7-tetrahydro-1,3-benzo-
thiazole-2,6-diamine dihydrobromide (331 mg, 1.00 mmol);
yield: 448 mg (77%), pale yellow amorphous solid; mp
116–119 °C. IR (KBr): m 3409, 2977, 1637, 1529, 1444,
1367, 1250, 1165, 1029, 753, 679 cm–1. MS (FAB): 582
(MH+, 49%), 70 (100%). Anal. C26H33Cl2N5O4S (C, H, N).
The crude 22 (400 mg, 0.69 mmol) was dissolved in
glacial acetic acid (10 ml), then dry HCl gas was introduced
into the solution for 1 h at room temperature. The mixture
was evaporated in vacuo and the residue triturated with
anhydrous ether to give 16 as pale yellow solid; yield:
366 mg (96%) mp 166–169 °C. IR (KBr): m 2973, 1643,
1469, 1358, 1261, 1132, 1032, 878, 826, 713, 586 cm–1. MS
(FAB): 483 [(MH-2HCl)+, 100%]. Anal. C21H25Cl2N5O2S ×
2HCl (C, H, N).
6.1.3.10. (2S)-N-(2-Amino-1,3-benzothiazol-6-yl)-1-{(2R)-
2-[(benzylsulfonyl)amino]-3-phenylpropanoyl}-2-pyrrolidi-
necarboxamide (19). Compound 19 was prepared by cou-
pling N-(benzylsulfonyl)-D-phenylalanine (180 mg,
0.56 mmol) and N-(2-amino-1,3-benzothiazol-6-yl)-
pyrrolidinecarboxamide dihydrochloride (23) (169 mg,
0.50 mmol) using a procedure described for the synthesis of
9; yield: 100 mg (35%), pale yellow solid; mp 130–133 °C.
IR (KBr): m 3314, 1637, 1526, 1466, 1407, 1307, 1227, 1150,
1123, 934, 819, 748, 698, 544, 488 cm–1. MS (FAB): 564
(MH+, 100%). Anal. C28H29N5O4S2 × H2O (C, H, N).
6.1.3.11. (2R)-N-{2-[(2-Amino-1,3-benzothiazol-6-yl)ami-
no]-2-oxoethyl}-2-[(benzylsulfonyl)amino]-3-phenylpropa-
namide (20). Using a procedure described for the synthesis
of 9, compound 20 was prepared from N-(benzylsulfonyl)-
D-phenylalanine (400 mg, 1.25 mmol) and 2-amino-N-(2-
amino-1,3-benzothiazol-6-yl)-acetamide dihydrochloride
(24) (336 mg, 1.14 mmol); yield: 170 mg (29%), yellow
solid; mp 127–130 °C. IR (KBr): m 3351, 1639, 1530, 1466,
1408, 1307, 1123, 958, 818, 697, 539 cm–1. MS (FAB): 524
(MH+, 100%). 1H-NMR (300 MHz, DMSO-d6): d = 2.80 (dd,
BX part of ABX system, 1H, JBX = 9.42 Hz, JAB = 13.66 Hz,
CH2–Phe), 3.02 (dd, AX part of ABX system, 1H,
JAX = 5.09 Hz, JAB = 13.66 Hz, CH2–Phe), 3.81–4.11 (m,
4H, 2 × CH2), 4.22 (m, 1H, CH), 7.15–7.40 (m, 13H, 2 × Ph,
3 × CH), 7.67 (d, 1H, J = 9.04 Hz, NHSO2), 7.95–8.00 (m,
1H, NH), 8.54 (t, 1H, J = 5.66 Hz, NH–Gly), 9.85 (s, 2H,
NH2) ppm. Anal. C25H25N5O4S2 × H2O (C, H, N).
6.1.3.8. (2R)-N-{2-[(2-Amino-4,5,6,7-tetrahydro-1,3-ben-
zo-thiazol-6-yl)amino]-2-oxoethyl}-2-[(benzylsulfonyl)-
amino]-3-phenylpropanamide (17). Using a general proce-
dure described for the synthesis of 9 above, 17 was prepared
from N-(benzylsulfonyl)-D-phenylalanyl-glycine (150 mg,
0.40 mmol) and 4,5,6,7-tetrahydro-1,3-benzothiazole-2,6-
diamine dihydrobromide (119 mg, 0.36 mmol); yield: 70 mg
(37%), white solid; mp 119-122 °C. IR (KBr): m 3346, 2926,
1655, 1522, 1455, 1371, 1314, 1233, 1126, 956, 750, 698,
543, 487 cm–1. MS (FAB): 528 (MH+, 88%), 91(100%).
