Thrombin-Inhibiting Anticoagulant Liposomes: Development and Characterization

Article abstract of DOI:10.1002/cmdc.201500489

Many peptides and peptidomimetic drugs suffer from rapid clearance in vivo; this can be reduced by increasing their size through oligomerization or covalent conjugation with polymers. As proof of principle, an alternative strategy for drug oligomerization

Full text of DOI:10.1002/cmdc.201500489

DOI: 10.1002/cmdc.201500489
Full Papers
Thrombin-Inhibiting Anticoagulant Liposomes:
Development and Characterization
Wegderes Endreas,[a] Jana Brüßler,[b] Doru Vornicescu,[a] Michael Keusgen,[a] Udo Bakowsky,[b]
and Torsten Steinmetzer*[a]
Many peptides and peptidomimetic drugs suffer from rapid
clearance in vivo; this can be reduced by increasing their size
through oligomerization or covalent conjugation with poly-
mers. As proof of principle, an alternative strategy for drug oli-
gomerization is described, in which peptidomimetic thrombin
inhibitors are incorporated into the liposome surface. For this
purpose, the inhibitor moieties were covalently coupled to
a palmitic acid residue through a short bifunctionalized ethyl-
ene glycol spacer. These molecules were directly added to the
lipid mixture used for liposome preparation. The obtained lipo-
somes possess strong thrombin inhibitory potency in enzyme
kinetic measurements and anticoagulant activity in plasma.
Their strong potency and positive z potential indicate that
large amounts of the benzamidine-derived inhibitors are locat-
ed on the surface of the liposomes. This concept should be ap-
plicable to other drug molecules that suffer from rapid elimina-
tion and allow covalent modification with a suitable fatty acid
residue.
Introduction
The therapeutic potential of peptidic drug candidates is often
hampered by their limited proteolytic stability and short half-
lives as a result of rapid renal clearance. Many unmodified pep-
tides are eliminated from plasma within a few minutes after
administration. Therefore, various strategies to extend the sta-
bility of peptides have been developed that also prolong their
half-lives.[1] This includes N- and C-terminal modifications, in-
corporation of d-configured or N-methylated amino acids, re-
placement of peptide bonds and labile amino acids with un-
natural residues, and numerous types of cyclizations.[2] Addi-
tional strategies are based on various types of conjugation to
increase the molecular mass of the drug molecules, which re-
sults in reduced renal clearance. This includes PEGylation
(PEG=polyethylene glycol), conjugation with polymers of N-
acetylneuraminic acid (polysialic acids) and hydroxyethyl starch
(HESylation), and conjugation with random polymers of pro-
line, alanine, and serine (PASylation).[3,4] The half-lives of drug
molecules can also be extended by conjugation with albumin
mediated by a special endosomal recycling mechanism by
using the high affinity of albumin to the neonatal Fc receptor
(FcRn).[5] An additional strategy is based on multimerization;
for instance, the plasma half-life of a dimeric erythropoietin,
obtained by chemical conjugation, was enhanced to 24 h com-
pared to 4 h for the monomeric form.[6]
Similar problems of rapid renal clearance and in some cases
also hepatobiliary clearance were found by us for various ben-
zamidine-derived peptidomimetic and substrate–analogue in-
hibitors of various trypsin-like serine proteases.[7–10] Therefore,
we investigated the possibility to achieve suitable oligomeriza-
tion of related protease inhibitors onto the surface of lipo-
somes for potential half-life prolongation. For this purpose, we
used previously described benzamidine-derived inhibitors
1 and 2 (Figure 1) that inhibit thrombin in the low nanomolar
Figure 1. Structures of used substrate analogue thrombin inhibitors.[11]
range and possess negligible potency against related proteases
such as factor Xa, plasmin, urokinase-type plasminogen activa-
tor (uPA), plasma kallikrein, and matriptase.[11] Thrombin is
a trypsin-like serine protease and a key enzyme of the blood
coagulation cascade. Various direct thrombin inhibitors, such
as oral dabigatran etexilate and parenteral r-hirudin, bivaliru-
din, and argatroban, have been approved and can be used as
anticoagulants for various prophylactic and acute applica-
tions.[12] Moreover, thrombin is also an excellent tool for study-
ing receptor–ligand interactions.[13–15] The protein data bank
contains nearly 400 crystal structures in complex with various
types of inhibitors; it is easily accessible and numerous test
systems to determine the potency of new inhibitors exist.
[a] W. Endreas, Dr. D. Vornicescu, Prof. M. Keusgen, Prof. T. Steinmetzer
Institute of Pharmaceutical Chemistry, Philipps University
Marbacher Weg 6, 35032 Marburg (Germany)
E-mail: steinmetzer@uni-marburg.de
[b] Dr. J. Brüßler, Prof. U. Bakowsky
Institute of Pharmaceutical Technology & Biopharmacy, Philipps University
Ketzerbach 63, 35032 Marburg (Germany)
Supporting Information and ORCID(s) from the author(s) for this article are
available on the WWW under http://dx.doi.org/10.1002/cmdc.201500489.
ChemMedChem 2016, 11, 340 – 349
340
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
The used inhibitor type contains an l-configured trifunction-
al amino acid in the P3 position. X-ray structure analysis of
a close analogue in complex with thrombin revealed that its
P3 side chain is directed into the solvent and should tolerate
further modification without losing potency.[11] In this manu-
script, we describe the covalent coupling of these inhibitors
with palmitic acid, which enables their incorporation into lipo-
somes. The fatty acid residue was attached either directly to
the P3 side chain or through a short bifunctional ethylene
glycol spacer. The inhibitory potency of the inhibitors before
and after incorporation into the liposomes was determined by
enzyme kinetic studies and by measurement of their anticoa-
gulant activities in plasma. The results will be described in this
publication.
Results
Synthesis
The new Lys- and Asn-derived thrombin inhibitors were pre-
pared as shown in Schemes 1 and 2, respectively. Inhibitor 1[11]
was coupled to 8-(benzyloxycarbonylamino)-3,6-dioxaoctanoic
acid (Cbz-Adoa-OH), and this was followed by removal of the
protecting group. This provided compound 3, which was fur-
ther coupled to palmitic acid or biotin to give inhibitor 4 or 5,
respectively. Furthermore, compound 1 was directly coupled
to palmitic acid to give inhibitor 6 lacking the ethylene glycol
spacer. Similarly, Asp compound 2 was treated with 1-(tert-bu-
toxycarbonylamino)-4,7,10-trioxa-13-tridecanamine (Boc-Tota-
Scheme 1. Synthesis of Lys-containing inhibitors 3–6. Reagents and conditions: a) BOP, DIPEA (2 equiv.) in DMF; b) 33% HBr in acetic acid, preparative HPLC;
c) palmitic acid or biotin, BOP, DIPEA (2 equiv.) in DMF, preparative HPLC; d) palmitic acid, BOP, DIPEA (2 equiv.) in DMF, preparative HPLC.
Scheme 2. Synthesis of Asn-containing inhibitors 7–9. Reagents and conditions: a) BOP, DIPEA (2 equiv.) in DMF; b) TFA, preparative HPLC; c) palmitic acid or
biotin, BOP, DIPEA (2 equiv.) in DMF, preparative HPLC.
