A Bisbenzamidine Phosphonate as a Janus-faced Inhibitor for Trypsin-like Serine Proteases

Article abstract of DOI:10.1002/cmdc.201500319

A hybrid approach was applied for the design of an inhibitor of trypsin-like serine proteases. Compound 16 [(R,R)- and (R,S)-diphenyl (4-(1-(4-amidinobenzylamino)-1-oxo-3-phenylpropan-2-ylcarbamoyl)phenylamino)(4-amidinophenyl)methylphosphonate hydrochloride], prepared in a convergent synthetic procedure, possesses a phosphonate warhead prone to react with the active site serine residue in a covalent, irreversible manner. Each of the two benzamidine moieties of 16 can potentially be accommodated in the S1 pocket of the target enzyme, but only the benzamidine close to the phosphonate group would then promote an irreversible interaction. The Janus-faced inhibitor 16 was evaluated against several serine proteases and caused a pronounced inactivation of human thrombin with a second-order rate constant (k_inac/K_i) of 59 500 M^-1 s^-1. With human matriptase, 16 showed preference for a reversible mode of inhibition (IC₅₀=2.6 μM) as indicated by linear progress curves and enzyme reactivation. Two fragments, two modes: We designed a Janus-faced serine protease inhibitor by merging two benzamidine fragments, which impart the molecule with either irreversible or reversible inhibitory activity. Unexpected differences in potency toward trypsin-like proteases were found; the compound exhibits remarkable inhibitory activity against human thrombin. This hybrid approach is a useful way to obtain potent and selective inhibitors.

Full text of DOI:10.1002/cmdc.201500319

DOI: 10.1002/cmdc.201500319
Communications
A Bisbenzamidine Phosphonate as a Janus-faced Inhibitor
for Trypsin-like Serine Proteases
Daniela Häußler,[a] Tamara Scheidt,[a] Marit Stirnberg,[a] Torsten Steinmetzer,[b] and
Michael Gütschow*[a]
A hybrid approach was applied for the design of an inhibitor
of trypsin-like serine proteases. Compound 16 [(R,R)- and (R,S)-
diphenyl (4-(1-(4-amidinobenzylamino)-1-oxo-3-phenylpropan-
2-ylcarbamoyl)phenylamino)(4-amidinophenyl)methylphospho-
nate hydrochloride], prepared in a convergent synthetic proce-
dure, possesses a phosphonate warhead prone to react with
the active site serine residue in a covalent, irreversible manner.
Each of the two benzamidine moieties of 16 can potentially be
accommodated in the S1 pocket of the target enzyme, but
only the benzamidine close to the phosphonate group would
then promote an irreversible interaction. The Janus-faced in-
hibitor 16 was evaluated against several serine proteases and
caused a pronounced inactivation of human thrombin with
a second-order rate constant (kinac/Ki) of 59500mÀ1 sÀ1. With
human matriptase, 16 showed preference for a reversible
mode of inhibition (IC50 =2.6 mm) as indicated by linear prog-
ress curves and enzyme reactivation.
generation of thrombin without affecting existing thrombin
levels, which are thought to be sufficient to ensure primary he-
mostasis via weak platelet activation.[5,6]
Matriptase, the eponymous enzyme of the matriptase sub-
family, is a member of the type II transmembrane serine pro-
teases (TTSPs). This enzyme processes and activates several
substrates, which themselves play critical roles in tumorigene-
sis, such as hepatocyte growth factor/scatter factor, urokinase
plasminogen activator, and protease-activated receptor 2.