1H-NMR (300 MHz, DMSO-d6): d = 1.55–1.87 (m, 2H,
CH2), 2.20–2.60 (m, 4H, 2 × CH2,), 2.80 (dd, BX part of
ABX system, 1H, JBX = 9.42 Hz, JAB = 13.66 Hz, CH2–Phe),
3.02 (dd, AX part of ABX system, 1H, JAX = 5.09 Hz,
JAB = 13.76 Hz, CH2–Phe), 3.82–4.20 (m, 5H, CH2–Gly,
PhCH2SO2, CH), 4.15–4.30 (m, 1H, CH), 6.65 (s, 2H, NH2),
7.15–7.40 (m, 10H, 2 × Ph), 7.67 (d, 1H, J = 9.04 Hz,
NHSO2), 7.95–8.00 (m, 1H, NH), 8.54 (t, 1H, J = 5.66 Hz,
NH–Gly) ppm. Anal. C25H29N5O4S2 × H2O (C, H, N).
6.1.3.12. (2S)-N-[(2-Amino-4,5,6,7-tetrahydro-1,3-benzo-
thiazol-6-yl)methyl]-1-{(2R)-2-[(benzylsulfonyl)amino]-3-
cyclohexylpropanoyl}-2-pyrrolidinecarboxamide (21). Us-
ing a general procedure described above for the synthesis of
9, the compound 21 was prepared from N-(benzylsulfonyl)-
D-(b-cyclohexyl)-alanyl-L-proline (186 mg, 0.44 mmol) and
6-aminomethyl-4,5,6,7-tetrahydro-1,3-benzothiazole-2,6-
diamine dihydrobromide (138 mg, 0.40 mmol); yield:

264
P. Marinko et al. / European Journal of Medicinal Chemistry 39 (2004) 257–265
183 mg (78%); white foam. IR (KBr): m 3356, 2924, 1648,
1523, 1444, 1311, 1141, 782, 699, 546 cm–1. MS (FAB): m/z
(%) 588 (MH+, 100). Anal. C29H41N5O4S2 (C, H, N).
6.1.4.2. Structure solution and refinement. Diffraction data
were indexed and integrated using the MOSFLM v. 6.11 pro-
gram [23] and scaled with the SCALA software [24]. The
exact orientation and positions were determined with the
Patterson search techniques using the AMORE program [25]
and a-thrombin–hirugen coordinates (1HGT.pdb) as a re-
placement model. Initial models of the inhibitors were con-
structed and minimized using the SYBYL 6.7. program
package. Subsequently, the inhibitors were fitted into the
electron density map manually and the crystallographic re-
finement was performed using the CNS version 1.0. [26].
6.1.3.13. 2-Amino-N-(2-amino-1,3-benzothiazol-6-yl)-ace-
tamide dihydrochloride (24). A stirred solution of N-(tert-
butoxycarbonyl)glycine (1.31 g, 7.5 mmol) in N,N-
dimethylformamide (6 ml) was cooled to –15 °C and treated
successively with ethyl chloroformate (0.79 ml, 8.25 mmol)
and triethylamine (1.15 ml, 8.25 mmol). After 1 h at –15 °C
benzothiazol-2,6-diamine (1.24 g, 7.5 mmol) was added and
the mixture was kept at 0–5 °C overnight. The precipitate was
filtered off and the filtrate was evaporated under reduced
pressure. The residual oil was triturated with ether to give
crude tert-butyl 2-[(2-amino-1,3-benzothiazol-6-yl)amino]-
2-oxoethylcarbamate (25) as a pale yellow solid foam; yield:
0.51 g (21%). IR (KBr): m 3303, 2974, 1672, 1531, 1472,
1411, 1230, 1159, 1124, 816 cm–1. MS (FAB): 323 (MH+,
6.1.5. Biological assays
6.1.5.1. Amidolytic enzyme assay. Spectrophotometric en-
zyme tests were performed in microtiter plates in a final
volume of 125 µl. Thrombin was tested at 1 nM with the
substrate S-2366 at 100 nM final concentration and factor Xa
at 3 nM with the substrate S-2222 at 200 nM final concentra-
tion. Trypsin was tested at 6 nM using the substrate S-2251 at
200 nM. Inhibitors were dissolved in DMSO (concentration
of stock solutions: 10 mmol/l) and tested at 100 µM as the
highest concentration. Inhibitors were diluted using HNPT
buffer pH 7.8 consisting of HEPES 100 mM, NaCl 140 mM,
PEG 6000 0.1% and Tween 80 0.02%. The reaction kinetics
were linear both with the time and the concentration of the
enzyme. Full concentration–response curves were run for
inhibitors showing >50% inhibition for any enzyme at
100 µM. Apparent Ki values were calculated according to
Cheng and Prusoff based on IC50 [17a], or from a relation
between reaction velocity in the absence and presence of
inhibitor using the relevant Km [17b]. The Km for the sub-
strates used were determined under the test conditions with at
least five substrate concentrations ranging from 0.5 to
15 times Km [27] according to Eadie [28]. The Km was
108 µM for the substrate S-2366, 613 µM for the substrate
S-2222 and 430 µM for the substrate S-2251.