ChemMedChem 2016, 11, 340 – 349
www.chemmedchem.org
341
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
H). Cleavage of the Boc group provided inhibitor 7, which was
converted into palmitoyl and biotinyl derivatives 8 and 9, re-
spectively.
proteases. As expected, further elongation of the P3 side chain
with the used ethylene glycol derivatives was well accepted
and resulted in low nanomolar inhibitors in both series. A
slightly decreased potency was only found for compound 6
obtained after direct coupling of the palmitoyl residue to the
Lys side chain. In contrast, the coupling of the Tota moiety to
the Asp side chain of inhibitor 2 provided Asn derivatives 7–9
with slightly enhanced thrombin affinity. This tendency con-
firms previous data, for which thrombin inhibition after incor-
poration of Asn in the P3 position was stronger than that of
the Asp analogue.[11]
Moreover, Boc-Tota-H was directly coupled with biotin,
which was followed by acidic deprotection. The obtained inter-
mediate was treated with palmitic acid to provide compound
13 [Me-(CH2)14-CO-Tota-Biotin; for the synthesis, see Scheme S1
in the Supporting Information], which was used for the prepa-
ration of bifunctionalized liposomes (Figure 2).
Biological assays
Enzyme kinetic studies
Preparation of anticoagulant liposomes
The inhibitory potencies of the synthesized inhibitors against
thrombin and a few related trypsin-like serine proteases are
summarized in Tables 1 and 2. The results indicate that all of
these compounds are highly selective thrombin inhibitors with
relatively weak affinity (Ki >3 mm) against the other studied
Owing to slightly better chemical accessibility, P3 Lys-contain-
ing palmitoylated inhibitors 4 and 6 were used for the prepa-
ration of liposomes made by the thin-film hydration
method.[16,17] The liposomes were obtained from a mixture of
Figure 2. Schematic presentation of bifunctionalized liposomes containing inhibitor 4 and biotin derivative 13 on their surface.
Table 1. Inhibitory potency of P3-Lys-derived inhibitors.
Compd
R
Ki [nm][a]
Ki [mm][a]
IC200 [mm]
Thrombin
FXa
Plasmin
uPA
aPTT
1
3
4
5
6
H-
Adoa-
Palmitoyl-Adoa-
Biotin-Adoa-
Palmitoyl-
0.73
3.92Æ0.21
7.28Æ0.93
5.93Æ0.49
3.52Æ0.49
6.13Æ0.99
3.74Æ0.31
3.12Æ0.69
4.16Æ0.48
3.59Æ0.57
7.41Æ0.18
25.31Æ4.28
30.32Æ5.37
44.49Æ3.18
31.36Æ4.25
51.01Æ6.53
0.10
0.12
0.32
0.71
5.29
0.82Æ0.07
1.39Æ0.06
0.89Æ0.04
2.36Æ0.62
[a] Data are the meanÆSD obtained for three measurements.
ChemMedChem 2016, 11, 340 – 349
www.chemmedchem.org
342
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Table 2. Inhibitory potency of P3-Asp/Asn-derived inhibitors.
Compd
R
Ki [nm][a]
Ki [mm][a]
IC200 [mm]
Thrombin
FXa
aPTT
uPA
aPTT
2
7
8
9
HO-
Tota-
Palmitoyl-Tota-
Biotin-Tota-
1.6
5.62Æ0.25
6.02Æ0.98
4.73Æ0.20
3.67Æ0.29
17.92Æ1.32
18.32Æ1.52
17.99Æ1.65
23.64Æ1.75
26.92Æ1.43
26.42Æ1.94
28.32Æ1.26
56.25Æ1.20
–
0.41Æ0.02
0.87Æ0.06
0.35Æ0.06
0.05
0.09
–
[a] Data are the meanÆSD obtained for three measurements.
DPPC (1,2-dipalmitoyl-sn-glycero-3-phosphocholine), CH (cho-
lesterol), and palmitoylated inhibitor (ratio 65:30:5 mol%).
Moreover, bifunctionalized liposomes containing a thrombin
inhibitor and an additional biotin moiety on the surface were
prepared by the same method (Figure 2). In this case, the
same mixture of DPPC, CH, and palmitoylated inhibitor 4 or 6
was used, which contained additional compound 13 (Palmi-
toyl-Tota-Biotin, 10% relative to thrombin inhibitor 4 or 6).
Analysis of the supernatants after liposome preparation re-
vealed that the inhibitor was most likely fully incorporated into
the liposomes. In the least, it was not possible to detect free
inhibitor 4 or 6 by HPLC or MS analysis. Moreover, no thrombin
inhibitory potency was observed upon analyzing the superna-
tants in enzyme kinetic assays.
allows visualization of formulations, and it confirmed the parti-
cle size measurements performed by DLS. For size determina-
tion, all objects within a representative scan area were individ-
ually evaluated (Figure 3). For the DPPC/CH liposomes (Fig-
ure 3A), spherical liposomes with a size range of 78Æ11 nm
could be visualized and a monomodal size distribution was de-
termined. The monofunctionalized liposomes in Figure 3B,C
also show spherical shapes and sizes similar to those obtained
from the DLS measurements (Table 3). For the liposomes con-
taining 5 mol% inhibitor 4, a size of 324Æ65 nm was mea-
sured, whereas inhibitor 6 led to liposomes that were 335Æ
87 nm in size. On the AFM images of these formulations, ag-
gregates of lipids are seen beside the liposomes. The addition
of biotinylated compound 13 to the formulation resulted in
a slight further increase in the liposome size (370Æ60 nm for
DPPC/CH/4/13, Figure 3D; 390Æ70 nm for DPPC/CH/6/13, Fig-
ure 3E), and again, some lipid aggregates are visible, especially
for the bifunctionalized liposomes containing inhibitor 6.
Physicochemical characterization of the liposomes
The z potential is the overall charge of a particle in a particular
medium, and it can be used to predict its interaction with
other molecules. The determined particle sizes and z potentials
are summarized in Table 3. For the unmodified DPPC/CH lipo-
somes, a negative z potential was found owing to the pres-
ence of negatively charged phosphoester groups. In contrast,
a positive potential was determined for all liposomes contain-
ing the positively charged thrombin inhibitors. Relative to that
of the pure DPPC/CH formulation, all modified liposomes pos-
sess an increased particle size.
Thrombin inhibition by the prepared liposomes
The inhibitory potency of the anticoagulant liposomes was de-
termined by enzyme kinetic measurements by using the fluo-
rogenic substrate Tos-Gly-Pro-Arg-AMC (Tos=tosyl, AMC=7-
amido-4-methylcoumarin) in the presence of thrombin
(Table 3). The provided inhibitor concentration on the x axis
(Figure 4) was calculated on the basis of the total inhibitor
amount used for the preparation of the liposomes, assuming
that it was completely incorporated and equally distributed on
In addition to measurements by dynamic light scattering
(DLS), atomic force microscopy (AFM) was conducted. AFM
Table 3. Thrombin inhibitory potency, anticoagulant activity, particle size, and z potential of the prepared liposomes.