Inhibition of matriptase is therefore thought to be a possible
strategy for cancer treatment.[7,8]
Matriptase-2, another member of the TTSPs, has attracted
considerable attention because of its critical role in iron ho-
meostasis.[9] Hepcidin, a hepatic peptide hormone that binds
to and down-regulates the iron transporter ferroportin, is af-
fected by matriptase-2, which cleaves hemojuvelin, the bone
morphogenetic protein (BMP) co-receptor, at the cell surface of
hepatocytes.[10] This leads to modulation of the BMP/SMAD sig-
naling pathway and to suppression of the expression of the
hepcidin-encoding gene. Accordingly, by applying matriptase-2
inhibitors, hepcidin expression will be promoted, and there-
fore, this enzyme has been considered a potential target for
the treatment of iron-overload diseases.[9,11,12]
Trypsin-like serine proteases possess primary substrate specific-
ity for basic amino acids. Several members of this family are
promising drug targets, and the first orally available inhibitors
of trypsin-like serine proteases have been approved for clinical
use.[1,2] For example, inhibitors of the coagulation proteases
thrombin and factor Xa are used as anticoagulant agents to
prevent deep-vein thrombosis and as therapeutics against
angina pectoris, stroke, and heart attack, which are major
causes of death and morbidity. Thrombin cleaves fibrinogen to
form fibrin and activates factors V, VIII, XI, XIII, as well as TAFI
and protein C.[2,3] Moreover, thrombin stimulates platelet secre-
tion and aggregation. Different types of antithrombotic drugs
have been designed.[2,4] Factor Xa is a much more specific pro-
tease than thrombin. Together with factor Va and calcium ions,
it forms the prothrombinase complex on a phospholipid sur-
face, which is responsible for the conversion of prothrombin
into thrombin. Thus, inhibitors of factor Xa attenuate the
The benzamidine moiety represents an arginine mimetic
which has been used extensively as a pharmacophore prone
to interact with protein regions bearing anionic and hydrogen
bond acceptor groups. In particular, benzamidines have been
frequently incorporated as P1 fragments of inhibitors for tryp-
sin-like serine proteases.[13] In the protease–inhibitor com-
plexes, the benzamidine groups were shown to be deeply
buried in the S1 specificity pocket.[14,15] Substrate analogue in-
hibitors can be derived from the particular substrate structure,
for example, from P3 to P1, by incorporating the benzamidine
moiety into the P1 side chain and removing the carboxyl
group of the P1 amino acid. Hence, the altered structure of
the N-terminal product of proteolytic cleavage leads to an im-
proved binding affinity for the active site. Several so-designed
peptidomimetics have been reported as inhibitors for trypsin-
like serine proteases, such as thrombin, factor Xa, matriptase,
and matriptase-2.[12,16,17] The structure of the reversible direct
thrombin inhibitor melagatran (1), a major milestone to this
structural class of antithrombotics, is shown in Figure 1.
[a] D. Häußler, T. Scheidt, Dr. M. Stirnberg, Prof. Dr. M. Gütschow
Pharmaceutical Institute, Pharmaceutical Chemistry I
University of Bonn, An der Immenburg 4, 53121 Bonn (Germany)
E-mail: guetschow@uni-bonn.de
[b] Prof. Dr. T. Steinmetzer
Institute of Pharmaceutical Chemistry
Philipps University Marburg, Marbacher Weg 6, 35032 Marburg (Germany)
a-Aminoalkylphosphonate diphenyl esters are phosphonic
analogues of naturally occurring amino acids. Their peptidyl
derivatives, such as the thrombin inhibitor 2 (Figure 1),[18] have
been reported as inactivators of serine proteases,[19,20] or activi-
ty-based probes.[21–23] The irreversible mode of action of such
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201500319: general methods and materials,
details of syntheses, analytical data, NMR spectra, LC–MS data, descriptions
of enzyme assays, inhibition of thrombin by melagatran, reactivation ex-
periments, and protein MS data.
ChemMedChem 2015, 10, 1641 – 1646
1641
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
lack of activity against cysteine or threonine proteases is an
advantage over other inhibitor classes such as chloromethyl
ketones.[19] The selectivity of the phosphonate esters toward
trypsin-like serine proteases can be enhanced by introducing
residues of basic amino acids such as arginine, lysine, homo-
lysine, and ornithine in the P1 position, or likewise, the benz-
amidine pharmacophore.[18,19,23,24]
Fragments 3 and 4 represent hypothetical prototypes of
either substrate analogue inhibitors (3) or irreversible inhibitors
(4) for trypsin-like proteases. In both cases, the benzamidine
moiety would be expected to be accommodated in the S1
specificity pocket of the protease. We performed a hybrid
design approach by combining both fragments to a single
inhibitor structure.[25,26] One phenyl ring present in both 3 and
4 (highlighted) was merged, and the resulting bisbenzamidine
phosphonate was synthesized and evaluated at the aforemen-
tioned proteases, i.e., thrombin, factor Xa, matriptase, and
matriptase-2. Inhibition of the reference enzyme trypsin as well
as leukocyte elastase and chymotrypsin was also examined.
This study is aimed at investigating the biological activity and
selectivity profile of such a Janus-faced molecule.
