Synthesis of highly potent and selective hetaryl ureas as integrin αVβ3-receptor antagonists

Article abstract of DOI:10.1016/S0960-894X(02)00161-0

Solid-phase synthesis and SAR of integrin α_Vβ₃-receptor antagonists containing a urea moiety as non-basic guanidine mimetic are described. The most potent compounds exhibited IC₅₀ values towards α_Vβ₃ in the nanomolar range and high selectivity versus related integrins like α_IIbβ₃. For selected examples efficacy in functional cellular assays is demonstrated.

Full text of DOI:10.1016/S0960-894X(02)00161-0

Bioorganic & Medicinal Chemistry Letters 12 (2002) 1379–1382
Synthesis of Highly Potent and Selective Hetaryl Ureas
as Integrin ꢀVꢁ3-Receptor Antagonists
Udo E. W. Lange,a,* Gisela Backfisch,b Jurgen Delzer,b Herve Geneste,b
Claudia Graef,a Wilfried Hornberger,b Andreas Kling,b,* Arnulf Lauterbach,a
Thomas Subkowskia and Christian Zechela
aBASF AG, D-67056 Ludwigshafen, Germany
bAbbott GmbH &Co. KG, D-67008 Ludwigshafen, Germany
Received 12 November 2001; revised 23 January 2002; accepted 6 March 2002
Abstract—Solid-phase synthesis and SAR of integrin aVb3-receptor antagonists containing a urea moiety as non-basic guanidine
mimetic are described. The most potent compounds exhibited IC50 values towards aVb3 in the nanomolar range and high selectivity
versus related integrins like aIIbb3. For selected examples efficacy in functional cellular assays is demonstrated. # 2002 Elsevier
Science Ltd. All rights reserved.
Integrins form a superfamily of heterodimeric trans-
membrane glycoprotein receptors. They are involved in
cell–cell and cell–matrix adhesion, cell migration, and
signaling. Currently 17 a- and 9 b-subunits and at least
20 combinations thereof with different affinities and spe-
cificities to matrix proteins like fibrinogen, fibronectin,
vitronectin, osteopontin and laminin are known.1 Over
the last years integrins have received much attention in
drug discovery research,2 in particular the b3-subfamily
with the fibrinogen receptor aIIbb3 mediating platelet
aggregation,3 and the vitronectin receptor aVb3.
reactivity to aVb3 and aMb2,13 revealed a beneficial
effect on restenosis.14 Selective antagonists of aIIbb3 did
not show this effect while the aVb3 specific monoclonal
antibody LM609 reduced the neointima formation in
animal models.15
Many integrin ligands share the tripeptide sequence
RGD depicted in Figure 1 as a common recognition
motif.16 Studies on cyclic peptides containing the RGD
sequence demonstrated that selective ligands for aVb3
could be obtained if the arginine and aspartate b-carbon
atoms were separated by a distance of 650–700 pm.17
The integrin aVb3 is specifically expressed in activated
proliferating and migrating endothelial or smooth muscle
cells, macrophages, and tumor cells.4 aVb3 is involved in
restenosis after percutaneous transluminal coronary
angioplasty (PTCA),5 angiogenesis and neovasculariza-
tion,6 rheumatoid arthritis,7 diabetic retinopathy and
age-related macular degeneration,8 tumor growth and
metastasis,9 osteoporosis,10 and acute renal failure.11
Therefore, the aVb3 receptor represents an interesting
therapeutic target.12
Most small molecule aVb3-receptor antagonists descri-
bed in the literature follow this pattern and mimic the
RGD motif of the natural ligands by displaying an
acidic and a basic moiety in a defined distance.18 In
these structures the aspartate is often replaced by a
b-amino acid.19
Studies with the antithrombotic abciximab (ReoPro1,
chimeric 7E3Fab), an antagonist of aIIbb3 with cross
*Corresponding authors. Fax: +49-621-60-21156; e-mail: udo.lange@
basf-ag.de (U. Lange); fax: +49-621-589-68916; e-mail: andreas.
kling@abbott.com (A. Kling).
Figure 1. RGD-motif of integrin ligands.