Composition
z potential [mV][a]
Particle size [nm][a]
IC50 [nm][a,b]
IC200 [mm] aPTT
DPPC/CH
À8.60Æ3.13
+29.74Æ10.05
+23.56Æ4.79
+29.95Æ11.43
+37.18Æ16.19
75.27Æ15.96
353.27Æ159.14
330.35Æ124.12
392.78Æ191.37
343.23Æ181.85
–
–
DPPC/CH/4
DPPC/CH/6
DPPC/CH/4/13
DPPC/CH/6/13
0.55Æ0.03
81Æ8.09
4.0Æ0.24
116 Æ14.4
1.61
7.26
2.16
7.97
[a] Data are the meanÆSD obtained for three measurements. [b] IC50 values were determined in enzyme kinetic measurements with a fluorogenic sub-
strate.
ChemMedChem 2016, 11, 340 – 349
www.chemmedchem.org
343
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Figure 3. Visualization of the size distribution and morphology of the liposomes by using AFM: A) DPPC/CH, B) DPPC/CH/4, C) DPPC/CH/6, D) DPPC/CH/4/13,
E) DPPC/CH/6/13.
of the liposomes to improve the accessibility for the target
enzyme. Despite slightly decreased thrombin affinity relative to
that of the monofunctionalized liposomes, a similar tendency
was found for both bifunctionalized formulations. Particles pre-
pared from inhibitor 4 and biotinylated compound 13 inhibit
thrombin with IC50 =4 nm, whereas those containing inhibitor
6 possess reduced activity (IC50 =116 nm, Table 3). Moreover,
measurements with inhibitor-free DPPC/CH or DPPC/CH/13 lip-
osomes show no inhibitory potency against thrombin. For
these control measurements, the liposome formulations were
diluted similar to that described for the more potent inhibitor-
containing DPPC/CH/4 liposomes.
Figure 4. Inhibition of the thrombin-catalyzed cleavage of the fluorogenic
substrate Tos-Gly-Pro-Arg-AMC (10.9 mm) by the prepared liposome formula-
Anticoagulant activity in plasma
~
tions: DPPC/CH/4 (~), DPPC/CH/6 ( ), DPPC/CH/4/13 ( ), and DPPC/CH/6/
*
The anticoagulant activities of the free Lys-derived inhibitors
and the prepared liposomes were determined by the activated
partial thromboplastin time (aPPT) clotting assay in human
plasma and are expressed as their IC200 values, which corre-
spond to the inhibitor concentrations required to double the
clotting time relative to that of the control (Figure 5). As de-
scribed above, the inhibitor concentration of the liposomes
was calculated by assuming complete incorporation of com-
pound 4 or 6. Free inhibitors 1 and 3 possess excellent antico-
agulant potency with IC200 values of 0.1 and 0.12 mm, respec-
tively. A slightly reduced activity is found for palmitoyl-Adoa
and biotin derivatives 4 and 5, whereas analogue 6 lacking the
Adoa spacer has relatively poor potency (Table 1). A similar
tendency is observed for the mono- and bifunctional liposome
*
13 ( ). The inhibitor concentrations in the liposomes were calculated on the
basis of the total inhibitor amount used for their preparation, assuming that
the inhibitors were completely incorporated and distributed on the surface.
the surface. However, it cannot be excluded that some amount
of the inhibitor is incorporated in the inner liposome mem-
brane.
The monofunctionalized liposomes prepared from inhibitor
4 containing the Adoa spacer possess excellent inhibitory po-
tency (IC50 =0.55 nm), whereas the liposomes containing com-
pound 6 lacking the ethylene glycol linker have strongly re-
duced activity (IC50 =81 nm). This indicates the importance of
the flexible spacer between the inhibitor head and the surface
ChemMedChem 2016, 11, 340 – 349
www.chemmedchem.org
344
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Figure 6. Avidin-mediated neutralization of thrombin inhibition. Different
Figure 5. aPPT-assay of the Lys-derived inhibitors and liposome preparations
~
*
)
avidin concentrations were added to the free biotinylated inhibitors 5 (
^
*
*
^
in plasma [free inhibitors: 1 ( ), 3 ( ), 5 ( ), 4 ( ), 6 ( ); monofunctionalized
!
~
&
*
and 9 ( ), to monofunctionalized DPPC/CH/4 liposomes ( ), and to bifunc-
liposomes: DPPC/CH/4 ( ), and DPPC/CH/6 ( ); bifunctionalized liposomes:
!
&
tionalized DPPC/CH/4/13 liposomes containing different concentrations of
DPPC/CH/4/13 ( ), and DPPC/CH/6/13 ( )]. The dotted line represents the
~
inhibitor 13 ( : 10 mol%, 3: 20 mol%, ": 50 mol%, ~: 100 mol% relative to
doubling of the clotting time in the absence of inhibitor.
inhibitor 4). Measurements were performed with 15 pm thrombin in the
presence of 11 mm Tos-Gly-Pro-Arg-AMC.
preparations, whereas the particles containing compound 4
with the Ado spacer are much more potent than their ana-
logues based on inhibitor 6 (Table 3).
tion-dependent effect, it was only possible to achieve partial
neutralization of thrombin inhibitory potency by the addition
of avidin (Figure 6). A stronger neutralizing effect was observed
for free biotinylated inhibitors 5 and 9, whereas no influence
was found for the control DPPC/CH/4 liposomes lacking biotin-
ylated derivative 13.
Neutralization of bifunctionalized DPPC/CH/4/13 liposomes by
avidin
In previous work, a chimeric anticoagulant that could be neu-
tralized was prepared[18] by coupling the active site directed
thrombin inhibitor CRC220[19] to an idraparinux-derived syn-
thetic pentasaccharide[20] through a short spacer consisting of
a short ethylene glycol unit attached to a Lys-containing di-
peptide segment. The thrombin inhibitor segment enabled the
inhibition of fluid phase and clot-bound thrombin, whereas
the pentasaccharide led to antithrombin-mediated factor Xa in-
hibition and a prolonged half-life of approximately 3 h with
more predictable pharmacokinetics. Moreover, the Lys side
chain was further coupled to biotin, which enabled neutraliza-
tion of these anticoagulants after injection of avidin.[21]
Discussion
Various groups have described the development of antithrom-
botic nanoparticles and micelles by using different antiplatelet
and anticoagulant strategies. For instance, antiplatelet activity
was reported for nanosilver particles that accumulate within
platelets and reduce interplatelet proximity.[22] Anticoagulant
micelles were obtained by mixing covalently coupled conju-
gates of DSPE-PEG2000-maleimide (DSPE=1,2-distearoyl-sn-
glycero-3-phosphoethanolamine) with a cysteine-containing
derivative of the bivalent 20-mer thrombin inhibitor bivalirudin
(hirulog) and with a clot-targeting pentapeptide sequence Cys-
To adapt this antidote strategy, bifunctionalized liposomes
were prepared by adding biotinylated compound 13 during
their preparation (10% compound 13 relative to inhibitor 4 or
6, Figure 2). The obtained liposomes (DPPC/CH/4/13 and
DPPC/CH/6/13) possess slightly reduced inhibitory potency
against thrombin relative to the monofunctionalized formula-
tion. Also in this case, the bifunctionalized liposomes contain-
ing inhibitor 4 are more potent than their analogues with in-
hibitor 6.
