Synthesis of sialyl LewisX mimics. Modifications of the 6-position of galactose

Article abstract of DOI:10.1016/S0960-894X(00)00692-2

Seven sLe^x mimics where the -CH₂OH group of the galactose moiety is replaced by -CH₂NH₃/⁺, -CH₂NHAc, -CH₂NHBz, -CH₂OSO₃Na, -COONa and -CONH₂ have been prepared and tested for their binding affinity to E-selectin.

Full text of DOI:10.1016/S0960-894X(00)00692-2

Bioorganic & Medicinal Chemistry Letters 11 (2001) 459±462
Synthesis of Sialyl Lewisx Mimics.
Modi®cations of the 6-Position of Galactose
Rolf Bantelia,* and Beat Ernstb
aNovartis Pharma AG, CH-4002 Basel, Switzerland
bInstitute of Molecular Pharmacy, University of Basel, CH-4051 Basel, Switzerland
Received 16 August 2000; accepted 30 November 2000
AbstractÐSeven sLex mimics where the -CH2OH group of the galactose moiety is replaced by -CH2NH3+, -CH2NHAc,
-CH2NHBz, -CH2OSO3Na, -COONa and -CONH2 have been prepared and tested for their binding anity to E-selectin. # 2001
Elsevier Science Ltd. All rights reserved.
The rolling of leukocytes on endothelial cells of blood
vessels is the initial stage in the recruitment of leuko-
cytes to in¯amed tissue.1 This process is mediated by the
interaction of complex carbohydrate ligands on leuko-
cytes with the carbohydrate binding receptors E- and P-
selectin on the endothelial cells. The minimal structure
recognized by both selectins is the tetrasaccharide sialyl
Lewisx (1, sLex, Fig. 1).2
interest to block this process by antagonizing the
binding of sLex to E- and P-selectin. In the course of the
search for simpli®ed and more potent E-selectin
antagonists we4,5 and others6 have found that N-ace-
tylglucosamine in sLex (1) can be replaced by (R,R)-1,2-
cyclohexanediol. Concomitant replacement of N-acetyl-
neuraminic acid by l-phenyl lactic acid or l-cyclohexyl
lactic acid led to compounds 25 (IC50=0.35 mM) and 3
(IC50=0.08 mM) which show a 3- and 12-fold increase
in their in vitro biological activity7 compared to the
parent compound 1 (IC50=1 mM)5 (Fig. 1). Literature
reports8a suggest that the 6-hydroxy group of galactose
as well as the fucose hydroxy groups and the sialic acid
carboxy group are a prerequisite for E-selectin binding.
This recognition process is involved in in¯ammatory
diseases, ischemia/reperfusion injury, metastasis and
angiogenesis.3 It is therefore of great pharmacological
In order to investigate whether additional binding can
be gained by selectively modifying the 6-position of
galactose, a series of modi®cations of 2 and 3, respec-
tively, with positively charged (4), negatively charged (7,
8 and 9), neutral (5 and 10) and bulky (6) replacements
for the 6-OH group were prepared (Fig. 2).
In order to prepare the 6-amino-6-deoxygalactose deri-
vatives 4, 5 and 6, the tetrol 115a was regioselectively
alkylated with p-methoxybenzylbromide at the 3-posi-
tion via the intermediate stannylene acetal9 to yield 12
(Scheme 1). Tosylation of the 6-hydroxy group and
azide replacement gave 14. Removal of the PMB group
under oxidative conditions (Ce(NH4)2(NO3)6) yielded
15 which was alkylated at the same position with the
tri¯ate 1610 again via the intermediate stannylene acetal.
Hydrogenation of 17 gave the ammonium salt 4.
Finally, Schotten-Baumann acylation using acetic
anhydride or benzoyl chloride in aqueous base led to the
acetamide 5 and the benzamide 6, respectively.
Figure 1. Mimics of sialyl Lewisx.
*Corresponding author. Tel.: +41-61-324-54-44; fax: +41-61-324-67-
35; e-mail: rolf.baenteli@pharma.novartis.com
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00692-2

