Diphenyl phosphonate inhibitors for the urokinase-type plasminogen activator: Optimization of the P4 position

Article abstract of DOI:10.1021/jm060622g

This paper describes the structure-activity relationship in a series of tripeptidyl diphenyl phosphonate irreversible urokinase plasminogen activator (uPA) inhibitors, originally derived from an arginyltripeptide. uPA is considered an interesting target i

Full text of DOI:10.1021/jm060622g

J. Med. Chem. 2006, 49, 5785-5793
5785
Diphenyl Phosphonate Inhibitors for the Urokinase-Type Plasminogen Activator: Optimization
of the P4 Position
Jurgen Joossens,† Pieter Van der Veken,† Georgiana Surpateanu,† Anne-Marie Lambeir,‡ Ibrahim El-Sayed,† Omar M. Ali,†
Koen Augustyns,*,† and Achiel Haemers†
Laboratory of Medicinal Chemistry and Laboratory of Medical Biochemistry, UniVersity of Antwerp, UniVersiteitsplein 1, BE-2610 Antwerp,
Belgium
ReceiVed May 24, 2006
This paper describes the structure-activity relationship in a series of tripeptidyl diphenyl phosphonate
irreversible urokinase plasminogen activator (uPA) inhibitors, originally derived from an arginyltripeptide.
uPA is considered an interesting target in anticancer drug design. The selectivity of these inhibitors for uPA
is enhanced by the optimization of the P4 position. The most interesting compound shows an IC50 of 5 nM,
with a selectivity index of more than 3000 toward other Arg/Lys-specific proteases such as tissue-type
plasminogen activator, plasmin, factor Xa, and thrombin. A synthetic strategy for the preparation of small
libraries of diphenyl phosphonate analogues of capped tripeptides is described. It is shown that uPA is
irreversibly inhibited, and interactions with the active site were modeled. Finally, a diparacetamol phosphonate
analogue was developed to circumvent the release of cytotoxic phenol.
Introduction
Urokinase plasminogen activator (uPA)a is a trypsin-like
serine protease and a pivotal enzyme in pericellular proteolysis.
It is produced as a pro-enzyme, pro-uPA, and activated by
plasmin and some other proteases. It is selectively bound to its
membrane-anchored receptor (uPAR) and activates plasminogen
into plasmin. Plasmin plays an important role in the breakdown
of extracellular matrix (ECM). It activates several matrix
metalloproteases, which in turn degrade several components of
the ECM including fibrin, laminin, and fibronectin. Proteolytic
degradation of this ECM is very important for tumor growth
and metastasis, and the uPA-uPAR system appears to be a key
player in the proteolytic degradation process in cancer invasion
and metastasis. Although proteolysis during cancer progession
is a very complex event, it becomes more and more evident
that plasminogen activation by uPA and the uPA-uPAR1-3
complex, together with other extracellular proteases,4 may be
an attractive target in the discovery of new antimetastasis
approaches.
Some recent data support uPA/uPAR as a valid therapeutic
target. A small synthetic amidinophenylalanine-type uPA inhibi-
tor (WX-UK1) suppresses rat breast cancer metastasis and
reduces primary tumor growth. This inhibitor is currently in
phase I/II trials.5 Another amidine-based, peptide-derived inhibi-
tor reduces the number of experimental lung metastasis in a
fibrosarcoma model in mice.6 uPA-deficient mice show reduced
metastasis of transgenic mammary cancer.7 Down regulation
of uPAR suppresses invasiveness in colon cancer cells.8
Antisense oligonucleotides for uPAR inhibit metastasis of
human melanoma9 and experimental prostate cancer bone
metastases10 in mice. The uPA and uPAR status is one of the
strongest prognostic factors in breast carcinoma.11
Figure 1. Development of the uPA inhibitors 3 from the peptidic
arginal 1 as lead compound.
Moreover, the physiological role of uPA can be taken over
by tissue plasminogen activator (tPA). tPA is also able to
activate plasminogen but is known to be mainly involved in
the fibrinolysis process. The majority of uPA-deficient mice
do not develop spontaneous phenotype. This is also true for
tPA-deficient mice but not for uPA/tPA- or plasminogen-
deficient animals.6 Hence, we could expect that targeting the
uPA system would not lead to mechanism-based toxicity.
Our research focuses on the discovery and development of
uPA inhibitors. The development of selective inhibitors for uPA
is hampered by its close resemblance to other trypsin-like serine
proteases, and several inhibitors suffer from a poor selectivity
toward tPA, plasmin, or thrombin. Because the latter enzymes
are members of the important blood coagulation cascade, the
development of highly selective uPA inhibitors is of utmost
importance. The structure of uPA is well-known, and several
* To whom correspondence should be addressed: Tel.: +32-3-8202703.
Fax: +32-3-8202739. E-mail: koen.augustyns@ua.ac.be.
† Laboratory of Medicinal Chemistry.
‡ Laboratory of Medical Biochemistry.
a Abbreviations: uPA, urokinase-type plasminogen activator; tPA, tissue-
type plasminogen activator; uPAR, urokinase-type plasminogen activator
receptor; FXa, factor Xa; ECM, extracellular matrix.
10.1021/jm060622g CCC: $33.50 © 2006 American Chemical Society
Published on Web 08/24/2006

5786 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 19
Joossens et al.
was replaced by a diphenyl phosphonate group (2), which is
known to react in a covalent way with the active site serine of
serine proteases, affording a phosphonylated and irreversibly
inhibited enzyme.17,18 Because the P1 position was not optimized
in the mentioned combinatorial library, we chose to first
optimize this position by replacing the arginine moiety by
various linear or cyclic guanidinylated side chains. A guanidi-
nylated benzyl group afforded the most interesting compound
(3a), providing high selectivity and high potency toward uPA
(IC50, 4.3 nM; selectivity toward plasmin, 470; tPA, 1000;
thrombin, 4700; trypsin, 70). To the best of our knowledge,
these diphenylphosphonate peptides were the first selective and
irreversible uPA inhibitors reported in the literature.15
In this paper we report the further optimization of this series
of phosphonate peptidic inhibitors for uPA (Figure 2). We
selected compound 3a as lead compound and started with the
optimization of the P4 postion because it was known from the
arginal analogues (1) that improvement at this position was
possible. We linked several amide (benzamide and substituted
benzamides, adamantyl) and sulfonamide (aromatic, benzylic)
groups at the terminal amino group (compounds 3b-o) and
made a small library of these compounds. Aliphatic groups were
not used, as previous investigations, confirmed by literature
data,14,16 showed less convincing results.
Finally, the release of cytotoxic phenol upon covalent binding
to the target protein constitutes a potential problem in biological
settings using diphenyl phosphonates. To circumvent this
problem, a diaryl phosphonate 4 containing a “safe” phenol,
p-acetylaminophenol (paracetamol), was prepared.
Chemistry
Diphenyl phosphonylated tripeptides were prepared by a fast
and convenient synthetic route, affording the opportunity to
obtain a diverse series of compounds. The peptide tail of the
target compounds was synthesized via standard Fmoc-solid-
phase peptide-synthesis protocols (Scheme 1). Using 2-chlo-
rotrityl resin as the solid support, the Fmoc-protected dipeptide
(D)-Ser(Ot-Bu)-Ala 6 was constructed and the N-terminus was
capped with different acyl and sulfonyl substituents, affording
resin-bound compounds 7. The capped dipeptide 8 was cleanly
cleaved from the resin using dichloromethane/methanol/acetic
acid (60:20:20), conditions not affecting the tert-butyl protecting
group on the serine side chain. The peptide tail was subsequently
coupled to the aminophosphonate intermediate 14 by means of
a polymer-assisted solution phase procedure, as reported earlier,
involving polymer-supported carbodiimide as coupling reagent
and a polyamine scavenger resin for purification.19 The protected
compounds 9 were isolated with flash chromatography, and tert-
butyl and Boc deprotection afforded compounds 3b-o.