To test the neutralization of the thrombin inhibitory potency
of the free biotinylated inhibitors and selected liposome for-
mulations, the inhibitor amount was adjusted to provide a re-
sidual thrombin activity of approximately 10% in the absence
of avidin (Figure 6). Addition of avidin to the DPPC/CH/4/13
liposomes had only a minor effect on the neutralization of
their thrombin inhibitory potency. To enhance the efficacy of
the antidote, additional liposomes with higher concentrations
of biotinylated derivative 13 (20, 50, and 100% relative to in-
hibitor 4) were prepared. However, despite a slight concentra-
Arg-Glu-Lys-Ala.[23]
Moreover,
d-Phe-Pro-Arg-chloromethyl
ketone (PPACK)-functionalized perfluorocarbon (PFC)-core
nanoparticles were prepared and were shown to possess sig-
nificant thrombin inhibitory potency in enzyme kinetic studies
and to inhibit clotting in plasma.[24] However, chloromethyl
ketone derived inhibitors have limited selectivity, and previous
studies revealed that the irreversible thrombin inhibitor PPACK
suffers from a very short half-life of approximately 2–3 min
in vivo.[25,26] In a following study, these PFC-core nanoparticles
were conjugated with bivalirudin covalently coupled to
a DSPE-carboxy-PEG2000 derivative.[27] Although no specific in-
formation was provided for the quality and yield of this reac-
tion, the described carbodiimide-mediated coupling of unpro-
tected bivalirudin without any pre-activation of the DSPE-car-
boxy-PEG2000 derivative might be challenging because of sev-
eral functional groups present in this relatively large peptidic
thrombin inhibitor. The same group also reported the prepara-
ChemMedChem 2016, 11, 340 – 349
www.chemmedchem.org
345
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
tion of anticoagulant liposomes obtained by coupling PPACK
to preformed carboxy-functionalized liposomes.[28] Compared
with free PPACK, these liposomes possess reduced inhibitory
potency against thrombin, as determined by using an enzyme
kinetic assay. However, they still showed significant anticoagu-
lant activity in an aPPT assay in plasma and decreased throm-
bus formation in vivo.
tions. Their significant anticoagulant activity and strongly in-
creased positive z potential indicate that a large amount of
the inhibitor is present on the surface. However, we cannot ex-
clude that some molecules are incorporated in an opposite di-
rection, whereby the inhibitory portion is located inside the
liposomes. The same strategy could be applied to other drug
molecules that tolerate modification with a linker coupled to
a suitable fatty acid. Although we did not perform any phar-
macokinetic studies, we assume that incorporation of these in-
hibitors into liposomes should increase their half-lives in circu-
lation relative to that of the unmodified peptidic drug mole-
cules. Until now, we have only used one standard mixture of
DPPC/CH (molar ratio ꢀ2:1) as the liposomal formulation. Cer-
tainly, their stability can be further improved, for example, by
adding small amounts of PEG-containing lipids such as 1,2-dis-
tearoyl-sn-glycero-3-phosphoethanolamine-N-[maleimide(poly-
ethylene glycol)-2000], which protect the liposomes from rapid
phagocytotic elimination.[29]
In our work, we investigated a different strategy for the
preparation of anticoagulant liposomes. For this purpose,
a highly potent and specific thrombin inhibitor was coupled
through a short ethylene glycol spacer to a palmitic acid. This
palmitoylated inhibitor is synthetically well accessible and can
be easily purified and characterized. Moreover, in contrast to
highly reactive chloromethyl ketones,[25] it is a chemically
stable compound. To obtain anticoagulant liposomes, the 4-
MeO-bs-Lys(Adoa-Palmitoyl)-Pro-4-Amba·TFA (4) conjugate was
simply mixed with standard reagents such as 1,2-dipalmitoyl-
sn-glycero-3-phosphocholine (DPPC) and cholesterol (CH)
during liposome preparation. This convenient strategy avoids
chemical modification of preformed and functionalized lipo-
somes, which can lead to side reactions on the inhibitor
moiety and requires further purification steps, for example, re-
moval of coupling reagents. On the basis of the IC50 values ob-
tained from enzyme kinetic studies, we found that liposomes
prepared from inhibitor 4 containing the flexible ethylene
glycol spacer are more than 100-fold more potent than lipo-
somes prepared from 4-MeO-bs-Lys(Palmitoyl)-Pro-4-Amba·TFA
(6) without the linker. This indicates that the potency of the
liposomes strongly depends on a sufficient distance between
the phospholipid surface and the thrombin inhibitor head. We
assume that the longer distance improves the accessibility of
the inhibitor for thrombin and probably enables simultaneous
binding of several thrombin molecules to a single particle. A
similar tendency with improved anticoagulant potency of the
liposomes containing inhibitor 4 was observed in the plasma
clotting assay.
Experimental Section
General methods
Analytical HPLC experiments were performed with a Shimadzu LC-
10 A system (analytical HPLC column: Nucleodur C18, 5 mm, 100 Š,
4.6”250 mm, Machery-Nagel, Düren, Germany) with UV detection
at l=220 nm. A linear gradient (1% increase in solvent B per
minute, flow rate 1 mLminÀ1) of 0.1% trifluoroacetic acid (TFA) in
acetonitrile (solvent B) in 0.1% TFA in water (solvent A) at different
starting concentrations of solvent B was used. The final inhibitors
were purified to more than 95% purity (based on detection at l=
220 nm) by preparative HPLC (pumps: Varian PrepStar Model 218
gradient system, detector: ProStar Model 320, fraction collector:
Varian Model 701) equipped with either a C18 column (Nucleosil,
5 mm, 300 Š, 32 mm”250 mm, Macherey–Nagel, Düren, Germany)
or a C8 column (Nucleodur, 5 mm, 100 Š, 32”250 mm) by using the
same solvents (gradient: 0.5% increase in solvent B per minute,
flow rate 20 mLminÀ1). All final compounds were obtained as
lyophilized TFA salts after preparative HPLC. The molecular masses
of the synthesized compounds were determined with
a QTrap 2000 ESI spectrometer (Applied Biosystems, now Life Tech-
Bifunctionalized liposomes containing the inhibitor 4 and
biotinylated Me-(CH2)14-CO-Tota-Biotin (13) on their surface
were prepared to adopt an avidin-mediated neutralization
strategy previously described for other types of thrombin in-
hibitors.[21] However, only a weak reduction of the inhibitory
potency could be achieved for bifunctionalized liposomes,
which contain 10% of compound 13 compared to inhibitor 4.
Although the efficacy of neutralization was slightly enhanced
for liposomes containing increasing amounts of derivative 13,
only partial neutralization was possible. Surprisingly, neutraliza-
tion of free biotinylated thrombin inhibitors 4-MeO-bs-Lys(A-
doa-Biotinyl)-Pro-4-Amba·TFA (5) and 4-MeO-bs-Asp(Tota-Bio-
tinyl)-Pro-4-Amba·TFA (9) by addition of avidin was more effec-
tive. At present, we have no explanation for the found discrep-
ancies and relatively poor efficacy of this antidote strategy.
1
nologies, Carlsbad, CA). H NMR and 13C NMR spectra were record-
ed by using a Jeol-ECX500 (Jeol Inc., Peabody, MA) and were refer-
enced to internal solvent signals. The reagents for synthesis, in-
cluding the amino acid derivatives, coupling reagents, and sol-
vents, were obtained from Orpegen, Bachem, Calbiochem, Alfa
Aesar, Fluka, Acros, or Aldrich. 1-(tert-Butyloxycarbonylamino)-
4,7,10-trioxa-13-tridecanamine (Boc-Tota-H) and 8-(Cbz-amino)-3,6-
dioxaoctanoic acid (Cbz-Adoa-OH) were purchased from Iris Bio-
tech GmbH (Marktredwitz, Germany).
