Selective cell adhesion inhibitors: Barbituric acid based α4β7-MAdCAM inhibitors

Article abstract of DOI:10.1016/j.bmcl.2007.07.068

A novel series of barbituric acid derivatives were identified as selective inhibitors of α4β7 MAdCAM (mucosal addressin cell adhesion molecule-1) interactions via a high throughput screening exercise. These inhibitors were optimized to submicromolar potencies in whole cell adhesion assays, retaining their selectivity over α4β1 VCAM.

Full text of DOI:10.1016/j.bmcl.2007.07.068

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 2509–2512
Selective cell adhesion inhibitors: Barbituric acid based
a4b7—MAdCAM inhibitors
a,
a
a
b
a
*
Geraldine C. Harriman, Matthias Brewer, Robert Bennett, Cyrille Kuhn,
b
a
a
a
Marc Bazin, Greg Larosa, Paul Skerker, Nancy Cochran, Debra Gallant,
a
a
a
a
a
Deborah Baxter, Dominic Picarella, Bruce Jaffee, Jay R. Luly and Michael J. Briskin
a
Millennium Pharmaceuticals, 35 Landsdowne Street, Cambridge, MA 02139, USA
Pfizer Global Research and Development, 3-9 Rue de la Loge, Fresnes 94265, France
b
Received 30 April 2007; revised 17 July 2007; accepted 18 July 2007
Available online 21 August 2007
Abstract—A novel series of barbituric acid derivatives were identified as selective inhibitors of a4b7 MAdCAM (mucosal addressin
cell adhesion molecule-1) interactions via a high throughput screening exercise. These inhibitors were optimized to submicromolar
potencies in whole cell adhesion assays, retaining their selectivity over a4b1 VCAM.
Ó 2007 Elsevier Ltd. All rights reserved.
The circulation and homing of leukocyte subsets is char-
acterized by an intricate series of cell–cell interactions
that collectively contribute to a tissue specific homing
MAdCAM is known to interact with its ligand a4b7
using a very select group of residues in the c–d loop of
its first immunoglobulin domain where the recognition
binding sequence is leucine–aspartic acid–threonine (1,
1
,2
event. This multistep process that leads to the infiltra-
tion of leukocytes into target tissues involves multiple
interactions, which include the association of carbohy-
drates with selectins, chemokine receptors with chemo-
kines, and integrins with immunoglobulin-like (Ig-like)
cell adhesion molecules. Lymphocytes bearing the inte-
grin a4b7, a heterodimeric protein expressed on T-lym-
phocytes, home specifically to tissues expressing the cell
adhesion molecule mucosal addressin cell adhesion mol-
9
,10
LDT).
modified tripeptides,
modified amino acids (phenylalanine derivatives)
as integrin antagonists. Close mimics of the recognition
The literature contains several examples of
1
1–13
14,15
tripeptide mimics,
and
1
6–21
2
2
binding motif including modified LDT analogs and
2
3
mannose scaffolded LDT mimics have proven to pro-
vide highly selective molecules. Most members of the
phenylalanine class of a4b7-MAdCAM inhibitors are
non-selective and inhibit both a4b7-MAdCAM and
a4b1-VCAM mediated cell adhesion. Recent reports in
the literature regarding potential undesirable immuno-
modulatory effects via the inhibition of a4b1-VCAM
illustrate the importance of selective inhibitors for use
in the treatment of diseases driven by a4b7/MAdCAM
3
–6
ecule (MAdCAM). This cell adhesion molecule is spe-
cifically expressed on specialized vascular endothelial
cells within gut associated lymphoid tissues and within
lamina propria as well. Under inflammatory activation
conditions the expression of MAdCAM is highly upreg-
ulated and results in increased recruitment of a4b7+
lymphocytes. This phenomenon is associated with tissue
damage in both murine and non-human primate models
of inflammatory bowel disease (IBD) where anti-a4b7
2
4
lymphocyte trafficking. In our efforts to develop small
molecule antagonists of leukocyte trafficking, we identi-
fied 2 as an inhibitor of cell adhesion mediated by the
leukocyte cell surface integrin a4b7 and the cell adhesion
molecule MAdCAM via a high throughput screen of
our small molecule collection.
7
,8
mAbs have shown efficacy.
We herein report on the chemistry involved in our
optimization efforts in this series of novel and selective
barbituric acid based integrin antagonists. The starting
molecule 2, proved to be a selective inhibitor of
Keywords: Integrin; Cell adhesion molecule; MAdCAM; a4b7; Barbi-
turic acid.
*
Corresponding author. Tel.: +1 6173741010; fax: +1 6173741320.;
e-mail: gharriman@galenea.com
0
960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.07.068

