Exploiting subsite S1 of trypsin-like serine proteases for selectivity: Potent and selective inhibitors of urokinase-type plasminogen activator

Article abstract of DOI:10.1021/jm010244+

A nonselective inhibitor of trypsin-like serine proteases, 2-(2-hydroxybiphenyl-3-yl)-lH-indole-5-carboxamidine (1) (Verner, E.; Katz, B. A.; Spencer, J.; Allen, D.; Hataye, J.; Hruzewicz, W.; Hui, H. C.; Kolesnikov, A.; Li, Y.; Luong, C.; Martelli, A.; Radika. K.; Rai, R.; She, M.; Shrader, W.; Sprengeler, P. A.; Trapp, S.; Wang, J.; Young, W. B.; Mackman, R. L. J. Med. Chem. 2001, 44, 2753-2771) has been optimized through minor structural changes on the S1 binding group to afford remarkably selective and potent inhibitors of urokinase-type plasminogen activator (uPA). The trypsin-like serine proteases¹ that comprise drug targets can be broadly categorized into two subfamilies, those with Ser190 and those with Ala190. A single-atom modification, for example, replacement of hydrogen for chlorine at the 6-position of the 5-amidinoindole P1 group on 1, generated up to 6700-fold selectivity toward the Ser190 enzymes and against the Alal90 enzymes. The larger chlorine atom displaces a water molecule (H₂O1_s1) that binds near residue 190 in all the complexes of 1, and related inhibitors, in uPA, thrombin, and trypsin. The water molecule, H₂O1_s1, in both the Ser190 or Ala190 enzymes, hydrogen bonds with the amidine N1 nitrogen of the inhibitor. When it is displaced, a reduction in affinity toward the Ala190 enzymes is observed due to the amidine N1 nitrogen of the bound inhibitor being deprived of a key hydrogen-bonding partner. In the Ser190 enzymes the affinity is maintained since the serine hydroxyl oxygen Oγ_ser190 compensates for the displaced water molecule. Highresolution crystallography provided evidence for the displacement of the water molecule and validated the design rationale. In summation, a novel and powerful method for engineering selectivity toward Ser190 proteases and against Ala190 proteases without substantially increasing molecular weight is described.

Full text of DOI:10.1021/jm010244+

3856
J . Med. Chem. 2001, 44, 3856-3871
Exp loitin g Su bsite S1 of Tr yp sin -Lik e Ser in e P r otea ses for Selectivity: P oten t
a n d Selective In h ibitor s of Ur ok in a se-Typ e P la sm in ogen Activa tor
Richard L. Mackman,*,† Bradley A. Katz,‡ J . Guy Breitenbucher,§ Hon C. Hui,† Erik Verner,† Christine Luong,‡
Liang Liu,| and Paul A. Sprengeler*,‡
Departments of Medicinal Chemistry, Structural Biology, and Preclinical Sciences, Axys Pharmaceuticals Inc.,
180 Kimball Way, South San Francisco, California 94080
Received J une 1, 2001
A nonselective inhibitor of trypsin-like serine proteases, 2-(2-hydroxybiphenyl-3-yl)-1H-indole-
5-carboxamidine (1) (Verner, E.; Katz, B. A.; Spencer, J .; Allen, D.; Hataye, J .; Hruzewicz, W.;
Hui, H. C.; Kolesnikov, A.; Li, Y.; Luong, C.; Martelli, A.; Radika. K.; Rai, R.; She, M.; Shrader,
W.; Sprengeler, P. A.; Trapp, S.; Wang, J .; Young, W. B.; Mackman, R. L. J . Med. Chem. 2001,
44, 2753-2771) has been optimized through minor structural changes on the S1 binding group
to afford remarkably selective and potent inhibitors of urokinase-type plasminogen activator
(uPA). The trypsin-like serine proteases1 that comprise drug targets can be broadly categorized
into two subfamilies, those with Ser190 and those with Ala190. A single-atom modification,
for example, replacement of hydrogen for chlorine at the 6-position of the 5-amidinoindole P1
group on 1, generated up to 6700-fold selectivity toward the Ser190 enzymes and against the
Ala190 enzymes. The larger chlorine atom displaces a water molecule (H2O1S1) that binds near
residue 190 in all the complexes of 1, and related inhibitors, in uPA, thrombin, and trypsin.
The water molecule, H2O1S1, in both the Ser190 or Ala190 enzymes, hydrogen bonds with the
amidine N1 nitrogen of the inhibitor. When it is displaced, a reduction in affinity toward the
Ala190 enzymes is observed due to the amidine N1 nitrogen of the bound inhibitor being
deprived of a key hydrogen-bonding partner. In the Ser190 enzymes the affinity is maintained
since the serine hydroxyl oxygen OγSer190 compensates for the displaced water molecule. High-
resolution crystallography provided evidence for the displacement of the water molecule and
validated the design rationale. In summation, a novel and powerful method for engineering
selectivity toward Ser190 proteases and against Ala190 proteases without substantially
increasing molecular weight is described.
In tr od u ction
S3, S2, S1, S1′, S2′, S3′, etc. which bind corresponding
peptides of the physiological substrate P4-P1, on one
side of the scissile bond, and P1′-P4′ on the other side.4
The trypsin-like serine proteases specifically bind argi-
nine or lysine residues in subsite S1. The remaining
pockets around the S1 subsite enable these enzymes to
refine their preference for a more specific sequence of
peptides and therefore a more defined substrate. Small-
molecule inhibitors of these particular enzymes have
been traditionally designed along very similar prin-
ciples, with most of the noncovalent, reversible, small-
molecule inhibitors possessing a basic charged moiety,
such as an amidine or guanidine, that binds in the S1
subsite, making strong interactions with the conserved
Asp189 residue. The very small S1 binding inhibitors
that have been reported,5-13 for example, 2-4 (Figure
1), rarely generate sufficient potency and selectivity to
be therapeutically useful. Consequently, potency and
selectivity are often enhanced by elaborating the small
inhibitors into larger molecules, designed to occupy
neighboring subsites, where the target and relevant
antitargets are different. For instance, many of the
potent and selective factor Xa inhibitors that have been
reported tend to span from the S1 to S4 subsites to
achieve sufficient potency and selectivity.14-18 A draw-
back of this design strategy to occupy several subsites
is that the final molecules can have higher molecular
weights, poorer absorption, distribution, metabolism,
The selectivity of small-molecule inhibitors toward a
protein or enzyme target is often of crucial importance
in the development of therapeutically useful molecules.
Engineering high selectivity can be especially challeng-
ing when the site of interaction between the inhibitor
and the enzyme is within a highly homologous region
of a large family of enzymes, such as the catalytic site
of trypsin-like serine proteases. Within this family of
enzymes there are several important therapeutic tar-
gets, for example, urokinase-type plasminogen activator
(uPA), which is implicated in tumor progression and
metastasis,2 and the coagulation enzyme factors Xa,
VIIa, and IIa (thrombin). The treatment of the disorders
in which these enzymes are involved requires selective
inhibitors to reduce the risk of nonspecific in vivo
toxicities, thereby providing a higher chance of clinical
success.3
The serine proteases select their substrates by using
a sequence of subsites or specificity pockets denoted S4,
* To whom correspondence should be addressed. (R.L.M.) Phone:
(650) 829-1191. Fax: (650) 829-1019. E-mail: richard•mackman@
axyspharm.com. (P.A.S.) Phone: (650) 829-1267. Fax: (650) 829-1019.
E-mail: paul•sprengeler@axyspharm.com.
† Department of Medicinal Chemistry.
‡ Department of Structural Biology.
§ Current address: RW J ohnson PRI, 3210 Merryfield Row, San
Diego, CA 92121.
| Department of Preclinical Sciences.
10.1021/jm010244+ CCC: $20.00 © 2001 American Chemical Society
Published on Web 10/06/2001

Exploiting S1 of Serine Proteases for Selectivity
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23 3857
Ta ble 1. Effect of Modifications on the P1 Group toward Enzyme Inhibition
Ki (µM)
FVIIa
compd
no.
R1
R2
formula
vola
uPA
plasmin
trypsin
FXa
tPA
p-kall.
thrombin
1
H
Cl
F
Me
OMe
OH
H
H
H
H
H
H
H
Cl
C21H17N3O
0.0
17.5
3.6
27.2
35.7
9.0
0.008
0.009
0.020
0.10
0.36
2.0
0.10
0.11
0.35
0.21
3.9
0.13
0.23
1.1
1.6
14
0.16
1.7
0.75
7.0
0.078
19
1.3
>75
26
0.035
8.8
3.6
39
18
46
0.0019
0.80
0.11
4.8
0.60
1.3
0.32
60
7.5
>75
>75
>75
65
10a
10b
10c
10d
10e
11
C
C
21H16ClN3O
21H16FN3O
C22H19N3O
C
C
22H19N3O2
21H17N3O2
16
>75
5.5
19
3.0
50
4.7
>75
C21H16ClN3O
17.5
0.49
5.2
3.3
1.3
a
Increase in van der Waals volume for H-R1 compared to H-H. The van der Waals volume of H-R1 is calculated using the program
MOE, and then the volume for H-H, calculated using the same software, is subtracted to estimate the overall increase in van der Waals
volume that is created by the substitution R1.
The ability to rationally design inhibitors capable of
distinguishing between the Ser190 and Ala190 sub-
classes of enzymes is a potentially powerful method for
generating selective inhibitors between physiologically
related enzymes such as uPA (Ser190) and tPA (Ala190)
or factor VIIa (Ser190) and factor Xa/thrombin (Ala190).
By applying our findings on small S1 amidine-based
binding inhibitors that interact with residue 190,13 onto
a novel, nonselective protease inhibitor, 1,1 we have
established a method for engineering remarkable se-
lectivity toward Ser190 enzymes and against Ala190
enzymes. A single-atom halogen substitution at the
6-position of the 5-amidinoindole group displaces a
F igu r e 1. Examples of nonpeptidic small-molecule inhibitors
of serine proteases.
hydrogen-bonded water molecule in S1 and produces up
to 6700-fold selectivity between these subclasses of
enzymes. The design rationale, synthesis, and structural
characterization using high-resolution crystallography
and excretion properties (ADME), and often a more
challenging synthesis.
are presented in this paper. Additionally, the relation-
Selectivity optimization within subsite S1, a site that
ship between binding affinity and the increase in the
is readily accessible and occupied by most, if not all,
van der Waals size of the group that is introduced is
serine protease inhibitors, could potentially alleviate
discussed, along with the general application of this
these drawbacks. Despite the high homologies within
novel method for developing selective serine protease
this subsite, there are some differences in shape, size,
inhibitors.
and hydrophobicity that provide each enzyme with a
distinct S1 subsite that can be exploited for specific-
ity.13,19 For instance, residue 190 is commonly repre-
Ra tion a le
The small-molecule inhibitor 1, previously reported,1
displays a potency ranging from 2 to 320 nM toward a
panel of eight serine proteases (Table 1). High-resolution
crystal structures of 1 and related inhibitors22 com-
plexed in uPA (Figures 2a and 3a(i)), trypsin (Figure
3b(i)), and thrombin (Figure 3c(i)) indicate that 1 binds
to the catalytic site of the protease and spans from S1
to S1′ across the catalytic machinery.1 Two key sites of
interaction between the inhibitor and the enzyme are
direct or water-mediated hydrogen bonds to Asp189 and
a cluster of short (<2.3 Å) and normal range hydrogen
sented as serine, alanine, or threonine within this family
of enzymes, the most common targets and antitargets
falling into either the Ser190 (uPA, plasmin, trypsin,
factor VIIa) or Ala190 (factor Xa, thrombin, plasma-
kallikrein (p-kallikrein), tPA) subclasses. In an earlier
paper we describe how a small set of amidine-based S1
binding inhibitors such as 2 and 3 achieve modest
selectivity toward the Ser190 enzymes, due, in part, to
the formation of an additional hydrogen bond between
13
one of the amidine nitrogens and OγSer190
.
Selectivity
can also be observed in small arylguanidine-based S1
inhibitors such as 4, which have good selectivity toward
uPA and against plasmin, tPA, thrombin, and factor
Xa.12,20,21 Although this selectivity preference toward
uPA has eluded satisfactory explanation, it may be due,
in part, to the difference in residue 190. Nevertheless,
the result remains consistent with the enzymes pos-
sessing distinct S1 pockets that provide opportunities
for selectivity optimization.
bonds involving the inhibitor, a water molecule (H2Ooxy
)
trapped in the oxyanion hole, a distally bound water
molecule (H2Odis), and OγSer195 (Figures 2a and 3a(i)).1
These interactions with residues that are common to
all trypsin-like serine proteases enable 1 to generate
high affinity toward many enzymes in the family while
maintaining a low molecular weight (327 amu). For
example, the affinity toward uPA (Ki ) 8 nM, Table 1)