Synthesis
Inhibitors 1 and 2: Synthesized as previously described.[11]
4-MeO-bs-Lys(Adoa)-Pro-4-Amba·2TFA (3): A mixture of com-
pound 1[11] (100 mg, 0.13 mmol) and Cbz-Adoa-OH (39 mg,
0.13 mmol) in DMF (2 mL) was cooled to 08C and then treated
with N,N-diisopropylethylamine (DIPEA; 44 mL, 0.26 mmol) followed
by the stepwise addition of 1-benzotriazolyloxytris(dimethylamino)-
phosphonium hexafluorophosphate (BOP; 57.5 mg, 0.13 mmol).
Conclusions
The described strategy for the preparation of anticoagulant lip-
osomes can be considered as proof of principle for the con-
venient incorporation of protease inhibitors into lipid formula-
ChemMedChem 2016, 11, 340 – 349
www.chemmedchem.org
346
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
The mixture was stirred for 15 min at 08C and for 3 h at room tem-
perature. The solvent was evaporated under reduced pressure, and
the remaining residue was treated with 33% HBr in acetic acid
(1 mL). The solvent was removed under reduced pressure, and the
product was purified by preparative HPLC and lyophilized from
water to afford a white solid (66.8 mg, 56%). 1H NMR (500 MHz,
[D6]DMSO): d=9.22 (s, 2H), 9.18 (s, 2H), 8.36 (t, J=6.1, 1H), 7.80 (s,
3H), 7.73–7.62 (m, 6H), 7.37 (d, J=8.5, 2H), 7.05–7.01 (m, 2H), 4.31
(dd, J=16.3, 6.0, 1H), 4.25 (dd, J=16.3, 6.1, 1H), 3.89–3.83 (m, 4H),
3.80 (s, 3H), 3.60–3.54 (m, 6H), 3.48–3.42 (m, 1H), 3.39–3.32 (m,
1H), 3.01–2.91 (m, 4H), 1.88–1.65 (m, 4H), 1.58–1.18 ppm (m, 6H).
13C NMR (126 MHz, [D6]DMSO): d=171.5, 169.1, 169.0, 165.4, 162.1,
158.4, 158.1, 145.9, 132.5, 128.7, 128.0, 127.1, 126.4, 113.9, 69.9,
69.9, 69.4, 66.9, 55.6, 53.5, 46.5, 41.5, 37.9, 31.4, 29.0, 28.6, 24.3,
22.1 ppm. MS: m/z (%): 690.29 [M+H]+. HPLC (start at 10% sol-
vent B): tR =18.4 min.
J=16.2, 6.1, 1H), 3.93 (dd, J=7.7, 3.8, 1H), 3.88 (td, J=8.5, 5.7,
1H), 3.83 (s, 3H), 3.52–3.36 (m, 2H), 2.91 (dd, J=12.4, 6.4, 2H), 2.00
(t, J=7.4, 2H), 1.94–1.67 (m, 4H), 1.64–1.33 (m, 4H), 1.33–1.09 (m,
28H), 0.85 ppm (t, J=6.9, 3H). 13C NMR (126 MHz, [D6]DMSO): d=
171.9, 171.5, 169.2, 165.4, 162.1, 158.5, 158.2, 145.9, 132.6, 128.7,
128.0, 127.1, 126.3, 113.9, 59.4, 55.6, 53.4, 46.6, 41.6, 38.1, 35.4,
31.3, 31.2, 29.0, 28.9, 28.7, 28.6, 28.5, 25.2, 24.3, 22.1, 22.0,
13.9 ppm. MS: m/z (%): 783.45 [M+H]+. HPLC (start at 10% sol-
vent B): tR =60.36 min.
4-MeO-bs-Asp(Tota)-Pro-4-Amba·2 TFA (7): Compound 2[11]
(100 mg, 0.15 mmol) and Boc-Tota-OH (48 mg, 0.15 mmol) were
coupled in DMF (2 mL) by using BOP (66 mg, 0.15 mmol) and
DIPEA (51 mL, 0.3 mmol) as described for the preparation of com-
pound 3. The solvent was evaporated under reduced pressure, and
the residue was treated for 30 min with TFA (1 mL). After concen-
tration under reduced pressure, the product was purified by prepa-
rative HPLC and lyophilized from water to afford a colorless solid
4-MeO-bs-Lys(Adoa-Palmitoyl)-Pro-4-Amba·TFA (4): Compound 3
(100 mg, 0.11 mmol) and palmitic acid (28 mg, 0.11 mmol) were
coupled by using BOP (48.6 mg, 0.11 mmol) and DIPEA (38 mL,
0.22 mmol) as described for the preparation of compound 3. DMF
was evaporated under reduced pressure, and the product was pu-
rified by preparative HPLC and lyophilized from tert-butyl alcohol
1
(60 mg, 42%). H NMR (500 MHz, [D6]DMSO): d=9.30 (s, 2H), 9.20
(s, 2H), 8.17 (t, J=6.2, 1H), 8.09 (d, J=9.2, 1H), 8.03 (t, J=5.5, 1H),
7.83–7.63 (m, 7H), 7.34 (d, J=8.5, 2H), 7.06 (d, J=9.0, 1H), 4.33–
4.18 (m, 3H), 3.85–3.78 (m, 4H), 3.59 (dd, J=11.9, 5.5, 1H), 3.51–
3.35 (m, 13H), 3.26–3.20 (m, 2H), 2.92–2.73 (m, 4H), 2.62 (dd, J=
15.4, 9.4, 1H), 2.32 (dd, J=15.5, 5.3, 1H), 1.83–1.69 (m, 4H),
1.41 ppm (quint., J=6.9, 2H). 13C NMR (126 MHz, [D6]DMSO): d=
171.1, 169.4, 169.0, 165.4, 162.2, 158.5, 158.3, 145.8, 132.6, 128.6,
127.9, 126.9, 126.3, 118.3, 115.9, 114.0, 69.6, 69.6, 69.4, 69.4, 67.8,
67.3, 59.9, 55.7, 50.2, 46.4, 41.6, 36.7, 35.7, 29.0, 28.8, 27.1,
23.7 ppm. MS: m/z (%): 734.37 [M+H]+. HPLC (start at 10% sol-
vent B): tR =19.65 min.
to afford
a
white solid (75 mg, 65.4%). 1H NMR (500 MHz,
[D6]DMSO): d=9.24 (s, 2H), 9.08 (s, 2H), 8.35 (t, J=6.1, 1H), 7.78 (t,
J=5.6, 1H), 7.75–7.65 (m, 5H), 7.62 (t, J=5.9, 1H), 7.41 (d, J=8.4,
2H), 7.06 (d, J=9.0, 2H), 4.35 (dd, J=16.3, 6.1, 1H), 4.29 (dd, J=
16.2, 6.1, 1H), 3.94–3.80 (m, 7H), 3.58–3.46 (m, 5H), 3.40 (t, J=6.0,
3H), 3.22–3.15 (m, 2H), 3.00 (dd, J=13.0, 6.4, 2H), 2.04 (t, J=7.5,
2H), 1.93–1.68 (m, 4H), 1.61–1.16 (m, 32H), 0.85 ppm (t, J=6.9,
3H). 13C NMR (126 MHz, [D6]DMSO): d=172.2, 171.5, 169.2, 169.0,
165.4, 162.1, 158.4, 158.1, 146.0, 132.5, 128.7, 128.0, 127.0, 126.3,
113.9, 70.1, 70.0, 69.3, 69.1, 59.4, 55.6, 53.5, 46.6, 41.5, 38.4, 37.8,
35.3, 31.2, 29.0, 29.0, 28.95, 28.9, 28.9, 28.7, 28.6, 28.6, 25.2, 24.3,
22.1, 22.0, 13.9 ppm. MS: m/z (%): 928.63 [M+H]+. HPLC (start at
50% solvent B): tR =20.88 min.