Figure 1. Structures (1, 2) or fragments (3, 4) of substrate analogue inhibi-
tors (1, 3) or irreversible inhibitors (2, 4) for trypsin-like serine proteases.
phosphonates relies on the nucleophilic attack of the active
serine at the phosphorus atom and simultaneous release of
one phenolic residue. The resulting serine phosphono diester
is known to undergo a slow ‘aging’ upon loss of the second
phenoxy group to form a phosphono monoester. A complete
The inhibitor 16 was produced in a convergent seven-
step synthetic route outlined in Scheme 1. To generate the
Scheme 1. Synthesis of the bisbenzamidine phosphonate 16. Reagents and conditions: a) Mg(ClO4)2, THF, reflux, 80 h, 55%; b) NMM, ClCO2iBu, TEA, THF,
À258C!RT, overnight, 98%; c) TFA, CH2Cl2, 1 h, RT; d) HATU, DIPEA, DMF, RT, 16 h, 89%; e) H2N-OH·HCl, DIPEA, EtOH, reflux, 2 h, 61%; f) Ac2O, AcOH, RT;
g) Pd/C/H2, AcOH, 275 kPa, RT, 2 h; h) preparative HPLC, lyophilization, 9%.
ChemMedChem 2015, 10, 1641 – 1646
www.chemmedchem.org
1642
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
a-aminophosphonate 8,
a
Kabachnik–Fields reaction was
the inhibitory activity toward the seven proteases, as: 1) an
active-site-directed mode of interaction was supposed, and
2) the substrate concentrations used were in the range of the
corresponding KM values (Table 1). Weak inhibition was found
for factor Xa, elastase, and trypsin. In contrast, we observed
a relatively strong inhibition of thrombin, and a moderate de-
crease in the activity of matriptase, matriptase-2, and chymo-
trypsin. These differences were unexpected and might result
from different binding modes. The bisbenzamidine 16 might
either occupy the S1 pocket with the benzamidine from its
irreversible inhibitor part or its substrate analogue inhibitor seg-
ment. In the first binding mode, inactivation would be facilitat-
ed, because the aforementioned benzamidine would direct the
electrophilic phosphonate close to the nucleophilic serine resi-
due. In the alternative binding mode, noncovalent inhibition
would occur. As a matter of course, each interaction might
lead to both situations and hence to a mixture of two different
enzyme–inhibitor complexes. Nevertheless, the different
shapes of the progress curves indicate a preference for either
irreversible or substrate analogue inhibition. For example, as
depicted in Figure 2, the reaction of 16 with thrombin resulted
in time-dependent inhibition, and the second-order rate con-
stant of inactivation, kinac/Ki, could be determined. This was
also the case for matriptase-2 and trypsin (Table 1). For the
other enzymes, the progress curves did not follow an expo-
nential equation and were instead rather linear, as shown for
the inhibition of matriptase in Figure 3. Such a behavior was
also observed for factor Xa, elastase, and chymotrypsin.[31] In
a control experiment, the established inhibitor melagatran
1 (Figure 1) was inspected toward thrombin. Linear time-inde-
pendent inhibition was observed, in agreement with the
known noncovalent binding mode, and a Ki value of 2.4 nm
was determined (lit.: Ki =4 nm).[15] The data are shown in
Figure S1 (Supporting Information).
chosen. This involves a three-component reaction between 4-
cyanobenzaldehyde (7), 4-aminobenzoic acid (6), and triphe-
nylphosphite (5). The conversion is usually promoted by Lewis
or Brønsted acids.[27] In our case, Mg(ClO4)2 was used, which
was reported to be an efficient catalyst for the formation of a-
aminophosphonates.[28] Compound 8, a diphenyl phosphonate
containing a nitrile functionality on the one hand and a carbox-
ylic acid moiety on the other hand, was obtained in good
yield. This building block represents the irreversible inhibitor
part of the desired final product.
The substrate analogue inhibitor segment of the target mole-
cule 16 was synthesized as follows: 4-(Aminomethyl)benzo-
nitrile (9) and Boc-protected phenylalanine (10) were coupled
by the mixed-anhydride method using isobutyl chloroformate
and N-methylmorpholine to yield the amino acid amide 11. To
remove the N-protecting group, a mixture of dichloromethane
and trifluoroacetic acid was used. The salt 12 was directly sub-
jected to a coupling reaction with the a-aminophosphonate 8
produced before. By applying HATU as reagent and DIPEA as
base,[22,29] the bisnitrile phosphonate 13 was obtained.
To simultaneously convert the two nitrile moieties of com-
pound 13 into amidines, the method of Judkins et al. was
used.[30] Hydroxylamine was reacted to form amidoxime 14.
The following treatment with acetic anhydride furnished acet-
amidoxime 15. After a palladium/carbon-catalyzed reduction,
the obtained crude product was purified by preparative HPLC
to give the final bisbenzamidine phosphonate 16.
