Synthesis and biological activity of peptidyl aldehyde urokinase inhibitors

Article abstract of DOI:10.1016/S0960-894X(00)00149-9

Solid- and solution-phase synthesis of peptidomimetric inhibitors of urokinase-type plasminogen activator based on the sequence dSerAlaArg-al are described. The biological activities of these unique inhibitors are reported herein. Carbonate prodrugs were prepared and tested as potential drug delivery systems. (C) 2000 Published by Elsevier Science Ltd.

Full text of DOI:10.1016/S0960-894X(00)00149-9

Bioorganic & Medicinal Chemistry Letters 10 (2000) 983±987
Synthesis and Biological Activity of Peptidyl Aldehyde
Urokinase Inhibitors
Susan Y. Tamuray, Michael I. Weinhouse, Christopher A. Roberts, Erick A. Goldman{,
Kevin Masukawa, Susanne M. Anderson, Cheryl R. Cohenx, Annette E. Bradbury,
Vernon T. Bernardino, Steven A. Dixon, Michael G. Ma, Thomas G. Nolan*
and Terence K. Bruncky
Corvas International, Inc., Department of Medicinal Chemistry, 3030 Science Park Road, San Diego, CA 92121, USA
Received 22 December 1999; accepted 29 February 2000
AbstractÐSolid- and solution-phase synthesis of peptidomimetic inhibitors of urokinase-type plasminogen activator based on the
sequence dSerAlaArg-al are described. The biological activities of these unique inhibitors are reported herein. Carbonate prodrugs
were prepared and tested as potential drug delivery systems. # 2000 Published by Elsevier Science Ltd.
Urokinase-type plasminogen activator (u-PA) is one of
the two major endogenous plasminogen activators that
catalyze the conversion of the zymogen plasminogen to
the ®brinolytic protease plasmin.1 The primary role of
u-PA is to generate plasmin in events involving the
degradation of the extracellular matrix.2 Localization of
u-PA on the cell surface is achieved by binding to uro-
kinase plasminogen activator receptor (u-PAR),3 which
is attached to the cell membrane via its glycosyl phos-
phatidyl inositol (GPI) anchor. Recent advances in the
elucidation of the function of the u-PA/u-PAR system
have led to an increased understanding of the role
played by this enzyme in angiogenesis,4 cell invasion,5
and cancer metastasis.6 Considerable e€orts are being
focused on the development of selective direct and
mechanism-based synthetic u-PA inhibitors.7 Inhibitors
of u-PA are potential therapeutic targets for cancer,
arthritis, and pathological angiopathies.8
this library based on in vitro activity against human
u-PA (Fig. 1). The solid- and solution-phase syntheses of
peptidomimetic inhibitors of u-PA based on the tripep-
tide sequence of 1a and 1b, d-SerAlaArg-al, is described
herein. The targets possess an interesting range of
topographical and physical properties including polarity
and lipophilicity, which further elucidated the require-
ments for potential inhibitors in the active site of u-PA.
Many of these peptidyl aldehydes possess signi®cant
levels of potency against urokinase and selectivity
towards other serine proteases, including t-PA.
Solid-Phase Chemistry9
Two solid-phase methodologies were used in the prep-
aration of 2±10 mg quantities of ®nal target argininals:
the argininal aminal linker methodology and the peptidyl
aldehyde-HCAM resin methodology.
A variety of tetrapeptide P1-argininals were prepared
using a library approach. The lead structures 1a and 1b
were discovered by screening individual compounds in
The argininal aminal linker methodology10 was utilized
in the preparation of the targets 2, 5±7 and 9. The
preparation of compound 2 is exempli®ed in Scheme 1.
Iterative TBTU/HOBt mediated couplings with the
appropriate N-a-Fmoc-amino acids and piperidine de-
blocking protocols produced P3±P2 intermediate 14, which
was capped with a sulfonamide P4 group to generate the
fully elaborated intermediate 15. The production of 2
from 15 was achieved in the following manner: catalytic
removal of the Ng-alloc moiety and subsequent acidic
hydrolysis to cleave the aldehyde from the resin while
*Corresponding author. Tel.: +1-858-455-9800; fax: +1-858-455-
7895; e-mail: tom_nolan@corvas.com
yPresent address: Bioreason, Inc., 150 Washington Ave., Suite 303,
Santa Fe, NM 87501, USA.