Figure 2. Structures of the diaryl phosphonates reported in this paper.
research groups are involved in the development of reversible
inhibitors for uPA. Most of these inhibitors are characterized
by a basic function that forms a specific ionic interaction with
the carboxylate group of Asp-189 of the S1 site in the deepest
pocket of the substrate binding groove of the enzyme.6,12-14
Peptide as well as nonpeptide inhibitors were developed.
We reported recently the first irreversible, selective and potent
diphenyl phosphonate peptidic inhibitors for uPA, based on the
Z-D-Ser-Ala-Arg structure.15 These inhibitors were designed
using Z-D-Ser-Ala-Arginal (1) as lead compound (Figure 1).16
This peptidic compound was derived from a combinatorial
library with diversity at P2, P3, and P4. The aldehyde function
Scheme 1. Synthesis of the Target Compounds 3
Conditions: (a) (i) piperidine, DMF, (ii) Fmoc-D-Ser(OtBu), TBTU, TEA, DMF; (b) (i) piperidine, DMF, (ii) R′SO2Cl, collidine, DCM or R′COOH,
TBTU, TEA, DMF; (c) AcOH, MeOH, DCM; (d) (i) HOBt, resin-bound carbodiimide, DCM, (ii) resin-bound polyamine; (e) TFA.

Diphenyl Phosphonate Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 19 5787
IC50 (µM) or % inhibition at given concentrationa (µM)
Table 1. Inhibitory Activities against UPA and Related Enzymes
IC50 (nM) uPA
kapp (M-1 s-1) uPA
tPA
plasmin
thrombin
FXa
trypsin
N.D.
2
3a
3b
3c
3d
3e
3f
61 ( 1
1.77 × 103
( 0.08 × 103
60 × 103
50% @ 2.5
(∼40)
10.8 ( 0.3
(300)
2.0 ( 0.2
(470)
1.9 ( 0.1
(740)
0.037 ( 0.003
(11)
15 ( 1
(3000)
0.99 ( 0.05
(155)
1.54 ( 0.17
(1000)
10 ( 1
(1400)
5 ( 1
(600)
1.7 ( 0.3
(293)
0.9 ( 0.1
(136)
0.9 ( 0.2
(145)
5.0 ( 0.4
(700)
8 ( 1
(1800)
7.2 ( 0.5
(1400)
4.2 ( 0.2
(700)
0.50 ( 0.02
(8)
20 ( 0.3
(4700)
4.9 ( 0.3
(1900)
0.77 ( 0.03
(220)
22 ( 1
(5400)
6 ( 1
(1000)
4.3 ( 0.2
(2800)
35 ( 4
(5000)
6 ( 0.1
(750)
1.67 ( 0.04
(288)
2.98 ( 0.03
(451)
7.5 ( 0.4
(1200)
67 ( 12
(9600)
34 ( 4
(7700)
15.9 ( 0.6
(3200)
6.3 ( 0.2
(1000)
65 ( 4
(1100)
43% @ 125
4.3 ( 0.2
2.6 ( 0.3
3.5 ( 0.7
5 ( 1
4.6 ( 0.7
(1000)
0.3 ( 0.1
(70)
0.16 ( 0.2
(60)
0.31 ( 0.03
(90)
0.40 ( 0.05
(80)
0.280 ( 0.04
(50)
0.098 ( 0.006
(65)
280 ( 30
(40)
0.16 ( 0.03
(20)
0.36 ( 0.04
(60)
0.24 ( 0.02
(36)
0.44 ( 0.06
(70)
0.16 ( 0.03
(23)
0.16 ( 0.03
(40)
0.14 ( 0.02
(30)
0.22 ( 0.06
(40)
0.20 ( 0.04
(30)
( 2 × 103
94 × 103
3.1 ( 0.7
(1200)
0.28 ( 0.02
(80)
15 ( 2
(3000)
4.4 ( 0.5
(76o)
3.4 ( 0.1
(2300)
13 ( 2
(1900)
8.6 ( 0.6
(1000)
4.3 ( 0.2
(740)
3.7 ( 0.2
(560)
6.6 ( 0.5
(1100)
16 ( 1
(2300)
18 ( 3
(4000)
5.5 ( 0.4
(1100)
3.6 ( 0.4
(600)
6.4 ( 0.3
(1000)
64% @ 250
( 4 × 103
97 × 103
33 ( 5
(9400)
45% @ 250
( 7 × 103
40 × 103
( 2 × 103
46 × 103
5.8 ( 1.2
1.5 ( 0.2
6.9 ( 0.8
8 ( 1
96 ( 14
(16 551)
68% @ 250
( 2 × 103
80 × 103
( 10 × 103
71 × 103
3g
3h
3i
∼250
( 0.2 × 103
36 × 103
62% @ 250
( 1 × 103
98 × 103
5.8 ( 0.4
6.6 ( 0.5
6.2 ( 0.3
7 ( 2
100 ( 18
(19 000)
58 ( 7
(9000)
53% @ 250
( 6 × 103
45 × 103
3j
( 1 × 103
74 × 103
3k
3l
( 4 × 103
70 × 103
53% @ 250
56% @ 250
57% @ 250
62% @ 250
58% @ 250
( 10 × 103
75 × 103
3m
3n
3b
4
4.4 ( 0.6
5 ( 1
( 10 × 103
83 × 103
( 11 × 103
62 × 103
6 ( 1
( 7 × 103
49 × 103
6.7 ( 0.3
0.9 ( 0.01
(130)
1.5 ( 0.04
(200)
( 1 × 103
a Selectivity index is given in parentheses (IC50/IC50 uPA).
Scheme 2. Synthesis of the Intermediate 14
Biochemical Evaluation
The diphenyl phosphonate tripeptides were evaluated for their
ability to inhibit various trypsin-like serine proteases using the
appropriate chromogenic substrates.15 The IC50 values were
determined for uPA and five other exemplatory trypsin-like
serine proteases. Four of them are involved in the blood
coagulation cascade (tPA, plasmin, thrombin, and FXa;
Table 1).
Compound 2 is the original lead diphenyl phosphonate with
arginine at P1 showing low activity and limited selectivity.15
Potency and selectivity is enhanced when the arginine analogue
at P1 is replaced with a benzylguanidine group (3a). When the
terminal benzylcarbamate function (3a) was substituted for a
sulfonyl or acyl group, all compounds remain equipotent toward
uPA with IC50 values of about 5 nM but show varying
selectivities as a result of enhanced or reduced affinities for
plasmin, tPA, thrombin, FXa, and trypsin. We found that amide
groups in general and a benzamide group in particular at the
P4 position were the most selective compounds (3d). Increases
in selectivity of 2- to 6-fold were observed compared to that of
3a. The selectivity index toward trypsin, however, is limited to
about 80. Although rather low, this index is still far better than
the corresponding index for related peptide-like inhibitors.15
Paracetamol ester 4 was found to be a very potent inhibitor
of uPA with low nanomolar activity, comparable to the parent
compound 3a. There was, however, a small decrease in
selectivity toward plasmin and a significant decrease in selectiv-
ity toward thrombin. The selectivity profile of this acetami-
dophenyl phosphonate 4, however, can still be considered as
promising.
Conditions: (a) (i) di-tert-butyl dicarbonate, TEA, dioxane, (ii) Dess-
Martin oxidation; (b) benzyl carbamate, triphenyl phosphite, Cu(OTf)2,
DCM; (c) (i) TFA, (ii) N,N′-bis(tert-butoxycarbonyl)-1-guanylpyrazole,
MeCN; (d) H2, Pd/C.