3858 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23
Mackman et al.
F igu r e 2. Stereoview X-ray density map of (a) 1 in uPA and (b) 10a in uPA. In (a) the hydrogen bonds near the catalytic Ser195
are colored as follows: short hydrogen bonds are in cyan, the intra-enzyme and intra-inhibitor in yellow, the other normal range
hydrogen bonds in magenta. In S1 the water molecule H2O1S1 is hydrogen bonded to OγSer190 and amidine nitrogen N1 (green).
Amidine nitrogen N1 also donates a hydrogen bond to OγSer190 (yellow). The amidine binds to Asp189 through bridging water
molecule H2O2S1. In (b), the density for the chlorine is clearly visible and the amidine N1 makes a full hydrogen bond to OγSer190
(white). The inhibitor also makes direct interactions with Asp189. Ser195 moves to occupy the oxyanion hole and is now involved
in normal range hydrogen bonds to Gly193 and the inhibitor hydroxyl. This movement of Ser195 has displaced two water molecules
that are clearly visible in (a) (H2Ooxy is the nearest water molecule, and H2Odis is the one furthest back).
is exceptional for a reversible inhibitor of this size and
makes 1 an excellent scaffold on which to build selectiv-
ity.
with the inhibitor 1 (Figures 2a and 5). It was reasoned
that a group introduced at the 6-position of the inhibitor,
such as a chlorine, designed to displace H2O1S1 (Figure
5), could impart selectivity since the residues that flank
H2O1S1 are different between closely related targets and
antitargets, e.g., uPA and tPA. Displacement of a tightly
bound water molecule can also be expected to provide
an entropic advantage toward binding.23 Furthermore,
the crystal structures of the Ser190 enzymes bound by
1 (Figures 2a and Figure 3a-c(i)) indicate that the
amidine nitrogen N1 has partial hydrogen bonds to the
cobound water (H2O1S1) and OγSer190, but, in contrast,
the Ala190 enzymes accommodate the lack of OγSer190
by the formation of a full hydrogen bond between the
amidine nitrogen N1 and H2O1S1. This subtle difference
in hydrogen bond length emphasizes the importance of
the water molecule toward binding in the Ala190
enzymes. The 6-chloro-substituted analogue of 1 and
also other 6-substituted 5-amidinoindoles were therefore
targeted to test this hypothesis.
To maintain the low molecular weight characteristic
of 1, attention was focused onto the S1 pocket binding
motif to improve selectivity. The overall homology and
sequence identity for the S1 site region between uPA
and several other serine protease enzymes are shown
in Figure 4 and Table 2. The highest homology is
between uPA and tPA, which is not surprising since
they share the same physiological substrate, plasmino-
gen. The homology is much higher (72-89%) when only
the residues that contribute to the architecture of the
S1 subsite are considered (Table 2). Despite these high
homologies within the S1 region, the size and shape of
the S1 subsite can differ even when the residues are
almost identical.13 For example, the depth of the uPA
S1 pocket along the Ser195-Asp189 axis is much
greater than that of trypsin notwithstanding the high
similarity (94%, Table 2). Two of the residues that
contribute to the homology variations in the S1 subsite
between uPA and the other enzymes are residues 190
and 213. In general, the trypsin-like serine proteases
can be divided into four subclasses on the basis of the
identity of residue 190: Ser190, Ala190, Thr190, and
other residues. All the common targets and antitargets,
mentioned herein, fall into either the Ser190 or Ala190
category (Table 2). The 190 and 213 residues both flank
a tightly bound water molecule, H2O1S1, that is cobound
Ch em istr y
The synthesis of 1, described earlier,1 involved the use
of the Fisher indolization procedure.24 Applying the
same strategy to the synthesis of the desired 2,5,6-
trisubstituted indoles yielded a mixture of regioisomers,
due to the asymmetric nature of the hydrazone inter-
mediate 9 (Scheme 1). For example, the condensation
of acetophenone 8 with hydrazine 7 to give the asym-

Exploiting S1 of Serine Proteases for Selectivity
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23 3859
F igu r e 3. Side by side binding in (a) uPA, (b) trypsin, and (c) thrombin (modeled from the isosteric benzimidazole version of 1
in thrombin32) of 1 (i) and 10a (ii). In (a)(ii), (b)(ii), and (c)(ii) note the loss of H2O1S1 in all the 10a complexes and also the tilt of
the inhibitor 10a compared to 1. In (a) note that three more water molecules are displaced. Ser195 moves into the oxyanion hole
and displaces H2Ooxy and H2Odis. Asp189 moves to make direct interactions with the amidine, thus displacing H2O2S1. In (b)
Ser195 moves into the oxyanion hole similar to (a) and displaces H2Ooxy and H2Odis. In (c) the complex of 1 is modeled on the basis
of the complex of the isosteric benzimidazole version of 1 in thrombin.32 Only one water molecule is displaced, notably H2O1S1
.
Ser195 does not move, and H2Ooxy and H2Odis are not displaced.
metric hydrazone 9, followed by polyphosphoric acid
(PPA) mediated Fisher indolization, resulted in an
equimolar mixture of 6-chloro-2-(2-hydroxybiphenyl-3-
yl)-1H-indole-5-carboxamidine (10a ) and 4-chloro-2-(2-
hydroxybiphenyl-3-yl)-1H-indole-5-carboxamidine (11)
(1:1 ratio determined by HPLC methods). These isomers
could only be separated by careful preparative HPLC,
resulting in a substantial loss of material, thus prompt-
ing the investigation of an alternative, more regiospe-
cific method for the synthesis of 2,5,6-trisubstituted
indoles. Buchwald and co-workers found that milder
Fisher indolization conditions can afford good regiocon-
trol of the indolization step following hydrazone forma-
tion.25 Unfortunately the milder conditions do not effect
cyclization when electron-withdrawing groups such as
amidine and chlorine are present. Substituted 2-arylin-
doles have been readily prepared using a palladium-
catalyzed coupling of an alkyne onto an o-iodoaniline
followed by cyclization of the resultant 2-ethynylaniline
to the indole heterocycle.26-30 For example, Yasuhara
et al. produced a variety of 2,5-substituted and 2,6-
substituted analogues very efficiently and in good yield
using this methodology.26 Furthermore, the conditions
appear to be compatible with many different functional
groups including halides and nitriles, which can be
readily transformed into amidines. The control of regio-
selectivity is provided by the substitution pattern on the
aryl halide, which can be easily controlled through the
selective synthesis of o-iodoanilines. A retrosynthetic
scheme in which the 6-substituent can be varied is
shown in Scheme 2. Synthesis would require a common

3860 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23
Mackman et al.
F igu r e 4. Sequence homology among several trypsin-like serine proteases compared to uPA. The chymotrypsin numbering system
was used throughout. OAH ) oxyanion hole, and BS ) â strand. Residues of the catalytic triad are marked by an asterisk.
Ta ble 2. Homology Comparisons between uPA and Related Enzymes
entire sequence
of residues (%)
residues within
S1 subsite
4.5 Å of 1b (%)
residuesc (%)
identical
similara
identical
similar
identical
similar
residue 190
tPA
plasmin
trypsin
p-kallikrein
factor Xa
factor VIIa
thrombin
45
37
37
33
30
30
25
54
48
47
45
43
39
36
72
76
76
72
72
80
64
84
76
80
76
76
84
72
78
89
89
78
83
89
72
83
89
94
78
89
89
72
alanine
serine
serine
alanine
alanine
serine
alanine
a
b
Residues were considered similar only if they had both similar charge/hydrophobicity and size. Twenty-five residues were included
in the comparison (see the boxed residues in Figure 4). c The 18 residues used for comparison are 189-195, 213-220, and 226-228.
phenylacetylene component, 13, and a variety of 2-sub-
stituted 4-amino-5-iodobenzonitriles, 12.
methyl group of nitrotoluene 15 was oxidized to acid
16 and then transformed to the nitrile 17 via the amide,
in a yield of 29% from 15. Reduction of the nitro group
then afforded the aniline 18 in high yield. Commercial
aniline 5 and aniline 18 were both converted to aryl
halides 12a and 12b in yields of 48% and 33%, respec-
tively, using N-iodosuccinimide (NIS) in acetic acid.
The MEM-protected phenol 13 has been previously
synthesized.1 The precursors 12a -e were obtained
according to Schemes 3-6. The 2-halogen-substituted
4-amino-5-iodobenzonitriles 12a and 12b were prepared
according to procedures outlined in Scheme 3. The

Exploiting S1 of Serine Proteases for Selectivity
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23 3861
Sch em e 3a
F igu r e 5. Close-up view of the base of the S1 subsite in the
uPA-1 complex. H2O1S1 is situated in a small pocket in the
upper right which can be accessed most readily from the
6-position of the heterocyclic ring. The size of the small pocket
occupied by H2O1S1 is close to the van der Waals volume of
chlorine.
a
Reagents and conditions: (a) CrO3, 18 N H2SO4, AcOH, 100
°C; (b) (COCl)2, THF, DMF; (c) NH3(g), THF, 0 °C; (d) TFAA, THF;
(e) H2, 10% Pd/C, MeOH; (f) NIS, AcOH.
Sch em e 1a
Sch em e 4a
a
Reactions and conditions: (a) NaNO2, 18 N H2SO4, AcOH; (b)
CuSO4, KCN, H2O, benzene, NaHCO3; (c) SnCl2‚2H2O, dioxane,
12 N HCl; (d) NIS, AcOH.
a
Reagents and conditions: (a) Me3Al, NH4Cl, p-xylene, reflux;
(Scheme 5). Reduction of the nitro group afforded the
aniline 25, which was then treated with NIS to give the
desired product 12d , in 29% yield. Alternatively, the
phenol 23 could be converted to benzyl ether 26 and
then reduced to provide aniline 27 in a high overall yield
(Scheme 6). Subsequent treatment with NIS then af-
forded 4-amino-5-iodo-2-(benzyloxy)benzonitrile (12e) in
80% yield.
A variety of N-substituted anilines such as carbam-
ates and mesylates have been used in the palladium-
mediated coupling of the alkyne onto the aryl halide
(Scheme 7). The work of Yasuhara et al.26 indicated that
(b) NaNO2, 6 N HCl, 0 °C, then SnCl2‚2H2O; (c) 8,1 EtOH, Et3N,
reflux; (d) PPA, 155 °C.
Aniline 19 was transformed into benzonitrile 20 in 91%
yield through cyanide displacement of the diazonium
salt (Scheme 4). Tin(II) chloride reduction of 20 afforded
aniline 21, which was then converted to 4-amino-5-iodo-
2-methylbenzonitrile (12c) in 53% overall yield using
NIS in acetic acid. To prepare 12d , the 1,2-methylene-
dioxy ring of 22 was opened using sodium cyanide in
HMPA,31 and treatment of the phenol 23 with methyl
iodide gave methyl ether 24, in 28% yield over two steps
Sch em e 2. Proposed Regiospecific Synthesis of 6-Substituted 5-Amidino-2-arylindoles

