Nonpeptide αvβ3 antagonists. Part 2: Constrained glycyl amides derived from the RGD tripeptide

Article abstract of DOI:10.1016/S0960-894X(01)00687-4

Mimetics of the RGD tripeptide are described that are potent, selective antagonists of the integrin receptor, α_vβ₃. The use of the 5,6,7,8-tetrahydro[1,8]naphthyridine group as a potency-enhancing N-terminus is demonstrated. Two 3-substituted-3-amino-propionic acids previously contained in α_IIbβ₃ antagonists were utilized to enhance binding affinity and functional activity for the targeted receptor. Further affinity increases were then achieved through the use of cyclic glycyl amide bond constraints.

Full text of DOI:10.1016/S0960-894X(01)00687-4

Bioorganic & Medicinal Chemistry Letters 12 (2002) 25–29
Nonpeptide ꢀvꢁ3 Antagonists. Part 2: Constrained Glycyl Amides
Derived from the RGD Tripeptide
Robert S. Meissner,a,* James J. Perkins,a Le T. Duong,b George D. Hartman,a
William F. Hoffman,a Joel R. Huff,a Nathan C. Ihle,a,y Chih-Tai Leu,b Rose M. Nagy,b
Adel Naylor-Olsen,c Gideon A. Rodan,b Sevgi B. Rodan,b David B. Whitman,a
Gregg A. Wesolowskib and Mark E. Duggana
aDepartment of Medicinal Chemistry, Merck Research Laboratories, West Point, PA 19486, USA
bDepartment of Bone Biology and Osteoporosis Research, Merck Research Laboratories, West Point, PA 19486, USA
cDepartment of Molecular Design and Diversity, Merck Research Laboratories, West Point, PA 19486, USA
Received 25 June 2001; accepted 12 October 2001
Abstract—Mimetics of the RGD tripeptide are described that are potent, selective antagonists of the integrin receptor, avb3. Th e
use of the 5,6,7,8-tetrahydro[1,8]naphthyridine group as a potency-enhancing N-terminus is demonstrated. Two 3-substituted-3-
amino-propionic acids previously contained in aIIbb3 antagonists were utilized to enhance binding affinity and functional activity
for the targeted receptor. Further affinity increases were then achieved through the use of cyclic glycyl amide bond constraints.
# 2001 Elsevier Science Ltd. All rights reserved.
The vitronectin receptor avb3 is a member of the integ-
rin superfamily of receptors that is highly expressed in
osteoclast cells.1,2 These cells are responsible for
resorption of bone in the remodeling cycle. Patients
withosteoporosis ahve decreased bone mass and
increased bone fragility due to an imbalance between
osteoclast-mediated bone resorption and bone forma-
tion.3 Proteins and peptides that contain the arginine-
glycine-aspartic acid (RGD) tripeptide, including vitro-
Herein, we describe the identification of a class of avb3
antagonists derived directly from the tripeptide RGD.
We have explored the use of the THN group as an N-
terminus in this series and have examined the effect of
substitution at the 3-position of the propionic acid on in
vitro potency. In addition, structure–activity relation-
ship (SAR) investigations of constraints about the gly-
cyl amide bond of these tripeptide mimetics are
described.
nectin, osteopontin, bone sialoprotein, and echistatin
4,5
bind to avb3 withhighaffinity.
Antibodies to avb3,
The synthesis of three peptidomimetics is described in
Scheme 1. The ketone 2 was subjected to a Friedlander
echistatin, and small-molecule avb3 antagonists have
been shown to inhibit bone resorption in vitro and pre-
vent bone loss in vivo.6ꢀ14 A recent report from these
laboratories described the design of a series of high-
affinity avb3 antagonists from members of the ‘centrally
constrained’ sulfonamide class of aIIbb3 fibrinogen
antagonists.15 Central to this communication was the
16
condensation with2-amino-3-formylpyridine
(1) to
produce the naphthyridine 3. Regioselective hydro-
genation of the distal pyridine ring followed by ester
hydrolysis gave the tetrahydronaphthyridine acid 4.
Diimide coupling of acid 4 withglycine ethyl ester, and
subsequent hydrolysis gave the acid 5. Finally, coupling
of 5 to three 3-amino-propionate esters bearing different
3-substituents, followed by hydrolysis provided the tar-
get compounds 6, 7, and 8.
finding that
a
5,6,7,8-tetrahydro[1,8]naphthyridine
(THN) moiety could be utilized as a novel N-terminus,
imparting potency, integrin selectivity, and increased
lipophilicity to this structural class.
Preparation of the pyrrolidinone-constrained antagonist
15 (Scheme 2) began with the conversion of 2-oxo-hex-
anoic (9) acid to its (R)-benzyl-oxazolidinone, and
subsequent ketalization to give 10. Stereoselective allyl-
ation17 followed by ozonolysis produced the aldehyde
*Corresponding author. Tel.: +1-215-652-4407; fax: +1-215-652-7310;
e-mail: robert_meissner@merck.com
yCurrent address: Celltech Chiroscience Inc., Bothell, WA 98021,
USA.
0960-894X/02/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00687-4

26
R. S. Meissner et al. / Bioorg. Med. Chem. Lett. 12 (2002) 25–29
11. A facile one-pot reductive amination/cyclization
withglycine ethyl ester, followed by ketal unmasking
provided the pyrrolidinone-ketone 12. A five-step
sequence utilizing transformations previously described
in Scheme 1 then produced the desired antagonist 15.
The corresponding 3-pyridyl analogue 30 was prepared
from 14 through the same sequence, utilizing the corre-
sponding 3(S)-pyridin-3-yl b-alanine ester. The diaster-
eoisomer 31 was produced withteh same protocol,
using the (S)-antipode of the benzyl-oxazolidinone
chiral auxiliary. The synthesis of a piperidinone variant
21 (Scheme 3) started with the alkylation of the enolate
of N-(trimethysilyl)piperidin-2-one (17) withthe iodide
16.18 Ketal removal then revealed ketone 18.
Friedlander condensation of 18 with2-amino-3-formyl-
pyridine was followed by N-alkylation of the piperidi-
none with ethyl iodoacetate, and then hydrogenation
produced the THN-ester 20. The standard three-step
sequence then yielded the piperidinone analogue 21 as a
mixture of diastereomers.
Preparation of an imidazolidone analogue began with
the conversion of parent heterocycle to its bis-BOC
carbamate 23 (Scheme 4). Mono-deprotection was then
carried out by treatment withmagnesium perchlorate in
refluxing acetonitrile, a novel variation of the Stafford
deprotection.19 Alkylation of the free NH group with
iodide 16 provided the urea 24. BOC solvolysis and
N-alkylation afforded the ketal-ester 25. The targeted
product 27 was then attained from 25 in six steps.
The compounds displayed in Table 1 were evaluated for
their ability to displace a non-peptide radioligand from
human recombinant avb3 (SPAV320), and to inhibit the
rate of ADP-stimulated aggregation of gel-filtered
human platelets mediated through the integrin aIIbb3
(PLAGGIN21). The RGD tripeptide 2822 and the ana-
logue 29,23 which lacks the aspartic acid a-carboxylate
and the arginine a-amino group, have IC50 values >10
mM in bothin vitro assays. Measurable binding affinity
to avb3 was attained when the guanidinomethylene
N-terminal portion of 29 was replaced with5,6,7,8-tet-
rahydro[1,8]naphthyridine (THN), producing com-
pound 6. This analogue had an IC50 of 111 nM in the
SPAV3 assay while remaining inactive in the PLAG-
GIN assay at concentrations up to 10 mM.
Scheme 1. (a) Proline, ethanol, reflux; (b) H2, Pd/C; (c) 6 N HCl; (d)
glycine ethyl ester hydrochloride, EDC, HOBT; (e) ethyl 3-aminopro-
pionate, EDC, HOBT; (f) ethyl-3(S)-3-aminopent-4-ynoate; (g) ethyl-3
(S)-3-amino-3-pyridin-3-ylpropanoate, EDC, HOBT.
Scheme 2. Xc=N-(R)-(+)-4-benzyl-2-oxazolidinone (a) ethylene gly-
col, cat. TsOH, toluene, reflux; (b) pivaloyl chloride, triethylamine,
THF, then lithium (R)-(+)-4-benzyl-2-oxazolidinone; (c) NaHMDS,
allyl bromide; (d) ozone, then PPh3; (e) glycine ethyl ester, sodium
triacetoxyborohydride, triethylamine; (f) acetone, cat. TsOH, reflux;
(g) 2-amino-3-formylpyridine, proline, ethanol, reflux; (h) H2, Pd/C,
ethanol; (i) 6 N HCl; (j) ethyl-3(S)-3-aminopent-4-ynoate, EDC, HOBT.
Scheme 3. (a) LDA; (b) acetone, cat. TsOH, reflux; (c) 2-amino-3-
formylpyridine, proline, ethanol, reflux; (d) NaHMDS, ethyl iodoace-
tate; (e) H2, Pd/C, ethanol; (f) 6 N HCl; (g) ethyl-3(S)-3-amino-3-pyr-
idin-3-ylpropanoate, BOP, NMM.

R. S. Meissner et al. / Bioorg. Med. Chem. Lett. 12 (2002) 25–29
27
Binding affinities of this class for avb3 could be further
improved by incorporating substituents at the 3-posi-
tion of the propionic acid that have been utilized in
potent fibrinogen receptor antagonists.24 The 3(S)-acety-
lene analogue 7 and 3(S)-3-pyridyl analogue 8 had IC50
values in the SPAV3 assay of 8.8 nM and 1.3 nM,
respectively. These two compounds also inhibited the
formation of mouse osteoclasts in vitro (OCFORM
reogenic center of 30 by conversion to the imidazoli-
done 27 resulted in a substantial loss of binding affinity
and cell-based potency (Table 2).
Modeling of the 3-pyridyl propionic acid analogues was
carried out to compare the tertiary structural differences
to the avb3 affinities.26 The amide 8 and (S)-pyrrolidi-
none 30 have low energy minima that share a high
degree of atom overlap throughout the chain. The
increased restriction of 30 to these conformations may
explain the increased affinity over 8, assuming that these
minima are similar to the integrin-bound state. For the
(R)-pyrrolidinone 31 and piperidinone 21, low energy
assay25) withIC
values of ꢁ100–200 nM. Although
50
bothcompounds demonstrated functional activity
mediated through avb3, they were poor inhibitors of
platelet aggregation. The aIIbb3 antagonists from which
these substituted propionic acids were drawn bore N-
termini significantly different from the THN, such as
guanidines, amidines, and piperidines. The finding that
7 and 8 are potent antagonists of avb3, yet lack activity
in the PLAGGIN assay, suggests that the N-terminal
moiety can play a critical role in modulating integrin
selectivity among compounds that bear identical C-ter-
mini.
Table 1. Glycine-containing antagonists
Having identified affinity-enhancing substituents at the
N- and C-termini, our efforts focused on exploring the
effects of constraints at the central region of the mole-
cule. Constraining the glycyl amide of 7 through cycli-
zation into the (S)-pyrrolidinone 15 yielded a 3-fold
affinity enhancement in the SPAV3 assay. A similar
increase was gained withthe pyridine-containing ( S)-
pyrrolidinone analogue 30, resulting in a highly potent
IC50 (nM)
Compd
X
Y
SPAV3
PLAGGIN
28
29
6
7
8
NH2
H
—
—
—
COOH
H
H
CCH
3-Pyridyl
>10,000
>10,000
111
>10,000
>10,000
>10,000
>10,000
>10,000
avb3 antagonist witha SPAV3 IC of 0.35 nM, and an
50
OCFORM IC50 of 62 nM. Selectivity over aIIbb3 was
maintained, as the PLAGGIN IC50 remained greater
than 10 mM. Interestingly, the (R)-pyrrolidinone anti-
pode 31 was only about 3-fold less potent in the SPAV3
assay than 30, making it essentially equipotent withthe
amide parent 8. Ring expansion to produce the piper-
idinone 21 offered no advantage over the amide 8.
Finally, an attempt to eliminate the pyrrolidinone ste-
8.8
1.3
Table 2. Glycyl-constrained antagonists
IC50 (nM)
SPAV3 OCFORM PLAGGIN
Compd
Z
Y
15
CCH
2.9
150
>10,000
30
31
3-Pyridyl
3-Pyridyl
0.35
0.94
62
79
>10,000
>10,000
21
27
3-Pyridyl
1.2
480
>10,000
Scheme 4. (a) BOC2O, DMAP; (b) Mg(ClO4)2, reflux; (c) LiHMDS,
16; (d) TFA, toluene; (e) LiHMDS, ethyl iodoacetate; (f) acetone, cat.
TsOH, reflux; (g) 2-amino-3-formylpyridine, proline, ethanol, reflux;
(h) H2, Pd/C, ethanol; (i) 6 N HCl; (j) ethyl-3(S)-3-amino-3-pyridin-3-
ylpropanoate, EDC, HOBT; (k) NaOH, ethanol.
3-Pyridyl 12
>1000
7100

28
R. S. Meissner et al. / Bioorg. Med. Chem. Lett. 12 (2002) 25–29
minima exist which place the THN, 3-pyridyl, and car-
boxylate groups in positions similar to those for the
minima of 8 and 30. Conversely, this premise does not
hold for compound 27. Models of this imidazolidone
showed preferred confirmations that present the
charged termini similarly to those for the above set, yet
the affinity of 27 for avb3 was significantly reduced. Of
interest is the observation that the imidazolidone 27
proved the most potent of the series in the PLAGGIN
assay, indicating an erosion of integrin selectivity. The
coplanar presentation of the two alkyl groups about the
imidazolidone ring is unique to this compound and may
be the source of its functional divergence.
10. Engleman, V. W.; Nickols, G. A.; Ross, F. P.; Horton,
M. A.; Griggs, D. W.; Settle, S. L.; Ruminski, P. G.; Tei-
telbaum, S. L. J. Clin. Invest. 1997, 99, 2284.
11. Lark, M. W.; Stroup, G. B.; Cousins, R. D.; James,
I. E.; Hwang, S. M.; Bradbeer, J.; Hoffman, S.; Lechowska,
B.; Vasko-Moser, J.; Zembryki, D.; Votta, B.; Lee-Rykac-
zewski, L.; Perng, D.; Salyers, K.; Smith, B.; Liang, X.;
Dodds, R.; Newlander, K. A.; Ross, S. T.; Erhard, K.;
Huffman, W. F.; Drake, F. H.; Miller, W. H.; Gowen, M.
Bone 1998, 22, 28S.
12. Lark, M. W.; Stroup, G. B.; Hwang, S. M.; James, I. E.;
Rieman, D. J.; Drake, F. H.; Bradbeer, J. N.; Mathur, A.;
Erhard, K. F.; Newlander, K. A.; Ross, S. T.; Salyers, K. L.;
Smith, B. R.; Miller, W. H.; Huffman, W. F.; Gowen, M. J.
Pharm. Exp. 1999, 91, 612.
13. Miller, W. H.; Alberts, D. P.; Bhatnagar, P. K.; Bondinell,
W. E.; Callahan, J. F.; Calvo, R. R.; Cousins, R. D.; Erhart,
K. F.; Heerding, D. A.; Keenan, R. M.; Kwon, C.; Manley,
P. J.; Newlander, K. A.; Ross, S. T.; Samanem, J. M.; Uzins-
kas, I. N.; Venslavsky, J. W.; Yuan, C. C.-T.; Haltiwanger,
R. C.; Gowan, M.; Hwang, S.-M.; James, I. E.; Lark, M. W.;
Rieman, D. J.; Stroup, G. B.; Azzarano, L. M.; Salyers, K. L.;
Smith, B. R.; Ward, K. W.; Johanson, K. O.; Huffman, W. F.
J. Med. Chem. 2000, 43, 22.
14. Lark, M. W.; Stroup, G. B.; Dodds, R. A.; Kapadia, R. J.;
Hoffman, S.; Hwang, S. M.; James, I. E.; Lechowska, B.;
Liang, X.; Rieman, D. J.; Salyers, K. L.; Ward, K.; Smith,
B. R.; Miller, W. H.; Huffman, W. F.; Gowen, M. J. Bone
Miner. Res. 2001, 16, 319.
With this series of compounds, we have demonstrated
the design of RGD tripeptide mimetics that are highly
potent antagonists of the avb3 receptor but poor inhibi-
tors of aIIbb3-mediated platelet aggregation in vitro. We
have extended the use of the 5,6,7,8-tetrahydro[1,8]-
naphthyridine group to an alternative structural class
and confirmed its ability to provide potency enhance-
ment and integrin selectivity for avb3. Similarly, C-
terminal 3-substituents of known aIIbb3 antagonists
were utilized to enhance binding affinity and functional
activity for the targeted receptor. Substantial potency
increases were then added through the identification of
optimal constraining elements of the glycyl amide bond.
Additional structural classes that build on these findings
will be the subject of future reports.
15. 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. R.; Hartman, G. D. J. Med.
Chem. 2000, 43, 3736.
16. Turner, J. A. J. Org. Chem. 1983, 48, 3401.
Acknowledgements
17. The diastereoselectivity of this allylation was >95:5.
18. Stowell, J. C.; King, B. T.; Hauck, H. F., Jr. J. Org. Chem.
1983, 48, 5381.
19. Stafford, J. A.; Brackeen, M. F.; Karanewsky, D. S.; Val-
vano, N. L. Tetrahedron Lett. 1993, 34, 7873.
The authors thank Robert Lynch for PLAGGIN IC50
determinations.
20. Scintillation proximity bead-based binding assay utilizing
displacement of 2(S)-(4-125Iodo-benzenesulfonylamino)-3-{4-
[2-(5,6,7,8-tetrahydro[1,8]naphthyridin-2-yl)-ethyl]-benzoyl-
amino}-propionic acid from purified, recombinant human
avb3 receptor. For a detailed assay protocol, see ref 15.
21. The PLAGGIN assay measures the aggregation of human
gel-filtered platelets activated withADP, in the presence of
fibrinogen and the antagonist. The value is reported as a per-
cent of the rate of aggregation without added antagonist. For
a detailed description of the PLAGGIN assay, see ref 15.
22. Obtained from Bachem California Inc., Torrance, CA,
USA.
23. (a) Greenspoon, N.; Hershkoviz, R.; Alon, R.; Varon, D.;
Shenkman, B.; Marx, G.; Federman, S.; Kapustina, G.; Lider,
O. Biochemistry 1993, 32, 1001. (b) Lider, O.; Greenspoon, N.;
Hershkoviz, R.; Ronen, A. PCT International Application
WO 9309795, 1993.
24. (a) Zablocki, J. A.; Rico, J. G.; Garland, R. B.; Rogers,
T. E.; Williams, K.; Schretzman, L. A.; Rao, S. A.; Bovy,
P. R.; Tjoeng, F. S. J. Med. Chem. 1995, 38, 2378. (b) Hutch-
inson, J. H.; Cook, J. J.; Brashear, K. M.; Breslin, M. J.;
Glass, J. D.; Gould, R. J.; Halczenko, W.; Holahan, M. A.;
Lynch, R. J. J. Med. Chem. 1996, 39, 4583. (c) Askew, B. C.;
McIntyre, C. J.; Hunt, C. A.; Claremon, D. A.; Baldwin, J. J.;
Anderson, P. S.; Gould, R. J.; Lynch, R. J.; Chang, C. C.-T.;
Cook, J. J.; Lynch, J. J.; Holahan, M. A.; Sitko, G. R.; Stra-
nieri, M. T. Bioorg. Med. Chem. Lett. 1997, 7, 1531.
25. (a) The OCFORM assay measures the ability of com-
pounds to inhibit the transformation of osteoblast-like cells
References and Notes
1. Rodan, S. B.; Rodan, G. A. J. Endocrinol. 1997, 154, S47.
2. Clover, J.; Dodds, R. A.; Gowen, M. J. Cell Sci. 1992, 103,
267.
3. Parfitt, A. M. J. Cell. Biol. 1994, 55, 273.
4. (a) Pytela, R.; Pierschbacher, M. D.; Ginsberg, M. H.;
Plow, E. F.; Ruoslahti, E. Science 1986, 231, 1559. (b)
Miyauchi, A.; Alvarez, J.; Greenfield, E. M.; Teti, A.; Grano,
M.; Colucci, S.; Zambonin-Zallowe, A.; Ross, F. P.; Tei-
telbaum, S. L.; Cheresh, D.; Hruska, K. J. Biol. Chem. 1991,
266, 20369.
5. Ross, F. P.; Chappel, J.; Alvarez, J. I.; Sander, D.; Butler,
W. T.; Farach-Carson, M. C.; Mintz, K. A.; Robey, P. G.;
Teitelbaum, S. L.; Cheresh, D. A. J. Biol. Chem. 1993, 268,
9901.
6. 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.
7. Yamamoto, M.; Fisher, J. E.; Gentile, M.; Seedor, J. G.;
Leu, C.-T.; Rodan, S. B.; Rodan, G. A. Endocrinology 1998,
139, 1411.
8. Chambers, T. J.; Fuller, K.; Darby, J. A.; Pringle, J. A.;
Horton, M. A. In vitro Bone Miner. 1986, 1, 127.
9. Crippes, B. A.; Engleman, V. W.; Settle, S. L.; Delarco, J.;
Ornberg, R. L.; Helfrich, M. H.; Horton, M. A.; Nickols,
G. A. Endocrinology 1996, 137, 918.

R. S. Meissner et al. / Bioorg. Med. Chem. Lett. 12 (2002) 25–29
29
(1.8 cells), originally derived from mouse calvaria, into osteo-
clasts. For a detailed assay protocol, see Duggan, M. E. U.S.
Patent 5,981,546, 1999. For the assay development, see Weso-
lowski, G.; Duong, L. T.; Lakkakorpi, P. T.; Nagy, R. M.;
Tezuka, K.; Tanaka, H.; Rodan, R. A.; Rodan, S. B. Exp. Cell
Res. 1995, 219, 679. (b) Nakamura, I.; Tanaka, H.; Rodan,
G. A.; Duong, L. T. Endocrinology 1998, 139, 5182.
26. For eachmolecule, 500 conformations were generated
using a distance geometry method (S. Kearsley, Merck & Co.,
unpublished). The conformations were energy minimized
using the MMFF force field as implemented in Batchmin
(Mohamadi, F.; Richard, N. G. J.; Guida, W. C.; Liskamp,
R.; Caufield, C.; Chang, G.; Hendrickson, T.; Still, W. C. J.
Comput. Chem. 1990, 11, 440) witha 4r distance-dependent
dielectric model. Alignments of the molecules were generated
using the following procedure. Three pharmacophore points,
the proximal THN nitrogen, the 3-aryl group, and the car-
boxylate, were assigned for eachconformation of eachmole-
cule. The resulting inter-point distances were used as a
molecular descriptor and clustering based on that descriptor
generated 18 clusters (unique presentations of the pharmaco-
phore points).