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00161-0

1380
U. E. W. Lange et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1379–1382
As a starting point for our own studies on aVb3-receptor
antagonists20 we chose to connect b-amino acids with a
urea moiety via a spacer element. The urea moiety
serves as a non-basic hydrogen bond donor mimicking
the guanidine, a replacement which has already been
applied in the development of integrin antagonists.21 As
spacer element between the two pharmacophores we
used a combination of an amino acid and a hetaryl
moiety that allowed us to adjust the relative positions
and the distance of the carboxylic group and the urea
moiety.
from commercially available or easily accessible com-
pounds according to Scheme 2.22 A small library of
antagonists was prepared using 10 b-amino acids, 3
Fmoc-amino acids (Gly, b-Ala, Sar), and 8 hetaryl urea
building blocks. The IC50 (aVb3) values in this set of
antagonists varied from more than 100 mM to 0.1 nM.23
One of the best combinations contained racemic b-3-
pyridyl-b-alanine and glycine at the C-terminus. We
used this subset of our library to determine the effect of
the hetaryl moiety on the activity of the antagonist. The
results are summarized in Table 1.24,25
The solid-phase synthesis of the aVb3-receptor antago-
nists described above is depicted in Scheme 1. The
required hetaryl urea building blocks were prepared
The nature of the hetaryl moiety had a strong influence
on the activity of the antagonists. While the 2,5-sub-
stituted thiophene 18a and thiazole 24a25b exhibited
micromolar activity, the corresponding 2,5-substituted
furan 20a and oxazole 22a had IC50 values in the sub-
nanomolar range. The 2,4-substituted thiophene 19a
and thiazole 25a also showed high activity. The N-
methylpyrrole 21a, the 2,4-substituted oxazole 23a and
in particular the triazole 26a were less effective. The low
activity of thiophene 18a and thiazole 24a can be
attributed to wider bond angles in these hetaryls com-
pared to furans26 resulting in a different display of the
two pharmacophores. The loss of activity in derivatives
21a, 23a and 26a compared to 19a, 20a, 22a and 25a
remains unclear. Especially the lower activity of 23a
Scheme 1. (a) Piperidine, DMF; (b) Fmoc-amino acid, TBTU, DIEA,
DMF, rt; (c) piperidine, DMF; (d) HO2C-hetaryl-NH-CO-NH-Bn,
TBTU, DIEA, DMF, rt; (e) AcOH, TFE, DCM. 2-chlorotrityl resin
was used as solid support.
Table 1. Effect of the hetaryl moiety on the activity of the aVb3
antagonists
Compd
Hetaryl
aVb3 IC50 (nM)a
18a
1000
19a
20a
21a
22a
23a
24a
25a
26a
2
0.1
16
0.3
14
10,000
0.3
72
Scheme 2. (a) SOCl2, MeOH; (b) Fe, AcOH; (c) Bzl-NCO; (d) LiOH,
THF, H2O; (e) H2SO4, AcOH, H2O; (f) CO(NH2)2, H2O, Á; (g) NaOH,
MeOH, H2O; (h) CO(NH2)2, EtOH, Á; (i) KOH, dioxane, H2O.
aValues are means of three experiments; intra-assay variation <10%,
inter-assay variation <factor 2.

U. E. W. Lange et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1379–1382
1381
Table 2. Variation of the b-amino acid
compared to 25a is somewhat surprising. Based on these
results the highly active furan derivatives and the 2,5-
substituted oxazoles were chosen24 for a more detailed
SAR study.
The substituents R1 and R2 next to the carboxylic
group were modified to probe their influence on the
antagonist activity. In the furan series most of these
variations from hydrophilic (20a, Table 1) to lipophilic
moieties of very different size (20b–h, Table 2) were well
tolerated with hardly any loss of activity. Oxazoles with
a 3,5-dichlorophenyl side chain (22b) and an a-Z-dia-
minopropionic acid (Z-DAP) moiety (22c) were slightly
less active than the corresponding furan derivatives 20b
and 20c (Table 2).
