αvβ3 antagonists based on a central thiophene scaffold

Article abstract of DOI:10.1016/S0960-894X(01)00357-2

A series of novel, highly potent α_vβ₃ antagonists based on a thiophene scaffold and containing an acylguanidine as an Arg-mimetic is described. A number of structural features, such as cyclic versus open guanidine and a variety of li

Full text of DOI:10.1016/S0960-894X(01)00357-2

Bioorganic & Medicinal Chemistry Letters 11 (2001) 2011–2015
ꢀvꢁ3 Antagonists Based on a Central Thiophene Scaffold
Anusch Peyman,a,* Karl-Heinz Scheunemann,a David W. Will,a Jochen Knolle,a
Volkmar Wehner,a Gerhard Breipohl,a Hans Ulrich Stilz,a Denis Carniato,b
Jean-Marie Ruxer,b Jean-Francois Gourvest,b Marielle Auberval,b Bernard Doucet,b
Roland Baron,b Martine Gaillard,b Thomas R. Gadekc and Sarah Bodaryc
aAventis Pharma Deutschland GmbH, D-65926 Frankfurt, Germany
b
´
Aventis Pharma S.A,. 102 Route de Noisy, 93235 Romainville Ce dex, France
cGenentech, 1 DNA Way, South San Francisco, CA 94080, USA
Received 21 March 2001; revised 3 May 2001; accepted 17 May 2001
Abstract—A series of novel, highly potent avb3 antagonists based on a thiophene scaffold and containing an acylguanidine as an
Arg-mimetic is described. A number of structural features, such as cyclic versus open guanidine and a variety of lipophilic side
chains, carbamates, sulfonamides and b-amino acids were explored with respect to inhibition of avb3 mediated cell adhesion and
selectivity versus aIIbb3 binding. In addition, compound 19 was found to be active in the TPTX model of osteoporosis. # 2001
Published by Elsevier Science Ltd.
Chemistry
The vitronectin receptor, also known as avb3, is a
member of the integrin family of cell surface receptors
which play a major role in cell adhesion and signal
transduction processes.1 Integrins are a/b heterodimers
composed of at least 15 a subunits and eight b subunits.
The b3 class of the integrin family, aIIbb3 (GPIIb/IIIa or
fibrinogen receptor) and avb3 (vitronectin receptor), has
received special attention in recent drug discovery
efforts.2 aIIbb3 is prevalent on platelets and plays a role
in thromboembolic disorders while avb3 has been
implicated in tumor progression, angiogenesis and rest-
enosis.2 In addition, avb3 is the dominant receptor for
mediating the attachment of osteoclasts to bone during
bone resorption3 which makes it an attractive target for
the treatment of osteoporosis. avb3 binds to extracellular
matrix adhesive proteins containing an RGD sequence
such as vitronectin, fibrinogen, bone sialo protein II, von
Willebrand factor, and osteopontin. Therefore, we4,5 and
others3,6 have investigated peptidomimetic approaches
to identify avb3 antagonists. In this communication we
describe a new series of highly potent and selective
RGD peptidomimetic avb3 antagonists which are based
on a central thiophene scaffold and which contain acyl-
guanidines as an Arg-replacement.
The synthesis of the avb3 antagonist 1 is outlined in
Scheme 1. 5-Bromothiophene-2-carboxylic acid is reac-
ted with acrylic acid methyl ester catalyzed by
PdCl2(PPh2Ferrocene)2/NEt3 in acetonitrile to give com-
pound 2 (82%) which is subsequently reduced by cataly-
tic hydrogenation to give 5-(2-methoxycarbonylethyl)-
thiophene-2-carboxylic acid 3. Compound 3 is coupled
with (S)-Cbz-diaminopropionic acid tert-butylester (Z-
DAP-OtBu) using TOTU in DMF to give 4 in 95%
yield. The acylguanidino compound 5 is formed by
reaction of the ester with guanidine in THF (35%),
which after deprotection with TFA yields antagonist 1.
Scheme 2 shows an alternative synthesis of the central
building block: 2-thiophencarbaldehyde is reacted with
N,N 0-dimethylethylenediamine to give 1,3-dimethyl-2-
(2-thienyl)-imidazolidine 6 which is lithiated and reacted
with Boc-anhydride to give compound 7 in 71% yield.
Compound 7 is converted to 5-formyl-thiophene-2-car-
boxylic acid tert-butyl ester 8 by treatment with methyl
iodide and subsequently condensed with ethyl (diethyl
phosphono)acetate to give compound 9 (53%). Cataly-
tic hydrogenation of 9 followed by cleavage of the tert-
butylester yields 10 which is again condensed with Z-DAP-
OtB u to11. The cyclic acylguanidino group in 12 is formed
by reaction of 11 with 2-amino-3,4,5,6-tetra-
*Corresponding author. Fax: +49-69-305-80261; e-mail: anusch.
peyman@aventis.com
0960-894X/01/$ - see front matter # 2001 Published by Elsevier Science Ltd.
PII: S0960-894X(01)00357-2

2012
A.Peyman et al./ Bioorg.Med.Chem.Lett.11 (2001) 2011–2015
Scheme 1. Reagents and conditions: (a) PdCl2(PPh2ferrocene)2, acrylic acid methyl ester, NEt3, MeCN/85 ꢀC, 20 h; (b) H2, Pd/C, MeOH/rt, 7 h; (c)
Z-DAP-OtBu, TOTU, DIEA, DMF/rt, 6 h; (d) guanidine, THF/reflux, 2.5 h; (e) TFA, DCM/rt, 3.5 h.
Scheme 2. Reagents and conditions: (a) MeNH(CH2)2NHMe, toluene/reflux(–H2O), 4 h; (b) nBuLi, TEMED, Boc2O, THF/ꢁ70 ꢀC for 4 h, rt for
16 h; (c) MeI, Et2O/rt, 16 h; (d) P(O)(EtO)2CH2COOEt, NaH, THF/60 ꢀC, 3 h; (e) (i) H2 (4 bar), Pd/C, EtOH/50 ꢀC, 7 h; (ii) TFA, DCM/rt, 5 h; (f)
Z-DAP-OtBu, TOTU, DIEA, DMF/rt, 8 h; (g) 2-amino-3,4,5,6-tetrahydropyrimidine, DMF/rt, 8 h; (h) TFA, DCM/rt, 4 h.
hydropyrimidine in DMF (45%) and final cleavage of
the tert-butylester with TFA yields avb3 antagonist 13.
after conversion to the acylguanidine (71%) and depro-
tection, gives the avb3 antagonist 19. Alternatively,
compounds containing this side chain can be obtained
by the use of 2-(1-adamantylmethyloxycarbonylamino)-
3-aminopropionic acid instead of Z-DAP-OtBu.
Antagonists containing a sulfonamide side chain are
obtained from the same intermediate 17 by reaction
with the corresponding sulfonylchlorides, followed by
conversion to the acylguanidine and deprotection.
Antagonists containing a b-amino acid component are
obtained from intermediate 3 by condensation with the
corresponding b-amino acid, followed again by conver-
sion to the acylguanidine and deprotection.
Scheme 3 illustrates the synthesis of compounds con-
taining a cyclopropyl group as spacer between the cen-
tral thiophene scaffold and the acylguanidino group.
Compound 2 is coupled with Z-DAP-OtBu to give 14.
The cyclopropyl group in 15 is formed by reaction with
diazomethane under Pd(OAc)2 catalysis and avb3
antagonist 16 is obtained as in Scheme 2 by reaction of
15 with 2-amino-3,4,5,6-tetrahydropyrimidine in DMF
and TFA cleavage of the tert-butylester. Other com-
pounds containing the cyclopropyl spacer were obtained
in an analogous manner. The preparation of com-
pounds containing the adamantylmethyl carbamate side
chain is outlined in Scheme 4. The benzyloxycarbamate
group in compound 4 (Scheme 1) is removed by cataly-
tic hydrogenation to give a free amino group in 17
which is reacted with N-(1-adamantylmethoxy-
carbonyloxy)-succinimide to give 18 (100%) which,
Results and Discussion
One of the most interesting challenges in this area is the
design of selective antagonists since both avb3 and
aIIbb3 bind to the same RGD recognition motif, and
Scheme 3. Reagents and conditions: (a) Z-DAP-OtBu, TOTU, DIEA, DMF/rt, 8 h; (b) Et2O/THF, 1:1, CH2N2, Pd(OAc)2 0 ꢀC, 2 h; (c) (i) 2-amino-
3,4,5,6-tetrahydropyrimidine, DMF/rt, 6 h; (ii) TFA, DCM/rt, 4 h.
Scheme 4. Reagents and conditions: (a) H2, Pd/C, 5 N HCl in MeOH/rt, 14 h; (b) N-(1-adamantylmethoxycarbonyloxy)-succinimide, NaHCO3,
H2O, THF/rt, 4 h; (c) 2-amino-3,4,5,6-tetrahydropyrimidine, DMF/rt, 5 h; (d) TFA, DCM/rt, 3 h.

A.Peyman et al./ Bioorg.Med.Chem.Lett.11 (2001) 2011–2015
2013
subtle differences in the ligand structure may determine
selectivity. The IC50 values for the inhibition of binding
of fibrinogen to aIIbb3 and of Kistrin7 (a disintegrin with
high affinity to avb3) to avb3 are summarized in Tables
1–3. In addition, a more stringent whole-cell assay was
used to differentiate between avb3 antagonists: human
293 cells were engineered to overexpress the avb3 recep-
tor (>106 copies/cell) and were used to assess the
potency of avb3 antagonists by the inhibition of their
binding to immobilized vitronectin.
acylguanidino group into an unfavorable position, since
the conjugation ranges from the thiophene ring through
the double bond through the carbonyl group and even
the guanidino group, which all end up in a single plane.
On the other hand, the ethylene spacer and the cyclo-
propyl group result in similar potencies as can be seen
Table 2. Thiophene-based avb3 antagonists with sulfonamide side
chains
Guanidine mimetic
The acylguanidines used in this study produce avb3
antagonists with excellent potency. The comparison of 1
and 13 (Table 1) shows that cyclic acylguanidines lead
to a further enhancement of potency, which becomes
even more evident in the 293-cell attachment assay. At
the same time the selectivity towards the fibrinogen
receptor aIIbb3 (GPIIb/IIIa) is increased. The same
ranking is observed for the b-amino acid containing
antagonists 25 and 26 (Table 3) and a similar SAR has
been observed earlier for normal guanidines.4,5
R
K/avb3
IC50 (nM)
Vn/293
IC50 (nM)
Fg/aIIbb3
IC50 (nM)
33
3
47
235
34
35
36
37
38
39
40
41
42
43
44
45
46
47
2
2
2
3
2
3
2
4
5
2
2
4
5
4
13
6
80
17
Spacer
The spacer separating the central thiophene scaffold and
the acylguanidine moiety is of great importance. Clearly
the vinyl spacer is less favorable as can be seen from a
comparison of 19, 20, and 21. Obviously it forces the
12
26
7
55
9
Table 1. Thiophene-based avb3 antagonists with lipophilic carbamate
side chains
35
12
10
90
313
80
91
167
69
22
65
11
R
X
K/avb3
IC50 (nM)
Vn/293
IC50 (nM)
Fg/aIIbb3
IC50 (nM)
925
500
40
1
39
7
837
63
2900
4500
13
16
19
20
21
60
2
41
26
2450
480
17
255
1350
105
1100
10
>10,000
69
1000
2800
The IC50 values denote the concentration required to reduce binding
of fibrinogen (Fg) to aIIbb3 and Kistrin to avb3 by 50% (K/avb3) or
293 cell attachment to vitronectin by 50% (Vn/293)

