Design of selective thrombin inhibitors based on the (R)-Phe-Pro-Arg sequence

Article abstract of DOI:10.1021/jm011133d

Potent and selective inhibitors of thrombin were sought based on the (R)-Phe-Pro-Arg sequence. The objective was to generate similar binding interactions to those achieved by potent competitive inhibitors of the argatroban type, so eliminating the need for covalent interaction with the catalytic serine function, as utilized by aldehyde and boronic acid type inhibitors. Improving the S₁ subsite interaction by substitution of arginine with a 4-alkoxybenzamidine residue provided potent lead 2 (K_i = 0.37 nM). Though an amide bond, which H-bonds to the active site, is lost, modeling indicated that a new H-bond is generated between the alkoxy oxygen atom and the catalytic Ser-195 hydroxyl group. Substitution of the benzamidine system by 1-amidinopiperidine then gave compound 4, which provided a further gain in selectivity over trypsin. However, previous work had shown that these compounds were likely to be too lipophilic (Log D +0.4 and +0.2, respectively) and to suffer rapid hepatic extraction, presumably via biliary elimination. Accordingly, both proved short-acting when administered intravenously to rats and showed poor activity when given intraduodenally. The aim was then to reduce lipophilicity below a log D of - 1.2, which in a previously reported series had been effective in preventing rapid clearance. It was anticipated that compounds of this type would rely on the cation selective paracellular route of absorption from the gastrointestinal tract. Potent polar analogues with selectivity >1000 over trypsin were obtained. The best in vivo activity was shown by compound 12. However, in the final analysis, its oral bioavilability proved poor, relative to analogues with similar physicochemical properties derived from argatroban, consistent with the hypothesis that molecular shape is an additional important determinant of paracellular absorption.

Full text of DOI:10.1021/jm011133d

2432
J . Med. Chem. 2002, 45, 2432-2453
Design of Selective Th r om bin In h ibitor s Ba sed on th e (R)-P h e-P r o-Ar g
Sequ en ce
J ohn C. Danilewicz,†,‡ Stuart M. Abel,§ Alan D. Brown,‡ Paul V. Fish,*,‡ Edward Hawkeswood,|
Stephen J . Holland,| Keith J ames,‡ Andrew B. McElroy,‡ J ohn Overington, Michael J . Powling,| and
David J . Rance§
Departments of Discovery Chemistry, Drug Metabolism, Discovery Biology, and Molecular Informatics Structure and Design,
Pfizer Global Research and Development, Sandwich, Kent CT13 9NJ , United Kingdom
Received December 21, 2001
Potent and selective inhibitors of thrombin were sought based on the (R)-Phe-Pro-Arg sequence.
The objective was to generate similar binding interactions to those achieved by potent
competitive inhibitors of the argatroban type, so eliminating the need for covalent interaction
with the catalytic serine function, as utilized by aldehyde and boronic acid type inhibitors.
Improving the S1 subsite interaction by substitution of arginine with a 4-alkoxybenzamidine
residue provided potent lead 2 (Ki ) 0.37 nM). Though an amide bond, which H-bonds to the
active site, is lost, modeling indicated that a new H-bond is generated between the alkoxy
oxygen atom and the catalytic Ser-195 hydroxyl group. Substitution of the benzamidine system
by 1-amidinopiperidine then gave compound 4, which provided a further gain in selectivity
over trypsin. However, previous work had shown that these compounds were likely to be too
lipophilic (Log D +0.4 and +0.2, respectively) and to suffer rapid hepatic extraction, presumably
via biliary elimination. Accordingly, both proved short-acting when administered intravenously
to rats and showed poor activity when given intraduodenally. The aim was then to reduce
lipophilicity below a log D of -1.2, which in a previously reported series had been effective in
preventing rapid clearance. It was anticipated that compounds of this type would rely on the
cation selective paracellular route of absorption from the gastrointestinal tract. Potent polar
analogues with selectivity >1000 over trypsin were obtained. The best in vivo activity was
shown by compound 12. However, in the final analysis, its oral bioavilability proved poor,
relative to analogues with similar physicochemical properties derived from argatroban,
consistent with the hypothesis that molecular shape is an additional important determinant
of paracellular absorption.
In tr od u ction
basic P1 side chain pack together to interact with the
hydrophobic S2 site.5-7 Napsagatran (Ro 46-6240),
developed by Hilpert et al.,8 though having a more
complex P1 residue, can nevertheless be viewed as
belonging to this group. The only interaction with the
catalytic serine residue is via a hydrogen bond to the
carboxylate function in both argatroban and napsagat-
ran. Unfortunately, none of these compounds is orally
active due to either poor absorption from the gas-
trointestinal tract and/or rapid clearance via the bile.4,9
A second inhibitor type is based on the substrate-
derived irreversible chloromethyl ketone inhibitor PPACK
and includes compounds such as DuP-71410,11 and
efegatran (GYKI-14 766).12 These compounds interact
covalently with the hydroxyl group of the catalytic
serine residue. The neighboring proline ring and (R)-
Phe side chain cooperate to fill the S2 site in a similar
fashion as the two distal lipophilic groups of the first
series.5 Though oral activity has been claimed for these
compounds, we were concerned that high enzyme se-
lectivity might not be obtainable when substantial
affinity is derived by interacting covalently with the
ubiquitous active site serine function. In the case of
aldehyde type inhibitors, there is also the potential
problem of achieving adequate optical and chemical
stability.
The search for potent selective and orally active
thrombin inhibitors has gathered momentum in recent
years.1 Thrombin is the last in a cascade of trypsin-like
plasma serine proteases, which by catalyzing the con-
version of fibrinogen to fibrin, activation of FXIII and
inducing platelet aggregation is a key enzyme in hae-
mostasis and thrombus formation. The inhibition of a
single enzyme in the cascade, and in particular throm-
bin, has been an attractive goal in that it could also
provide superior antithrombotic therapy by increasing
efficacy and safety as compared to heparin and the
coumarins. Additionally, by keeping molecular size
small, the opportunity exists for obtaining oral activ-
ity.1,2
Two small molecular weight inhibitor types are
emerging as structure-activity relationships are ex-
plored. The first is of the argatroban3 and NAPAP4 type
(Chart 1), where lipophilic groups on either side of the
* To whom correspondence should be addressed. Tel.: 44 1304
644589. Fax: 44 1304 656433. E.mail: paul_fish@sandwich.pfizer.com.
† Senior author.
‡ Department of Discovery Chemistry.
§ Department of Drug Metabolism.
| Department of Discovery Biology.
Molecular Informatics Structure and Design.
10.1021/jm011133d CCC: $22.00 © 2002 American Chemical Society
Published on Web 05/11/2002

Design of Selective Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12 2433
Ch a r t 1
Subnanomolar potency and oral bioavailability have
been obtained with “folded” inhibitors of the argatroban
type, e.g., UK-154 606.13 We considered that these
properties might be equally well-combined in “linear”
PPACK-derived structures but ones in which the reac-
tive functionality was deleted to ensure competitive
kinetics and to enhance the potential for high selectivity.
While our work was in progress,14 two series, which
included inogatran15 and compound 1,16 were published.
Inogatran is reported to have a Ki of 15 nM against
human thrombin, which gave encouragement that the
above objective should be achievable. We describe here
our own work on this approach.
Com p ou n d Design
In their early studies on aryl amidines, Markwardt
et al.17 showed that simple 4-alkoxyphenylamidines
possess significant thrombin and trypsin inhibitory
activity, with Ki values in the 5 × 10-5 M region. We,
therefore, sought to determine how such a P1 side chain
might be incorporated into a linear inhibitor type. The
published complex of human thrombin with PPACK18
(PDB19 code 1PPB) was used as a basis for modeling
studies. The covalent linkage to Ser 195 and His 57
along with the bridging methylene was removed by
modeling, and the resulting noncovalently bound com-

2434 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12
Danilewicz et al.
Ta ble 1
a
Calculated from the measured value for 5 using Rekker fragmental constants.
plex was minimized using the CHARMM program.20
The histidine and serine residues adopted conformations
closer to those observed for the remainder of the serine
proteinase family and to those observed in thrombin
complexes with noncovalently bound inhibitors,5,6 while
the inhibitor backbone and side chains maintained
similar interactions with the enzyme. When the complex
of benzamidine with human thrombin (PDB code
1DWB18) was examined with respect to this model, it
showed that a 4-oxy substituent would be best joined
via a two carbon link to the proline pyrrolidine ring.
Such a modification would result in loss of an amide
linking group, the NH of which, in PPACK, is H-bonded
to the Ser-214 carbonyl function of thrombin. Initially,
we argued that loss of this interaction might, in large
measure, be compensated for by a reduction in desol-
vation energy on active site entry. However, it appeared
from our model that a new H-bond between the alkoxy
oxygen atom and the catalytic Ser-195 hydroxyl group
was also possible. These considerations led to the design
of compound 2, but because racemic 2-(2-hydroxyethyl)-
piperidine was used as starting material, the diastereo-
isomer 3 was also obtained. The more active isomer 2
proved highly potent and was assigned the S configu-
ration on the basis of the modeling studies.
Very interestingly, Hilpert et al.8 had shown that
1-amidinopiperidine is intrinsically more selective than
benzamidine for thrombin and used this finding to
produce their novel napsagatran series, which has high
thrombin/trypsin selectivity. 1-Amidinopiperidine was
therefore docked into the P1 pocket, which showed that
this system could effectively replace the benzamidine
fragment. This resulted in the design of compound 4.
High potency was indeed retained, with a further small
gain in selectivity, taking into account that 4 is a
mixture of diastereoisomers (Table 1). It is of note that
a related alkoxy 1-amidinopiperidine P1 group has been
F igu r e 1. Prolongation in TT on iv administration of com-
pound 2 to rats at 1 mg/kg; means ( SEM, n ) 4.
successfully employed by Soll et al. in a different
inhibitor template.21
In a previous paper,13 it was shown that compounds
of this type require a degree of polarity, with log D
values less than or equal to -1.2, to minimize hepatic
extraction and so allow some oral bioavailability. Ab-
sorption of such polar compounds from the gastrointes-
tinal tract is then presumably achieved by the paracel-
lular route. When given intravenously (iv) to rats,
compound 2 showed only a moderate duration of action,
as indicated by the relatively rapid decline in thrombin
time (TT) in Figure 1, when compared with similarly
potent inhibitors that have good pharmacokinetics.13
Prolongation of TT was routinely used for in vivo
compound evaluation as a more sensitive measurement
of activity than the more commonly used activated
partial thromboplastin time (APTT). Little activity was
also evident at a standard dose of 10 mg/kg given
intraduodenally (idd), as shown in Figure 2. Thus, the
in vivo profile was indeed indicative of too rapid a
clearance, which was consistent with a measured log D

Design of Selective Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12 2435
F igu r e 3. Prolongation in TT on iv administration of com-
pound 12 to rats at 1 mg/kg; means ( SEM, n ) 4.
F igu r e 2. Prolongation in TT on idd administration of
compound 2 to rats at 10 mg/kg; means ( SEM, n ) 4.
as in the cyclohexylmethyl and benzyl analogues 15 and
16, reduced potency. Similarly, introduction of a het-
eroatom as in compounds 19-21 was detrimental.
Extension of the carboxylic acid side chain as in the
glutamic acid analogue 22 (Table 3), which more closely
mimicked compound 4, reduced selectivity as compared
to the corresponding aspartic acid analogue 12 and
offered no in vivo advantage.
In their pioneering work on thrombin inhibitors, the
group at Mitsubishi23,24 showed that the incorporation
of a carboxyl group, as in argatroban, was important to
improve compound toleration. We showed a similar
effect in a related series of phenylamidine-derived
inhibitors,13 the carboxyl group being in the same
relative position in both series, forming an H bond to
the catalytic serine residue. The same phenomenon
appeared to operate in this current series. Compound
6 (Table 3) was lethal when given iv to rats at 1 mg/kg.
In contrast, compounds 2 and 12 were tolerated at 30
mg/kg when given iv to mice. It is apparent that the
carboxyl group in these latter compounds is positioned
quite differently than in the argatroban-derived series.
Thus, all that may be required is the mere physical
presence of a carboxyl group anywhere in the molecule,
that is compatible with thrombin inhibitory activity, to
improve toleration.
The more active compounds were evaluated in vivo,
and the best responses, shown in Figures 3 and 4, were
obtained with compound 12. Compound 13 gave a less-
sustained response when given idd to rats. Therefore,
despite reduced in vitro potency, compound 12 exhibited
an in vivo profile that is clearly superior to that of
compound 4 (Log D +0.2). The log D of 12 was found to
be -1.8. Thus, again, a reduction in lipophilicity is seen
to be beneficial in improving in vivo activity.
The selectivity of compound 12 was examined against
a representative selection of thrombin-related enzymes,
as can be seen from Table 4. Relatively low activity was
evident against factor (F) Xa, plasmin, and FVIIa, with
selectivities of >1000 being achieved. Selectivity against
trypsin was slightly lower (900×), though values >1000×
were achieved with compounds 13 and 14.
of +0.4. Compound 4, with a log D of +0.2, showed a
similarly poor duration of action when given iv to rats
at 1 mg/kg. Replacement of the cyclohexyl ring by a
phenyl ring as in compound 5 reduced log D to -1.1,
which is in our desired range, but potency fell sharply
showing the importance of the hydrophobic interaction
with the S2 site.
Our objective was then to modify the N-carboxy-
methyl-(R)-cyclohexylalanine (Cha) fragment so as to
achieve a substantial reduction in lipophilicity. N-alkyl
aspartic acid derivatives were selected for study (Table
2), as this residue satisfied four requirements. First, the
N atom would provide a proton (even when tertiary, if
protonated), which modeling predicted could still inter-
act with the Gly-216 residue. The thrombin-PPACK
X-ray crystal structure5,18 shows that the (R)-Phe car-
bonyl and amino groups form antiparallel H bonds to
the corresponding amido and carbonyl functions of Gly-
216. Second, the total lipophilicity contribution of the
alkyl substituent would be reduced if brought under the
close influence of a protonated N atom. Third, a free
carboxyl group, which appears to be important for good
in vivo toleration, as discussed below, would be retained,
and fourth, a net positive charge would be maintained,
so favoring absorption by the paracellular route, which
is cation selective.22 An important consideration, how-
ever, is that it would be necessary to invert stereochem-
istry and to use (S)-aspartic acid, so that the relative
orientation of lipophilic and carboxyl groups in com-
pound 4 would be preserved.
The N-cyclohexyl analogue 7 was the first to be
synthesized and showed a promising level of activity.
The cyclopentyl analogue 8 was inferior, but further
small gains in potency were possible by increasing ring
size as in compounds 9 and 10, taking into account that
some compounds were prepared as mixtures of diaste-
reoisomers, one of which was assumed to be only weakly
active (cf. compounds 2 and 3). Activity was retained
on “opening” the ring as with the 3-pentyl analogue 11.
Because the N substituent is presumed to extend into
the hydrophobic S2 site, the effect of further substitution
was examined. N-methylation gave a small increace in
potency (compounds 12 and 13), but introduction of an
ethyl group resulted in loss of activity (compound 17).
Large polar groups such as dimethylaminoethyl were
detrimental. The introduction of a double bond into the
cycloalkyl group maintained potency as in compound 14.
Attempts to further extend the hydrophobic interaction,
En zym e-In h ibitor Bin d in g
Details of the interactions of PPACK with thrombin
have been described in detail.18 However, briefly, as
shown in Figure 5, aside from the covalent bonds formed
to Ser 195 and His 57, a substratelike, antiparallel

2436 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12
Danilewicz et al.
Ta ble 2
a
b
Inhibition at 10-6 M. Inhibition at 10-5 M. c Inhibition at 10-4 M.
â-sheet type interaction is formed between the peptide
backbone of PPACK and the mainchain of the enzyme
(NH of P1 Arg to carbonyl of Ser 214, carbonyl of P3 (R)-
Phe to amide of Gly 216, and amino terminus of P3 with
carbonyl of Gly 216). The carbonyl of P2 Pro is exposed
to solvent in the complex. The guanidine of the P1 Arg
residue forms an “end-on” salt bridge to Asp 189 of the
enzyme and, further, hydrogen bonds to the carbonyl
of Gly 219 and also to a “conserved” buried water
molecule at the base of the P1 pocket. The methylenes
of the pyrrolidine ring of P2 Pro are cradled under a
hydrophobic lid formed by the aryl portions of Tyr 60A
and Trp 60D, from the loop insertion unique to throm-
bin. It is this unique loop that is thought to be

Design of Selective Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12 2437
Ta ble 4. Enzyme Inhibition Profile of Compound 12
Ta ble 3
enzyme
calcd Ki
thrombin
trypsin
FXa
plasmin
FVIIa
2.17 ( 0.09 × 10-9
1.93 ( 0.12 × 10-6
4.97 ( 0.24 × 10-6
3.30 ( 0.15 × 10-5
M
M
M
M
38.7 ( 1.9% inhibition at 10-4
M
restrained molecular dynamics simulations, selected
water molecules from the thrombin-PPACK complex
structure that were included in the calculations (Figure
6). As in PPACK, a series of important “backbone”
interactions are formed, despite the fact that 12 only
contains one peptide bond. The proton associated with
the tertiary amine (which we assume is protonated at
physiological pH) interacts with the carbonyl of Gly 216,
while the amide carbonyl of 12 forms a hydrogen bond
with the amide proton of Gly 216. The hydrogen bond
corresponding to that from the P2 residue of PPACK is
absent in 12, with the carbonyl of Ser 214 now partially
exposed to solvent; however, a new hydrogen bond
would appear to be formed between the ether oxygen of
12 and the active site residue, Ser 195. As discussed
above, it is possible that the formation of this hydrogen
bond compensates for the loss of the preceding main
chain interaction. A further interaction to the main
chain of thrombin is formed by one of the carboxylate
oxygens, corresponding to a substrate-like main chain
interaction for a P4 residue.
The guanidine of 12 forms an end-on salt bridge
interaction with Asp 189 at the base of the P1 pocket,
with further required hydrogen bonds observed to the
backbone carbonyl of Gly 219 and a water molecule
bound at the base of the P1 pocket, as with PPACK. The
methylene atoms of the piperidine ring, however, form
extensive van der Waals interactions with the remain-
ing hydrophobic parts of the P1 pocket. It is interesting
to note that there is a sequence difference between
thrombin
Ki (nM)
trypsin
Ki (nM)
selectivity
tryp/throm
compd
*
22
6
S
RS
1.8
56
550
1600
305
28
F igu r e 4. Prolongation in TT on idd administration of
compound 12 to rats at 10 mg/kg; means ( SEM, n ) 3.
responsible for a large part of the selectivity of thrombin
in vivo. The (R)-Phe residue binds into a hydrophobic
pocket formed by Leu 99, Ile 174, and Trp 215.
A model of compound 12 bound to the active site of
human thrombin was built using PPACK as the tem-
plate and the modeling program QUANTA.25 The model
was refined by energy minimization techniques and
F igu r e 5. Stereoview of active site of human thrombin-PPACK thrombin complex (PDB code 1PPB), showing binding mode of
PPACK (thick lines). Hydrogen bonds are indicated by dotted gray lines. Residues in thrombin are labeled according to established
numbering conventions.