3862 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23
Mackman et al.
Sch em e 5a
yielded indoles 29a and 29c in yields of 52% and 26%,
respectively (Table 3, method A). The same palladium-
(II)-mediated coupling reaction employing the BOC
derivatives 28b, 28d , and 28e initially gave the coupled
biarylalkyne intermediate prior to formation of the
indole heterocycle.26 These intermediates were then
treated with TBAF30 to induce cyclization and simul-
taneous removal of the BOC group, resulting in good
yields (>44%) of indoles 29b, 29d , and 29e (method B).
In the examples shown in Table 3, the BOC-substituted
anilines were generally easier to prepare and tended to
give better overall yields of the final products. The
conversion of the nitrile groups to amidines was per-
formed using two general procedures (methods C and
D) involving hydroxylamine addition, acetylation of the
hydroxyamidine, and reduction. The difference between
the two procedures was the use of 10% palladium on
carbon catalyst rather than zinc dust for the simulta-
neous reduction of the acetoxyamidine intermediate and
the benzyl group in the preparation of the hydroxyl
analogue 10e. In all examples the final step was clean
and efficient removal of the MEM protecting group by
treatment with hydrochloric acid to yield indoles 10a -
e. On the basis of the results obtained (Table 3), the
route has general applicability toward the efficient
synthesis of a variety of 2,5,6-trisubstituted indoles.
a
Reagents and conditions: (a) HMPA, NaCN, 150 °C; (b) DMF,
CH3I, K2CO3; (c) SnCl2‚2H2O, dioxane, 6 N HCl, 0 °C; (d) NIS,
AcOH.
Sch em e 6a
Resu lts a n d Discu ssion
The analogues 10a -e and 11 were tested for activity
toward a panel of eight human serine proteases (Table
1). The enzymes represent a cross section of targets and
antitargets from the coagulation cascade and the uPA
system, and also the Ala190 and Ser190 subclasses of
enzymes. Compared to 1, the incorporation of a chlorine
atom adjacent to the amidine, 10a , only marginally
changed the affinity toward the Ser190 enzymes uPA,
plasmin, trypsin, and factor VIIa (all <11-fold), but
toward the Ala190 enzymes factor Xa, p-kallikrein,
thrombin, and, importantly, tPA, the affinity was
dramatically reduced by 244-, 421-, 188-, and 251-fold,
respectively. The magnitude of these changes in binding
affinity creates a “switchlike” effect whereby the chlo-
a
Reagents and conditions: (a) BnBr, DMF, K2CO3; (b) SnCl2‚
2H2O, dioxane, 6 N HCl; (d) NIS, CH3CN.
both BOC- and mesyl-substituted anilines can afford
2-arylindoles in high yields. The benzonitriles 12a -e
were therefore converted into either the mesyl- or BOC-
substituted anilines 28a -e in yields ranging from 42%
to 64% (Scheme 7). The palladium(II)-mediated coupling
of the alkyne with the mesyl-substituted anilines 28a
and 28c generated intermediate alkynes which cyclized
in situ to the respective N-mesylindoles. Removal of the
mesyl groups by treatment with sodium hydroxide
Sch em e 7a
a
Reagents and conditions: (a) MsCl, DIEA, CH2Cl2, DMAP; (b) di-tert-butyl dicarbonate, DIEA, THF or DMF, then 50% NaOH; (method
A) DMF, Pd(PPh3)2Cl2, Et3N, CuI, 13; then NaOH; (method B) DMF, Pd(PPh3)2Cl2, Et3N, CuI, 13; then TBAF; (method C) 50% NH2OH(aq),
EtOH, reflux; then Ac2O, AcOH; then zinc dust; then 4 N HCl, dioxane, MeOH; (method D) 50% NH2OH(aq), EtOH, reflux; then Ac2O,
AcOH; then H2, 10% Pd/C; then 4 N HCl, dioxane, MeOH.
Ta ble 3. Formation of Indoles Using Palladium-Mediated Coupling of Alkynes with Aryl Halides
aniline
R1
X
method
product
yield (%)
method
product
X
yield (%)
28a
28b
28c
28d
28e
SO2Me
CO2 Bu
Cl
F
Me
OMe
OBn
A
B
A
B
B
29a
29b
29c
29d
29e
52
44
26
69
54
C
C
C
C
D
10a
10b
10c
10d
10e
Cl
F
Me
OMe
OH
20
20
9
58
27
t
SO2Me
t
CO2 Bu
t
CO2 Bu

Exploiting S1 of Serine Proteases for Selectivity
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23 3863
rine atom essentially turns “off” (or significantly re-
duces) the inhibitory activity of 10a toward all the
Ala190 proteases. The single-atom change has therefore
produced a highly potent, low molecular weight, selec-
tive inhibitor of uPA, with between 90- and 6700-fold
selectivity against the Ala190 proteases, and between
12- and 200-fold selectivity against the other Ser190
proteases. The 4-position chlorine analogue 11 had
reduced affinity toward all the enzymes including uPA.
It is worth noting that 1 is a 2-fold less effective
inhibitor of the Ser190 protease factor VIIa, compared
to Ala190 protease factor Xa, but the chlorine analogue
10a is a 10-fold better inhibitor of factor VIIa compared
to factor Xa. This moderate 20-fold effect may be of
importance in the development of potent and selective
factor VIIa inhibitors.
structural changes, the entropy gain of displacing four
water molecules offsets any unfavorable effects on
potency and results in a highly potent inhibitor toward
uPA (Table 1).
In the thrombin-10a complex model (Figure 3c(ii)),
the water molecule H2O1S1 that is cobound in the
complexes of thrombin with two related inhibitors and
therefore presumed to be present in the binding of 1 to
thrombin1,32 is displaced, leaving the amidine without
an important hydrogen bond partner. In contrast to the
uPA-10a complex, the water molecules H2Ooxy and
H2Odis are not displaced in the thrombin-10a complex
model, and the short hydrogen bonds between the
inhibitor hydroxyl, OγSer195, and H2Ooxy remain intact.
Therefore, in thrombin only one water molecule is
displaced by 10a , compared to 1 (Figure 3c(i)), and no
major structural changes occur in the position of resi-
dues Asp189 and Ser195.
Com p a r ison s betw een th e Bin d in g Mod es of
In h ibitor s 10a a n d 1. The complexes of 10a in the
Ser190 enzymes uPA and trypsin and Ala190 enzyme
thrombin were determined by X-ray methods and
compared to the uPA and trypsin complexes of 1, and a
model of the thrombin-1 complex. The thrombin-1
complex has not been solved, so the complex of the
isosteric 5-amidinobenzimidazole analogue of 1 was
used to model the thrombin-1 complex.32 Earlier results
comparing benzimidazole- and indole-based S1 binding
groups1 indicate that the binding modes of inhibitors
in which only this group is changed are essentially
identical. The interactions in the uPA-10a complex
(Figures 2b and 3a(ii)) differ quite significantly from
those in the uPA-1 complex (Figures 2a and 3a(i)). In
the uPA-1 complex two water molecules are cobound
in S1, the water molecule targeted for displacement, H2-
O1S1, and a second water molecule, H2O2S1, that is
recruited to bridge the interaction between the amidine
group of the inhibitor and Asp189. This second water
molecule is only found in the uPA-1 complex since the
S1 subsite of uPA is deeper than that of trypsin and
thrombin. The inhibitor is prevented from moving
deeper into the S1 pocket to form a direct interaction
with Asp189 by the unique cluster of short and normal
range hydrogen bonds between the inhibitor and
Ser195.1,22 In the uPA-10a complex, both of the water
molecules cobound in S1 are displaced, leaving one full
hydrogen bond between the amidine nitrogen N1 and
OγSer190. This is in contrast to the two longer range
hydrogen bonds that exist in the uPA-1 complex
between N1 and H2O1S1 and OγSer190. Rather unexpect-
edly, the short hydrogen bond network outside S1 is
sacrificed to allow the inhibitor to bind 0.8-1.0 Å deeper
into the S1 pocket. Residue Asp189 shifts 0.5 Å toward
the inhibitor to form direct interactions with the ami-
dine, thereby displacing H2O2S1 (Figures 2b and 3a(ii)).
The breakdown of the short hydrogen bond cluster
outside S1 is also accompanied by the movement of
Ser195 to occupy the oxyanion hole, thus displacing
H2Ooxy and also H2Odis. The short hydrogen bond that
existed between the inhibitor phenolate and OγSer195 in
the uPA-1 complex is now an ordinary length (2.60 Å)
hydrogen bond in the uPA-10a complex. In summation,
a total of four ordered water molecules are displaced,
while the protein experiences significant shifts of two
key residues, Asp189 and Ser195, leading to a break-
down of the short hydrogen bond cluster. Despite these
The complex of 10a in trypsin also confirms that the
H2O1S1 water molecule has been displaced (Figure
3b(ii)). Similar to the uPA-10a complex, H2Ooxy and
H2Odis are also displaced, resulting in the absence of a
total of three water molecules. The complex of 1 in
trypsin displays direct interactions between the amidine
and the Asp189 due to its shallower S1 subsite relative
to that of uPA. All three complexes involving 10a
therefore confirm the displacement of H2O1S1 as in-
tended in the design of the inhibitor.
The combined structural information clearly reveals
the basis for the selectivity enhancements that are
observed in the enzyme assays. In the Ser190 proteases
the loss of H2O1S1 is accompanied by the formation of a
full hydrogen bond between nitrogen N1 and OγSer190
.
In contrast, inhibitors that displace H2O1S1 in the
Alal90 enzymes such as thrombin leave the amidine in
an environment devoid of both the water molecule and
OγSer190 as hydrogen-bonding partners (Figure 6). This
unfavorable situation is manifested in significantly
reduced affinities toward the Ala190 enzymes, and
hence selective inhibition toward the Ser190 enzymes
is generated. Although X-ray structures are not avail-
able for all the Ala190 enzymes, it is highly likely that
the water molecule is displaced in tPA, factor Xa, and
p-kallikrein. Therefore, the displacement of H2O1S1 by
the appropriate chemical modification of the indole ring
is a powerful way of tuning the selectivity of amidine-
based protease inhibitors toward Ser190 enzymes and
against Ala190 enzymes.
Rela tion sh ip of Su bstitu en t Size tow a r d In h ibi-
tor Bin d in g. All the proteases have an S1 subsite that
is subtly different in size and shape. It has already been
noted that the S1 subsites of uPA and trypsin have
different sizes upon inhibitor binding, despite both
enzymes having similar residues lining the S1 pocket
(Table 2).13 Furthermore, the movement of residue 189,
in the case of the uPA-10a complex, also implies that
there is some flexibility in the position of the S1 subsite
residues, which will affect the overall size of the
inhibitor bound S1 pocket. The introduction of groups
both smaller (fluorine, hydroxyl) and larger (methyl,
methoxy) than chlorine provides a method for probing
the relationship between the size of the P1 group on the
inhibitor and binding affinity. The chemistry scheme,
used to prepare 10a , was easily modified to introduce

3864 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23
Mackman et al.
toward the Ala190 proteases was reduced by margins
of 17-fold (factor Xa), 103-fold (tPA), 116-fold (p-kal-
likrein), and 23-fold (thrombin). Although not as im-
pressive as the results with the chlorine analogue, these
differences still produce a molecule with selectivities
ranging from 6- to 441-fold against all the other
enzymes while maintaining a high potency toward uPA
(Ki ) 20 nM). The uPA-10b complex is markedly
different from the uPA-10a complex as a result of the
reduced size of the fluorine group (Figure 7). Neither of
the two water molecules cobound in subsite S1, H2Ooxy
nor H2Odis, is displaced. The mode of binding in the
uPA-10b complex is therefore very similar to that of
the uPA-1 complex. This is in stark contrast to the
trypsin-10b complex in which the fluorine does displace
32
the water molecule H2O1S1
.
The difference between
the uPA and trypsin structures regarding the inclusion
or exclusion of H2O1S1 is a reflection of the greater depth
of the uPA pocket compared to trypsin, which allows
uPA to accommodate both the fluorine and water
simultaneously. It is presumed that, in the thrombin-
10b complex, the water molecule H2O1S1 would also be
absent. Therefore, in uPA, which has the deepest and
largest S1 subsite, the transition between the H2O1S1
cobound and unbound states occurs upon changing from
fluorine to chlorine. This transition occurs upon moving
from hydrogen to fluorine in trypsin and, most likely,
thrombin. Since selectivity is enhanced in the fluorine
analogue 10b, even though H2O1S1 remains in uPA, it
follows that it is only necessary to displace the water
molecule from the Ala190 enzymes and not the Ser190
enzymes to achieve the desired selectivity effect.
F igu r e 6. Schematic displaying the basis for selectivity
enhancement based on the binding of 10a in trypsin and
thrombin. The typical binding of a 6-unsubstituted amidine,
e.g., 1, to a Ser190 protease such as trypsin is shown in the
upper left panel. The amidine N1 atom (blue) has hydrogen
bonds to both H2O1S1 and OγSer190 (red). The upper right panel
shows the typical binding of 1 to an Ala190 protease such as
thrombin in which the amidine nitrogen N1 has only one
hydrogen bond to H2O1S1. The lower panels show the situation
in which the 6-chlorine substituted analogue 10a displaces H2-
O1S1 in both the Ser190 (lower left) and Ala190 enzymes (lower
right). In the lower left panel the binding of the 6-chlorine
analogue 10a displaces H2O1S1 and forms one (full) hydrogen
bond to OγSer190. In the lower right panel the binding of the
6-chlorine analogue 10a to the Ala190 proteases leaves the
amidine N1 without a full complement of hydrogen-bonding
partners, just the Asp189. The hydrogen bond from amidine
nitrogen N1 to the Asp189 oxygen remains in all cases.
Investigation of larger groups such as methyl, 10c,
and methoxy, 10d , allowed more trends to be observed
between inhibitor size and affinity. It is assumed in the
following discussion that the water molecule H2O1S1 is
always displaced by the groups larger than chlorine on
the basis of the finding that chlorine displaces the water
molecule in uPA, which has a large S1 subsite compared
to the other proteases. The van der Waals volume of
H-F, H-Cl, H-OH, H-OMe, and H-Me was calcu-
lated [using the software MOE 2000.02 from Chemical
Computing Group, Inc.] and compared to that of hydro-
gen H-H (Table 1) to estimate the relative increase in
inhibitor size in the region of residue 190 within S1. In
general, a clear trend of reduced affinity with increasing
several different groups at the 6-position (see the
Chemistry section).
The smaller fluorine analogue 10b, compared to the
chlorine analogue 10a , had approximately the same
affinity toward all the Ser190 proteases. The affinity
F igu r e 7. uPA complexes of (i) 1 and (ii) 10b. The inhibitors have essentially the same mode of binding to uPA. The fluorine
analogue 10b does not displace H2O1S1 and H2O2S1, within subsite S1. Similarly the hydrogen bond cluster involving Ser195 and
His57, the inhibitor hydroxyl group, and water molecules H2Ooxy and H2Odis is also unaffected by the fluorine substituent.