Compound 16 was evaluated at selected proteases; the ki-
netic parameters are listed in Table 1. We used fluorogenic and
chromogenic peptide substrates for these assays. Because dif-
ferent shapes of the progress curves were observed, an estima-
tion of the respective inhibitory capacity was based on product
formation after 15 min in the presence of five different inhibi-
tor concentrations, and the IC50 values were obtained on the
basis of these end-point determinations. The IC50 values were
considered to be meaningful for a comparative assessment of
From these findings, it was concluded that thrombin prefer-
entially forms a covalent enzyme–inhibitor complex with 16.
To illustrate the resulting irreversible mode of interaction,
Table 1. Protease inhibition by bisbenzamidine phosphonate 16: assay conditions and parameters of inhibition.
[c]
Enzyme
Substrate
[S] [mm][a]
K
M [mm]
IC50 [mm][b]
k ]
inac/Ki [mÀ1 sÀ1
Human thrombin
Cbz-Gly-Gly-Arg-AMC
Mes-d-Arg-Pro-Arg-AMC
Boc-Gln-Ala-Arg-AMC
Boc-Ile-Glu-Gly-Arg-AMC
MeO-Suc-Ala-Ala-Pro-Val-pNA
Boc-Gln-Ala-Arg-AMC
40
10
40
100
100
40
40[d]
0.075Æ0.019
2.6Æ0.1
59500Æ3800
ND[f]
Human matriptase
Human matriptase-2
Bovine factor Xa
Human neutrophil elastase
Bovine trypsin
8.00Æ0.65[e]
32[d]
1.5Æ0.5
1060Æ10
59.0Æ3.2[g]
120Æ10
ND[f]
54[h]
23Æ7
ND[f]
22[d]
9.8Æ1.1
0.85Æ0.08
121Æ11
Human chymotrypsin
Suc-Ala-Ala-Pro-Phe-AMC
40
15.9Æ1.7[i]
ND[f]
[a] Substrate concentration. [b] IC50 values were obtained from duplicate measurements in the presence of five different inhibitor concentrations. Reactions
were followed over 15 min, and the formation of the product was plotted versus inhibitor concentration, [I]. IC50 values were determined by nonlinear re-
gression using equation [P]=[P0]/(1+[I]/IC50), in which [P] is the product concentration after 15 min in the presence of various inhibitor concentrations,
and [P0] is the product concentration after 15 min in the absence of inhibitor. [c] The reactions were followed over 45 min. The second-order rate constant
of inactivation (kinac/Ki) was obtained from linear regression of plots of the kobs values versus [I]. Under conditions of linear dependency of kobs on [I] (i.e.,
[I]!Ki (1+[S]/KM)), the equation kobs =kinac[I]/(Ki(1+[S]/KM)+[I]) simplifies to kinac/Ki =kobs/[I](1+[S]/KM), and kinac/Ki values were calculated accordingly.
[d] Data from ref. [16]. [e] The KM value was obtained by triplicate measurements at nine different substrate concentrations (2, 4, 6, 8, 10, 15, 20, 25, and
30 mm). [f] Not determined; time-dependent inhibition was not observed. [g] The KM value was obtained by triplicate measurements at 12 different sub-
strate concentrations (10, 20, 40, 60, 80, 100, 150, 200, 250, 300, 400, and 500 mm). [h] Data from refs. [12,32]. [i] The KM value was obtained by duplicate
measurements at 14 different substrate concentrations (4, 6, 8, 12, 16, 20, 40, 60, 80, 100, 120, 140, 160, and 180 mm).
ChemMedChem 2015, 10, 1641 – 1646
www.chemmedchem.org
1643
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
Figure 2. Inhibition of human thrombin by bisbenzamidine phosphonate 16.
Measurements were performed with a plate reader. From top to bottom:
uninhibited reaction, [I]=25 nm, [I]=50 nm, [I]=75 nm, [I]=100 nm,
[I]=125 nm. The first-order rate constants (kobs) were obtained by nonlinear
Figure 4. Reactivation of human matriptase, inhibited by bisbenzamidine
phosphonate 16. Reactions were performed with a plate reader. Dark grey:
the enzyme was incubated with 16 (4 mm) for 30 min in a volume of
200 mL. Assay buffer was added to a total volume of 4 mL, which then was
decreased to ~150 mL by centrifugation to remove unbound inhibitor. The
sample was filled to 1 mL, and the centrifugation was repeated. The ob-
tained volume of ~160 mL was filled to a volume of 180 mL. Finally, substrate
(20 mL, 100 mm) was added, and the product formation was assayed. The
concentration of matriptase was 0.0005 mgmLÀ1. Black: reaction in the
absence of 16. Inset: reactions were performed using 162 mL assay buffer
and 8 mL DMSO (black) or inhibitor solution in DMSO (dark grey). Matriptase
was added and was incubated at 378C for 30 min, without subsequent filtra-
tion. The reaction was initiated by the addition of substrate. Final concentra-
tions: [matriptase]=0.0005 mgmLÀ1, [S]=10 mm, [I]=4 mm.
