Design, synthesis, and biological evaluation of α4β1 integrin antagonists based on β-D-mannose as rigid scaffold

Article abstract of DOI:10.1002/1521-3773(20011015)40:20<3870::AID-ANIE3870>3.0.CO;2-X

A cyclic peptide role model was used for the design and synthesis of a new class of biologically active and α₄-selective integrin antagonists (e.g. 1) based on β-D-mannose. These carbohydrate-based peptidomimetics were synthesized to include th

Full text of DOI:10.1002/1521-3773(20011015)40:20<3870::AID-ANIE3870>3.0.CO;2-X

COMMUNICATIONS
reaction mixture was stirred for 6 h. After the solid material had been
filtered off, the filtrate was condensed to 5 mL, to which excess diethyl
ether was added to obtain the solid product. Slow evaporation of the
methanol solution gave crystals of 2 in 81.3% yield. M.p. 1478C (decomp).
1H NMR (500 MHz, [D7]DMF, TMS): d ˆ 0.95 (s, 3H), 1.02 (s, 3H), 2.19
(br, 4H), 2.96 (s, 3H), 6.84 ± 7.18 (m, 4H), 7.30 (t, 2H, J ˆ 7.3 Hz), 7.45 (t,
2H, J ˆ 7.4 Hz), 7.91 (d, 2H, J ˆ 7.5 Hz), 8.19 (d, 2H, J ˆ 7.8 Hz); 13C NMR
(125.76 MHz, [D7]DMF, TMS): d ˆ 23.4, 23.6, 35.7, 50.3, 55.8 (OCH3),
drick, University of Göttingen, Germany, 1997. Crystallographic data
(excluding structure factors) for the structures reported in this paper
have been deposited with the Cambridge Crystallographic Data
Centre as supplementary publication nos. CCDC-163484 (2) and
CCDC-163485 (3). Copies of the data can be obtained free of charge
on application to CCDC, 12 Union Road, Cambridge CB21EZ, UK
(fax: (‡44)1223-336-033; e-mail: deposit@ccdc.cam.ac.uk).
[18] S. V. Lindeman, D. Kosynkin, J. K. Kochi, J. Am. Chem. Soc. 1998, 120,
13268.
ˆ
ˆ
120.1, 126.2, 127.8, 129.9, 132.8 (C C), 136.0 (C C), 137.0, 141.1, 170.9
ˆ
(C O); IR (KBr): nÄ ˆ 1666, 1652, 1634 (n(COO)asym); 1324
[19] C. F. J. Barnard, J. F. Vollano, P. A. Chaloner, S. Z. Dewa, Inorg.
Chem. 1996, 35, 3280.
1
(n(COO)sym) cm
; elemental analysis (C, H, N) gave erratic results
presumably due to the easy evaporation of solvate methanol molecules.
[20] G. C. Pimentel, A. L. McClellan, The Hydrogen Bond, Freeman, San
Francisco, 1960, p. 202.
[21] R. T. Morrison, R. N. Boyd, Organic Chemistry, 3rd ed., Allyn and
Bacon, Boston, 1973, p. 145.
3: Compound 2 (1.0 mmol) was dissolved in acetic anhydride (20 mL), and
the resulting solution was then stirred for 3 h. After the solid residue had
been filtered off, the filtrate was evaporated to dryness. Slow evaporation
of the DMF solution gave crystals of 3 in 80.3% yield. M.p. 1688C
1
(decomp). H NMR (500 MHz, [D7]DMF, TMS): d ˆ 0.84 (s, 3H), 0.92 (s,
3H), 0.99 (s, 3H), 1.77 ± 1.93 (m, 2H), 2.39 ± 2.52 (m, 2H), 2.58 (s, 3H,
3JPt,H ˆ 27.10 Hz), 7.16 ± 7.38 (br, 2H), 7.25 (t, 2H, J ˆ 7.6 Hz), 7.39 (t, 2H,
J ˆ 7.6 Hz), 7.84 (d, 2H, J ˆ 7.5 Hz), 7.98 ± 8.22 (br, 2H), 8.15 (d, 2H, J ˆ
7.8 Hz); 13C NMR (125.76 MHz, [D7]DMF, TMS): d ˆ 22.2, 23.0, 24.8
ˆ
(OOCCH3), 35.8, 50.2, 57.3 (OCH3), 120.0, 126.5, 127.7, 129.9, 133.9 (C C),
ˆ
ˆ
ˆ
134.6 (C C), 137.2, 141.2, 170.6 (C O), 180.2 (C O); IR (KBr): nÄ ˆ 1666,
1601 (n(COO)asym); 1302 (n(COO)sym) cm 1; elemental analysis calcd (%)
for C24H28N2O7Pt ´ C3H7NO: C 44.75, H 4.87, N 5.80; found: C 44.60, H 4.85,
N 5.88.
