Thiophene-based vitronectin receptor antagonists

Article abstract of DOI:10.1016/S0960-894X(02)00942-3

A series of α_vβ₃ antagonists based on a thiophene scaffold were synthesized via two routes and evaluated for in vitro biological activity. We have identified several structurally similar antagonists with different selectivities towards α_IIbβ₃, α_vβ₅ and α₅β₁ at the cellular level. In addition, these antagonists exerted an antiangiogenic effect in the chick chorioallantoic membrane (CAM) assay.

Full text of DOI:10.1016/S0960-894X(02)00942-3

Bioorganic & Medicinal Chemistry Letters 13 (2003) 503–506
Thiophene-Based Vitronectin Receptor Antagonists
Monica Bubenik,* Karen Meerovitch, Frederic Bergeron,
Giorgio Attardo and Laval Chan
´
Shire BioChem Inc., 275 Armand-Frappier Blvd., Laval, Quebec, Canada H7V4A7
Received 4 July 2002; accepted 7 October 2002
Abstract—A series of avb3 antagonists based on a thiophene scaffold were synthesized via two routes and evaluated for in vitro
biological activity. We have identified several structurally similar antagonists with different selectivities towards aIIbb3, avb5 and
a5b1 at the cellular level. In addition, these antagonists exerted an antiangiogenic effect in the chick chorioallantoic membrane
(CAM) assay.
# 2002 Elsevier Science Ltd. All rights reserved.
The survival of vascular endothelial cells during angio-
genesis is dependent on the interaction with extra-
cellular matrices (ECM), which are mediated by the
integrin family of cell adhesion receptors. Integrin avb3
is minimally expressed on resting blood vessels, but is
significantly up-regulated on vascular cells within
human tumors such as breast, and in response to
growth factors in the chick chorioallantoic membrane
(CAM).1,2 Integrin avb3 shares a common b3 subunit
with the platelet fibrinogen receptor, aIIbb3, which is a
key player of platelet aggregation. Both these integrins
bind to ECM proteins through the tripeptide sequence,
arginine-glycine-aspartic acid (RGD). Disruption of
avb3-ECM interactions with cyclic RGD peptides causes
apoptosis of angiogenic endothelial cells, resulting in the
disruption of neovascularization and inhibition of
tumor growth.3,4 Thus, the criteria for potential cancer
therapeutics include potent avb3 compounds which are
selective towards aIIbb3, since antagonists of aIIbb3 are
associated with undesirable bleeding complications.
Integrin avb5 is also an RGD dependent vitronectin
adhesion receptor that plays a critical role in angiogen-
esis and is up-regulated in neuroblastoma.5,6 The fact
that the RGD tripeptide sequence serves as a recogni-
tion motif in multiple ligands for several different
integrins (such as avb3, aIIbb3, avb5) has prompted con-
siderable drug discovery efforts in designing peptidomi-
metics.7 Most of the attention has been directed towards
blockade of the avb3 but there is increasing evidence
supporting the beneficial use of dual avb3/avb5 antago-
nists for antiangiogenesis therapy.8À10 It is clear, how-
ever, that avb3 and avb5 are not the only integrins
involved in angiogenesis since it has been recently
shown that tumors in av-null mice displayed normal
vascular development and even enhanced growth.11
Recent experiments have highlighted the importance of
integrins in the b1 family (such as a5b1) in angiogenesis.12
Structural features important for both integrin affinity
and specificity to avb3 have been derived from studies of
cyclic RGD peptide antagonists13 and other non-pep-
tides.14,15 Much of the patent literature describes avb3
antagonists that incorporate one or more rings as a
central constraint,7 including several disclosures cover-
ing the use of thiophenes.16 The central constraint bears
basic and acidic side chains, the 2,3-diaminopropanoic
acid being the most extensively used,17 which mimic the
guanidine and carboxylate of the RGD sequence. The
overall separation between the guanidine and carboxylic
acid termini, and conformation of the antagonist are
critical in achieving potency and specificity. With these
considerations in mind, along with a survey of the lit-
erature for potent avb3 antagonists as our starting point,
we focused our efforts on identifying potent and selec-
tive avb3 antagonists which could potentially prove
useful in cancer therapy. Herein, we describe the syn-
thesis and biological profile of a series of RGD peptido-
mimetic antagonists consisting of a guanidine linked via
an aryl moiety to a central thiophene scaffold.16c
Our key strategy was to vary the positioning of the
guanidinyl residue relative to the central thiophene
*Corresponding author. Tel.: +1-450-978-7805; fax: +1-450-978-
7777; e-mail: mbubenik@ca.shire.com
0960-894X/03/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00942-3

504
M. Bubenik et al. / Bioorg. Med. Chem. Lett. 13 (2003) 503–506
scaffold so as to mimic different RGD conformations,
as well as find the optimal separation of the guanidinyl
residue from the carboxylic acid terminus and determine
their effects on integrin affinity and selectivity. Based on
earlier work done in our laboratories, we decided to
conformationally restrict the basic side chain by intro-
duction of an aryl group in hopes of improving potency.
Hence a series of o-, m-, and p-substituted phenyl or
benzyl guanidines were prepared. The aryl-linked thio-
phenes were rapidly accessed using a palladium cata-
lyzed cross-coupling with in situ generation of the
stannanes, which has been cited in the literature for the
preparation of various biaryls18 (Scheme 1). The cou-
pled products were obtained in modest yields along with
homo-coupled side-products. The guanidines were pre-
pared according to literature procedures19 in varying
yields, with the ortho giving the lowest yield, pre-
sumably due to steric crowding. Simultaneous cleavage
of the tert-butyl ester and the bis-BOC guanidine afforded
the final compounds, 2–9 (Table 1), in quantitative
yields as the TFA salts.
Antagonists (3, 5, 7, 9), which were further profiled,
were efficiently prepared on a multi-gram scale employ-
ing a Suzuki cross-coupling.20 We found it convenient
to use a reusable polymer supported Pd(PPh3)4 catalyst
as described by Fenger and Le Drian to carry out the
cross-coupling.21 Thus, commercially available 3-amino-
phenyl boronic acid hemisulfate (12), thiophene 1,
polymer-supported Pd(PPh3)4 catalyst and aqueous
sodium carbonate in DME, was degassed and heated
at 80 ꢀC affording the coupled aniline-thiophenes
quantitatively (Scheme 2). Guanidinylation and depro-
tection was effected as before giving the final com-
pounds 7 and 9 in 52and 46% overall yields,
respectively. In the case of the benzylamine-thiophene
antagonists, intermediate 11 was prepared in two steps
by guanidinylation followed by conversion to the
boronic acid (Scheme 2). Thus, to 10 was added one
equivalent of methyl lithium to abstract the NH protons
followed by one equivalent of t-butyl lithium to effect
lithium-halogen exchange.22 Quenching with triisopro-
pyl borate and subsequent hydrolysis afforded the
boronic acids in good yields. The cross-coupling pro-
ceeded in acceptable yields giving a mixture of bis-BOC
and mono-BOC compounds. After final deprotection, 3
and 5 were obtained in 55 and 65% overall yields,
respectively.
Compounds 2–9 were tested for their ability to inhibit
biotinylated-fibrinogen from binding to avb3 and aIIbb3
purified proteins in a solid-phase receptor binding assay.
The IC50 values are summarized in Table 1. Once the
IC50 values fell in the nM range, this assay did not prove
useful in differentiating compounds. Therefore, a cell
adhesion assay was chosen to assess the potency of the
avb3 antagonists as this was observed to be a more
stringent assay for differentiating between them. The IC50
values for the compounds to inhibit the binding of K562
cells engineered to overexpress the avb3 receptor from
attaching to immobilized fibrinogen are summarized in
Scheme 1. (a) 3-Bromo-BOC benzyl amine or 3-bromo-BOC aniline,
(SnBu3)2, (Ph3P)2PdCl2, dioxane, 90 ꢀC, (27–38%); (b) HCl (4 M in
1,4-dioxane), 3 min, rt (100%); (c) (tert-butoxycarbonylimino-pyrazol-
1-yl-methyl)-carbamic acid tert-butyl ester, diisopropylethylamine,
DMAP, DMF, 55 ꢀC, (14–83%); (d) TFA/CH2Cl2 (1:1), rt (60–99%).
Table 1. Thiophene-based antagonists 2–9
Compd
n
R1
avb3 Fg IC50 (nMÆSD)a
K562/avb3-Fg IC50 (nMÆSD)a
aIIbb3 Fg IC50 (nMÆSD)a
aIIbb3/avb3
2
3
4
5
6
7
8
9
1
1
1
1
0
0
0
0
o
m
p
m
o
m
p
m
Ph
Ph
Ph
Mes
Ph
Ph
Ph
0.04Æ0.001
2.2Æ1.7
2.1Æ0.07
3.4Æ4.1
5500
1.0Æ0.9
71.5Æ40.3
1.4Æ1.3
21Æ9.9
0.74Æ0.45b
730Æ480
0.0058Æ0.0036
0.34Æ0.13b
6.1Æ5.0
<0.1
<0.5
<0.1
<0.1
<0.1
3
11Æ7.4
0.16Æ0.07b
0.25Æ0.18
5330Æ4490
0.19Æ0.13
0.59Æ0.28b
0.0013Æ0.00014
0.31Æ0.16b
3790Æ2920
<0.1
300
Mes
0.00092Æ0.00083b
aThe IC50 values denote the concentration required to reduce the binding of biotinylated fibrinogen (Fg) to purified avb3 and aIIbb3 and K562cells
expressing avb3 (K562/avb3) to immobilized fibrinogen by 50%. Results are shown as a meansÆsd based on a minimum of at least two independent
experiments performed in triplicate. See ref 24 for full experimental details.
bStandard average error of mean (SEM) was determined.

M. Bubenik et al. / Bioorg. Med. Chem. Lett. 13 (2003) 503–506
505
shortening the space between the guanidinyl and phenyl
moieties by one carbon (comparing 5 and 9), there is a
dramatic enhancement in potency and selectivity where
compound 9 displays an activity against avb3 in the
picomolar range and is 300-fold selective. We attribute
the high potency to the combined effects of the mesityl
group and the shorter overall separation between the
guanidine and carboxylic acid, observations which have
been documented in the literature as giving improved
potency in compounds derived from 2,3-diaminopropa-
noic acid.14,15,23
Compounds 3, 5, 7 and 9 may additionally represent an
interesting class of dual antagonists since they display
good activities against avb5 directed cell adhesion to
vitronectin (Table 2). Interestingly, compounds 5, 7 and
9 were also found to antagonize a5b1 mediated cell
adhesion to fibronectin whereas 3 did not. There is an
enhancement in a5b1 potency when changing the phenyl
for a mesityl group on the sulfonyl (comparing 3 to 5
and 7 to 9), an observation which, on a different scaf-
fold, has been noted in the literature.16
Scheme 2. (a) As in (d), (55%); (b) MeLi (1 equiv), t-BuLi (1 equiv),
B(iOPr)3, ether, À78 ꢀC to rt, o/n (65%); (c) polymer supported
Pd(PPh3)4, NaCO3 (aq, 2M), DME, 80 ꢀC (55–99%); (d) (tert-
butoxycarbonylimino-pyrazol-1-yl-methyl)-carbamic acid tert-butyl
ester, DIPEA, DMAP, DMF, 55 ꢀC, (51–64%); (e) TFA/CH2Cl2 (1:1),
rt (91–99%).
To assess whether selected compounds were able to
exert an anti-angiogenic activity, they were tested in the
chick mesh CAM assay using VEGF as an angiogenic
stimulus (Table 3). All compounds inhibited VEGF-
induced angiogenesis in a dose-dependent manner, with
5 being particularly potent at low concentrations com-
pared to 3 and 7.
Table 2. Thiophene-based antagonists 3, 5, 7 and 9
Compd
HT29/VN avb5
avb5/
avb3
K562/FN a5b1
a5b1/
avb3
IC50 (nMÆSEM)a
IC50 (nMÆSEM)a
We have identified potent avb3 antagonists by system-
atic positioning of the guanidinyl moiety around the
aryl-thiophene, giving compounds with varying integrin
affinities as determined by cell adhesion studies. Com-
pound 9 is a highly potent avb3 antagonist which
exhibited the best selectivity against aIIbb3 of the series.
Moreover, 9 showed activity against avb5 and a5b1, yet
still exhibited preferential selectivity for avb3. In con-
trast, while compounds 3, 5 and 7 showed activity
against avb5 and a5b1, 5 was the most non-selective for
all integrins tested, and 3 and 7 were somewhat selective
for avb3. Antagonists 3, 5 and 7 demonstrated an anti-
angiogenic effect in the CAM assay. We suggest that the
ability for compound 5 to non-selectively inhibit avb3/
avb5/a5b1 compared to the more selective inhibition of
compounds 3 and 7 towards avb3, contributes to its
increased potency in the CAM assay. These inhibitors
have been further characterized in endothelial cell in
vitro assays to determine the integrin selectivity required
for anti-angiogenic activity and the results will be pre-
sented in due course.24 Having identified potent vitro-
nectin receptor antagonists, it still remains a significant
challenge to identify those exhibiting good pharmaco-
kinetics. This series of compounds are not expected to
exhibit a good pharmacokinetics profile due to the pres-
ence of the guanidino function. However, there have
been significant advances reported by numerous groups
who have identified effective guanidine equivalents,
which in combination with simple ester prodrugs, pro-
vide compounds with acceptable oral bioavailability.25
Further work is required to address these pharmaco-
kinetic issues.
3
5
7
9
72Æ36
11Æ5.7
6.4Æ3.7
2.5Æ1.0
97
1
33
2717
640Æ261
71Æ36
95Æ47
26Æ13
864
6
500
28,260
aThe IC50 values denote the concentration required to reduce the
binding of HT29 cells expressing avb5 to vitronectin (VN) and K562
cells expressing a5b1, to fibronectin (FN) by 50%. Results are shown
as a meansÆSEM based on a minimum of at least three independent
experiments performed in triplicate. See ref 24 for full experimental
details.
Table 3. Dose-dependent inhibition of VEGF-mediated angiogenesis
in the chick CAM assay by 3, 5, and 7
[Inhibitor] (mg/mesh)
% Inhibition of capillary formation
3
5
7
9
3
17
33
2
9
65
nd
98
2
9
a
4291
57
66
95
aNot determined.
Table 1. For both series, (n=0,1), it was found that
when the guanidinylaryl moiety is positioned meta to
the thiophene, the compounds are exceptionally potent
against avb3 (3, 7 compared to 2, 4, 6, 8). In the phenyl
sulfonyl series the ortho and para compounds were very
active against aIIbb3 hence the complete lack of selec-
tivity for avb3, whereas the meta compounds were
slightly selective. Interestingly, in the mesityl series, by

506
M. Bubenik et al. / Bioorg. Med. Chem. Lett. 13 (2003) 503–506
R. M.; Perkins, J. J.; Rodan, S. B.; Wesolowski, G.; Whitman,
Acknowledgements
D. B.; Zartman, A. E.; Rodan, G. A.; Hartman, G. D. J. Med.
Chem. 2000, 43, 3736.
The authors would like to thank Luisa Iruela-Arispe
(University of California, Los Angeles, CA) for per-
forming the CAM experiments, Therese Godbout for
help in the preparation of this manuscript, and Serge
Lamothe for helpful discussions.
16. (a) Peyman, A.; Scheunemann, K.-H.; Will, D. W.;
Knolle, J.; Wehner, V.; Breipohl, G.; Stilz, H. U.; Carniato,
D.; Ruxer, J.-M.; Gourvest, J.-F.; Auberval, M.; Doucet, B.;
Baron, R.; Gaillard, M.; Gadek, T. R.; Bodary, S. Bioorg.
Med. Chem. Lett. 2001, 11, 2011. (b) Scheunemann, K.;
Knolle, J.; Peyman, A.; Will, D. W.; Carniato, D.; Gourvest,
J.; Gadek, T.; McDowell, R.; Bodary, S. C.; Cuthbertson, R.
A. WO 9959992, 1999. (c) Duggan, M. E.; Hartman, G. D.
WO 0006169, 2000. Merck USA have described a similar ser-
ies of compounds in the patent literature including one shown
below which closely resembles 7.
References and Notes
1. Gasparni, G.; Brooks, P. C.; Biganzoli, E.; Vermeulen,
P. B.; Bonoldi, E.; Dirix, L. Y.; Ranieri, G.; Miceli, R.; Cher-
esh, D. A. Clin. Cancer Res. 1998, 4, 2625.
2. Brooks, P. C.; Clark, R. A. F.; Cheresh, D. A. Science
1994, 264, 569.
3. Brooks, P. C.; Montgomery, A. M. P.; Rosenfeld, M.;
Reisfeld, R. A.; Hu, T.; Cheresh, D. A. Cell 1994, 79, 1157.
4. Brooks, P. C.; Strombland, S.; Klemke, R.; Visscher, D.;
Sarker, F. H.; Cheresh, D. A. J. Clin. Invest. 1995, 96, 1815.
5. Friedlander, M.; Brooks, P. C.; Shaffer, R. W.; Kincaid,
C. M.; Varner, J. A.; Cheresh, D. A. Science 1995, 270, 1500.
6. Erdreich-Epstein, A.; Shimada, H.; Groshen, S.; Liu, M.;
Metelitsa, L. S.; Kim, K. S.; Stins, M. F.; Seeger, R. C.; Dur-
den, D. L.; Cancer Res. 2000, 60, 712.