The aminophosphonate building block 14 was prepared from
tert-butylcarbamate-protected 4-aminophenylacetaldehyde (11),
prepared from the corresponding alcohol (10) with Dess-Martin
periodinane, as described earlier15 (Scheme 2). For the ami-
doalkylation, we used a previously developed procedure with
copper triflate as catalyst, affording N-benzyloxycarbonyl-
protected diphenyl phosphonate 12.20 These softer conditions
allow the use of Boc protecting groups, which is not possible
with the standard acetic acid conditions. Acidolysis removed
the tert-butyl carbamate protecting group. N,N′-bis(tert-butoxy-
carbonyl)-1-guanylpyrazole was used to introduce the protected
guanidine group. Compound 13 was subsequently deprotected
under hydrogenolytic conditions (Scheme 2). The p-acetylami-
nophenyl phosphonate 4 was prepared analogously using tri-
p-acetylaminophenyl phosphite.18

5788 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 19
Joossens et al.
Figure 3. Inhibitors 3a and 5-7 in the active site of uPA with ionic- and hydrogen bonds (carbons of inhibitors, green; carbons of uPA, gray;
nitrogen, blue; oxygen, red; sulfur, yellow; phosphorus, orange). Compound 3a was modeled as a flexible ligand in a flexible uPA pocket.
The hypothesis that these compounds are indeed irreversible
inhibitors of uPA is supported by their kinetic behavior. The
inhibition was time-dependent and the rate of inactivation
increased with the inhibitor concentration. When the reaction
was allowed to proceed to completion, the enzyme was fully
inhibited, even at nM concentrations. The inactivation rate
decreased in the presence of substrate, indicating competition
between the inhibitors and the substrate for the enzyme’s active
site. Lowering the inhibitor concentration by diluting the
inhibited enzyme with substrate solution did not cause a rise in
uPA activity. We selected compound 4 to evaluate whether it
is also an irreversible inhibitor for the other investigated
enzymes. Compound 4 showed irreversible binding properties
for uPA, tPA, and trypsin. Surprisingly, the slow recovery of

Diphenyl Phosphonate Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 19 5789
that R-stereochemistry at the carbon, corresponding to the natural
L-amino acids, and the R-stereochemistry at the phosphorus gave
the most plausible result. The phosphonate oxygen is directed
toward the oxyanion hole where it forms two hydrogen bonds
with the backbone NH groups of Gly193 and Ser195. The
phenol oxygen forms a hydrogen bond with the protonated
His57 from the catalytic triad. The phenyl ring of the phenol is
put into the hydrophobic S1′ pocket. The P2 alanine is placed
in the small S2 pocket. The hydroxyl group of the P3 D-Ser
interacts with the carbonyl oxygen of Leu97B. Together with
Thr97A, the residue Leu97B is part of an uPA-specific insertion
loop, which is not present in the other related trypsin-like serine
proteases. The CO and NH of the P3 residue form a hydrogen
bond with, respectively, the NH and CO of Gly216, and the
carbonyl oxygen of the benzyl carbamate forms a hydrogen bond
with the backbone NH of Gly218. These contacts compare well
with those of the reported crystal structures of 5-7 (Figure 3).
Figure 4. Inhibitors 3a (yellow) and 4 (pink) shown in the active site
of uPA. The molecular surface of uPA is visualized (red ) solvent
accessible, green ) hydrophobic, blue ) hydrophilic).
The highly specific interactions in the S1 pocket explain why
the benzylguanidine inhibitors (3) are much more potent than
the previously described diphenyl phosphonates.15 Modeling of
the arginine analogue 2 in uPA showed that, when the phosphor
atom is covalently linked to Ser195, there is no possibility to
form the salt bridge with Asp189. Indeed, the distance between
the phosphor atom and the carbon atom of the guanidine is 9.24
Å in 3a bound to uPA, whereas it is only 7.76 Å in 2 bound to
uPA. Similar modeling conclusions could be obtained for the
other P1 modifications.15
the enzymatic activity of thrombin, after dilution of the
preformed enzyme-inhibitor complex, indicated that 4 is either
a reversible slow-binding thrombin inhibitor or that the phos-
phonylated active site serine is hydrolyzing. A similar behavior
(but less obvious) was observed with plasmin.
To check the possibility to evaluate our compounds in a
mouse model, we tested compound 4 toward the murine
enzymes uPA, plasmin, and tPA. The IC50 value for muPA is
410 ( 60 nM, and there is only a factor 5 in selectivity toward
the two other murine enzymes. These data were in agreement
with formerly published results comparing the species specificity
of amidine-based urokinase inhibitors.21 It is known that there
is a small but important difference in the sequence of the active
site of mouse uPA compared to human uPA. This result shows
that it is not beneficial to test these compounds in an all-mouse
tumor model.
We also studied why compounds 3 are highly selective with
respect to the related trypsin-like enzymes. The general order
of potency of these compounds toward the studied enzymes is
uPA > trypsin > plasmin > tPA > thrombin > FXa. There
are several factors that contribute to this selectivity. The first
one is the much deeper S1 pocket of uPA. The distance from
the side-chain oxygen of Ser195 to the carboxylic carbon atom
of Asp189 is about 0.5 Å longer compared to that of trypsin,
plasmin, tPA, and FXa and about 0.9 Å longer compared to
that of thrombin. These smaller pockets hamper the correct
positioning of the benzylguanidine. A second factor is the above-
mentioned difference between the Ser190 enzymes (uPA,
trypsin, and plasmin) and the Ala190 enzymes (tPA, thrombin,
and FXa). The oxygen of the side chain of Ser190 forms an
extra hydrogen bond with one of the nitrogens of the guanidine
of the inhibitor and explains the higher potency toward the
Ser190 enzymes. A third contributing factor is the presence of
a unique insertion loop in uPA (Thr97A-Leu97B) forming a
hydrogen bond with the hydroxyl of the D-Ser in the inhibitors.
Further arguments for the selectivity order can be found in the
sequence identity. The overall sequence identity with uPA is
highest for tPA (44%), lower for trypsin and plasmin (34%),
and lowest for thrombin and FXa (29%). Apart from tPA, this
agrees well with the general order of potency. The lower potency
for tPA can be explained by the much lower sequence identity
with uPA when only the 28 amino acids are considered that
have at least one atom within a distance of 4.5 Å of 3a. This
sequence identity is 81 and 74% for trypsin and plasmin,
respectively, whereas it is only 59% for tPA.
Molecular Modeling
A molecular modeling study was performed to explain the
high potency and remarkable selectivity of the here reported
uPA inhibitors. A first question was how compound 3a fits in
the active site of uPA. We decided to model the complex after
reaction of the diphenyl phosphonate with the active site serine
because we believe this reflects best the transition state of the
reaction. However, we were also able to model the diphenyl
phosphonate in the active site before the reaction (results not
shown).
The model of 3a in uPA was constructed from the superposi-
tion of the crystal structures of three different inhibitors (5-7),
followed by energy minimization.6,22,23 The benzylguanidine
moiety fits nicely in the S1 specificity pocket of uPA (Figure
3). The terminal guanidine forms a symmetrical salt bridge with
the carboxylate of Asp189. One nitrogen atom makes a hydrogen
bond with the carbonyl oxygen of Gly218, and another nitrogen
atom forms a hydrogen bond with Ser190 OH. This Ser190 in
the S1 pocket is one of the important specificity factors because
it is present in uPA, plasmin, and trypsin but is replaced by
Ala190 in tPA, thrombin, and factorXa.24 The phenyl ring is
sandwiched between the planes of the amide bonds of Trp215-
Gly216 and Ser190-Cys191 and Tyr228 at the back. Because
the synthesized compounds are diastereomeric mixtures at the
carbon connected to the phosphorus atom and because the
phosphorus atom becomes chiral after covalent linkage with the
oxygen of Ser195, we investigated which of the four possible
diastereomers gave the best fit in the active site. It turned out
In several models we noticed the same interactions irrespec-
tive of the nature of the P4 substituent, which explains their
equal potency toward uPA. Some differences in selectivity are
observed, but this could not be explained by our molecular
models.