Received: May 17, 2001 [Z17129]
Design, Synthesis, and Biological Evaluation of
a4b1 Integrin Antagonists Based on
b-d-Mannose as Rigid Scaffold
[1] G. R. Desiraju, Angew. Chem. 1995, 107, 2541; Angew. Chem. Int. Ed.
Engl. 1995, 34, 2311.
[2] J. Frey, Z. Rappoport, J. Org. Chem. 1997, 62, 8372.
[3] T. N. Parac, D. L. Caulder, K. N. Raymond, J. Am. Chem. Soc. 1998,
120, 8003.
[4] K. Biradha, M. J. Zaworotko, J. Am. Chem. Soc. 1998, 120, 6431.
[5] O.-S. Jung, S. H. Park, Y.-A Lee, Y. Cho, K. M. Kim, S. Lee, H. K.
Chae, Y. S. Sohn, Inorg. Chem. 1996, 35, 6899.
[6] C. Peter, X. Daura, W. F. Gunsteren, J. Am. Chem. Soc. 2000, 122,
7461.
[7] K. Doclo, U. Rothlisberger, J. Phys. Chem. A 2000, 104, 6464.
[8] E. Yashima, K. Maeda, Y. Okamoto, Nature 1999, 399, 449.
[9] R. B. Hannak, G. Färber, R. Kornat, B. Kräutler, J. Am. Chem. Soc.
1997, 119, 2313.
[10] O.-S. Jung, Y. J. Kim, Y.-A Lee, J. K. Park, H. K. Chae, J. Am. Chem.
Soc. 2000, 122, 9921.
[11] O.-S. Jung, Y.-A Lee, S. H. Park, K. H. Yoo, Bull. Chem. Soc. Jpn.
1999, 72, 2091.
[12] Y. S. Sohn, K. M. Kim, S.-J. Kang, O.-S. Jung, Inorg. Chem. 1996, 35,
4274.
[13] K.-T. Aye, L. Gelmini, N. C. Payne, J. J. Vittal, R. J. Puddephatt, J.
Am. Chem. Soc. 1990, 112, 2464.
Jürgen Boer, Dirk Gottschling, Anja Schuster,
Bernhard Holzmann, and Horst Kessler*
The tuning of protein ± protein interactions by small non-
peptidic molecules remains one of the great challenges in
medicinal chemistry. Although already proposed in 1980 by
Farmer,[1] only a few successful examples on the synthesis of
peptidomimetics based on rigid scaffolds such as cyclohexane
and pyranose sugars have been reported so far.[2] Starting
from the b-d-mannose scaffold we developed peptidomimet-
ics in a rational combinatorial approach focusing on the
interaction of the a4b1 and a4b7 integrins with their ligands.
The basis of this research were cyclic hexapeptides as potent
and selective a4b7 integrin antagonists recently developed by
our group using the ªspatial screeningº procedure.[3]
[14] S. O. Dunham, R. D. Larsen, E. H. Abbott, Inorg. Chem. 1993, 32,
2049.
[15] U. Bierbach, T. W. Hambley, J. D. Roberts, N. Farrell, Inorg. Chem.