7. Duggan, M. E.; Hutchinson, J. H. Exp. Opin. Ther. Pat.
2000, 10, 1367.
8. Lode, H. N.; Moehler, T.; Xiang, R.; Jonczyk, A.; Gillies,
S. D.; Chersh, D. A.; Reisfeld, R. A. Proc. Natl. Acad. Sci.
U.S.A. 1999, 96, 1591.
9. Van Waes, C.; Enamorado-Ayala, I.; Hecht, D.; Sulica, L.;
Chen, Z.; Batt, D. G.; Mousa, S. A. Int. J. Oncol. 2000, 16, 1189.
10. Kumar, C. C.; Malkowski, M.; Yin, Z.; Tanghhetti, E.;
Yaremko, B.; Nechuta, T.; Varner, M. L.; Smith, E. M.; Neus-
tadt, B.; Presta, M.; Armstrong, L. Cancer Res. 2000, 61, 2232.
11. Reynolds, L. E.; Wyder, L.; Lively, J. C.; Taverna, D.;
Robinson, S. D.; Huang, X.; Sheppard, D.; Hynes, R. O.;
Hodivala-Dilke, K. Nature. Med. 2002, 8, 2 7.
12. Kim, S.; Bell, K.; Mousa, S. A.; Varner, J. A. Amer. J.
Path. 2000, 156, 1345.
13. Haubner, R.; Schmitt, W.; Holzemann, G.; Goodman,
S. L.; Jonczyk, A.; Kessler, H. J. Am. Chem. Soc. 1996, 118,
7881.
.
17. Hartman, G. D.; Prugh, J. D.; Egbertson, M. S.; Duggan,
M. E.; Hoffman, W. WO 948577.
18. Kelly, T. R.; Jagoe, C. T.; Gu, Z. Tetrahedron Lett. 1991,
32, 4263.
19. (a) Bernatowicz, M. S.; Wu, Y.; Matsueda, G. R. Tetra-
hedron Lett. 1993, 34, 3389. (b) Drake, B.; Patek, M.; Lebl, M.
Synthesis 1994, 579.
20. For a recent review, see: Miyaura, N.; Suzuki, A. Chem.
Rev. 1995, 95, 2457.
21. Fenger, I.; Le Drian, C. Tetrahedron Lett. 1998, 39, 4287.
22. Wakefield, B. J. Organolithium Methods; Academic: New
York, 1988.
23. It is interesting that the IC50 for compound 9 is
extremely low on avb3 cell adhesion when compared to the
other analogues. In addition, we suggest that the low IC50
might be due to a combination of a low affinity of the
avb3 receptor for fibrinogen combined with a high affinity
of the compound towards the receptor. Taken together,
this compound would prevent avb3 from binding to fibri-
nogen at very low concentrations.
24. Meerovitch, K.; Bergeron, F.; Grouix, B.; Poirier, C.;
Bubenik, M.; Chan, L.; Leblond, L.; Bowlin, T.; Goudreau,
H.; Attardo, G. Vasc. Pharmaco. In press.
25. Miller, W. H.; Alberts, D. P.; Bhatnagar, P. K.; Bondinell,
W. E.; Callahan, J. F.; Calvo, R. R.; Cousins, R. D.; Erhard,
K. F.; Heerding, D. A.; Keenan, R. M.; Kwon, C.; Manley,
P. J.; Newlander, K. A.; Ross, S. T.; Samanen, J. M.; Uzins-
kas, I. N.; Venslavsky, J. W.; Yuan, C. C.; Haltiwanger, R. C.;
Gowen, M.; Hwang, S. M.; James, I. E.; Lark, M. W.; Rie-
man, D. J.; Stroup, G. B. J. Med. Chem. 2000, 43, 2 2 .
14. Pitts, W. J.; Wityak, J.; Smallheer, J. M.; Tobin, A. E.;
Jetter, J. W.; Buynitsky, J. S.; Harlow, P. P.; Solomon, K. A.;
Corjay, M. H.; Mousa, S. A.; Wexler, R. R.; Jadhav, P. K. J.
Med. Chem. 2000, 43, 2 7.
15. Duggan, M. E.; Duong, L. T.; Fisher, J. E.; Hamill, T. G.;
Hoffman, W. F.; Huff, J. R.; Ihle, N. C.; Chih-Tai, L.; Nagy,