regression of the progress curves using the equation [P]=[P ](1-exp(Àkobst),
1
for which [P] is the product concentration, and [P ] is the product concen-
1
tration at infinite time. Time-dependent inhibition was analyzed as noted in
Table 1. The inset shows a plot of kobs values of duplicate measurements
versus inhibitor concentrations, [I]. Linear regression gave an apparent
second-order rate constant (kobs/[I]) of 29800Æ1900mÀ1 sÀ1
.
Figure 3. Inhibition of human matriptase by bisbenzamidine phosphonate
16. The measurements were performed with a plate reader. From top to
bottom: uninhibited reaction, [I]=1 mm, [I]=2 mm, [I]=3 mm, [I]=4 mm,
[I]=5 mm. Rates (v) were obtained by linear regression of the progress
curves. Assay conditions are noted in Table 1. The inset shows a plot of the
rates of duplicate measurements versus inhibitor concentrations, [I].
Nonlinear regression using the equation v=v0/(1+([I]/IC50)) gave an IC50
value of 2.6Æ0.1 mm.
Figure 5. Reactivation of human thrombin, inhibited by bisbenzamidine
phosphonate 16. The reactions were performed with a plate reader. Dark
grey: the enzyme was incubated with 16 (125 nm) for 30 min in a volume of
200 mL. Assay buffer was added to a total volume of 4 mL, which was then
decreased to ~150 mL by centrifugation to remove unbound inhibitor. The
sample was filled to 1 mL, and the centrifugation was repeated. The
obtained volume of ~160 mL was filled to a volume of 190 mL. Finally, the
substrate (10 mL, 800 mm) was added, and product formation was assayed.
The concentration of thrombin was 2 UmLÀ1. Black: reaction in the absence
of 16. Inset: reactions were performed using 160 mL assay buffer and 10 mL
DMSO (black) or inhibitor solution in DMSO (dark grey). Thrombin was
added and was incubated at 258C for 30 min, without subsequent filtration.
The reaction was initiated by the addition of substrate. Final concentrations:
[thrombin]=2 UmLÀ1, [S]=40 mm, [I]=125 nm.
reactivation experiments were performed with thrombin and
compared with matriptase, for which a preferred noncovalent
interaction was assumed. Both enzymes were incubated with
16 at a concentration of ~2”IC50. After 30 min, the samples
were centrifuged to remove unbound inhibitor. The appropri-
ate substrate was added to each enzyme concentrate and
product formation was monitored.
Whereas matriptase recovered activity (Figure 4), thrombin
was shown to remain inactivated by 16 (Figure 5). A further
ChemMedChem 2015, 10, 1641 – 1646
www.chemmedchem.org
1644
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
genated at 275 kPa for 2 h. The final product 16 was initially puri-
fied by preparative reversed-phase HPLC. A few drops of 1n HCl
were added to form the salt 16, which was obtained after lyophili-
zation.
independent experiment was carried out in which both pro-
teases were inhibited with 16, and after a significant increase
in the substrate concentration, subsequent product formation
was compared for the inhibited and non-inhibited enzyme. In
this case, we observed only marginally enhanced thrombin ac-
tivity (Supporting Information Figure S2), whereas matriptase
was reactivated to a significant extent (Figure S3). These results
indicate that the favored accommodation of 16 differs at the
active sites of thrombin and matriptase.
Acknowledgements
D.H. holds a fellowship from the Bonn International Graduate
School of Drug Sciences (BIGS DrugS). D.H. was also supported
by a fellowship from the NRW International Graduate Research
School Biotech-Pharma. M.S. is supported by the German
Research Foundation (STI 660/1-1). The authors thank Anna-
Christina Schulz-Fincke and Jens Fuchs, both University of
Bonn, for assistance. The authors also thank Dr. Marc Sylvester
(Mass Spectrometry Unit, IBMB, University of Bonn) for per-
forming MS experiments.
