β-D-Glucosyl Conjugates of Highly Potent Inhibitors of Blood Coagulation Factor Xa Bearing 2-Chorothiophene as a P1 Motif

Article abstract of DOI:10.1002/cmdc.201200224

We synthesized a novel O-glucoside of the recently reported potent factor Xa (fXa) inhibitor 1, which bears a 5-chlorothien-2-yl moiety and 1-isopropylpiperidine as fragments that bind the S1 and S4 enzyme pockets, respectively. A β-D-glucosyl unit was conjugated through an ether-linked C3-alkyl spacer to the central phenyl ring of 1. The synthesized β-D-glucose-based compound 16 achieved picomolar inhibitory potency against human fXa (K_i=60pM) and high selectivity over thrombin and other serine proteases. In addition to the chlorothienyl S1 binder, a large gain in ΔG resulted from the addition of protonated 1-isopropylpiperidine (ΔΔG=29.7-30.5kJmol^-1), which should bind to the aromatic S4 pocket through efficient cation-π and CH{dot operator}{dot operator}{dot operator}π interactions. Instead, the C3-alkyl-linked glucose fragment, which is likely directed toward the solvent outside the enzyme binding site, improves ΔG by an average of 2.9-3.8kJmol^-1. Compound 16 showed sub-micromolar invitro anticoagulant activity, as assessed by prothrombin time (PT) and activated thromboplastin time (aPTT) clotting assays in pooled human plasma (PT₂ and aPTT₂ equal to 0.135 and 0.389μM, respectively). Although compound 16 was 1.4-fold less active than parent compound 1 in the exvivo anticoagulant assay in mice, it showed a significant (1.6-fold) prolongation of PT relative to controls (P<0.05) 60min after oral dosing (75mgkg^-1).

Full text of DOI:10.1002/cmdc.201200224

MED
DOI: 10.1002/cmdc.201200224
b-d-Glucosyl Conjugates of Highly Potent Inhibitors of
Blood Coagulation Factor Xa Bearing 2-Chorothiophene as
a P1 Motif
Gianfranco Lopopolo,[a] Modesto de Candia,[a] Luigi Panza,[b] Maria Rosaria Romano,[c]
Marcello Diego Lograno,[c] Francesco Campagna,[a] and Cosimo Altomare*[a]
We synthesized a novel O-glucoside of the recently reported
potent factor Xa (fXa) inhibitor 1, which bears a 5-chlorothien-
2-yl moiety and 1-isopropylpiperidine as fragments that bind
the S1 and S4 enzyme pockets, respectively. A b-d-glucosyl
unit was conjugated through an ether-linked C3-alkyl spacer to
the central phenyl ring of 1. The synthesized b-d-glucose-
based compound 16 achieved picomolar inhibitory potency
against human fXa (Ki =60 pm) and high selectivity over
thrombin and other serine proteases. In addition to the chloro-
thienyl S1 binder, a large gain in DG resulted from the addition
through efficient cation–p and CꢀH···p interactions. Instead,
the C3-alkyl-linked glucose fragment, which is likely directed
toward the solvent outside the enzyme binding site, improves
DG by an average of 2.9–3.8 kJmolꢀ1. Compound 16 showed
sub-micromolar in vitro anticoagulant activity, as assessed by
prothrombin time (PT) and activated thromboplastin time
(aPTT) clotting assays in pooled human plasma (PT2 and aPTT2
equal to 0.135 and 0.389 mm, respectively). Although com-
pound 16 was 1.4-fold less active than parent compound 1 in
the ex vivo anticoagulant assay in mice, it showed a significant
(1.6-fold) prolongation of PT relative to controls (P<0.05)
60 min after oral dosing (75 mgkgꢀ1).
of
protonated
1-isopropylpiperidine
(DDG=29.7–
30.5 kJmolꢀ1), which should bind to the aromatic S4 pocket
Introduction
The serine protease factor Xa (fXa), located at the confluence
of the intrinsic and extrinsic pathways of the blood coagula-
tion cascade, catalyzes the cleavage of prothrombin to form
thrombin, thus initiating the common pathway that leads to
the formation of fibrin clots. Among the newly developed anti-
coagulant drugs with better safety profiles than those currently
used, thus allowing fixed oral dosing and not requiring time-
and resource-intensive monitoring, fXa inhibitors have
emerged as attractive options for venous thromboembolism
(VTE) treatment and stroke prevention in patients with atrial fi-
brillation (AF).[1] As fXa inhibitors should prevent the genera-
tion of new thrombin without affecting the basal thrombin
level that ensures primary hemostasis,[2] they should have
a lower risk of bleeding than current antithrombotic therapy
(heparins and warfarin),[3] and an even higher therapeutic ratio
than direct thrombin inhibitors (DTIs)[4] such as dabigatran
etexilate.[5]
ing techniques, and 3D QSAR studies.[8] FXa contains a serine
protease domain in a trypsin-like closed b-barrel fold, encom-
passing the catalytic triad Ser195-His57-Asp102 and two prox-
imal binding pockets, namely S1 and S4. Potent fXa inhibitors
are usually L-shaped, with a preorganized scaffold orienting
two nearly orthogonal molecular fragments toward the proxi-
mal S1 and S4 pockets in the enzyme binding site. Due to the
high shape retention of the active site and the well-defined
subpockets, fXa is a good model for molecular recognition
studies.[9,10]
We recently reported N-(2-hydroxyphenyl)acetamide-based
inhibitors of fXa bearing 5-chlorothien-2-yl and 1-isopropylpi-
peridin-4-yl moieties as P1 and P4 fragments, respectively.[11]
The most potent compound 1 (Figure 1) inhibited fXa with a Ki
value of 0.3 nm and very high selectivity over thrombin and
other serine proteases, achieving in vitro anticoagulant activity
in the low micromolar range, as assessed by the prothrombin
time clotting assay (PT2 =3.30 mm). Based on these data, com-
Intensive efforts are underway toward the pharmaceutical
development of direct fXa inhibitors as new antithrombotic
agents, as evidenced by the growing number of publications
and patents filed in the last decade.[6] Large clinical trial pro-
grams have established the effectiveness of fXa inhibitors, such
as apixaban and rivaroxaban, in patients experiencing (or at
risk of) VTE in orthopedic surgery,[1] and these drugs are pre-
dicted to have a great impact on the anticoagulant market.[7]
The rational design of orally bioavailable small-molecule fXa
inhibitors, achieving good enzyme potency and selectivity, has
been considerably supported by the X-ray co-crystal structures
of protein–ligand complexes, sophisticated molecular model-
[a] Dr. G. Lopopolo, Dr. M. de Candia, Prof. F. Campagna, Prof. C. Altomare
Dipartimento Farmaco Chimico
University of Bari A. Moro, Via E. Orabona 4, 70125 Bari (Italy)
E-mail: altomare@farmchim.uniba.it
[b] Prof. L. Panza
Dipartimento di Scienze del Farmaco
University of Piemonte Orientale A. Avogadro
Via Bovio 6, 28100 Novara (Italy)
[c] Dr. M. R. Romano, Prof. M. D. Lograno
Dipartimento di Bioscienze, Biotecnologie e Scienze Farmacologiche
University of Bari A. Moro, Via E. Orabona 4, 70125 Bari (Italy)
ChemMedChem 0000, 00, 1 – 10
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

C. Altomare et al.
MED
tional CꢀH···p interactions with the isopropyl group as well
with the side chains of Phe174, Tyr99, and Trp215.
An optimal oral anticoagulant should have a predictable
pharmacokinetic (PK) profile which allows fixed oral dosing
and a relatively wide therapeutic index. From a PK viewpoint,
it should be characterized by low plasma protein binding, low
volume of distribution, low systemic clearance for decreasing
drug–drug interactions, rapid onset of action, and ability to
bind clot-bound coagulation factors.[14] As a trend, small-mole-
cule inhibitors of fXa and thrombin possessing similar enzyme
binding affinities show better PK profiles and in vivo anticoa-
gulant activity when their logP values are less than 4.[15] The
high lipophilicity of compound 1 (logP>4), for example, may
unfavorably affect its plasma protein binding, distribution
volume, and systemic clearance. This prompted us to decrease
its lipophilicity through glucosidation, which is a molecular op-
timization strategy that allows improvement of both drug solu-
bility and targeting via glucose transporters.