4-MeO-bs-Asp(Tota-Palmitoyl)-Pro-4-Amba·TFA (8): Compound 7
(50 mg, 52 mmol) and palmitic acid (13.3 mg, 52 mmol) were cou-
pled by using BOP (23 mg, 52 mmol) and DIPEA (17.7 mL, 0.1 mmol)
as described for the preparation of compound 3. The product was
purified by preparative HPLC and lyophilized from tert-butyl alco-
hol to afford a colorless solid (30 mg, 53%). 1H NMR (500 MHz,
[D6]DMSO): d=9.28–9.16 (m, 2H), 9.07 (s, 2H), 8.16 (t, J=6.2, 1H),
8.10 (d, J=9.1, 1H), 8.07–7.99 (m, 1H), 7.79–7.63 (m, 5H), 7.45 (d,
J=8.4, 1H), 7.34 (d, J=8.4, 1H), 7.09–6.98 (m, 2H), 4.36–4.21 (m,
2H), 3.85–3.77 (m, 3H), 3.59 (dd, J=14.6, 8.1, 1H), 3.49–3.19 (m,
14H), 3.05–2.73 (m, 4H), 2.64 (dd, J=15.5, 9.8, 1H), 2.32 (dd, J=
15.5, 5.0, 1H), 2.04–1.65 (m, 5H), 1.61–1.36 (m, 6H), 1.20 (s, 24H),
0.82 ppm (t, J=6.9, 3H). 13C NMR (126 MHz, [D6]DMSO): d=172.0,
171.0, 169.4, 169.0, 165.3, 162.2, 158.4, 158.1, 145.9, 132.7, 128.5,
128.1, 127.9, 127.2, 126.9, 126.2, 114.3, 114.0, 69.7, 69.5, 69.4, 68.1,
67.8, 60.0, 55.7, 50.2, 46.4, 41.7, 35.6, 35.3, 31.2, 29.4, 29.2, 29.0,
28.7, 28.6, 25.2, 23.6, 22.0, 13.9 ppm. MS: m/z (%): 972.41 [M+H]+.
HPLC (start at 10% solvent B): tR =62.57 min.
4-MeO-bs-Lys(Adoa-Biotinyl)-Pro-4-Amba·TFA (5): Inhibitor
3
(100 mg, 0.11 mmol) and biotin (27 mg, 0.11 mmol) were suspend-
ed in DMF (2 mL) and coupled with BOP (48 mg, 0.11 mmol) in the
presence of DIPEA (37 mL, 0.22 mmol) as described for the prepara-
tion of compound 3. The product was purified by HPLC and
lyophilized from water to afford a white powder (68 mg, 66%).
1H NMR (500 MHz, [D6]DMSO): d=9.24 (s, 2H), 9.01 (s, 2H), 8.35 (t,
J=6.0, 1H), 7.80 (t, J=5.6, 1H), 7.75–7.65 (m, 5H), 7.62 (t, J=5.9,
1H), 7.41 (d, J=8.4, 2H), 7.10–7.02 (m, 2H), 6.44–6.26 (m, 2H),
4.36–4.28 (m, 2H), 4.12 (dd, J=7.7, 4.5, 1H), 3.93–3.81 (m, 7H),
3.57–3.46 (m, 6H), 3.43–3.37 (m, 3H), 3.21–3.07 (m, 3H), 3.00 (dd,
J=13.4, 6.7, 2H), 2.82 (dd, J=12.5, 5.1, 1H), 2.60–2.56 (m, 1H), 2.07
(t, J=7.4, 2H), 1.90–1.24 ppm (m, 16H). 13C NMR (126 MHz,
[D6]DMSO): d=172.1, 171.5, 169.2, 169.0, 165.3, 162.7, 162.1, 158.2,
157.9, 146.0, 132.5, 128.7, 128.0, 127.0, 126.3, 113.9, 70.1, 70.0, 69.3,
69.1, 61.0, 59.2, 55.6, 55.4, 53.6, 53.5, 46.6, 41.8, 41.5, 38.4, 37.8,
36.4, 35.1, 31.3, 29.0, 28.6, 28.1, 28.0, 25.2, 24.3, 22.1 ppm. MS: m/z
(%): 916.47 [M+H]+. HPLC (start at 10% solvent B): tR =24.52 min.
4-MeO-bs-Asp(Tota-Biotinyl)-Pro-4-Amba·TFA (9): Compound 7
(100 mg, 0.1 mmol) and biotin (24 mg, 0.1 mmol) were suspended
in DMF (2 mL) and coupled by treatment with BOP (44 mg,
0.1 mmol) and DIPEA (34 mL, 0.2 mmol) as described for the prepa-
ration of compound 3. The product was purified by preparative
HPLC and lyophilized from water to afford a colorless solid (68 mg,
1
4-MeO-bs-Lys(Palmitoyl)-Pro-4-Amba·TFA (6): 4-MeO-bs-Lys-Pro-4-
Amba (1; 100 mg, 0.13 mmol) and palmitic acid (33.3 mg,
0.13 mmol) were coupled by using BOP (57.5 mg, 0.13 mmol) and
DIPEA (44 mL, 0.26 mmol) as described for the preparation of com-
pound 3. The compound was purified by preparative HPLC and
lyophilized from tert-butyl alcohol to afford a colorless solid
(91.6 mg, 78.6%). 1H NMR (500 MHz, [D6]DMSO): d=9.25 (s, 2H),
9.20–9.11 (m, 2H), 8.38 (t, J=6.1, 1H), 7.76–7.63 (m, 6H), 7.41 (d,
J=8.4, 2H), 7.10–7.04 (m, 2H), 4.35 (dd, J=16.3, 6.0, 1H), 4.29 (dd,
63%). H NMR (500 MHz, [D6]DMSO): d=9.22 (s, 2H), 9.12 (s, 2H),
8.20 (t, J=6.1, 1H), 8.14 (d, J=9.1, 1H), 8.06 (t, J=5.4, 1H), 7.74–
7.67 (m, 5H), 7.38 (d, J=8.2, 2H), 7.09 (d, J=8.9, 2H), 6.36 (d, J=
24.2, 2H), 4.37–4.21 (m, 4H), 4.16–4.10 (m, 1H), 3.89–3.81 (m, 4H),
3.63 (t, J=7.8, 1H), 3.52–3.35 (m, 13H), 3.25 (t, J=6.3, 2H), 3.12–
3.03 (m, 3H), 2.90 (dq, J=12.9, 6.6, 1H), 2.86–2.76 (m, 2H), 2.67
(dd, J=15.5, 9.8, 1H), 2.61–2.55 (m, 1H), 2.35 (dd, J=15.4, 5.0, 1H),
2.04 (t, J=7.4, 2H), 1.85–1.68 (m, 4H), 1.63–1.41 (m, 8H), 1.34–
1.24 ppm (m, 2H). 13C NMR (126 MHz, [D6]DMSO): d=171.9, 171.0,
ChemMedChem 2016, 11, 340 – 349
www.chemmedchem.org
347
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
169.4, 169.0, 165.3, 162.7, 162.2, 158.4, 158.1, 132.7, 128.5, 127.9,
126.9, 126.2, 114.0, 69.7, 69.5, 69.4, 68.1, 67.8, 61.0, 60.0, 59.2, 55.7,
55.4, 50.2, 46.4, 41.7, 35.7, 35.2, 29.4, 28.8, 28.2, 28.0, 25.2,
23.7 ppm. MS: m/z (%): 960.53 [M+H]+. HPLC (start at 10% sol-
vent B): tR =26.03 min.
percentage relative to a control without inhibitor) were plotted
against the avidin concentration.