The covalent modification of thrombin by 16 was further
supported by MS peptide map fingerprinting of a tryptic
digest of the excised SDS-PAGE protein band after incubation
of thrombin (20 UmLÀ1) with 16 (1.25 mm) for 30 min. Thus,
compared with the activity assay, a tenfold concentration of
the enzyme and the inhibitor was chosen. Two different pep-
tide fragments of 15 or 16 amino acids containing the active
site serine were obtained after tryptic digestion. In the course
of the LC–MS sample preparation, methionine was oxidized to
the sulfone, and cysteine was alkylated by iodoacetamide. The
peptides originating from thrombin treated with compound
16 showed two additional modifications with high peak inten-
sity. The first mass difference (685.2566) was due to the forma-
tion of the serine phosphono diester, and the second differ-
ence (609.2253) was due to the subsequent ‘aging‘, that is, hy-
drolysis to the phosphono monoester (Supporting Information
Table S1). Hence, in the sample resulting from incubation with
16, the peptides were predominately modified by the bisbenz-
amidine phosphonate, confirming the covalent modification of
thrombin by 16.
In conclusion, we have designed the Janus-faced bisbenz-
amidine 16 by merging two benzamidine fragments. The frag-
ments introduce the capacity of either irreversible or reversible
inhibition into the molecule. Unexpected differences in the po-
tency of 16 toward trypsin-like protease were found. This com-
pound exhibited remarkable inhibitory activity against human
thrombin, with a second-order rate constant of inactivation of
59500mÀ1 À1. Thus, our hybrid approach appears to be useful
s
for obtaining potent and even selective inhibitors and will be
extended by combining other benzamidine fragments in
future studies.
Keywords: benzamidines
·
enzyme inhibition
·
hybrid
compounds · phosphonates · serine proteases
[1] C. H. Yeh, K. Hogg, J. I. Weitz, Arterioscler. Thromb. Vasc. Biol. 2015, 35,
1056–1065.
[2] For a review, see D. Gustafsson, R. Bylund, R. Antonsson, I. Nilsson, J. E.
Nystrçm, U. Eriksson, U. Bredberg, A. C. Teger-Nilsson, Nat. Rev. Drug
Discovery 2004, 3, 649–659.
[3] L. O. Mosnier, J. Biol. Chem. 2011, 286, 502–510.
[4] For reviews, see: a) T. Steinmetzer, J. Stürzebecher, Curr. Med. Chem.
´
ˇ
2004, 11, 2297–2321; b) M. Ilic, D. Kikelij, J. Ilas, Curr. Top. Med. Chem.
2011, 11, 2834–2848.
[5] For reviews on thrombin and factor Xa inhibitors, see: a) B. I. Eriksson,
D. J. Quinlan, J. W. Eikelboom, Annu. Rev. Med. 2011, 62, 41–57; b) D. J.
Pinto, J. M. Smallheer, D. L. Cheney, R. M. Knabb, R. R. Wexler, J. Med.
Chem. 2010, 53, 6243–6274; c) A. Straub, S. Roehrig, A. Hillisch, Angew.
Chem. Int. Ed. 2011, 50, 4574–4590; Angew. Chem. 2011, 123, 4670–
4686.
[6] S. Roehrig, A. Straub, J. Pohlmann, T. Lampe, J. Pernerstorfer, K. H.
Schlemmer, P. Reinemer, E. Perzborn, J. Med. Chem. 2005, 48, 5900–
5908.
[7] For a review, see K. List, Future Oncol. 2009, 5, 97–104.
[8] a) T. Steinmetzer, A. Schweinitz, A. Stürzebecher, D. Dçnnecke, K.
Uhland, O. Schuster, P. Steinmetzer, F. Müller, R. Friedrich, M. E. Than, W.
Bode, J. Stürzebecher, J. Med. Chem. 2006, 49, 4116–4126; b) M. Ham-
mami, E. Rühmann, E. Maurer, A. Heine, M. Gütschow, G. Klebe, T. Stein-
metzer, MedChemComm 2012, 3, 807–813.
[9] For reviews, see: a) M. Stirnberg, M. Gütschow, Curr. Pharm. Des. 2013,
19, 1052–1061; b) C. Y. Wang, D. Meynard, H. Y. Lin, Front. Pharmacol.
2014, 19, 114–129; c) D. Meynard, J. L. Babitt, H. Y. Lin, Blood 2014, 123,
168–176.
Experimental Section
[10] L. Silvestri, A. Pagani, A. Nai, I. De Domenico, J. Kaplan, C. Camaschella,
Cell Metab. 2008, 8, 502–511.