An examination of the preferred docking pose of 1 into the
fXa binding site (Figure 2) suggested that the position ortho to
the amide moiety on the central phenyl ring could be a suita-
ble site for functionalization, likely allowing the hydrophilic
sugar moiety to be oriented toward the solvent and avoiding
any clash with the S1 and S4 enzyme pockets.
Figure 1. Structures of factor Xa inhibitors. Compound 1 was recently de-
scribed by us as a potent and selective fXa inhibitor.[11] Compounds 2
(MCM09)[16,17] and 3[18] are two patented glycan conjugates of phenylglycina-
mide- and anthranilamide-based fXa inhibitors, respectively.
pound 1 could be considered a promising lead for further
physicochemical optimization and pharmacological studies.
As predicted by our docking calculations (Figure 2), and in
good agreement with the X-ray crystallographic structures of
fXa–inhibitor complexes,[12,13] compound 1 preferably directs
As a sugar unit, we chose b-d-glucose which, in addition to
enhancing solubility, may be recognized by glucose transport-
ers (GLUTs) localized in the gastrointestinal (GI) tract. Our ap-
proach was also supported by recent patents disclosing two
glycan-bearing fXa inhibitors (Figure 1). Researchers at Mor-
phochem Chemie[16,17] reported a number of 3-benzamidino
derivatives of phenylglycine (PhGly), including MCM09 (2)
which has a b-d-glucose unit at the ortho position of the
phenyl side chain, whereas Astellas Pharma scientists disclosed
the glucuronic acid derivative (3) of an anthranilamide-based
fXa inhibitor.[18] In particular, compound 2 inhibited human fXa
with an in vitro IC50 value of 2.4 nm and also showed good
ex vivo anticoagulant properties in mice. Interestingly, 2 inhib-
ited melanoma lung cancer colonies in mice in a dose-depen-
dent manner.[19] The mechanism underlying this anticancer
effect has not yet been well-elucidated; however, it is known
that tumor cells express several pro-coagulant activities which
may be involved in hemostasis disturbances and metastasis,
mediated by the action of serine proteases on the proteinase-
activated receptors (PARs), suggesting that inhibitors of coagu-
lation factors could interfere with cellular proliferation in
cancer cell lines.[20] Moreover, it is known that GLUTs are over-
expressed in malignant cells,[21–24] such as colon, breast, and
lung carcinomas, not only contributing to glucose uptake but
also allowing the targeted delivery of glycan conjugates of an-
ticancer agents.[25]
Figure 2. Schematic representation of the binding mode of compound
1 into the binding site of human fXa, according to our docking calcula-
tions.[11] Key residues in the S1 and S4 binding sites are highlighted.
In this study, we modified compound 1 by connecting a b-
d-glucose unit to the central phenyl ring (16) through a propyl
spacer. Herein we report the synthesis, inhibition of blood co-
agulation factors and related serine proteases, and in vitro and
ex vivo anticoagulant activity of glycan 16 and related deriva-
tives.
the neutral 2-chlorothiophene group into the S1 pocket, with
the chlorine atom projecting toward the center of the Tyr228
phenyl ring. The protonated N1-isopropyl residue (the tertiary
ammonium head) resides at the center of the S4 aromatic box,
where it undergoes efficient cation–p interactions and addi-
&2
&
www.chemmedchem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 10
ÝÝ
These are not the final page numbers!

b-d-Glucosyl Conjugates of Factor Xa Inhibitors
MED
Results and Discussion
Synthesis
b-d-Glucose was conjugated at
the ortho position of the amide
group on the central phenyl
ring of compound 1 through an
ether C3-alkyl linker. This mini-
mum spacer length was pre-
ferred in order to avoid as many
unfavorable intramolecular and
intermolecular interactions as
possible that might arise from
directly linking to the glucose
moiety and/or from using short-
er spacers, such as a steric clash-
es between the sugar moiety
and the enzyme binding sites,
as well as to avoid possible for-
Scheme 2. Compounds 5, 6, and 7 were reported previously.[11] Reagents and conditions: a) compound 8, K2CO3,
dry DMF, RT, overnight; b) K2CO3, dry DMF, RT, overnight; c) SnCl2·2H2O, EtOAc, reflux, 48 h; d) MeONa, MeOH, RT,
3 h.
mation of b-elimination side
products during synthesis. The
peracetyl-protected glucose was
used as the starting material,
taking into account the compatibility of the acetyl ester cleav-
age conditions for preserving the other functional groups pres-
ent in the target molecule. Moreover, due to the anchimeric
assistance of the C2’ oxycarbonyl group, the 2’-OAc group
preferentially affords 1,2-trans-glycosidation, in our case, b-
anomer formation.
yield), overcoming side reactions and preserving as much of
the integrity of the entire molecule as possible. Cleavage of
the acetyl-protecting groups was achieved with MeONa/MeOH
at room temperature for 3 h to give 11 in almost quantitative
yield.
Compound 10 was coupled with N-Boc-isonipecotic acid in
DCC/HOBt/THF (12, 40% yield), followed by removal of the
Boc-protecting group to yield 13 (Scheme 3). In a one-pot pro-
cedure, Boc-deprotection and reductive amination of 12 with
acetone afforded N-isopropyl derivative 15 (86% yield). Deace-
tylation of 13 and 15 under basic conditions (MeONa/MeOH)
at room temperature provided compounds 14 and 16, respec-
tively, in greater than 90% yields.
The synthesis pathways for the target glucosyl conjugate 16
and related compounds are shown in Schemes 1–3. Com-
pound 4 was prepared in 53% yield by direct glycosidation of
3-bromo-1-propanol with 1,2,3,4,6-penta-O-acetyl-b-d-gluco-
pyranose, using BF3·Et2O as a promoter (Scheme 1).
3-Bromopropyl-2,3,4,6-tetra-O-acetyl-b-d-glucopyranoside 4,
the bromomethyl derivative 6 (prepared according to previ-
ously reported procedures[11]), and 2-nitroresorcinol were used
as starting materials for the synthesis of compound 10
(Scheme 2). Monoalkylation of a large excess of 2-nitroresorci-
nol with 3-(bromomethyl)-5-(2-chlorothiophen-5-yl)isoxazole
(6) afforded compound 8 (58% yield), which subsequently re-
acted with 4, allowing nitrophenylether 9 to be obtained in
almost quantitative yield. Reduction of the nitro to an amino
group in compound 10 was accomplished using SnCl2·2H2O in
EtOAc at reflux, which proved the best conditions among
those explored (N2H4/Ni-Raney in MeOH at reflux, Fe/NH4Cl in
EtOH at reflux, H2/Pt), to obtain the desired compound (52%
Inhibition of factor Xa and related enzymes
To explore fragment-growing effects, the newly synthesized
C3-alkyl-linked glucosyl conjugates 14–16, along with com-
pounds 5, 7, 8, and 11, were evaluated in vitro for inhibition of
fXa. The half maximal inhibitory concentration (IC50) was deter-
mined for each compound using fixed amounts of hfXa (2 nm)
and the chromogenic substrate Z-d-Arg-Gly-Arg-p-NA (40 mm)
and varying the concentrations of the inhibitor. Each IC50 was
determined in at least triplicate, and the values were averaged.
The IC50 values were then used to calculate the inhibition con-
stants (Ki values; Table 1) using the Cheng–Prusoff equation.[26]
Target compound 16 achieved sub-nanomolar fXa inhibition
activity (Ki =0.06 nm), displaying a fivefold increase in potency
relative to parent compound 1. Lineweaver–Burk plots were
generated for 16 using a fixed amount of hfXa (2 nm) and
varying concentrations of substrate (25–200 mm) in the ab-
sence or presence of inhibitor at four concentrations (0.075–
0.6 nm), which displayed a competitive inhibition mechanism.
