Isoxazolyl, oxazolyl, and thiazolylpropionic acid derivatives as potent α4β1 integrin antagonists

Article abstract of DOI:10.1016/S0960-894X(01)00511-X

A series of isoxazolyl, oxazolyl, and thiazolylpropionic acid derivatives derived from LDV was found to be a potent antagonist of the α₄β₁ integrin. The synthesis and SAR leading up to 3-[3-(1-{2-[3-methoxy-4-(3-o-tolyl-ureido)-pheny

Full text of DOI:10.1016/S0960-894X(01)00511-X

Bioorganic & Medicinal Chemistry Letters 11 (2001) 2593–2596
Isoxazolyl, Oxazolyl, and Thiazolylpropionic Acid Derivatives as
Potent ꢀ4ꢁ1 Integrin Antagonists
Allen J. Duplantier,* Gretchen E. Beckius, Robert J. Chambers, Louis S. Chupak,
Teresa H. Jenkinson, Anne S. Klein, Kenneth G. Kraus, Elizabeth M. Kudlacz,
Michael W. McKechney, Martin Pettersson, Carrie A. Whitney and Anthony J. Milici
Pfizer Global Research and Development, Groton Labs, Eastern Point Road, Groton, CT 06340, USA
Received 8 June 2001; accepted 19 July 2001
Abstract—A series of isoxazolyl, oxazolyl, and thiazolylpropionic acid derivatives derived from LDV was found to be a potent
antagonist of the a4b1 integrin. The synthesis and SAR leading up to 3-[3-(1-{2-[3-methoxy-4-(3-o-tolyl-ureido)-phenyl]-acetyl-
amino}-3-methyl-butyl)-isoxazol-5-yl]-propionic acid (22) are reported. In an allergic mouse model, compound 22 was efficacious
delivered systemically (58% inhib @ 10 mg/kg, sc) as well as by intra-tracheal instillation (ED50=2 mg/kg). # 2001 Elsevier Science
Ltd. All rights reserved.
The integrin a4b1 or very late antigen-4 (VLA-4) is
constitutively expressed on the surface of all leukocytes
(lymphocytes, monocytes, eosinophils, and basophils).1
It binds to vascular cell adhesion molecule-1 (VCAM-1)
on activated endothelial cells resulting in migration of
cells out of the vasculature at sites of inflammation. In
addition, once cells have reached the extravascular
space, VLA-4 can bind to the CS-1 region on fibro-
nectin resulting in enhanced leukocyte activation and
release of inflammatory mediators.2,3 Antibodies to a4
have blocked leukocyte infiltration in a variety of ani-
mal models of human disease (asthma, multiple sclero-
sis, inflammatory bowel disease, etc.)4À6 as well as in
multiple sclerosis in man.7 Small peptide and nonpep-
tide a4b1 antagonists have similarly shown efficacy in
disease models of inflammation (BIO12118 and TR-
140359). Thus, antagonism of a4b1 represents a well
validated target for the treatment of inflammatory dis-
orders.
BIO1211).11 In efforts to obtain an orally available a4b1
antagonist we modified the LDV trimer by tying the
aspartic acid moiety into an isoxazole ring and remov-
ing the valine group, thus reducing the number of
metabolically labile amide bonds (1). The evaluation of
several libraries in which the amino terminus of 1 was
coupled with various carboxylic acids led us to the same
diphenylurea substituent (5) that others have reported.11
Humphries reported LDV to be the minimal essential
binding sequence of the CS-1 region on fibronectin.10
Subsequent reports by others have revealed potent a4b1
antagonists by capping the amino terminus of LDV
with substituted phenylacetyl groups, and capping the
carboxy terminus with various amino acids (e.g.,
*Corresponding author. Fax: +1-860-686-605; e-mail: allen_j_du-
plantier@groton.pfizer.com
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00511-X

2594
A. J. Duplantier et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2593–2596
The preparation of isoxazole 5 began with the reaction
of oxime 112 with sodium hypochlorite13 in the presence
of the terminal alkyne 2 to form the appropriate iso-
xazole ring, which was subsequently deprotected with
HCl in dioxane to give amine 3 as a hydrochloride salt
(Scheme 1). EDCI coupling of 3 with carboxylic acid 4
followed by saponification with 1 N NaOH in tert-
BuOH provided 5 in good yield. In a similar fashion,
oxime 1 was coupled to various other olefins and
alkynes to give compounds 6–10 (Scheme 2).
converted to carboxamide 17 by reaction with isobutyl
chloroformate followed by ammonia.15
The preparation of thiazolylacetic acid 19 from 1816 was
analogous to the synthesis of thiazolylpropionic acid 16
from 12, and is presented in Scheme 4. The synthesis of
the regioisomeric analogues of 16 and 19 (21 and 20,
respectively) is presented in Scheme 5. The reaction of
the thioamide of N-Cbz-l-leucine with ethyl 4-chloro-3-
oxobutyrate in acetone followed by the addition of tri-
fluoroacetic anhydride and pyridine17 resulted in the
desired thiazolyacetate precursor to 20, but in low yield.
Likewise, reaction with ethyl 5-chloro-4-oxopropionate
in ethanol in the presence of potassium carbonate at
reflux18 gave the desired thiazolylpropionate precursor
to 21 in 33% yield.
The synthesis of the oxazole and thiazole analogues of 5
is presented in Scheme 3. N-Cbz-l-leucine was coupled
with 5-aminolevulinic acid methyl ester hydrochloride
(11) in the presence of EDCI to provide the amide
intermediate 12, which was then treated with either
phosphorus oxychloride14 to give oxazole 13, or Law-
esson’s reagent to give thiazole 15. Compounds 13 and
15 were then converted to 14 and 16, respectively, by Pd
catalyzed hydrogenation to remove the Cbz group fol-
lowed by EDCI coupling of the resulting amine with 4,
and subsequent saponification. Carboxylic acid 16 was
Scheme 3. (a) HOBT, EDCI, DIEA, DMF, rt, 16 h; (b) POCl3, tolu-
ene, reflux (35%); (c) Lawesson’s reagent, toluene, reflux (41%); (d)
H2, Pd/C, EtOAc; (e) 4, HOBT, EDCI, DIEA, DMF, rt, 16 h; (f) 1:1
1 N NaOH, tert-BuOH; (g) N-methylmorpholine, DME, isobutyl
chloroformate, À10 ꢀC, 15 min; then NH3(g), 5 min; * see Scheme 1 for
definition of R.
Scheme 1. (a) 5% NaOCl, TEA, CH2Cl2 (29%); (b) 4 M HCl in
dioxane, rt, 1 h (99%); (c) HOBT, EDCI, DIEA, DMF, rt, 16 h; (d)
1:1 1 N NaOH, tert-BuOH.
Scheme 4. (a) Lawesson’s reagent, toluene, reflux; (b) 4 M HCl in
dioxane, rt, 1 h; (c) 4, HOBT, EDCI, DIEA, DMF, rt, 16 h; (d) 1:1 1 N
NaOH, tert-BuOH; * see Scheme 1 for definition of R .
Scheme 5. (a) Acetone, rt, 4 h, then trifluoroacetic anhydride, pyri-
dine, À10 ꢀC, rt, 16 h (18%); (b) H2, Pd/C, EtOAc; (c) 4, HOBT,
EDCI, DIEA, DMF, rt, 16 h; (d) 1:1 1 N NaOH, tert-BuOH; (e)
K2CO3, EtOH, reflux, 5 days (33%); * see Scheme 1 for definition of R.
Scheme 2. (a) 5% NaOCl, TEA, CH2Cl2; (b) 4 M HCl in dioxane, rt, 1
h; (c) 4, HOBT, EDCI, DIEA, DMF, rt, 16 h; (d) 1:1 1 N NaOH, tert-
BuOH; * see Scheme 1 for definition of R.

A. J. Duplantier et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2593–2596
Table 1. In vitro potency of a4b1 integrin antagonists
2595
Compd
Receptor/ligand ELISA IC50 (nM)a
Jurkat cell IC50 (nM)a
Human eosinophil IC50 (nM)a
5
6
7
8
9
10
14
16
17
19
20
21
48Æ20 (9)
8 (2)
20 (1)
500 (1)
nt
27Æ7 (13)
67 (2)
88 (1)
1800 (2)
na
na
45Æ13 (8)
16Æ5 (5)
nt
2400 (1)
nt
143Æ72 (6)
347Æ115 (3)
495Æ158 (3)
nt
nt
nt
nt
171Æ20 (3)
180Æ57 (8)
nt
3Æ1 (3)
0.25Æ0.06 (4)
na
nt
nt
nt
nt
800 (1)
800 (1)
10.3 (1)
2.6Æ1.0 (3)
nt
2040 (1)
0.69 (2)
4.3Æ2.0 (4)
9.8 (1)
22
BIO1211
TR-14035
1.0 (2)
97Æ47 (3)
140Æ110 (3)
aValuesÆstandard error are means of the number of experiments in parentheses (na=0% inhib at 1 mM; nt=not tested).
The synthesis of 22, an ortho-methoxyphenylurea ana-
logue of 5, is shown in Scheme 6. 4-Chloro-2-methoxy-
nitrobenzene was added to a preformed mixture of
sodium hydride and diethylmalonate in DMF to give
the desired 4-nitro-3-methoxyphenylmalonate, which
was reduced by means of Pd catalyzed hydrogenation.
The resulting amine was treated with o-tolyl isocyanate
followed by saponification, EDCI coupling with 3, and
final saponification to provide 22.
As shown in Table 1, isoxazole 5 was found to be a
potent a4b1 antagonist in the ELISA (IC50=48 nM),
Jurkat cell (IC50=27 nM), and human eosinophil
(IC50=143 nM) assays. The isoxazole ring (5) could be
replaced with other five-membered heterocyclic rings
such as isoxazoline (7), isoxazole (14), and thiazole (16)
without effecting the potency. Interestingly, isoxazolines
6 and 7 were equipotent, implying that one of the
propionic acid groups in 6 is not involved with binding.
The regiochemical placement of the propionic acid was
found to be important, as can be seen with the greatly
reduced potency of thiazolylpropionic acid 21. Mod-
ification of the length of the propionic acid side chain
was detrimental to potency (8 and 19). Compound 22, a
methoxy analogue of 5, was found to be quite potent
(IC50=1 nM) in both the Jurkat cell and human eosino-
phil assays. Moreover, 22 was found to be 100Â more
potent than both BIO1211 and TR-14035 against
human eosinophils.
Compounds were screened for VLA-4 antagonist activ-
ity in a cell-based assay in which calcein labeled Jurkat
cells were allowed to bind to human recombinant
sVCAM-1 in the presence of serial dilutions of test
compounds. All active compounds were then profiled in
a receptor/ligand ELISA assay to confirm that the
compounds site of action is on VLA-4. In this assay,
VLA-4 isolated from Jurkat cells was plated in 96-well
plates and incubated with serial dilutions of compound
together with the biotinylated CS-1 fragment of fibro-
nectin. Binding was detected with streptavidin-alkaline
phosphatase. Compounds that were active in both
assays were then profiled in a human eosinophil cell-
binding scintillation proximity assay (SPA). In this
assay, eosinophils were freshly isolated from healthy
donors, labeled with 3H-arachidonic acid and incubated
with VCAM-1 coated SPA beads in the presence of
serial dilutions of compound.
Compounds 5, 16, and 22 were evaluated in an allergic
mouse model,19 measuring the compound’s effect on
BALF eosinophil influx in response to antigen chal-
lenge.20 Compound 22 was found to be efficacious fol-
lowing systemic administration (58% inhib @ 10 mg/kg,
sc).21 When dosed by intra-tracheal instillation, com-
pounds 5, 16, and 22 had ED50’s of 20, 100, and 2 mg/kg,
respectively.21 At similar doses, BIO1211 was ineffective.
In conclusion, we have developed a novel series of a4b1
integrin antagonists that were derived from LDV.
Compound 22 is a representative analogue that has
nanomolar potency in the human eosinophil assay, is
stable in human plasma and is efficacious in an in vivo
allergic mouse model. Future directions continue to
focus on improving oral bioavailability.
References and Notes
1. Hemler, M. Annu. Rev. Immunol. 1990, 8, 365.
2. Guan, J.; Hynes, R. Cell 1990, 60, 53.
3. Wayer, E.; Garcia-Pardo, A.; Humphries, M.; McDonald,
J.; Carter, W. J. Cell Biol. 1989, 109, 1321.
Scheme 6. (a) Diethylmalonate, NaH, DMF, 90 ꢀC, 16 h; (b) H2, Pd/
C, EtOAc; (c) o-tolyl isocyanate, dichloromethane, TEA; (d) NaOH,
tert-BuOH, reflux; (e) 3, HOBT, EDCI, DIEA, DMF, rt, 16 h; (f)
NaOH, tert-BuOH.

2596
A. J. Duplantier et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2593–2596
4. Abraham, W.; Sielczak, M.; Ahmed, A.; Cortes, A.; Lau-
redo, I.; Kim, J.; Pepinsky, B.; Benjamin, C.; Leone, D.; Lobb,
R.; Weller, P. J. Clin. Invest. 1994, 93, 776.
5. Kent, S.; Karlik, S.; Cannon, C.; Hines, D.; Yednock, T.;
Fritz, L.; Horner, H. J. Neuroimmunol. 1995, 58, 1.
L.; Pepinsky, R.; Leone, D.; Sprague, A.; Abraham, W.; Gill,
A.; Lobb, R.; Adams, S. J. Med. Chem. 1999, 42, 920.
12. Jones, R.; Dawson, C.; O’Mahony, M. Synlett 1999, S1, 873.
13. Lee, G. Synthesis 1982, 508.
14. Dow, R. J. Org. Chem. 1990, 55, 386.
6. Podolsky, D.; Lobb, R.; King, N.; Benjamin, C.; Pepinsky,
B.; Sehgal, P.; deBeaumont, M. J. Clin. Invest. 1993, 92, 372.
7. Tubiridy, N.; Behan, P.; Capildeo, R.; Chaudhuri, A.;
Forbes, R.; Hawkins, C.; Hughes, R.; Palace, J.; Sharrack, B.;
Swingler, R.; Young, C.; Moseley, I.; MacManus, D.; Don-
oghue, S.; Miller, D. Neurology 1999, 53, 466.
8. Abraham, W.; Gill, A.; Ahmed, A.; Sielczak, M.; Lauredo,
I.; Botinnikova, Y.; Lin, K.-C.; Pepinsky, B.; Leone, D.;
Lobb, R.; Adams, S. Am. J. Respir. Crit. Care Med. 2000, 162,
603.
9. Sircar, I.; Gudmundsson, K.; Martin, R.; Nomura, S.;
Jayakumar, H.; DM, N.; Cardarelli, P.; Mah, J.; Castro, M.;
Cao, Y.; Griffiths, R.; Lazarides, E. 218th National Meeting
of the American Chemical Society, New Orleans, LA; Amer-
ican Chemical Society: Washington, DC, 1999; MEDI 59.
10. Komoriya, A.; Green, L.; Mervic, M.; Yamada, S.;
Yamada, K.; Humphries, M. J. Biol. Chem. 1991, 266, 15075.
11. Lin, K.; Ateeq, H.; Hsiung, S.; Chong, L.; Zimmerman,
C.; Castro, A.; Lee, W.; Hammond, C.; Kalkunte, S.; Chen,
15. Travecchia, P.; Gentili, P.; Kurz, M.; Sottani, C.; Bonfi-
chi, R.; Selva, E.; Lociuro, S.; Restelli, E.; Ciabatti, R. Tetra-
hedron 1995, 51, 4867.
16. Hashiguchi, S.; Kawada, A.; Natsugari, H. Synthesis
1992, 403.
17. Schmidt, U.; Gleich, P.; Griesser, H.; Utz, R. Synthesis
1986, 992.
18. Videnov, G.; Kaiser, D.; Kempter, C.; Jung, G. Angew.
Chem., Int. Ed. Engl. 1996, 35, 1503.
19. For a general reference on mouse models of asthma, see
Herz, U.; Lumpp, U.; Da Palma, J. C.; Enssle, K.; Takatsu,
K.; Schnoy, N.; Daser, A.; Kottgen, E.; Wahn, U.; Renz, H.
Immunol. Cell Biol. 1996, 74, 209.
20. Kudlacz, E. M.; Andresen, C. J.; Salafia, M.; Whitney,
C. A.; Naclerio, B.; Changelian, P. S. Am. J. Respir. Cell Mol.
Biol. 2001, 24, 469.
21. Kudlacz, E.; Whitney, C.; Conklyn, M.; Duplantier, A.;
Milici, A. J. Am. J. Respir. Crit. Care Med. 2001, 163,
A196.