{Present address: COR Therapeutics Inc., 256 E. Grand Ave., San
Francisco, CA 94080, USA.
xPresent address: Allergan Pharmaceuticals, 2525 Dupont Drive,
Irvine, CA 92612, USA.
0960-894X/00/$ - see front matter # 2000 Published by Elsevier Science Ltd.
PII: S0960-894X(00)00149-9

984
S. Y. Tamura et al. / Bioorg. Med. Chem. Lett. 10 (2000) 983±987
removing the t-butyl protecting group from the serine
hydroxyl.
Solution-Phase Chemistry9
The preparation of milligram to multigram quantities of
argininal 1a for in vivo evaluation was facilitated by the
solution-phase aminal methodology previously described
wherein a protected argininal synthon, nitroargininal
ethyl cyclol,13 was coupled to a peptide surrogate
(Scheme 3). After d-serine t-butyl ether was capped with
isobutylchloroformate, the resulting iBoc-d-serine t-
butyl ether was coupled to alanine t-butyl ester to a€ord
compound 17. The N-capped dipeptide 17 was depro-
tected with TFA, then coupled to nitroargininal ethyl
cyclol. The nitro group of the fully elaborated target 18
was hydrogenolyzed, and the aminal was hydrolyzed
under mild conditions to reveal the aldehyde. RP HPLC
puri®cation yielded the ®nal target.
The peptidyl aldehyde-HCAM resin methodology11
incorporates an aldehyde protected as a semicarbazone
that is directly attached to aminomethylpolystyrene
resin. The peptidyl-argininal synthesis via the HCAM
resin, similar in practice to the above mentioned argin-
inal aminal solid phase iterative method, was utilized in
the preparation of 3, 4, 8, and 10±11; however, the ®nal
targets were released from the resin using a di€erent
cleavage protocol. Beginning with Fmoc-Arg(Boc)2-
HCAM resin 16,11 piperidine deblocking protocols and
iterative TBTU/HOBt mediated couplings produced the
resin-bound intermediates. Final hydrolysis with 90%
aqueous TFA concomitantly removed the peptide
protecting groups and cleaved the target aldehydes 10
and 11 from the resin. In the case of compound 4
(Scheme 2), the capped P3 amino acid, 2-phenethyl
SO2-dSer-OH, was prepared in solution before being
coupled in this iterative manner.12 Overall, the di€er-
ences in the argininal aminal linker and argininal
HCAM methodologies were found in both the ease of
guanidine deprotection and higher yields using the
HCAM method.
In Vitro Pharmacology: Results and Discussion
Eleven peptidyl argininals were prepared and evaluated
for their ability to inhibit various serine proteases
(Table 1). Potent and selective inhibitors of u-PA were
discovered. From this focused set of compounds, we
were able to discover several determinants for the
Figure 1. Design of new peptidyl original targets based on structures 1a and 1b.
Scheme 1. Reagents and conditions: (a) Fmoc-Ala-OH, TBTU, HOBt, DMF, DIEA; (b) 10% piperidine in DMF; (c) Fmoc-dSer(t-Bu)OH, TBTU,
HOBt, DMF, DIEA; (d) benzenesulfonyl chloride, 2,4,6-collidine, CH2Cl2; (e) Pd(PPh3)4, morpholine, THF, DMSO, 0.5M HCl; (f) TFA:
CH2Cl2:H2O (6:3:1), 3±4 h; preparative RP HPLC.
Scheme 2. Reagents and conditions: (a) 10% piperidine in DMF; (b) Fmoc-Ala-OH, TBTU, HOBt, DMF, DIEA; (c) phenethylSO2dSer(O-t-Bu)OH,
TBTU, HOBt, DMF, DIEA; (d) TFA:H2O (9:1); preparative RP HPLC.

S. Y. Tamura et al. / Bioorg. Med. Chem. Lett. 10 (2000) 983±987
985
Scheme 3. Reagents and conditions: (a) i-BuOCOCl, aq Na2CO3, 99%; (b) Ala t-butyl ester, EDC, HOBt, CH3CN, DIEA, quant.; (c) TFA,
CH2Cl2, quant.; (d) Ng-nitroargininal ethyl cyclol, HCl salt, EDC, HOBt, CH3CN, DIEA, 50%; (e) 50 psi H2, 10% Pd/C, EtOH, HOAc, H2O, 4 h;
(f) 3.0 M HCl; preparative RP HPLC, 62% for two steps.
Table 1. In vitro IC50 values (nM) of peptidyl P1 argininals against urokinase, plasmin, and t-PAa
Compound
Synthetic methodb
P4
P3
P2
Urokinase
Plasmin
t-PA
1a
1b
2
C
A
A
B
B
A
A
A
B
A
B
B
i-Boc
Cbz
PhSO2
BnSO2
2-phenethylSO2
Cbz
d-Ser
d-Ser
d-Ser
d-Ser
d-Ser
d-Thr
d-alloThr
d-Ser
d-Ser
d-Ser
d-Ser
d-Me-d-Ser
Ala
Ala
Ala
Ala
Ala
Ala
Ala
Pip
Pro
Aze
d-Ala
Ala
23.1
19.0
10.3
5.1
3.1
261
1460
904
1170
275
367
699
282
Ð
125
482
>2500
>2500
>2500
>2500
>2500
>2500
>2500
>2500
>2500
>2500
>2500
>2500
>2500
>2500
314
4
5
6
7
814
9
Cbz
Cbz
Cbz
Cbz
Cbz
Cbz
33.4
101
12.4
11.0
>2500
>2500
10
11
aConcentration of compounds 1a, 1b, 2±12 necessary to inhibit human enzyme (urokinase, plasamin, and t-PA) cleavage of the chromogenic sub-
strates described in ref 15 by 50%.
bMethod A: aminal-linker solid-phase synthesis; Method B: semicarbazone solid-phase synthesis; Method C: solution-phase synthesis.
potency and selectivity of peptidyl argininal u-PA inhi-
bitors. Changes in the P4 position to sulfonamides from
carbamates produced potent inhibitors, the best being
the phenethylsulfonamide 4. The stereochemistry of the
hydroxyl group of P3 is critical to binding when threo-
nine replaces the serine: d-allo-Thr containing target 6
is 5-fold more potent than d-Thr derivative 5. Surpris-
ingly, replacement of d-Ser with an a-methyl-d-Ser in
the P3 position resulted in the abolished activity of tar-
get 11. This may be due to di€erences in the ground
state conformation of the quarternary C-a of 11,
including undesirable changes in the f and c angles at
the P3 position. When cyclic amino acids were placed in
the P2 position, the following general trend in activities
was observed: the smaller the ring size, the greater the
potency of the target against u-PA. In this series, the P2
analogues 8 and 9 were the most potent against u-PA;
however, these compounds exhibited a decreased selectiv-
ity towards plasmin.
measured over 24 h. Pharmacokinetic analysis revealed
an apparent terminal elimination half-life (t1/2) of
300‹31 min, with a Cmax of 4.7‹0.7 mg/mL (see Table
2). Further biological evaluation of this candidate was
limited due to the short plasma half-life following oral
or sc administration. Mechanisms to improve the phar-
macokinetics, while retaining the high level of bioavail-
ability of this general class of compounds, are under
active study. The highly polar and hydrophilic nature of
1a was proposed to have a detrimental e€ect on the
plasma half-life of the inhibitors. Carbonate moieties
possessing a broad range of physical properties includ-
ing lipophilicity and polarizability may impart desirable
pharmacokinetic properties upon prodrugs of this class
of inhibitors. In addition, variations in the carbonate
may a€ect its rate of hydrolysis or enzymatically cata-
lyzed cleavage in vivo.
Prodrugs 19a±c were prepared17 (Scheme 4) in order to
test the e€ect of carbonates on the in vivo plasma half-
life of 1a. The simple preparation of these compounds
required the derivatization of the dSer hydroxyl of
compound 18 with the appropriate chloroformate in
pyridine. Hydrogenolysis of the nitro group followed
by hydrolysis of the ethyl aminal produced the desired
targets.
Prodrugs
Although compound 1a displayed a Cmax of 22.9‹2.3
mg/mL and an overall AUC of 2333‹361 mgꢀ min/mL
in conscious dogs (20 mg/kg po, n=3), its apparent
terminal elimination half-life was only 51‹3.3 min.16
Compound 1a was administered sc to rats at a dose of
50 mg/kg (n=3), and its plasma concentration was
Prodrugs 19a±c were easily converted to 1a in plasma,
where the half-life of the prodrugs was determined by

986
S. Y. Tamura et al. / Bioorg. Med. Chem. Lett. 10 (2000) 983±987
Table 2. Pharmacokinetic parameters of 1a (mean‹sem): comparison of 1a and 19a dosed in rat (n=3, 50 mg/kg sc)
Compound
Apparent terminal
half-life of 1a (h)
Cmax (mg/mL)
Tmax (h)
AUC (mgꢀ min/mL)
1a
Prodrug 19a
5.0‹0.52
10.7‹2.4
4.7‹0.7
2.3‹0.1
2.33‹0.88
2.67‹0.67
2672‹351
2316‹49
Scheme 4. Reagents and conditions: (a) ROCOCl, pyr, 45±62%; (e) 50 psi H2, 10% Pd/C, EtOH, HOAc, H2O, 4 h; (f) 3.0 M HCl; preparative RP
HPLC, 15±33% for two steps.
analytical RP HPLC analysis. The half-life of 19a
ranged from a few minutes in rat and mouse plasma to
45±75 min in dog and human plasma. Facile cleavage of
19c to 1a was also observed in rat plasma. Surprisingly,
it was found that prodrug moiety of 19b was unstable at
neutral pH in aqueous solution.
Acknowledgements
The authors wish to thank L. Truong and E. Ganio for
technical support in performing in vitro assays. The
authors also wish to thank G. P. Vlasuk for critical
reading of the manuscript.
Prodrug 19a was administered sc to rats at a dose of 50
mg/kg (n=3), and plasma concentration of prodrug 19a
and 1a were measured over 30 h.16 As predicted, no 19a
was observed. The apparent terminal elimination half-
life (t1/2) of 1a was 10.7‹2.4 h, with a Cmax of 2.3‹0.1
mg/mL. The comparison of pharmacokinetic data of
compound 1a derived from dosing both 1a (plasma
concentration was measured over 24 h) and 19a in rats
is shown in Table 2. The major di€erences in the plasma
concentration-time pro®le of 1a following dosing with
19a compared to 1a are an extended duration and
attenuated Cmax with little e€ect on the overall exposure
(relative bioavailability approximately 87%). The
apparent terminal elimination half-life appeared to be
longer after dosing with 19a; however, the elimination
phase was not linear and, since the mechanistic details
of pro-drug conversion in vivo are not well understood,
the apparent terminal elimination half-life should be
interpreted with caution.
References and Notes
1. Kucinski, C. S.; Fletcher, A. P.; Sherry, S. J. Clin. Invest.
1968, 47, 1238.
2. Saksela, O. Biochem. Biophys. Acta 1985, 823, 35.
3. (a) Vassalli, J.-D.; Baccino, D.; Belin, D. J. Cell Biol. 1985,
100, 86. (b) Stoppelli, M. P.; Corti, A.; Soentini, A.; Cassani,
G.; Blasi, F.; Assoian, R. K. Proc. Natl. Acad. Sci. USA 1985,
82, 4939.
4. (a) Rabbani, S. In Vivo 1998, 12, 135. (b) Tkachuk, V.;
Stepanova, V.; Little, P. J.; Bobik, A. Clin. Expl. Pharmacol.
Physiol. 1996, 23, 759. (c) Mignatti, P.; Rifkin, D. B. Enzyme
Prot. 1994, 49, 117. (d) Auerbach, W.; Auerbach, R. Pharmac.
Ther. 1994, 63, 265.
5. (a) Kim, J.; Yu, W.; Kovalski, K.; Ossowski, L. Cell 1998,
94, 353. (b) Ossowski, L. Cancer Res. 1992, 52, 6754. (c)
Ossowski, L. Cell Biol. 1988, 107, 2437. (d) Quigley, J. P.;
Gold, L. I.; Schwimmer, R.; Sullivan, L. M. Proc. Natl. Acad.
Sci. USA 1987, 84, 2776.
6. Andreasen, P. A.; Kjoller, L.; Christensen, L.; Du€y, M. J.
Int. J. Cancer 1997, 72, 1.
7. (a) Bridges, A. J.; Lee, A.; Schwartz, C. E.; Towle, M. J.;
Little®eld, B. A. Bioorg. Med. Chem. 1993, 1, 403. (b) Towle, M.
J.; Lee, A.; Maduakor, E. C.; Schwartz, C. E.; Bridges, A. J.;
Little®eld, B. A. Cancer Res. 1993, 53, 2553. (c) Wakselman, M.;
Xie, J.; Mazaleyrat, J.-P.; Boggetto, N.; Vilain, A.-C.; Mon-
tagne, J.-J.; Reboud-Ravaux, M. J. Med. Chem. 1993, 36, 1539.