Preparation of antithrombotic liposomes
Liposomes containing thrombin inhibitors 4 and 6 were prepared
by using a thin-film hydration method as described previously.[16,17]
The membrane was composed of a mixture of DPPC, CH, and pal-
mitoylated inhibitors 4 and 6 in a ratio of 65:30:5 mol%. Stock sol-
utions of the lipids and inhibitors were prepared in a mixture of
chloroform/methanol (2:1 v/v). For the preparation of the anticoa-
gulant liposomes, the components were mixed from the stock sol-
utions in a round-bottomed flask to a total amount of 10 mg lipid
and inhibitor. The lipid and inhibitor mixture was dried to a thin
film by using a rotary evaporator under vacuum at 408C. The re-
sulting film was rehydrated with phosphate-buffered saline (PBS,
1 mL), pH 7.4 (0.15m). After vigorous shaking, the lipid dispersions
were sonicated in a bath-type sonicator at 558C for 2 min and
stored by 48C. This method provided liposomes with an anticoagu-
lant surface. Bifunctionalized liposomes used for the neutralization
assay with avidin were obtained in the same manner by the addi-
tion of biotinylated compound 13 in the lipid and inhibitor mix-
ture. The concentration of compound 13 in the mixture was 10%
of inhibitor 4 or 6, respectively. Additional bifunctionalized lipo-
somes with increasing concentrations of biotin derivative 13 (20,
50, and 100%) were prepared from inhibitor 4.
Biological evaluation
Enzyme kinetic measurements: Kinetic measurements were per-
formed with a Fluoroskan Ascent Microplate reader (Thermo Fisher
Scientific Inc.) by using 50 mm Tris·HCl buffer (pH 8.0 at room tem-
perature, containing 0.1m NaCl and the inhibitor, 100 mL), substrate
(20 mL) diluted with water and enzyme (20 mL) (total assay volume
140 mL). Measurements with bovine a-thrombin (prepared accord-
ing to Walsmann,[30] 15 pm in assay) and uPA (Medac, ꢀ1.7 nm in
assay) were performed with the substrate Tos-Gly-Pro-Arg-AMC (KM
for thrombin: 4 mm; KM for uPA: 41 mm), whereas Mes-d-Ser(Bzl)-
Phe-Arg-AMC (Mes=methylsulfonyl, Bzl=benzyl) was used for
human plasmin (Calbiochem, ꢀ350 pm in assay, KM =25 mm) and
Mes-d-Arg-Pro-Arg-AMC[31] was used for factor Xa (FXa; Enzyme Re-
search South Bend, IN, USA, ꢀ100 pm in assay, KM =22 mm). In all
measurements, linear progress curves were observed. The Ki values
of the inhibitors summarized in Tables 1 and 2 were calculated
from Dixon plots.
The inhibitory potency of the prepared liposomes against throm-
bin, given as IC50 values, was determined as described above by
using a single substrate concentration of 11 mm Tos-Gly-Pro-Arg-
AMC. The data shown in the IC50 curves (Figure 4) were fitted to
the three-parameter Equation (1),[32] in which v is the steady-state
velocity at different inhibitor concentrations, v0 is the constant ve-
locity in the absence of inhibitor, I is the inhibitor concentration,
and s is a slope factor. The inhibitor concentration on the x axis
was calculated on the basis of the total inhibitor amount used for
the preparation of the liposomes, assuming that it was completely
incorporated and distributed on the surface.
Physicochemical characterization of liposomes
The physicochemical properties of the liposomes including size,
charge, shape, and morphology were investigated by using dy-
namic light scattering (DLS), laser Doppler velocimetry (LDV), and
atomic force microscopy (AFM).
Dynamic light scattering: The hydrodynamic diameter of lipid for-
mulations, prepared and stored as described above, were deter-
mined by DLS by using a Zetasizer Nano ZS (Malvern Instruments,
Herrenberg, Germany) equipped with a 10 mW HeNe laser at
a wavelength of 633 nm at 258C, as described previously.[33] Scat-
tered light was detected at an angle of 1738 with laser attenuation
and measurement position adjusted automatically by the Malvern
software. Values given are the meansÆstandard deviation of three
independent experiments with each experiment comprising three
measurements of the same sample with at least 10 runs, as deter-
mined by the Zetasizer. The average value was calculated with the
volume distribution data of the samplesÆstandard deviation.
Laser Doppler velocimetry: The z potential was measured with
a Zetasizer Nano ZS (Malvern Instruments, Herrenberg, Germany)
at 258C and a scattering angle of 178 by measuring electrophoretic
mobility with LDV. Values given are the meansÆstandard deviation
of three independent experiments with each experiment compris-
ing three measurements of the same sample with at least 10 runs,
as determined by the Zetasizer. The average value was calculated
with data of the samplesÆstandard deviation.
v0
ꢀ
ꢁ
v ¼
s
I
ð1Þ
1 þ
IC50
Anticoagulant activity in plasma: The aPTT measurements were per-
formed in a Thrombotrack 8 (Immuno GmbH, Heidelberg, Germa-
ny) by using a commercial test kit (Diagnostica Stago, Düsseldorf,
Germany). Platelet-poor plasma (100 mL, German Red Cross, Frank-
furt, prior incubated at 378C), a mixture of phospholipid cephalin
and contact activator Kaolin (100 mL), and a inhibitor- or liposome-
containing solution in 50 mm Tris-buffer (pH 7.8 containing 0.1m
NaCl, 80 mL) was incubated at 378C for 6 min and clotting was initi-
ated by adding 0.02m CaCl2 (100 mL) prewarmed to 378C. The de-
termined clotting times were plotted as a function of the inhibitor
concentrations.