1996, 35, 4865.
a4b1 and a4b7 integrins play an important role in numerous
inflammatory and autoimmune disorders.[4] The most impor-
tant biological ligands for these a4 integrins are fibronectin
[16] G. Bandoli, P. A. Caputo, F. P. Intini, M. F. Sivo, G. Natile, J. Am.
Chem. Soc. 1997, 119, 10370.
[*] Prof. Dr. H. Kessler, Dr. J. Boer, D. Gottschling
Institut für Organische Chemie und Biochemie
Technische Universität München
[17] Crystal data for 2: C22H26N2O6Pt ´ 2CH3OH: monoclinic, P21/n, a ˆ
10.371(3), b ˆ 10.857(5), c ˆ 23.417(8) Š, b ˆ 100.71(3)8, Vˆ
2591(2) Š3, 1calcd ˆ 1.727 Mgm 3, F(000) ˆ 1336, l ˆ 0.71073 Š, m ˆ
Lichtenbergstrasse 4, 85747 Garching (Germany)
Fax : (‡49)89-289-13210
1
5.465 mm
,
crystal size 0.15  0.20  0.30 mm, Z ˆ 4, R(wR2) ˆ
0.0787 (0.1932) on 2313 unique reflections with I > 2s(I), GOF ˆ
E-mail: kessler@ch.tum.de
1.102, 316 parameters refined. Crystal data for 3: C24H24N2O7Pt ´
A. Schuster, Prof. Dr. B. Holzmann
Institut für Medizinische Mikrobiologie, Immunologie und Hygiene
Technische Universität München
Å
C3H7NO: triclinic, P1, a ˆ 9.982(1), b ˆ 10.267(2), c ˆ 14.633(2) Š,
a ˆ 76.69(1), b ˆ 77.179(9), g ˆ 76.411(1)8, Vˆ 1396.1(3) Š3, 1calcd
ˆ
1.724 Mgm 3, F(000) ˆ 720, l ˆ 0.71073 Š, m ˆ 5.079 mm 1, crystal
size 0.30  0.45  0.30 mm, Z ˆ 2, R(wR2) ˆ 0.0426 (0.1040) on 4508
unique reflections with I > 2s(I), GOF ˆ 1.109, 352 parameters
refined. Programs used: SHELXS-97 and SHELXL-97: G. M. Shel-
Trogerstrasse 4a, 81675 München (Germany)
Supporting information for this article is available on the WWW under
http://www.angewandte.com or from the author.
3870
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4020-3870 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2001, 40, No. 20

COMMUNICATIONS
(Fn), vascular cell adhesion molecule-1 (VCAM-1), and
mucosal addressin cell adhesion molecule-1 (MAdCAM-1),
the latter being an exclusive ligand for a4b7 integrin under
physiological conditions.[4, 5]
propylene, carboxylmethylene, and (2R)-hydroxypropylene
groups of the mannose derivative 11c with the LDT side
chains of the potent and selective cyclo(-Leu-Asp-Thr-Ala-d-
Pro-Phe-) P2 revealed a good match of these key pharmaco-
phoric groups (Figure 2). An energy minimized conformation
of compound 11c was used for the superposition representing
one of the accessible structures in solution.
The cyclic hexapeptide cyclo(-Leu-Asp-Thr-Ala-d-Pro-
Ala-) P1 and several related peptides with the general
formula cyclo(-Leu-Asp-Thr-Xaa-d-Pro-Xbb-), which include
the bioactive Leu-Asp-Thr (LDT) motif
of the MAdCAM-1 ligand,[6] were re-
cently developed in our group as potent
and selective inhibitors of the MAd-
CAM-1/a4b7 ligand/integrin interac-
tion.[3] The bioactive conformation of
these constrained cyclic peptides con-
sists of two b-turns with the d-proline in
the i ‡ 1 position of a bII'-turn and
aspartic acid in the i ‡ 1 position of a bI-
and/or bII-turn.[7]
The elucidation of the bioactive con-
formation of these peptides led us to the
replacement of the whole peptidic back-
bone by a sugar scaffold which should
present the pharmacophoric LDT side
chain mimetics in similar spatial orien-
tation. A successful implementation of
Farmerꢁs approach[1] would then imply
that the amide backbone is not neces-
sarily involved in receptor binding while
all essential functional groups are main-
tained in their active conformation by
the scaffold.