2438 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12
Danilewicz et al.
F igu r e 6. Stereoview of modeled binding of 12 to active site of human thrombin. See legend to Figure 5 for further details.
trypsin and thrombin (Ser 190 in trypsin is Ala in
thrombin) in the P1 pocket that favors the binding of
more lipophillic ligands in the P1 pocket of thrombin,
and this gives rise to some of the observed thrombin
selectivity of 12.
2 and 3. The more active was assigned the S configu-
ration on the basis of the computer modeling studies
described above, the R diastereoisomer showing the
much poorer fit. Diimide coupling of 25 with N-Cbz-(R)-
Phe-OH afforded compound 60. Treatment of 60 with
ethanolic HCl, to form the imidate, resulted only in
partial simultaneous removal of the Cbz-protecting
group. Thus, when the product was treated with am-
monia, a mixture of 6 and its N-Cbz derivative, which
were readily separable by chromatography, was ob-
tained.
The guanidine analogues 4 and 5 were prepared as
shown in Schemes 2 and 3. N-Cbz-2-(2-hydroxyethyl)-
piperidine 35b was converted to the methanesulfonate
36b, which was reacted with the Na alkoxide of N-Boc-
4-hydroxypiperidine in dimethylformamide (DMF) to
give the ether 37b (Scheme 2). Hydrogenolysis of the
Cbz-protecting group and reaction of amine 41 with the
acid chloride of N-Troc-(R)-Cha gave 42 (Scheme 3).
Removal of the Boc-protecting group and reaction of
amine 43 with N,N′-di-t-butoxycarbonyl-S-methylisothio-
urea29 then gave bis-Boc-protected guanidine 44. We
found, however, that inclusion of mercuric chloride
greatly facilitated this reaction and improved yields.
Removal of the Troc-protecting group with zinc, as
described above, gave 45, which on reaction with tert-
butyl bromoacetate and acid treatment, gave compound
4 as a mixture of diastereoisomers.
The binding in the aryl pocket is quite distinct to that
observed in PPACK. The cyclohexyl group shows ex-
tensive hydrophobic contacts with Trp 215, Leu 99, and
Ile 174, and the N-methyl group extends this pattern
of hydrophobic interactions, primarily increasing inter-
actions with Ile 174. This area is more polar in trypsin
(the residue corresponding in position to Ile 174 is a
glutamine residue), and so, the additional hydrophobic
interactions possible in thrombin contribute to the
observed enhanced potency and selectivity. Extending
the methyl group to larger alkyl groups would affect the
conformation of the essential acid moiety and start to
shield polar atoms of thrombin from contact with
solvent, consistent with the observed loss of potency on
extending this group.
Ch em istr y
The methods for the synthesis of the inhibitors
disclosed in Tables 1-3 (compounds 2-22) are described
in Schemes 1-6. Phenylamidines 2, 3, and 6 were
prepared as outlined in Scheme 1. Compound 23 was
reacted with 4-cyanophenol under Mitsunobu condi-
tions26 to give the phenyl ether 24. Removal of the Boc-
protecting group using trifluoroacetyl (TFA) in the
presence of anisole, as a carbonium ion scavenger, gave
amine 25. Coupling with the acid chloride of N-Troc-
(R)-Cha then gave the diastereoisomers 26 and 27,
which were readily separated by chromatography on
silica. Removal of the Troc-protecting group with zinc27
afforded amines 28 and 29. Reaction with ethyl bro-
moacetate, followed by the sequential treatment with
ethanolic HCl and ethanolic ammonia (Pinner reac-
tion28), gave amidines 32 and 33. Hydrolysis with NaOH
followed by acidification then gave the diastereoisomers
Removal of the Boc-protecting group from 37b (Scheme
2), followed by reaction of the resultant amine 38b with
N,N′-di-t-butoxycarbonyl-S-methylisothiourea gave bis-
Boc-protected guanidine 39b in good yield. Hydro-
genolysis of the Cbz-protecting group gave 40b, which
on coupling with N-Fmoc-(R)-Phe-OH, using PyBroP,30
gave 47. Removal of the Fmoc-protecting group with
piperidine and reaction of 48 with tert-butyl bromoac-
etate, followed by acid treatment, gave 5 as the hydro-
chloride salt.
Compound 34b was resolved as the (1S)-(+)-10-
camphorsulfonic acid salt,31,32 the absolute configuration

Design of Selective Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12 2439
Sch em e 1a
a
Reagents: (a) Boc2O, EtOAc; (b) 4-cyanophenol, DEAD, PPh3, THF; (c) TFA, PhOMe, CH2Cl2; (d) Troc-D-Cha-Cl, DIPEA, CH2Cl2; (e)
chromatographic separation of diastereoisomers; (f) Zn, KH2PO4, THF; (g) BrCH2CO2Et, K2CO3, MeCN; (h) HCl and then NH3; (i) NaOH
then HCl; (j) Cbz-D-Phe-OH, WSCDI, HOBt, NMM, CH2Cl2.
of which was confirmed as S by single-crystal X-ray
crystallographic analysis. The free base 34a was then
elaborated, in a manner similar to 34b, to give 40a .
Both 40a ,b were coupled with N-Fmoc-(S)-aspartic acid
â-tert-butyl ester to give 50a ,b (Scheme 4). Removal of
the Fmoc-protecting group, followed by reductive ami-
nation with a range of ketones and aldehydes, gave
52a -l. Acid treatment then gave 7-11, 15, 16, 19, and

2440 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12
Danilewicz et al.
Sch em e 2a
a
Reagents: (a) CbzOSuc, TEA; (b) MsCl, TEA, CH2Cl2; (c) N-Boc-4-hydroxypiperidine, NaH, DMF; (d) HCl, CH2Cl2; (e) MeSC(dNBoc)NHBoc,
HgCl2, TEA, CH2Cl2; (f) H2, Pd/C, EtOH; (g) Fmoc-D-Phe-OH, PyBroP, DIPEA, CH2Cl2; (h) piperidine, THF; (i) BrCH2CO2tBu, K2CO3,
MeCN; (j) HCl, CH2Cl2.
21. Methylation of 52c,f,g,k , followed by acid treatment,
gave 12-14 and 20.
Compound 52a was also alkylated with acetaldehyde
and 2-chloroethyldimethylamine to give 54 and 55,

Design of Selective Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12 2441
Sch em e 3a
a
Reagents: (a) Pd/C, 1,4-cyclohexadiene, EtOH; (b) Troc-D-Cha-Cl, DIPEA, CH2Cl2; (c) HCl, CH2Cl2; (d) MeSC(dNBoc)NHBoc, HgCl2,
TEA, CH2Cl2; (e) Zn, AcOH; (f) BrCH2CO2tBu, K2CO3, MeCN.
respectively, which on acid treatment afforded the acids
17 and 18 (Scheme 5). Reaction of 40a with N-Fmoc-
(S)-glutamic acid γ-tert-butyl ester gave 56 as shown
in Scheme 6. Reaction of 56 in a manner similar to that
shown in Scheme 4 then gave the glutamic acid ana-
logue 22.
substantial in both rat (39% of the iv dose) and dog
(21%). The overall iv pharmacokinetic profiles for 12 and
61 are therefore very similar (Table 5). In particular,
non-renal clearance is low with respect to hepatic blood
flow for both compounds in both species, indicating very
little first-pass extraction. However, after oral admin-
istration, the bioavailability of compound 12 was found
to be only 2% in the rat and 9% in the dog, indicative of
very low absorption. These poor results contrasted
markedly with the corresponding values of 16 and 37%
obtained with compound 61. We were particularly
concerned as our experience indicated that for polar
compounds absorbed by the paracellular route, the rat
is likely to be the more predictive species for man.
Compounds 12 and 61 have similar physicochemical
properties [12: Mw 465.6, Log D7.4 -1.8, pKa 8.8 (second
basic center); 61: Mw 553.7, Log D7.4 -1.8, pKa 8.0
P h a r m a cok in etics a n d Disp osition
The pharmacokinetics of compound 12 were examined
in detail and compared with one of our best benzamidine
analogues, 61 (UK-179 094).13 As can be seen in Table
5, following iv administration of 12, volumes of distribu-
tion at steady state were 0.31 and 0.53 L/kg for rat and
dog, respectively. Corresponding plasma clearance val-
ues were 21.1 and 6.4 mL/min/kg, and â-elimination
half-life values were 0.8 and 1.0 h in rat and dog,
respectively. Renal elimination of unchanged drug was

2442 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12
Danilewicz et al.
Sch em e 4a
a
Reagents: (a) Fmoc-L-Asp(OtBu)-OH, PyBroP, DIPEA, CH2Cl2; (b) piperidine, THF; (c) ketone, NaBH(OAc)3, AcOH, THF; (d) aldehyde,
NaBH(OAc)3, THF; (e) 37% HCHO(aq), NaBH(OAc)3, CH2Cl2; (f) HCl, CH2Cl2.
(second basic center)]; they have the same Log D values,
and 12 even has the potential advantage of the lower
molecular weight. However, the second basic pKa of 12
is a little greater than that of 61, but because we believe
that these inhibitors are absorbed by the paracellular
route, that is through aqueous channels, we would not
expect that small differences in the degree of ionization
would influence absorption from the gastrointestinal
tract. We conclude that molecular shape, which is
influenced by folding, may be the determining factor.
Con clu sion
We have successfully developed a series of highly
potent and selective thrombin inhibitors, based on the
(R)-Phe-Pro-Arg sequence, without recourse to covalent
receptor interaction. By adjusting polarity and the
incorporation of a carboxyl group, presystemic clearance
has been minimized and toleration has been improved.
However, the best compound of the series, 12 (UK-
285 954), showed low oral bioavailability as compared

Design of Selective Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12 2443
Sch em e 5a
a
Reagents: (a) MeCHO, NaBH(OAc)3Cl2, THF; (b) HCl‚Me2NCH2CH2Cl, K2CO3, MeCN; (c) HCl, CH2Cl2.
Ta ble 5. Pharmacokinetic Parameters for Compounds 12 and
61 in Rat and Dog
alanine; DEAD, diethyl azodicarboxylate; DIPEA, diisopropyl-
ethylamine; DMAP, 4-(dimethylamino)pyridine; Fmoc-Asp-
(OtBu)-OH, N-R-fluorenylmethoxycarbonyl-(S)-aspartic acid
â-tert-butyl ester; Fmoc-Glu(OtBu)-OH, N-R-fluorenylmethoxy-
carbonyl-(S)-glutamic acid â-tert-butyl ester; Fmoc-D-Phe-OH,
N-R-fluorenylmethoxycarbonyl-(R)-phenylalanine; HOBt, N-
hydroxybenzotriazole; MsCl, methanesulfonyl chloride; NMM,
N-methylmorpholine; PyBroP, bromotris(pyrrolidino)phos-
phonium hexafluorophosphate; TEA, triethylamine; TFA, tri-
fluoroacetic acid; Troc-D-Cha-OH, N-(2,2,2-trichloroethoxy-
carbonyl)-(R)-Cha; WSCDI, 1-(3-dimethylaminopropyl)-3-eth-
ylcarbodiimide hydrochloride; 0.88NH3, ammonia solution
specific gravity 0.88.
parameter
compd
rat
dog
12
10
21.2
0.31
0.12
0.8
61
10
11
0.60
12
61
Melting points were determined on a Buchi 510 apparatus
and are uncorrected. 1H nuclear magnetic resonance (NMR)
spectroscopic data were recorded on a Varian Unity 300 or a
Bruker AC-300 NMR instrument. The 1H NMR spectra of
advanced intermediates were somewhat complex due to the
occurrence of rotamers, and in some cases diastereoisomers,
but were thoroughly consistent with the assigned structure.
Mass spectra were obtained with a Fisons Trio-1000 spec-
trometer using thermospray ionization. Column chromatog-
raphy was accomplished on Kieselgel 60 (230-400 mesh)
obtained from E. Merck, Darmstadt. Kieselgel 60 F254 plates,
also from E. Merck, were used for thin-layer chromatography
(TLC), and compounds were visualized with UV light and
sprayed sequentially with iodoplatinate reagent (chloroplatinic
acid, KI, and H2O), Dragendorff’s reagent (bismuth oxynitrate,
AcOH-H2O; KI, H2O), and aqueous NaNO2. Analar or high-
performance liquid chromatography (HPLC) grade solvents
iv dose level (mg/kg)
clearance (mL/min/kg)
volume of dist (L/kg)
R-elim half-life (h)
â-elim half-life (h)
po dose level (mg/kg)
oral bioavailability (%)
1
1
6.9
0.53
6.4
0.53
0.16
1.0
10
9
0.8
50
16
0.90
2
37
50
2
to our most successful examples of benzamidine type
inhibitors, such as 61 (UK-179 094). We conclude that
the optimization of physicochemical parameters is in-
sufficient and that molecular size and shape also appear
to be important in allowing absorption of polar com-
pounds from the gastrointestinal tract by the para-
cellular route.
Exp er im en ta l Section
were used, and evaporation was conducted on
a rotary
evaporator at 25-50 °C under vacuum.
N-t-Bu t yloxyca r b on yl-2(RS)-(2-h yd r oxyet h yl)p ip er i-
d in e (23). A solution of Boc2O (10.91 g, 50 mmol) in EtOAc
Ch em istr y. The abbreviations used are as follows: Boc2O,
di-tert-butyl dicarbonate; CbzOSuc, N-(benzyloxycarbonyloxy)-
succinimide; Cbz-D-Phe-OH, N-benzyloxycarbonyl-(R)-phenyl-

2444 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12
Danilewicz et al.
Sch em e 6a
a
Reagents: (a) Fmoc-L-Glu(OtBu)-OH, PyBroP, DIPEA, CH2Cl2; (b) piperidine, THF; (c) cyclohexanone, NaBH(OAc)3, AcOH, THF; (d)
37% HCHO(aq), NaBH(OAc)3, CH2Cl2; (e) HCl, CH2Cl2.
(15 mL) was added to a stirred solution of 2(RS)-(2- hydroxy-
ethyl)piperidine (34b) (5.06 g, 50 mmol) in EtOAc (25 mL) at
0 °C, and the mixture was then warmed to 23 °C. After a
further 1.5 h, the reaction mixture was evaporated under
reduced pressure and the residue was purified by chromatog-
raphy on silica gel using hexanes-Et2O (1:1) as eluant to
provide BOC-amine 23 (8.46 g, 74%) as a clear oil. Rf 0.25
(hexane:Et2O, 1:1). 1H NMR (CDCl3): δ 1.30-1.85 (m, 8 H),
1.48 (s, 9 H), 1.92 (t, 1 H), 2.68 (t, 1 H), 3.16 (m, 1 H), 3.56-
3.66 (m, 1 H), 3.94 (d, 1 H), 4.22 (s, 1 H).
amine 24 (2.83 g, 8.56 mmol) and anisole (1.88 mL, 17.3 mmol)
in dry CH2Cl2 (15 mL). After a further 1 h, the reaction mixture
was evaporated under reduced pressure, the residue was
dissolved in water, and the solution was washed with Et2O,
basified with aqueous NaOH (2 M), and extracted with Et2O
(×3). The combined extracts were washed with brine, dried
(Na2SO4), and evaporated under reduced pressure to afford
amine 25 (1.43 g, 75%) as an oil, which solidified on standing.
1
Rf 0.15 (CH2Cl2:MeOH:0.88NH3, 93:7:1). H NMR (CDCl3): δ
1.06-1.28 (m, 1 H), 1.28-1.51 (m, 3 H), 1.51-1.76 (m, 2 H),
1.76-1.94 (m, 3 H), 2.56-2.80 (m, 2 H), 3.07 (d, 1 H), 4.02-
4.10 (m, 2 H), 6.94 (d, 2 H), 7.58 (d, 2 H). LRMS: m/z 231.3
(M + H)+.
N-t-Bu toxyca r bon yl-2(RS)-[2-(4-cya n op h en oxy)eth yl]-
p ip er id in e (24). DEAD (1.73 mL, 11 mmol) was added to a
stirred, ice-cooled solution of alcohol 23 (2.29 g, 10 mmol),
triphenylphosphine (2.88 g, 11 mmol), and 4-cyanophenol (1.31
g, 11 mmol) in tetrahydrofuran (THF; 75 mL), and then, the
cooling bath was removed. After an additional 18 h, the
reaction mixture was evaporated under reduced pressure and
the residue was dissolved in ether. The resulting solution was
washed with aqueous NaOH solution (1 M, ×2) and brine,
dried (Na2SO4), and evaporated under reduced pressure to give
crude product, which was purified by chromatography on silica
gel, using hexanes-Et2O (1:1) as eluant, to furnish the ether
24 (2.86 g, 87%) as a clear oil, which solidified on standing. Rf
N-(2,2,2-Tr ich lor oet h oxyca r b on yl)-(R)-Ch a . Aqueous
NaHCO3 (82 mL, 1 M) was added to an ice cold stirred solution
of H-D-Cha-OH (11.91 g, 41 mmol) dissolved in aqueous NaOH
(41 mL, 1 M). The resulting thick suspension was stirred
vigorously as a solution of N-(2,2,2-trichloroethoxycarbony-
loxy)succinimide (7.02 g, 41 mmol) in dioxan (82 mL) was
added. After 1 h, most of the dioxan was evaporated and the
residue was extracted with EtOAc. The aqueous phase was
then acidified with concentrated HCl and extracted with
EtOAc. The extract was washed with brine and dried (MgSO4).
Evaporation gave an oil (13.79 g), which was dissolved in ether,
and the solution was filtered through a short column of silica
to give Troc-D-Cha-OH (12.3 g, 87%) as a white foam, which
solidified on standing. Rf 0.20 (Et2O:hexane, 6:4). 1H NMR
(CDCl3) (85:15 distribution of rotamers): δ 0.96 (m, 2 H), 1.22
1
0.30 (hexane:Et2O, 1:1). H NMR (CDCl3): δ 1.20-1.80 (m, 6
H), 1.35 (s, 9 H), 1.78-1.94 (m, 1 H), 2.20-2.34 (m, 1 H), 2.80
(t, 1 H), 3.90-4.10 (m, 3 H), 4.44-4.56 (m, 1 H), 6.90 (d, 2 H),
7.58 (d, 2 H). Anal. (C19H26N2O3) C, H, N.
2(RS)-[2-(4-Cya n op h en oxy)eth yl]p ip er id in e (25). TFA
(10 mL) was added to a stirred, ice-cooled solution of BOC-

Design of Selective Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12 2445
(m, 3 H), 1.45 (m, 1 H), 1.55-1.90 (m, 7 H), 4.48 (m, 1 H),
(m, 21 H), 2.24 (m, 2 H), 2.68 (t, 0.15 H), 3.12 (m, 2.7 H), 3.38
(d, 0.15 H), 3.59 (dd, 0.85 H), 3.72 (m, 1.14 H), 4.05 (m, 3.85
H), 4.37 (m, 0.15 H), 4.60 (d, 0.15 H), 5.04 (m, 0.85 H), 6.92
(d, 2 H), 7.55 (dd, 2 H). LRMS: m/z 470 (M + H)+.
N-[N-(Eth oxyca r bon ylm eth yl)-(R)-cycloh exyla la n yl]-
2(R)-[2-(4-cya n op h en oxy)eth yl]p ip er id in e (31). This com-
pound was prepared from 29 in a manner similar to 30. Ester
31 (94%) was obtained a a gum. Rf 0.28 (EtOAc:hexane, 6:4).
1H NMR (CDCl3) (1:1 distribution of rotamers): δ 2.83 (m, 2
H), 1.05-2.11 (mm, 21 H), 2.22 (m, 2 H), 2.63 (t, 0.5 H), 2.74
(d, 0.5 H), 3.05 (d, 0.5 H), 3.11 (t, 0.5 H), 3.20 (d, 0.5 H), 3.36
(d, 0.5 H), 3.62 (m, 1.5 H), 3.85-4.25 (mm, 4.5 H), 4.60 (dd,
0.5 H), 5.04 (q, 0.5 H), 6.90 (d, 1 H), 6.94 (d, 1 H), 7.56 (d, 1
H), 7.60 (d, 1 H). LRMS: m/z 470 (M + H)+.
4.69 (d, 1 H), 4.82 (d, 1 H), 5.32 (d, 0.85 H), 5.58 (d, 0.15 H).
N-[N-(2,2,2-Tr ich lor oeth oxyca r bon yl)-(R)-cycloh exyl-
alan yl]-2(S)-[2-(4-cyan oph en oxy)eth yl]piper idin e (26) an d
N-[N-(2,2,2-Tr ich lor oet h oxyca r b on yl)-(R)-cycloh exyl-
a la n yl]-2(R)-[2-(4-cya n op h en oxy)eth yl]p ip er id in e (27).
Oxalyl chloride (1.5 mL, 17.2 mmol) was added to a stirred
solution of Troc-D-Cha-OH (1.49 g, 4.3 mmol) in CH2Cl2 (20
mL) followed by one drop of DMF. After 1.5 h, the solution
was evaporated, azeotroped with CH2Cl2 (×2), and dried under
high vacuum. The residual oil was dissolved in CH2Cl2 (10 mL)
and stirred with ice-cooling, and the amine 25 (1.0 g, 4.3 mmol)
was added dissolved in CH2Cl2 (10 mL), followed by DIPEA
(1.5 mL, 8.6 mmol). After 1 h, the solvent was evaporated and
the residue was partitioned between EtOAc and H2O. The
organic phase was washed with HCl (2 M), H2O, saturated
aqueous NaHCO3, and brine. Drying over MgSO4 and evapo-
ration gave a foam (2.60 g). The above reaction was repeated
on twice the scale, and the crude products were combined (7.55
g) and chromotographed on silica. Elution with increasing
proportions of Et2O-hexane (1:1 to 6:4) gave initially the S,R
diastereoisomer 26 (3.04 g, 42%) as a foam. Rf 025 (Et2O:
hexane, 6:4). 1H NMR (CDCl3) (90:10 distribution of rotam-
ers): δ 0.60-1.85 (mm, 18 H), 1.92 (m, 2 H), 2.75 (m, 1 H),
2.70 (t, 0.1 H), 3.22 (t, 0.9 H), 3.70 (d, 0.9 H), 3.98 (m, 2.1 H),
4.44 (m, 0.1 H), 4.57 (d, 2 H), 4.68 (m, 1H), 4.79 (m, 1.1 H),
4.97 (m, 0.9 H), 5.63 (d, 0.1 H), 5.71 (d, 0.9 H). LRMS: m/z
559 (M + H)+. Anal. (C26H34Cl3N3O4) C, H, N. Continued
elution gave the R,R diastereoisomer 27 (3.17 g, 44%) as a
foam. Rf 0.20 (Et2O:hexane, 6:4). 1H NMR (CDCl3) (1:1
distribution of rotamers): δ 0.72-1.85 (mm, 8 H), 2.00 (m, 2
H), 2.28 (m, 1 H), 2.70 (t, 0.5 H), 3.17 (t, 0.5 H), 3.68 (d, 0.5
H), 3.88-4.23 (mm, 2.5 H), 4.47 (d, 0.5 H), 4.53 (dd, 0.5 H),
4.62 (d, 0.5 H), 4.79 (m, 2 H), 5.00 (m, 0.5 H), 5.77 (t, 1 H),
6.89 (d, 1 H), 7.57 (d, 2 H). LRMS: m/z 559 (M + H)+. Anal.
(C26H34Cl3N3O4) C, H, N.
N-[(R)-Cycloh exyla la n yl]-2(S)-[2-(4-cya n op h en oxy)-
eth yl]p ip er id in e (28). This compound was prepared from 26
in a similar manner as 29. Amine 28 (71%) was obtained as a
gum. Rf 0.50 (CH2Cl2:MeOH:0.88NH3, 93:7:1). 1H NMR (CDCl3)
(80:20 distribution of rotamers): δ 0.63-2.07 (mm, 22 H), 2.23
(m, 1 H), 2.68 (t, 0.2H), 316 (t, 0.8 H), 3.70 (m, 1.8 H), 4.03
(m, 2 H), 4.37 (m, 0.2 H), 4.58 (d, 0.2 H), 4.98 (m, 0.8 H), 6.91
(d, 2 H), 7.56 (d, 2 H). LRMS: m/z 384 (M + H)+, 767 (2M +
H)+.
N-[(R)-Cycloh exyla la n yl]-2(R)-[2-(4-cya n op h en oxy)-
eth yl]p ip er id in e (29). Zn dust (18 g, 0.28 mol) was added to
a stirred solution of Troc-amine 27 (3.1 g, 55 mmol) in THF
(56 mL) followed by KH2PO4 (18 mL, 1 M). After 1.5 h, the
mixture was filtered and evaporated to a small volume. The
pH was lowered to 2 with HCl (2 M), H2O was added, and the
mixture was extracted with EtOAc (100 mL). The aqueous
phase was separated, and the organic phase was extracted
with H2O (×8). The combined aqueous extracts were basified
to pH 11 with NaOH (2 M) and extracted with CH2Cl2. The
extract was washed with brine and dried over Na2SO4 to give
amine 29 (1.63 g, 77%) as a gum. Rf 0.43 (CH2Cl2:MeOH:
0.88NH3, 93:7:1). 1H NMR (CDCl3) (1:1 distribution of rota-
mers): δ 0.7-2.08 (mm, 22 H), 2.29 (m, 1 H), 2.65 (t, 0.5 H),
3.13 (t, 0.5 H), 3.65 (d, 0.5 H), 3.80 (m, 1 H), 4.00 (m, 2 H),
4.23 (m, 0.5 H), 4.60 (d, 0.5 H), 5.03 (m, 0.5 H), 6.89 (d, 1 H),
6.95 (d, 1 H), 7.57 (t, 2 H). LRMS: m/z 384 (M + H)+.
N-[N-(Eth oxyca r bon ylm eth yl)-(R)-cycloh exyla la n yl]-
2(S)-[2-(4-a m id in op h en oxy)et h yl]p ip er id in e Dih yd r o-
ch lor id e (32). An ice cold solution of nitrile 30 (780 mg, 1.66
mmol) in absolute EtOH (10 mL, dried over 3 Å molecular
sieves) was saturated with HCl gas and left at 0 °C for 18 h.
The solvent was evaporated under vacuum, and the residue
was azeotroped with EtOH to give the intermediate ethyl
imidate, Rf 0.6 (CH2Cl2:MeOH:0.88NH3, 93:7:1). A solution of
NH3 in EtOH (3.4 mL, 1.96 M, 6.6 mmol) was added to the
foregoing imidate, and the resulting suspension was heated
at 50 °C for 2 h. A second portion of ethanolic ammonia (2
mL, 3.9 mM) was added, and heating was continued for a
further 1 h. The solvent was evaporated, and the residue was
partitioned between ether and H2O. The aqueous phase was
then basified with NaOH (1 M) and extracted with CH2Cl2.
The extract was washed with brine, dried over Na2SO4, and
filtered. The filtrate was acidified with etherial HCl and
evaporated to dryness to give amidine 32 (804 mg, 79%) as a
white powder. Rf 0.33 (CH2Cl2:MeOH:0.88NH3, 80:20:5). 1H
NMR (DMSO-d6) (90:10 distribution of rotamers): δ 0.84 (m,
2 H), 1.00 (mm, 20 H), 1.93 (m, 1 H), 2.22 (m, 1 H), 2.79 (t, 0.1
H), 3.20 (m, 0.9 H), 3.29-3.91 (mm, 5 H), 4.11 (m, 4 H), 4.30
(d, 0.1 H), 4.48 (m, 0.9 H), 4.60 (m, 0.1 H), 4.79 (m, 0.9 H),
7.08 (d, 2 H), 7.88 (d, 2 H), 9.15 (s, 2 H), 9.30 (s, 2 H). LRMS:
m/z 487 (M + H)+, 470 (M + H - NH3)+. Anal. (C27H42N4O4‚
2HCl‚3.2H2O) C, N; H: calcd, 8.23; found, 7.78.
N-[N-(Eth oxyca r bon ylm eth yl)-(R)-cycloh exyla la n yl]-
2(R)-[2-(4-a m id in op h en oxy)et h yl]p ip er id in e Dih yd r o-
ch lor id e (33). This compound was prepared from 31 in a
manner similar to 32. Amidine 33 (83%) was obtained as a
white powder. Rf 0.35 (CH2Cl2:MeOH:0.88NH3, 80:20:5). 1H
NMR (DMSO-d6) (6:4 distribution of rotamers): δ 0.83 (m, 2
H), 1.00-2.29 (mm, 22 H), 2.76 (t, 0.4 H), 3.22 (t, 0.6 H), 2.28-
4.25 (mm, 9 H), 4.30 (br s, 0.4 H), 4.34 (br s, 0.4 H), 4.55 (br
s, 0.6 H), 4.90 (br s, 0.6 H), 7.05 (d, 1.2 H), 7.17 (d, 0.8 H),
7.87 (t, 2 H), 9.05 (s, 2 H), 9.27 (d, 2 H). LRMS: m/z 487 (M +
H)+, 470 (M + H - NH3)+. Anal. (C27H42N4O4‚2HCl‚1.0H2O‚0.1-
CH2Cl2) C, H, N.
2(S)-(2-Hydr oxyeth yl)piper idin e (34a). 2(RS)-(2-Hydroxy-
ethyl)piperidine (34b) (110 g) was resolved using (1S)-(+)-10-
camphorsulfonic acid as described31 via the intermediate (S,S)-
10-camphor-sulfonate salt, mp 167 °C (Literature31 166-167
°C); [R]D 25 + 32.5° (c ) 2.2, CHCl3) (Literature32 [R]D + 32.4°
(c ) 2, CHCl3)). The absolute configuration of the salt was
confirmed by single-crystal X-ray crystallographic analysis.
The amine 34a (13.56 g) was obtained as fine needles, mp 69-
70 °C (Literature31 68-69 °C), Rf 0.25 (isobutyl methyl ketone:
AcOH:H2O, 2:1:1, upper phase). GC analysis of the bis-TFA
derivative, employing a Chiraldex B-TA No C70 column,
showed an enantiomeric excess (ee) of >98%.
N -B e n z y lo x y c a r b o n y l-2(S )-(2-h y d r o x y e t h y l)p i p -
er id in e (35a ). To a stirred, ice-cooled solution of amine 34a
(13.4 g, 0.104 mol) in CH2Cl2 (250 mL) were added, sequen-
tially, Et3N (15.9 mL, 0.114 mol) and CbzOSuc (27.21 g, 0.109
mol). The cooling bath was removed, and the reaction mixture
was stirred at 23 °C for 18 h, washed with brine, dried (Na2-
SO4), and evaporated under reduced pressure to give an oil
(30.6 g), which was purified by chromatography on silica gel,
using hexanes-EtOAc (1:1) as eluant, to provide carbamate
35a (27.5 g, 100%) as an oil. Rf 0.48 (hexane:EtOAc, 1:1);
N-[N-(Eth oxyca r bon ylm eth yl)-(R)-cycloh exyla la n yl]-
2(S)-[2-(4-cya n op h en oxy)eth yl]p ip er id in e (30). K2CO3 (1.1
g, 8 mmol) was added to a solution of amine 28 (1.53 g, 4 mmol)
in CH3CN (20 mL), followed by ethyl bromoacetate (0.49 mL,
4.4 mmol). The suspension was stirred at 23 °C for 19 h, and
most of the solvent was evaporated under vacuum. The residue
was partitioned between EtOAc and H2O, and the organic
phase was washed with brine and dried over Na2SO4. Evapo-
ration gave a gum that was chromotographed on silica, with
EtOAc-hexane as eluant (6:4), giving ester 30 (1.79 g, 95%)
1
as a gum. Rf 0.40 (EtOAc:hexane, 6:4). H NMR (CDCl3) (85:
15 distribution of rotamers): δ 0.62-1.05 (m, 2 H), 1.05-2.0

2446 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12
Danilewicz et al.
[R]D25-24.5° (c ) 1.06, MeOH). 1H NMR (CDCl3): δ 1.50-
1.86 (m, 8 H), 1.86-2.05 (m, 1 H), 2.70-2.85 (m, 1 H), 3.14-
3.50 (br s, 2 H), 3.50-3.68 (m, 1 H), 4.06(d, 1 H), 4.40-4.58
(m, 1 H), 5.16 (s, 2 H), 7.20-7.43 (m, 5 H).
mmol) derived from N-t-butoxycarbonyl-4-hydroxypiperidine
(2.42 g, 12.0 mmol) and NaH (0.48 g, 60% dispersion in oil,
12.0 mmol) by the procedure described for 37a gave 37b (2.62
g, 59%) as a colorless oil. Rf 0.30 (hexane:EtOAc, 7:3). 1H NMR
(CDCl3): consistent with that of 37a .
N-Ben zyloxyca r b on yl-2(S)-[2-(4-p ip er id yloxy)et h yl]-
p ip er id in e h yd r och lor id e (38a ). A stirred, ice-cooled solu-
tion of BOC-amine 37a (26.16 g, 58.6 mmol) in CH2Cl2 (300
mL) was saturated with hydrogen chloride gas. After a further
1.25 h, the solvent was removed by evaporation under reduced
pressure and residual HCl was removed azeotropically using
CH2Cl2 (×3) to give amine hydrochloride 38a (22.59 g, 100%)
as a white foam. Rf 0.60 (CH2Cl2:MeOH:0.88NH3, 85:15:2). 1H
NMR (CDCl3): δ 1.28-2.15 (m, 12 H), 2.85 (t, 1 H), 2.95-
3.54 (m, 7 H), 4.06 (d, 1 H), 4.43 (br s, 1 H), 5.10 (dd, 2 H),
7.20-7.40 (m, 5 H), 9.41 (br s, 2 H). LRMS: m/z 347.3 (M +
H)+.
N-Ben zyloxyca r bon yl-2(RS)-[2-(4-p ip er id yloxy)eth yl]-
p ip er id in e Hyd r och lor id e (38b). HCl gas deprotection of
BOC-amine 37b (15.27 g, 34.2 mmol) by the procedure
described for 38a gave 38b (13.4 g, 100%) as a white foam. Rf
0.50 (CH2Cl2:MeOH:0.88NH3, 84:14:2). 1H NMR (CDCl3): con-
sistent with that of 38a . LRMS: m/z 347.0 (M + H)+. Anal.
(C20H30N2O3‚HCl) C, H, N.
N-Ben zyloxyca r b on yl-2(RS)-(2-h yd r oxyet h yl)p ip er i-
d in e (35b). Protection of 2(RS)-(2-hydroxyethyl)piperidine
(34b) (25.0 g, 193 mmol) with CbzOSuc (48.1 g, 193 mmol)
and Et3N (29.6 mL, 212 mmol) by the procedure described for
35a gave Cbz-amine 35b (22.0 g, 43%) as a colorless oil. Rf
1
0.48 (hexane:EtOAc, 1:1). H NMR (CDCl3): consistent with
that of 35a . LRMS: m/z 264.4 (M + H)+. Anal. (C15H21NO3)
C, H, N.
N-Ben zyloxyca r bon yl-2(S)-(2-m eth a n esu lfon yloxyeth -
yl)p ip er id in e (36a ). MsCl (16.1 mL, 0.208 mol) was added
dropwise over 15 min to a stirred, ice-cooled solution of alcohol
35a (27.43 g, 0.104 mol) and Et3N (29 mL, 0.208 mol) in CH2-
Cl2 (350 mL); the temperature of the reaction mixture was
allowed to rise to 17 °C during the addition. After an additional
25 min, the reaction mixture was washed with aqueous citric
acid solution (1 M), water, and saturated aqueous NaHCO3
solution, dried (Na2SO4), and evaporated under reduced pres-
sure to give a light yellow oil (40 g), which was purified by
chromatography on silica gel, using hexanes-EtOAc (1:1) as
eluant, to furnish mesylate 36a (33.5 g, 93%) as a clear oil,
1
which solidified on standing. Rf 0.59 (hexane:EtOAc, 1:1). H
N-Ben zyloxyca r bon yl-2(S)-{2-[N-(N,N′-d i-t-bu toxyca r -
b on yla m id in o)-4-p ip er id yloxy]et h yl}p ip er id in e (39a ).
Et3N (24.5 mL, 0.176 mol) was added to a stirred, ice-cooled
solution of amine hydrochloride 38a (22.59 g, 58.6 mmol) in
CH2Cl2 (280 mL), the mixture was allowed to warm to 23 °C,
and then, N,N′-di-t-butoxycarbonyl-S-methylisothiourea29 (18.7
g, 64.4 mmol) and HgCl2 (15.91 g, 58.6 mmol) were added
sequentially. The reaction mixture was stirred for 18 h, for
an additional 2 h under reflux, and then filtered through
Celite. The filtrate was washed with water, (further filtration
of this two phase mixture was necessary to remove precipitated
material), dried (Na2SO4), and evaporated under reduced
pressure to give crude product, which was purified by chro-
matography on silica gel, using an elution gradient of hex-
anes-EtOAc (7:3 to 1:1), to provide guanidine 39a (30.37 g,
88%) as a sticky foam. Rf 0.20 (hexane:EtOAc, 1:1); [R]D25-
5.7° (c ) 1.1, MeOH). 1H NMR (CDCl3): δ 1.30-1.87 (m, 11
H), 1.45 (s, 18 H), 1.92-2.09 (m, 1 H), 2.85 (t, 1 H), 3.23-3.45
(m, 5 H), 3.68 (br s, 2 H), 4.06 (d, 1 H), 4.42 (br s, 1 H), 5.11
(dd, 2 H), 7.25-7.38 (m, 5 H), 10.09 (s, 1 H). LRMS: m/z 589.4
(M + H)+. Anal. (C31H48N4O7‚0.15EtOAc) C, H, N.
NMR (CDCl3): δ 1.34-1.86 (m, 6 H), 1.86-1.95 (m, 1 H), 2.16-
2.32 (m, 1 H), 2.74-2.95 (m, 4 H), 4.01-4.18 (m, 1 H), 4.06(t,
2 H), 4.41-4.56 (m, 1 H), 5.10 (s, 2 H), 7.22-7.42 (m, 5 H). A
trace of EtOAc was removed from the product azeotropically,
using hexane (×2), before the next step of the reaction
sequence.
N-Ben zyloxyca r b on yl-2(RS)-(2-m et h a n esu lfon yloxy-
eth yl)p ip er id in e (36b). Mesylation of alcohol 35b (21.3 g,
80.9 mmol) with MsCl (12.5 mL, 0.162 mol) and Et3N (22.6
mL, 0.162 mol) by the procedure described for 36a gave 36b
(23.24 g, 84%) as a colorless oil. Rf 0.5 (hexane:EtOAc, 1:1).
1H NMR (CDCl3): consistent with that of 36a . Anal. (C16H23
NO5S‚0.15CH2Cl2) C, H, N.
-
N-t-Bu toxyca r bon yl-4-h yd r oxyp ip er id in e. Boc2O (35.58
g, 0.163 mol) was added to a stirred, ice-cooled solution of
4-hydroxypiperidine (15.0 g, 0.148 mol) in CH2Cl2 (250 mL).
The cooling bath was removed, and then, the reaction mixture
was stirred at 23 °C for 56 h, washed with aqueous citric acid
solution (1 M), dried (MgSO4), and evaporated under reduced
pressure to give a yellowish oil; treatment of which with
hexane (20 mL), followed by chilling, promoted crystallization.
Filtration and washing of the product with cold hexane
afforded N-t-butoxycarbonyl-4-hydroxypiperidine (25.72 g, 86%).
Rf 0.37 (hexane:EtOAc, 1:1). 1H NMR (CDCl3): δ 1.28-1.56
(m, 2 H), 1.42 (s, 9 H), 1.72-1.83 (m, 2 H), 2.25 (br s, 1 H),
2.92-3.08 (m, 2 H), 3.74-3.88 (m, 3 H). Anal. (C10H19NO3) H,
N; C: calcd, 59.68; found, 59.27.
N-Ben zyloxyca r b on yl-2(S)-[2-(N-t-b u t oxyca r b on yl-4-
p ip er id yloxy)eth yl]p ip er id in e (37a ). N-t-Butoxycarbonyl-
4-hydroxypiperidine (19.67 g, 97.5 mmol) was added to a
stirred suspension of NaH (3.9 g, 60% dispersion in oil, 97.5
mmol) in dry DMF (150 mL) under N2. After 2 h, a solution of
mesylate 36a (32.9 g, 96.4 mmol) in dry DMF (100 mL) was
added and the resulting reaction mixture was stirred for 18
h. The bulk of the solvent was removed under reduced
pressure, the residue was diluted with water, and the resulting
oily suspension was extracted with EtOAc (×3). The combined
extracts were washed with brine, dried (Na2SO4), and evapo-
rated under reduced pressure to give an oil, which was purified
by chromatography on silica gel, using hexanes-EtOAc (7:3)
as eluant, to afford ether 37a (26.26 g, 61%) as an oil. Rf 0.35
(hexane:EtOAc, 7:3); [R]D25-12.3° (c ) 1.01, MeOH). 1H NMR
(CDCl3): δ 1.30-1.80 (m, 11 H), 1.39 (s, 9 H), 1.94-2.12 (m,
1 H), 2.86 (t, 1 H), 2.94-3.10 (m, 2 H), 3.20-3.45 (m, 3 H),
3.60-3.77 (m, 2 H), 4.07 (d, 1 H), 4.43 (br s, 1 H), 5.10 (s, 2
H), 7.23-7.40 (m, 5 H). Anal. (C25H38N2O5) C, H, N.
N-Ben zyloxyca r b on yl-2(RS)-{2-[N-(N,N′-d i-t-b u t oxy-
car bon ylam idin o)-4-piper idyloxy]eth yl}piper idin e (39b).
Guanidinylation of 38b (13.09 g, 34.2 mmol) with N,N′-di-
t-butoxycarbonyl-S-methylisothiourea
(10.86
g,
37.4
mmol), HgCl2 (9.29 g, 34.2 mmol), and Et3N (14.3 mL, 0.103
mol) by the procedure described for 39a gave 39b (15.7 g, 78%)
as a sticky foam. Rf 0.35 (hexane:EtOAc, 6:4). 1H NMR
(CDCl3): consistent with that of 39a . LRMS: m/z 589.4 (M +
H)+.
2(S )-{2-[N -(N ,N ′-D i-t -b u t o x y c a r b o n y la m id in o )-4-
p ip er id yloxy]eth yl}p ip er id in e (40a ). A solution of Cbz-
amine 39a (26.82 g, 45.55 mmol) in absolute EtOH (400 mL)
was hydrogenated (60 psi) over 10% palladium on charcoal (5.0
g) at 23 °C for 2.5 h. The resulting mixture was filtered
through a pad of Arbocel, and the filtrate was evaporated
under reduced pressure, azeotroping with EtOAc (×2), to
furnish amine 40a (20.36 g, 94%) as a gum. Rf 0.13 (CH2Cl2:
MeOH:0.88NH3, 93:7:1); [R]D25-1.69° (c ) 1.3, MeOH). 1H
NMR (CDCl3): δ 1.00-1.81 (m, 11 H), 1.44 (s, 18 H), 1.81-
1.94 (m, 2 H), 2.52-2.66 (m, 2 H), 3.05 (d, 1 H), 3.39 (br s, 1
H), 3.44-3.57 (m, 3 H), 3.72 (br s, 1 H), 9.8 (vbr s, 1 H).
LRMS: m/z 455 (M + H)+. Anal. (C23H42N4O5‚0.15EtOAc) C,
H, N.
2(R S )-{2-[N -(N ,N ′-Di-t -b u t oxyca r b on yla m id in o)-4-
p ip er id yloxy]eth yl}p ip er id in e (40b). Hydrogenation of
amine 39b (15.68 g, 26.6 mmol) by the procedure described
for 40a gave 40b (12.0 g, 99%) as a gum. Rf 0.22 (CH2Cl2:
MeOH:0.88NH3, 93:7:1). 1H NMR (CDCl3): consistent with
that of 40a .
N-Ben zyloxyca r bon yl-2(RS)-[2-(N-t-bu toxyca r bon yl-4-
p ip er id yloxy)eth yl]p ip er id in e (37b). Reaction of the mes-
ylate 36b (3.41 g, 10.0 mmol) with the sodium alkoxide (12.0

Design of Selective Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12 2447
2(RS)-[2-(N-t-Bu t oxyca r b on yl-4-p ip er id yloxy)et h yl]-
p ip er id in e (41). To a stirred solution of Cbz-amine 37b (5.39
g, 12.1 mmol) in absolute EtOH (300 mL) was added 10%
palladium on charcoal (5.4 g), followed by 1,4-cyclohexadiene
(11.4 mL, 121.1 mmol). The reaction mixture was stirred for
1 h under an atmosphere of N2 and then filtered through a
pad of Arbocel. The filtrate was evaporated under reduced
pressure to give crude product, which was purified by chro-
matography on silica gel, using CH2Cl2-MeOH-0.88NH3
(90:10:1) as eluent, to give piperidine 41 (2.75 g, 73%) as a
pressure to give the crude product. Purification by chroma-
tography on silica gel, using an elution gradient of CH2Cl2-
MeOH (100:0 to 95:5), gave tert-butyl ester 46 (299 mg, 72%)
as a white foam. Rf 0.55 (CH2Cl2:MeOH:0.88NH3, 90:10:1). 1H
NMR (CDCl3): δ 0.73-1.02 (m, 2 H), 1.05-2.12 (m, 50 H),
3.00-3.86 (m, 12 H), 4.08-4.23 (m, 1 H), 4.51-4.62 (m, 1 H),
4.85-4.96 (m, 1 H), 10.10-10.20 (m, 1H). Anal. (C38H67N5O8)
C, H, N.
N-[N-(R-Flu or en ylm eth oxycar bon yl)-(R)-ph en ylalan yl]-
2(R S )-{2-[N -(N ,N ′-d i-t -b u t o x y -c a r b o n y la m id in o )-4-
p ip er id yloxy]eth yl}p ip er id in e (47). DIPEA (0.53 mL, 3
mmol) was added to an ice cold stirred solution of amine 40b
(681 mg, 1.5 mmol), Fmoc-D-Phe-OH (620 mg, 1.6 mmol), and
PyBroP (769 mg, 1.65 mmol) in CH2Cl2 (5 mL). After 2 h, a
second portion of PyBroP (100 mg, 0.21 mmol) was added and
the reaction was stirred for a further 2 h. The solvent was
evaporated, and the residue was partitioned between EtOAc
and H2O. The organic phase was washed with citric acid (1
M), H2O, saturated aqueous NaHCO3, and brine and then
dried over MgSO4. Evaporation gave a foam, which was
purified by chromotography on silica, using EtOAc-hexane
(1:1) as eluant, to give amide 47 (790 mg, 64%) as a white foam.
Rf 0.3 (EtOAc:hexane, 1:1). 1H NMR (CDCl3): δ 1.30-1.96
(mm, ∼30 H), 2,62 (m, 0.75 H), 3.00 (m, 2 H), 3.36 (m, 5 H),
3.71 (m, 2.25 H), 3.95 (m, 0.25 H), 4.11-4.57 (mm, 3.75 H),
4.88 (m, 1.25 H), 5.71 (m, 0.75 H), 7.14-7.44 (mm, ∼10 H),
7.57 (d, 2 H), 7.77 (d, 2 H), 10.11 (d, 1 H). LRMS: m/z 602 (M
+ H - Fmoc)+. Anal. (C47H61N5O8) C, H, N.
1
brown oil. Rf 0.63 (CH2Cl2:MeOH:0.88NH3, 90:10:1). H NMR
(CDCl3): δ 1.00-1.86 (m, 20 H), 2.05 (br s, 1 H), 2.58-2.65
(m, 2 H), 3.00-3.13 (m, 3 H), 3.38-3.45 (m, 1 H), 3.45-3.58
(t, 2 H), 3.62-3.80 (m, 2 H). LRMS: m/z 313.3 (M + H)+. Anal.
(C17H32N2O3‚0.1CH2Cl2) C, N; H: calcd, 10.11; found, 10.76.
N-[N-(2,2,2-Tr ich lor oeth oxyca r bon yl)-(R)-cycloh exyl-
a la n yl]-2(RS)-[2-(N-t-bu toxy-ca r bon yl-4-p ip er id in yloxy)-
eth yl]p ip er id in e (42). Reaction of amine 41 (1.85 g, 5.91
mmol) with the acid chloride prepared from Troc-D-Cha-OH
(2.05 g, 5.91 mmol) and DIPEA (2.06 mL, 11.82 mmol) by the
procedure described for 26 furnished amide 42 (3.29 g, 87%)
as a yellow gum. Rf 0.73 (CH2Cl2:MeOH:0.88NH3, 90:10:1). 1H
NMR (CDCl3): δ 0.80-2.18 (m, 34 H), 2.62-3.25 (m, 3 H),
3.30-3.57 (m, 3 H), 3.60-3.83 (m, 2 H), 4.45-4.88 (m, 5 H),
5.68-5.92 (m, 1 H). LRMS: m/z 644.1 (M + H)+. Anal.
(C29H48N3O6Cl3‚0.4CH2Cl2) C, H, N.
N-[N-(2,2,2-Tr ich lor oeth oxyca r bon yl)-(R)-cycloh exyl-
a la n yl]-2(RS)-[2-(4-p ip er id in yl-oxy)eth yl]p ip er id in e Hy-
d r och lor id e (43). HCl gas deprotection of BOC-amine 42
(3.23 g, 5.04 mmol) by the procedure described for 38a gave
43 (3.00 g, 100%) as a yellow foam. Rf 0.83 (CH2Cl2:MeOH:
N-[(R)-P h en yla la n yl]-2(RS)-{2-[N-(N,N′-d i-t-bu toxyca r -
b on yla m id in o)-4-p ip er id yloxy]et h yl}p ip er id in e
(48).
Piperidine (0.85 mL, 9 mmol) was added to a solution of Fmoc-
amine 47 (750 mg, 0.91 mmol) in THF (4 mL). After 35 min,
the solvent was evaporated and the residue was purified by
chromatography on silica, using CH2Cl2-MeOH-0.88NH3
(193:7:1) as eluant, to give amine 48 (480 mg, 88%) as a white
foam. Rf 0.5 (CH2Cl2:MeOH:0.88NH3, 93:7:1). 1H NMR
(CDCl3): δ 0.78 (m, 0.25 H), 0.96 (m, 0.25 H), 1.32-1.78 (m,
∼29 H), 1.86 (m, 3 H), 2.45-3.17 (mm, 3 H), 3.17-3.87 (mm,
∼7.5 H), 3.73 (t, 0.6 H), 4.08 (m, 0.6 H), 4.29 (m, 0.15 H), 4.55
(m, 0.5 H), 4.78 (m, 0.15 H), 4.86 (m, 0.5 H), 5.03 (m, 0.15 H),
7.24 (m, 5 H), 10.12 (m, 1 H). LRMS: m/z 602 (M + H)+.
N -[N -(t -Bu t oxyca r b on ylm e t h yl)-(R )-p h e n yla la n yl]-
2(R S )-{2-[N -(N ,N ′-d i-t -b u t o x y -c a r b o n y la m id in o )-4-
p ip er id yloxy]eth yl}p ip er id in e (49). tert-Butyl bromoace-
tate (126 µL, 0.86 mmol) was added to a stirred suspension of
K2CO3 (215 mg, 1.56 mmol) and amine 48 (470 mg, 0.78 mmol)
in CH3CN (5 mL). After 14 h at 23 °C, the solvent was
evaporated and the residue was partitioned between EtOAc
and H2O. The organic phase was washed with brine and dried
over MgSO4. Purification by chromatography on silica, using
EtOAc-hexane (7:3) as eluant, gave ester 49 (480 mg, 86%)
as a white foam. Rf 0.28 (EtOAc:hexane, 7:3). 1H NMR
(CDCl3): δ 0.24 (m, 0.32 H), 0.51 (m, 0.28 H), 1.03-1.97 (mm,
∼38 H), 2.33 (br s, 1 H), 2.52 (q, 0.6 H), 2.73-3.58 (mm, ∼10
H), 3.72 (m, 2 H), 3.90 (m, 1 H), 4.57 (d, 0.4 H), 4.85 (m, 0.6
H), 7.20 (m, 5 H), 10.15 (d, 1 H). LRMS: m/z 716 (M + H)+.
N-[N-(R-F lu or en ylm et h oxyca r b on yl)-O-t-b u t yl-(S)-r-
a sp a r tyl]-2(S)-{2-[N-(N,N′-d i-t-bu toxyca r bon yla m id in o)-
4-p ip er id yloxy]eth yl}p ip er id in e (50a ). DIPEA (25 mL, 143
mmol) was added to a stirred solution of amine 40a (22.9 g,
47.7 mmol), Fmoc-Asp(OtBu)-OH (19.6 g, 47.7 mmol), and
PyBroP (23.4 g, 50.1 mmol) in dry CH2Cl2 (160 mL) under N2
at 0 °C, and the mixture was stirred for 3 h. After it was stirred
for an additional 1 h at 23 °C, the reaction mixture was diluted
with EtOAc (1 L) and washed sequentially with water, aqueous
citric acid (1 M), saturated aqueous NaHCO3 solution, water,
and brine (200 mL each). The organic solution was dried
(MgSO4) and evaporated under reduced pressure to give crude
product (2.87 g), which was purified by chromatography on
silica gel, using hexanes-EtOAc (6:4 to 1:1) as eluant, to
furnish amide 50a (33.0 g, 77%) as a white foam. Rf 0.38
(hexanes:EtOAc, 1:1). 1H NMR (CDCl3): δ 1.47 (2xs, 29 H),
1.57-2.12 (mm, 10 H), 2.48 (m, 0.8 H), 2.70 (m, 1.2 H), 3.15
(m, 0.4 H), 3.23-3.60 (mm, 5.6 H), 3.95 (m, 3 H), 4.13-4.58
1
0.88NH3, 80:20:5). H NMR (CDCl3): δ 0.80-2.19 (m, 25 H),
2.58-3.68 (m, 8 H), 4.42-4.90 (m, 5 H), 5.81-5.88 (m, 1 H),
9.20-9.68 (br s, 2 H). LRMS: m/z 540.1 (M + H)+. Anal.
(C24H40N3O4Cl3‚HCl‚0.4CH2Cl2) C, H, N.
N-[N-(2,2,2-Tr ich lor oeth oxyca r bon yl)-(R)-cycloh exyl-
a la n yl]-2(RS)-{2-[N-(N,N′-d i-t-bu toxyca r bon yla m id in o)-
4-p ip er id yloxy]eth yl}p ip er id in e (44). Guanidinylation of
43 (2.97 g, 5.15 mmol) with N,N′-di-t-butoxycarbonyl-S-me-
thylisothiourea (1.49 g, 5.15 mmol), HgCl2 (3.53 g, 10.29 mmol),
and Et3N (2.15 mL, 15.44 mmol) by the procedure described
for 39a gave 44 (3.76 g, 83%) as a beige foam. Rf 0.63 (CH2-
Cl2:MeOH, 95:5). 1H NMR (CDCl3): δ 0.79-2.18 (m, 43 H),
2.60-3.85 (m, 8 H), 4.42-4.90 (m, 5 H), 5.82-5.92 (m,1 H),
10.02-10.19 (br s, 1 H). Anal. (C35H58N5O8Cl3‚1.1CH2Cl2) C,
H, N.
N -[(R )-Cycloh e xyla la n yl]-2(R S )-{2-[N -(N ,N ′-d i-t -b u -
toxyca r bon yla m id in o)-4-p ip er id yloxy]eth yl}p ip er id in e
(45). Zn dust (1.25 g, 19.02 mmol) was added to a stirred
solution of Troc-amine 44 (745 mg, 0,95 mmol) in AcOH (35
mL), and the resulting heterogeneous solution was stirred at
room temperature for 4 h. The reaction mixture was then
neutralized by the cautious addition of saturated aqueous
NaHCO3 solution. The resulting aqueous solution was ex-
tracted with ethyl acetate (×1), the organic extract was washed
with brine and dried (MgSO4), and the solvent was evaporated
under reduced pressure to give the crude product. Purification
by chromatography on silica gel, using an elution gradient of
CH2Cl2:MeOH (100:0 to 95:5), gave amine 45 (367 mg, 63%).
Rf 0.52 (CH2Cl2:MeOH:0.88NH3, 90:10:1). 1H NMR (CDCl3):
δ 0.75-2.13 (m, 43 H), 3.05-3.65 (m, 5 H), 3.66-3.88 (m, 2
H), 3.90-4.33 (m, 2 H), 4.46-4.90 (m, 2 H). LRMS: m/z 608.7
(M + H)+.
N-[N-(t-Bu toxyca r bon ylm eth yl)-(R)-cycloh exyla la n yl]-
2(R S )-{2-[N -(N ,N ′-d i -t -b u t o x y c a r b o n y la m i d i n o )-4-
p ip er id yloxy]eth yl}p ip er id in e (46). tert-Butyl bromoace-
tate (94 µL, 0.64 mmol) was added to a stirred suspension of
K2CO3 (0.16 g, 1.16 mmol) in a solution of amine 45 (352 mg,
0.58 mmol) in MeCN (5.0 mL). The reaction mixture was
stirred at room temperature under N2 for 30 h. The solvent
was then evaporated under reduced pressure, and the result-
ant semisolid was partitioned between water and EtOAc. The
organic layer was separated, washed with brine, and dried
(MgSO4), and the solvent was evaporated under reduced

2448 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12
Danilewicz et al.
1
(mm, 3.4 H), 4.85 (m, 0.6 H), 5.01 (m, 1 H), 5.65 (d, 0.6 H),
5.85 (d, 0.4 H), 7.30 (t, 2 H), 7.40 (t, 2 H), 7.59 (d, 2 H), 7.76
(d, 2 H), 10.10 (d, 1 H). LRMS: m/z 849.3 (M + H)+, 670, 626.7
(M-Fmoc + 2H)+. Anal. (C46H65N5O10‚0.4EtOAc) C, H, N.
N-[N-(R-F lu or en ylm et h oxyca r b on yl)-O-t-b u t yl-(S)-r-
aspar tyl]-2(RS)-{2-[N-(N,N′-di-t-bu toxycar bon ylam idin o)-
4-p ip er id yloxy]eth yl}p ip er id in e (50b). Coupling of amine
40b (5.00 g, 11.0 mmol) with Fmoc-Asp(OtBu)-OH (4.75 g, 11.5
mmol) in the presence of PyBroP (5.38 g, 11.5 mmol) and
DIPEA (3.83 mL, 22 mmol) by the procedure described for 50a
gave 50b (6.05 g, 65%) as a gum. Rf 0.68 (CH2Cl2:MeOH, 90:
10). 1H NMR (CDCl3): consistent with that of 50a . LRMS: m/z
849 (M + H)+.
N -(O -t -B u t y l-(S )-r-a s p a r t y l)-2(S )-{2-[N -(N ,N ′-d i-t -
b u t oxyca r b on yla m id in o)-4-p ip er id yloxy]et h yl}p ip er i-
d in e (51a ). Piperidine (1.7 mL, 18.3 mmol) was added to a
stirred solution of Fmoc-amine 50a (1.55 g, 1.83 mmol) in THF
(7.5 mL). After 1 h, the reaction mixture was evaporated under
reduced pressure and the residue was purified by chromatog-
raphy on silica gel, using CH2Cl2-MeOH-0.88NH3 (193:7:1)
as eluant, to afford amine 51a (1.05 g, 92%) as a white foam.
Rf 0.50 (CH2Cl2:MeOH:0.88NH3, 93:7:1). 1H NMR (CDCl3):
1.30-2.17 (mm, 41 H), 2.42 (m, 1 H), 2.55 (m, 1.5 H), 3.12 (t,
0.5 H), 3.36 (m, 3 H), 3.47 (m, 2 H), 3.72 (m, 2.5 H), 4.08 (m,
0.5 H), 4.23 (m, 1 H), 4.52 (d, 0.5 H), 4.85 (m, 0.5 H), 10.09 (s,
1 H). LRMS: m/z 626 (M + H)+. Anal. (C31H55N5O8‚0.3H2O)
C, H, N.
-5.7° (c ) 2.45, MeOH). H NMR (CDCl3): δ 1.20-2.05 (mm,
26 H), 1.44 (s, 9 H), 1.49 (2xs, 18 H), 2.26-2.52 (mm, 3 H),
2.68 (t, 0.4 H), 3.10 (t, 0.6 H), 3.30-3.58 (mm, 5 H), 3.66-
4.04 (mm, 4 H), 4.18 (br s, 0.4 H), 4.53 (br d, 0.4 H), 4.88 (br
s, 0.6 H), 10.1 (br d, 1 H). LRMS: m/z 722.9 (M + H)+.
N-(N-Cyclooctyl-O-t-bu tyl-(S)-r-a sp a r tyl)-2(RS)-{2-[N-
(N,N′-di-t-bu toxycar bon yl-am idin o)-4-piper idyloxy]eth yl}-
p ip er id in e (52d ). Alkylation of amine 51b (280 mg, 0.45
mmol) with cyclooctanone (282 mg, 2.24 mmol) in the presence
of NaBH(OAc)3 (474 mg, 2.24 mmol) and AcOH (0.128 mL, 2.24
mmol) by the general procedure gave sec-amine 52d (160 mg,
49%) as an oil. Rf 0.58 (CH2Cl2:MeOH:0.88NH3, 90:10:1). 1H
NMR (CDCl3): δ 1.20-2.08 (mm, 27 H), 1.42 (s, 9 H), 1.45 (s,
18 H), 2.10-2.73 (mm, 3 H), 3.01-4.37 (mm, 10 H), 4.50 (br
d, 0.4 H), 4.88 (br s, 0.6 H), 10.1 (br s, 1 H). LRMS: m/z 737
(M + H)+.
N-(N-3-P en tyl-O-t-bu tyl-(S)-r-aspar tyl)-2(S)-{2-[N-(N,N′-
d i-t -b u t oxyca r b on yl-a m id in o)-4-p ip e r id yloxy]e t h yl}-
p ip er id in e (52e). Alkylation of amine 51a (300 mg, 0.479
mmol) with 3-pentanone (273 mg, 3.16 mmol) in the presence
of NaBH(OAc)3 (456 mg, 2.15 mmol) and AcOH (0.081 mL, 1.44
mmol) by the general procedure gave sec-amine 52e (68 mg,
20%). 1H NMR (CDCl3): δ 0.88 (m, 6 H), 1.2-2.46 (mm, 19
H), 1.43 (s, 9 H), 1.49 (18 H), 2.67 (t, 0.4 H), 3.11 (t, 0.6 H),
3.26-4.28 (mm, 10 H), 4.51 (br d, 0.4 H), 4.86 (br s, 0.6 H),
10.06 (br d, 1 H). LRMS: m/z 696.2 (M + H)+.
N-(N-Cycloh exyl-O-t-b u t yl-(S)-r-a sp a r t yl)-2(S)-{2-[N-
(N,N′-di-t-bu toxycar bon yl-am idin o)-4-piper idyloxy]eth yl}-
p ip er id in e (52f). Alkylation of amine 51a (688 mg, 1.10
mmol) with cyclohexanone (0.137 mL, 1.32 mmol) in the
presence of NaBH(OAc)3 (350 mg, 1.65 mmol) and AcOH (0.069
mL, 1.21 mmol) by the general procedure gave sec-amine 52f
(745 mg, 96%) as a white foam. Rf 0.80 (CH2Cl2:MeOH:
0.88NH3, 93:7:1). 1H NMR (CDCl3): δ 0.94-2.10 (mm, ∼50
H), 2.10-2.52 (mm, 2.6 H), 2.68 (t, 0.4 H), 3.11 (t, 0.6 H), 3.40
(2xm, 5 H), 3.55 (2xm, 2.6 H), 4.03 (t, 1 H), 4.20 (m, 0.4 H),
4.52 (d, 0.4 H), 4.88 (m, 0.6 H), 10.11 (d, 1 H). LRMS: m/z
708 (M + H)+.
N-(N-Cycloh ep t -4-en yl-O-t-b u t yl-(S)-r-a sp a r t yl)-2(S)-
{2-[N-(N,N′-d i-ter t-b u t oxy-ca r b on yla m id in o)-4-p ip er id -
yloxy]eth yl}p ip er id in e (52g). Alkylation of amine 51a (600
mg, 0.959 mmol) with 4-cycloheptenone33 (320 mg, 2.90 mmol)
in the presence of NaBH(OAc)3 (610 mg, 2.90 mmol) and AcOH
(0.165 mL, 2.90 mmol) by the general procedure gave sec-
amine 52g (600 mg, 87%) as a foam. Rf 0.39 (hexanes:EtOAc,
N -(O-t -Bu t yl-(S )-r-a sp a r t yl)-2(R S )-{2-[N -(N ,N ′-d i-t -
b u t oxyca r b on yla m id in o)-4-p ip er id yloxy]et h yl}p ip er i-
d in e (51b). Treatment of Fmoc-amine 51b (6.05 g, 7.10 mmol)
with piperidine by the procedure described for 51a gave 51b
(3.71 g, 84%) as a foam. Rf 0.52 (CH2Cl2:MeOH, 90:10). 1H
NMR (CDCl3): consistent with that of 51a . LRMS: m/z 626
(M + H)+.
Gen er a l P r oced u r e for th e P r ep a r a tion of Am in es 52.
NaBH(OAc)3 (15.0 mmol) was added, in one portion, to a
stirred solution of amine 51 (10.0 mmol), ketone/aldehyde (12.0
mmol), and AcOH (11.0 mmol) in THF (65 mL) under N2 at
23 °C, and the mixture was stirred for 2 h (monitored by TLC).
The reaction mixture was evaporated in vacuo, diluted with
EtOAc (100 mL), washed with saturated aqueous NaHCO3
solution and brine (100 mL each), dried (MgSO4), and evapo-
rated under reduced pressure to leave the crude product.
Purification by chromatography on silica gel, using CH2Cl2-
MeOH or hexanes-EtOAc as eluant, gave amine 52.
N-(N-Cycloh exyl-O-t-bu tyl-(S)-r-a sp a r tyl)-2(RS)-{2-[N-
(N,N′-di-t-bu toxycar bon yl-am idin o)-4-piper idyloxy]eth yl}-
p ip er id in e (52a ). Alkylation of amine 51b (420 mg, 0.67
mmol) with cyclohexanone (73 mg, 0.74 mmol) in the presence
of NaBH(OAc)3 (213 mg, 1.01 mmol) and AcOH (0.038 mL, 0.67
mmol) by the general procedure gave sec-amine 52a (412 mg,
87%) as a white foam. Rf 0.70 (CH2Cl2:MeOH:0.88NH3, 90:10:
1
1:1); [R]D -0.2° (c ) 0.50, MeOH). H NMR (CDCl3): δ 1.20-
2.06 (m, 20 H), 1.40 (s, 9 H), 1.50 (s, 18 H), 2.14-2.66 (m, 4
H), 2.68 (t, 0.4 H), 3.13 (t, 0.6 H), 3.31-3.58 (m, 5 H), 3.75 (br
s, 2 H), 3.86 (d, 0.6 H), 4.04 (t, 1 H), 4.21 (br s, 0.4 H), 4.53 (d,
0.4 H), 4.87 (br s, 0.6 H), 5.72 (s, 2 H), 10.02 (vbr s, 1 H).
LRMS: m/z 720.4 (M + H)+, 578.4, 503.0, 478.3. Anal.
(C38H65N5O8) C, H, N.
1
1). H NMR (CDCl3): δ 0.95-2.04 (mm, 20 H), 1.41 (s, 9 H),
N -[(N -Cycloh e xylm e t h yl)-O-t -b u t yl-(S )-r-a sp a r t yl]-
2(RS)-{2-[N-(N,N′-di-t-bu toxy-car bon ylam idin o)-4-piper id-
yloxy]eth yl}p ip er id in e (52h ). Alkylation of amine 51b (276
mg, 0.441 mmol) with cyclohexanecarboxaldehyde (0.107 mL,
0.88 mmol) in the presence of NaBH(OAc)3 (185 mg, 0.88
mmol) by the general procedure gave cyclohexylmethylamines
52h (137 mg, 43%) as a foam. Rf 0.30 (hexanes:EtOAc, 1:1).
1H NMR (CDCl3): δ 0.78-0.97 (m, 2 H), 1.07-2.07 (m, 23 H),
1.44 (s, 9 H), 1.48 (s, 18 H), 2.12-2.72 (m, 3.4 H), 3.13 (q, 0.6
H), 3.28-3.60 (m, 4.8 H), 3.66-3.97 (m, 3.2 H), 4.22 (br s, 0.4
H), 4.36 (br s, 0.4 H), 4.55 (br s, 0.6 H), 4.87 (br s, 0.6 H),
10.10 (vbr s, 1 H). LRMS: m/z 722.3 (M + H)+, 622.7, 580.6,
480.6.
N -(N -B e n zy l-O -t -b u t y l-(S )-r-a s p a r t y l)-2(S )-{2-[N -
(N,N′-di-t-bu toxycar bon ylam idin o)-4-piper idyloxy]eth yl}-
p ip er id in e (52i). Alkylation of amine 51a (502 mg, 0.802
mmol) with benzaldehyde (0.90 mL, 0.88 mmol) in the presence
of NaBH(OAc)3 (258 mg, 1.22 mmol) by the general procedure
gave benzylamine 52i (394 mg, 66%) as a foam. Rf 0.31
1.45 (s, 18 H), 2.12-2.75 (mm, 4 H), 3.10 (m, 0.6 H), 3.24-
3.90 (mm, 9 H), 4.03-4.31 (mm, 2 H), 4.50 (br d, 0.4 H), 4.87
(br s, 0.6 H), 10.12 (br s, 1 H). LRMS: m/z 708.7 (M + H)+.
N-(N-Cyclop en t yl-O-t-b u t yl-(S)-r-a sp a r t yl)-2(RS)-{2-
[N-(N,N′-d i-t-bu toxyca r bon yl-a m id in o)-4-p ip er id yloxy]-
eth yl}p ip er id in e (52b). Alkylation of amine 51b (200 mg,
0.32 mmol) with cyclopentanone (60 mg, 0.70 mmol) in the
presence of NaBH(OAc)3 (102 mg, 0.48 mmol) and AcOH (0.018
mL, 0.32 mmol) by the general procedure gave sec-amine 52b
(185 mg, 83%) as a gum. Rf 0.56 (CH2Cl2:MeOH:0.88NH3, 90:
1
10:1). H NMR (CDCl3): δ 1.20-2.05 (mm, 21 H), 1.41 (2xs, 9
H), 1.46 (s, 18 H), 2.24-4.37 (mm, 13 H), 4.50 (br d, 0.4 H),
4.88 (br s, 0.6 H), 10.12 (br s, 1 H). LRMS: m/z 694.5 (M +
H)+.
N-(N-Cycloh ep tyl-O-t-bu tyl-(S)-r-a sp a r tyl)-2(S)-{2-[N-
(N,N′-d i-ter t-bu toxyca r bon yl-a m id in o)-4-p ip er id yloxy]-
eth yl}p ip er id in e (52c). Alkylation of amine 51a (700 mg,
1.12 mmol) with cycloheptanone (376 mg, 3.36 mmol) in the
presence of NaBH(OAc)3 (711 mg, 3.36 mmol) and AcOH (0.192
mL, 3.36 mmol) by the general procedure gave sec-amine 52c
(740 mg, 92%). Rf 0.56 (CH2Cl2:MeOH:0.88NH3, 90:10:1); [R]D
1
(hexanes:EtOAc, 1:1); [R]D -11.3° (c ) 0.16, MeOH). H NMR
(CDCl3): δ 1.30-2.02 (m,13 H), 1.44 (s, 9 H), 1.47 (s, 18 H),
2.33-2.54 (m, 2 H), 2.67 (t, 0.4 H), 3.05 (t, 0.6 H), 3.28-3.83

Design of Selective Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12 2449
(m, 9.4 H), 3.93 (t, 0.6 H), 4.05 (dd, 0.4 H), 4.15 (br s, 0.4 H),
4.55 (d, 0.6 H), 4.88 (br s, 0.6 H), 7.15-7.38 (m, 5 H), 10.10
(vbr s, 1 H). LRMS: m/z 716.6 (M + H)+, 616.6, 574.6, 499.6,
322.3, 238.3. Anal. (C38H61N5O8‚0.3CH2Cl2) C, H, N.
yloxy]eth yl}p ip er id in e (53g). Methylation of amine 52g
(346 mg, 0.481 mmol) with 37% aqueous formaldehyde (0.065
mL, 0.80 mmol) in the presence of NaBH(OAc)3 (204 mg, 0.96
mmol) by the general procedure gave tert-amine 53g (309 mg,
87%). Rf 0.66 (CH2Cl2:MeOH, 90:10); [R]D -20.6° (c ) 0.53,
MeOH). 1H NMR (CDCl3): δ 1.30-2.4 (mm, ∼23 H), 1.44 (s, 9
H), 1.50 (s, 18 H), 2.25 (s, 3 H), 2.68 (m, 0.6 H), 2.92 (dd, 0.6
H), 3.05 (t, 0.6 H), 3.3-3.54 (mm, 5 H), 3.74 (m, 2 H), 4.05 (br
d, 1.6 H), 4.7 (br s, 0.6 H), 5.73 (s, 2 H), 10.04 (br s, 1 H).
LRMS: m/z 734 (M + H)+. Anal. (C39H67N5O8‚0.5H2O) C, H,
N.
N-[N-(4-Tetr a h yd r op yr a n yl)-O-t-bu tyl-(S)-r-a sp a r tyl]-
2(R S )-{2-[N -(N ,N ′-d i-t -b u t o x y -c a r b o n y la m id in o )-4-
p ip er id yloxy]eth yl}p ip er id in e (52j). Alkylation of amine
51b (200 mg, 0.32 mmol) with 4-tetrahydropyranone (35 mg,
0.48 mmol) in the presence of NaBH(OAc)3 (102 mg, 0.48
mmol) and AcOH (0.018 mL, 0.32 mmol) by the general
procedure gave sec-amine 52j (180 mg, 79%). Rf 0.60 (CH2Cl2:
MeOH:0.88NH3, 90:10:1). 1H NMR (CDCl3): δ 1.22-2.08 (mm,
16 H), 1.42 (s, 9 H), 1.47 (s, 18 H), 2.1-2.74 (mm, 4 H), 3.03-
4.32 (mm, 14 H), 4.5 (br d, 0.25 H), 4.8 (br s, 0.75 H), 10.1 (br
s, 1 H). LRMS: m/z 710.5 (M + H)+.
N-[N-(3-Tetr a h yd r op yr a n yl)-O-t-bu tyl-(S)-r-a sp a r tyl]-
2 (S )-{2 -[N -(N ,N ′-d i -t -b u t o x y c a r b o n y l a m i d i n o )-4 -
p ip er id yloxy]eth yl}p ip er id in e (52k ). Alkylation of amine
51a (300 mg, 0.479 mmol) with 3-tetrahydropyranone34 (116
mg, 1.15 mmol) in the presence of NaBH(OAc)3 (297 mg, 1.40
mmol) and AcOH (0.027 mL, 0.32 mmol) by the general
procedure gave sec-amine 52k (221 mg, 65%) as a white foam.
Rf 0.79 (CH2Cl2:MeOH, 90:10). 1H NMR (CDCl3): δ 1.20-2.05
(m, 43 H), 2.17-2.75 (m, 3 H), 2.92-3.20 (m, 2 H); 3.22-3.60
(m, 6 H), 3.68-4.08 (m, 5 H), 4.10-4.56 (m, 1 H), 4.82-4.92
(m, 1 H), 10.08 (br s, 1 H). Anal. (C36H63N5O9‚0.2CH2Cl2) C,
H, N.
N-[N-(N-Meth yl-4-p ip er id yl)-O-t-bu tyl-(S)-r-a sp a r tyl]-
2(R S )-{2-[N -(N ,N ′-d i-t -b u t o x y -c a r b o n y la m id in o )-4-
p ip er id yloxy]eth yl}p ip er id in e (52l). Alkylation of amine
51b (500 mg, 0.80 mmol) with N-methyl-4-piperidone (0.11
mL, 0.90 mmol) in the presence of NaBH(OAc)3 (254 mg, 1.12
mmol) and AcOH (96 µL, 1.66 mmol) by the general procedure
gave sec-amine 52l (510 mg, 86%) as a white foam. Rf 0.50
(CH2Cl2:MeOH:0.88NH3, 93:7:2). 1H NMR (CDCl3) δ 1.20-1.98
(m, 45 H), 2.25-2.35 (m, 3 H), 2.50-4.90 (m, 15 H). Anal.
(C37H66N6O8‚0.2CH2Cl2) C, H, N.
Gen er a l P r oced u r e for th e P r ep a r a tion of N-Meth yl
Am in es 53. A solution of amine 52 (1.04 mmol) in CH2Cl2 (15
mL) was vigorously stirred with aqueous formaldehyde (37%,
w/v; 0.34 mL, 4.15 mmol) for 1 h. NaBH(OAc)3 (2.08 mmol)
was added, and stirring was continued for 1 h. The reaction
mixture was diluted with CH2Cl2 (25 mL), washed sequentially
with saturated aqueous NaHCO3 solution (10 mL), dried
(MgSO4), and evaporated under reduced pressure to leave the
crude product. Purification by chromatography on silica gel,
using CH2Cl2-MeOH or hexanes-EtOAc as eluant, gave
amine 53.
N-[N-Meth yl-N-(3-tetr a h yd r op yr a n yl)-O-t-bu tyl-(S)-r-
a sp a r tyl]-2(S)-{2-[N-(N,N′-d i-t-bu toxyca r bon yla m id in o)-
4-piper idyloxy]eth yl}piper idin e (53k). Methylation of amine
52k (208 mg, 0.293 mmol) with 37% aqueous formaldehyde
(0.095 mL, 1.17 mmol) in the presence of NaBH(OAc)3 (124
mg, 0.586 mmol) by the general procedure gave tert-amine 53k
(185 mg, 87%) as a white solid. Rf 0.83 (CH2Cl2:MeOH:
1
0.88NH3, 90:10:1). H NMR (CDCl3): δ 1.22-2.12 (m, 43 H),
2.21-2.23 (m, 3 H), 2.50-2.68 (m, 1 H), 2.90-3.55 (m, 9 H),
3.68-3.90 (m, 4 H), 3.91-4.53 (m,3 H), 4.76-4.88 (m, 1 H),
10.02 (br s, 1 H). LRMS: m/z 723.8 (M + H)+. Anal.
(C37H65N5O9‚0.2CH2Cl2) C, H, N.
N-(N-Cycloh exyl-N-eth yl-O-t-bu tyl-(S)-r-aspar tyl)-2(RS)-
{2-[N-(N,N′-di-t-bu toxy-car bon ylam idin o)-4-piper idyloxy]-
eth yl}p ip er id in e (54). A solution of acetaldehyde in THF (1
M; 0.97 mL, 0.97 mmol) was added to a stirred solution of
amine 52a (328 mg, 0.463 mmol) in THF (7 mL) at 23 °C, and
after 30 min, NaBH(OAc)3 (216 mg, 1.02 mmol) was added.
After 18 h, a second portion of the acetaldehyde solution (1
M; 0.34 mL, 0.34 mmol) and NaBH(OAc)3 (72 mg, 0.34 mmol)
were added, the reaction mixture was stirred for 3 h and then
evaporated under reduced pressure. The residual solid was
partitioned between EtOAc and water, and the organic phase
was separated, washed sequentially with saturated aqueous
NaHCO3 solution and brine, dried (MgSO4), and evaporated
under reduced pressure. The crude product was purified by
chromatography on silica gel, using hexanes-EtOAc (1:1) as
eluant, to provide ethylamine 54 (193 mg, 56%) as a white
foam. Rf 0.87 (EtOAc). 1H NMR (CDCl3): δ 1.04 (t, 3 H), 1.12-
2.18 (m, 49 H), 2.28-3.11 (m, 5 H), 3.26-3.88 (m, 7 H), 4.08-
4.19 (m, 2H), 4.43-4.86 (m, 1 H). LRMS: m/z 736.3 (M + H)+.
Anal. (C39H69N5O8‚0.2CH2Cl2) C, H, N.
N-{N-Cycloh exyl-N-[2-(d im eth yla m in o)eth yl]-O-t-bu t-
yl-(S)-r-a sp a r tyl}-2(RS)-{2-[N-(N,N′-d i-t-bu toxyca r bon yl-
a m id in o)-4-p ip er id yloxy]eth yl}p ip er id in e (55). Solid K2-
CO3 (228 mg, 1.64 mmol) and then 2-(dimethylamino)ethyl
chloride hydrochloride (119 mg, 0.826 mmol) were added to a
solution of amine 52a (328 mg, 0.55 mmol) in MeCN (10 mL)
at 23 °C, and the mixture was heated at 50 °C for 18 h and
then evaporated under reduced pressure. The residual solid
was partitioned between EtOAc and water, and the organic
phase was separated, washed with brine, dried (MgSO4), and
evaporated under reduced pressure. The crude product was
purified by chromatography on silica gel, using an elution
gradient of CH2Cl2-MeOH (100:0 to 95:5), to provide diamine
55 (400 mg, 93%) as an oil. Rf 0.39 (CH2Cl2:MeOH:0.88NH3,
95:5:0.1). 1H NMR (CDCl3): δ 0.95-2.18 (m, 49 H), 2.20-2.75
(m, 10 H), 3.03-4.98 (m, 14 H). LRMS: m/z 779 (M + H)+.
N-(N-Cycloh ep tyl-N-m eth yl-O-t-bu tyl-(S)-r-a sp a r tyl)-
2(S )-{2-[N -(N ,N ′-d i -t -b u t o x y -c a r b o n y la m i d i n o )-4-
p ip er id yloxy]eth yl}p ip er id in e (53c). Methylation of amine
52c (420 mg, 0.58 mmol) with 37% aqueous formaldehyde
(0.065 mL, 0.80 mmol) in the presence of NaBH(OAc)3 (185
mg, 0.87 mmol) by the general procedure gave tert-amine 53c
1
(390 mg, 91%). Rf 0.74 (CH2Cl2:MeOH:0.88NH3, 90:10:1). H
NMR (CDCl3): δ 1.2-2.1 (mm, 24 H), 1.42 (s, 9 H), 1.48 (s, 18
H), 2.20 (s, 3 H), 2.34 (dd, 1 H), 2.65 (br s, 1 H), 2.89 (dd, 1 H),
3.05 (t, 1 H), 3.29-3.56 (mm, 5 H), 3.74 (br s, 2 H), 4.00 (m, 2
H), 4.79 (br s, 1 H), 10.08 (br s, 1 H). LRMS: m/z 736.6 (M +
H)+.
N-[N-(r-F lu or en ylm et h oxyca r b on yl)-O-t-b u t yl-(S)-r-
glu ta m yl]-2(S)-{2-[N-(N,N′-d i-t-bu toxyca r bon yla m id in o)-
4-p ip er id yloxy]eth yl}p ip er id in e (56). Coupling of amine
40a (1.074 g, 2.36 mmol) with Fmoc-Glu(OtBu)-OH (1.012 g,
2.38 mmol) in the presence of PyBroP (1.171 g, 2.51 mmol)
and DIPEA (1.23 mL, 7.06 mmol) by the procedure described
for 50a gave 56 (1.368 g, 67%) as a white foam. Rf 0.47
N-(N-Cycloh exyl-N-m et h yl-O-t-b u t yl-(S)-r-a sp a r t yl)-
2(S)-{2-[N-(N,N′-d i-t-bu toxy-ca r bon yla m id in o)-4-p ip er id -
yloxy]eth yl}p ip er id in e (53f). Methylation of amine 52f (735
mg, 1.04 mmol) with 37% aqueous formaldehyde (0.34 mL, 4.15
mmol) in the presence of NaBH(OAc)3 (440 mg, 2.08 mmol)
by the general procedure gave tert-amine 53f (720 mg, 96%)
as a white foam. Rf 0.32 (CH2Cl2:MeOH:0.88NH3, 193:7:1). 1H
NMR (CDCl3): δ 1.00-1.94 (mm, ∼50 H), 2.02 (m, 1 H), 2.19
(s, 3 H), 2.31 (m, 2 H), 2.64 (m, 0.2 H), 2.97 (m, 1.8 H), 3.37
(m, 4 H), 3.48 (m, 1 H), 3.73 (m, 2 H), 4.08 (m, 2 H), 4.30 (m,
0.2 H), 4.48 (m, 0.2 H), 4.80 (m, 1 H), 10.00 (s, 1 H). LRMS:
m/z 722 (M + H)+. Anal. (C38H67N5O8‚0.5H2O) C, H, N.
N-(N-Cycloh ept-4-en yl-N-m eth yl-O-t-bu tyl-(S)-r-aspar t-
yl)-2(S)-{2-[N-(N,N′-di-t-butoxycarbonylamidino)-4-piperid-
1
(hexanes:EtOAc, 1:1); [R]D -8.6° (c ) 0.78, MeOH). H NMR
(CDCl3): δ 1.30-2.14 (m, 15 H), 1.43 (s, 18 H), 1.50 (s, 9 H),
2.18-2.42 (m, 2 H), 2.68 (t, 0.5 H), 3.15 (t, 0.5 H), 3.24-3.55
(m, 4 H), 3.63-3.88 (m, 2 H), 4.15-4.54 (m, 4 H), 4.73 (br s, 1
H), 4.87 (br s, 1 H), 5.75 (br s, 1 H), 7.30 (t, 2 H), 7.39 (t, 2 H),
7.60 (d, 2 H), 7.75 (d, 2 H), 10.02 (br s, 1 H). LRMS: m/z 864,
862 (M + H)+, 721, 640 (M-Fmoc + 2H)+. Anal. (C47H67N5O10
0.2CH2Cl2) C, H, N.
‚

2450 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12
Danilewicz et al.
N-(O-t-Bu tyl-(S)-r-glu ta m yl)-2(S)-{2-[N-(N,N′-d i-t-bu t-
oxyca r b on yla m id in o)-4-p ip er id yloxy]et h yl}p ip er id in e
(57). Treatment of Fmoc-amine 56 (1.325 g, 1.54 mmol) with
piperidine by the procedure described for 51a gave 57 (0.90 g,
91%) as a white foam. Rf 0.34 (CH2Cl2:MeOH, 90:10); [R]D
2. Amidine 3 (96%) was obtained as a white powder. Rf 0.1
(CH2Cl2:MeOH:0.88NH3, 80:20:5). 1H NMR (DMSO-d6) (6:4
distribution of rotamers): δ 0.8 (m, 2 H), 0.94-1.83 (mm, ∼19
H), 1.99 (m, 0.8 H), 2.15 (m, 1.2 H), 2.74 (t, 0.4 H), 3.12-3.84
(mm, ∼3.6 H), 3.96 (m, 1.4 H), 4.13 (m,1 H), 4.30 (m, 1 H),
4.50 (m, 0.6 H), 4.90 (m, 0.6 H), 7.04 (d, 1.2 H), 7.16 (m, 0.8
H), 7.86 (t, 2 H), 9.10 (s, 2 H), 9.25 (s, 2 H). LRMS: m/z 459
(M + H)+, 442 (M + H - NH3)+. Anal. (C25H38N4O4‚2HCl‚
0.4i-PrOH) C, H, N.
Gen er a l P r oced u r e of th e P r ep a r a tion of Ta r gets 4,
5, 7-22. A stirred, ice-cooled solution of the tert-butyl ester
(0.5 mmol) in CH2Cl2 (10 mL) was saturated with HCl gas,
the cooling bath was removed, and the resulting solution was
stirred at 23 °C until total deprotection was complete (typically
2-6 h at 23 °C). The solvent was removed by evaporation
under reduced pressure, and the residual HCl was removed
azeotropically using CH2Cl2 (×3) to give the amine di(tri)-
hydrochloride.
N-[N-Ca r b oxym et h yl-(R)-cycloh exyla la n yl]-2(RS)-[2-
(N-a m id in o-4-p ip er id yloxy)et h yl]p ip er id in e Dih yd r o-
ch lor id e (4). HCl gas deprotection of tert-butyl ester 46 (286
mg, 0.40 mmol) by the general procedure gave 4 (209 mg, 90%)
as a white solid. Rf 0.15, 0.12 (consistent with two diastere-
oisomers) (CH2Cl2:MeOH:0.88NH3, 80:20:5). 1H NMR (DMSO-
d6): δ 0.78-2.00 (m, 25 H), 2.95-4.06 (m, integral obscured
by HOD), 4.12-4.85 (m, 3 H), 7,52 (s, 4 H), 8.80-9.58 (br s, 2
H). LRMS: m/z 466.5 (M + H)+. Anal. (C24H43N5O4‚2HCl‚
0.45CH2Cl2‚0.5H2O) C, H, N.
N-[N-Ca r b oxym et h yl-(R)-p h en yla la n yl]-2(RS)-[2-(N-
a m id in o-4-p ip er id yloxy)eth yl]p ip er id in e Dih yd r och lo-
r id e (5). An ice cold solution of ester 49 (470 mg, 0.66 mmol)
in CH2Cl2 (15 mL) was saturated with HCl gas over 0.5 h. The
solution was then left at 23 °C for 2 h, and the solvent was
evaporated initially with a stream of N2 and then under
vacuum. The residue was azeotroped with CH2Cl2 to give 5
(336 mg, 89%) as a white powder. Rf 0.48 (MeOH:EtOAc:AcOH:
0.88NH3:H2O, 60:12:4:4:8). 1H NMR (DMSO-d6): δ 0.18 (m,
0.25 H), 0.14 (m, 0.15 H), 0.92-1.90 (mm, ∼12 H), 2.21 (m,
0.25 H), 2.92 (m, 1 H), 3.10-3.68 (m, ∼10 H), 2.81 (m, 2 H),
4.27 (m, 0.25 H), 4.64 (mm, 1.5 H), 7.27 (m, 5 H), 7.54 (m, 4
H). LRMS: m/z 460 (M + H)+. Anal. (C24H37N5O4‚2HCl‚
1.6H2O‚0.25CH2Cl2) C, H, N.
N-[(R)-P h en yla la n yl]-2(RS)-[2-(4-a m id in op h en oxy)eth -
yl]p ip er id in e Dih yd r och lor id e (6). This compound was
prepared from 60 in a manner similar to 32. The crude reaction
product was chromotographed on silica. Elution with mixtures
of CH2Cl2-MeOH-0.88NH3 (92.5:7.5:1 to 85:15:2) gave the
Cbz derivative of compound 6 (30%). Rf 0.47 (CH2Cl2:MeOH:
0.88NH3, 80:20:5). LRMS: m/z 520 (M + H)+, followed by 6
(21%). Rf 0.3 (CH2Cl2:MeOH:0.88NH3, 80:20:5). LRMS: m/z
395 (M + H)+. Compound 6 was treated with etherial HCl to
give the hydrochloride as a white powder. 1H NMR (DMSO-
d6): δ 0.11 (m, 0.5 H), 1.00-1.71 (mm, 5 H), 1.83 (m, 1 H),
2.12 (m, 1 H), 2.62 (m, 0.5 H), 3.00 (mm, 2.5 H), 3.45 (m, 1 H),
3.90 (m, 0.5 H), 4.04 (m, 2 H), 4.57 (m, 1 H), 4.72 (m, 1 H),
7.08 (m, 2 H), 7.23 (m, 5 H), 7.81 (m, 2 H), 8.42 (br s, 3 H),
9.04 (d, 2 H), 9.25 (d, 2 H). Anal. (C23H30N4O2‚2HCl‚1.4H2O‚
0.2Et2O‚0.2 EtOAc) C, H, N.
N-(N-Cycloh exyl-(S)-r-a sp a r tyl)-2(RS)-[2-(N-a m id in o-
4-piper idyloxy)eth yl]piper idin e dih ydr och lor ide (7). HCl
gas deprotection of tert-butyl ester 52a (280 mg, 0.39 mmol)
by the general procedure gave 7 (170 mg) as a white foam. Rf
0.57 (MeOH:EtOAc:AcOH:0.88NH3:H2O, 60:12:4:4:8). 1H NMR
(DMSO-d6): δ 0.95-2.10 (mm, 20 H), 2.62-3.82 (mm, 13 H),
3.92-4.36 (mm, 1 H), 4.57-4.83 (m, 2 H), 7.50 (s, 4 H), 8.75
(vbr s, 1 H), 9.35 (vbr s, 1 H), 12.9 (vbr s, 1 H). LRMS: m/z
452 (M + H)+. Anal. (C23H41N5O4‚2HCl‚CH2Cl2) C, H, N.
1
+11.4° (c ) 0.71, MeOH). H NMR (CDCl3): δ 1.30-2.10 (m,
16 H), 1.42 (s, 9 H), 1.50 (s, 18 H), 2.25-2.68 (m, 2.5 H), 3.10
(t, 0.5 H), 3.25-3.55 (m, 5 H), 3.65-3.88 (m, 3.5 H), 4.21 (br
s, 0.5 H), 4.52 (d, 0.5 H), 4.86 (br s, 0.5 H), 10.0 (vbr s, 1 H).
LRMS: m/z 640 (M + H)+. Anal. (C32H57N5O8‚0.1CH2Cl2) C,
H, N.
N -(N -Cycloh e xyl-O-t -b u t yl-(S )-r-glu t a m yl)-2(S )-{2-
[N-(N,N′-d i-t-bu toxyca r bon yl-a m id in o)-4-p ip er id yloxy]-
eth yl}p ip er id in e (58). Alkylation of amine 57 (853 mg, 1.33
mmol) with cyclohexanone (0.414 mL, 3.99 mmol) in the
presence of NaBH(OAc)3 (861 mg, 3.99 mmol) and AcOH (0.233
mL, 3.99 mmol) by the general procedure gave amine 58 (847
mg, 88%) as a white foam. Rf 0.32 (hexanes:EtOAc, 1:1); [R]D
-6.3° (c ) 0.79, MeOH). 1H NMR (CDCl3): δ 0.94-2.02 (m,
23 H), 1.42 (s, 9 H), 1.46 (s, 18 H), 2.08-2.73 (m, 4 H), 3.10 (t,
0.5 H), 3.28-3.63 (m, 7 H), 3.66-3.88 (m, 3 H), 4.22 (br s, 0.5
H), 4.53 (d, 0.5 H), 4.87 (br s, 0.5 H), 10.0 (vbr s, 1 H). LRMS:
m/z 723 (M + H)+. Anal. (C38H67N5O8) H, N; C: calcd, 63.22;
found, 62.75.
N-(N-Cycloh exyl-N-m et h yl-O-t-b u t yl-(S)-r-glu t a m yl)-
2(S)-{2-[N-(N,N′-d i-t-bu toxy-ca r bon yla m id in o)-4-p ip er id -
yloxy]eth yl}p ip er id in e (59). Methylation of amine 58 (598
mg, 0.828 mmol) with 37% aqueous formaldehyde (0.12 mL,
1.46 mmol) in the presence of NaBH(OAc)3 (0.350 mg, 1.66
mmol) by the general procedure gave amine 59 (0.572 mg, 94%)
as a white foam. Rf 0.42 (hexanes:EtOAc, 1:1); [R]D +0.5° (c )
1
0.65, MeOH). H NMR (CDCl3): δ 0.95-2.15 (m, 22 H), 1.42
(s, 9 H), 1.46 (s, 18 H), 2.20-2.53 (m, 3 H), 2.28 (s, 3 H), 2.62-
(t, 0.2 H), 3.00 (t, 0.8 H), 3.30-3.82 (m, 10 H), 3.99 (d, 0.8 H),
4.23 (br s, 0.2 H), 4.51 (d, 0.2 H), 4.86 (br s, 0.8 H), 10.0 (br s,
1 H). LRMS: m/z 735 (M + H)+. Anal. (C39H69N5O8) C, H, N.
N-[N-(Ben zyloxyca r bon yl)-(R)-p h en yla la n yl]-2(RS)-[2-
(4-cya n op h en oxy)eth yl]p ip er id in e (60). WSCDI (1.53 g, 8
mmol) was added to an ice cold stirred solution of amine 25
(921 mg, 4 mmol), Cbz-D-Phe-OH (1.32 g, 4 mmol), HOBt (540
mg, 4 mmol), and NMM (809 mg, 8 mmol) in CH2Cl2 (20 mL).
The mixture was allowed to warm to 23 °C, and after 18 h,
the solvent was evaporated. The residue was partitioned
between EtOAc and H2O, and the organic phase was washed
with HCl (2 M), brine, saturated aqueous NaHCO3, and brine,
and then dried over MgSO4. Evaoporation gave a foam, which
was purified by chromatography on silica, using Et2O-hexane
(2:1) as eluant, to give amide 60 (910 mg, 44%) as a white solid.
Rf 0.38 and 0.46 (diastereoisomers) (Et2O:hexane, 4:1). 1H
NMR (CDCl3): δ 0.6 (m, 0.6 H), 1.10-1.75 (mm, 6 H), 1.75-
2.35 (mm, 2 H), 2.60 (m, 0.4 H), 2.97 (m, 2.6 H), 3.56 (m, 0.6
H), 3.90 (m, 2.4 H), 4.50 (m, 0.4 H), 4.77-5.14 (m, 4 H), 5.50
(4xd, 1 H), 6.89 (m, 2 H), 7.20 (m, 10 H), 7.48 (m, 2 H). Anal.
(C31H33N3O4‚0.2Et2O) C, H, N.
N-[N-Ca r boxym eth yl-(R)-cycloh exyla la n yl]-2(S)-[2-(4-
a m id in op h en oxy)eth yl]p ip er id in e Dih yd r och lor id e (2).
NaOH (8 mL, 1 M) was added to a suspension of the ester 32
(740 mg, 1.3 mmol) in dioxan (8 mL). After 35 min, the
resulting solution was acidified to pH 2 with HCl (1 M) and
evaporated to dryness, and the residue was dried azeotropi-
cally with i-PrOH. The residue was then extracted with hot
i-PrOH, the suspension was filtered, and the filtrate was
evaporated to dryness. The residue on trituration with Et2O
gave 2 (630 mg, 86%) as a white powder. Rf 0.15 (CH2Cl2:
MeOH:0.88NH3, 80:20:5). 1H NMR (DMSO-d6) (75:25 rotamer
ratio): δ 0.65-1.84 (mm, ∼20 H), 1.94 (m, 0.75 H), 2.21 (m,
0.75 H), 2.79 (t, 0.25 H), 3.12-3.89 (mm, ∼6 H), 4.07 (m, 2
H), 4.33 (m, 0.25 H), 4.45 (m, 0.75 H), 4.55 (m, 0.25 H), 4.79
(m, 0.75 H), 7.08 (dd, 2 H), 7.85 (dd, 2 H), 9.06 (s, 2 H), 9.26
(s, 2 H). LRMS: m/z 459 (M + H)+, 442 (M + H - NH3)+. Anal.
(C25H38N4O4‚2HCl‚0.3H2O‚0.41i-PrOH) C, H, N.
N-(N-Cyclop en tyl-(S)-r-a sp a r tyl)-2(RS)-[2-(N-a m id in o-
4-piper idyloxy)eth yl]piper idin e Dih ydr och lor ide (8). HCl
gas deprotection of tert-butyl ester 52b (143 mg, 0.21 mmol)
by the general procedure gave 8 (95 mg) as a white foam. Rf
0.44 (MeOH:EtOAc:AcOH: 0.88NH3:H2O, 60:12:4:4:8). 1H
NMR (DMSO-d6): δ 1.20-2.04 (mm, 16 H), 2.64-3.80 (mm,
15 H), 3.94-4.86 (mm, 3 H), 7.43 (s, 4 H), 8.92 (vbr s, 1 H),
N-[N-Ca r boxym eth yl-(R)-cycloh exyla la n yl]-2(R)-[2-(4-
a m id in op h en oxy)eth yl]p ip er id in e Dih yd r och lor id e (3).
This compound was prepared from 33 in a manner similar to