Finally, the diparacetamol phosphonate 4 and the diphenyl
phosphonate 3a are equally potent toward uPA. The pocket
accommodating the phenyl or paracetamol is large enough, and

5790 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 19
Joossens et al.
added. The solution was stirred at room temperature for 4 h. DCM
was removed in vacuo. The crude product was dissolved in MeOH,
and the solution was kept at 4 °C until precipitation of diphenyl
phosphonate was complete. The precipitated product was filtered
off and crystallized from MeOH to give 12 as a white solid. Yield:
30%. H NMR (CDCl3): δ 1.5 (s, 9H), 3.0-3.5 (m, 2H), 4.7 (m,
1H), 5.1 (m, 2H), 5.2 (d, 1H), 6.5 (s, 1H), 7.1-7.4 (m, 19H). MS
(ESI): m/z 625 [M + Na].
no extra interactions are observed with the p-acetylamino group
(Figure 4). With respect to the selectivity of 4, the most
significant difference is the 13-fold higher potency of 4 toward
thrombin. This can be explained by an extra hydrogen bond
between the nitrogen of the p-acetylamino group and a car-
boxylic oxygen of Glu192.
1
Conclusions
Di-(4-acetamidophenyl)-1-(benzyloxycarbonylamino)-2-(4-
tert-butyloxycarbonylaminophenyl)ethanephosphonate (Parac-
etamol Analogue of 12). The procedure as given for 12 was
followed, except using tri-4-acetamidophenyl phosphite. The crude
product was purified by flash column chromatography (EtOAc/
An efficient method to make a small library of potent and
highly selective irreversible uPA inhibitors was developed. The
best compounds show uPA inhibiting activities of about 5 nM
(IC50, 15 min, 37 °C) and remarkable in vitro selectivity indices
of 3000 or more toward related enzymes from the fibrinolytic
and blood coagulation systems. Compound 3d belongs to the
most potent and selective uPA inhibitors ever reported. We have
shown that these diphenyl phosphonates irreversibly inhibit uPA.
A potent and selective p-acetylaminophenol analogue was
developed to circumvent the release of cytotoxic phenol.
Differences in potency and selectivity of these inhibitors could
be explained by molecular modeling.
1
MeOH) to give pure 12 as a white foam. Yield: 20%. H NMR
(CD3OD): δ 1.5 (s, 9H), 2.0 (s, 6H), 4.6 (m, 1H), 5.1 (m, 2H),
7.1-7.3 (m, 13H), 7.5 (m, 4H). MS (ESI): m/z 739 [M + Na].
Diphenyl 1-(Benzyloxycarbonylamino)-2-(4-N,N′-bis(tert-bu-
tyloxycarbonyl)guanidino)phenyl)ethanephosphonate (13). The
Boc-protected compound 12 was dissolved in 50% TFA in DCM
(2-5 mL). After stirring for 3 h at room temperature, the solvent
was evaporated. The crude oil was washed with cold ether, and a
precipitate was formed. N,N′-Bis(tert-butyloxycarbonyl)-guanidi-
nopyrazole (1 equiv) and TEA (1 equiv) in ACN (5 mL) were
added, and the mixture was stirred at reflux temperature overnight.
The solvent was evaporated, and the residue was dissolved in
EtOAc. The organic layer was washed with 1 N HCl, saturated
NaHCO3 solution, and brine. The organic layer was dried with Na2-
SO4 and evaporated. Yield 65%. 1H NMR (CDCl3): δ 1.5 (s (2×),
18H), 3.0-3.4 (m, 2H), 4.7 (m, 1H), 5.1 (m, 2H), 7.1-7.4 (m,
19H). MS (ESI): m/z 745 [M + Na].
Experimental Section
Chemistry. Amino acids and TBTU were purchased from
Novabiochem. Other reagents were obtained from Sigma-Aldrich
1
or Acros. Characterization of all compounds was done with H
1
NMR and mass spectrometry. H NMR spectra were recorded on
a 400 MHz Bruker Avance DRX-400 spectrometer. ES Mass
spectra were obtained from an Esquire 3000plus iontrap mass
spectrometer from Bruker Daltonics. Purity was verified using two
diverse HPLC systems using, respectively, a mass and UV detector.
Water (A) and ACN (B) were used as eluents. LC-MS spectra were
recorded on an Agilent 1100 series HPLC system using an Alltech
Prevail C18 column (2.1 × 50 mm, 3 µm) coupled with an Esquire
3000plus as MS detector and a 5-100% B, 20 min gradient was
used with a flow rate from 0.2 mL/min. Formic acid (0.1%) was
added to solvents A and B. Reversed phase HPLC was run on a
Gilson instrument equipped with an Ultrasphere ODS column (4.6
× 250 mm, 5 µm). A 10-100% B, 35 min gradient was used with
a flow rate from 1 mL/min. TFA (0.1%) was added to solvents A
and B. A 214 nm wavelength was used. When necessary, the
products were purified with flash chromatography on a Flashmaster
II (Jones Chromatography) with a 30 min gradient of 0-50%
EtOAc in hexane or 0-25% MeOH in EtOAc.
4-(tert-Butyloxycarbonylamino)phenylacetaldehyde (11). To
a solution of the amino alcohol 10 in 40 mL of dioxane were added
triethylamine (1 equiv) and di-tert-butyl dicarbonate (1.1 equiv).
The mixture was stirred at room temperature for 1-8 h (reaction
was monitored by TLC). The solution was concentrated in vacuo
and acidified with 2 N HCl. The acidified aqueous layer was
extracted with EtOAc (3×). The combined organic layers were dried
over Na2SO4 and evaporated under reduced pressure yielding the
Boc-protected N-substituted alcohol, which was purified by flash
column chromatography (EtOAc/hexane).
To a stirred suspension of this alcohol (1 equiv) in DCM (80
mL) at -78 °C, a solution of Dess-Martin periodane (1.5 equiv
from a 15 wt % solution) was added. The suspension was stirred
for 3 h at room temperature. The resulting solution was poured
into a vigorously stirred saturated NaHCO3 and Na2S2O3 solution
(1:1/100 mL). The organic layer was separated and washed with
brine and dried over Na2SO4. The crude aldehyde 11 was obtained
by removing the solvent in vacuo and used in the next step without
further purification. Yield: 58%. 1H NMR (CDCl3): δ 1.5 (s, 9H),
3.75 (d, 2H), 7.10 (d, 2H), 7.30 (m, 2H), 9.6 (t, 1H). MS (ESI):
m/z 258 [M + Na], 290 [M + MeOH + Na].
Di-(4-acetamidophenyl) 1-(Benzyloxycarbonylamino)-2-{4-
[N,N′-bis(tert-butyloxy-carbonyl)guanidino]phenyl}ethane-
phosphonate (Paracetamol Analogue of 13). This compound was
prepared in the same way as 13 using the corresponding paracetamol
1
analogue of 12. Yield: 40%. H NMR (CDCl3): δ 1.5 (d, 18H),
2.0 (d, 6H), 2.9-3.2 (m, 2H), 4.6 (m, 1H), 5.1 (s, 2H), 6.7-7.3
(m, 17H), 8.7 (s, 1H), 8.8 (s, 1H), 10.2 (s, 1H), 11.7 (s, 1H).
Diphenyl 1-Amino-2-(4-N,N′-bis(tert-butyloxycarbonyl)guani-
dino)phenyl)ethanephosphonate (14). The Z-protected phospho-
nate (1 equiv) 13 was dissolved in 20 mL MeOH. Pd/C (10%) was
added. After removing the oxygen with N2, H2 was bubbled through
the solution. The solution was stirred for 8 h at room temperature.
The solution was filtered over Celite, and the filtrate was evaporated.
Compound 14 was used without further purification.