1
100%). H-NMR (300 MHz, DMSO-d6): d = 1.40 (s, 9H,
tert-Bu), 3.70 (d, 2H, J = 6.03 Hz, CH2–Gly), 7.00 (t, 1H,
J = 6.03 Hz, NH–Gly), 7.20–7.40 (m, 3H, 3 × CH–Ar), 7.99
(s, 1H, NH), 9.83 (s, 2H, NH2) ppm.
The crude 25 (487 mg, 1.51 mmol) was dissolved in
glacial acetic acid (20 ml) and then dry HCl gas was intro-
duced into the solution for 1 h at room temperature. The
mixture was evaporated in vacuo and the residue triturated
with anhydrous ether to give 24 as pale yellow solid; yield:
360 mg (81%) mp 189–192 °C. IR (KBr): m 3240, 3050.
1697, 1647, 1596, 1560, 1491, 1439, 1318, 1237, 946, 836,
1
670 cm–1. MS (FAB): 223 [(MH-2HCl)+, 100%]. H-NMR
(300 MHz, DMSO-d6): d = 3.81 (s, 2H, CH2–Gly), 7.50 (m,
1H, CH–Ar), 7.60 (m, 1H, CH–Ar), 7.95 (s, 1H, NH), 8.22
+
(m, 1H, CH–Ar), 8.32 (s br, 3H, NH3 ), 9.60 (s br, 2H, NH2)
ppm. Anal. C9H10N4OS × 2HCl (C, H, N).
Using the same procedure, (2S)-N-(2-amino-1,3-
benzothiazol-6-yl)-pyrrolidinecarboxamide dihydrochloride
(23) was prepared from N-(tert-butoxycarbonyl)-L-proline
and benzothiazol-2,6-diamine.
6.1.5.2. Coagulation assays. The anticoagulant effect of
thrombin inhibitors was measured by TT, APTT and PT
assays. Inhibitors were added to normal pooled plasma over a
range of concentrations and clotting was recorded in an
automated coagulometer (Fibrintimer, Dade/Behring). The
concentration of inhibitor that doubled the clotting time was
determined for each assay. Normal pooled plasma: Venous
blood from 11 apparently healthy volunteers was collected
(one part 0.11 mol/l sodium citrate solution with nine parts of
blood) and centrifuged immediately at 2000 × g for 30 min at
4 °C. Plasma was removed, pooled and stored in aliquots at
–70 °C until use. For TT determination, inhibitors (10 µl
working solution, concentrations from 2.5 to 100 µM) and
pooled plasma (90 µl) were incubated at 37 °C for 5 min
whereupon thrombin (200 µl) was added. Clot formation was
measured in a coagulometer in duplicate. For APPT determi-
nation, inhibitors (10 µl working solution, concentrations
from 5 to 100 µM) and pooled plasma (90 µl) were incubated
6.1.4. X-ray crystallography
6.1.4.1. Crystallization and data collection. Human
a-thrombin was obtained from Roche (Mannheim, Ger-
many) as a lyophilisate. Sulfated hirugen (54–65) with
the sequence H-Gly-Asp-Phe-Glu-Glu-Ile-Pro-Glu-Glu-Tyr
(SO3H)-Leu-Gln-OH was obtained from Bachem (Heidel-
berg, Germany) as a lyophilisate. Thrombin crystals were
prepared as described previously [22]. The crystallization
was carried out using the hanging drop vapor diffusion tech-
nique. Macroseeding was necessary to obtain crystals of a
size adequate for the diffraction experiment. Crystals were
soaked in a solution containing 1 mM of inhibitor overnight
and then mounted on a loop for data collection. A single
crystal was shock frozen at 100 K and a complete data set
was taken using the MAR image plate system.