Scheme 1. Reagents and conditions: a) BF3·Et2O, MS 4 ꢁ, dry CH2Cl2, 08C!RT,
overnight.
ChemMedChem 0000, 00, 1 – 10
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
&
3
&
ÞÞ
These are not the final page numbers!

C. Altomare et al.
MED
Scheme 3. Reagents and conditions: a) DCC, HOBt, dry THF, RT, 48 h; b) CF3COOH, CH2Cl2, RT, 3 h; c) acetone, Na(CN)BH3, MeOH, RT, 24 h; d) MeONa, MeOH, RT,
overnight.
which were in full agreement with X-ray co-crystal structures
Table 1. Evaluation of fragment-growing effects on fXa inhibition.
of fXa in complex with structurally similar inhibitors,[12,13] as-
Compd
Ki [nm][a]
suming the 5-(5-chlorothien-2-yl)isoxazol-3-yl group as the S1
binding fragment (5) to assess the stepwise free energy gain
(DDG) of the linker (2-aminophenyl), S4 binder (1-isopropylpi-
peridine), and hydrophilic carrier (C3-alkyl-linked glucose).
The 5-chlorothien-2-yl-bearing compound 5 showed a Ki
1
5
7
0.3[b]
93000
45000
70000
14700
37[c]
8
11
14
15
16
value of 93 mm, corresponding to a DG value of ꢀ23.0 kJmolꢀ1
;
0.2
0.06
this fragment has 13 heavy atoms and a ligand efficiency (LE,
namely the ratio of DG to the number of non-hydrogen
atoms) of 0.42, which is the same LE as the highly potent fXa
inhibitor 1 (DG=ꢀ54.4 kJmolꢀ1). Adding a 2-aminophenyl
linker to fragment 5 resulted in 7, with a small DG gain
(DDG5!7 =ꢀ1.7 kJmolꢀ1). The binding affinity further increases
upon connection of 7 with a propyloxy-glucose fragment at
[a] Inhibition constant values are means of three duplicate determinations
(SEM<5% of the mean). [b] Value taken from previous study.[11] [c] Ki
value for thrombin is 17.6 mm.
Re-plotting the slopes of the above double reciprocal plots
against inhibitor concentration gave a Ki value (0.11ꢁ0.06 nm)
which was not significantly different from that calculated by
the Cheng–Prusoff equation.
the position ortho to NH2, improving DG for 11 by 2.9 kJmolꢀ1
.
A slightly higher DG gain, due to the C3-alkyl-linked glucose
fragment, was calculated by comparing 16 with the parent
compound 1 (DDG1!16 =ꢀ3.8 kJmolꢀ1).
The peracetylated glycoside derivative 15, assayed for fXa in-
hibition, proved to be almost equipotent with 1 (Ki =0.2 nm)
and three times less potent than 16. Despite its lower inhibito-
ry activity in vitro, it is possible that peracetylated compound
15 could have a cell permeability (and likely GI absorption)
better than that of compound 16, as demonstrated for perace-
tylated xylosides endowed with antiproliferative activity.[27]
Compound 14, the des-isopropyl analogue of 16, showed
much lower fXa affinity (Ki =37 nm) and selectivity; the com-
parison between the respective fXa/thrombin selectivity ratios
showed that 16 is 30-fold more selective than 14 toward fXa.
The data in Table 1 allowed us to quantify the contributions
of single fragments to the measured free energy of binding of
compound 16. We took into account the binding mode of the
parent compound 1, supported by our docking calculations,[11]
The largest increase in affinity relative to 7 resulted from ad-
dition of the piperidine fragments (S4 binders). The piperidine
moiety was linked to the aromatic NH2 in 11 through an amide
bond, with the secondary amino group fully protonated under
the conditions of the enzyme assay (pH 8), improving DG of
14 by 14.6 kJmolꢀ1. Alkylation of the piperidine nitrogen in 14
with an isopropyl group improves the affinity of 16 still further
(DDG14!16 =ꢀ15.9 kJmolꢀ1). Globally, the measured contribu-
tion of the protonated 1-isopropylpiperidine (DDG11!16) equals
ꢀ30.5 kJmolꢀ1, which is a DG gain similar to that calculated
when comparing compounds
1
and
7
(DDG7!1 =
ꢀ29.7 kJmolꢀ1). With the exception of the deep S1 pocket, the
active site of fXa, including the S4 aromatic binding pocket, is
located on the surface of the enzyme. The crucial role in fXa
molecular recognition of the cation–p interactions in the S4
&4
&
www.chemmedchem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 10
ÝÝ
These are not the final page numbers!

b-d-Glucosyl Conjugates of Factor Xa Inhibitors
MED
binding site has been well-established by a series of high-affin-
ity ligands with excellent inhibitory potencies.[13] The inhibitors
examined herein should bind to the fXa active site, with the
protonated piperidine N1-isopropyl residue (i.e., the tertiary
ammonium head) at the center of the S4 aromatic box, where
the isopropyl group can undergo cation–p interactions and ad-
ditional CꢀH···p interactions. Our data indicate that the gain in
the binding DG amounts to increments of 14.6 and
15.9 kJmolꢀ1, corresponding to efficient NꢀH···p interactions
(11!14) and CꢀH···p interactions of the isopropyl group (14!
16) with the side chains of Phe174, Tyr99, and Trp215. The 2-
aminophenyl linker effect was estimated to be only
1.7 kJmolꢀ1, and the fragment deconstruction of sugar-bearing
compounds 11 and 16 resulted in a rather small DG gain (2.9–
3.8 kJmolꢀ1) for the C3-alkyloxy-linked glucose, which is unlike-
ly to arise from favorable interactions of the glucose moiety
with specific binding sites. Most likely, the hydrophilic sugar
moiety is directed out of the fXa binding site into the solvent
environment, as shown in literature for other hydrophilic moi-
eties installed on the central linker connecting S1 and S4 bind-
ing groups in fXa inhibitors.[28]
In vitro and ex vivo anticoagulant activity
Compounds 15 and 16 were characterized in vitro using clot-
ting assays in human plasma, namely, the measurements of
prothrombin time (PT) and activated partial thromboplastin
time (aPTT). in vitro anticoagulant activity was expressed as
the concentrations of test compounds required to double PT
and aPTT of the uninhibited clotting times. Both glycans
showed in vitro anticoagulant activity in the nanomolar range,
that is, one order of magnitude higher than that of parent
compound 1. In particular, 16 showed PT2 and aPTT2 values of
0.135 and 0.389 mm, respectively. The observed anticoagulant
activity was higher than that expected from inhibition constant
values of the purified fXa enzyme. In fact, compound 16 dis-
played a fivefold increase in fXa inhibitory potency relative to
1, whereas the corresponding anticoagulant activity increased
by 24- and 13-fold in PT and aPTT clotting assays, respectively.
This difference in anti-clotting potencies may result from a sub-
stantial decrease in lipophilicity for 16 (ClogP=2.05, as calcu-
lated by Bio-Loom software, Biobyte, Claremont, CA, USA) rela-
tive to 1 (ClogP=4.02), with a consequent lowering of non-
specific interactions of 16 with plasma proteins and other
plasma components and, ultimately, an increase in the concen-
tration available to bind the target coagulation factors (fXa,
thrombin). The in vitro PT2 and aPTT2 values for 15 were deter-
mined to be 0.495 and 0.840 mm, respectively. Although less
potent in vitro than 16, the peracetylated derivative 15 may
undergo in vivo hydrolysis, catalyzed by nonspecific esterases
able to cleave the acetyl groups and release the more active
compound 16.[27]
The most potent glycan fXa inhibitors, 15 and 16, were as-
sayed for their inhibitory activity against blood coagulation fac-
tors, namely thrombin and factor VIIa (fVIIa), and other serine
proteases, such as a-chymotrypsin, trypsin, and leukocyte elas-
tase, with the results compared with those obtained with the
reference compound 1 (Table 2). Both compounds 15 and 16
displayed higher anti-thrombin activity than that of parent
compound 1, with Ki values in the nanomolar range. Addition-
ally, their fXa/thrombin selectivity ratios remained high (775
and 1500 for 15 and 16, respectively). Compound 16 demon-
strated excellent selectivity (>1.7ꢂ106-fold) against fVIIa of the
coagulation cascade and the three other serine proteases
tested.