Compd
R1
R2
aVb3 IC50 (nM)a,b
20b
20c
20d
20e
20f
20g
20h
22b
22c
H
NH-Z
H
H
H
H
H
H
NH-Z
3,5-Dichlorophenyl
H
4-Methylphenyl
3,4-Dimethoxyphenyl
b-Naphthyl
Cyclohexyl
p-Biphenyl
3,5-Dichlorophenyl
H
0.2
0.4
1.9
0.5
0.3
0.9
1.5
3.8
1.8
aCf. Table 1.
Modification of the spacer unit revealed the optimal
distance of the pharmacophores and the essential
hydrogen bond donors in this type of antagonist (Table
3). Replacement of the Z-DAP-Gly unit in 20c by
Z-Orn (27),27 Z-Lys (28) or a b-Z-diaminobutyric acid-
b-alanine unit (31) resulted in substantial loss of activ-
ity. The b-Z-diaminobutyric acid-glycine 32 and -sarco-
sine derivatives 33 were about 100-fold less potent than
the Z-DAP derivative 20c. Exchange of the glycine
building block against sarcosine (30) or b-alanine (29)
resulted in less active compounds with the remarkable
exception of the Z-DAP derivatives 29c and 30c which
still exhibited activity in the nanomolar range.
bAssay results for R2
6¼H refer to racemic compounds.
Table 3. Variation of the amino acid spacer
In summary, the optimal distance between the urea
moiety and the carboxylic group in this series appeared
to be about 11 bonds (cf. 29a–d, 31–33). The sec-
ondary amide groups next to the hetaryl moiety and
the b-amino acid, respectively, are crucial for potency
(cf. 30a–d and 28). The Z-DAP derivatives 29c and 30c
are exceptional in this regard. Here normally unfavor-
able modifications in the spacer element are much better
tolerated.
Compd
R1
R2
R3
aVb3 IC50
(nM)a,b
27
28
1000
5000
144
10,000
3.1
1000
311
50.5
8.9
29a
29b
29c
29d
30a
30b
30c
30d
31
H
H
NH-Z
H
H
H
3-Pyridyl
Cyclohexyl
H
H
H
H
Based on these results our subsequent studies focused
on DAP derivatives, assuming that in this class of
antagonists the pharmacokinetic profile could be mod-
ified by changes in the spacer unit without loss of
activity.
4-Methylphenyl
Cyclohexyl
3,5-Dichlorophenyl
H
H
Me
Me
Me
Me
H
NH-Z
H
4-Methylphenyl
100
2330
43.5
88.2
The Z-DAP derivatives 20c and 22c exhibited high
selectivity versus related integrins like aIIbb3, a5b1, and
a4b1 (Table 4). In early ADME studies 20c and 22c dis-
played medium permeation in the Caco-228 model and
good metabolic stability versus human liver microsomes.
32
33
H
Me
aCf. Table 1.
bAssay results for R2
6
Compd
aVb3/VN
ELISAa
IC50 (nM)
a
IIbb3/Fg
a5b1/FN
Adhesionb
Inh. @ 10ꢀ5
a4b1/FN
Adhesionb
Inh. @ 10ꢀ5
aVb3/OPN
Adhesionb
Inh. @ 10ꢀ5
M
SMC Migrationb
Caco-2 Permeation
assay Pappc, cm/sꢁ10ꢀ7
ELISAa
IC50 (nM)
IC50 (mM)
M
M
20c
22c
0.4
1.8
na
na
na
na
2%
na
86%
92%
1.2
0.5
6
2
aCf. Table 1.
bCell adhesion and migration: values are means of 4 experiments; intra-assay variation <20%, inter-assay variation <factor 2.
cPapp=apparent permeability coefficient;29 n=2, intra-assay variation <20%. Abbreviations: na=not active, VN=vitronectin, FN=fibronectin,
Fg=fibrinogen, OPN=osteopontin.

1382
U. E. W. Lange et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1379–1382
Both compounds showed medium to high efficacy in
functional cellular assays, the cell adhesion of recombi-
nant aVb3 transfected CHO-K1 cells and the cell
migration of primary human smooth muscle cells.23 The
data of the cellular assays did not fully reflect the in
vitro potency of the compounds. Similar observations
have been reported for other aVb3 antagonists.30
16. Ruoslahti, E. Annu. Rev. Cell Dev. Biol. 1996, 12, 697.
17. Haubner, R.; Finsinger, D.; Kessler, H. Angew. Chem.,
Int. Ed. Engl. 1997, 36, 1374.
18. (a) Duggan, M. E.; Hutchinson, J. H. Exp. Opin. Ther.
Pat. 2000, 10, 1367. (b) Solid phase synthesis of aVb3-receptor
antagonists: Hoekstra, W. J.; Poulter, B. L. Curr. Med. Chem.
1998, 5, 195. (c) Rockwell, A. L.; Rafalski, M.; Pitts, W. J.;
Batt, D. G.; Petraitis, J. J.; DeGrado, W. F.; Mousa, S.; Jad-
hav, P. K. Bioorg. Med. Chem. Lett. 1999, 9, 937. (d) Gopal-
samy, A.; Yang, H.; Ellingboe, J. W.; Kees, K. L.; Yoon, J.;
Murrills, R. Bioorg. Med. Chem. Lett. 2000, 10, 1715. (e)
Sulyok, G. A.; Gibson, C.; Goodman, S. L.; Holzemann, G.;
Wiesner, M.; Kessler, H. J. Med. Chem. 2001, 44, 1938.
19. Scarborough, R. M. Curr. Med. Chem. 1999, 6, 971.
20. (a) Kling, A.; Backfisch, G.; Delzer, J.; Geneste, H.;
Graef, C.; Holzenkamp, U.; Hornberger, W.; Lange, U. E.
W.; Lauterbach, A.; Mack, H.; Seitz, W.; Subkowski, T.
Bioorg. Med. Chem. Lett. 2002, 12, 441. (b) Zechel, C.; Back-
fisch, G.; Delzer, J.; Geneste, H.; Graef, C.; Hornberger, W.;
Kling, A.; Lange, U. E. W.; Lauterbach, A.; Seitz, W.; Sub-
kowski, T. manuscript in preparation.
Further pharmacokinetic assays will reveal whether these
antagonists are suitable for relevant efficacy studies in
animal models, e.g., for the prevention of restenosis.31
Acknowledgements
We thank Ulrike Lowen, Marion Stephan, Tanja Zim-
pelmann, Ralf Hogenmuller, Egon Fleischer, Stephanie
Meyer, Karlpeter Orth, Hermann Schulke, Carsten
Thiem, and Sonja Triebel for supporting chemical
synthesis, and Guido Fickartz, Ramona Hoffmann,
Sabine Jockel, Michael Lang, Dirk Mayer and Katja
Schmidt for assay development and screening.
21. Ruminski, P. G.; Clare, M.; Collins, P. W.; Desai, B. N.;
Lindmark, R. J.; Rico, J. G.; Rogers, T. E.; Russell, M. A.
WO 9708145; Chem. Abstr. 1997, 126, 264011.
22. Kling, A.; Lange, U.; Lauterbach, A.; Geneste, H.; Sub-
kowski, T.; Zechel, J.-C.; Graef, C. I.; Hornberger, W. WO
0066618; Chem. Abstr. 2000, 133, 350516.
23. For detailed protocols of assay setup and screening pro-
cedures see ref 22.
References and Notes
1. Eble, J. A.; Kuhn, K. Integrin-Ligand Interaction; Springer-
Verlag: Heidelberg, 1997.
2. (a) Cox, D. Drug News &Perspect. 1995, 8, 197. (b) Hillis,
G. S.; MacLeod, A. M. Clin. Sci. 1996, 91, 639. (c) Kerr, J. S.;
Mousa, S. A.; Slee, A. M. Drug News &Perspect. 2001, 14,
143.
3. Coller, B. S. J. Clin. Invest. 1997, 99, 1467.
4. Brooks, P. C.; Clark, R. A. F.; Cheresh, D. A. Science
1994, 264, 569.
24. (a) A similar approach has been pursued independently by
others: Chandrakumar, N. S.; Desai, B. N.; Devadas, B.; Huff,
R.; Khanna, I. K.; Rao, S. N.; Rico, J. G.; Rogers, T. E.;
Ruminski, P. G.; Russell, M. A.; Gasiecki, A. F.; Malecha, J.