Figure 2. Stereoview of the superposition between an energy minimized structure of 11c and the
structure of P2 as determined by NMR spectroscopy.
The orientation of the LDT side
chain mimetics attached to a b-d-man-
nose core at positions 6, 1, and 2 (all facing upwards)
resembles the configuration of the pharmacophoric side
chains in our lead peptides (Figure 1).[3] The free hydroxyl
groups at positions 3 and 4 of the carbohydrate scaffold were
modified to methyl ethers in order to introduce further
hydrophobicity.
Molecular modeling studies (InsightII, DISCOVER,
CVFF) confirmed that b-d-mannopyranose is indeed a proper
scaffold as the LDT-mimetic sequence is presented in the
bioactive arrangement. The superposition of the 2-methyl-
In Figure 2, the leucine, aspartic acid, and threonine side
chains and the respective mimetics attached to the mannose
core are highlighted using the stick model. The peptidic and
sugar backbone are given as thin lines. The relative distance
between the aspartic acid and threonine pharmacophores in
P2 (Cb Cb ˆ 4.60 Š), which is controlled by the positioning of
these two amino acids in the i ‡ 1 and i ‡ 2 position of the b-
turn, is being maintained in the mannose derivative (4.56 Š).
The leucine side chain mimetic at the primary hydroxyl group
of the mannose core exhibits high flexibility as in its peptidic
precursor and it is therefore likely to mediate the binding to
the receptor by an induced fit mechanism.
Based upon our model a small biased library of peptidomi-
metics with b-d-mannose as the rigid core was synthesized.
The anchoring of aspartic as well as glutamic acid-type side
chains at the hydroxyl group in position 1 and the attachment
of the serine-type side chain and its homologue together with
the use of a racemic threonine mimetic at position 2 led to
eight different mannose derivatives.
The synthetic protocol is shown in Scheme 1. The desired
products 11a ± f were synthesized in eleven steps starting from
ethyl thio-a-d-mannopyranoside (1). The key compound in
our synthetic strategy was the fully protected mannose
derivative 4.[8] The orthogonal protecting groups of compound
4 were successively cleaved and replaced by the pharmaco-
phoric side chains. Starting with desilylation of the primary
COOH
HOOC
O
O
H
O
O
N
N
H
O
i+1
NH
HN
O
O
O
O
H
N
O
N
O
O
HO
HO
P1
11c
Figure 1. Cyclic peptide P1 and carbohydrate-based peptidomimetic 11c
(see also Scheme 1). In the peptide, the pharmacophoric LDT motif is
localized within a b-turn with the aspartic acid in the i ‡ 1 position. The
carbohydrate scaffold presents the essential pharmacophoric groups in the
same relative orientation as the lead peptide.
Angew. Chem. Int. Ed. 2001, 40, No. 20
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4020-3871 $ 17.50+.50/0
3871

COMMUNICATIONS
OR1
afforded the mannose derivative 8 in 87% yield. Compound 8
was hydrolyzed after activation with N-bromosuccinimide/H
‡
OR2
1 R1, R2, R3 = H
a
b
R3O
O
2 R1, R2 = H, R3 = CDA-protected
and subsequently dissolved in thionyl chloride leading to the
corresponding mannosyl chloride. Coupling of the activated
mannose derivative with glycolaldehyde dimethyl acetal and
3-hydroxypropionaldehyde diethyl acetal in CH2Cl2 using
solid silver carbonate as the halogenophil led to compounds
9a and 9b as the required b-d-mannoglycosides in 72% and
62% yields, respectively.[9] After cleavage of the remaining
benzyl ether protecting group at position 2, 2-benzyl-3-
bromopropyl ether, and 2-benzyl tosylpropyl ether were
introduced by ether formation using potassium bistrimethyl-
silylamide as base. The products 10a ± f were isolated in yields
between 58% and 60%. The reaction of the mannose
derivative 9a and 9b with racemic 2-benzyl tosylpropyl ether
resulted in two pairs of diastereomeric products 10c and 10 f
as a 1/1 mixtures. The acetal protected aldehyde functions at
the anomeric position of compounds 10a ± f serve as precur-
sors of carboxylic acid groups. This strategy was chosen to
avoid by-products during the ether formation as a result of the
Ca acidity of the carboxylic acid derivatives. After cleavage of
the acetals the free aldehydes were immediately oxidized to
the corresponding carboxylic acids using sodium chlorite and
2-methyl-2-butene as the scavenger.[10] The remaining benzyl
protecting groups of the side chains of position 2 were
removed by hydrogenation with Pd/C under neutral condi-
tions. The final products 11a ± f were purified by HPLC.