For an early assessment of their ability as orally active anti-
coagulants, we examined the anticoagulant potency of com-
pounds 1 and 16 after oral dosing in mice. Preliminary results
showed no significant effects of peracetylated derivative 15
(data not shown). Figure 3 depicts the PT prolongation effects
of 1 and 16, administered via oral gavage at equimolar doses
(50 and 75 mgkgꢀ1, respectively), in blood samples withdrawn
at the 1 h time point.
Table 2. Inhibition data against blood coagulation factors and other
serine proteases, and in vitro anticoagulant activities of compounds 1, 15,
and 16.
Enzyme Ki values[a] and
clotting assay data[b]
Compd
15
1
16
0.06
90
>100
>100
>100
>100
FXa [nm]
Thrombin [nm]
FVIIa [mm]
a-Chymotrypsin [mm]
Trypsin [mm]
Leukocyte elastase [mm]
PT2 [mm]
0.3[c]
0.2
155
NT
>100
NT
NT
11000[c]
>100
60[c]
>100
>100
3.30[c]
5.03
0.495
0.840
0.135
0.389
aPTT2 [mm]
[a] Inhibition constants against human activated factor Xa, bovine throm-
bin, human recombinant activated fVIIa, bovine a-chymotrypsin and tryp-
sin, and human leukocyte elastase; maximum concentration tested was
100 mm. Data are means of three duplicate determinations (SEM<5% of
the mean); NT=not tested. [b] Concentrations of test compounds re-
quired to double the prothrombin time (PT) and activated partial throm-
boplastin time (aPTT) of the uninhibited clotting times in human plasma.
[c] Data taken from previous study.[11]
Figure 3. Ex vivo anticoagulant activity of compounds 1 (50 mgkgꢀ1) and 16
(75 mgkgꢀ1) after oral dosing in mice. Animals were dosed via oral gavage
with either vehicle (control) or test compounds at equimolar doses
(0.11 mmolkgꢀ1), and prothrombin time (PT) was measured 60 min after ad-
ministration. PT values are expressed as means ꢁSEM of duplicate determi-
nations for three mice per group (**P<0.005, *P<0.05 using Dunnett’s
test).
ChemMedChem 0000, 00, 1 – 10
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
&
5
&
ÞÞ
These are not the final page numbers!

C. Altomare et al.
MED
Compound 1 (50 mgkgꢀ1) significantly prolonged plasma PT
60 min after treatment (27.7ꢁ3.5 s) versus controls (12.4ꢁ
0.5 s) (P<0.005 with Dunnett’s test). The mean PT value of the
group receiving compound 16 (75 mgkgꢀ1) was also signifi-
cantly higher than that of the controls (19.4ꢁ2.3 s; P<0.05),
but its ex vivo effect in mice was 1.4-fold lower than that ob-
served for the group receiving compound 1. In contrast with
the results from clotting assays in vitro, compounds 1 and 16
showed 2.2- and 1.6-fold PT prolongations, respectively, com-
pared with controls, evaluated 1 h after oral administration of
one high single dose (0.11 mmolkgꢀ1). To better understand
the origin of this apparent contrast, further ex vivo anticoagu-
lant activity experiments (PT measurements at different time
points and doses), and pharmacokinetic studies (oral bioavaila-
bility) are needed.
Experimental Section
Chemistry
Melting points were determined using the capillary method on
a Stuart Scientific SMP3 electrothermal apparatus and are uncor-
rected. IR spectra were recorded using KBr disks on a PerkinElmer
Spectrum One FT-IR spectrophotometer (PerkinElmer, Buckingham-
shire, UK), and the most significant absorption bands are expressed
1
in cmꢀ1. H NMR spectra were recorded at 300 MHz on a JEOL ECP
300 instrument. Chemical shift values are expressed in d and the
coupling constants J in hertz. Abbreviations are as follows: s, sin-
glet; d, doublet; t, triplet; q, quartet, dd, doublet of doublets; m,
multiplet; br, broad. Signals due to NH and OH protons were locat-
ed by deuterium exchange with D2O. Mass spectra were recorded
on an Agilent GC–MS 689-973. Elemental analyses (C, H, N) were
performed on an Euro EA3000 analyzer (Eurovector, Milan, Italy)
using the Analytical Laboratory Service of the Dipartimento Farma-
co Chimico of the University of Bari, and the results were within
0.40% agreement of theoretical values. HPLC analyses were carried
out using a Symmetry 5 mm C18 column (3.9ꢂ150 mm) from
Waters (Milford, MA, USA), eluted at different mobile phase compo-
sitions of MeOH/H2O mixtures. All HPLC measurements were car-
ried out at 25ꢁ18C with a flow rate of 1.0 mLminꢀ1 at both 254
and 320 nm on a Waters HPLC 1525 multisolvent delivery system,
equipped with a Waters 2487 variable wavelength UV detector.
Chromatographic separations were performed on silica gel 60 for
column chromatography (Merck 70–230 mesh or, alternatively, 15–
40 mesh for flash chromatography). Several compounds were syn-
thesized according to known procedures with slight modifications;
their melting points and spectral data were in full agreement with
those reported in literature, and no effort was made at this stage
to optimize the yields. Unless otherwise stated, all chemicals and
solvents were purchased from Sigma–Aldrich.
Conclusions
In this study, glucose-based derivatives of recently reported
highly potent fXa inhibitors were synthesized, in which a d-
glucose unit was linked, through a b-O-glycoside bond, to the
central phenyl group (linker) of the inhibitor using an ether
C3-alkyl spacer. The O-glucoside 16, which bears 2-chlorothio-
phene and 1-isopropylpiperidine as fragments binding the S1
and S4 enzyme pockets, respectively, achieved in vitro excel-
lent inhibition potency against human fXa, with a Ki value in
the picomolar range (60 pm) and high selectivity over throm-
bin, fVIIa, and other serine proteases of the trypsin family. As-
suming the 2-chlorothiophene S1 binder was the minimal frag-
ment, we deconstructed b-d-glucose-bearing compound 16
and analyzed fragment-growing effects on its affinity to fXa.
Our analysis revealed that the largest DG gain results from ad-
dition of the protonated 1-isopropylpiperidine (DDG=29.7–
30.5 kJmolꢀ1), which should undergo efficient cation–p and Cꢀ
H···p interactions within the aromatic S4 pocket by the piperi-
dine NꢀH and the isopropyl group, respectively. The C3-alkyl-
linked glucose improves DG by 2.9–3.8 kJmolꢀ1, indicating that
the hydrophilic sugar moiety most likely extends outside of
the enzyme binding site toward the solvent.
3-Bromopropyl 2,3,4,6-tetra-O-acetyl-b-d-glucopyranoside (4): A
solution of 1,2,3,4,6-penta-O-acetyl-b-d-glucopyranose (2.50 g,
6.40 mmol) and 3-bromo-1-propanol (0.87 mL, 9.61 mmol) in dry
CH2Cl2 (32 mL) containing 4 ꢁ molecular sieves (250 mg) was
cooled to 08C, and BF3·Et2O (4.10 mL, 32.0 mmol) was added over
2 min. The mixture was stirred overnight, neutralized with NEt3,
and filtered. The resulting solution was washed with saturated
NaHCO3, H2O, and brine, then dried (Na2SO4) and concentrated
under reduced pressure to provide a residue which was purified by
silica gel flash chromatography (EtOAc/hexane, 1:1 v/v), providing
compound 4 as a white foam (1.58 g, 53%). Spectral data were in
full agreement with those reported:[32] 1H NMR (300 MHz, CDCl3):
d=5.21 (t, J=9.5 Hz, 1H), 5.07 (t, J=9.8 Hz, 1H), 4.97 (dd, J=9.8,
8.0 Hz, 1H), 4.51 (d, J=8.0 Hz, 1H), 4.26 (dd, J=12.2, 4.9 Hz, 1H),
4.14 (dd, J=12.2, 2.4 Hz, 1H), 4.01–3.94 (m, 1H), 3.73–3.65 (m, 2H),
3.45 (t, J=5.2 Hz, 2H), 2.15 (m, 2H), 2.09 (s, 3H), 2.06 (s, 3H),
2.01(s, 3H), 1.99 ppm (s, 3H).