Exploiting S1 of Serine Proteases for Selectivity
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23 3865
F igu r e 9. Plasma concentration vs time curve for 1 and 10a
following intravenous administration in male Sprague-Daw-
ley rats.
F igu r e 8. Alternative uPA inhibitors.
substituent size exists toward the Ser190 enzymes. This
trend is not surprising on the basis of the limited space
within the S1 subsite regardless of the flexibility of some
of the residues. In contrast, a subtly different trend can
be observed in affinity toward the Ala190 enzymes.
From fluorine substitution through methyl substitution,
a good correlation can be observed between affinity and
increased size of the substituent. However, in the case
of the methoxy analogue 10d , three of the four Ala190
enzymes and possibly even the fourth have a greater
affinity than the methyl analogue. This is most likely a
reflection of the smaller size of the Ala190 group,
compared to Ser190, which allows the S1 subsite to
interact more favorably with the larger methoxy group.
A consequence of the increased affinity of the methoxy
analogue toward Ala190 enzymes and the corresponding
decrease toward the Ser190 enzymes is that the selec-
tivity is reduced. Therefore, for optimal selectivity the
size of the substituent is also important, in addition to
displacement of the water molecule H2O1S1 in the
Ala190 enzymes.
The hydroxyl analogue 10e stands out as being the
poorest inhibitor toward many of the enzymes despite
a calculated van der Waals size between those of
fluorine and chlorine. This indicates that affinity toward
the target enzyme is also dependent on other subtle
factors. One prior example of hydroxyl substitution
adjacent to an amidine was reported by Choi-Sledeski
et al.17 and resulted in a 21-fold drop in potency toward
factor Xa. There are several potential effects that
6-position substitutions can have on the binding of the
inhibitor toward a particular enzyme. The nature of the
substituent and the close proximity to the amidine
group generate different desolvation effects upon bind-
ing. The hydroxyl analogue is presumed to be highly
solvated in solution23 and therefore experiences a large
desolvation effect when placed in the relatively nonpolar
S1 subsite. The crystallographic data of the 10a com-
plexes also indicate that the dihedral angle between the
amidine and the indole ring changes upon 6-substitution
due to steric factors. In the uPA-1 complex the amidine
dihedral angle is -18°, but this changes to +32° in the
uPA-10a complex. In the thrombin-10a complex the
amidine is coplanar, experiencing steric strain which
is offset by the formation of a hydrogen bond from the
chlorine to the nitrogen proton.32 Finally, it is also
expected that the pKa of the amidine group, and
potentially the phenol group, will change due to the
different electronegative effects exerted by each differ-
ent substituent. Despite these factors, it is apparant
that relatively small, nonpolar groups, e.g., fluorine and
chlorine, possess the appropriate characteristics for
maintaining optimal potency toward the Ser190 en-
zymes while simultaneously displacing H2O1S1 in the
Ala190 enzymes to generate selectivity.
Gen er a l Ap p lica tion to th e Design of Selective
P r otea se In h ibitor s. Two of the most highly re-
searched Ser190 enzyme targets for drug intervention
are uPA and factor VIIa. Since it was clear that the
fluorine and chlorine analogues held the most promise
toward distinguishing between Ser190 and Ala190
proteases such as uPA/tPA and factor VIIa/factor Xa,
these groups were translated onto other amidine-based
inhibitors of uPA (Figure 8) and factor VIIa [unpub-
lished results]. The 5-amidinobenzimidazole-based uPA
inhibitor 30 was found to be amenable to introduction
of a chlorine or fluorine atom and has yielded another
class of highly potent and selective uPA inhibitors.32
Introduction of a fluorine or chlorine atom onto a
benzamidine class of uPA inhibitors based upon 31 was
met with limited success (Figure 8).32 Although selectiv-
ity was generated, a reduction in potency toward uPA
masked much of the effect. Naphthamidine-based uPA
inhibitors have also been considered as candidates for
selectivity enhancement. For instance, the recently
reported inhibitor 3233 (Figure 8) has good overall
selectivity against several enzymes but in our hands has
limited selectivity against the Ala190 protease p-kal-
likrein. Modeling of the 7-fluorine analogue of 32 in uPA
indicates that this would probably displace H2O1S1 and
lead to an inhibitor with improved selectivity.
In Vivo P r op er ties of 10a Com p a r ed to 1. The
pharmacokinetic profiles of 1 and 10a were measured
in Sprague-Dawley rats following intravenous admin-
istration to assess the effects on clearance by the
halogen substitution. The plasma concentration-time
curves are shown in Figure 9, and the pharmacokinetic
parameters are summarized in Table 4. The addition
of the chlorine atom favorably altered the pharmacoki-
netic parameters and allowed plasma concentrations
well above the Ki to be attained for periods of greater
than 6 h, even with low doses of compound. Inhibitor 1
was eliminated from plasma in a biphasic fashion with
a clearance of 121 (mL/min)/kg and a terminal half-life

3866 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23
Mackman et al.
Ta ble 4. Pharmacokinetics of 1 and 10a Following Intravenous Administration in Male Sprague-Dawley Rats
PK parametera
nb
dosing route
dose (mg/kg)
1
10a
PK parametera
1
10a
3
IV
1
3
IV
2
294 ( 7
59 ( 1
Vcd (mL/kg)
Vsse (mL/kg)
MRTf (min)
AUCg (mM·min)
t1/2h (min)
1560 ( 1190
11900 ( 3200
106 ( 49
24 ( 8
102 ( 41
346 ( 81
7700 ( 950
130 ( 14
85 ( 2
body weight (g)
286 ( 12
121 ( 40
CLpc (mL/min/kg)
189 ( 32
a
b
d
PK parameters ) mean ( SD values from three rats. n ) number of animals. c CLp ) plasma clearance. Vc ) volume of distribution
in the central compartment. Vss ) volume of distribution at steady state. f MRT ) mean residence time. AUC ) integrated area under
the plasma concentration vs time curve from time 0 to infinity. t1/2 ) terminal half-life.
e
g
h
cal microanalyses were carried out by Robertson Microlit
Laboratories Inc., Madison, NJ , and were within 0.4 unit of
the theoretical values unless otherwise indicated.
of 102 min, whereas the modified inhibitor 10a was
cleared from plasma more slowly in a triphasic fashion,
with a clearance of 59 (mL/min)/kg and a terminal half-
life of 189 min. In both cases the predominant metabo-
lite that was detected in the plasma was the glucu-
ronide. It is thought that these pharmacokinetic
parameters are sufficient to address the therapeutic
potential of a uPA inhibitor in a variety of tumor models,
especially on a twice daily dosing regimen. These studies
are ongoing and will be reported in due course.
2-Ch lor o-4-h yd r a zin oben za m id in e (7). Ammonium chlo-
ride (7.5 g, 140 mmol) was added to p-xylene (100 mL) and
the mixture cooled to 0 °C using an ice bath. A 2 M solution of
trimethylaluminum in toluene (70 mL, 140 mmol) was added
dropwise. The mixture was stirred at 0 °C for 30 min and then
allowed to warm to room temperature. The nitrile 5 (8.0 g, 52
mmol) was added portionwise to the solution and the mixture
refluxed for 24 h before being allowed to cool. The mixture
was poured onto a slurry of silica gel (200 g) and CHCl3 (200
mL), and then the silica gel was removed by filtration. The
silica gel was washed with 50% MeOH/CHCl3, and the
combined organic solutions were then concentrated under
reduced pressure. A sample was subjected to preparative
HPLC chromatography to give 6 as a pale yellow solid. 1H
NMR (DMSO-d6): δ 9.10 (br s, 3H), 7.26 (dd, J ) 8.7, 2.0 Hz,
1H), 6.74 (d, J ) 2.0 Hz, 1H), 6,62 (d, J ) 8.7 Hz, 1H).
MS-ESI: m/z 170 [35Cl, MH+]. The crude product 6 was added
to 6 N HCl (50 mL) and cooled to 0 °C using an ice/salt bath.
A solution of NaNO2 (4.2 g, 61 mmol) in water (15 mL) was
added dropwise over 10 min. The mixture was stirred for 20
min and then added to a solution of tin(II) chloride dihydrate
(16.6 g, 74 mmol) in 6 N HCl (25 mL) at -5 °C. The mixture
was stirred for 30 min and then concentrated under reduced
pressure to give a crude residue. Water was added and the
solution concentrated under reduced pressure. The crude
hydrazine product 7 was used in subsequent reactions without
further purification.
6-Ch lor o-2-(2-h yd r oxybip h en yl-3-yl)-1H-in d ole-5-ca r -
b oxa m id in e (10a ) a n d 4-Ch lor o-2-(2-h yd r oxyb ip h en yl-
3-yl)-1H-in d ole-5-ca r boxa m id in e (11). The crude hydrazine
7 (2.0 g) was added to a solution of acetophenone 81 (0.48 g,
2.26 mmol) in EtOH (20 mL) followed by the addition of
triethylamine (3.6 mL, 26 mmol). The solution was refluxed
for 3 h and then allowed to cool. A white solid formed which
was removed by filtration. The eluant was concentrated under
reduced pressure to give the crude hydrazone 9. The hydrazone
was heated in PPA (10 mL) at 155 °C for 24 h to give two
close-running products by analytical HPLC. The solution was
allowed to cool and was then treated with 6 N HCl (10 mL)
and CH3CN (10 mL). The solid that formed was removed by
filtration and the solution subjected to preparative HPLC to
give the title products 10a (11 mg, <2%) and 11 (6 mg, <1%).
The spectral data for 10a are shown in the alternative
synthesis shown later. Insufficient material was obtained to
allow for a full structural characterization of 11 apart from
mass spectrometry analysis. MS-ESI: m/z 361.8 [35Cl, MH+].
Con clu sion
In recent years much effort has been made to generate
novel S1 binding groups for trypsin-like serine proteases
with the main goal of increasing oral bioavailability.
Relatively little attention has been paid to this particu-
lar subsite for selectivity optimization, yet this study
reveals that manipulation of the S1 binding group can
also afford excellent selectivities toward subfamilies of
serine proteases. Structure-based design has led to the
discovery that a halogen atom insertion adjacent to the
amidine of a potent and generic 5-amidinoindole-based
protease inhibitor generates remarkable selectivity
toward Ser190 proteases and against Ala190 proteases
without substantially increasing the molecular weight.
This single-atom chemical modification has transformed
a potent nonselective inhibitor of uPA and tPA into a
highly potent and selective inhibitor of uPA. The ap-
plication of this novel and powerful method for enhanc-
ing selectivity is expected to accelerate the discovery of
more potent and selective inhibitors of uPA, factor VIIa,
and novel Ser190 proteases uncovered by the human
genome sequencing.
Exp er im en ta l Section
Reactions were carried out under an inert atmosphere at
room temperature unless indicated otherwise. Solvents and
chemicals were reagent grade or better. All reagents were
obtained from commercial sources unless otherwise indicated.
Silica gel chromatography was performed using Merck Kie-
selgel 60 (230-400 mesh). Preparative HPLC purification was
carried out on a Gilson HPLC fitted with a VYDAC C18 (5 ×
25 cm) column eluting at 50 mL/min with a gradient of 2%-
90% solvent B in solvent A over 60 min (solvent A was 20 mmol
of aqueous HCl, and solvent B was CH3CN). Analytical HPLC
analysis was performed using an HP1100 eluting with a
gradient of 2-90% solvent B in solvent A (solvent A was 0.05%
TFA in water, and solvent B was CH3CN). NMR spectroscopy
was determined on a GE 300 MHz NMR or a J EOL ECLIPSE
270 MHz NMR. Chemical shifts are reported in units of δ
(ppm), and peak multiplicities are expressed as follows; s,
singlet; d, doublet; t, triplet; br, broad peak; q, quartet; m,
multiplet; dd, doublet of doublets; ddd, doublet of doublets of
doublets. Mass spectra were determined on a Perkin-Elmer
Sciex API-150EX mass spectrometer equipped with a turbo
ion spray source (MS-Sciex) or a Finnigan TSQ7000 equipped
with an electrospray ionization source (MS-ESI). The analyti-
4-Am in o-2-ch lor o-5-iod oben zon itr ile (12a ). 4-Amino-2-
chlorobenzonitrile (5) (50 g, 0.3 mol) and NIS (77.4 g, 0.34 mol)
were combined in AcOH (350 mL), and the resultant mixture
was stirred overnight. The brown solid that formed was
collected by filtration, washed with hexanes, and dried to give
the title product 12a (43.5 g, 48%) as a pale brown solid. MS-
Sciex: m/z 278.3 [35Cl, MH+].
2-F lu or o-4-n itr oben zoic Acid (16). To a solution of
2-fluoro-1-methyl-4-nitrobenzene (15) (25 g, 0.16 mol) in AcOH
(250 mL) was added 18 N sulfuric acid (40 mL) dropwise over
10 min. The light yellow solution was heated to 100 °C and
treated over 1 h with chromium(VI) oxide (56.4 g, 0.56 mol)
dissolved in water (45 mL). The temperature of the green