Di(4-acetamidophenyl) 1-Amino-2-(4-N,N′-bis(tert-butyloxy-
carbonyl)guanidino)phenyl)ethanephosphonate (Paracetamol
Analogue of 14). This compound was prepared in the same way
as 14 using the corresponding paracetamol analogue of 13.
Deprotection of the Compounds 9 to the Final Compounds
3. Compounds 9 were dissolved in 50% TFA in DCM (2-5 mL).
After stirring for 3 h at room temperature, the solvent was
evaporated. The crude oil was washed with cold ether, and a
precipitate was formed.
Diphenyl 1-[(N-Benzoyloxycarbonyl-D-seryl)-L-alanyl]amino-
2-(4-guanidinophenyl)ethanephosphonate Trifluoroacetate (3a).
Yield: 85%. 1H NMR (CDCl3): δ 1.3 (m, 3H), 3.0-3.4 (m, 2H),
3.6 (m, 2H), 4.25 (m, 1H), 4.4 (m, 1H), 5.1 (m, 3H), 7.1-7.4 (m,
19H). HPLC: 214 nm, tr 19.98 min, 100%. LC/MS: tr 14.4 min,
100%. MS (ESI): m/z 703 (M+1).
Diphenyl 1-[(N-Benzenesulfonyl-D-seryl)-L-alanyl]amino-2-(4-
guanidinophenyl)ethanephosphonate Trifluoroacetate (3b).
Yield: 71%. 1H NMR (CD3OD): δ 1.1 (m, 3H), 3.0 (m, 2H), 3.6
(m, 2H), 4.0 (m, 0.5HA), 4.1 (m, 0.5HB), 4.9 (m, 0.5HA), 5.1 (m,
0.5HB), 7.0-7.8 (m, 19H). HPLC: 214 nm, tr 19.24 min, 97.27%;
254 nm, tr 19.30 min, 100%. LC/MS: tr 13.0 min, 100%. MS
(ESI): m/z 709 [M + Na].
Diphenyl 1-[(N-R-Toluenesulfonyl-D-seryl)-L-alanyl]amino-2-
(4-guanidinophenyl)ethanephosphonate Trifluoroacetate (3c).
Yield: 76%. 1H NMR (CD3OD): δ 1.1 (d, 3H), 3.0 (m, 1H), 3.2-
3.5 (m, 3H), 3.6 (s, 2H), 3.7 (m, 1H), 4.1 (m, 1H), 4.9 (m, 0.5HA),
5.1 (m, 0.5HB), 7.0-7.4 (m, 19H). HPLC: 214 nm, tr 19.54 min,
Diphenyl 1-Benzyloxycarbonylamino-2-(4-tert-butyloxycar-
bonylaminophenyl)ethanephosphonate (12). To a solution of
crude aldehyde 11, triphenyl phosphite (1 equiv), and benzyl
carbamate (1 equiv) in DCM (50 mL), 0.1 equiv Cu(OTf)2 was

Diphenyl Phosphonate Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 19 5791
100%; 254 nm, tr 19.56 min, 100%. LC/MS: tr 13.3 min, 93.5%.
MS (ESI): m/z 723 [M + Na].
Yield: 92%. 1H NMR (CDCl3): δ 1.26 (dd, J ) 2.4 and J ) 7.2,
1H), 1.28 (dd, J ) 2.4 and J ) 7.2), 2.4 (s, 3H), 3.2 (m, 1H), 3.4
(m, 1H), 3.8 (m, 2H), 4.25 (m, 1H), 4.4 (m, 1H), 5.1 (m, 1H),
7.0-7.9 (m, 18H). HPLC: 214 nm, tr 18.56 min, 100%. HPLC:
254 nm, tr 18.56 min, 46.32%, tr 19.00 min, 46.32%. LC/MS: tr
13.5 min, 100%. MS (ESI): m/z 687 (M+1).
Diphenyl 1-[(N-o-Methylbenzoyl-D-seryl)-L-alanyl]amino-2-
(4-guanidinophenyl)ethanephosphonate Trifluoroacetate (3n).
Yield: 91%. 1H NMR (CDCl3): δ 1.28 (dd, J ) 4.4 and J ) 7.2,
1H), 1.30 (dd, J ) 4.4 and J ) 7.2), 2.4 (m, 3H), 3.25 (m, 1H),
3.45 (m, 1H), 3.85 (dd, J ) 5.2 and J ) 21.2, 1H), 3.92 (dd, J )
5.2 and J ) 21.2, 1H), 4.3 (m, 1H), 4.45 (m, 0.5H), 4.52 (m, 0.5H),
5.1 (m, 1H), 7.0-7.9 (m, 18H). HPLC: 214 nm, tr 18.46 min,
43.41%; tr 18.82 min, 43.81%; 254 nm, tr 18.80 min, 40.74%; tr
19.04 min, 39.77%. LC/MS: tr 13.6 min, 100%. MS (ESI): m/z
687 (M+1).
Diphenyl 1-[(N-o,o-Dimethylbenzoyl-D-seryl)-L-alanyl]amino-
2-(4-guanidinophenyl)ethanephosphonate Trifluoroacetate (3o).
Yield: 94%. 1H NMR (CDCl3): δ 1.28 (dd, J ) 7.2 and J ) 12.4,
1H), 1.30 (dd, J ) 7.2 and J ) 12.4), 2.28 (s, 3H), 2.31 (s, 3H),
3.25 (m, 1H), 3.45 (m, 1H), 3.82 (dd, J ) 6.4 and J ) 21.2, 1H),
3.88 (dd, J ) 6.4 and J ) 21.2, 1H), 4.3 (m, 1H), 4.55 (m, 0.5H),
4.58 (m, 0.5H), 5.1 (m, 1H), 7.0-7.9 (m, 17H). HPLC: 214 nm, tr
18.97 min, 39.36%; tr 19.13 min, 39.54%; 254 nm, tr 19.00 min,
39.78%; tr 19.20 min, 38.64%. LC/MS: tr 13.7 min, 100%. MS
(ESI): m/z 701 (M+1).
Di-(4-acetamidophenyl) 1-[(N-Benzyloxycarbonyl-D-seryl)-L-
alanyl]amino-2-[4-(guanidino)phenyl]- ethanephosphonate Tri-
fluoroacetate (4). Yield: 93%. 1H NMR (CDCl3): δ 1.3 (m, 3H),
2.0 (s, 6H), 3.0-3.4 (m, 2H), 3.7 (m, 2H), 4.1-4.4 (m, 2H), 5.0
(m, 1H), 5.1 (m, 2H), 7.1-7.6 (m, 17H). HPLC: 214 nm, tr 14.3
min, 42.1%; tr 14.0 min, 44.3%. HPLC: 254 nm, tr 13.8 min,
45.1%; tr 14.0 min, 49.0%. LC/MS: tr 12.0 min. Purity: 92.0%.
MS (ESI): m/z 817 (M+1).
uPA Inhibition: in Vitro Evaluation. Enzymatic activity was
measured at 37 °C in a Spectramax 340 (Molecular Devices)
microtiter plate reader using the chromogenic substrate S-2444 (L-
pyroGlu-Gly-L-Arg-p-NA‚HCl), with a Km of 80 µM. The substrate
was obtained from Chromogenix. The human enzyme was obtained
from Sigma-Aldrich and the mouse uPA was obtained from
Molecular Innovations, Inc. (U.S.A.). The reaction was monitored
at 405 nm, and the initial rate was determined between 0 and 0.25
absorbance units in 20 min. The reaction mixture contained 250
µM substrate and approximately 1 mU of enzyme in 145 µL of
buffer in a final volume of 200 µL. A 50 mM Tris buffer, pH 8.8,
was used. From each inhibitor concentration, 5 µL was added,
obtaining a final concentration from 0 to 250 µM in a total volume
of 0.2 mL. Activity measurements were routinely performed in
duplicate. The IC50 value is defined as the concentration of inhibitor
required to reduce the enzyme activity to 50% after a 15-min
preincubation with the enzyme at 37 °C before addition of the
substrate. IC50 values were obtained by fitting the data with the
four-parameter logistics equation using Grafit 5.
Diphenyl 1-[(N-Benzyl-D-seryl)-L-alanyl]amino-2-(4-guanidi-
nophenyl)ethanephosphonate Trifluoroacetate (3d). Yield: 46%.
1H NMR (CD3OD): δ 1.3 (d, 3H), 3.3 (d, 1H), 3.5 (m, 1H), 4 (m,
2H), 4.2 (2m, 1H), 4.5 (dm, 1H), 7.0-7.6 (m, 19H). HPLC: 214
nm, tr 18.12 min, 46.90%; tr 17.72 min, 43.45%; 254 nm, tr 17.99
min, 48.69%; tr 17.59 min, 42.39%. LC/MS: tr 12.7 min, 99.9%.
MS (ESI): m/z 695 [M + Na].
Diphenyl 1-[(N-Naftalenesulfonyl-D-seryl)-L-alanyl]amino-2-
(4-guanidinophenyl)ethanephosphonate Trifluoroacetate (3e).
1
Yield: 55%. H NMR (CD3OD): δ 1.0 (d, 3H), 3.2 (m, 1H), 3.5
(m, 2H), 3.6-3.8 (dm, 2H), 4.1 (dm, 1H), 5.0 (m, 1H), 7.0-7.7
(m, 18 H), 8.0 (m, 1H), 8.1 (m, 1H), 8.3 (d, 1H), 8.7 (m, 1H).
HPLC: 214 nm, tr 20.20 min, 100%; 254 nm, tr 20.13 min, 100%.
LC/MS: tr 14.4 min, 97.7%. MS (ESI): m/z 781 [M + Na].
Diphenyl 1-[(N-2-Thiophenesulfonyl-D-seryl)-L-alanyl]amino-
2-(4-guanidinophenyl)ethanephosphonate Trifluoroacetate (3f).
1
Yield: 35%. H NMR (CDCl3): δ 1.3 (m, 3H), 3.1 (m, 1H), 3.4
(m, 1H), 3.6 (m, 2H), 3.9 (m, 1H), 4.25 (m, 1H), 4.4 (m, 1H),
7.1-7.4 (m, 15H), 7.6 (m, 1H), 7.7 (m, 1H). HPLC: 214 nm, tr
18.62 min, 100%. HPLC: 254 nm, tr 18.23 min, 100%. LC/MS: tr
12.9 min, 94.9%. MS (ESI): m/z 715 (M+1).
Diphenyl 1-[(N-2-Thienyl-D-seryl)-L-alanyl]amino-2-(4-guani-
dinophenyl)ethanephosphonate Trifluoroacetate (3g). Yield:
55%. 1H NMR (CDCl3): δ 1.3 (m, 3H), 3.4 (m, 2H), 3.9 (m, 2H),
4.2-4.3 (m, 2H), 5.1 (m, 1H), 7.1-7.8 (m, 17H). HPLC: 214 nm,
tr 17.02 min, 42.13%; 17.50 min, 57.87%. HPLC: 254 nm, tr 16.89
min, 41.55%; 17.45 min, 58.45%. LC/MS: tr 12.9 min, 94.9%. MS
(ESI): m/z 679 (M+1).
Diphenyl 1-[(N-1-Adamantanyl-D-seryl)-L-alanyl])amino-2-(4-
guanidinophenyl)ethanephosphonate Trifluoroacetate (3h).
1
Yield: 67%. H NMR (CDCl3): δ 1.28 (dd, J ) 8.4 and J ) 16,
1H), 1.30 (dd, J ) 8.4 and J ) 16), 1.7 (m, 6H), 1.9 (m, 5H), 2.0
(m, 4H), 3.25 (m, 1H), 3.45 (m, 1H), 3.75 (m, 2H), 4.1 (m, 1H),
4.3 (m, 1H), 5.1 (m, 1H), 7.0-7.9 (m, 14H). HPLC: 214 nm, tr
19.96 min, 77.42%; 254 nm, tr 20.89 min, 35.77%; 21.22 min,
38.96%. LC/MS: tr 13.9 min, 98.8%. MS (ESI): m/z 731 (M+1).
Diphenyl 1-[(N-p-Methoxybenzenesulfonyl-D-seryl)-L-alanyl]-
amino-2-(4-guanidinophenyl)ethane-phosphonate Trifluoroac-
1
etate (3i). Yield: 23%. H NMR (CD3OD): δ 1.2 (m, 3H), 3.1
(m, 1H), 3.4 (m, 3H), 3.5-3.6 (m, 2H), 3.8 (m, 2H), 4.0 (m, 1H),
4.2 (m, 1H), 5.1 (m, 1H), 7.0-8.0 (m, 18H). HPLC: 214 nm, tr
19.62 min, 100%; 254 nm, tr 19.23 min, 100%. LC/MS: tr 13.8
min, 94.9%. MS (ESI): m/z 761 [M + Na].
Diphenyl 1-[(N-p-Bromobenzenesulfonyl-D-seryl)-L-alanyl]-
amino-2-(4-guanidinophenyl)ethane-phosphonate Trifluoroac-
etate (3j). Yield: 55%. 1H NMR (CDCl3): δ 1.2 (m, 3H), 3.1 (m,
1H), 3.2 (s, 2H), 3.5 (m, 1H), 3.6-3.7 (dm, 2H), 4.2 (m, 2H), 7.0-
8.0 (m, 18H). HPLC: 214 nm, tr 21.09 min, 94.07%; 254 nm, tr
20.10 min, 94.61%. LC/MS: tr 14.1 min, 96.4%. MS (ESI): m/z
809 [M + Na].
Diphenyl 1-[(N-p-Cyanobenzenesulfonyl-D-seryl)-L-alanyl]-
amino-2-(4-guanidinophenyl)ethane-phosphonate Trifluoroac-
1
etate (3k). Yield: 43%. H NMR (CD3OD): δ 1.3 (d, 3H), 3.2
V ) (V range)/(1 + exp(s × ln(abs(I0/IC50)))) + background
(m, 1H), 3.5 (m, 2H), 3.6-3.9 (dm, 2H), 4.2 (m, 1H), 5.1 (m, 1H),
7.1-7.4 (dm, 14H), 7.8-8.0 (dm, 4H). HPLC: 214 nm, tr 18.86
min, 92.64%; 254 nm, tr 18.79 min, 95.59%. LC/MS: tr 13.1 min,
100%. MS (ESI): m/z 757 [M + Na].
where s ) slope factor, V ) rate, I0 ) inhibitor concentration, and
range ) the fitted uninhibited value minus the background. The
equation assumes the y falls with increasing x.
Diphenyl 1-[(N-p-Methylbenzoyl-D-seryl)-L-alanyl]amino-2-
(4-guanidinophenyl)ethanephosphonate Trifluoroacetate (3l).
Yield: 93%. 1H NMR (CDCl3): δ 1.28 (dd, J ) 2.8 and J ) 7.2,
1H), 1.30 (dd, J ) 2.4 and J ) 7.2), 2.38 (d, 3H), 3.25 (m, 1H),
3.45 (m, 1H), 3.85 (dd, J ) 5.6 and J ) 14.8, 1H), 3.92 (dd, J )
5.6 and J ) 14.8, 1H), 4.2 (m, 0.5H), 4.35 (m, 0.5H), 4.45 (m,
0.5H), 4.53 (m, 0.5H), 5.1 (m, 1H), 7.0-7.9 (m, 18H). HPLC: 214
nm, tr 18.46 min, 47.58%; tr 18.82 min, 44.80%. HPLC: 254 nm,
tr 18.59 min, 47.64%; tr 19.04 min, 43.96%. LC/MS: tr 13.4 min,
100%. MS (ESI): m/z 687 (M+1).