Compound 16 showed good in vitro anticoagulant activity,
as assessed by the clotting assays in pooled human plasma
(PT2 and aPTT2 values of 0.135 and 0.389 mm, respectively). In
a preliminary ex vivo anticoagulant assay in mice, compound
16 showed a 1.6-fold PT prolongation compared with controls
(P<0.05) 60 min after oral dosing (75 mgkgꢀ1), although it was
1.4-fold less active than the parent compound 1 at equimolar
doses (2.2 PT prolongation; P<0.005). Unlike 1, compound 16
does not fully comply with the Lipinski’s guidelines for an oral
drug with regard to molecular weight (696 Da) and number of
hydrogen bond acceptors (12 nitrogen and oxygen atoms).
Nevertheless, compound 16, in addition to being a lead struc-
ture for further optimization of highly potent fXa inhibitors as
orally bioavailable or parenteral compounds for use in inten-
sive care medicine, may have potential for improving the deliv-
ery of anticoagulants to malignant cells overexpressing GLUTs,
ultimately strengthening its possible role in cancer thera-
py.[19,29–31]
3-{[5-(5-Chlorothiophen-2-yl)isoxazol-3-yl]methoxy}-2-nitrophe-
nol (8): K2CO3 (196 mg, 1.42 mmol) was added to a stirred solution
of 2-nitroresorcinol (2.00 g, 12.9 mmol) in dry DMF (14 mL). After
30 min, a solution of 3-(bromomethyl)-5-(5-chlorothiophen-2-yl)i-
soxazole 5 (333 mg, 1.29 mmol), prepared as previously reported,[11]
in dry DMF (3 mL) was added dropwise in 1 h. The reaction mixture
was stirred overnight at room temperature, diluted with H2O, and
extracted with EtOAc. The combined organic layers were sequen-
tially washed with saturated Na2CO3 and brine, dried (Na2SO4), and
concentrated under reduced pressure to afford a residue, which
was purified by silica gel chromatography (EtOAc/cyclohexane, 1:1
v/v) to provide compound 8 as yellow solid (266 mg, 58%); mp:
&6
&
www.chemmedchem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 10
ÝÝ
These are not the final page numbers!

b-d-Glucosyl Conjugates of Factor Xa Inhibitors
MED
1
152–1548C; H NMR (300 MHz, CDCl3): d=10.12 (s, 1H), 7.41 (t, J=
carbodiimide (DCC, 101 mg, 0.49 mmol) were added to a 08C solu-
tion of 1-(tert-butoxycarbonyl)piperidine-4-carboxylic acid (123 mg,
0.54 mmol) in THF (5 mL). Compound 10 (347 mg, 0.49 mmol) was
added after 30 min of stirring. The reaction mixture was stirred for
48 h at room temperature, then the DCU was removed by filtra-
tion, and the filtrate was concentrated under reduced pressure.
The residue was dissolved in EtOAc and sequentially washed with
saturated 5% NaHCO3, 1 n HCl, and brine. The organic phase was
dried (Na2SO4), filtered, and concentrated under reduced pressure.
Purification by silica gel flash chromatography (EtOAc/cyclohexane,
8:2 v/v) provided 12 as a white solid (180 mg, 40%), which was
used in the next reaction without further purification: 1H NMR
(300 MHz, CDCl3): d=7.26 (d, J=4.0 Hz, 1H), 7.14 (t, J=8.6 Hz, 1H),
6.93 (d, J=4.0 Hz, 1H), 6.82 (br, 1H), 6.63 (d, J=8.3 Hz, 1H), 6.59
(d, J=8.3 Hz, 1H), 6.51 (s, 1H), 5.18 (t, J=9.2 Hz, 1H), 5.15 (s, 2H),
5.06 (t, J=9.8 Hz, 1H), 4.97 (dd, J=9.5, 8.0 Hz, 1H), 4.52 (d, J=
8.0 Hz, 1H), 4.23 (dd, J=12.2, 4.3 Hz, 1H), 4.14–3.95 (m, 6H), 3.71–
3.65 (m, 2H), 2.89–2.72 (m, 2H), 2.52–2.40 (m, 1H), 2.09–1.98 (m,
2H), 2.04 (s, 3H), 2.00 (s, 3H), 1.97 (s, 3H), 1.89 (s, 3H), 1.81–1.62
(m, 4H), 1.44 ppm (s, 9H); IR (KBr): v˜ =1756, 1693, 1618, 1232,
8.3 Hz, 1H), 7.30 (d, J=3.7 Hz, 1H), 6.95 (d, J=3.9 Hz, 1H), 6.77
(dd, J=8.3, 1.2 Hz, 1H), 6.64 (dd, J=8.6, 1.2 Hz, 1H), 6.58 (s, 1H),
5.27 ppm (s, 2H); IR (KBr): v˜ =3391, 3146, 3110, 1613, 1593, 1530,
1482, 1468, 1361, 1192, 1164, 1101, 797, 787 cmꢀ1; Anal. calcd for
C14H9ClN2O5S: C 47.6, H 2.57, N 7.94, found: C 48.04, H 2.82, N 7.83.
3-(3-{[5-(5-Chlorothiophen-2-yl)isoxazol-3-yl]methoxy}-2-nitro-
phenoxy)propyl 2,3,4,6-tetra-O-acetyl-b-d-glucopyranoside (9):
K2CO3 (821 mg, 5.94 mmol) was added to a stirred solution of 8
(1.40 g, 2.83 mmol) in dry DMF (20 mL). After 30 min, a solution of
3-bromopropyl 2,3,4,6-tetra-O-acetyl-b-d-glucopyranoside 4 (1.46 g,
3.12 mmol) in dry DMF (10 mL) was added dropwise. The reaction
mixture was stirred overnight at room temperature, diluted with
H2O, and extracted with EtOAc. The combined organic layers were
washed with brine, dried (Na2SO4), and concentrated under re-
duced pressure to provide a residue which was purified by silica
gel flash chromatography (EtOAc/hexane, 1:1 v/v), to provide 9 as
a white solid (2.03 g, 97%): mp: 84–868C; 1H NMR (300 MHz,
CDCl3): d=7.34–7.28 (m, 2H), 6.95 (d, J=4.0 Hz, 1H), 6.73 (d, J=
8.3 Hz, 1H), 6.67 (d, J=8.3 Hz, 1H), 6.44 (s, 1H), 5.25 (s, 2H), 5.19 (t,
J=9.5 Hz, 1H), 5.06 (t, J=9.5 Hz, 1H), 4.96 (dd, J=9.5, 8.0 Hz, 1H),
4.49 (d, J=8.0 Hz, 1H), 4.24 (dd, J=12.2, 4.6 Hz, 1H), 4.15–4.05 (m,
3H), 4.00–3.93 (m, 1H), 3.72–3.64 (m, 2H), 2.10–1.87 (m, 2H), 2.06
(s, 3H), 2.01 (s, 3H), 1.99 (s, 3H), 1.75 ppm (s, 3H); IR (KBr): v˜ =
1754, 1535, 1479, 1370, 1268, 1228, 114, 1046, 807, 792 cmꢀ1; Anal.
calcd for C31H33ClN2O5SꢂH2O: C 49.05, H 4.65, N 3.69, found: C
49.36, H 4.60, N 3.61.
1168, 1111, 1040, 903, 798 cmꢀ1
.
3-(3-{[5-(5-Chlorothiophen-2-yl)isoxazol-3-yl]methoxy}-2-(piperi-
dine-4-carboxamido)phenoxy)propyl b-d-glucopyranoside (14):
Redistilled TFA (0.15 mL, 1.95 mmol) was added to a 08C solution
of 12 (180 mg, 0.20 mmol) in CH2Cl2 (5 mL), and the mixture was
stirred at room temperature for 3 h. After concentration under re-
duced pressure, the residue was partitioned between EtOAc and
1 n NaOH, and the aqueous layer was extracted twice with EtOAc.
The combined organic layers were dried (Na2SO4), filtered, and con-
centrated under reduced pressure to provide the crude product 13
(150 mg, 0.18 mmol), which was cooled to 08C in MeOH (10 mL).
MeONa (25 mg, 0.46 mmol) was added, and the mixture was stirred
overnight at room temperature. After concentration under reduced
pressure, the resulting residue was purified by silica gel flash chro-
matography (CH2Cl2/MeOH, 8:2 v/v), providing 14 as a white solid
3-(2-Amino-3-{[5-(5-chlorothiophen-2-yl)isoxazol-3-yl]methoxy}-
phenoxy)propyl 2,3,4,6-tetra-O-acetyl-b-d-glucopyranoside (10):
SnCl2·2H2O (1.01 g, 4.49 mmol) was added to a stirred solution of 9
(665 mg, 0.90 mmol) in EtOAc (15 mL). The mixture was stirred at
reflux for 48 h, diluted with EtOAc, and washed four times with
cold 1 n NaOH. The organic phase was washed with brine, dried
(Na2SO4), and concentrated under reduced pressure. Purification by
silica gel flash chromatography (CH2Cl2/EtOAc/cyclohexane, 7:2:1 v/
v/v) provided 10 as a colorless amorphous solid (347 mg, 52%),
which was used in the next step without further purification:
1H NMR (300 MHz, CDCl3): d=7.26 (d, J=4.0 Hz, 1H), 6.92 (d, J=
4.0 Hz, 1H) 6.66–6.49 (m, 3H), 6.46 (s, 1H), 5.18 (t, J=9.2 Hz, 1H),
5.17 (s, 2H), 5.06 (t, J=9.8 Hz, 1H), 4.98 (dd, J=9.5, 8.0 Hz, 1H),
4.52 (d, J=8.0 Hz, 1H), 4.24 (dd, J=12.2, 4.6 Hz, 1H), 4.14–4.01 (m,
6H), 3.75–3.63 (m, 2H), 2.08–1.95 (m, 2H), 2.05 (s, 3H), 2.00 (s, 3H),
1.98 (s, 3H), 1.92 ppm (s, 3H).
1
(100 mg, 83%): mp: 165–1678C; H NMR (300 MHz, D2O): d=7.25
(d, J=6.0 Hz, 1H), 7.10 (t, J=8.3 Hz, 1H), 6.95 (d, J=6.0 Hz, 1H),
6.80 (s, 1H), 6.76 (d, J=8.3 Hz, 1H), 6.60 (d, J=8.6 Hz, 1H), 5.10 (s,
2H), 4.85 (m, 1H), 4.07 (d, J=7 Hz, 1H), 4.01 (t, J=5.5 Hz, 2H),
3.92–3.85 (m, 4H), 3.64 (d, J=11.3 Hz, 1H), 3.62–3.55 (m, 1H), 3.43
(dd, J=11.3, 4.6 Hz, 1H), 3.14–2.99 (m, 1H), 2.92 (t, J=7.5 Hz, 1H),
2.85–2.70 (m, 2H), 2.65–2.55 (m, 1H), 2.30–2.19 (m, 1H), 2.10–2.02
(m, 1H), 1.95–1.80 (m, 2H), 1.76–1.45 ppm (m, 2H); IR (KBr): v˜ =
3405, 3210, 1660, 1260, 1110, 1030, 790 cmꢀ1; MS (ESI+): m/z: 654
[M+H]+; Anal. calcd for C29H38ClN3O10SꢂH2O: C 51.67, H 5.98, N
6.23, found: C 51.77, H 5.84, N 6.39.
3-(2-Amino-3-{[5-(5-chlorothien-2-yl)isoxazol-3-yl]methoxy}phen-
oxy)propyl b-d-glucopyranoside (11): MeONa (35 mg, 0.60 mmol)
was added to a 08C solution of 10 (200 mg, 0.28 mmol) in MeOH
(10 mL), and the reaction mixture was stirred overnight at room
temperature. After concentration under reduced pressure, the ob-
tained residue was purified by silica gel flash chromatography
(CH2Cl2/MeOH, 9:1 v/v), providing compound 11 in good purity (>
95% by HPLC) as a pale yellow solid (130 mg, 85%): mp: 100–
3-(2-{[5-(5-Chlorothiophen-2-yl)isoxazol-3-yl]methoxy}-3-[1-(iso-
propyl)piperidine-4-carboxamido]phenoxy)propyl 2,3,4,6-tetra-
O-acetyl-b-d-glucopyranoside (15): Redistilled TFA (0.15 mL,
1.95 mmol) was added to
a 08C solution of 12 (180 mg,
0.20 mmol) in CH2Cl2 (5 mL), and the mixture was stirred at room
temperature for 3 h. After concentration under reduced pressure,
the residue was dissolved in MeOH (5 mL) at 08C, and acetone
(1 mL) and Na(CN)BH3 (15 mg, 0.23 mmol) were added. The mixture
was stirred at room temperature for 24 h, then concentrated under
reduced pressure. The residue was partitioned between EtOAc and
1 n NaOH, and the aqueous layer was extracted twice with EtOAc.
The combined organic phases were dried (Na2SO4), filtered, and
concentrated under reduced pressure. Purification by silica gel
flash chromatography (EtOAc/MeOH, 9:1 v/v), provided 15 as
a white solid (149 mg, 86%): mp: 154–1568C; 1H NMR (300 MHz,
CDCl3): d=7.25 (d, J=3.7 Hz, 1H), 7.10 (t, J=8.3 Hz, 1H), 6.90 (d,
1
1028C; H NMR (300 MHz, D2O): d=7.26 (d, J=4.0 Hz, 1H), 6.92 (d,
J=4.0 Hz, 1H) 6.66–6.49 (m, 3H), 6.46 (s, 1H), 5.18 (t, J=9.2 Hz,
1H), 5.17 (s, 2H), 5.06 (t, J=9.8 Hz, 1H), 4.98 (dd, J=9.5, 8.0 Hz,
1H), 4.52 (d, J=8.0 Hz, 1H), 4.24 (dd, J=12.2, 4.6 Hz, 1H), 4.14–
4.01 (m, 6H), 3.75–3.63 (m, 2H), 2.08–1.95 ppm (m, 2H); IR (KBr):
v˜ =3300, 1601, 1225, 1118, 1060, 790 cmꢀ1
.
3-(3-[1-(tert-Butoxycarbonyl)piperidine-4-carboxamido]-2-{[5-(5-
chlorothiophen-2-yl)isoxazol-3-yl]methoxy}phenoxy)propyl
2,3,4,6-tetra-O-acetyl-b-d-glucopyranoside (12): N-hydroxybenzo-
triazole hydrate (HOBt, 66 mg, 0.49 mmol) and N,N’-dicyclohexyl-
ChemMedChem 0000, 00, 1 – 10
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
&
7
&
ÞÞ
These are not the final page numbers!