[11] M. Stirnberg, E. Maurer, A. Horstmeyer, S. Kolp, S. Frank, T. Bald, K.
Arenz, A. Janzer, K. Prager, P. Wunderlich, J. Walter, M. Gütschow, Bio-
chem. J. 2010, 430, 87–95.
[12] M. T. Sisay, T. Steinmetzer, M. Stirnberg, E. Maurer, M. Hammami, J. Ba-
jorath, M. Gütschow, J. Med. Chem. 2010, 53, 5523–5535.
[13] For examples, see refs. [8,12,16,17].
[14] M. Renatus, W. Bode, R. Huber, J. Stürzebecher, M. T. Stubbs, J. Med.
Chem. 1998, 41, 5445–5456.
[15] F. Dullweber, M. T. Stubbs, D. Musil, J. Stürzebecher, G. Klebe, J. Mol.
Biol. 2001, 313, 593–614.
[16] S. Dosa, M. Stirnberg, V. Lülsdorff, D. Häußler, E. Maurer, M. Gütschow,
Bioorg. Med. Chem. 2012, 20, 6489–6505.
[17] a) 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; b) G. De Nanteuil,
Synthesis: Compound 8 was synthesized in a one-pot reaction of
5 (1 equiv), 6 (1 equiv), and 7 (1 equiv) with Mg(ClO4)2 (0.05 equiv)
as catalyst in THF at reflux for 80 h. Boc-Phe-OH (10, 1 equiv) was
dissolved in THF and cooled to À258C. It was then coupled to 4-
(aminomethyl)benzonitrile hydrochloride (9, 1 equiv) using NMM
(1.1 equiv), ClCO2iBu (1.1 equiv), and TEA (2 equiv) at RT overnight.
The Boc protecting group was cleaved with TFA/CH2Cl2 (1:1) for
1 h at RT. Coupling of compounds 8 (1 equiv) and 12 (1 equiv) oc-
curred in DMF using HATU (1 equiv) and DIPEA (2 equiv). The mix-
ture was stirred for 16 h at RT. Amidoxime 14 was furnished by
holding 13 at reflux with H2N-OH·HCl (6 equiv) and DIPEA (6 equiv)
in EtOH for 2 h. Compound 14 (1 equiv) was dissolved in glacial
acetic acid; acetic anhydride (3 equiv) was added, and the mixture
was stirred at RT. After 2 h, the solution of the resulting acetyl-
amidoxime 15 was added to 10% palladium on carbon and hydro-
ChemMedChem 2015, 10, 1641 – 1646
www.chemmedchem.org
1645
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
P. Gloanec, S. BØguin, O. L. Giesen, H. C. Hemker, P. Mennecier, A. Rupin,
T. L. Verbeuren, J. Med. Chem. 2006, 49, 5047–5050; c) A. Schweinitz, A.
Stürzebecher, U. Stürzebecher, O. Schuster, K. Stürzebecher, T. Steinmet-
zer, Med. Chem. 2006, 2, 349–361; d) A. Stürzebecher, D. Dçnnecke, A.
Schweinitz, O. Schuster, P. Steinmetzer, U. Stürzebecher, J. Kotthaus, B.
Clement, J. Stürzebecher, T. Steinmetzer, ChemMedChem 2007, 2, 1043–
Acad. Sci. USA 2014, 111, 2518–2523; g) S. Serim, P. Baer, S. H. Verhelst,
Org. Biomol. Chem. 2015, 13, 2293–2299.
[22] E. Sabidó, T. Tarragó, E. Giralt, Bioorg. Med. Chem. 2010, 18, 8350–8355.
[23] C. M. Brown, M. Ray, A. A. Eroy-Reveles, P. Egea, C. Tajon, C. S. Craik,
Chem. Biol. 2011, 18, 48–57.
[24] a) L. M. Ni, J. C. Powers, Bioorg. Med. Chem. 1998, 6, 1767–1773; b) D. S.
Jackson, S. A. Fraser, L. M. Ni, C. M. Kam, U. Winkler, D. A. Johnson, C. J.
Froelich, D. Hudig, J. C. Powers, J. Med. Chem. 1998, 41, 2289–2301.
[25] For reviews, see: a) D. Munoz-Torrero, P. Camps, Curr. Med. Chem. 2006,
13, 399–422; b) M. Decker, Mini-Rev. Med. Chem. 2007, 7, 221–229;
c) M. C. Carreiras, E. Mendes, M. J. Perry, A. P. Francisco, J. Marco-Con-
telles, Curr. Top. Med. Chem. 2013, 13, 1745–1770.