8. (a) Rabbani, S. A.; Harakidas, P.; Davidson, D. J.; Henkin,
J.; Mazar, A. P. Int. J. Cancer 1995, 63, 840. (b) Alonso, D. F.;
Farõas, E. F.; Ladeda, V.; Davel, L.; Puricelli, L.; Bal de Kier
Jo€e, E. Breast Cancer Res. Treat. 1996, 40, 209. (c) Jankun,
J.; Keck, R. W.; Skrzypczak-Jankun, E.; Swiercz, R. Cancer
Res. 1997, 57, 559. (d) Xing, R. H.; Mazar, A.; Henkin, J.;
Rabbani, S. A. Cancer Res. 1997, 57, 3585. (e) Mullins, D. E.;
Rohlich, S. T. Biochim. Biophys. Acta 1983, 695, 177.
Conclusion
Solid-phase synthesis of peptidyl argininal targets has
resulted in further delineation of the active site require-
ments of urokinase plasminogen activator and in the
discovery of potent and selective inhibitors. Scaleup
of 1a and preparation of its prodrugs has allowed
pharmacokinetic analysis of compound 1a in rats. This
information will be useful in dosing rats for in vivo e-
cacy studies. Further investigation in animal models of
angiogenesis, tumor angiogenesis and tumor invasion is
ongoing and will be reported in due course.

S. Y. Tamura et al. / Bioorg. Med. Chem. Lett. 10 (2000) 983±987
987
9. All new compounds were characterized by spectroscopic
(1H NMR, low resolution MS) data. Yields refer to spectro-
scopically and chromatographically homogeneous (595% RP
Richard, B. M.; Nolan, T. G.; Hakanson, K.; Tulinsky, A.;
Nutt, R. F.; Ripka, W. C. J. Med. Chem. 1996, 39, 4531. (b)
Urokinase (Abbokinase), manufactured by Abbott Labora-
tories, was obtained from Priority Pharmaceuticals (San
Diego, CA). The potency of inhibitors (IC50) was determined
as described in ref 15a, using the chromogenic substrate S-
2444 (l-Pyroglutamyl-glycyl-l-arginine-p-nitroaniline hydro-
chloride), obtained from DiaPharma Group, Inc. (West Chester,
OH). All reactions contained, at ®nal concentration, 0.75 nM
urokinase and 250 mM S-2444. The concentrations of plasmin
and t-PA and their substrates were as follows: plasmin (1 nM), S-
2366 (300 mM), t-PA (1 nM), Pefachrome t-PA (800 mM).
16. The determination of plasma levels in animals was accom-
plished using solid-phase extraction sample preparation with
HPLC mass spectrometry to quantify compound 1a and its
prodrugs using methodologies that will be published else-
where. The data were evaluated for pharmacokinetic para-
meters using WinNonlin noncompartmental analysis.
1
HPLC, H NMR or TLC) materials.
10. Siev, D. V.; Gaudette, J. A.; Semple, J. E. Tetrahedron
Lett. 1999, 40, 5123.
11. Siev, D. V.; Semple, J. E. Organic Lett. 2000, 2, 19.
12. (a) d-Ser methyl ester, HCl salt, was converted to 2-
phenylethylSO2-dSer(O-t-Bu)-OH in two steps: capping with
phenethylsulfonyl chloride and collidine in acetonitrile, and
hydrolysis of the methylester with 1.0 N LiOH in aqueous
dioxane. (b) Compound 3 was prepared in an analogous manner.
13. Tamura, S. Y.; Semple, J. E.; Ardecky, R. A.; Leon, P.;
Carpenter, S.; Ge, Y.; Shamblin, B. M.; Weinhouse, M. I.;
Ripka, W. C.; Nutt, R. F. Tetrahedron Lett. 1996, 37, 4109.
14. Compounds 3 and 8 were spectroscopically and chroma-
tographically homogeneous (590% RP HPLC).
15. (a) Semple, J. E.; Rowley, D. C.; Brunck, T. K.; Ha-Uong,
T.; Minami, N. K.; Owens, T. D.; Tamura, S. Y.; Goldman, E.
A.; Siev, D. V.; Ardecky, R. A.; Carpenter, S. H.; Ge, Y.;
17. Compounds 19a±c are isolated as an 85:15 mixture of
l-Ala to d-Ala isomers.