Exploiting S1 of Serine Proteases for Selectivity
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23 3867
mixture was maintained at 100 °C for 1 h, and then the
mixture was stirred at ambient temperature overnight. The
green mixture was dissolved in water (1 L) and extracted with
Et2O (4 × 200 mL). The organic extracts were combined and
concentrated under reduced pressure to give a yellow solid.
Water (400 mL) and K2CO3 (25 g) were added, and the bright
yellow solution was washed with Et2O (2 × 200 mL). HCl (12
N) was added until pH 1, and the solid that formed was
collected by filtration. The solid was washed with water and
and 12 N HCl (3 mL) was added followed by tin(II) chloride
dihydrate (20.9 g, 90 mmol). The mixture was stirred for 4 h,
and then EtOAc was added (50 mL). The organic layer was
washed with water (100 mL). The aqueous layer was then
extracted with EtOAc (2 × 50 mL). The combined organic
extracts were washed with Na2CO3 solution and then dried
over Na2SO4. The organic solution was filtered and concen-
trated under reduced pressure to give 21 (5.0 g) as a crude
solid. The crude solid was dissolved in AcOH (50 mL) and
treated with NIS (8.5 g, 38 mmol). The mixture was stirred
for 3 h, then diluted with EtOAc (300 mL), and washed with
sodium thiosulfate solution (3 × 150 mL) and then 1 N NaOH
solution. The organic solution was dried over Na2SO4, filtered,
and concentrated under reduced pressure. The residue was
subjected to chromatography over silica gel eluting with 20%
EtOAc in hexanes to give the title product 12c (4.1 g, 53%
1
dried to give 16 (15.4 g, 52%) as a light yellow solid. H NMR
(DMSO-d6): δ 8.22-8.07 (m, 3H).
2-F lu or o-4-n itr oben zon itr ile (17). A solution of 2-fluoro-
4-nitrobenzoic acid (16) (15.4 g, 83 mmol) in THF (150 mL) at
0 °C was treated with oxalyl chloride (8.1 mL, 92 mmol) and
a few drops of DMF. The solution was stirred for 1 h at 0 °C
and then for 2 h at ambient temperature. The mixture was
concentrated under reduced pressure, and the crude solid was
dried. The solid was dissolved in THF (200 mL) and cooled to
0 °C. Ammonia gas was slowly bubbled through the mixture
for 30 min, and the mixture was then capped and stirred at 0
°C for 2 h. The mixture was added to saturated NaHCO3 (200
mL) and extracted with EtOAc (2 × 200 mL). The organic layer
was dried over MgSO4, filtered, and concentrated under
reduced pressure to give the benzamide (15.5 g, 100%) as a
1
from 20). H NMR (CDCl3): δ 7.79 (s, 1H), 6.56 (s, 1H), 4.50
(br s, 2H), 2.37 (s, 3H).
2-Hydr oxy-4-n itr oben zon itr ile (23). 1,2-(Methylenedioxy)-
4-nitrobenzene (22) (10.0 g, 60 mmol) was dissolved in HMPA
(50 mL) and heated to 150 °C. The stirred solution was treated
with NaCN (4.4 g, 90 mmol) over 20 min and then heated at
150 °C for a further 30 min. The solution was allowed to cool
and diluted with water and 1 N NaOH until pH 10. The
solution was washed with Et2O (2 × 200 mL). The aqueous
layer was treated with 1 N HCl until pH 6 and then extracted
with EtOAc (2 × 200 mL). The organic extracts were washed
with saturated NaHCO3 and brine, dried over Na2SO4, filtered,
and concentrated under reduced pressure. The residual solid
was subjected to chromatography over silica gel eluting with
30% EtOAc in hexanes to give the title product 23 (3.0 g, 31%).
1H NMR (DMSO-d6): δ 7.98-7.94 (m, 1H), 7.75-7.71 (m, 2H).
MS-Sciex: m/z 162.8 [M - H+].
2-Meth oxy-4-n itr oben zon itr ile (24). The phenol 23 (3.0
g, 18 mmol) was dissolved in anhydrous DMF (50 mL) and
treated with K2CO3 (5.0 g, 36 mmol) followed by MeI (1.4 mL,
22 mmol). The solution was stirred for 2 h and then diluted
with EtOAc (100 mL). The solution was washed with water
(100 mL) and then brine, dried over Na2SO4, filtered, and
concentrated under reduced pressure to give 24 (2.9 g, 89%).
1H NMR (CDCl3): δ 7.88 (dd, J ) 8.3, 2.1 Hz, 1H), 7.82 (d, J
) 2.0 Hz, 1H), 7.75 (dd, J ) 8.4, 0.5 Hz, 1H), 4.05 (s, 3H).
1
light yellow solid. H NMR (DMSO-d6): δ 8.21-8.05 (m, 3H),
7.91-7.83 (m, 2H). The benzamide was dissolved in THF (400
mL) and treated with triethylamine (16.6 mL, 0.12 mol) and
trifluoroacetic anhydride (14 mL, 99 mmol). The solution was
stirred for 16 h at ambient temperature and then added to
EtOAc (400 mL). The organic solution was washed with 0.25
N HCl (300 mL), saturated NaHCO3, and finally brine. The
organic solution was then dried over MgSO4, filtered, and
concentrated under reduced pressure to give 17 (14.7 g, 55%
1
from 15) as a white solid. H NMR: (DMSO-d6) δ 8.48-8.42
(m, 1H), 8.30-8.20 (m, 2H).
4-Am in o-2-flu or oben zon itr ile (18). 2-Fluoro-4-nitroben-
zonitrile (17) (4.74 g, 29 mmol) was stirred with 10% Pd/C
catalyst (500 mg) in MeOH (250 mL) under hydrogen at
atmospheric pressure for 8 h. The catalyst was then removed
by filtration over Celite, and the eluant was concentrated
under reduced pressure to give 18 (3.9 g, 100%). 1H NMR
(DMSO-d6): δ 7.38 (t, J ) 8.6 Hz, 1H), 6.52 (br s, 2H), 6.44-
6.40 (m, 2H).
4-Am in o-2-flu or o-5-iod oben zon itr ile (12b). The aniline
18 (3.9 g, 29 mmol) was dissolved in AcOH (60 mL) and cooled
by immersion in an ice bath. NIS (6.4 g, 29 mmol) was added,
and the reaction mixture was stirred for 2 h at ambient
temperature. The reaction mixture was then concentrated to
about one-quarter of the volume, and the solid that formed
was collected by filtration. The solid was washed with hexanes
and dried to give 12b (2.45 g, 33%) as a tan solid. 1H NMR
(DMSO-d6): δ 8.01 (d, J ) 7.7 Hz, 1H), 6.61-6.55 (m, 3H).
4-Am in o-5-iod o-2-m eth oxyben zon itr ile (12d ). The ni-
troarene 24 (2.9 g, 16 mmol) was dissolved in dioxane (45 mL)
and 6 N HCl (5 mL). The stirred solution was cooled to 0 °C
and treated with tin(II) chloride dihydrate (9.2 g, 49 mmol).
The mixture was stirred for 3 h, allowed to warm to room
temperature, and then diluted with EtOAc (150 mL). The
mixture was washed with water (3 × 100 mL) and brine, dried
over Na2SO4, filtered, and concentrated under reduced pres-
sure to give 25 (1.8 g, 75%) as a crude solid. The solid was
dissolved in AcOH (30 mL) and treated with NIS (2.7 g, 12
mmol). The mixture was stirred for 2 h and then diluted with
EtOAc (200 mL). The solution was washed with water and
brine, dried over Na2SO4, filtered, and concentrated under
reduced pressure to give a crude residue. The residue was
subjected to chromatography over silica gel eluting with 30%
EtOAc in hexanes to give the product 12d (1.3 g, 29% from
2-Meth yl-4-n itr oben zon itr ile (20). The aniline 19 (15.2
g, 0.1 mol) was dissolved in water (50 mL) and AcOH (50 mL).
Concentrated sulfuric acid (10.7 mL, 0.2 mol) was added and
the solution cooled to 0 °C using an ice bath. NaNO2 (7.6 g,
0.11 mol) dissolved in water (20 mL) was added dropwise over
5 min to give a mixture containing the diazonium salt. In a
separate flask, copper sulfate hydrate (30 g, 0.12 mol) in water
(100 mL) was cooled to 0 °C and then treated with KCN (32.5
g, 0.5 mol) in water (100 mL) over 10 min. NaHCO3 (84 g, 1.0
mol) and benzene (100 mL) were then added to this solution
followed by the diazonium mixture over 60 min. After the
resultant mixture was stirred for 30 min, EtOAc was added
(500 mL) and the organic layer was separated. The organic
layer was washed with saturated NaHCO3 and then dried over
Na2SO4. The organic solution was filtered and concentrated
under reduced pressure. The residual solid was resuspended
in petroleum ether and collected by filtration. The solid was
1
24). H NMR (DMSO-d6): δ 7.77 (s, 1H), 6.44 (s, 1H), 6.20 (br
s, 2H), 3.77 (s, 3H).
2-(Ben zyloxy)-4-n itr oben zon itr ile (26). The phenol 23
(2.0 g, 12 mmol) was dissolved in anhydrous DMF (30 mL)
and treated with K2CO3 (3.3 g, 24 mmol) followed by benzyl
bromide (1.6 mL, 13 mmol). The solution was stirred for 2 h
and then diluted with EtOAc (100 mL). The solution was
washed with brine (3×), dried over Na2SO4, filtered, and
concentrated under reduced pressure to give 26 (2.2 g, 72%).
1H NMR (DMSO-d6): δ 8.11 (m, 2H), 7.93 (ddd, J ) 8.4, 2.0,
1.0 Hz, 1H), 7.52-7.35 (m, 5H), 5.45 (s, 2H).
4-Am in o-2-(ben zyloxy)ben zon itr ile (27). The nitroarene
26 (2.7 g, 10.7 mmol) was dissolved in dioxane (20 mL) and
treated with tin(II) chloride (6.1 g, 32 mmol) and 6 N HCl (2
mL). The mixture was stirred overnight and then diluted with
EtOAc (100 mL). The solution was washed with 1 N NaOH,
1
dried to yield 20 (14.6 g, 91%). H NMR (CDCl3): δ 8.18 (d, J
) 0.7 Hz, 1H), 8.13 (d, J ) 8.4 Hz, 1H), 7.79 (d, J ) 8.4 Hz,
1H), 2.67 (s, 3H).
4-Am in o-5-iod o-2-m et h ylb en zon it r ile (12c). The ni-
trobenzene 20 (5 g, 30 mmol) was dissolved in dioxane (27 mL),