[26] J. Wang, S. O’Sullivan, S. Harmon, R. Keaveny, M. W. Radomski, C.
Medina, J. F. Gilmer, J. Med. Chem. 2012, 55, 2154–2162, and references
therein.
ˇ
ˇ
1053; e) J. Ilas, Z. Jakopin, T. Borstnar, M. Stegnar, D. Kikelj, J. Med.
ˇ
ˇ ´
Chem. 2008, 51, 5617–5629; f) J. Ilas, T. Tomasic, D. Kikelj, J. Med. Chem.
2008, 51, 2863–2867; g) F. Sielaff, E. Bçttcher-Friebertshäuser, D. Meyer,
S. M. Saupe, I. M. Volk, W. Garten, T. Steinmetzer, Bioorg. Med. Chem.
Lett. 2011, 21, 4860–4864; h) P. Bach, L. Knerr, O. Fjellstrçm, K. Hansson,
C. Mattsson, D. Gustafsson, Bioorg. Med. Chem. Lett. 2014, 24, 821–827.
[18] J. Oleksyszyn, B. Boduszek, C. M. Kam, J. C. Powers, J. Med. Chem. 1994,
37, 226–231.
´
[19] For a review, see: M. Sienczyk, J. Oleksyszyn, Curr. Med. Chem. 2009, 16,
1673–1687.
[27] a) S. T. Disale, S. R. Kale, S. S. Kahandal, T. G. Srinivasan, R. V. Jayaram, Tet-
rahedron Lett. 2012, 53, 2277–2279; b) B. Kaboudin, K. Moradi, Tetrahe-
dron Lett. 2005, 46, 2989–2991.
[20] a) L. Cheng, C. A. Goodwin, M. F. Scully, V. V. Kakkar, G. Claeson, Tetrahe-
dron Lett. 1991, 32, 7333–7336; b) A. Belyaev, X. Zhang, K. Augustyns,
A. M. Lambeir, I. De Meester, I. Vedernikova, S. ScharpØ, A. Haemers, J.
[28] S. Bhagat, A. K. Chakraborti, J. Org. Chem. 2007, 72, 1263–1270.
´
ˇ
Med. Chem. 1999, 42, 1041–1052; c) M. Sienczyk, J. Oleksyszyn, Bioorg.
[29] M. D. Mertens, J. Schmitz, M. Horn, N. Furtmann, J. Bajorath, M. Mares,
Med. Chem. Lett. 2006, 16, 2886–2890; d) Ł. Winiarski, J. Oleksyszyn, M.
M. Gütschow, ChemBioChem 2014, 15, 955–959.
[30] B. D. Judkins, D. G. Allen, T. A. Cook, B. Evans, T. E. Sardharwala, Synth.
Commun. 1996, 26, 4351–4367.
[31] The unexpected chymotrypsin inhibition might be attributed to an ac-
commodation of the substrate analogue inhibitor part of 16 in the S1
pocket.
´
Sienczyk, J. Med. Chem. 2012, 55, 6541–6553.
[21] a) E. Sabidó, T. Tarragó, S. Niessen, B. F. Cravatt, E. Giralt, ChemBioChem
2009, 10, 2361–2366; b) S. Nickel, F. Kaschani, T. C. Colby, R. A. van der
Hoorn, M. Kaiser, Bioorg. Med. Chem. 2012, 20, 601–606; c) S. Serim,
S. V. Mayer, S. H. Verhelst, Org. Biomol. Chem. 2013, 11, 5714–5721; d) R.
Grzywa, E. Burchacka, M. Łe˛cka, Ł. Winiarski, M. Walczak, A. Łupicka-
Słowik, M. Wysocka, T. Burster, K. Bobrek, K. Csencsits-Smith, A. Lesner,
[32] M. Gütschow, M. Pietsch, A. Themann, J. Fahrig, B. J. Schulze, J. Enzyme
Inhib. Med. Chem. 2005, 20, 341–347.
´
M. Sienczyk, ChemBioChem 2014, 15, 2605–2612; e) U. R. Haedke, S. C.
Frommel, F. Hansen, H. Hahne, B. Kuster, M. Bogyo, S. H. Verhelst, Chem-
BioChem 2014, 15, 1106–1110; f) P. Kasperkiewicz, M. Poreba, S. J.
Snipas, H. Parker, C. C. Winterbourn, G. S. Salvesen, M. Drag, Proc. Natl.
Received: July 21, 2015
Published online on August 25, 2015
ChemMedChem 2015, 10, 1641 – 1646
www.chemmedchem.org
1646
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim