Selective urokinase-type plasminogen activator (uPA) inhibitors. Part 2: (3-substituted-5-halo-2-pyridinyl)guanidines

Article abstract of DOI:10.1016/S0960-894X(01)00702-8

Based on previous modeling predictions, a series of (3-substituted-5-chloro-2-pyridinyl)guanidines have been designed with good potency and selectivity for urokinase-type plasminogen activator (uPA). Compound 36 has a K_i of 0.17 μM and greater than 300-fold selectivity with respect to tPA and plasmin.

Full text of DOI:10.1016/S0960-894X(01)00702-8

Bioorganic & Medicinal Chemistry Letters 12 (2002) 185–187
Selective Urokinase-Type Plasminogen Activator (uPA) Inhibitors.
Part 2: (3-Substituted-5-halo-2-pyridinyl)guanidines
Christopher G. Barber* and Roger P. Dickinson
Department of Discovery Chemistry, Pfizer Global Research and Development, Sandwich, Kent CT139NJ, UK
Received 9 July 2001; revised 10 October 2001; accepted 19 October 2001
Abstract—Based on previous modeling predictions, a series of (3-substituted-5-chloro-2-pyridinyl)guanidines have been designed
with good potency and selectivity for urokinase-type plasminogen activator (uPA). Compound 36 has a Ki of 0.17 mM and greater
than 300-fold selectivity with respect to tPA and plasmin. # 2002 Elsevier Science Ltd. All rights reserved.
The previous paper1 described the identification of
2-pyridinylguanidine derivatives as selective inhibitors
of urokinase-type plasminogen activator (uPA). Such
compounds are of potential use in disease indications
where degradation of extracellular matrix by uPA may
be a contributing factor, such as tumor growth and
metastasis, angiogenesis and tissue remodeling.2,3 uPA
acts through activation of the zymogen plasminogen to
plasmin. Related enzymes such as tissue plasminogen
activator (tPA) also act via generation of plasmin from
plasminogen,4,5 so an adequate degree of selectivity over
both tPA and plasmin is an important requirement.
The required guanidines were prepared from a 3-sub-
stituted-2-amino-5-chloropyridine derivative, most of
which were obtained from the amine 3 using Pd(0)-cat-
alyzed cross-coupling as the key step (Schemes 1and
2). It was decided to restrict SAR exploration to the
5-chloro series to enable use of chemoselective reac-
tions at the 3 position of the readily-available 2-amino-
3-bromo-5-chloropyridine (3).6 Thus, reaction of 3
with t-butyl propenoate under Heck conditions gave
the propenoic ester 4 which was hydrolysed (TFA), and
the resulting acid 5 was converted to the amide deriva-
tives 6–9 via activation as the HOBT esters (Scheme 1).
Reduction of 4 gave the propanoate derivative 10.
Arylalkenyl and cycloalkenyl derivatives were prepared
similarly except that the Heck reactions were carried out
under microwave irradiation (Scheme 2). In the case of
11 (R=3,4-methylenedioxyphenyl), the product was
prepared using a Stille coupling with [(E)-2-(1,3-benzo-
dioxol-5-yl)ethenyl]tributyl)stannane.7 A Heck reaction
using phenylethyne was used to prepare the phenylethy-
nyl derivative 12, and the phenyl derivative 13 was pre-
pared via a Suzuki coupling with benzeneboronic acid.
3,5-Dihalo derivatives such as 1 and 2 were the most
potent inhibitors in the previous investigation of simple
substitution in the pyridine ring. Modeling studies sug-
gested that the inhibitors bind in the S1 pocket of the
enzyme with the guanidine system interacting with
Asp189, and the 5-halo substituent (chlorine or bro-
mine) directed towards the catalytic serine.1 The model-
ing results also suggested that introduction of suitable
substituents at the 3-position could potentially capitalize
on interactions in the binding groove of the enzyme. This
paper describes identification of 3-substituted analogues
with submicromolar inhibitory activity against uPA and a
high degree of selectivity over tPA and plasmin.
Starting amines 15, in which the 3-substituent is an ether,
were accessed either by alkylation of 2-amino-5-chloro-
3-hydroxypyridine 148 or, in the case of R=methoxy-
ethyl or C6H5, by nucleophilic displacement of the bro-
mine in 3 with the corresponding alcohol or phenol
under basic conditions (Scheme 3).
Final products (17–35) were prepared by reaction of the
amines with N,N0-bis(t-butoxycarbonyl)-S-methyliso-
thiourea (16) in the presence of mercury (II) chloride,
followed by deprotection with trifluoroacetic acid
(Scheme 4). In the case compounds with a t-butyl ester
*Corresponding author. Fax:+44-1304-651987; e-mail: christopher_
barber@sandwich.pfizer.com
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00702-8

186
C. G. Barber, R. P. Dickinson / Bioorg. Med. Chem. Lett. 12 (2002) 185–187
3-substituent, the final product was the corresponding
carboxylic acid. Acid hydrolysis of 35 [R=2-(3-cyano-
phenylethenyl)] gave the corresponding carboxylic acid
product 36 (Scheme 4).
Carboxylic acid derivative 36 was the most active pyri-
dylguanidine that we identified within this series. In the
published X-ray crystal structure of uPA complexed
with the inhibitor Glu-Gly-Arg chloromethyl ketone, a
salt bridge is observable between the terminal Glu car-
boxylate and Arg217 of the enzyme.9 It is possible that
the higher potency of 36 compared with other compounds
in the series results from a combination of hydrophobic
binding in the main binding groove of the enyme and for-
mation of an equivalent salt bridge with Arg217.
Compounds were tested for their ability to inhibit uPA,
tPA and plasmin as described previously.1 Results for
simple heterocyclic guanidine derivatives, expressed as a
calculated Ki, are listed in Table 1, together with com-
parative figures for 1.
Compared with 1, introduction of propenoic acid (17)
or amide 3-substituents (18–21) gives a small increase in
potency, most notably in the case of the benzylamide
20. Increasing the flexibility of the side chain as in the
propanoic acid analogue 22 or the ether analogues 23
and 24 is unfavorable. Introduction of a phenyl sub-
stituent (26) is detrimental, suggesting an unfavorable
steric effect, but activity is restored with the phenyl ether
27. Further extension of the side chain as in 28 is
favorable. Comparison of 28 and 25 clearly demon-
strates the desirability of an aryl substituent.
Some activity against tPA and plasmin is evident with
the analogues bearing lipophilic 3-substituents, but in
all cases excellent selectivity for uPA is maintained. In
the case of 36, selectivity is at least 300-fold with respect
to tPA and plasmin.
In summary, we have used modeling predictions as a basis
for the design of (3-substituted-5-chloro-2-pyridinyl)-
guanidines with increased potency against uPA compared
with the earlier 3,5-dihalo-substituted analogues. Lipo-
philic 3-substituents such as arylalkenyl are the most
favorable, and the (3-carboxy)phenylethenyl analogue 36
is the most potent inhibitors in this series. Extension of this
work to other heterocyclic guanidine derivatives with
greater potency will be reported in due course.
The most potent analogues result from increasing the
rigidity of the aryl side chain, for example the aryl-
ethenyl derivatives 30 and 34.
Scheme 1. Reagents and conditions: (a) H2C¼CHCO2But, Pd(OAc)2, P(o-Tol)3, Et3N, 150 ꢀC (pressure vessel); (b) TFA, 20 ꢀC; (c) R1R2NH,
.
HOBT, Hunig’s base, 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide HCl; (d) NaBH4, EtOH.
Scheme 2. Reagents and conditions: (a) H2C¼CHR or (Bu)3SnCH¼CHR, Pd(OAc)2, P(o-Tol)3, Et3N, DMF, microwave oven (sealed vessel, 800W,
ca. 30 s); (b) PhCꢁCH, CuCl, PdCl2(PPh3)2, Et3N, DMF, microwave oven (sealed vessel, 800W, ca. 30 s); (c) PhB(OH)2, (Ph3P)4Pd, Na2CO3,
MeO(CH2)2OMe.
Scheme 3. Reagents and conditions: (a) BrCH2CO2But or PhCH2Br, MeONa, DMSO, 20 ꢀC; (b) ClCH2CONHCH2Ph, MeONa, DMSO; (c) ROH,
KOH, CuSO4 (anhydrous), MeO(CH2)2OMe, 140–150 ꢀC.

C. G. Barber, R. P. Dickinson / Bioorg. Med. Chem. Lett. 12 (2002) 185–187
187
Scheme 4. Reagents and conditions: (a) HgCl2, Et3N (3 equiv), CH2Cl2; (b) satd HCl/CH2Cl2 or TFA; (c) concd HCl, AcOH, reflux.
Table 1. Enzyme inhibition data for 2-pyridinylguanidine derivatives
Compd
R
Ki (mM) or % inhibition
uPA
tPA
Plasmin
a
1
Br
4.83
1.50
1.28
2.10
0.77
2.00
9.31
90.2
16.1
16.4
166.3
2.13
0.92
0.77
0.55
0.73
1.60
0.69
0.49
0.17
272.0
a
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
36
(E)-CH¼CHCO2H
(E)-CH¼CHCONHMe
(E)-CH¼CHCO(1-morpholino)
(E)-CH¼CHCONHCH2Ph
(E)-CH¼CHCON(Me)CH2Ph
CH2CH2CO2H
278.5
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
OCH2CO2H
OCH2CONHCH2Ph
OCH2CH2OCH3
C6H5
OC6H5
OCH2C6H5
CꢁCC6H5
342.0
a
65.9
53.8
71.5
258.5
218.0
a
(E)-CH¼CHC6H5
(E)-CH¼CHC6H11
(E)-CH¼CH(2-pyridinyl)
(E)-CH¼CH(4-C6H4OMe)
(E)-CH¼CH[(3,4-OCH2O)C6H4]
(E)-CH¼CH(3-C6H4CO2H)
a
a
60.142.5
50% @ 300 mM
52.0
a
b
a<50% inhibition at 1mM.
b<50% inhibition at 100 mM.
Acknowledgements
2. Fazioli, F.; Blasi, F. Trends Pharmacol. Sci. 1994, 15, 25.
3. Evans, D. M.; Sloan-Stakleff, K. D. Drug News Perspect.
1997, 10, 85.
4. Madison, E. L. Fibrinolysis 1994, 8 (Suppl. 1), 221.
5. Collen, D.; Lijnen, H. R. Thromb. Haemost. 1995, 74,
161.
6. Murtiashaw, C. W.; Breitenbach, R.; Goldstein, S. W.;
Pezzullo, L.; Quallich, J.; Sarges, R. J. Org. Chem. 1992, 57,
1930.
We thank N. Smith and R. Strang for their assistance in
preparing the compounds, M. F. Burslem, G. Easter
and B. Williams-Jones for the biological data, and the
staff of the Physical Sciences Department, Sandwich,
UK, for analytical data.
7. Bridges, A. J.; Lee, A.; Schwartz, C. E.; Towle, M. J.; Lit-
tlefield, B. A. Bioorg. Med. Chem. 1993, 1, 403.
8. Mattern, G. Helv. Chim. Acta 1977, 60, 2062.
9. Spraggon, G.; Phillips, C.; Nowak, U. K.; Ponting, C. P.;
Saunders, D.; Dobson, C. M.; Stuart, D. I.; Jones, E. Y.
Structure 1995, 3, 681.
References and Notes
1. Barber, C. G.; Dickinson, R. P.; Horne, V. A. Bioorg. Med.
Chem. Lett. 2002, 12, 181.