460
R. Banteli, B. Ernst / Bioorg. Med. Chem. Lett. 11 (2001) 459±462
È
13
For the synthesis of the 6-O-sulfate 7, the pseudo-
trisaccharide 11 was selectively silylated to 18 and al-
kylated to 19 (Scheme 1). Desilylation led to 20,
sulfation of which was selective for the primary 6-posi-
tion. After hydrogenation of 21 and ion exchange chro-
matography the disodium salt 7 was obtained. The
uronic acid compounds 8 and 9 were prepared by oxi-
dation of the primary 6-hydroxy group of the galactose
moiety of a precursor with a fully assembled carbon
framework. In a ®rst attempt, treatment of the triol 20
with TEMPO11 (2,2,6,6-tetramethyl-piperidin-1-oxyl
radical, which is known to selectively oxidize primary
hydroxy groups in the presence of secondary ones)
failed. In a second attempt, the secondary hydroxy
groups in the 2- and 4-position of 19 were protected by
benzoylation. By desilylation, the primary alcohol 23
(Scheme 2) was obtained. Subsequent oxidation with
Dess±Martin reagent12 gave the aldehyde 24 which
decomposed upon chromatography on silica (!b-elim-
ination of the 4-benzoyl group). Without puri®cation it
was therefore further oxidized using sodiumchlorite
to yield the uronic acid derivative 25. Subsequent
debenzoylation could only be accomplished under harsh
conditions (70 ꢀC, 24 h) suggesting that the 2- and the 4-
positions are sterically hindered. Finally, the desired
compared 8 was obtained by hydrogenation over Pd/C.
Further hydrogenation over Rh/Al2O3 eventually led to
the cyclohexyllactic acid derivative 9.
In order to prepare the target 10, the uronic acid 25
was transformed into the corresponding uronic amide
(! -COOH ! -COCl ! -CONH2). However, all
attempts to subsequently remove the benzoyl protecting
groups were acompanied by b-elimination of the 4-ben-
zoyl group. In an alternative approach benzylation of
the 2- and 4-position of 19 failed due to unreactivity of
the 4-position. Since this position was so unreactive
towards protection it was argued that it might also be
unreactive towards oxidation and that protection
thereof might not be necessary. Compound 2714 was
therefore monoprotected with ClCH2COCl to give 28
(Scheme 3). Desilylation and two-step oxidation was
performed as above to give 31. Conversion of the
sodiumsalt to the acid and reaction with chloro-
enamine15 followed by treatment with NH3/MeOH led
to the amide 32. Deacylation with thiourea16 to give 33
followed by hydrogenation ®nally yielded the urona-
mide 10.
All mimics were inactive (IC50 >1 0 mM ) . 8 The fact that
the acetamide 5, the benzamide 6 and the sulfate 7 are
inactive can be rationalized by the following argument:
Figure 2. Modi®cation of the galactose moiety.
Scheme 1. (a) 1. Bu2SnO, benzene, re¯ux, 5 h; 2. p-CH3O-BnCl (18 equiv), Bu4NBr (1.5 equiv), 60 ꢀC, 6 h, 81%; (b) TsCl (1.2 equiv), CH2Cl2, pyr-
idine, re¯ux, 17 h, 75%; (c) NaN3 (4 equiv), DMF, 60 ꢀC, 46 h, 64%; (d) Ce(NH4)2(NO3)6 (3 equiv), CH3CN/H2O 9:1, 0 ꢀC, 3 h, 81%; (e) 1. Bu2SnO
(1.5 equiv), MeOH, re¯ux, 2 h, evaporation; 2. CsF (5 equiv), 16 (5 equiv), DME, rt, 22 h, 65%; (f) H2, Pd(OH)2, 48 h, 27% and 11% (slightly
impure); (g) 1. Ac2O (10.5 equiv), H2O, 2. adjust pH to 9±10 with NaOH, quant. (h) BzCl (1.1 equiv), NaOH, H2O, rt, 3 h, 85%; (i) TBDPSCl,
imidazole, DMF, 16 h, 87%; (j) 1. Bu2SnO (1.5 equiv), MeOH, re¯ux, 2 h, 2. CsF (5 equiv), 16 (5 equiv), DME, 24 h, 60% (16% recovered starting
ꢀ
.
material); (k) TBAF 1 M in THF (1.1 equiv), THF, AcOH (>1.1 equiv), 7 h, 77%; (l) 1. SO3 py, (1.5 equiv), pyridine, 2 h, 0 C, 1.5 h rt, 2. Dowex
50+, >50%; (m) 1. H2, Pd(OH2)/C, MeOH, 6 h, 2. Dowex 50 Na+
.