2510
G. C. Harriman et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2509–2512
a4b7 binding to MAdCAM, had an IC50 = 2.2 lM
Fig. 1).
in a a4b7-MAdCAM mediated whole cell adhesion
30
assay.
(
The chemistry leading to the barbituric acids was initi-
ated by the formation of the appropriately substituted
urea (Scheme 1, synthesis of compound 2 used as exam-
ple) via the reaction of potassium cyanate and a substi-
Initial analogs were synthesized systematically holding
R and R constant (R = hydrogen, R = phenyl) and
1
3
1
3
2
varying the substituent at R . From these data we were
2
able to identify certain functionality for R that resulted
2
5,26
tuted amine 3.
This provided the urea 4 in 90% yield.
in more potent molecules. Examples of this are shown in
Table 1. The scope of the analogs ranged from small al-
kyl groups (exemplified by 11) to larger heterocyclic
groups. The optimal functional groups identified were
in general heterocyclic, with the analog containing the
3-indolyl group (9) being one of the most potent early
molecules. Position of the heteroatom proved important
as the 2-indolyl 10 was less potent than the 3-indolyl 9.
Aromatic substitutions affecting the electronics of the
phenyl containing analogs were inconsequential to the
potency of the compounds (exemplified by 14, 15, and
16).
This was followed by a two component reaction involv-
ing the condensation of a diacid equivalent with the
appropriately substituted urea 4 in refluxing sodium eth-
oxide in ethanol producing the barbituric acid 5 in 40%
2
7,28
yield.
benzaldehyde in the presence of a catalytic amount of
Reaction of the activated methylene with
acetic acid produced the exocyclic olefin 2 in 85%
2
yield.
9
We investigated the structure–activity relationships of
the barbituric acid for their ability to inhibit the interac-
tion between a4b7 and MAdCAM. Our initial approach
included the modifications of three functional groups
outlined in Figure 2. These molecules were evaluated
Once the 3-indolyl was identified as an important phar-
macophore component, the relative importance of the
free N–H hydrogens was investigated, Table 2. Methyl-
ation at the indole nitrogen resulted in a decrease in
activity (exemplified by 17). In addition methylation at
the barbituric N–H also resulted in decreased potency
O
(exemplified by 18).
OH
O
HN
H
N
OH
H N
N
H
Selected analogs were then evaluated for their selectiv-
ity against two additional integrin/cell adhesion mole-
cule pairs to determine the structure–activity trends in
selectivity (Table 3). Since we were interested in
advancing selective a4b7-MAdCAM selective inhibitors
only, we were pleased to find that our molecules
proved to be inactive against both the related inte-
grin/cell adhesion molecule pair a4b1-VCAM and the
non-related integrin/cell adhesion molecule pair a5b1-
fibronectin.
2
O
N
O
O
O
OH
O
1
, LDT
2
Figure 1. The recognition binding motif sequence 1 of MAdCAM and
the barbituric acid 2.
O
O
NH2
NH
NH2
KOCN
EtO
OEt
O
ACOH/H O
NaOEt, EtOH
2
Δ
Table 1. Potency of inhibitors
3
4
O
O
O
N
O
H
N
R2
HN
H
HN
O
N
O
O
N
O
O
O
R3
AcOH
2
R
3
Compound
R
a4b7/MAdCAM
IC50 (lM)
2
2
6
Ph
Cynamoyl
2-Furyl
Ph
Ph
Ph
2.2
8.6
1.3
4.1
0.9
10
5
Scheme 1. General schematic of the synthesis of barbituric acids.
7
8
3
N-CH -3-indoyl Ph
9
3-Indoyl
2-Indoyl
3-Indoyl
3-Indoyl
3-Indoyl
Phenyl
Ph
Ph
O
10
11
R1
t-Butyl
>20
0.26
0.06
1.0
1.3
N
R2
1
1
1
1
2
3
4
5
2-Biphenyl
Naphthyl
4-ChloroPh
Ph
O
N
R3
O
3
4-CF Ph
1
2
Figure 2. Three sites of diversity (R , R , and R ) were investigated.
3
16
Ph
3,5-DichloroPh 1.6