Thereby, the two diastereomeric compounds 11 f could be
separated yielding products 11 f1 and 11 f2. However, com-
pounds 11c1 and 11c2 were not resolved (see Table 1).
To evaluate the biological activity of the carbohydrate-
based peptidomimetics 11a ± f, we tested whether these
compounds interfered with the binding of integrin a4b1 to its
VCAM-1 ligand or with the binding of integrin a4b7 to the
VCAM-1 and MAdCAM-1 ligand. The integrin ligands
VCAM-1 and MAdCAM-1 were immobilized on tissue
culture plates and adhesion of the lymphoid cell lines 38-b7
(a4b7pos, a4bn1eg† and Jurkat (a4b1pos, a4bn7eg† was analyzed in the
presence or absence of compounds 11a ± f at 5 mm con-
centration as described.[11] The results are summarized in
Table 1. The inhibitory activity of the a4b1 integrin antagonist
cyclo[1,7](C-Q-I-D-S-P-C) was measured at 1 mm as a con-
trol.[12]
R3O
3 R1 = TBDPS, R2 = H, R3 = CDA-protected
SEt
c
OR1
OR2
OMe
4 R1 = TBDPS, R2 = Bn
5 R1 = H, R2 = Bn
d
e
O
O
O
6 R1 = CH2CHMe2, R2 = Bn
OMe
SEt
f, g
OCH2CHMe2
OBn
O
MeO
8
MeO
SEt
h,i
OCH2CHMe2
9a R4 = CH2CH(OMe)2
9b R4 = CH2CH2CH(OEt)2
OBn
O
MeO
MeO
OR4
j
10a R4 = CH2CH(OMe)2, R5 = CH2CH2(OBn)
10b R4 = CH2CH(OMe)2, R5 = CH2CH2CH2(OBn)
10c R4 = CH2CH(OMe)2, R5 = CH2CH(OBn)CH3
10d R4 = CH2CH2CH(OEt)2, R5 = CH2CH2(OBn)
10e R4 = CH2CH2CH(OEt)2, R5 = CH2CH2CH2(OBn)
OCH2CHMe2
OR5
O
MeO
MeO
OR4
10f
R
4 = CH2CH2CH(OEt)2, R5 = CH2CH(OBn)CH3
11a R6 = CH2CH2OH, n = 1
11b R6 = CH2CH2CH2OH, n = 1
11c R6 = CH2CH(OH)CH3, n = 1
11d R6 = CH2CH2OH, n = 2
11e R6 = CH2CH2CH2OH, n = 2
k
OCH2CHMe2
OR6
O
11f
R
6 = CH2CH(OH)CH3, n = 2
MeO
MeO
O(CH2)nCOOH
Scheme 1. a) 1,1,2,2-Tetramethoxycyclohexane[8] (1.5 equiv), CSA, TMOF,
MeOH (43%); b) TBDPSCl (1.5 equiv), imidazole, DMF (85%);
c) KN(SiMe3)2, BnBr (96%); d) TBAF, THF (96%); e) KN(SiMe3)2,
BrCH2CHMe2 (89%); f) 90% TFA (73%); g) KN(SiMe3)2, MeI (87%);
h) NBS, cat. HCl, CH3CN/H2O (91%); i) SOCl2, CH2Cl2 then Ag2CO3,
CH2Cl2, RT, 9a: HOCH2CH(OMe)2 (72%), 9b: HOCH2CH2CH(OEt)2
(62%); j) 10% Pd/C, H2 then KN(SiMe3)2, THF, 10a: BrCH2CH2OBn
(60%), 10b: BrCH2CH2CH2OBn (58%), 10c: BrCH2CH2(OBn)CH3
(60%), 10d: BrCH2CH2OBn (60%), 10e: BrCH2CH2CH2OBn (58%),
10 f: BrCH2CH2(OBn)CH3 (60%); k) 1) 3n HCl, THF, RT; 2) tBuOH,
2-methyl-2-butene, NaClO2 (1 equiv), NaH2PO4, H2O, RT; 3) 10% Pd/C,
H2, MeOH, products 11a (62%), 11b (92%), 11c (60%), 11d (65%), 11a
(71%), 11 f1 (35%), 11 f2 (21%). Bn ˆ Benzyl, CSA ˆ camphorsulfonic
acid, NBS ˆ N-bromosuccinimide, TBAF ˆ tetra-n-butylammonium fluo-
ride, TBDPSCl ˆ tert-butyldiisopropylsilyl chloride, TFA ˆ trifluoroacetic
acid, TMOF ˆ trimethyl orthoformate.
Although our approach started with potent and selective
a4b7 antagonists, no activity was found for this integrin.
However, variation of the pharmacophoric groups led to an
enhanced activity towards the cognate a4b1 integrin. Com-
pound 11a inhibited integrin a4b1-mediated binding of Jurkat
cell to VCAM-1 by 70% (Table 1) at 5 mm concentration.
This compound was also tested at the concentration of
2.5 mm. In this case the cell adhesion was inhibited by 34%.
In contrast, a4b7 integrin-dependent adhesion of 38-b7 cells to
MAdCAM-1 or VCAM-1 was not affected by compound 11a;
this indicates selective antagonist activity for the a4b1 integrin
(Table 1). Compounds 11b ± e, 11 f1, and 11 f2 did not exhibit
significant biological activity in any of the assays investigated.
The side chains at positions 6, 1 and 2 of the active mannose
derivative 11a mimic the LDS peptide sequence. The
inhibitory activity reported in Table 1 can therefore be
alcohol function with tetra-n-butylammonium fluoride
(TBAF), the leucine side chain mimetic was introduced by
ether formation using 1-bromo-2-methylpropane and potas-
sium bistrimethylsilylamide as base. The resulting compound
6 was isolated in 89% yield. Subsequently, cleavage of the
cyclohexane-1,2-diacetal (CDA) protecting group and con-
version of the free trans vicinal diol into methyl ethers
3872
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4020-3872 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2001, 40, No. 20

COMMUNICATIONS
Table 1. Effect of mannose-based peptidomimetics at 5 mm concentration on Jurkat and 38-b7 lymphoma cell binding to immobilized VCAM-1 and
MAdCAM-Ig.
Jurkat (a4b1) VCAM-1
38-b7 (a4b7) VCAM-1
38-b7 (a4b7) MAdCAM-Ig
[%][a]
n[b]
[%]
n
[%]
n
11a
30 Æ 15
66 Æ 3
75 Æ 9
79 Æ 23
93 Æ 10
81 Æ 19
68 Æ 8
n.t.
7
3
9
8
3
78 Æ 9
n.t.[d]
8
104 Æ 26
n.t.
4
11a[c]
11b
89 Æ 5
85 Æ 3
79 Æ 1
85 Æ 8
88 Æ 4
6
6
2
6
3
108 Æ 18
122 Æ 15
130 Æ 10
132 Æ 19
104 Æ 11
126 Æ 26
n.t.