2014
A.Peyman et al./ Bioorg.Med.Chem.Lett.11 (2001) 2011–2015
from various examples 19–21 or 13–16. Considering
that the cyclopropyl-containing antagonists consist of
diastereomeric mixtures, one may speculate that one of
the pure forms could show some advantages over the
open chain ethylene spacer.
porated in the ‘C-terminus’ of RGD peptidomimetics.
On the other hand, this series shows excellent selectiv-
ities versus aIIbb3. Bulky aromatic side chains are pre-
ferred; see, for example, the increase in potency from 22
to 25. However, ortho substitution of the aromatic side
chain, as in 32 containing the 1-naphthyl group, does
not seem to enhance potency. The SAR for the spacer
and the guanidino group are the same as in the other
series. Two different side chains, benzyl and 1-ada-
mantylmethyl, were incorporated in the carbamate ser-
ies. The bulky adamantyl moiety offers a slight
advantage over the flat benzyl group, enhancing
potency by a factor of 2 (cf. 13 and 19) when used in
combination with the ethylene spacer. The opposite is
the case when the cyclopropyl spacer is used. The selec-
tivity versus aIIbb3 is also excellent in this series and
ranges from 40 to 650, with the exception of compound
20 which contains the vinyl-spacer. An interesting SAR
is observed in the sulfonamide side chain series (Table
2). No major differences in potency are observed in the
K/avb3 binding assay, however, when the 293-cell
attachment assay is used, the SAR becomes more evi-
dent. Clearly, aromatic side chains are preferred over
alkyl chains. IC50 values for antagonists containing
alkyl side chains are in the 100-nM range (41–46)
with the exception of the n-propyl group, which has
an IC50 value of 22 nM. Antagonists with aromatic
sulfonamide side chains (34–39) are extremely potent
with IC50 values in the low nanomolar range. How-
ever, the affinity for aIIbb3 is also increased. Com-
pound 37, which contains the biphenyl sulfonamide
side chain, shows an IC50 of 3 nM for K/avb3 binding
and 9 nM for Fg/aIIbb3 binding and is a representative
for a dual inhibitor.
Lipophilic side chain
The influence of the lipophilic side chain was investi-
gated in three different series: the carbamate side chains
(Table 1); the sulfonamide side chains (Table 2); and the
b-amino acids (Table 3). Of those series, the b-amino
acid containing avb3 antagonists are less active in the
293-cell attachment assay, despite good activity in the
standard binding assay. We have also observed this dif-
ference in other series when b-amino acids were incor-
Table 3. Thiophene-based avb3 antagonists with b-amino acid side
chains. All compounds are 1:1 racemic mixtures with the exception of
24, which is the pure S-enantiomer
R
X
K/avb3
IC50(nM)
Vn/293
IC50(nM)
Fg/aIIbb3
IC50(nM)
22
23
175
35
—
>10000
>10000
3610
avb3 antagonist 19 was chosen to evaluate the potency
of this series in bone resorption in the in vivo thyr-
oidectomized-parathyroidectomized (TPTX) rat model,
which measures the ability of a compound to inhibit the
parathyroid hormone stimulated calcemic response in
hypocalcemic TPTX rats.8 In this model, upon sub-
cutaneous administration, 19 showed a dose dependent
effect, the inhibition was ꢁ6% (0.1 mg/kgꢂ2), ꢁ26%
(0.3 mg/kgꢂ2) and ꢁ56% (1.0 mg/kgꢂ2).
24
25
26
27
28
29
30
31
32
9
10
33
32
8
1430
1730
>10000
>10000
10000
>10000
>5000
41000
>5000
15000
10500
20900
>10000
>10000
>10000
>10000
>10000
>100000
In conclusion, we have discovered a new class of potent
and selective avb3 antagonists based on a central thio-
phene scaffold and the acylguanidine Arg-mimetic. The
compounds are capable of inhibiting bone resorption in
vivo and therefore are promising drug substances for
the treatment of osteoporosis.
16
120
69
175
References and Notes
1. Hynes, R. O. Cell 1992, 69, 11.
2. Miller, W. H.; Keenan, R. M.; Willette, R. N.; Lark, M. W.
Drug Disc.Today 2000, 5, 397.
3. Hartman, G. D.; Duggan, M. E. Expert Opin.Invest.Drugs
2000, 9, 1281.
4. Peyman, A.; Wehner, V.; Knolle, J.; Stilz, H. U.; Breipohl,
G.; Scheunemann, K. H.; Carniato, D.; Ruxer, J. M.; Gourvest,

A.Peyman et al./ Bioorg.Med.Chem.Lett.11 (2001) 2011–2015
2015
J. F.; Gadek, T. R.; Bodary, S. Bioorg.Med.Chem.Lett. 2000,
10, 179.
7. Green, L. J.; Brierley, V. H.; Humphries, M. J.Biochem.
Soc.Trans. 1995, 23, 505 S.
5. Peyman, A.; Gourvest, J.-F.; Gadek, T. R.; Knolle, J.
Angew.Chem,. Int.Ed. 2000, 39, 2874.
6. Excellently reviewed by: Duggan, M. E.; Hutchinson, J. H.
Expert Opin.Ther.Pat. 2000, 10, 1367.
8. Fisher, J. E.; Caulfield, M. P.; Sato, M.; Quartuccio, H. A.;
Gould, R. J.; Garsky, V. M.; Rodan, G. A.; Rosenblatt, M.
Endocrinology 1993, 132, 1411.