C. Altomare et al.
MED
J=3.7 Hz, 1H), 6.80 (br, 1H), 6.60 (d, J=8.6 Hz, 1H), 6.56 (d, J=
8.6 Hz, 1H), 6.52 (s, 1H), 5.16 (t, J=9.8 Hz, 1H), 5.12 (s, 2H), 5.04 (t,
J=9.5 Hz, 1H), 4.95 (dd, J=9.5, 8.0 Hz, 1H), 4.50 (d, J=8.0 Hz, 1H),
4.22 (dd, J=12.2, 4.3 Hz, 1H), 4.12–3.90 (m, 4H), 3.71–3.62 (m, 2H),
2.90–2.88 (m, 2H), 2.79–2.62 (m, 1H), 2.38–2.11 (m, 3H), 2.09–1.72
(m, 6H), 2.03 (s, 3H), 1.98 (s, 3H), 1.96 (s, 3H), 1.88 (s, 3H),
1.02 ppm (d, J=6.7 Hz, 6H); 13C (300 MHz, [D6]DMSO): d=20.0
(2C), 22.3, 22.4 (2C), 22.5, 31.2 (2C), 50.1 (3C), 56.4, 63.6, 64.2, 66.6,
67.9, 70.2, 73.0, 73.5, 74.6, 100.8, 102.6, 107.5, 107.7, 116.9, 128.1,
129.0 (2C), 129.3 (2C), 129.4, 135.1, 155.8 (2C), 156.6, 165.2, 171.2
(2C), 171.9, 172.4 ppm; IR (KBr): v˜ =2995, 1745, 1669, 1462, 1383,
1227, 1116, 1045, 790 cmꢀ1; MS (ESI+): m/z: 864 [M+H]+; Anal.
calcd for C40H50ClN3O14S: C 55.58, H 5.83, N 4.86, found: C 55.84, H
5.84, N 5.08.
ies were performed at pH 8 for fXa and thrombin, and at pH 7.5
for a-chymotrypsin, as previously reported.[33] Kinetics for recombi-
nant human fVIIa, pancreas bovine trypsin, and human leukocyte
elastase (200 mm Tris buffer,150 mm NaCl, and 0.1% PEG6000,
pH 7.5), were performed by monitoring the absorbance increase at
405 nm and 258C with a microplate BioRad spectrophotometer.
The enzyme solution (50 mL) and buffer (50 mL) were mixed with
2 mL of DMSO solution containing the test compound or DMSO
alone as the control, and incubated (15 min for fVIIa, 30 min for
trypsin, and 10 min for elastase). Reactions were initiated by
adding 100 mL of substrate solutions, and the increase in absorb-
ance was monitored for 5 min. Initial velocities were determined,
and the concentrations of the inhibitors required to diminish the
control velocity by 50% (IC50) were calculated by nonlinear (sigmoi-
dal) regression. At least three independent IC50 values were deter-
mined to calculate inhibition constants (Ki) using the Cheng–Prus-
off equation.[26]
3-(2-{[5-(5-Chlorothiophen-2-yl)isoxazol-3-yl]methoxy}-3-[(1-iso-
propyl)piperidine-4-carboxamido]phenoxy)propyl b-d-glucopy-
ranoside (16): MeONa (20 mg, 0.37 mmol) was added to a 08C so-
lution of 15 (149 mg, 0.172 mmol) in MeOH (5 mL), and the mix-
ture was stirred overnight at room temperature. After concentra-
tion under reduced pressure, the resulting residue was purified by
silica gel flash chromatography (CH2Cl2/MeOH, 9:1 v/v), providing
16 as a white solid (115 mg, 96%): mp: 164–1668C; 1H NMR
(300 MHz, [D6]DMSO): d=8.72 (br, 1H), 7.55 (d, J=4.0 Hz, 1H), 7.30
(d, J=4.0 Hz, 1H), 7.16 (t, J=8.3 Hz, 1H), 6.86 (s, 1H), 6.76 (d, J=
8.3 Hz, 1H), 6.70 (d, J=8.6 Hz, 1H), 5.14 (s, 2H), 4.93 (br, 4H), 4.10
(d, J=7.7 Hz, 1H), 4.01 (t, J=5.5 Hz, 2H), 3.91–3.84 (m, 1H), 3.64
(d, J=11.3 Hz, 1H), 3.62–3.55 (m, 1H), 3.43 (dd, J=11.3, 4.6 Hz,
1H), 3.14–2.99 (m, 3H), 2.92 (t, J=8.0 Hz, 1H), 2.79–2.70 (m, 2H),
2.65–2.57 (m, 1H), 2.38–2.19 (m, 1H), 2.09–2.02 (m, 2H), 1.96–1.85
(m, 2H), 1.76–1.49 (m, 4H), 0.92 ppm (d, J=6.4 Hz, 6H); 13C
(300 MHz, CDCl3): d=20.2 (2C), 31.3 (2C), 31.6, 50.1 (2C), 56.1, 63.3,
64.2, 67.5, 67.6, 72.2, 75.7, 78.9, 79.1, 102.2, 105.3, 108.3, 108.7,
118.2, 129.6 (2C), 129.7, 129.9, 130.8 (2C), 135.1, 156.8, 157.6 (2C),
165.9 ppm; IR (KBr): v˜ =3405, 3210, 1661, 1601, 1262, 1109, 1029,
In vitro plasma clotting time assays: Clotting time(s) of the test
compounds, namely prothrombin time (PT) and activated partial
thromboplastin time (aPTT), were measured using a coagulometer
(Behnk Electronic, Norderstedt, Germany) and compared with
those from human control plasma. Pooled lyophilized human
plasma (100 mL; Futura Systems, Formello, Rome, Italy) was incu-
bated for 3 min at 378C with test compound solution (10 mL) or
solvent (DMSO maximum of 1%), followed by the addition of PT
reagent (200 mL), or aPTT reagent (100 mL) and 0.025m CaCl2
(100 mL; Futura Systems), to trigger clot formation. Each measure-
ment was performed in triplicate, and the concentrations of test
compound which caused twofold prolongation of the basal clot-
ting times (PT2 and aPTT2) were calculated from each individual
concentration–response curve.
Ex vivo anticoagulant assays in mice: All animal experiments were
performed in agreement with the Italian law on animal care
No. 116/1992 and the EEC/609/86, and all efforts were made to
minimize the number of animals used. CD1 male mice (Harlan Lab-
oratories) weighing 25–35 g were used. The mice were housed
under standard laboratory conditions with open access to standard
food and tap water. After overnight fasting, mice were anesthe-
tized by intraperitoneal injection with urethane (1.2 gkgꢀ1), and
test compounds 1 and 16, suspended in a 0.5% methylcellulose
solution, were administered orally at equimolar concentrations
(0.11 mmolkgꢀ1) using a gastric tube (FTP-20-30, InstechLabs). At
1 h after inhibitor administration, blood (0.2 mL) was collected
from the inferior vena cava into syringes containing sodium citrate
(3.8% v/v). Platelet poor plasma (PPP) was then prepared by centri-
fugation for 20 min at 1200 g to measure PT. Data (mean ꢁSEM)
were compared with those of the vehicle group. One way ANOVA
and multiple comparison tests (Dunnett’s test) were performed ac-
cording to a standard statistical package. P values less than 0.05
were considered statistically significant.
794 cmꢀ1
;
MS (ESI+): m/z: 696 [M+H]+; Anal. calcd for
C32H42ClN3O10Sꢂ0.5 H2O: C 54.50, H 6.14, N 5.96, found: C 54.28, H
6.10, N 6.03.
Biological assays
Inhibition assays for factor Xa, thrombin, and other serine proteases:
The test compounds were assayed in vitro for their inhibitory activ-
ity toward fXa, thrombin, and other serine proteases, determining
the hydrolysis rates of the synthetic chromogenic substrates moni-
tored at 405 nm. Enzymes and substrates were used as follows
(final concentrations): 2 nm human factor Xa and 0.04 mm S-2765
(Z-d-Arg-Gly-Arg-p-NA) from Chromogenix AB-Instrumentation Lab-
oratories (Milan, Italy); 0.41 UmLꢀ1 bovine thrombin from Sigma–
Aldrich (Milan, Italy) and 0.05 mm S-2238 (d-Phe-Pip-Arg-p-NA)
from Chromogenix ABInstrumentation Laboratories; 30 nm human
recombinant factor VIIa (reconstituted with 0.2 mL of deionized
water) and 1 mm Spectrozyme fVIIa from American Diagnostics
(Stamford, CT, USA); 0.4 mgmLꢀ1 bovine a-chymotrypsin and
0.185 mm N-succinyl-Ala-Ala-Pro-Phe-p-NA from Sigma–Aldrich;
2 nm bovine pancreas trypsin from Calbiochem (Darmstadt, Germa-
ny) and 4 mm S-2238; 5 nm human neutrophil leukocyte elastase
and 0.66 mm N-(methoxysuccinyl)-Ala-Ala-Pro-Val- p-NA from
Sigma–Aldrich. All buffer salts were purchased from Sigma–Aldrich.
Acknowledgements
C.A. and M.d.C. thank Prof. Raimondo De Cristofaro (Institute of
Internal Medicine and Geriatrics, and Hemostasis Research
Center, Catholic University School of Medicine, Rome, Italy) for
use of laboratory devices and helpful discussions on enzymatic
assay data pertaining to blood coagulation factors. C.A. acknowl-
edges financial support from the Italian Ministry for Education
Universities and Research (MIUR, Rome, Italy; PRIN 2007, grant
no. 2007T9HTFB 003).
Enzyme solutions were incubated with DMSO solutions of the test
inhibitors (DMSO did not exceed 1%) in various concentrations
(0.1–100 nm or 0.1–100 mm), before the respective chromogenic
substrates were added to initiate the enzyme kinetics. Kinetic stud-
&8
&
www.chemmedchem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 10
ÝÝ
These are not the final page numbers!

b-d-Glucosyl Conjugates of Factor Xa Inhibitors
MED
[18] F. Hirayama, J. Fujiyasu, D. Kaga, K. Negoro, D. Sasuga, N. Seki, K.
Suzuki, (Astellas Pharma Inc., Tokyo, Japan), WO2007055183, 2007.
[19] C. Rossi, S. Hess, R. W. Eckl, A. Di Lena, A. Bruno, O. Thomas, A. Poggi, J.
Thromb. Haemostasis 2006, 4, 608–613.
Keywords: factor Xa · glycosides · inhibitors · thrombin ·
thromboembolism
[20] M. Wojtukiewicz, E. Sierko, W. Kisiel, Semin. Thromb. Hemostasis 2007,
33, 621–642.
[21] M. M. Abouzied, E. S. Crawford, H. A. Nabi, J. Nucl. Med. Technol. 2005,
33, 145–155.
[22] X. C. Nguyen, W. W. Lee, J. H. Chung, S. Y. Park, S. W. Sung, Y. K. Kim, Y.
So, D. S. Lee, J. K. Chung, M. C. Lee, S. E. Kim, Eur. J. Radiol. 2007, 62,
214–219.
[23] J. Reinhard, W. E. Hull, C. W. von der Lieth, U. Eichhorn, H. C. Kliem, B.
Kaina, M. Wiessler, J. Med. Chem. 2001, 44, 4050–4061.
[24] C. D. Young, A. S. Lewis, M. C. Rudolph, M. D. Ruehle, M. R. Jackman,
U. J. Yun, O. Ilkun, R. Pereira, E. D. Abel, S. M. Anderson, PLOS One 2011,
6, e23205.
[25] Y.-S. Lin, R. Tungpradit, S. Sinchaikul, F. M. An, D. Z. Liu, S. Phutrakul, S. T.
Chen, J. Med. Chem. 2008, 51, 7428–7441.
[26] Y. Cheng, W. H. Prusoff, Biochem. Pharmacol. 1973, 22, 3099–3108.
[27] F. Cheng, R. Johnsson, J. Nilsson, L. ꢁ. Fransson, U. Ellervik, K. Mani,
Cancer Lett. 2009, 273, 148.
[28] T. Ishihara, N. Seki, F. Hirayama, M. Orita, H. Koshio, Y. Taniuchi, Y. Sakai-
Moritani, Y. Iwatsuki, S. Kaku, T. Kawasaki, Y. Matsumoto, S. Tsukamoto,
Bioorg. Med. Chem. 2007, 15, 4175–4192.
[1] D. J. P. Pinto, J. M. Smallheer, D. L. Cheney, R. M. Knabb, R. R. Wexler, J.
Med. Chem. 2010, 53, 6243–6274.
[2] R. J. Leadley, Jr., Curr. Top. Med. Chem. 2001, 1, 151–159.
[3] L.-A. Linkins, Annu. Rev. Med. 2005, 56, 63–77.
[4] P. C. Wong, E. J. Crain, C. A. Watson, B. Xin, J. Thromb. Haemostasis
2009, 7, 1313–1320.
[5] N. H. Hauel, H. Nar, H. Priepke, U. Ries, J.-M. Stassen, W. Wienen, J. Med.
Chem. 2002, 45, 1757–1766.
[6] M. de Candia, G. Lopopolo, C. Altomare, Expert Opin. Ther. Pat. 2009, 19,
1535–1580.
[7] I. Melnikova, Nat. Rev. Drug Discovery 2009, 8, 353–354.
[8] S. Maignan, V. Mikol, Curr. Top. Med. Chem. 2001, 1, 161–174.
[9] M. Nazarꢃ, H. Matter, D. W. Will, M. Wagner, M. Urmann, J. Czech, H.
Schreuder, A. Bauer, K. Ritter, V. Wehner, Angew. Chem. 2012, 124, 929–
935; Angew. Chem. Int. Ed. 2012, 51, 905–911.
[10] L. M. Salonen, M. C. Holland, P. S. J. Kaib, W. Haap, J. Benz, J. L. Mary, O.
Kuster, W. B. Schweizer, D. W. Banner, F. Diederich, Chem. Eur. J. 2012,
18, 213–222.
[11] G. Lopopolo, F. Fiorella, M. de Candia, O. Nicolotti, S. Martel, P. A. Car-
rupt, C. Altomare, Eur. J. Pharm. Sci. 2011, 42, 180–191.
[12] M. Nazarꢃ, D. W. Will, H. Matter, H. Schreuder, K. Ritter, M. Urmann, M.
Essrich, A. Bauer, M. Wagner, J. Czech, M. Lorenz, V. Laux, V. Wehner, J.
Med. Chem. 2005, 48, 4511–4525.
[13] L. M. Salonen, C. Bucher, D. W. Banner, W. Haap, J.-L. Mary, J. Benz, O.
Kuster, P. Seiler, W. B. Schweizer, F. Diederich, Angew. Chem. 2009, 121,
825–828; Angew. Chem. Int. Ed. 2009, 48, 811–814.
[29] L. R. Zacharski, Cancer Lett. 2002, 186, 1–9.
[30] J. T. Loynes, L. R. Zacharski, Expert Opin. Ther. Targets 2003, 7, 399–404.
[31] A. K. Kakkar, Cancer Treat. Rev. 2003, 29, 23–26.
[32] M. E. Jung, E. C. Yang, B. T. Vu, M. Kiankarimi, E. Spyrou, J. Kau, J. Med.
Chem. 1999, 42, 3899–3909.
[33] M. de Candia, F. Liantonio, A. Carotti, R. De Cristofaro, C. Altomare, J.
Med. Chem. 2009, 52, 1018–1028.
[14] S. Haas, Vasa 2009, 38, 13–29.
[15] M. Remko, J. Mol. Struct.: THEOCHEM 2009, 916, 76–85.
[16] Morphochem Chemie, Munich (Germany), US2008051688, 2008.
[17] T. Fuchs, R. Eckl (Morphochem Chemie, Munich, Germany), WO/2006/
008141, 2006.
Received: April 30, 2012
Revised: July 11, 2012
Published online on && &&, 0000
ChemMedChem 0000, 00, 1 – 10
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
&
9
&
ÞÞ
These are not the final page numbers!

FULL PAPERS
G. Lopopolo, M. de Candia, L. Panza,
M. R. Romano, M. D. Lograno,
F. Campagna, C. Altomare*
O-Glucosides of fXa inhibitors: b-d-
Glucose was conjugated through an
ether-linked C3-alkyl spacer to the cen-
tral phenyl ring of compound 1, provid-
ing 16, which showed picomolar inhibi-
tory potency (Ki =60 pm) against factor
Xa (fXa), good selectivity over thrombin,
sub-micromolar in vitro activity in the
prothrombin time (PT) clotting assay,
and a significant (1.6-fold) prolongation
of the basal PT in an ex vivo assay in
mice.
&& – &&
b-d-Glucosyl Conjugates of Highly
Potent Inhibitors of Blood Coagulation
Factor Xa Bearing 2-Chorothiophene
as a P1 Motif
&10
&
www.chemmedchem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 10
ÝÝ
These are not the final page numbers!