G. C. Harriman et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2509–2512
2511
Table 2. Potency of certain 3-indoyl substituted inhibitors
11. Shroff, H.; Schwender, C.; Baxter, Brookfield A.; Payne,
L.; Cochran, N.; Gallant, D.; Briskin, M. Bioorg. Med.
Chem. Lett. 1998, 8, 1601.
O
R1
N
1
2. Shroff, H.; Schwender, C.; Dottavio, D.; Yang, L.;
Briskin, M. Bioorg. Med. Chem. Lett. 1996, 6, 2495.
3. Dubree, N.; Artis, D.; Castanedo, G.; Marsters, J.;
Sutherlin, D.; Caris, L.; Clark, K.; Keating, S.; Beresini,
M.; Chiu, H.; Fong, S.; Lowman, H.; Skelton, N.;
Jackson, D. J. Med. Chem. 2002, 45, 3451, cyclic
peptide.
N
R2
O
N
O
1
1
R
2
R
Compound
a4b7/MAdCAM IC50 (lM)
1
4. Boer, J.; Gottschling, D.; Schuster, A.; Semmrich, M.;
Holzmann, B.; Kessler, H. J. Med. Chem. 2001, 44, 2586.
5. Georgette, M. C.; Fredrick, C. S.; Nathan, J. P. D.; John,
B. N.; Caris, L.; Clark, K.; Keating, S.; Beresini, M.; Chiu,
H.; Fong, S.; Marsters, J.; Jacksona, D.; Sutherlina, D.
Bioorg. Med. Chem. Lett. 2002, 12, 2913.
13
17
18
H
H
H
0.06
0.18
5.5
CH
H
3
1
CH
3
1
6. Harriman, G.; Schwender, C.; Gallant, D.; Cochran, N.;
Briskin, M. Bioorg. Med. Chem. Lett. 2000, 10, 1497.
7. Lawson, E. C.; Kinney, W. A.; Santulli, R. J.; Fisher, C.
M.; Damiano, B. P.; Maryanoff, B. E. Lett. Drug Des.
Discov. 2005, 2, 601.
Table 3. Selectivity of certain potent inhibitors a4b1/VCAM
1
Compound
a4b7/MAdCAM
IC50 (lM)
a4b1/VCAM
IC50 (lM)
a5b1/FN
IC50 (lM)
7
1.3
>20
>20
>20
>20
>20
>20
1
1
8. Jackson, D. Curr. Pharm. Des. 2002, 8, 99.
1
1
2
3
0.26
0.06
9. Dyatkin, A. B.; Gong, Y.; Miskowski, T. A.; Kimball, E.
S.; Prouty, S. M.; Fisher, M. C.; Santulli, R. J.; Schneider,
C. R.; Wallace, N. H.; Hornby, P. J.; Diamond, C.;
Kinney, W. A.; Maryanoff, B. E.; Damiano, B. P.; He, W.
Bioorg. Med. Chem. 2005, 13, 6693, phe analogs.
0. Castanedo, G. M.; Sailes, F. C.; Dubree, N. J. P.;
Nicholas, J. B.; Caris, L.; Clark, K.; Keating, S. M.;
Beresini, M. H.; Chiu, H.; Fong, S.; Marsters, J. C.;
Jackson, D. Y.; Sutherlin, D. P. Bioorg. Med. Chem. Lett.
2
The pharmacokinetic properties in mouse of compound
3 were examined, and it showed an oral bioavailability
of 62%, a t1/2 of 4.1 h, a clearance of 71 mL/min/kg, a
1
2
002, 12, 2913.
volume of distribution of 24.8 L/kg, and an AUC of
1
2
1. Gong, Y.; Barbay, J. K.; Dyatkin, A. B.; Miskowski, T.
A.; Kimball, E. S.; Prouty, S. M.; Fisher, E. C.;
Santulli, R. J.; Schneider, C. R.; Wallace, N. H.;
Ballentine, S. A.; Hageman, W. E.; Masucci, J. A.;
Maryanoff, B. E.; Damiano, B. P.; Andrade-Gordon, P.;
Hlasta, D. J.; Hornby, P. J.; He, W. J. Med. Chem.
70 mg min/mL. The pharmacokinetics of this molecule
was not suitable for its evaluation in the DSS (dextran
sulfate sodium) derived mouse model of inflammatory
3
1
bowel disease. Further optimization of this chemical
series is ongoing and may result in molecules with more
suitable exposure profiles to enable a proof of concept
study with this class of molecules.
2
22. Carson, K.; Schwender, C.; Shroff, H.; Cochran, N.;