Inhibitor stock solutions were prepared in DMSO and stored at
-20 °C. Because the compounds described in this paper completely
inactivate uPA following pseudo-first-order kinetics, the IC50 value
is inversely correlated with the second-order rate constant of
inactivation. For a simple pseudo-first-order inactivation process,
the activity after incubation with inhibitor (Vi) varies with the
inhibitor concentration (i), as described in the following equation:
Vi ) V0‚e-kit, where V0 is the activity in absence of inhibitor, k is
the second-order rate constant of inactivation, and t is the time.
The inactivation rate constant was determined from the time
course of inhibition.
Diphenyl 1-[(N-o-Methylbenzoyl-D-seryl)-L-alanyl]amino-2-
(4-guanidinophenyl)ethanephosphonate Trifluoroacetate (3m).

5792 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 19
Joossens et al.
The inhibitor was mixed with the substrate (250 µM final
concentration), and the buffer solution with the enzyme was added
at the time zero. The inhibitor concentrations were chosen to obtain
total inhibition of the enzyme within 20 min. The progress curves
show the absorbance of p-nitroanilide produced as a function of
time. Initially, no inhibitor is bound to the enzyme, and the tangent
to the progress curve (dA/dt) is proportional to the concentration
of the free enzyme. The concentration of free enzyme decreases
over time due to the kinetics of inhibitor binding, as described
above. Progress curves were recorded in pseudo-first-order condi-
tions ([I]0 . [E]0) and with less than 10% conversion of the
substrate during the entire time course. In these conditions, dA/dt
decreases exponentially with time. The progress curves were fitted
with the integrated rate equation to yield a value for kobs, a pseudo-
first-order rate constant.
oxygen of Ser195. Then the complex was minimized using
MMFF94x force field with a flexible ligand in a rigid enzyme.
This was followed by minimization using MMFF94x force field
of a flexible ligand in a flexible enzyme pocket. This pocket was
defined as the amino acids with at least one atom within a distance
of 6 Å of the ligand. All other inhibitors were built from 3c,
followed by analogous minimization. The following PDB files were
used to study the selectivity toward related enzymes: tPA (1A5H,
1RTF), thrombin (1DWB, 1GJ4, 1H8I), plasmin (1BUI), FXa
(1FAX, 1KYE, 1MQ6), and trypsin (1H4W, 1EB2, 1MAX).
Acknowledgment. This work received support from the
Fund for Scientific Research-Flanders (Belgium; FWO). P.V.
is a postdoctoral fellow of the FWO and G.S. is a visiting
postdoc of the FWO. J.J. is a fellow of the Institute of Promotion
of Innovation in Science and Technology of Flanders (IWT).
The excellent technical assistance of W. Bollaert is greatly
appreciated.
At ) V0[1 - exp(-kobst)]/kobs + A0
where At ) absorbance at time t, A0 ) absorbance at time zero,
and V0 ) uninhibited initial rate.
Supporting Information Available: Experimental details for
the intermediates 8 and 9, extra modeling pictures, figures
demonstrating irreversible inhibition, and purity data for target
compounds. This material is available free of charge via the Internet
at http://pubs.acs.org.
The apparent second-order rate constant (kapp) was calculated
from the slope of the linear part of the plot of kobs versus the
inhibitor concentration ([I]0). In case of competition between the
inhibitor and the substrate, kapp is smaller than the “real” second-
order rate constant k discussed above because a certain fraction of
the enzyme is present as an enzyme-substrate complex. kapp
depends on the substrate concentration used in the experiment, as
described by Lambeir et al.25
References
(1) Dano, K.; Behrendt, N.; Hoyer-Hansen, G.; Johnsen, M.; Lund, L.
R.; Ploug, M.; Romer, J. Plasminogen Activation and Cancer.
Thromb. Haemostasis 2005, 93, 676-681.
Determination of the Selectivity for uPA. The IC50 values for
plasmin (from human plasma, Sigma), tPA (recombinant, Boe-
hringer Ingelheim), thrombin (from human plasma, Sigma), FXa,
and trypsin (bovine, Roche) were determined in the same way as
for uPA. S-2288 (H-D-Ile-Pro-Arg-pNa‚2HCl) for tPA (Km: 1 mM),
S-2366 (pyroGlu-Pro-Arg-pNA‚HCl) for plasmin (Km: 400 µM)
and thrombin (Km: 150 µM), and S-2772 (Boc-D-Arg-Gly-Arg-
pNA‚2HCl) for FXa (Km: 1.5 mM) and BAPNA (NR-benzoyl-
D,L-Arg-pNA‚HCl) for trypsin (Km: 1mM) were used as substrates.
FX (Sigma) is activated with Russel’s viper venom (Sigma).
The mixture contained 580 µM substrate for thrombin and
plasmin, 1.25 mM for tPA, 522 µM for FXa, and 425 µM for trypsin
and approximately 5 mU of enzyme and 145 µL of buffer. For
tPA and thrombin, Tris buffer, pH 8.3, was used, for FXa, Tris
buffer, pH 8.3, with 2.5 mM CaCl2 was used, for plasmin, Tris
buffer, pH 7.4, was used, and for trypsin, Tris buffer, pH 8.2, was
used. The selectivity index was calculated as (IC50 enzyme X)/
IC50 uPA, where X is tPA, thrombin, plasmin, FXa, or trypsin.
Determination of the Inhibition Type. To follow the dissocia-
tion of the inhibitor-enzyme complex, aliquots of enzyme were
incubated without inhibitor and in the presence of a concentration
of the inhibitor 50 times the IC50 at 37 °C. The enzyme concentra-
tion was 2.5 times higher than the concentration used for the IC50
determinations. After 15 min, the aliquots were diluted 50-fold into
the substrate concentration and assay buffer used for the IC50
determination. The dissociation of the enzyme-inhibitor complex
was monitored by substrate hydrolysis by measuring the absorbance
every 20 s.
Molecular Modeling. The molecular modeling studies were
performed using MOE 2005.06 software (Chemical Computing
Group). All software necessary to build compounds and perform
minimizations, alignments, and superpositions is available in the
MOE package and was used with standard settings, unless otherwise
mentioned. Crystal structures were downloaded from the Protein
Databank (PDB), and the ligands were checked and corrected when
necessary. Residues not belonging to the enzyme or ligand were
removed, including water molecules. Hydrogen atoms were added
and energy minimized with MMFF94x force field (standard
parameters), keeping all heavy atoms fixed. The following PDB
structures of uPA were used: 1VJ9 (6), 1LMW (5), and 1EJN (7).
The latter two were aligned and superposed on the first. Compound
3c was built using different parts of the three ligands in the three
mentioned enzymes, followed by covalent linkage to the side-chain
(2) Mazzieri, R.; Blasi, F. The Urokinase Receptor and the Regulation
of Cell Proliferation. Thromb. Haemostasis 2005, 93, 641-646.
(3) Duffy, M. J. The Urokinase Plasminogen Activator System: Role
in Malignancy. Curr. Pharm. Des. 2004, 10, 39-49.
(4) Lee, M.; Fridman, R.; Mobashery, S. Extracellular Proteases As
Targets for Treatment of Cancer Metastases. Chem. Soc. ReV. 2004,
33, 401-409.
(5) Setyono-Han, B.; Sturzebecher, J.; Schmalix, W. A.; Muehlenweg,
B.; Sieuwerts, A. M.; Timmermans, M.; Magdolen, V.; Schmitt, M.;
Klijn, J. G. M.; Foekens, J. A. Suppression of Rat Breast Cancer
Metastasis and Reduction of Primary Tumour Growth by the Small
Synthetic Urokinase Inhibitor WX-UKI. Thromb. Haemostasis 2005,
93, 779-786.
(6) Schweinitz, A.; Steinmetzer, T.; Banke, I. J.; Arlt, M. J. E.;
Sturzebecher, A.; Schuster, O.; Geissler, A.; Giersiefen, H.; Zeslaw-
ska, E.; Jacob, U.; Kruger, A.; Sturzebecher, J. Design of Novel and
Selective Inhibitors of Urokinase-Type Plasminogen Activator With
Improved Pharmacokinetic Properties for Use As Antimetastatic
Agents. J. Biol. Chem. 2004, 279, 33613-33622.
(7) Almholt, K.; Lund, L. R.; Rygaard, J.; Nielsen, B. S.; Dano, K.;
Romer, J.; Johnsen, M. Reduced Metastasis of Transgenic Mammary
Cancer in Urokinase-Deficient Mice. Int. J. Cancer 2005, 113, 525-
532.
(8) Ahmed, N.; Oliva, K.; Wang, Y.; Quinn, M.; Rice, G. Downregulation
of Urokinase Plasminogen Activator Receptor Expression Inhibits
Erk Signaling With Concomitant Suppression of Invasiveness Due
to Loss of UPAR-Beta 1 Integrin Complex in Colon Cancer Cells.
Br. J. Cancer 2003, 89, 374-384.
(9) D'Alessio, S.; Margheri, F.; Pucci, M.; Del Rosso, A.; Monia, B. P.;
Bologna, M.; Leonetti, C.; Scarsella, M.; Zupi, G.; Fibbi, G.; Del
Rosso, M. Antisense Oligodeoxynucleotides for Urokinase-Plasmi-
nogen Activator Receptor Have Anti-Invasive and Anti-Proliferative
Effects in Vitro and Inhibit Spontaneous Metastases of Human
Melanoma in Mice. Int. J. Cancer 2004, 110, 125-133.
(10) Margheri, F.; D'Alessio, S.; Serrati, S.; Pucci, M.; Annunziato, F.;
Cosmi, L.; Liotta, F.; Angeli, R.; Angelucci, A.; Gravina, G. L.;
Rucci, N.; Bologna, M.; Teti, A.; Monia, B.; Fibbi, G.; Del Rosso,
M. Effects of Blocking Urokinase Receptor Signaling by Antisense
Oligonucleotides in a Mouse Model of Experimental Prostate Cancer
Bone Metastases. Gene Ther. 2005, 12, 702-714.
(11) Weigelt, B.; Peterse, J. L.; van’t Veer, L. J. Breast Cancer Metasta-
sis: Markers and Models. Nat. ReV. Cancer 2005, 5, 591-602.
(12) Rockway, T. W.; Nienaber, V.; Giranda, V. L. Inhibitors of the
Protease Domain of Urokinase-Type Plasminogen Activator. Curr.
Pharm. Des. 2002, 8, 2541-2558.
(13) Mackman, R. L.; Katz, B. A.; Breitenbucher, J. G.; Hui, H. C.; Verner,
E.; Luong, C.; Liu, L.; Sprengeler, P. A. Exploiting Subsite S1 of
Trypsin-Like Serine Proteases for Selectivity: Potent and Selective
Inhibitors of Urokinase-Type Plasminogen Activator. J. Med. Chem.
2001, 44, 3856-3871.

Diphenyl Phosphonate Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 19 5793
(14) Zeslawska, E.; Jacob, U.; Schweinitz, A.; Coombs, G.; Bode, W.;
Madison, E. Crystals of Urokinase Type Plasminogen Activator
Complexes Reveal the Binding Mode of Peptidomimetic Inhibitors.
J. Mol. Biol. 2003, 328, 109-118.
(15) Joossens, J.; Van der Veken, P.; Lambeir, A. M.; Augustyns, K.;
Haemers, A. Development of Irreversible Diphenyl Phosphonate
Inhibitors for Urokinase Plasminogen Activator. J. Med. Chem. 2004,
47, 2411-2413.
(16) Tamura, S. Y.; Weinhouse, M. I.; Roberts, C. A.; Goldman, E. A.;
Masukawa, K.; Anderson, S. M.; Cohen, C. R.; Bradbury, A. E.;
Bernardino, V. T.; Dixon, S. A.; Ma, M. G.; Nolan, T. G.; Brunck,
T. K. Synthesis and Biological Activity of Peptidyl Aldehyde
Urokinase Inhibitors. Bioorg. Med. Chem. Lett. 2000, 10, 983-987.
(17) Oleksyszyn, J.; Boduszek, B.; Kam, C. M.; Powers, J. C. Novel
Amidine-Containing Peptidyl Phosphonates As Irreversible Inhibitors
for Blood-Coagulation and Related Serine Proteases. J. Med. Chem.
1994, 37, 226-231.
(18) Belyaev, A.; Zhang, X. M.; Augustyns, K.; Lambeir, A. M.; De
Meester, I.; Vedernikova, I.; Scharpe, S.; Haemers, A. Structure-
Activity Relationship of Diaryl Phosphonate Esters As Potent
Irreversible Dipeptidyl Peptidase IV Inhibitors. J. Med. Chem. 1999,
42, 1041-1052.
N-Protected Diphenyl 1-Aminoalkylphosphonates. Synthesis 2005,
634-638.
(21) Klinghofer, V.; Stewart, K.; McGonigal, T.; Smith, R.; Sarthy, A.;
Nienaber, V.; Butler, C.; Dorwin, S.; Richardson, P.; Weitzberg, M.;
Wendt, M.; Rockway, T.; Zhao, X. M.; Hulkower, K. I.; Giranda,
V. L. Species Specificity of Amidine-Based Urokinase Inhibitors.
Biochim. Biophys. Acta 2001, 40, 9125-9131.
(22) Spraggon, G.; Phillips, C.; Nowak, U. K.; Ponting, C. P.; Saunders,
D.; Dobson, C. M.; Stuart, D. I.; Jones, E. Y. The Crystal Structure
of the Catalytic Domain of Human Urokinase-Type Plasminogen-
Activator. Structure 1995, 3, 681-691.
(23) Sperl, S.; Jacob, U.; de Prada, N. A.; Sturzebecher, J.; Wilhelm, O.
G.; Bode, W.; Magdolen, V.; Huber, R.; Moroder, L. (4-Aminom-
ethyl)Phenylguanidine Derivatives As Nonpeptidic Highly Selective
Inhibitors of Human Urokinase. Proc. Natl. Acad. Sci. U.S.A. 2000,
97, 5113-5118.
(24) Katz, B. A.; Mackman, R.; Luong, C.; Radika, K.; Martelli, A.;
Sprengeler, P. A.; Wang, J.; Chan, H. D.; Wong, L. Structural Basis
for Selectivity of a Small Molecule, S1-Binding, Submicromolar
Inhibitor of Urokinase-Type Plasminogen Activator. Chem. Biol.
2000, 7, 299-312.
(19) Senten, K.; Daniels, L.; Van der Veken, P.; De Mester, I.; Lambeir,
A. M.; Scharpe, S.; Haemers, A.; Augustyns, K. Rapid Parallel
Synthesis of Dipeptide Diphenyl Phosphonate Esters As Inhibitors
of Dipeptidyl Peptidases. J. Comb. Chem. 2003, 5, 336-344.
(20) Van der Veken, P.; El Sayed, I.; Joossens, J.; Stevens, C. V.;
Augustyns, K.; Haemers, A. Lewis Acid-Catalyzed Synthesis of
(25) Lambeir, A. M.; Borloo, M.; DeMeester, I.; Belyaev, A.; Augustyns,
K.; Hendriks, D.; Scharpe, S.; Haemers, A. Dipeptide-Derived
Diphenyl Phosphonate Esters: Mechanism-Based Inhibitors of
Dipeptidyl Peptidase IV. Biochim. Biophys. Acta 1996, 1290, 76-82.
JM060622G