P. Marinko et al. / European Journal of Medicinal Chemistry 39 (2004) 257–265
265
[6] S. Bajusz, E. Szell, D. Bagdy, E. Barabas, G. Horvath, M. Dioszegi,
Z. Fittler, G. Szabo, A. Juhasz, J. Med. Chem. 33 (1990) 1729–1735.
[7] L. Peterlin-Mašicˇ, D. Kikelj, Tetrahedron 57 (2001) 7073–7105.
[8] C.M. Timmers, J.B.M. Rewinkel, WO Patent Application 0024718,
2000, Chem. Abstr. 132 (2000) 322143.
[9] S.D. Lepore, M.R. Wiley, WO Patent Application 0027199, 2000,
Chem. Abstr. 132 (2000) 347944.
[10] A.D. Brown, J.D. Smith, J. Blagg, E.C. Gautier, A.B. McElroy, Euro-
pean Patent Application 0997474, 2000, Chem. Abstr. 134 (2000)
131552.
at 37 °C for 5 min whereupon Pathromtin (Dade/Behring)
(100 µl) was added and the sample was incubated for another
2 min at 37 °C. The addition of calcium chloride (100 µl)
triggered the coagulation process and clot formation was
detected with a coagulometer in duplicate. For PT determi-
nation, inhibitors (10 µl working solution, concentrations
from 5 to 100 µM) and pooled plasma (90 µl) were incubated
at 37 °C for 5 min whereupon thromboplastin (Thromborel S,
Dade/Behring) (200 µl) was added. Clot formation was mea-
sured in a coagulometer in duplicate.
[11] J.Z. Ho, T.S. Gibson, E. Semple, Bioorg. Med. Chem. Lett. 12 (2002)
743–748.
[12] A.M. Naylor-Olsen, P.E. Sanderson, WO PatentApplication 9961442,
1999, Chem. Abstr. 132 (2000) 12328.
[13] M. Morrissette, P.G. Nanterment, A.M. Naylor-Olsen, M.G. Stanton,
C. Coburn, B.D. Dorsey, R. Isaacs, T.A. Lyle, D. Staas, C. Burgey,
P.E. Sanderson, H.G. Selnick, K.J. Stauffer, P.D. Williams, WO Patent
Application 0026211, 2000, Chem. Abstr. 132 (2000) 334473.
[14] P. Marinko, A. Obreza, L. Peterlin-Mašicˇ, A. Krbavcˇicˇ, D. Kikelj, J.
Heterocycl. Chem. 37 (2000) 405–409.
[15] D. Kikelj, U. Urleb, in: E. Schaumann (Ed.), Science of
Synthesis—Five-Membered Hetarenes with One Chalcogen and One
Additional Heteroatom, II/11, Georg Thieme Verlag, Stuttgart, 2002,
pp. 627−833.
[16] L. Peterlin-Mašicˇ, A. Jurca, P. Marinko, A. Jancˇar, D. Kikelj, Tetrahe-
dron 58 (2002) 1557–1563.
[17] (a) Y. Cheng, W.H. Prusoff, Biochem. Pharmacol 22 (1973) 3099–
3108; (b) R.B. Brandt, J.E. Laux, S.W. Yates, Biochem. Med. Metab.
B. 37 (1987) 344–349.
Acknowledgements
The authors thank Dr. T.B. Tschopp and Dr. B. Steiner
(Pharma Division, Preclinical Research, F. Hoffmann La
Roche Ltd, Basel, Switzerland) for biological testing of some
compounds, Prof. W. Bode for offering us the possibility to
work in his group at the Max Planck Institut für Biochemie,
Martinsried, Germany, R. Friedrich for the assistance with
solving the X-ray structures and Dr. Roger Pain (Jožef Stefan
Institute, Ljubljana, Slovenia) for critical reading of the
manuscript. Financial support of this work by the Ministry of
Education, Science and Sport of the Republic of Slovenia
(Grant No P0787-502 and L3-2034) and by Lek Pharmaceu-
ticals d.d., Ljubljana is gratefully acknowledged.