3868 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23
Mackman et al.
water, and brine. The organic solution was dried over Na2-
SO4, filtered, and concentrated under reduced pressure to give
27 (2.4 g, 100%). 1H NMR (CDCl3): δ 7.44-7.28 (m, 6H), 6.21
(dd, J ) 8.2, 2.0 Hz, 1H), 6.18 (dd, J ) 7.6, 2.0 Hz, 1H), 5.12
(s, 2H). MS-Sciex: m/z 225.0 [MH+].
The mesylate 28c (672 mg, 2.0 mmol) was dissolved in dry
DMF (20 mL). Triethylamine (2.8 mL, 20 mmol) was added
followed by dichlorobis(triphenylphosphine)palladium(II) (140
mg, 0.2 mmol). The mixture was then treated with alkyne 131
(564 mg, 2.0 mmol). The resultant mixture was stirred for 30
min, and then CuI (76 mg, 0.4 mmol) was added. The mixture
was stirred overnight, then diluted with EtOAc (100 mL), and
washed with brine (4 × 100 mL). The organic solution was
dried over Na2SO4, filtered, and concentrated under reduced
pressure. The residue was subjected to chromatography over
silica gel eluting with 25% EtOAc in hexanes to give the
4-Am in o-5-iod o-2-(ben zyloxy)ben zon itr ile (12e). The
amine 27 (2.4 g, 10.7 mmol) was dissolved in CH3CN (25 mL)
and treated with NIS (2.4 g, 10.7 mmol). The mixture was
stirred for 2 h, and then another portion of NIS (0.48 g, 2.1
mmol) was added. The mixture was stirred overnight and then
diluted with EtOAc (50 mL). The organic solution was washed
with brine, dried over Na2SO4, filtered, and concentrated under
reduced pressure. The residue was subjected to chromatogra-
phy over silica gel eluting with 30% EtOAc in hexanes to give
1
mesylated indole (300 mg, 31%). H NMR (CDCl3): δ 7.96 (d,
J ) 0.7 Hz, 1H), 7.87 (s, 1H), 7.55-7.51 (m, 2H), 7.46-7.24
(m, 6H), 6.65 (d, J ) 0.7 Hz, 1H), 4.51 (s, 2H), 3.35 (s, 3H),
3.11 (s, 3H), 2.98-2.86 (br m, 4H), 2.69 (s, 3H). The mesylated
indole (0.30 g, 0.86 mmol) was dissolved in dioxane (10 mL),
and MeOH (10 mL) and then treated with 1 N NaOH until
pH 13. The mixture was stirred for 3 h and then diluted with
EtOAc (100 mL). The solution was washed with brine, dried
over Na2SO4, filtered, and concentrated under reduced pres-
sure. The crude residue was subjected to chromatography over
silica gel eluting with 25% EtOAc in hexanes to give 29c (0.21
g, 26%). 1H NMR (CDCl3): δ 10.18 (br s, 1H), 7.91 (s, 1H),
7.72 (dd, J ) 7.4, 2.2 Hz, 1H), 7.57-7.54 (m, 2H), 7.47-7.24
(m, 7H), 6.85 (d, J ) 2.1, 0.8 Hz, 1H), 4.66 (s, 2H), 3.29-3.26
(m, 2H), 3.21-3.17 (m, 5H), 2.62 (s, 3H). MS-Sciex: m/z 411.0
[M - H+].
1
12e (3.0 g, 80%). H NMR (DMSO-d6): δ 7.80 (s, 1H), 7.46-
7.32 (m, 5H), 6.54 (s, 1H), 6.20 (br s, 2H), 5.12 (s, 2H). MS-
Sciex: m/z 351.0 [MH+].
N -(5-Ch lor o-4-cya n o-2-iod op h e n yl)m e t h a n e su lfon -
a m id e (28a ). The amine 12a (5.6 g, 20 mmol) was dissolved
in CH2Cl2 (100 mL) and treated with N,N-diisopropylethyl-
amine (DIEA) (17.4 mL, 100 mmol) and N,N-(dimethylamino)-
pyridine (DMAP) (0.25 g, 2 mmol) at 0 °C. Mesyl chloride (4.6
mL, 60 mmol) was added slowly over 15 min and the mixture
stirred for 1.5 h. A further portion of DIEA (20 mL, 100 mmol)
and mesyl chloride (3.0 mL, 39 mmol) were added, and the
solution was stirred overnight. The mixture was concentrated
under reduced pressure and the residue dissolved in EtOAc
(200 mL). The organic solution was extracted with 1 N NaOH
(5 × 200 mL), and the combined aqueous extracts were then
acidified with 12 N HCl until pH 2. The aqueous solution was
then extracted with EtOAc (3 × 500 mL), and the combined
organic extracts were dried over Na2SO4. The mixture was
filtered and the eluant concentrated under reduced pressure
to give a crude residue. The residue was suspended in Et2O
and collected by filtration to give the title product 28a (3.4 g,
47%). 1H NMR (CDCl3): δ 8.05 (s, 1H), 7.76 (s, 1H), 6.97 (br s,
1H), 3.13 (s, 3H).
Met h od B: E xa m p le, 6-Met h oxy-2-[2-[(2-m et h oxy-
e t h oxy)m e t h oxy]b ip h e n yl-3-yl]-1H -in d ole -5-ca r b on i-
tr ile (29d ). The alkyne 13 (377 mg, 1.33 mmol) and aryl halide
29d (500 mg, 1.33 mmol) were dissolved in dry DMF (20 mL).
Triethylamine (0.93 mL, 6.65 mmol) was added followed by
dichlorobis(triphenylphosphine)palladium(II) (93 mg, 0.13 mmol)
and the resultant solution stirred for 30 min. CuI (51 mg, 0.27
mmol) was added and the reaction mixture stirred for a further
2 h. The mixture was diluted with EtOAc (100 mL) and washed
with brine (4 × 100 mL). The organic solution was dried over
Na2SO4, filtered, and concentrated under reduced pressure.
The residue was subjected to chromatography over silica gel
eluting with 20% EtOAc in hexanes to give the coupled alkyne
(4-Cyan o-2-flu or o-5-m eth oxyph en yl)car bam ic Acid ter t-
Bu tyl Ester (28b). 4-Amino-2-fluoro-5-iodobenzonitrile (12b)
(5.0 g, 19.1 mmol) and DMAP (233 mg, 1.91 mmol) were
dissolved in THF (50 mL). A solution of di-tert-butyl dicar-
bonate (9.16 g, 42.0 mmol) in THF (10 mL) was added followed
by DIEA (13.3 mL, 76 mmol). The mixture was stirred for 3
h, and then the solvent was removed under reduced pressure.
The residue was dissolved in 1:1 THF/EtOH (40 mL). NaOH
(50%) (2.53 mL, 31.3 mmol) was added and the mixture stirred
for 2 h. The solvent was removed under reduced pressure and
the residue dissolved in EtOAc (250 mL). The solution was
washed with water (3 × 100 mL) and then brine (100 mL),
dried over Na2SO4, filtered, and concentrated under reduced
pressure. The resultant light brown residue was subjected to
silica gel chromatography eluting with an increasing gradient
of 10-40% EtOAc in hexanes to give 28b (3.27 g, 47%) as a
1
(0.55 g, 79%). H NMR (CDCl3): δ 8.06 (s, 1H), 7.93 (s, 1H),
7.62 (s, 1H), 7.55-7.33 (m, 7H), 7.19 (t, J ) 7.8 Hz, 1H), 4.96
(s, 2H), 3.98 (s, 3H), 3.35-3.32 (m, 2H), 3.18 (s, 3H), 3.16-
3.12 (m, 2H), 1.55 (s, 9H). The coupled alkyne was then
dissolved in THF (25 mL) and treated with a 1 N solution of
tetrabutylammonium fluoride in THF (2.1 mL, 2.1 mmol). The
mixture was refluxed for 5 h, allowed to cool, and then diluted
with EtOAc (150 mL). The solution was washed with brine,
dried over Na2SO4, filtered, and concentrated under reduced
pressure. The crude residue was subjected to chromatography
over silica gel eluting with 40% EtOAc in hexanes to give 29d
1
(395 mg, 69%). H NMR (CDCl3): δ 10.20 (br s, 1H), 7.83 (s,
1H), 7.71 (dd, J ) 7.2, 2.5 Hz, 1H), 7.57-7.54 (m, 2H), 7.48-
7.36 (m, 3H), 7.32-7.23 (m, 3H), 6.94 (s, 1H), 6.81 (dd, J )
2.1, 0.7 Hz, 1H), 4.67 (s, 2H), 3.94 (s, 3H), 3.37-3.33 (m, 2H),
3.28-3.25 (m, 2H), 3.23 (s, 3H). MS-Sciex: m/z 428.8 [MH+].
1
white solid. H NMR (DMSO-d6): δ 8.55 (s, 1H), 8.39 (d, J )
8.3 Hz, 1H), 7.73 (d, J ) 13.2 Hz, 1H), 1.47 (s, 1H).
N -(5-Me t h yl-4-cya n o-2-iod op h e n yl)m e t h a n e su lfon -
a m id e (28c). 28c was prepared from amine 12c in 64% yield
6-Ch lor o-2-[2-[(2-m eth oxyeth oxy)m eth oxy]bip h en yl-3-
yl]-1H-in d ole-5-ca r bon itr ile (29a ). 29a was prepared in a
yield of 52% according to method A from 28a and alkyne 13.
MS-Sciex: m/z 433.4 [35Cl, MH+].
1
using the procedure described for the preparation of 28a . H
NMR (CDCl3): δ 7.99 (s, 1H), 7.59 (s, 1H), 6.88 (br s, 1H), 3.08
(s, 3H).
(4-Cya n o-2-iod o-5-m eth oxyp h en yl)ca r ba m ic Acid ter t-
Bu tyl Ester (28d ). 28d was prepared from aniline 12d in 42%
yield (with DMF as solvent) using the procedure described for
the preparation of 28b. H NMR (CDCl3): δ 8.01 (s, 1H), 7.83
(s, 1H), 7.06 (br s, 1H), 3.93 (s, 3H), 1.54 (s, 9H).
6-F lu or o-2-[2-[(2-m eth oxyeth oxy)m eth oxy]bip h en yl-3-
yl]-1H-in d ole-5-ca r bon itr ile (29b). 29b was prepared in a
yield of 44% according to method B from 28b and alkyne 13.
1H NMR (DMSO-d6): δ 8.13 (d, J ) 6.5 Hz, 1H), 7.72 (dd, J )
6.5, 2.6 Hz, 1H), 7.58-7.55 (m, 2H), 7.50-7.32 (m, 7H), 7.02
(d, J ) 1.5 Hz, 1H), 4.51 (s, 2H), 2.96-2.94 (m, 5H), 2.89-
2.86 (m, 2H). MS-Sciex: m/z 415.1 [M - H+].
6-(Ben zyloxy)-2-[2-[(2-m eth oxyeth oxy)m eth oxy]biph en -
yl-3-yl]-1H-in d ole-5-ca r bon itr ile (29e). 29e was prepared
in a yield of 54% according to method B from 28e and alkyne
13. 1H NMR (CDCl3): δ 10.16 (br s, 1H), 7.85 (s, 1H), 7.69
(dd, J ) 7.2, 2.2 Hz, 1H), 7.56-7.23 (m, 13H), 6.97 (s, 1H),
6.80 (dd, J ) 2.1, 0.9 Hz, 1H), 5.23 (s, 2H), 4.66 (s, 2H), 3.34-
3.30 (m, 2H), 3.25-3.21 (m, 2H), 3.20 (s, 3H).
1
(4-Cyan o-2-iodo-5-ben zyloxyph en yl)car bam ic Acid ter t-
Bu tyl Ester (28e). 28e was prepared from aniline 12e in 60%
yield (with DMF as solvent) using the procedure described for
1
the preparation of 28b. H NMR (CDCl3): δ 8.12 (s, 1H), 7.84
(s, 1H), 7.48 (dd, J ) 8.4, 1.8 Hz, 2H), 7.41-7.31 (m, 3H), 7.04
(br s, 1H), 5.20 (s, 2H), 1.53 (s, 9H). MS-Sciex: m/z 450.8
[MH+].
Meth od A: Exam ple, 2-[2-[(2-Meth oxyeth oxy)m eth oxy]-
bip h en yl-3-yl]-6-m eth yl-1H-in d ole-5-ca r bon itr ile (29c).