W.; Miyashiro, J. M. WO 9952896; Chem. Abstr. 1999, 131,
2999691 (b) Alig, L.; Edenhofer, A.; Hilpert, K.; Weller, T.
EP928790; Chem. Abstr. 1999, 131, 116229.
25. (a) For synthesis of the thiazole building blocks cf.: ref
24b (2,4-substituted).
5. Bishop, G. G.; McPherson, J. A.; Sanders, J. M.; Hessel-
bacher, S. E.; Feldman, M. J.; McNamara, C. A.; Gimpel,
L. W.; Powers, E. R.; Mousa, S. A.; Sarembock, I. J. Circula-
tion 2001, 103, 1906.
6. Mousa, S. A. Exp. Opin. Ther. Pat. 1999, 9, 1237.
7. Storgard, C. M.; Stupack, D. G.; Jonczyk, A.; Goodman,
S. L.; Fox, R. I.; Cheresh, D. A. J. Clin. Invest. 1999, 103, 47.
8. Hammes, H. P.; Brownlee, M.; Jonczyk, A.; Sutter, A.;
Preissner, K. T. Nature Med. 1996, 2, 529.
(b) 24a was prepared separately in addition to the described
library Thomsen, I.; Pedersen, U.; Rasmusen, P. B.; Yde, B.;
Andersen, T. P.; Lawesson, S.-O. Chem. Lett. 1983, 809 (2,5-
substituted; prepared analogously from H2NCO-CH2-NH-
CO-CO2Et followed by urea formation with benzylisocyanate
and saponification; overall yield: 16%).
26. Bird, C. W.; Cheeseman, G. W. H. In Comprehensive
Heterocyclic Chemistry; Bird, C. W.; Cheeseman, G. W. H.
Ed.; Pergamon: Oxford, 1984; Vol. 4, p 3.
9. Marshall, J. F.; Hart, I. R. Sem. Cancer Biol. 1996, 7, 129.
10. Chorev, M.; Dresner-Pollak, R.; Eshel, Y.; Rosenblatt, M.
Biopolymers (Pept. Sci.) 1995, 37, 367.
27. The Orn derivative 27 was prepared in solution from fur-
anoic acid 9 and Z-Orn-OtBu using standard peptide coupling
methods followed by acidic ester cleavage.
11. Noiri, E.; Gailit, J.; Sheth, D.; Magazine, H.; Gurrath,
M.; Muller, G.; Kessler, H.; Goligorsky, M. S. Kidney Int.
1994, 46, 1050.
28. (a) Artursson, P. Crit. Rev. Ther. Drug Carrier System
1991, 8, 105. (b) Hilgers, A. R.; Conradi, R. A.; Burton, P. S.
Pharm. Res. 1990, 7, 902.
12. (a) Brooks, P. C. Drug News &Perspect. 1997, 10, 456. (b)
Miller, W. H.; Keenan, R. M.; Willette, R. N.; Lark, M. W.
Drug Disc. Today 2000, 5, 397.
29. Papp values >2e-7 cm/s are considered as medium, >2e-6
cm/s as high transport rate.
30. (a) Srivatsa, S. S.; Fitzpatrick, L. A.; Tsao, P. W.; Reilly,
T. M.; Holmes, D. R., Jr.; Schwartz, R. S.; Mousa, S. A.
Cardiovasc. Res. 1997, 36, 408. (b) Mousa, S. A.; Lorelli, W.;
Mohamed, S.; Batt, D. G.; Jadhav, P. K.; Reilly, T. M. Car-
diovasc. Pharmacol. 1999, 33, 641.
13. Coller, B. S. Am. Heart J. 1999, 138, S1.
14. Blindt, R.; Bosserhoff, A. K.; Zeiffer, U.; Krott, N.; Han-
rath, P.; vom Dahl, J. J. Mol. Cell. Cardiol. 2000, 32, 2195 and
references cited therein.
15. van der Zee, R.; Murohara, T.; Passeri, J.; Kearney, M.;
Cheresh, D. A.; Isner, J. M. Cell Adhes. Commun. 1998, 6, 371.
31. Miller, D. D.; Karim, M. A.; Edwards, W. D.; Schartz,
R. S. Atherosclerosis 1996, 124, 145.