Design of Selective Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12 2451
9.34 (vbr s, 1 H), 12.8 (vbr s, 1 H). LRMS: m/z 438 (M + H)+.
Anal. (C22H39N5O4‚2HCl‚1.5H2O‚0.5CH2Cl2) C, H, N.
(m, 2 H), 7.4 (s, 4 H), 7.5 (s, 5 H), 9.3 (br s, 1 H), 9.98 (br s, 1
H), 12.9 (br s, 1 H). LRMS: m/z 460 (M + H)+, 418, 255, 213,
108. Anal. (C24H37N5O4‚2HCl‚0.5H2O‚0.66CH2Cl2) C, H, N.
N-(N-Cycloh exyl-N-et h yl-(S)-r-a sp a r t yl)-2(RS)-[2-(N-
a m id in o-4-p ip er id yloxy)eth yl]p ip er id in e Dih yd r och lo-
r id e (17). HCl gas deprotection of tert-butyl ester 54 (186 mg,
0.253 mmol) by the general procedure gave 17 (156 mg) as a
white foam. Rf 0.54 (MeOH:EtOAc:AcOH:0.88NH3:H2O, 60:12:
4:4:8). 1H NMR (DMSO-d6): δ 0.85-2.20 (m, 25 H), 2.58-4.40
(m, integral obscured by solvent), 4.62-4.93 (m, 2 H), 7.35-
7.63 (m, 4 H), 9.30-9.90 (br m, 1 H). LRMS: m/z 480.7 (M +
H)+. Anal. (C25H45N5O4‚2HCl‚0.1H2O‚1.0CH2Cl2) C, H, N.
N-{N-Cycloh exyl-N-[2-(d im eth yla m in o)eth yl]-(S)-r-a s-
p a r tyl}-2(RS)-[2-(N-a m id in o-4-p ip er id yloxy)eth yl]p ip er i-
d in e Tr ih yd r och lor id e (18). HCl gas deprotection of tert-
butyl ester 55 (390 mg, 0.501 mmol) by the general procedure
gave 18 (0.31 mg) as a sticky white foam. Rf 0.15 (MeOH:
EtOAc:AcOH:0.88NH3:H2O, 60:12:4:4:8). 1H NMR (DMSO-
d6): δ 0.98-2.20 (m, 24 H), 2.60-4.82 (m, integral obscured
by HOD), 7.85-8.20 (m, 4 H). Anal. (C27H50N6O4‚3HCl‚
1.0H2O‚0.7CH2Cl2) C, H, N.
N-(N-Cycloh ep tyl-(S)-r-a sp a r tyl)-2(S)-[2-(N-a m id in o-4-
p ip er id yloxy)eth yl]p ip er id in e Dih yd r och lor id e (9). HCl
gas deprotection of tert-butyl ester 52c (320 mg, 0.44 mmol)
by the general procedure gave 9 (200 mg) as a white powder.
Rf 0.50 (MeOH:EtOAc:AcOH:0.88NH3:H2O, 60:12:4:4:8); [R]578
+6.3° (c ) 0.8, MeOH). 1H NMR (DMSO-d6): δ 1.19-2.15 (mm,
22 H), 2.64-3.05 (mm, 3 H), 3.05-3.8 (mm, 10 H), 3.92-4.07
(mm, 3 H), 7.50 (s, 4 H), 8.6 (vbr s, 1 H), 9.2 (vbr s, 1 H), 13.0
(vbr s, 1 H). LRMS: m/z 466.5 (M + H)+. Anal. (C24H43N5O4‚
2HCl‚1.1H2O‚0.6CH2Cl2) C, H, N.
N-(N-Cyclooct yl-(S)-r-a sp a r t yl)-2(RS)-[2-(N-a m id in o-
4-p ip er id yloxy)et h yl]p ip er id in e Dih yd r och lor id e (10).
HCl gas deprotection of tert-butyl ester 52d (160 mg, 0.22
mmol) by the general procedure gave 10 (89 mg) as a white
powder. Rf 0.58 (MeOH:EtOAc:AcOH:0.88NH3:H2O, 60:12:4:
1
4:8). H NMR (DMSO-d6): δ 1.1-2.06 (mm, 24 H), 2.61-3.8
(mm, 13 H), 3.92-4.93 (mm, 3 H), 7.48 (s, 4 H), 8.48-9.25
(br, 2 H), 13.0 (vbr s, 1 H). LRMS: m/z 480 (M + H)+. Anal.
(C25H45N5O4‚2HCl‚2.0H2O‚0.25CH2Cl2) C, H, N.
N-(N-3-P en tyl-(S)-r-a sp a r tyl)-2(S)-[2-(N-a m id in o-4-p i-
p er id yloxy)eth yl]p ip er id in e Dih yd r och lor id e (11). HCl
gas deprotection of tert-butyl ester 52e by the general proce-
N-[N-(4-Tetr a h yd r op yr a n yl)-(S)-r-a sp a r tyl]-2(R,S)-[2-
(N-a m id in o-4-p ip er id yloxy)et h yl]p ip er id in e Dih yd r o-
ch lor id e (19). HCl gas deprotection of tert-butyl ester 52j (180
mg, 0.253 mmol) by the general procedure gave 19 (125 mg)
as a white foam. Rf 0.48 (MeOH:EtOAc:AcOH:0.88NH3:H2O,
60:12:4:4:8). 1H NMR (DMSO-d6): δ 1.19-2.05 (mm, 16 H),
2.68-4.37 (mm, 16 H), 4.58-4.82 (m, 2 H), 7.37 (s, 4 H), 8.92
(vbr s, 1 H), 9.25 (vbr s, 1 H), CO2H not detected. LRMS: m/z
454 (M + H)+. Anal. (C22H39N5O5‚2HCl‚0.5H2O‚1.0CH2Cl2) C,
H, N.
N-[N-Meth yl-N-(3-tetr a h yd r op yr a n yl)-(S)-r-a sp a r tyl]-
2(S)-[2-(N-a m id in o-4-p ip er id yl-oxy)eth yl]p ip er id in e Di-
h yd r och lor id e (20). HCl gas deprotection of tert-butyl ester
53k (165 mg, 0.228 mmol) by the general procedure gave 20
(124 mg) as an off-white solid. Rf 0.52 (MeOH:EtOAc:AcOH:
0.88NH3:H2O, 60:12:4:4:8). 1H NMR (DMSO-d6): δ 1.08-2.18
(m, 21 H), 2.68 (br s, 3 H), 3.70-4.94 (m, integral obscured by
HOD), 7.38-7.59 (m, 5 H). Anal. (C23H41N5O5‚2HCl‚3.1H2O‚
0.6CH2Cl2) C, H, N.
N-[N-(N-Met h yl-4-p ip er id yl)-(S)-r-a sp a r t yl]-2(RS)-[2-
(N-a m id in o-4-p ip er id yloxy)eth yl]p ip er id in e Tr ih yd r o-
ch lor id e (21). HCl gas deprotection of tert-butyl ester 52l (230
mg, 0.32 mmol) by the general procedure gave 21 (130 mg) as
a white foam. Rf 0.13 (MeOH:EtOAc:AcOH:0.88NH3:H2O, 60:
12:4:4:8). 1H NMR (DMSO-d6): δ1.20-2.32 (m, 17 H), 2.63 (s,
3 H), 2.64-4.35 (m, integral obscured by HOD), 4.58-4.83 (m,
1 H). LRMS: m/z 467.5 (M + H)+. Anal. (C23H42N6O4‚3HCl‚
1.4H2O‚0.6CH2Cl2) C, H, N.
1
dure gave 11 as a white foam. H NMR (DMSO-d6): δ 0.82-
0.99 (m, 6 H), 1.20-2.08 (mm, 16 H), 2.6-4.34 (mm, 12 H),
4.59-4.80 (m, 2 H), 7.50 (s, 4 H), 8.38 (vbr s, 1 H), 9.10 (vbr s,
1 H), CO2H not detected. LRMS: m/z 460.6 (M + H)+. Anal.
(C22H41N5O4‚2HCl‚3.0H2O‚0.5CH2Cl2) C, H, N.
N-(N-Cycloh exyl-N-m et h yl-(S)-r-a sp a r t yl)-2(S)-[2-(N-
a m id in o-4-p ip er id yloxy)eth yl]p ip er id in e Dih yd r och lo-
r id e (12). HCl gas deprotection of tert-butyl ester 53f (690
mg, 0.95 mmol) by the general procedure gave 12 (523 mg) as
a white powder. Rf 0.50 (MeOH:EtOAc:AcOH:0.88NH3:H2O,
60:12:4:4:8). 1H NMR (DMSO-d6): δ 1.00-2.30 (mm, 22 H),
2.55-4.40 (mm, ∼15 H), 4.80 (m, 2 H), 7.52 (d, 4 H), 9.50-
10.40 (br m, 1 H), 13.00 (vbr s, 1 H). LRMS: m/z 466 (M +
H)+. Anal. (C24H43N5O4‚2HCl‚1.3H2O‚0.8CH2Cl2) C, H, N.
N-(N-Cycloh ep tyl-N-m eth yl-(S)-r-a sp a r tyl)-2(S)-[2-(N-
a m id in o-4-p ip er id yloxy)eth yl]p ip er id in e Dih yd r och lo-
r id e (13). HCl gas deprotection of tert-butyl ester 53c (310
mg, 0.42 mmol) by the general procedure gave 13 (117 mg) as
a white solid. Rf 0.52 (MeOH:EtOAc:AcOH:0.88NH3:H2O, 60:
1
12:4:4:8). H NMR (DMSO-d6): δ 1.20-2.05 (mm, 22 H), 2.68
(s, 3 H), 2.74-4.4 (mm, 12 H), 4.64-4.94 (m, 2 H), 7.42 (s, 4
H), 9.43 (vbr s, 1 H), 13.02 (vbr s, 1 H). LRMS: m/z 480.2 (M
+ H)+. Anal. (C25H45N5O4‚2HCl‚2.0H2O‚0.25CH2Cl2) C, H, N.
N-(N-Cycloh ep t-4-en yl-N-m eth yl-(S)-r-a sp a r tyl)-2(S)-
[2-(N-a m id in o-4-p ip er id yloxy)eth yl]p ip er id in e Dih yd r o-
ch lor id e (14). HCl gas deprotection of tert-butyl ester 53g
(281 mg, 0.38 mmol) by the general procedure gave 14 (231
mg) as a white solid. Rf 0.54 (MeOH:EtOAc:AcOH:0.88NH3:
H2O, 60:12:4:4:8); [R]D -19.1° (c ) 0.79, H2O). 1H NMR
(DMSO-d6): δ 1.2-2.4 (mm, 16 H), 2.68 (s, 3 H), 2.7-4.35 (mm,
16 H), 4.74-4.94 (m, 2 H), 5.75 (s, 2 H), 7.6 (s, 4 H), 9.5-10.1
(vbr s, 1 H), 13.0 (vbr s, 1 H). LRMS: m/z 478 (M + H)+. Anal.
(C25H43N5O4‚2HCl‚0.5H2O‚1.0CH2Cl2) C, H, N.
N-(N-Cycloh exyl-N-m eth yl-(S)-r-glu ta m yl)-2(S)-[2-(N-
a m id in o-4-p ip er id yloxy)eth yl]p ip er id in e Dih yd r och lo-
r id e (22). HCl gas deprotection of tert-butyl ester 59 (538 mg,
0.730 mmol) by the general procedure gave 22 (412 mg) as a
white foam. Rf 0.67 (MeOH:EtOAc:AcOH:0.88 NH3:H2O, 60:
1
12:4:4:8); [R]D -12.7° (c ) 0.55, H2O). H NMR (DMSO-d6): δ
1.0-2.5 (m, 23 H), 2.5-2.95 (m, 1 H), 2.66 (s, 3 H), 3.02-3.68
(m, 11.5 H), 3.75-4.9 (m, 2.5 H), 7.45 (s, 4 H), 9.75-9.05 (m,
1 H), 12.3 (br s, 1 H). LRMS: m/z 480.2 (M + H)+, 438.2 (M-
CH3N2 + 2H)+. Anal. (C25H45N5O4‚2HCl‚1.9H2O‚0.35CH2Cl2)
C, H, N.
N-(N-Cycloh exylm eth yl-(S)-r-aspar tyl)-2(RS)-[2-(N-am i-
d in o-4-p ip er id yloxy)et h yl]p ip er id in e Dih yd r och lor id e
(15). HCl gas deprotection of tert-butyl ester 52h (132 mg,
0.183 mmol) by the general procedure gave 15 (100 mg) as a
white foam. Rf 0.56 (MeOH:EtOAc:AcOH:0.88NH3:H2O, 60:12:
4:4:8). 1H NMR (DMSO-d6): δ 0.8-2.0 (mm, 23 H), 2.6-3.8
(mm, 13 H), 3.96-4.48 (mm, 2 H), 7.52 (s, 4 H), 8.68 (vbr s, 1
H), 9.48 (vbr s, 1 H), 13.0 (vbr s, 1 H). LRMS: m/z 466 (M +
H)+, 424, 335, 255. Anal. (C24H43N5O4‚2HCl‚1.5H2O‚0.4CH2-
Cl2) C, H, N.
Biology
Deter m in a tion of In h ibitor P oten cy a n d Selec-
tivity. The inhibition of thrombin, trypsin, FXa, FVIIa,
and plasmin were measured spectrophotometrically in
96 well plates using chromogenic substrates. All assays
were carried out in a total reaction volume of 200 µL.
Compound dilutions in water were preincubated with
enzyme at room temperature for 15 min prior to
addition of chromogenic substrate. After 30 min incuba-
tion at 30 °C, the optical density was measured at 405
nm in a Thermomax plate reader (Molecular Devices,
N-(N-Ben zyl-(S)-r-a sp a r tyl)-2(S)-[2-(N-a m id in o-4-p ip -
er id yloxy)eth yl]p ip er id in e Dih yd r och lor id e (16). HCl gas
deprotection of tert-butyl ester 52i (173 mg, 0.24 mmol) by the
general procedure gave 16 (107 mg) as a white solid. Rf 0.55
(MeOH:EtOAc:AcOH: 0.88NH3:H2O, 60:12:4:4:8); [R]D -5.3°
1
(c ) 0.15, MeOH). H NMR (DMSO-d6): δ 1.14-1.94 (mm, 12
H), 2.62-3.70 (mm, 10 H), 3.88-4.37 (mm, 3 H), 4.55-4.80

