2908
N. Xi et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2905–2909
analogue 2d, being unable to form a hydrogen bond
around the aromatic ring, shows significantly loss of
binding potency (Ki ¼ 71 nM).
results indicated the presence of a hydrophobic pocket
near the guanidine binding site and hydrogen bonding
around the aromatic moiety of the b-amino acid, which
are in agreement with the findings from other groups.
The effects of the conformation and configuration of the
c-lactam on the binding were also assessed. Conforma-
tional changes in the central scaffold affect the binding
significantly to polar ligands, such as compounds 1a, 3a,
and 3b, but are less influential in more hydrophobic
ligands. The favored stereochemistry on the c-lactam is
the R-configuration, while on the b-amino acid is the
S-configuration. These SAR results provide a ligand
avb3 binding model and are useful for designing new
avb3 antagonists.
The N-aryl-c-lactam scaffold has a well-defined confor-
mation because of the hindered rotation between the
central phenyl and lactam ring.18 The dihedral angle ð/Þ
between the two rings is dictated by the ortho-sub-
stituent (relative to the lactam) on the phenyl ring.
Different angles ð/Þ guide the guanidine and carboxylic
acid groups to different relative geometries, hence
influence the binding to avb3. This is manifested by the
pyridyl analogues 1a, 3a, and 3b, where a fluorine at the
ortho-position in 3a provides the best binding, while a
smaller H in 1a or a bigger Cl in 3b leads to less
potency.16 Here, the guanidine and carboxylate groups
act like an electrostatic clamp, holding the entire mole-
cule in the binding pocket.19 Only the correct confor-
mation in the central link allows the ligand to interact
with the receptor effectively. Interestingly, the fluorine
adds no benefit to binding when comparing 3c
(Ki ¼ 0:5 nM) with 1d (Ki ¼ 0:7 nM), suggesting the
important contribution of the hydrophobic groups (i.e.,
the methylenes in the cyclic guanidine and 1,3-benzo-
dioxolyl moiety) to the binding. As a result, compounds
1d and 3c are among the most potent avb3 antagonists in
the c-lactam series.
Acknowledgements
We are grateful to Dr. Gilbert Rishton and Dr. Wenge
Zhong for proofreading the manuscript.
References andnotes
1. Giancotti, F. G.; Ruoslahti, E. Science 1999, 285, 1028.
2. For recent reviews, see: Miller, W. H.; Keenan, R. M.;
Willette, R. N.; Lark, M. W. Drug Discov. Today 2000, 5,
397; Coleman, P. J.; Duong, L. T. Exp. Opin. Ther. Pat.
2002, 12, 1009.
Examination of the four individual stereoisomers of
compound 2c in the avb3 assay revealed that the
R-configuration on the c-lactam is required for effective
binding. The corresponding S-configuration is detri-
mental, as seen with 1S,2S-2c and 1S,2R-2c (Ki > 1 lM
for both compounds). This result agrees with the notion
that the relative geometry of the guanidine and car-
boxylic acid groups is an important structural parameter
in the antagonistic binding (vide supra). Apparently,
optimal orientations of the benzylurea (as a guanidine
mimetic) and carboxylate groups are vital for the
antagonists to bind with the receptor.19 Localized
structural modifications such as the chirality change on
the b-amino acid do not alter the relative orientation of
the two ionic groups or their mimetic groups, therefore
exert less dramatic effects on the binding. This is illus-
trated in the compounds with the R-configured c-lactam
core. For example, compound 1R,2S-2c is about 15
times more active than its diastereomer 1R,2R-2c,
favoring the S-configuration on the b-amino acid. This
outcome is substantiated in more potent guanidine
analogues. Thus, the favored isomer 1R,2S-1d binds to
the receptor in picomolar range (Ki ¼ 0:1 nM) while its
diastereomer 1R,2R-1d is a less potent avb3 antagonist
(Ki ¼ 7:3 nM).
3. For review, see: Duong, L. T.; Rodan, G. A. Rev.
Endocrinol. Metab. Dis. 2001, 2, 95.
4. Duggan, M. E.; Duong, L. T.; Fisher, J. E.; Hamill, T. G.;
Hoffman, W. F.; Huff, J. R.; Ihle, N. C.; Leu, C. T.; Nagy,
R. M.; Perkins, J. J.; Rodan, S. B.; Wesolowski, G.;
Whitman, D. B.; Zartman, A. E.; Rodan, G. A.; Hartman,
G. D. J. Med. Chem. 2000, 43, 3736.
5. Xiong, J.-P.; Stehle, T.; Diefenbach, B.; Zhang, R.;
Dunker, R.; Scott, D. L.; Joachimiak, A.; Goodman,
S. L.; Arnaout, M. A. Science 2001, 294, 339.
6. Feuston, B. P.; Culberson, J. C.; Duggan, M. E.; Hart-
man, G. D.; Leu, C.-T.; Rodan, S. B. J. Med. Chem. 2002,
45, 5640.
7. Marinelli, L.; Lavecchia, A.; Gottschalk, K.-E.; Novellino,
E.; Kessler, H. J. Med. Chem. 2003, 46, 4393.
8. Paytash, P. L.; Sparrow, E.; Gathe, J. C. J. Am. Chem.
Soc. 1950, 72, 1415.
9. For the preparation of b-arylamino esters, see: Cardillo,
G.; Gentilucci, L.; Melchiorre, P.; Spampinato, S. Bioorg.
Med. Chem. Lett. 2000, 10, 2755.
10. Zafar, A.; Melendez, R.; Geib, S. J.; Hamilton, A. D.
Tetrahedron 2002, 58, 683.
11. Evans, D. A. Asymmetric Synth. 1984, 3, 1.
12. The single crystal of compound 12a was obtained from
50% EtOAc in hexane and its structure was determined by
a Rigaku AFC7R diffractometer with graphite monochro-
mated Cu-Ka radiation and a rotating generator: crystal
system, orthorhombic; lattice type, primitive; lattice
Generally, the c-lactam derivatives are highly selective
for avb3 versus aIIbb3. Among all the compounds listed
in Table 2, only compound 1R,2S-1d shows marginal
binding (Ki ¼ 2100 nm) toward aIIbb3, affording more
than 21,000-fold selectivity favoring avb3. None of the
other compounds bind to aIIbb3 (Ki >25,000 nM).
ꢀ
ꢀ
parameters: a ¼ 14:817(4) A; b ¼ 20:717(4) A; c ¼
3
ꢀ
ꢀ
6:278(3) A; V ¼ 1927:0(9) A ; space group P212121 (#19);
residuals: R, 0.088; Rw, 0.067.
13. Soloshonok, V. A.; Fokina, N. A.; Rybakova, A. V.;
Shishkina, I. P.; Galushko, S. V.; Sorochinsky, A. E.;
Kukhar, V. P. Tetrahedron: Asymmetry 1995, 6, 1601.
14. Ligand binding assay. Ruthenylation of Vitronectin and
Fibrinogen: Purified human vitronectin (Yatohgo, T.;
Izumi, M.; Kashiwagi, H.; Haiyashi, M. Cell Struct. Funct.
In conclusion, we found that the c-lactam derivatives
were potent and selective avb3 antagonists. Our SAR