5
3
5
4
4
4
11c1,2[e]
11d
11e
11 f1
11 f2
Ref.[f]
10
5
94 Æ 10
3
9 Æ 6
5
n.t.
[a] Cell adhesion is presented as% of medium control. The data represent the mean values Æ the standard deviation; [b] number of experiments;
[c] adhesion at 2.5 mm concentration of the inhibitor; [d] n.t.: not tested; [e] racemic mixture not resolved; [f] Ref.: reference peptide at 1 mm concentration.
[4] G. Kilger, B. Holzmann, J. Mol. Med. 1995, 73, 347 ± 354.
[5] C. Berlin, E. L. Berg, M. J. Briskin, D. P. Andrew, P. J. Kilshaw, B.
explained by the similarity with the IDS(P) sequence which
represents the key motif in VCAM-1 necessary for binding to
a4b1 integrin.[13] These data demonstrate that inhibition of the
Holzmann, I. L. Weissman, A. Hamann, E. C. Butcher, Cell 1993, 74,
185 ± 195.
VCAM-1/a4b1 integrin interaction does not necessarily re-
quire amide bonds. Based upon this result physiologically
more stable and bioavailable drugs might be accessible, which
overcome the well known restriction of conventional peptide
or peptide-related compounds. Further investigations are
currently in progress in our group.
In summary, we have reported on the design and synthesis
of a biological active sugar derivative which can be considered
as a new lead structure for rational combinatorial develop-
ment of anti-inflammatory drugs. Starting from large proteins
the cell adhesion molecules VCAM-1 and MAdCAM-1 highly
potent cyclic peptides were designed[3] which served as a basis
for the sugar derivatives presented here. This class of
peptidomimetics have all requirements necessary for orally
available drugs. Furthermore, active compound 11a fulfills
Lipinskiꢁs rules[14] for bioavailability (MG ˆ 366.4; logP ˆ
0.60;[15] number of hydrogen donors: 2; number of hydrogen
acceptors: 9). Therefore, it represents a promising starting
platform for lead optimization.
[6] J. L. Viney, S. Joenes, H. H. Chiu, B. Lagrimas, M. E. Renz, L. G.
Presta, D. Jackson, K. J. Hillan, S. Lew, S. Fong, J. Immunol. 1996, 157,
2488 ± 2497.
[7] R. Haubner, D. Finsinger, H. Kessler, Angew. Chem. 1997, 109, 1440 ±
1456; Angew. Chem. Int. Ed. Engl. 1997, 36, 1374 ± 1389; D. F. Mierke,
M. Kurz, H. Kessler, J. Am. Chem. Soc. 1994, 116, 1042 ± 1049.
[8] S. V. Ley, H. W. M. Priepke, S. T. Warriner, Angew. Chem. 1994, 106,
2410 ± 2412; Angew. Chem. Int. Ed. Engl. 1994, 33, 2290; P. Grice, S. V.
Ley, J. Pietruska, H. W. A. Priepke, S. L. Warriner, J. Chem. Soc.
Perkin Trans. 1 1997, 351 ± 364.
[9] H. Paulsen, O. Lockhoff, Chem. Ber. 1981, 114, 3102 ± 3114; P. J.
Garegg, T. Iversen, Carbohydr. Res. 1979, 70, C13 ± C14.
[10] E. Dalcanale, J. Org. Chem. 1986, 51, 567 ± 569.
[11] Cell adhesion assay: 96-well plates were coated with donkey anti-
human IgG (0.5 mg in 100 mL PBS per well) over night at 48C. The
plates were washed with adhesion assay buffer (Clickꢁs RPMI, 1%
BSA, 1.0 mm Ca2‡, 1.0 mm Mg2‡) and incubated with 100 ng per well
MAdCAM-Ig for 30 min at room temperature. After washing with
adhesion assay buffer, the plates were blocked with adhesion assay
buffer for 1 h at 378C. Alternatively, the plates were coated over night
at 48C with 400 ng per well VCAM-1 in 50mm carbonate buffer
(pH 9.4). 38-b7 lymphoma cells or Jurkat lymphoma cells were
labelled for 30 min at 378C with 12 mgmL 1 H33342 dye (Calbiochem,
LA Jolla, CA) in cell adhesion buffer and washed with PBS containing
1
1 mm EDTA and with PBS. The cells (8 Â 105 cellsmL
) were
Received: May 21, 2001 [Z17151]
resuspended in cell adhesion buffer containing 1.0 mm Ca2‡ and
1.0 mm Mg2‡. The mannose derivatives were preincubated with the
1
cells at the final concentration of 2 mgmL for 10 min at 378C.