2452 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12
Danilewicz et al.
Inc). The percentage inhibition and IC50 were calculated
from triplicate assays of an eight concentration response
curves. From the substrate Km (previously determined
by standard methods) and the IC50, the Ki for each
inhibitor was calculated from the formula Ki ) IC50/
{(1 + S/Km)} according to the method of Cheng and
Prusoff.35
in the cotralateral saphenous vein in conscious dogs and
from the vena cava in rats anaesthetized by halothane.
Urine samples from dog (0-7 h) were collected via a
urethral bladder catheter and from rat (0-24 h) by
housing in metabolism cages. Plasma and urine drug
concentrations were quantified utilizing LCMS (Perkin-
Elmer Sciex API III+ mass spectrometer equipped with
an articulated IonSpray interface). Pharmacokinetic
parameters were derived by computational analysis
using the programs PIVKIN and POPKIN (Dr. B. A.
Wood, Sandwich) for intravenous and oral data, respec-
tively.
Dr u g Assa y. Plasma or urine (1 mL) was fortified
with compound 8 (100 ng) as internal standard. Cali-
bration standards were prepared covering the range
0.005-1.0 µg 12/mL. Drug and internal standards were
extracted from plasma or urine by mixing with pH 6.0
potassium phosphate buffer (1 mL) and percolation
through an activated Certify Bond Elut cartridge [3 mL
cartridge, prerinsed with methanol (2 mL) and pH 6.0
potassium phosphate buffer (2 mL)]. Following washing
with methanol (2 mL) and then water (2 mL) and
aspiration to dryness, the cartridge was eluted with 5%
methanolic ammonia (3 mL) and the residue was
evaporated to dryness. Samples were reconstituted in
50:50 methanol/water, vortexed, centrifuged (13 000
rpm), and submitted for mass spectral analysis using
LCMS.
Enzyme and chromogenic substrate concentration and
supplier for each assay were as follows: thrombin
(human or bovine plasma; Sigma) at final concentra-
tions of 0.04 and 0.08 U/mL, respectively; thrombin
substratesS2238 (H-D-Phe-Pip-Arg-pNA, Quadratech,
U.K.), final concentration 0.1 mM; trypsin (bovine
pancreas; Sigma), final concentration 0.5 U/mL; trypsin
substratesS2222 (Benz-Ile-Glu-Gly-Arg-pNA, Kabi,
Quadratech), final concentration 0.1 mM; plasmin (bo-
vine plasma; Boehringer Mannheim, U.K.), final con-
centration 0.01 U/mL; plasmin substrateschromozyme
PL (tosyl-Gly-Pro-Lys-pNA; Boehringer Mannheim),
final concentration 0.2 mM; FXa (bovine plasma; Boe-
hringer Mannheim), final concentration 0.02 U/mL; FXa
substratesS2222, final concentration 0.2 mM; FVIIa
(human plasma; Diagnostica Stago, Shield Diagnostics,
U.K.), final concentration 0.06 µg/mL; FVIIa substrates
chromozyme t-PA (N-methylsulfonyl-D-Phe-Pro-Arg-
pNA; Boehringer Mannheim). Recombinant tissue factor
(American Diagnostica, Alpha Labs, U.K.) was added
to the FVIIa assay at a final concentration of 0.12 µg/
mL. The thrombin, trypsin, and plasmin assays were
performed in 50 mM HEPES and 150 mM NaCl buffer
(pH 8.0) and at pH 7.5 for the FXa assay. For the FVIIa
assay, 50 mM TRIS and 100 mM NaCl buffer (pH 7.5)
was used.
Log D7.4 Deter m in a tion s. These were carried out
using the method developed by Stopher and McClean.36
Concentrations of drug in the organic and aqueous
phases were determined by HPLC using a Kromasil
Silica column (25 cm × 4.6 mm id), with ultraviolet
detection at 210 nm. A mixture of ammonium phospate
buffer (pH 3.0, 0.05 M) and acetonitrile (3:1, v/v) was
used as the mobile phase with a flow rate of 1 mL/min.
Measured Ki values for human thrombin and human
trypsin (pancreas; Calbiochem, U.K.) were determined
using the same substrates/buffers as described above.
Kinetic assays were performed over 15 min, after the
addition of substrate (at four different concentrations)
to start the reaction. Ki was determined from a plot of
1/V against 1/S.
Ack n ow led gm en t. We thank our following col-
leagues for their expert and enthusiastic technical
assistance: Yvonne Ailwood, David Hardstone, Gwen
Easter, Kirsty McCracken, Drew Gibson, Alan McInnes,
Brian McRitchie, Leena Patel, Michael Pitcher, and
Anthony Pomeroy. We thank Dr. J . Bordner (Pfizer
Global Research and Development, Groton, U.S.A.) and
his staff for determining the crystal structure of 34b,
(1S)-10-camphorsulfonate, by X-ray defraction. We also
thank members of the Department of Physical Sciences
for spectroscopic services and elemental analyses.
Deter m in a tion of Ex Vivo An ticoa gu la n t Activ-
ity. Anaesthetized male Sprague-Dawley rats were
given compound either by bolus iv injection dissolved
in saline or by delivery directly idd dissolved in water.
Appropriate vehicle controls were performed. Serial
arterial blood samples were collected into one-tenth
volume of 3.2% trisodium citrate and centrifuged, and
plasma was collected and stored frozen until assay.
Thrombin clotting times were determined using an
Automated Coagulation Laboratory (ACL-300; Instru-
mentation Laboratories [IL], U.K.) and the IL Test TT
reagent at a thrombin concentration of 6 U/mL. The
change in clotting time induced by the inhibitor was
recorded as a multiple of the predose TT.
Refer en ces
(1) For reviews on thrombin inhibitors, see (a) Lyle, T. A. Small-
molecule Inhibitors of Thrombin. Perspect. Drug Discovery Des.
1993, 1, 453-460. (b) Das, J .; Kimball, S. D.; Thrombin Active
Site Inhibitors. Bioorg. Med. Chem. 1995, 3, 999-1007. (c) Vacca,
J . P. Thrombosis and Coagulation. Ann. Rep. Med. Chem. 1998,
33, 81-90. (d) Sanderson, P. E. J . Small, Noncovalent Serine
Protease Inhibitors. Med. Res. Rev. 1999, 19, 179-197. (e)
Menear, K. Direct Thrombin Inhibitors: Current Status and
Future Prospects. Exp. Opin. Invest. Drugs 1999, 8, 1373-1384.
(f) Fevig, J . M.; Wexler, R. R. Anticoagulants: Thrombin and
Factor Xa Inhibitors. Ann. Rep. Med. Chem. 1999, 34, 81-100.
(g) Vacca, J . P. New Advances in the Discovery of Thrombin and
Factor Xa Inhibitors. Curr. Opin. Chem. Biol. 2000, 4, 394-400.
(2) Adang, A. E. P.; Rewinkel, J . B. M. A New Generation of Orally
Active Antithrombotics: Comparing Strategies in the GPIIb/IIIa,
Thrombin and Factor Xa Areas. Drugs Future 2000, 25, 369-
383.
P h a r m a cok in et ics. Male rats (Sprague-Dawley,
n ) 3 per sampling time) and dogs (Beagle, n ) 2) were
used to determine in vivo pharmacokinetic parameters.
Drugs were formulated in saline for iv administration
(1 mL/kg) by rapid bolus via the caudal vein in rat and
by constant rate infusion (over 15 min) via the saphe-
nous vein in dog. Drug solutions in water were admin-
istered orally by gavage to rat (10 mL/kg) and dog
(1 mL/kg). Plasma was prepared from blood collected
into lithium heparin tubes from an indwelling cannula
(3) (a) Bush, L. R. Argatroban, A Selective, Potent Thrombin
Inhibitor. Cardiovasc. Drug Rev. 1991, 9, 247-263. (b) J eske,
W.; Walenga, J . M.; Lewis, B. E.; Fareed, J . Pharmacology of
Argatroban. Exp. Opin. Invest. Drugs 1999, 8, 625-654.

Design of Selective Thrombin Inhibitors
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 12 2453
(4) Markwardt, F. Pharmacological Approaches to Thrombin Regu-
lation. Ann. N. Y. Acad. Sci. 1986, 485, 204-214.
(18) Bode, W.; Mayr, I.; Baumann, U.; Huber, R.; Stone, S. R.;
Hofsteenge, J . The Refined 1.9 Å Crystal Structure of Human
R-Thrombin: Interaction with D-Phe-Pro-Arg Chloromethyl
ketone and Significance of the Tyr-Pro-Pro-Trp Insertion Seg-
ment. EMBO J . 1989, 8, 3467-3475.
(19) Bernstein, F. C.; Koetzle, T. F.; Williams, G. J . B.; Meyer, E. F.,
J r.; Brice, M. D.; Rodgers, J . R.; Kennard, O.; Shimanouchi, T.;
Tasumi, M. The Protein Data Bank: A Computer Based Archival
File for Macromolecular Structures. J . Mol. Biol. 1977, 112,
535-542.
(20) Brooks, B. R.; Bruccoleri, R. E.; Olafson, B. D.; States, D. J .;
Swaminathan, S.; Karplus, M. CHARMM: A Program for
Macromolecular Energy, Minimization, and Dynamics Calcula-
tions J . Comput. Chem. 1982, 4, 187-217.
(21) Lu, T.; Tomczuk, B.; Illig, C. R.; Bone, R.; Murphy, L.; Spurlino,
J .; Salemme, F. R.; Soll, R. M. In Vitro Evaluation and Crystal-
lographic Analysis of a New Class of Selective, Nonamide-based
Thrombin Inhibitors. Bioorg. Med. Chem. Lett. 1998, 8, 1595-
1600.
(22) Gustke, R. F.; McCormick, P.; Ruppin, H.; Soergel, K. H.;
Whalen, G. E.; Wood, C. M. Human Intestinal Potential Differ-
ence: Recording Method and Biophysical Implications. J . Phys-
iol. 1981, 321, 571-582.
(23) Kikumoto, R.; Tamao, Y.; Ohkubo, K.; Tezuka, T.; Tonomura,
S.; Okomoto, S.; Hijicata, A. Thrombin Inhibitors. 3. Carboxyl-
Containing Amide Derivatives of NR-Substituted L-Arginine. J .
Med. Chem. 1980, 23, 1293-1299.
(24) Kikumoto, R.; Tamao, Y.; Tezuka, T.; Tonomura, S.; Hara, H.;
Ninomiya, K.; Hijikata, A.; Okomoto, S. Selective Inhibition of
Thrombin by (2R,4R)-4-Methyl-1-[N2-[(3-methyl-1,2,3,4-tetra-
hydro-8-quinolinyl)sulfonyl]-L-arginyl]-2-piperidinecarboxylic Acid.
Biochemistry 1984, 23, 85-90.
(5) Brandstetter, H.; Turk, D.; Hoeffken, H. W.; Grosse, D.; Stu¨r-
zebecher, J .; Martin, P. D.; Edwards, B. F. P.; Bode, W. Refined
2.3 Å X-ray Crystal Structure of Bovine Thrombin Complexes
Formed with the Benzamidine and Arginine-based Trombin
Inhibitors NAPAP, 4-TAPAP and MQPA. J . Mol. Biol. 1992, 226,
1085-1099.
(6) Turk, D.; Stu¨rzebecher, J .; Bode, W.; Geometry of Binding of
NR-Tosylated Piperidides of m-Amidino-, p-Amidino-, and p-
Guanidino Phenylalanine to Thrombin and Trypsin. Fed. Eur.
Biochem. Soc. 1991, 287, 133-138.
(7) Bode, W.; Turk, D.; Stu¨rzebecher, J . Geometry of Binding of the
Benzamidine- and Arginine-Based Inhibitors NR-(2-Naphthyl-
sulphonyl-glycyl)-DL-p-amidinophenyl-piperidine (NAPAP) and
(2R,4R)-4-methyl-1-[NR-(3-Methyl-1,2,3,4-tetrahydro-8-quino-
linesulphonyl)-L-argininyl]-2-piperidine carboxylic acid (MQPA)
to Human R-Thrombin. Eur. J . Biochem. 1990, 193, 175-182.
(8) Hilpert, K.; Ackermann, J .; Banner, D. W.; Gast, A.; Gubernator,
K.; Hadva´ry, P.; Labler, L.; Muller, K.; Schmid, G.; Tschopp, T.
B.; van de Waterbeemd, H. Design and Synthesis of Potent and
Highly Selective Thrombin Inhibitors. J . Med. Chem. 1994, 37,
3889-3901.
(9) Hauptmann, J .; Kaiser, B.; Paintz, M.; Markwardt, F. Pharma-
cological Characterization of a New Structural Varient of 4-Ami-
nophenylalanine Amide-type Synthetic Thrombin Inhibitor.
Pharmazie 1989, 44, 282-284.
(10) Kettner, C.; Mersinger, L.; Knabb, R. The Selective Inhibition
of Thrombin by Peptides of Boroarginines. J . Biol. Chem. 1990,
265, 18 289-18 297.
(11) Wityak, J .; Earl, R. A.; Abelman, M. M.; Bethel, Y. B.; Fisher,
B. N.; Kauffman, G. S.; Kettner, C. A.; Ma, P.; McMillan, J . L.;
Mersinger, L. J .; Pesti, J .; Pierce, M. E.; Rankin, F. W.; Chorvat,
R. J .; Confalone, P. N. Synthesis of Inhibitor DuP 714. J . Org.
Chem. 1995, 60, 3717-3722.
(12) Bajusz, S.; Szell, E.; Bagdy, D.; Barabas, E.; Horvath, G.;
Dioszegi, M.; Fittler, Z.; Szabo, G.; J uhasz, A.; Tomori, E.;
Szilagyi, G. Highly Active and Selective Anticoagulants: D-Phe-
Pro-Arg-H, a free Tripeptide Aldehyde Prone to Spontanious
Inactivation, and Its Stable N-Methyl Derivative, D.-MePhe-Pro-
Arg-H. J . Med. Chem. 1990, 33, 1729-1735.
(13) (a) Danilewicz, J . C.; Ellis, D.; Kobylecki, R. J . Preparation of
N-[N-(Ar ylsu lfon yl)-a m idin oph en yla la n yl]h ydr opyr idin e-
carboxylic acids and analogues as Antithrombotics. World Patent
95/13274. (b) Allen, M.; Abel, S. M.; Barber, C. G.; Cussans, N.
J .; Danilewicz, J . C.; Ellis, D.; Hawkeswood, E.; Herron, M.;
Holland, S.; Fox, D. N. A.; J ames, K.; Kobylecki, R. J .; Overing-
ton, J . P.; Pandit, J .; Palmer, H.; Powling, M.; Rance, D. J .;
Taylor, W.; Shepperson, N. S. The Discovery of UK-156,406s
An orally active thrombin inhibitor. Book of Abstracts, 215th
ACS National Meeting, March 29-April 2, 1998, Dallas, TX;
American Chemical Society: Washington, DC, 1998; Abst.
MEDI200. In summary, UK-156,406 is a potent inhibitor of
thrombin (Ki ) 0.46 nM) and shows selectivity against related
serine proteases trypsin (26 nM) and plasmin (3400 nM).
Pharmacokinetic studies in dogs shows UK-156,406 to have a
half-life of 0.8 h, volume of distribution of 0.9 L/Kg, clearance of
11 mL/min/kg, and oral bioavailability of 45%. For a synopsis of
clinical pharmacokinetic and toleration data, see refs 1d,e.
(14) Brown, A. B.; Danilewicz, J . D.; Fish, P. V. Antithrombotic
N-Amidinopiperidine and Benzamidine Derivatives. World Patent
97/16444.
(15) (a) Teger-Nilsson, A.-C. E.; Bylund, R. E. World Patent 93/11152.
(b) Teger-Nilsson, A.-C.; Gyzander, E.; Andersson, S.; Englund,
M.; Mattsson, C.; Ulvinger, J .-C.; Gustafsson, D. In Vitro
properties of Inogatran, a New Low Molecular Weight Inhibitor
of Thrombin. Abstracts: XVth Congress of the International
Society on Thrombosis and Haemostasis, J erusalem, Israel, J une
11-16, 1995. Thromb. Haemostasis 1995, 73, 1633. (c) Preville,
P.; He, J . X.; Tarazi, M.; Siddiqui, M. A.; Cody, W. L.; Doherty,
A. M. An Efficient Preparation of the Potent and Selective
Pseudopeptide Thrombin Inhibitor, Inogatran. Bioorg. Med.
Chem. Lett. 1997, 7, 1563-1566.
(25) QUANTA 4.1; Molecular Simulations, Inc., 555 Oakmead Park-
way, Sunnyvale, CA.
(26) Freedman, J .; Vaal, M. J .; Huber, E. W. The Mitsunobu Reaction
of Some Indan Amino Alcohols. J . Org. Chem. 1991, 56, 670-
672.
(27) J ust, G.; Grozinger, K. A Selective, Mild Cleavage of Trichloro-
ethyl Esters, Carbamates and Carbonates to Carboxylic acids,
Amines, and Phenols using Zinc/THF/pH 4.2-7.2 Buffer. Syn-
thesis 1976, 457-458.
(28) Gautier, J .-A.; Miocque, M.; Farnoux, C. C. Preparation and
Synthetic Uses of Amidines. In The Chemistry of Functional
Groups. The Chemistry of Amidines and Imidates; Patai, S., Ed.;
J ohn Wiley and Sons: London, 1975; pp 292-296.
(29) Bergeron, R. J .; McManis, J . S. Total Synthesis of (()-15-
Deoxyspergualin. J . Org. Chem. 1987, 52, 1700-1703.
(30) Fre´rot, E.; Coste, J .; Pantaloni, A.; Dufour, M.-N.; J ouin, P.
PyBOP and PyBroP: Two Reagents for the difficult Coupling of
the R,R-dialkyl Amino Acid, Aib. Tetrahedron 1991, 47, 259-
270.
(31) Beyerman, H. C.; Van Den Bosch, S.; Breuker, J . H.; Maat, L.
Absolute Configuration and Optical Rotatory Dispersion of
2-Ethylpiperidine and Some of Its Derivatives in Connection
with a Sector Rule. Recueil 1971, 90, 755-764.
(32) Toy, M. S.; Price, C. C. D- and L-Polyconidine. J . Am. Chem. Soc.
1960, 82, 2613-2616.
(33) Marshall, J . A.; Royce, R. D., J r. Synthetic Studies on Cembra-
nolides. Stereoselective Synthesis of a Crassin Acetate Synthon.
J . Org. Chem. 1982, 47, 693-698.
(34) Zweifel, G.; Plamondon, J . Hydroboration of Dihydropyrans and
Dihydrofurans. J . Org. Chem. 1970, 47, 898-902.
(35) Cheng, Y.-C.; Prusoff, W. H. Relationship Between the Inhibition
Constant (Ki) and the Concentration of the Inhibitor that Causes
50 Percent Inhibition (IC50) of an Enzymatic Reaction. Biochem.
Pharmacol. 1973, 22, 3099-3108.
(36) Stopher, D.; McClean, S. An Improved Method for the Deter-
mination of Distribution Coefficients. J . Pharm. Pharmacol.
1990, 42, 144.
(16) Kimble, S. D.; Das, J .; Lau, W. F. European Patent 0601459.
(17) Markwardt, F.; Walsmann, P.; Kazmirowski, H. G. Untersu-
chungen u¨ber den Einfluâ von Ring-Substitutionen auf die
Thrombinhemmende Wirkung von Benzylamin- und Benzamid-
inderivaten. Pharmazie 1969, 24, 400-402.
J M011133D