[18] Log P and pKa values were calculated with the packageACD/pKa/log
P version 5.0. Available from http://www.acdlabs.com.
[19] Coordinates have been deposited in the Protein Data Bank (accession
code 15B1).
[20] M. Breznik, S. Golicˇ Grdadolnik, G. Giester, I. Leban, D. Kikelj, J.
Org.Chem. 66 (2001) 7044–7050.
References
[21] (a) D.M. Feng, S.J. Gardell, S.D. Lewis, M.G. Bock, Z. Chen,
R.M. Freidinger, A.M. Naylor-Olsen, H.G. Ramjit, R. Woltmann,
E.P. Baskin, J.J. Lynch, R. Lucas, J.A. Shafer, K.B. Dancheck,
I.W. Chen, S.S. Mao, J.A. Krueger, T.R. Hare, A.M. Mulichak,
J.P. Vacca, J. Med. Chem. 40 (1997) 3726–3733; (b) V.J. Klimkowski,
M.R. Wiley, US Patent 5811402 (1998).
[22] E. Skrzypczak-Jankun, V.E. Carperos, K.G. Ravichandran, A. Tulin-
sky, M. Westbrook, J.M. Maraganore, J. Mol. Biol. 221 (1991) 1379–
1393.
[1] E.W. Davie, K. Fujikawa, W. Kisiel, Biochemistry 30 (1991) 10363–
10370.
[2] J.W. Fenton II, F.A. Ofosu, D.G. Moon, J.M. Maraganore, Blood
Coagul. Fibrin. 2 (1991) 69–75.
[3] For recent reviews, see (a) K. Menear, Curr. Med. Chem. 5 (1998)
457–468; (b) J.B.M. Rewinkel, A.E.P. Adang, Curr. Pharm. Design 5
(1999) 1043–1075; (c) J.P. Vacca, Curr. Opin. Chem. Biol. 4 (2000)
394–400; (d) J.M. Fevig, R.R. Wexler, Annu. Rep. Med. Chem. 34
(1999) 81; (e) T. Steinmetzer, J. Hauptmann, J. Stürzebecher, Exp.
Opin. Invest. Drugs 10 (2001) 845–864; (f) C.A. Coburn, Exp. Opin.
Ther. Patents 11 (2001) 721–738.
[23] A.G.W. Leslie, Newsletter on protein crystallography, No. 26, Dares-
bury Laboratory, Warrington, UK, 1992 Science and Engineering
Research Council.
[4] (a) D. Gustafsson, M. Elg, S. Lenfors, I. Borjesson, A.C. Teger
Nilsson, Blood Coagul. Fibrin. 7 (1996) 69–79; (b) A.C. Teger Nils-
son, R. Bylund, D. Gustafsson, E. Gyzander, U. Eriksson, Thromb.
Res. 85 (1997) 133–145 ; (b).
[5] (a) B.I. Eriksson, S. Carlsson, M. Halvarsson, B. Risberg, C. Matts-
son, Thromb. Haemostasis 78 (1997) 1404–1407; (b) D. Gustaffson,
T. Antonsson, R. Bylund, U. Eriksson, E. Gyzander, I. Nilsson,
M. Elg, C. Mattsson, J. Deinyúm, S. Pehrsson, A. Karlsson,
H. Sörensen, Thromb. Haemostasis 79 (1998) 110–118.
[24] W. Kabsch, J. Appl. Crystallogr. 21 (1988) 916–924.
[25] J. Navaza, Acta Cryst. A50 (1994) 157–163.
[26] A.T. Brunger, P.D. Adams, G.M. Clore, W.L. DeLano, P. Gros,
R.W. Grosse-Kunstleve, J.S. Jiang, J. Kuszewski, M. Nilges,
N.S. Pannu, R.J. Read, L.M. Rice, T. Simonson, G.L. Warren, Acta
Cryst. D54 (1998) 905–921.
[27] R. Lottenberg, J.A. Hall, M. Blinder, E.P. Binder, C.R. Jackson,
Biochem. Biophys. Acta 742 (1983) 539–557.
[28] G.S. Eadie, J. Biol. Chem. 146 (1942) 85–90.