Neutralization of bifunctional liposomes by avidin: Avidin-based neu-
tralization of the thrombin inhibitory potency was performed in
microtiter plates in the same buffer as that described above for
the enzyme kinetic measurements by using 15 pm thrombin and
11 mm Tos-Gly-Pro-Arg-AMC in the assay. The amounts of liposomes
and inhibitors used were adjusted to provide approximately 90%
thrombin inhibition. Briefly, the substrate (20 mL) dissolved in
water, liposomes or inhibitors in buffer (50 mL), and avidin (50 mL,
0.055–3.6 mm in assay, avidin obtained from Calbiochem) in buffer
were mixed and the measurement was started by the addition of
thrombin (20 mL). The determined residual thrombin activities (in
Atomic force microscopy: The lipid formulations (20 mL) were
transferred onto a silicon chip and left to dry. AFM was performed
with a vibration-damped Nanoscope IV Bioscope (Veeco Instru-
ments, Mannheim, Germany) as described in detail elsewhere.[34]
Commercial pyramidal Si3N4 tips (NSC16 AlBS, Micromash, Estonia)
were mounted on a cantilever (length 230 mm, resonance frequen-
cy 170 kHz, and nominal force constant 40 NmÀ1) and measure-
ments were performed in tapping mode to minimize damage to
ChemMedChem 2016, 11, 340 – 349
www.chemmedchem.org
348
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[17] H. Bakowsky, T. Richter, C. Kneuer, D. Hoekstra, U. Rothe, G. Bendas, C.
Ehrhardt, U. Bakowsky, Biochim. Biophys. Acta, Gen. Subj. 2008, 1778,
242–249.
[18] M. Petitou, V. Nancy-Portebois, G. Dubreucq, V. Motte, D. Meuleman, M.
de Kort, C. A. A. van Boeckel, G. M. T. Vogel, J. A. J. Wisse, Thromb. Hae-
mostasis 2009, 102, 804–810.
the specimens. The scan speed was proportional to the scan size
and the scan frequency was between 0.5 and 1.5 Hz. Images were
obtained by displaying the amplitude signal of the cantilever in
the trace direction, and the height signal in the retrace direction,
both signals being simultaneously recorded. The results were vi-
sualized either in height or in amplitude mode.
[19] M. Reers, R. Koschinsky, G. Dickneite, D. Hoffmann, J. Czech, W. Stüber,
J. Enzyme Inhib. 1995, 9, 61–72.
[20] M. Petitou, C. A. van Boeckel, Angew. Chem. Int. Ed. 2004, 43, 3118–
3133; Angew. Chem. 2004, 116, 3180–3196.
[21] S. T. Olson, R. Swanson, M. Petitou, Blood 2012, 119, 2187–2195.
[22] S. Shrivastava, T. Bera, S. K. Singh, G. Singh, P. Ramachandrarao, D. Dash,
ACS Nano 2009, 3, 1357–1364.
[23] D. Peters, M. Kastantin, V. R. Kotamraju, P. P. Karmali, K. Gujraty, M. Tirrell,
E. Ruoslahti, Proc. Natl. Acad. Sci. USA 2009, 106, 9815–9819.
[24] J. Myerson, L. He, G. Lanza, D. Tollefsen, S. Wickline, J. Thromb. Haemo-
stasis 2011, 9, 1292–1300.
[25] N. Vyavahare, N. A. Scott, S. R. Hanson, J. Kohn, J. Controlled Release
1993, 27, 165–173.
[26] S. R. Hanson, L. A. Harker, Proc. Natl. Acad. Sci. USA 1988, 85, 3184–
3188.
[27] J. W. Myerson, L. He, J. S. Allen, T. Williams, G. Lanza, D. Tollefsen, S. Car-
uthers, S. Wickline, Nanotechnology 2014, 25, 395101.
[28] R. U. Palekar, J. W. Myerson, P. H. Schlesinger, J. E. Sadler, H. Pan, S. A.
Wickline, Mol. Pharm. 2013, 10, 4168–4175.
Keywords: anticoagulants
peptidomimetics · thrombin
·
inhibitors
·
liposomes
·
[1] M. Werle, A. Bernkop-Schnurch, Amino Acids 2006, 30, 351–367.
[2] J. S. Davies, J. Pept. Sci. 2003, 9, 471–501.
[3] V. M. Ahrens, K. Bellmann-Sickert, A. G. Beck-Sickinger, Future Med.
Chem. 2012, 4, 1567–1586.
[4] L. Pollaro, C. Heinis, MedChemComm 2010, 1, 319–324.
[5] B. Elsadek, F. Kratz, J. Controlled Release 2012, 157, 4–28.
[6] A. J. Sytkowski, E. D. Lunn, K. L. Davis, L. Feldman, S. Siekman, Proc. Natl.
Acad. Sci. USA 1998, 95, 1184–1188.
[7] A. Schweinitz, T. Steinmetzer, I. J. Banke, M. J. Arlt, A. Stürzebecher, O.
Schuster, A. Geissler, H. Giersiefen, E. Zeslawska, U. Jacob, A. Krüger, J.
Stürzebecher, J. Biol. Chem. 2004, 279, 33613–33622.
[8] J. Hauptmann, T. Steinmetzer, H. Vieweg, P. Wikstrçm, J. Stürzebecher,
Pharm. Res. 2002, 19, 1027–1033.
[9] T. Steinmetzer, A. Schweinitz, S. Künzel, P. Wikstrçm, J. Hauptmann, J.
Stürzebecher, J. Enzyme Inhib. 2001, 16, 241–249.
[10] J. Stürzebecher, D. Prasa, J. Hauptmann, H. Vieweg, P. Wikstrçm, J. Med.
Chem. 1997, 40, 3091–3099.
[29] I. Levacheva, O. Samsonova, E. Tazina, M. Beck-Broichsitter, S. Levachev,
B. Strehlow, M. Baryshnikova, N. Oborotova, A. Baryshnikov, U. Bakow-
sky, Colloids Surf. B 2014, 121, 248–256.
[30] P. Walsmann, Pharmazie 1968, 23, 401–402.
[11] T. Steinmetzer, B. Baum, A. Biela, G. Klebe, G. Nowak, E. Bucha, Chem-
MedChem 2012, 7, 1965–1973.
[31] D. Meyer, F. Sielaff, M. Hammami, E. Bçttcher-Friebertshäuser, W. Garten,
T. Steinmetzer, Biochem. J. 2013, 452, 331–343.
[12] A. Straub, S. Roehrig, A. Hillisch, Angew. Chem. Int. Ed. 2011, 50, 4574–
4590; Angew. Chem. 2011, 123, 4670–4686.
[13] B. Baum, M. Mohamed, M. Zayed, C. Gerlach, A. Heine, D. Hangauer, G.
Klebe, J. Mol. Biol. 2009, 390, 56–69.
[14] B. Baum, L. Muley, M. Smolinski, A. Heine, D. Hangauer, G. Klebe, J. Mol.
Biol. 2010, 397, 1042–1054.
[32] K. Ukraincik, W. Piknosh, Methods Enzymol. 1981, 74, 497–508.
[33] E. E. Marxer, J. Brüssler, A. Becker, J. Schümmelfeder, R. Schubert, C.
Nimsky, U. Bakowsky, Eur. J. Pharm. Biopharm. 2011, 77, 430–437.
[34] J. Sitterberg, A. Ozcetin, C. Ehrhardt, U. Bakowsky, Eur. J. Pharm. Bio-
pharm. 2010, 74, 2–13.
[15] A. Biela, F. Sielaff, F. Terwesten, A. Heine, T. Steinmetzer, G. Klebe, J. Med.
Chem. 2012, 55, 6094–6110.
Received: October 19, 2015
[16] S. Anabousi, M. Laue, C. M. Lehr, U. Bakowsky, C. Ehrhardt, Eur. J. Pharm.
Biopharm. 2005, 60, 295–303.
Published online on December 14, 2015
ChemMedChem 2016, 11, 340 – 349
www.chemmedchem.org
349
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim