Potent, selective, orally active 3-oxo-1,4-benzodiazepine GPIIb/IIIa integrin antagonists

Full text of DOI:10.1021/jm960558a

©
Copyright 1996 by the American Chemical Society
Volume 39, Number 25
December 6, 1996
Communications to the Editor
Ta ble 1. In Vitro Activitiesa of Peptide and Nonpeptide
GPIIb/IIIa Antagonists
P oten t, Selective, Or a lly Active
3
-Oxo-1,4-ben zod ia zep in e GP IIb/IIIa
In tegr in An ta gon ists
,
†
†
J ames M. Samanen,* Fadia E. Ali,
†
†
Linda S. Barton, William E. Bondinell,
J oelle L. Burgess, J ames F. Callahan,
Raul R. Calvo, Wenting Chen, Lichong Chen,
Karl Erhard, Giora Feuerstein, Richard Heys,
Shing-Mei Hwang, Dalia R. J akas,
Richard M. Keenan, Thomas W. Ku, Chet Kwon,
Chao-Pin Lee, William H. Miller,
Kenneth A. Newlander, Andrew Nichols,
Michael Parker, Catherine E. Peishoff,
Gerald Rhodes, Steven Ross, Arthur Shu,
Richard Simpson, Dennis Takata, Tobias O. Yellin,
Irene Uzsinskas, J oseph W. Venslavsky,
†
†
†
‡
†
†
|
‡
§
†
†
†
†
∇
†
†
|
†
⊥
#
†
‡
#
†
†
†
†
†
†
Chuan-Kui Yuan, and William F. Huffman
Department of Medicinal Chemistry, Radiochemistry
Department, Department of Cellular Biochemistry,
Pharmaceutical Sciences, Pharmacology Department,
Department of Physical/ Structural Chemistry, and
Department of Pharmacokinetics, SmithKline Beecham
Pharmaceuticals, King of Prussia, Pennsylvania 19406
Received J uly 29, 1996
The role of GPIIb/IIIa antagonists in the prevention
of thrombosis has gained increasing attention in recent
years.1 Since chronic oral administration of GPIIb/IIIa
antagonists could, in principle, address myocardial
infarction and stroke, the number one and number three
,2
3
leading causes of death in the United States, intense
interest has developed over the discovery of orally active
4
GPIIb/IIIa antagonists. The peptide SK&F 107260 (1,
a
Methods are described in Bondinell et al.8 b Patelet aggrega-
depicted in Table 1), shown to contain a turn-extended-
turn conformation about the critical residues (NMe)Arg-
Gly-Asp, was one of a series of cyclic peptide antago-
nists that defined a peptide pharmacophore model
tion in human platelet-rich plasma induced by ADP, as in
Bondinell et al. Binding of [ H]-1 to GPIIb/IIIa purified from
8
c
3
5
human platelets, reconstituted in liposomes, as in Bondinell et
8
6
al.
7
which was employed in the design of a series of
8
-substituted 3-oxo-1,4-benzodiazepine GPIIb/IIIa an-
tagonists, typified by the 8-(p-amidinophenyl)amido
†
Department of Medicinal Chemistry.
Radiochemistry Department.
Department of Cellular Biochemistry.
Pharmaceutical Sciences.
Pharmacology Department.
Department of Physical/Structural Chemistry.
Department of Pharmacokinetics.
7,8
compounds 2 and 3.
Extensive investigations of the
‡
§
8-substituted analogs have led to the discovery of
several analogs which displayed oral activity but with
limited oral duration of action.
∇
|
⊥
#
We have concurrently explored a series of 7-substi-
S0022-2623(96)00558-4 CCC: $12.00 © 1996 American Chemical Society

4
868 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25
Communications to the Editor
Ta ble 2. Effects of Intravenous and Intraduodenal Doses of GPIIb/IIIa Antagonists on ex Vivo Platelet Aggregation in the Conscious
a
Dog
compd
no.
dose
(mg/kg)
route of
admin
maximal
inhibition (%)
time (min) to
maximal inhibition
duration (min) of
maximal inhibition
inhibition (%)
at 390 min
b
c
7
8
9
0.3
iv
id
iv
id
iv
id
90-100
70-90
90-100
90-100
90-100
90-100
10
30
5
10
5
∼140
∼140
∼240
∼240
∼240
∼240
controld
control
50
60
50
3
.0
0.3
.0
0.1
.0
3
1
30
90
a
8
b
Methods are described in Bondinell et al. Route of administration: iv ) intravenous, bolus dose; id ) intraduodenal catheter, short
infusion. Maximal inhibition of platelet aggregation ex vivo, canine whole blood/collagen. d Returned to control level of platelet aggregation
prior to 390 min.
c
F igu r e 1. Overlay of 7-substituted 1,4-benzodiazepine 4 and
8
-substituted 1,4-benzodiazepine 2, showing that the amidine
F igu r e 3. Overlay of 7-substituted 1,4-benzodiazepine 7 and
8-substituted 1,4-benzodiazepine 2, showing that the bipiper-
idine in 7 is capable of accessing the same space as the amidine
in 2, as described in Figure 2.
in 4 cannot readily access the same space as the amidine in 2
when both benzodiazepine templates are superimposed, as
9
described in Ku et al. Models of 2 and 4 were prepared in
SYBYL (Tripos Associates) from the X-ray crystal structure
5
of the benzodiazepine nucleus (Ku et al.). The conformations
shown for 2 and 4 were minimized via SYBYL.
F igu r e 4. Percent platelet aggregation (canine whole blood
stimulated with collagen) ex vivo after (O) 0.1 mg/kg, (9) 0.3
mg/kg, or (b) 1.0 mg/kg oral (po) administration of 9 to the
conscious dog.
F igu r e 2. Overlay of 7-substituted 1,4-benzodiazepine 6 and
8
-substituted 1,4-benzodiazepine 2, showing that the piperi-
an expanded cationic site), or (b) there are two distinct
sets of receptor binding groups that bind the cationic
groups. While N-methylation of the 8-(p-amidinophen-
yl)amido compound 2 gave rise to the orally active
N-methylamide 3, the equivalent N-methylation of the
dine in 6 is capable of accessing the same space as the amidine
in 2. The conformation of 2 had been previously shown to
allow for an overlay of the amidine and carboxylate groups in
2
1
with the guanidine and carboxylate groups in cyclic peptide
5
, described in Ku et al. Models of 2 and 6 were prepared as
described in Figure 1.
7
-(p-amidinophenyl)amido compound 4 afforded the
tuted 3-oxo-1,4-benzodiazepines, keeping in mind our
initial overlay hypothesis as a basic design guide. In
an earlier publication, we showed that an amidinophen-
yl group in position 7 of the potent benzodiazepine
GPIIb/IIIa antagonist 4 occupies a different region of
space than the amidinophenyl group in position 8 of 2,
potent antagonist 5, Table 1, which displayed in vivo
activity after intravenous administration at 0.3 mg/kg
but displayed no in vivo activity after intraduodenal
administration at 3 mg/kg (data not shown).
We sought a new series of compounds in which a
7-substituent could position a cationic group in the same
region of space accessed by the cationic 8-substituent
of 4. In this way it might be possible to merge aspects
of the structure-activity relationships (SAR) of the 7-
and 8-substituted series, allowing for the discovery of
9
Figure 1, suggesting two hypotheses: (a) The receptor
group(s) that bind the cationic groups in RGD-related
peptides and nonpeptide mimetics can occupy two
different regions of space due to receptor flexibility (i.e.,

Communications to the Editor
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25 4869
Sch em e 1. Homochiral Synthesis12 of 3-Oxo-1,4-benzodiazepine 9a
a
(a) 0.1 M 11 in anhydrous DMSO, 125 °C (47% of 12, 28% of 13); (b) 1:1 TFA/CH2Cl2, anisole (95%); (c) 1-BOC-4,4′-bipiperidine, EDC,
(i-Pr)2NEt, DMF (94%); (d) 2.0 N NaOH (2 equiv), 1:1 MeOH/THF, then AcOH (81%); (e) 4 M HCl in dioxane, CHCl3, then neutralization
of excess reagent with ca. 1.0 N KOH in EtOH to give 9‚HCl (75%) and precipitation from aqueous solution at pH 6.8 (83%).
nophenyl group in 2, albeit in a somewhat contorted
conformation, Figure 2. Other orientations are also
possible via isomerism of the amide linkage. The more
rigid bipiperidinylcarbonyl group in the 7-substituted
compound 7 could hold the terminal piperidine into a
more fixed orientation and avoid the ambiguity of amide
bond isomerism. As can be seen in the overlay shown
in Figure 3, the 7-bipiperidinyloxo group in 7 is capable
of positioning its cationic amine in the same region of
space as the cationic amine in the 8-(amidinophenyl)-
amido group of 2. Compound 7 proved to be a potent
GPIIb/IIIa antagonist in vitro, Table 1, as well as in the
conscious dog, Table 2, comparable now to the 8-sub-
3
15
F igu r e 5. Reversible binding of [ H]-9 to purified GPIIbIIIa.
8
3
3
stituted benzodiazepine 2.
GPIIb/IIIa was incubated with radioligand [ H]-1 or [ H]-9 in
the presence or absence of unlabeled 1 or 9 for 2 h. Alterna-
tively, GPIIb/IIIa was pre-equilibrated with radioligand for 1
h, 2 µM unlabeled 1 or 9 was then added, and the samples
were incubated for an additional 1 h; 0.5 and 1 µg of the lentil
lectin-purified GPIIb/IIIa was used in these radioligand bind-
ing assays for 1 and 9, respectively. In experiments A-E, the
conditions were varied as follows: (A) 2 nM [ H]-9 incubated
with GPIIb/IIIa for 2 h (showing total binding of [ H]-9), (B) 2
nM [ H]-9 and 1 µM of cold 9 incubated together with GPIIb/
IIIa for 2 h (showing reversibility of binding by 9), (C) 2 nM
H]-9 and 1 µM 1 incubated together with GPIIb/IIIa for 2 h
showing reversibility of binding by 1), (D) 2 nM [ H]-9
incubated with GPIIb/IIIa for 1 h and then 1 µM 9 for 1 h
again, showing reversibility of binding by 9), and (E) 2 nM
H]-9 incubated with GPIIb/IIIa for 1 h and then 1 µM 1 for
h (again, showing reversibility of binding by 1).
Continued exploration of the 7-bipiperidinyloxo series
led to an investigation of position 4. Although replace-
ment of the 4-phenethyl group with 4-Me, as in 8,
resulted in a decrease in potency in vitro, Table 1, the
4
-Me modification resulted in an unexpected gain in
3
duration of antiaggregatory activity in vivo, Table 2.
3
Both enantiomers of 8 were prepared by homochiral
3
12
synthesis from (R)- and (S)-aspartic acid. The key step
in the synthesis of both enantiomers involved the
intramolecular displacement of the aryl fluoride 11 to
construct the seven-membered ring of the 1,4-benzodi-
azepine system, as exemplified in the synthesis of the
3
[
(
3
(
[
3
13
S-enantiomer, Scheme 1. By this reaction, the desired
1
cyclization product 12 was obtained without racemiza-
tion, together with the elimination product 13. To
complete the synthesis, the carboxylic acid 14 was
new compounds that could display high affinity for
GPIIb/IIIa but contain structures that would possess
enhanced pharmacokinetic properties. One solution
from that search was the (piperidinylpropyl)amide 6. A
piperidinyl GPIIb/IIIa antagonist has been described by
Hartman et al. and in our own work where 8-(pi-
peridinylethyl)amide 1,4-benzodiazepine analogs were
found to display potent GPIIb/IIIa antagonist activity.
Compound 6 displayed high affinity for GPIIb/IIIa and
high potency in the platelet aggregation assay, Table
. The cationic amine of the piperidinylpropyl group
could access the same region of space as the 8-amidi-
1
4
coupled with 1-BOC-4,4′-bipiperidine to afford, after
protecting group removal, zwitterionic SB 214857 9, in
good overall yield and high enantiomeric purity. As
seen in Table 1, the S-enantiomer 9 is considerably more
1
0
1
5
potent in vitro than the R-enantiomer 10.
1
1
SB 214857, 9, is a potent antiaggregatory agent after
intravenous and intraduodenal administration to con-
scious dogs, more potent than either 7 or 8, Table 2,
and displays an extended duration of action. The
compound is also orally active, Figure 4. After po
1

4
870 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 25
Communications to the Editor
administration of compound 9 at 0.1 mg/kg, platelet
aggregation ex vivo was inhibited by 80% between 90
min and 3 h. After po administration of compound 9
at 1.0 mg/kg, however, platelet aggregation was inhib-
ited by >80% for the duration of the experiment, 8 h.
This increased activity in vivo for the 4-Me com-
pounds 8 and 9 over the 4-phenethyl compound 7 is
counterintuitive, since the decreased affinity and lipo-
philicity of the smaller methyl group would suggest
diminished in vivo activity. In hindsight, we believe
that the smaller size of the 4-Me group and diminished
clearance are the factors that favor the 4-Me over the
(3) Gunby, P. Cardiovascular diseases remain nation’s leading cause
of death. J AMA 1992, 267, 335-336.
(
4) Ali, F. E.; Samanen, J . M.; Calvo, R.; Romoff, T.; Yellin, T.;
Vasko, J .; Powers, D.; Stadel, J .; Bennett, D.; Berry, D.; Nichols,
A. Potent Fibrinogen Receptor Antagonists Bearing Conforma-
tional Constraints. In Peptides, Chemistry and Biology, Pro-
ceedings 12th American Peptide Symposium; Smith, J . A., Rivier,
J . E., Eds.; ESCOM: Leiden, 1992; pp 761-762.
(
5) Kopple, K. D.; Baures, P. W.; Bean, J . W.; D’Ambrosio, C. A.;
Hughes, J . L.; Peishoff, C. E.; Eggleston, D. S. Conformations
of Arg-Gly-Asp Containing Heterodetic Cyclic Peptides: Solution
and Crystal Studies. J . Am. Chem. Soc. 1992, 114, 9615-9623.
6) Peishoff, C. E.; Ali, F. E.; Bean, J . W.; Calvo, R.; D’Ambrosio, C.
A.; Eggleston, D. S.; Hwang, S. M.; Kline, T. P.; Koster, P. F.;
Nichols, A.; Powers, D.; Romoff, T.; Samanen, J . M.; Stadel, J .;
Vasko, J . A.; Wong, A.; Kopple, K. D. Investigation of Confor-
mational Specificity at GPIIb/IIIa: Evaluation of Conformation-
ally Constrained RGD Peptides. J . Med. Chem. 1992, 35, 3962-
(
4
-phenethyl group.
3
969.
The several modifications to the benzodiazepine 2 that
(
7) Ku, T. W.; Ali, F. E.; Barton, L. S.; Bean, J . W.; Bondinell, W.
E.; Burgess, J . L.; Callahan, J . F.; Calvo, R. R.; Chen, L.;
Eggleston, D. S.; Gleason, J . G.; Huffman, W. F.; Hwang, S. M.;
J akas, D. R.; Karash, C. B.; Keenan, R. M.; Kopple, K. D.; Miller,
W. M.; Newlander, K. A.; Nichols, A.; Parker, M. F.; Peishoff,
C. E.; Samanen, J . M.; Uzinskas, I.; Venslavsky, J . W. Direct
Design of a Potent Non-Peptide Fibrinogen Receptor Antagonist
Based on the Structure and Conformation of a Highly Con-
trained Cyclic RGD Peptide. J . Am. Chem. Soc. 1993, 115,
led to 9 raise the question of whether 9 is still a
nonpeptide mimetic of peptide 1. That 9 may be
regarded as a nonpeptide mimetic of the peptide 1 comes
from the fact that both compounds (a) position cationic
amine and anionic carboxylate groups in the same
regions of space and (b) require an (S)-acetic acid side
chain. We have also found that a derivative of 9 bearing
tritium in the benzo group displays specific, saturable
binding to human GPIIb/IIIa with a Kd of 2 nM,
8
861-8862.
(8) Bondinell, W. E.; Keenan, R. M.; Miller, W. H.; Ali, F. E.; Allen,
A. C.; DeBrosse, C. W.; Eggleston, D. S.; Erhard, K. F.;
Haltiwanger, R. C.; Huffman, W. F.; Hwang, S.-M.; J akas, D.
R.; Koster, P. F.; Ku, T. W.; Lee, C. P.; Nichols, A. J .; Ross, S.
T.; Samanen, J . M.; Valocik, R. E.; Vasko-Moser, J . A.; Ven-
slavsky, J . W.; Wong, A. S.; Yuan, C.-K. Design of a Potent and
Orally Active Nonpeptide Platelet Fibrinogen Receptor (GPIIb/
IIIa) Antagonist. Bioorg. Med. Chem. 1994, 2, 897-908.
3
identical with the Ki of inhibition of [ H]-1 binding to
GPIIb/IIIa by cold 9.16 Tritiated 9 binding, furthermore,
is completely reversed by either cold 9 or cold 1, even
after preincubation for 1 h, Figure 5. We have also
found that the Ki values for a series of GPIIb/IIIa
(
9) Ku, T. W.; Miller, W. H.; Bondinell, W. E.; Erhard, K. F.; Keenan,
R. M.; Nichols, A. J .; Peishoff, C. E.; Samanen, J . M.; Wong, A.
S.; Huffman, W. F. Potent Nonpeptide Fibrinogen Receptor
Antagonists Which Present an Alternative Pharmacophore. J .
Med. Chem. 1995, 38, 9-12.
3
3
antagonists, determined in either a [ H]-9 or [ H]-1
competition binding assay, are similar.16 Thus, we
observe no appreciable difference between the binding
of SB 214857 and SK&F 107260 to purified human
GPIIb/IIIa. Site-directed mutagenesis experiments in
(
10) Hartman, G. D.; Egbertson, M. S.; Halszenko, W.; Laswell, W.
L.; Duggan, M. E.; Smith, R. L.; Naylor, A. M.; Manno, P. D.;
Lynch, R. J .; Zhang, G.; Chang, C. T. C.; Gould, R. J . Non-peptide
Fibrinogen Receptor Antagonists. 1. Discovery and Design of
Exosite Inhibitors. J . Med. Chem. 1992, 35, 4640-4642.
(11) (a) Bondinell, W. E. Potent and Orally Active Nonpeptide Platelet
Fibrinogen Receptor (GPIIb/ IIIa) Antagonists Designed from
Cyclic Peptides. SCI Conference on Chemical Lead Generation
in Drug Discovery, Cambridge, England, J une 25-27, 1995. (b)
Bondinell, W. E.; et al. Manuscript in preparation.
12) Miller, W. H.; Ku, T. W.; Ali, F. E.; Bondinell, W. E.; Calvo, R.
R.; Davis, L. D.; Erhard, K. E.; Hall, L. B.; Keenan, R. M.; Kwon,
C.; Newlander, K. A.; Ross, S. T.; Samanen, J . M.; Takata, D.
T.; Yuan, C.-K. Enantiospecific Synthesis of SB 214857, a Potent,
Orally Active Nonpeptide Fibrinogen Receptor Antagonist,
Tetrahedron Lett. 1995, 36, 9433-9436.
the G-protein-coupled receptor area suggest that, in
some cases,1
7,18
nonpeptide ligands can bind to sites that
are similar to the peptide binding site or, in other
1
9,20
cases,
distinct from the peptide binding site. SB
2
14857, however, appears to be an example where the
(
nonpeptide ligand binds to the same receptor binding
site as the peptide and, as such, is a true peptidomi-
metic.