Exploiting S1 of Serine Proteases for Selectivity
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23 3869
Meth od C: Exa m p le, 2-(2-Hyd r oxybip h en yl-3-yl)-6-
m eth oxy-1H-in d ole-5-ca r boxa m id in e (10d ). The indole
29d (0.39 g, 0.91 mmol) was dissolved in absolute EtOH (20
mL) and excess 50% aqueous hydroxylamine (5 mL). The
mixture was refluxed for 4 h and then concentrated under
reduced pressure. The crude solid was dissolved in EtOAc (50
mL) and washed with water followed by brine. The organic
solution was dried over Na2SO4, filtered, and concentrated
under reduced pressure. The crude residue containing the
hydroxyamidine was dissolved in AcOH (10 mL) and treated
with acetic anhydride (95 µL, 1 mmol). The mixture was stirred
for 1 h, and then zinc dust (∼250 mg) was added. The mixture
was stirred for another 4 h, and then the zinc dust was
removed by filtration. The organic solution was concentrated
under reduced pressure, and the crude residue containing the
amidine was then dissolved in MeOH (15 mL). The solution
was treated with 4 N HCl in dioxane (10 mL) and stirred for
2 h. The reaction mixture was diluted with EtOAc (100 mL)
and washed with brine. The organic solution was dried over
Na2SO4, filtered, and concentrated under reduced pressure.
The crude residue was subjected to preparative HPLC to give
11.91 (s, 1H), 9.22 (s, 2H), 9.17 (s 2H), 8.93 (s, 1H), 7.86 (d, J
) 7.7 Hz, 1H), 7.68 (dd, J ) 8.5, 1.7 Hz, 1H), 7.50-7.28 (m,
6H), 7.17 (dd, J ) 8.3, 1.7 Hz, 1H), 7.06-7.00 (m, 2H). 13C NMR
(DMSO-d6): δ 163.5, 150.7, 138.2, 131.9, 130.5, 129.3, 128.4,
127.6, 127.2, 124.5, 121.8, 121.6, 121.1, 109.2, 108.9, 102.1,
98.7, 98.2. MS-Sciex: m/z 346.1 [MH+]. Anal. (C21H16FN3O‚
HCl) C, H, N, F.
6-Meth yl-2-(2-h yd r oxybip h en yl-3-yl)-1H-in d ole-5-ca r -
boxa m id in e (10c). 10c was prepared from indole 29c in a
yield of 9% according to method C. 1H NMR (DMSO-d6): δ
11.64 (s, 1H), 9.17 (br s, 4H), 8.90 (s, 1H), 7.74 (d, J ) 7.5 Hz,
1H), 7.70 (s, 1H), 7.56-7.34 (m, 6H), 7.21 (d, J ) 7.2 Hz, 1H),
7.11-7.06 (m, 2H), 2.48 (s, 3H). 13C NMR (DMSO-d6): δ 168.2,
150.5, 138.3, 138.1, 136.8, 131.8, 130.1, 129.2, 128.3, 127.5,
127.4, 127.0, 125.8, 121.9, 121.4, 120.9, 120.5, 112.9, 101.8,
19.7. MS-Sciex: m/z 341.8 [MH+]. Anal. (C22H19N3O‚HCl‚
0.25H2O) C, H, N.
En zym ology. Enzymes and substrates were purchased as
follows: high molecular weight (HMW) uPA (American Diag-
nostica), tPA (Sigma), trypsin (Worthington), thrombin (Cal-
biochem), plasmin (Enzyme Research Labs), and factor Xa
(Haematological Technologies, Inc.). The substrates for trypsin,
plasmin, and thrombin (tosyl-Gly-Pro-Lys-pNA (pNA ) p-
nitroanilide)) and for factor Xa (CH3OCO-D-CHA-Gly-Arg-
pNA-AcOH, “Pefachrome Xa”) were purchased from Cen-
terchem, Inc. The substrate for HMW uPA (Glt-Gly-Arg-AMC)
was from the Peptide Institute, and the substrate for tPA (CH3-
SO2-D-HHT-Gly-Arg-pNA, “Spectrozyme tPA”) was from Ameri-
can Diagnostica.
1
10d (0.19 g, 58%). H NMR (DMSO-d6): δ 11.57 (s, 1H), 8.97
(br s, 2H), 8.90 (s, 1H), 8.80 (br s, 2H), 7.82 (s, 1H), 7.70 (dd,
J ) 7.5, 1.6 Hz, 1H), 7.55 (m, 2H), 7.48 (t, J ) 7.6 Hz, 2H),
7.38 (m, 1H), 7.22-7.19 (m, 2H), 7.09 (t, J ) 7.5 Hz, 1H), 7.03
(s, 1H), 3.89 (s, 3H). 13C NMR (DMSO-d6): δ 165.2, 153.2,
150.5, 139.6, 138.4, 136.4, 131.8, 130.0, 129.3, 128.4, 127.1,
122.1, 121.9, 121.8, 121.0, 110.5, 102.0, 94.2, 55.9. MS-Sciex:
m/z 358.0 [MH+]. Anal. (C22H19N3O2‚C2HF3O2) C, H, N.
Meth od D: Exa m p le, 2-(2-Hyd r oxybip h en yl-3-yl)-6-
h yd r oxy-1H-in d ole-5-ca r boxa m id in e (10e). The indole 29e
(0.76 g, 1.5 mmol) was dissolved in dry EtOH (20 mL), and
excess 50% aqueous hydroxylamine (5 mL) was added. The
mixture was refluxed for 4 h and then concentrated under
reduced pressure. The crude solid was dissolved in EtOAc (50
mL) and washed with water followed by brine. The organic
solution was dried over Na2SO4, filtered, and concentrated
under reduced pressure. The crude residue containing the
hydroxyamidine was dissolved in AcOH (10 mL) and treated
with acetic anhydride (147 µL, 1.6 mmol). The mixture was
stirred for 30 min, and then 10% Pd/C catalyst (∼250 mg) was
added followed by MeOH (10 mL). The mixture was stirred
for another 4 h under hydrogen, and then the catalyst was
removed by filtration over Celite. The organic solution was
concentrated under reduced pressure, and the crude residue
containing the amidine was then dissolved in MeOH (15 mL).
The solution was treated with 4 N HCl in dioxane (10 mL)
and stirred for 2 h. The reaction mixture was diluted with
EtOAc (100 mL) and washed with brine. The organic solution
was dried over Na2SO4, filtered, and concentrated under
reduced pressure. The crude residue was subjected to prepara-
En zym e Assa ys. uPA was assayed spectrofluorometrically,
and all the others were assayed spectrophotometrically. All
enzymes were assayed at ambient temperature in 50 mM Tris,
150 mM NaCl, 0.05% (v/v) Tween-20, 10% (v/v) DMSO, 0.002%
antifoam, and EDTA (1 mM) at pH 7.4. The thrombin and
factor Xa assays also contained 5.0 mM CaCl2. The final
concentrations of uPA, tPA, trypsin, thrombin, plasmin, and
factor Xa in the respective assays were 4 (or 5), 11 (or 15), 15,
10, 6, and 4 nM. The substrate concentrations of 85 µM (uPA),
300 µM (tPA), 25 or 45 µM (trypsin), 300 µM (thrombin), 130
µM (plasmin), and 550 µM (factor Xa) were chosen on the basis
of the determined Km values with EDTA of 85, 500, 45, 300,
100, and 600 µM, respectively. For uPA activity, the rate of
change in relative fluorescence units (RFUs) emitted at 460
nm (excitation at 355 nm) was measured immediately after
addition of enzyme. For all other enzymatic activities, the rate
of change in absorbance at 405 nm was measured immediately
after addition of enzyme. Apparent inhibition constants, Ki′,
were calculated from the velocity data generated at the various
inhibitor concentrations using the software package BatchKi
(developed and provided by Dr. Petr Kuzmic, Biokin Ltd.,
Madison, WI) using methodology similar to that described for
tight-binding inhibitors.34 BatchKi applies nonlinear regres-
sion analysis to return nominal likelihood intervals at the 65%
confidence, using the logic described by Bates and Watts.35 The
Ki′ values are accepted where the nominal likelihood intervals
at 65% confidence vary within 0.5-1.5-fold from the Ki′ values.
These values were converted to Ki values by the formula Ki )
Ki′/(1 + S/Km).
1
tive HPLC to give 10e (0.14 g, 27%). H NMR (DMSO-d6): δ
11.41 (s, 1H), 10.68 (s, 1H), 8.87 (br s, 2H), 8.82 (s, 1H), 8.60
(br s, 2H), 7.88 (s, 1H), 7.65 (dd, J ) 7.6, 1.6 Hz, 1H), 7.55-
7.53 (m, 2H), 7.47 (t, J ) 7.5 Hz, 2H), 7.38 (d, J ) 6.9 Hz,
1H), 7.20 (dd, J ) 7.4, 1.5 Hz, 1H), 7.12 (s, 1H), 7.07 (t, J )
7.7 Hz, 1H), 6.96 (s, 1H). 13C NMR (DMSO-d6): δ 164.9, 152.3,
150.5, 140.4, 138.4, 136.3, 131.7, 129.9, 129.3, 128.3, 127.1 (2
· C), 122.0, 121.6, 120.9, 107.4, 102.2, 97.5. MS-Sciex: m/z 344.0
[MH+]. Anal. (C21H17N3O2‚HCl‚0.25H2O) C, H, N, O.
Cr ysta lliza tion a n d P r ep a r a tion of P r otea se-In h ibi-
tor Com p lexes. Procedures and conditions for crystallization
have been reported by Katz et al.32
6-Ch lor o-2-(2-h yd r oxybip h en yl-3-yl)-1H-in d ole-5-ca r -
boxa m id in e (10a ). 10a was prepared from indole 29a in a
P h a r m a cok in etic Stu d y. Male Sprague-Dawley rats
purchased from the Charles River Laboratory, weighing ap-
proximately 280-300 g, were used in the pharmacokinetic
study. Compounds were formulated as an aqueous solution in
40% PEG400/H2O (1 mg/mL or 2 mg/mL) and administered
intravenously via bolus injection into the tail vein. Blood was
collected from the jugular vein in a heparinized tube at several
time points up to 24 h after dosing. Plasma samples were
separated by centrifugation and kept frozen at -80 °C until
analysis. Plasma concentrations were determined by LC-MS/
MS on a SCIEX API 300 tandem mass spectrometer equipped
with an electrospray ionization source. Samples were prepared
for analysis by mixing three volume equivalents of acetonitrile
1
yield of 20% according to method C. H NMR (DMSO-d6): δ
11.85 (s, 1H), 9.35 (br s, 2H), 9.14 (br s, 2H), 9.07 (s, 1H), 7.88
(s, 1H), 7.76 (dd, J ) 8.5, 1.7 Hz, 1H), 7.69 (s, 1H), 7.58-7.37
(m, 5H), 7.26 (dd, J ) 8.5, 1.7 Hz, 1H), 7.15 (s, 1H), 7.12 (t, J
) 8.5 Hz, 1H). 13C NMR (DMSO-d6): δ 167.3, 150.5, 139.4,
138.4, 138.0, 131.3, 130.6, 129.2, 128.5, 127.5, 127.4, 127.1,
122.8, 121.4, 120.7, 120.0, 119.9, 112.6, 101.1. MS-ESI: m/z
361.8 [35Cl, MH+]. Anal. (C21H16ClN3O‚HCl‚0.25H2O) C, H, N,
Cl.
6-F lu or o-2-(2-h yd r oxybip h en yl-3-yl)-1H-in d ole-5-ca r -
boxa m id in e (10b). 10b was prepared from indole 29b in a
1
yield of 20% according to method C. H NMR (DMSO-d6): δ