R. Banteli, B. Ernst / Bioorg. Med. Chem. Lett. 11 (2001) 459±462
È
461
Scheme 2. (a) BzCl (4.8 equiv), pyridine, rt, 16 h, 87%; (b) TBAF (1.1 equiv), AcOH (ca. 1.3 equiv), THF, rt, 16 h, 95%; (c) Dess±Martin period-
inane (1.2 equiv), CH2Cl2, rt, 2 h, (used crude); (d) NaClO2 (30 equiv), 2-methyl-2-butene, i-PrOH, NaH2PO4, H2O, rt, 16 h, 88% (over two steps);
(e) 1 M NaOH (9 equiv), 40 ꢀC, 18 h, 70 ꢀC, 24 h, 78%; (f) H2, Pd(OH)2/C, dioxane/H2O/AcOH 24:12:1, 18 h, 70%; (g) H2, Rh/Al2O3 5%, dioxane/
H2O 1:1, 4 h, 93%.
Scheme 3. (a) ClCH2COCl, pyridine, CH2Cl2, 10 min, 87%; (b) TBAF, AcOH, THF, 16 h, 65%; (c) Dess±Martin periodinane (1.2 equiv), CH2Cl2,
rt, 2 h, (used crude); (d) NaClO2 (30 equiv), 2-methyl-2-butene, i-PrOH, NaH2PO4, H2O, rt, 16 h, 82% (two steps); (e) 1. Dowex W X 8 H+-form; 2.
(CH3)2CˆCClN(CH3)2 (1.5 equiv), CH2Cl2, 0 ꢀC, 1 h, evaporate to dryness then add CH2Cl2 again; 3. NH3/MeOH, 5 min, 75%; (f) H2NCSNH2 (3
equiv), EtOH, 48 h, 50 ꢀC, 83%; (g) H2, Pd(OH)2/C, dioxane/water 2:1, 3 h, 90%.
2. (a) Patel, T. P.; Goelz, S. E.; Lobb, R. R.; Parekh, R. B.
as the 6-hydroxy is in close contact with the protein
Biochemistry 1994, 33, 14815. (b) Phillips, M. L.; Nudelman,
these substituents might be too bulky thereby prevent-
E.; Gaeta, F. C. A.; Perez, M.; Singhal, A. K.; Hakomori, S.;
ing the close interaction between sLex and E-selectin.
Also, the ammonium salt 4, as well as the uronic acid
Paulson, J. C. Science 1990, 250, 1130.
3. (a) Gallin, J. I. et. al. In¯ammation: Basic Principles and
compounds 8 and 9 which contain replacements of the
6-OH with comparable steric demand as 6-OH itself but
Clinical Correlates, 2nd Edition: Raven Press: New York,
1992, p 407. (b) Giannis, A. Angew. Chem., Int. Ed. Engl.
1994, 33, 178.
with positive or negative charges, proved to be inactive.
Surprisingly, even the uronic amide 10 with an
uncharged 6-OH replacement which remains a hydro-
gen bond donor showed no activity. These results sug-
gest that the 6-hydroxy group of the galactose moiety is
optimally suited for the direct binding to E-selectin.
4. Banteli, R.; Ernst, B. Tetrahedron Lett. 1997, 38, 4059.
5. (a) Kolb, H. C.; Ernst, B. Chem. Eur. J. 1997, 3, 1571. (b)
Kolb, H. C.; Ernst, B. Pure Appl. Chem. 1997, 69, 1879.
6. (a) Bamford, M. J.; Bird, M.; Gore, P. M.; Holmes, D. S.;
Priest, R.; Prodger, J. C.; Saez, V. Bioorg. Med. Chem. Lett.
1996, 6, 239. (b) Prodger, J. C.; Bamford, M. J.; Bird, M.;
Gore, P. M.; Holmes, D. S.; Priest, R.; Saez, V. Bioorg. Med.
Chem. 1996, 4, 793.
7. IC50 values are determined by GlycoTech Corp., Rockville,
Maryland 20850, USA with a standard ELISA assay using an
E-selectin-IgG and biotinylated polymer containing sLea. IC50
values greater than 10 mM are no longer detectable and the
compounds are considered inactive. For more information see
footnote 7 in ref 4.
References and notes
1. (a) Boschelli, H. Drugs Future 1995, 20, 805. (b) Spertini,
O.; Luscinskas, F. W.; Gimbrone, M. A.; Tedder, T. F. J. Exp.
Med. 1992, 175, 1789.

462
R. Banteli, B. Ernst / Bioorg. Med. Chem. Lett. 11 (2001) 459±462
È
8. (a) Stahl, W.; Sprengard, U.; Kretzschmar, G.; Kunz, H.
Angew. Chem. 1994, 106, 2186. (b) Ernst, B.; Banteli, R.;
Kolb, H.; Oehrlein, R. unpublished results.
12. Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4156.
13. Nakahara, Y.; Ogawa, T. Carbohydr. Res. 1990, 200,
363.
9. (a) Kovac, P.; Glaudemans, C. P. J. Carbohydr. Res. 1985,
142, 158. (b) Chowdhury, U. S. Tetrahedron 1996, 52, 12775.
10. Degerbeck, F.; Fransson, B.; Grehn, L.; Ragnarsson, U.
J. Chem. Soc., Perkin Trans. 1 1993, 11.
11. (a) Davis, N. J.; Flitsch, S. L. Tetrahedron Lett. 1993, 34,
1181. (b) Gyorgydeak, Z.; Thiem, J. Carbohydr.Res. 1995, 268, 85.
(c) de Nooy, A. E. J.; Besm er, A. C.Tetrahedron 1995, 51, 8023.
14. 27 was prepared in analogy to 19 using cyclohexyllactic
acid tri¯ate in the alkylation step instead of phenyllactic acid
tri¯ate 16.
15. Ernst, B.; De Mesmaeker, A.; Wagner, B.; Winkler, T.
Tetrahedron Lett. 1990, 31, 6167.
16. Cook, A. F.; Maichuk, D. T. J. Org. Chem. 1970, 35,
1940.