In conclusion, from our series of 3-oxo-1,4-benzodiaz-
epines, we have discovered a potent, selective, orally
active GPIIb/IIIa antagonist that displays an extended
(13) All new compounds gave satisfactory spectroscopic and analytical
data as described in ref 14.
14) Bondinell, W. E.; Callahan, J . F.; Huffman, W. F.; Keenan, R.
M.; Ku, T. W.; Newlander, K. A.; Samanen, J . M.; Uzinskas, I.
N. WO 9414776, J uly 7, 1994.
(
2
0
duration of action. Furthermore, with (a) the discov-
ery of high affinity in 7-(bipiperidinylamido)-1,4-benzo-
diazepines and (b) their apparent mimicry of the original
peptide ligand, we have shown that the original design
hypothesis can continue to serve as a guide in analog
design during subsequent studies, such as the search
for analogs with enhanced oral duration. SB 214857 is
currently undergoing clinical investigation.
(
15) Compound 9 is also selective for GPIIb/IIIa over Rvâ3 (K
i
10 340
(
1200 nM) in a binding assay employing purified human
3
placental Rvâ3 and [ H]-1.
(16) Wong, A.; et al. Manuscript in preparation.
(
17) Lee, J . A.; Sutiphong, J . A.; Longton, E. D.; Peishoff, C. E.;
Stadel, J . M.; Kumar, C.; Ohlstein, E. H.; Gleason, J . G.; Elliott,
J . D. Lysine 182 of endothelin B receptor modulates agonist
selectivity and antagonist affinity: evidence for the overlap of
peptide and non-peptide ligand binding sites. Biochemistry
1995, 33, 14543-14549.
(18) Lee, J . A.; Elliott, J . D.; Sutiphong, J . A.; Friesen, W. J .; Ohlstein,
E. H.; Stadel, J . M.; Gleason, J . G.; Peishoff, C. E. Tyr-129 is
important to the peptide ligand affinity and selectivity of human
endothelin type A receptor. Proc. Natl. Acad. Sci. U.S.A. 1994,
Su p p or tin g In for m a tion Ava ila ble: Elemental analyti-
cal data and additional figures showing the effects of com-
pounds 3, 5, and 7-9 on ex vivo platelet aggregation in the
conscious dog (1 page). Ordering information is given on any
current masthead page.
9
1, 7164-7168.
(
19) Gether, U.; J ohansen, T. E.; Snider, R. M.; Lowe, J . A., III;
Nakanishi, S.; Schwartz, T. W. Different Binding Epitopes on
the NK1 Receptor for Substance P and a Non-peptide Antago-
nist. Nature 1993, 362, 345-348.
Refer en ces
(
1) Nichols, A. J .; Ruffolo, R. R., J r.; Huffman, W. F.; Poste, G.;
Samanen, J . Development of GPIIb/IIIa Antagonists as Anti-
thrombotic Drugs. Trends Pharm. Sci. 1992, 13, 413-417.
2) Nichols, A. J .; Vasko, J . A.; Koster, P. F.; Valocik, R. E.;
Samanen, J . M. GPIIb/IIIa Antagonists as Novel Antithrombotic
Drugs: Potential Therapeutic Agents. In Cellular Adhesion:
Molecular Definition to Therapeutic Potential; Metcalf, B. W.,
Dalton, B. J ., Poste, G., Eds.; Plenum Press: New York, 1994;
pp 213-237.
(20) Fong, T. M.; Cascieri, M. A.; Yu, H.; Bansai, A.; Swain, C.;
Strader, C. A. Amino-Aromatic Interaction Between Histidine
197 of the Neurokinin-1 Receptor and CP 96345. Nature 1993,
362, 350-353.
(21) This work was first communicated at the 14th American Peptide
Symposium, Columbus, OH, J une 18-23, 1995.
(
J M960558A