3870 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23
Mackman et al.
(14) Brandstetter, H.; Ku¨hne, A.; Bode, W.; Huber, R.; von der Saal,
W.; Wirthensohn, K.; Engh, R. A. X-Ray Structure of Active Site-
inhibited Clotting Factor XasImplications for Drug Design and
Substrate Recognition. J . Biol. Chem. 1996, 271, 29988-29992.
(15) Zhao, Z.; Arniaz, D. O.; Griedel, B.; Sakata, S.; Dallas, J . L.;
Whitlow, M.; Trinh, L.; Post, J .; Liang, A.; Morrissey, M. M.;
Shaw, K. J . Design, Synthesis, and In Vitro Biological Activity
of Benzimidazole Based Factor Xa Inhibitors. Bioorg. Med.
Chem. Lett. 2000, 10, 963-966.
(16) Ewing, W. R.; Becker, M. R.; Manetta, V. E.; Davis, R. S.; Pauls,
H. W.; Mason, H.; Choi-Sledeski, Y. M.; Green, D.; Cha, D.;
Spada, A. P.; Cheney, D. L.; Mason, J . S.; Maignan, S.; Guillo-
teau, J .-P.; Brown, K.; Colussi, D.; Bentley, R.; Bostwick, J .;
Kasiewski, C. J .; Morgan, S. R.; Leadley, R. J .; Dunwiddie, C.
T.; Perrone, M. H.; Chu, V. Design and Structure-Activity
Relationships of Potent and Selective Inhibitors of Blood Co-
agulation Factor Xa. J . Med. Chem. 1999, 42, 3557-3571.
(17) Choi-Sledeski, Y. M.; McGarry, D. G.; Green, D. M.; Mason, H.
J .; Becker, M. R.; Davis, R. S.; Ewing, W. R.; Dankulich, W. P.;
Manetta, V. E.; Morris, R. L.; Spada, A. P.; Cheney, D. L.; Brown,
K. D.; Colussi, D. J .; Chu, V.; Heran, C. L.; Morgan, S. R.;
Bentley, R. G.; Leadley, R. J .; Maignan, S.; Guilloteau, J .-P.;
Dunwiddie, C. T.; Pauls, H. W. Sulfonamidopyrrolidinone Factor
Xa Inhibitors: Potency and Selectivity Enhancements via P-1
and P-4 Optimization. J . Med. Chem. 1999, 42, 3572-3587.
(18) (a) Rai, R.; Sprengeler, P. A.; Elrod, K. C.; Young, W. B.
Perspectives on Factor Xa Inhibition. Curr. Med. Chem. 2000,
1, 1-25. (b) Fevig, J . M.; Wexler, R. R. Anticoagulants: Thrombin
and Factor Xa Inhibitors. Annu. Rep. Med. Chem. 1999, 34, 81-
100.
(19) Rijken, D. C.; Groeneveld, E. Substrate Specificity of tissue-type
and urokinase-type plasminogen activators. Biochem. Biophys.
Res. Commun. 1991, 174, 432-438.
(20) Zeslawska, E.; Schweinitz, A.; Karcher, A.; Sondermann, P.;
Sperl, S.; Stu¨rzebecher, J .; J acob, U. Crystals of the Urokinase
Type Plasminogen Activator Variant âc-uPA in Complex with
Small Molecule Inhibitors Opens the Way towards Structure-
based Drug Design. J . Mol. Biol. 2000, 301, 465-475.
(21) Nienaber, V.; Wang, J .; Davidson, D.; Henkin, J . Re-engineering
of Human Urokinase Provides a System for Structure-based
Drug Design at High Resolution and Reveals a Novel Structural
Subsite. J . Biol. Chem. 2000, 275, 7239-7248.
(22) Katz, B. A.; Elrod, K. C.; Luong, C.; Rice, M.; Mackman, R. L.;
Sprengeler, P. A.; Spencer, J .; Hataye, J .; J anc, J .; Link, J .;
Litvak, J .; Rai, R.; Rice, K.; Sideris, S.; Verner, E.; Young, W. A
and 50 µL of internal standard solution with the plasma
sample. The samples were centrifuged and the supernatant
transferred to another vial for drying by heated nitrogen gas.
The sample was reconstituted in 200 µL of mobile phase (25%
acetonitrile and 0.2% formic acid in water) and injected onto
the LC-MS/MS system. Chromatographic separation was
performed using a C18 column (Phaenomenex, 5 µm, 2 × 100
mm) with a flow rate of 0.4 mL/min and a linear gradient of
solvent B in solvent A (solvent A was 0.2% formic acid in water,
and solvent B was 0.2% formic acid in acetonitrile). Concentra-
tions were calculated using previously established standard
curves of the inhibitors, ranging from 1 to 10000 ng/mL, and
the internal standard in rat plasma. The limit of quantitation
was 6 ng/mL plasma.
Pharmacokinetic parameters were estimated using Win-
Nonlin Version 1.5 (Pharsight Corp., California). Pharmaco-
kinetic calculations were performed using a nominal dose and
collection time.
Ack n ow led gm en t. We thank Drs. M. Venuti, J .
Knolle, and R. McDowell for helpful discussions and
insights. We also thank Dr. J ames J anc, Dr. Kesavan
Radika, and J ing Wang for the Ki determinations.
Refer en ces
(1) Verner, E.; Katz, B. A.; Spencer, J . R.; Allen, D.; Hataye, J .;
Hruzewicz, W.; Hui, H. C.; Kolesnikov, A.; Li, Y.; Luong, C.;
Martelli, A.; Radika. K.; Rai, R.; She, M.; Shrader, W.; Spren-
geler, P. A.; Trapp, S.; Wang, J .; Young, W. B.; Mackman, R. L.
Development of Serine Protease Inhibitors Displaying a Multi-
centered Short (<2.3 Å) Hydrogen Bond Binding Mode: Inhibi-
tors of Urokinase-type Plasminogen Activator and Factor Xa.
J . Med. Chem. 2001, 44, 2753-2771.
(2) Magill, C.; Katz, B. A.; Mackman, R. L. Emerging therapeutic
targets in oncology: urokinase-type plasminogen activator sys-
tem. Emerging Ther. Targets 1999, 3, 109-133. Andreasen, P.
A.; Kjøller, L.; Christensen, L.; Duffy, M. J . The Urokinase-type
Plasminogen Activator System in Cancer Metastasis: A Review.
Int. J . Cancer 1997, 72, 1-22. Rabbani, S. A.; Xing, R. H. Role
of urokinase (uPA) and its receptor (uPAR) in invasion and
metastasis of hormone-dependent malignancies (Review) J . Int.
Oncol. 1998, 12, 911-920.
(3) Deadman, J . J .; Elgendy, S.; Goodwin, C. A.; Green, D.; Baban,
J . A.; Patel, G.; Skordalakes, E.; Chino, N.; Claeson, G.; Kakkar,
V. V.; Scully, M. F. Characterization of a Class of Peptide
Boronates with Neutral P1 Side Chains as Highly Selective
Inhibitors of Thrombin. J . Med. Chem. 1995, 38, 1511-1522.
(4) Schechter, I.; Berger, A. On the size of the active site in
proteases. I. Papain. Biochem. Biophys. Res. Commun. 1967, 27,
157-162.
(5) Mares-Guia, M.; Shaw, E. Studies on the Active Center of
Trypsin. J . Biol. Chem. 1965, 240, 1579-1585.
(6) Markwardt, F. Synthetic, Low Molecular Thrombin Inhibitors.
A New Concept of Anticoagulants. Haemostasis 1974, 3, 185-
202.
(7) Geratz, J . D.; Shaver, S. R.; Tidwell, R. R. Inhibitory Effect of
Amidino-substituted Heterocyclic Compounds on the Amidase
Activity of Plasmin and of High and Low Molecular Weight
Urokinase and on Urokinase-induced Plasminogen Activation.
Thromb. Res. 1981, 24, 73-83.
(8) Geratz, J . D.; Stevens, F. M.; Polakoski, K. L.; Parrish, R. F.;
Tidwell, R. R. Amidino-Substituted Aromatic Heterocycles as
Probes of the Specificity Pocket of Trypsin-Like Proteases. Arch.
Biochem. Biophys. 1979, 197, 551-559.
(9) Walsmann, P.; Horn, H.; Landmann, H.; Markwardt, F.; Stru-
zebecher, J .; Wagner, G. Synthetische Inhibitoren der Serinepro-
teinasen. Pharmazie 1975, 30, 386-389.
(10) Loeffler, L. J .; Mar, E.-C.; Geratz, J . D.; Fox, L. B. Synthesis of
Isosteres of p-Amidinophenylpyruvic Acid. Inhibitors of Trypsin,
Thrombin and Pancreatic Kallikrein. J . Med. Chem. 1975, 18,
287-292.
(11) Stu¨rzebecher, J .; Markwardt, F. Synthetische Inhibitoren der
Serineproteinasen. Pharmazie 1978, 33, 599-602.
(12) Yang, H.; Henkin, J . Selective Inhibition of Urokinase by
Substituted Phenylguanidines: Quantitative Structure-Activity
Relationship Analyses. J . Med. Chem. 1990, 33, 2956-2961.
Yang, H.; Henkin, J . Competitive Inhibitors of Human Uroki-
nase. Fibrinolysis 1992, 6 (Suppl. 1), 31-34.
(13) Katz, B. A.; Mackman, R.; Luong, C.; Radika, K.; Martelli, A.;
Sprengeler, P. A.; Wang, J .; Chan, H.; 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.
Novel Serine Protease Inhibition Motif Involving
a Multi-
Centered Short Hydrogen Bonding Network at the Active Site.
J . Mol. Biol. 2001, 307, 1451-1486.
(23) Dunitz, J . D. The Entropic Cost of Bound Water in Crystals and
Biomolecules. Science 1994, 264, 670.
(24) Robinson, B. The Fischer Indole Synthesis; J ohn Wiley and
Sons: New York, 1982.
(25) Wagaw, S.; Yang, B. H.; Buchwald, S. L. A Palladium-Catalyzed
Strategy for the Preparation of Indoles: A Novel Entry into the
Fisher Indole Synthesis. J . Am. Chem. Soc. 1998, 120, 6621-
6622.
(26) Yasuhara, A.; Kanamori, Y.; Kaneko, M.; Numata, A.; Kondo,
Y.; Sakamoto, T. Convenient synthesis of 2-substituted indoles
from 2-ethynylanilines with tetrabutylammonium fluoride. J .
Chem. Soc., Perkin Trans. 1 1999, 529-534.
(27) Fagnola, M. C.; Candiani, I.; Visentin, G.; Cabri, W.; Zarini, F.;
Mongelli, N.; Bedeschi, A. Solid-Phase Synthesis of Indoles Using
the Palladium-Catalysed Coupling of Alkynes with Iodoaniline
Derivatives. Tetrahedron Lett. 1997, 38, 2307-2310.
(28) Rudisill, D. E.; Stille, J . K. Palladium-Catalyzed Synthesis of
2-Substituted Indoles. J . Org. Chem. 1989, 54, 5856-5866.
(29) Larock, R. C.; Yum, E. K. Synthesis of Indoles via Palladium-
Catalyzed Heteroannulation of Internal Alkynes. J . Am. Chem.
Soc. 1991, 113, 6689-6690.
(30) Yasuhara, A.; Sakamoto, T. Deprotection of N-sulfonyl Nitrogen-
Heteroaromatics with Tetrabutylammonium Fluoride. Tetrahe-
dron Lett. 1998, 39, 595-596.
(31) Imakura, Y.; Okimoto, K.; Konishi, T.; Hisazumi, M.; Yamazaki,
J .; Kobayashi, S.; Yamashita, S. Regioselective Cleavage Reac-
tion of the Aromatic Methylenedioxy Ring. V.1) Cleavage with
Sodium Alkoxides-Alcohols, Potassium tert-Butoxide-Alcohols,
Dimsyl Anion-Methyl Alcohol, Metallic Sodium-Alcohols and
Sodium Cyanide in Dipolar Aprotic Solvents. Chem. Pharm.
Bull. 1992, 40, 1691-1696.
(32) Katz, B. A.; Sprengeler, P. A.; Luong, C.; Verner, E.; Elrod, K.;
Kirtley, M.; J anc, J .; Spencer, J .; Breitenbucher, J . G.; Hui, H.
C.; McGee, D.; Allen, D.; Martelli, A.; Mackman, R. L. Engineer-
ing Inhibitors Highly Specific for the S1 sites of Ser190 Trypsin-
Like Serine Protease Drug Targets. Chem. Biol., in press.

Exploiting S1 of Serine Proteases for Selectivity
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23 3871
(33) 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.; Hulkower, K. I.; Giranda,
V. L. Species Specificity of Amidine-Based Urokinase Inhibitors.
Biochem. 2001, 40, 9125-9131.
Automatic Determination of Tight-Binding Inhibition Constants.
Anal. Biochem. 2000, 281, 62-67.
(35) Bates, D. M.; Watts, D. W. Nonlinear Regression and its
Applications; J ohn Wiley and Sons: New York, 1988.
(34) Kuzmic, P.; Sideris, S.; Cregar, L. M.; Elrod, K. C.; Rice, K. D.;
J anc, J . W. High-Throughput Screening of Enzyme Inhibitors:
J M010244+