006, 49, 3402.
Gallant, D.; Briskin, M. Bioorg. Med. Chem. Lett. 1997, 7,
11.
7
2
3. Locardi, E.; Boer, J.; Modlinger, A.; Schuster, A.;
Holzmann, B.; Kessler, H. J. Med. Chem. 2003, 46, 5752.
4. Engelhardt, B.; Briskin, M. J. Eur. J. Immunol. 2005, 35,
2268.
References and notes
2
1
2
3
. Butcher, E. C. Cell 1991, 67, 1033.
. Springer, T. A. Cell 1994, 65, 301.
. Brezinschek, R. I.; Brezinschek, H.-P.; Lazarovits, A. I.;
Lipsky, P. E.; Oppenheimer-Marks, N. Am. J. Pathol.
25. Yogeeswari, P.; Sriram, D.; Brahmandam, A.; Sridharan,
I.; Thirumurugan, R.; Stables, J. P. Med. Chem. Res. 2003,
12, 57.
1
996, 149, 1651.
26. Laudien, R.; Mitzner, R. J. Chem. Soc., PT 2 2001, 11,
2226.
27. Oguz, S. F.; Dogan, I. Tetrahedron: Asymmetry 2003, 14,
1857.
4
. Berlin, C.; Berg, E. L.; Briskin, M. J.; Andrew, D.;
Kilshaw, P. J.; Holzmann, B.; Weissman, I. L.; Hainann,
A.; Butcher, E. C. Cell 1993, 74.
5
. Bargatze, R. F.; Jutila, M. A.; Butcher, E. C. Immunity
1
28. Jursic, B. S.; Neumann, D. M. Tetrahedron Lett. 2001, 42,
8435.
995, 3, 99.
6
. Briskin, M.; Winsor-Hines, D.; Shyan, A.; Cochran, N.;
Bloom, S.; Wilson, J.; Butcher, E. C.; Kassam, N.;
Mackay, C. R.; Newman, W.; Ringler, D. J. Am. J.
Pathol. 1997, 151, 97.
29. Naya, S.; Nitta, M. Tetrahedron 2003, 59, 3709.
30. To evaluate these compounds, a cell adhesion assay
involving the B-cell lymphoma RPMI 8866 cells, and
soluble human MAdCAM-IG chimera was used in a 96-
well plate format. RPMI 8866 cells have previously been
shown to bind avidly to MAdCAM. RPMI 8866 cells were
fluorescently labeled by preincubation with BCECF stain
(Molecular Probes), washed, and resuspended in assay
7
. Picarella, D.; Hurlbut, P.; Rottman, J.; Shi, X.; Butcher,
E.; Ringler, D. J. Immunol. 1997, 158, 2099.
8
. Hesterberg, P. E.; Winsor-Hines, D.; Briskin, M. J.;
Soler-Ferran, D.; Merrill, C.; Mackay, C. R.; New-
man, W.; Ringler, D. J. Gastroenterology 1996, 111,
2
+
buffer containing 2% fetal calf serum and 2 mM Mg
.
1
373.
. Briskin, M. J.; McEvoy, L. M.; Butcher, E. C. Nature
993, 363, 461.
0. Briskin, M. J.; Rott, L.; Butcher, E. C. J. Immunol. 1996,
56, 719.
Compounds were tested in HBSS/2% FCS/25 mM Hepes
buffer at 2.5 106 cells per mL. The typical assay consisted
of a final volume of 200 mL containing 50 mL of cells at
1.25 10 cells per well. Adhesion assays for MAdCAM
were washed on an automatic plate washer using a buffer
9
1
5
1
1

2512
G. C. Harriman et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2509–2512
consisting of 50 mM Tris/150 mM NaCl/2 mM MnCl
2
,
presence and absence of an inhibitor and IC50 values were
determined using Kaleidagraph (Adelbeck Software) and
are reported as an average of several determinations.
31. Cooper, H. S.; Murthy, S. N. S.; Shah, R. S.; Sedergran,
D. J. Lab. Invest. 1993, 69, 238.
pH 7.2, in a wash volume of 500 mL for two wash cycles.
Assays were then read on an Idexx fluorescent plate reader
at 485/535 nm. Inhibition was determined by the number
of cells adhering to the MAdCAM lined plates in the