Subsequently, cells were allowed to adhere for 30 min at 378C and
non-adherent cells were removed by inverse centrifugation for 10 min
at 50 g. Adhesion assays were quantified by fluorimetry using a
Cytofluor 2300 (Millipore, Bedford, MA). BSA ˆ bovine serum
albumin, PBS ˆ phosphate buffered saline (solution).
[12] J. H. Wang, R. B. Pepinsky, T. Stehle, J. H. Liu, M. Karpusas, B.
Browning, L. Osborn, Proc. Natl. Acad. Sci. USA 1995, 92, 5714 ±
5718.
[13] J. M. Clements, P. Newham, M. Shepherd, R. Gilbert, T. J. Dudgeon,
L. A. Needham, R. M. Edwards, L. Berry, A. Brass, M. J. Humphries,
J. Cell Sci. 1994, 107, 2127 ± 2135.
[14] C. A. Lipinski, F. Lombardo, B. W. Dominy, P. J. Feeney, Adv. Drug
Delivery Rev. 1997, 23, 3 ± 25.
[1] P. S. Farmer in Drug Design, Vol. 10 (Ed.: E. J. Ariens), Academic
Press, New York, 1980, pp. 119 ± 143.
[2] R. Hirschmann, K. C. Nicolaou, S. Pietranico, E. M. Leahy, J. Salvino,
B. Arison, B. M. A. Cichy, P. G. Spoors, W. C. Shakespeare, P. A.
Sprengeler, P. Hamley, A. B. Smith III, T. Reisine, K. Raynor, L.
Maechler, C. Donaldson, W. Vale, R. M. Freidinger, M. R. Cascieri,
C. D. Strader, J. Am. Chem. Soc. 1993, 115, 12550 ± 12568; R.
Hirschmann, P. A. Sprengeler, T. Kawasaki, J. W. Leahy, W. C.
Shakespeare, A. B. Smith III, Tetrahedron 1993, 49, 3665 ± 3676;
G. L. Olson, D. R. Bolin, M. P. Bonner, M. Bös, C. M. Cook, D. C.
Fry, B. J. Graves, M. Hatada, D. E. Hill, M. Kahn, V. S. Madison, V. K.
Rusiecki, R. Sarabu, J. Sepinwall, G. P. Vincent, M. E. Voss, J. Med.
Â
Chem. 1993, 36, 3039 ± 3049; P. C. Belanger, C. Dufresne, Can. J.
[15] A. K. Ghose, G. M. Crippen, J. Chem. Inf. Comput. Sci. 1987, 27, 21 ±
35.
Chem. 1986, 64, 1514 ± 1520; R. Hirschmann, J. Hynes, M. A. Cichy-
Knight, R. D. v. Rijn, P. A. Sprengeler, P. G. Spoors, W. C. Shake-
speare, S. Pietranico-Cole, J. Barbosa, J. Liu, W. Yao, S. Rohrer, A. B.
Smith III, J. Med. Chem. 1998, 41, 1382 ± 1391; C. Papageorgiou, R.
Haltiner, C. Bruns, T. J. Petcher, Bioorg. Med. Chem. Lett. 1992, 2,
135 ± 140.
[3] J. Boer, D. Gottschling, A. Schuster, M. Semmrich, B. Holzmann, H.
Kessler, J. Med. Chem. 2001, 44, 2586 ± 2592.
Angew. Chem. Int. Ed. 2001, 40, No. 20
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4020-3873 $ 17.50+.50/0
3873