Solid-Phase Synthesis of Complex Oligosaccharides Using A Novel Capping Reagent

Article abstract of DOI:10.1021/jo0354239

Solid-phase-supported oligosaccharide synthesis of a core N-glycan tetrasaccharide and of a trisaccharide containing the Galili antigen is reported. The synthesis is based on a hydroxymethylbenzyl benzoate spacer-linker system attached to the Merrifield r

Full text of DOI:10.1021/jo0354239

Solid -P h a se Syn th esis of Com p lex Oligosa cch a r id es Usin g a Novel
Ca p p in g Rea gen t
Xiangyang Wu and Richard R. Schmidt*
Fachbereich Chemie, Universita¨t Konstanz, Fach M 725, D-78457, Konstanz, Germany
richard.schmidt@uni-konstanz.de
Received September 29, 2003
Solid-phase-supported oligosaccharide synthesis of a core N-glycan tetrasaccharide and of a
trisaccharide containing the Galili antigen is reported. The synthesis is based on a hydroxymeth-
ylbenzyl benzoate spacer-linker system attached to the Merrifield resin, O-Fmoc-protected
O-glycosyl trichloroacetimidates as glycosyl donors, and benzoyl isocyanate as a capping reagent
for low-reactivity hydroxy groups. In this way, the target molecules could be efficiently obtained
with little byproduct formation, and hence final purification was convenient.
In tr od u ction
groups as glycosyl acceptors, a capping step in the
synthetic cycle is required in order to reduce the ac-
cumulation of product mixtures. Thus, particularly n -
1 side product formation will be reduced, hence leading
to greatly simplified purification of the target molecules.
Obviously, such capping methods can also be combined
with monitoring of desired products or undesired byprod-
ucts or with tagging of the target molecule.
Oligosaccharides play an important role in various
biological processes.1-3 Therefore, many innovative meth-
ods for their synthesis have been recently developed.4
Hence, efforts have also been devoted to increase the
efficiency of polymer-support-based approaches,5 which
have become standard for the synthesis of oligopeptides6
and oligonucleotides.7 Progress in this challenging task
would provide several advantages over solution-phase-
based oligosaccharide synthesis: (i) standardized build-
ing blocks will become available, (ii) use of excess building
blocks and/or reagents will increase the reaction yields
and thus the overall efficiency, (iii) synthesis will be
automated and thus become much faster, and (iv) much
less sophisticated purification steps will be required.8
However, with the existing glycosylation procedures,
completion can hardly be reached for all glycosylation
steps. Therefore, similar to solid-phase synthesis of
peptides9 and oligonucleotides10 for unreactive hydroxy
Application of the common capping concept to polymer-
supported oligosaccharide synthesis has so far received
little attention. Seeberger and co-workers11a successfully
utilized the capping and tagging (cap-tag) concept for
the automated synthesis of (1-6)-linked trisaccharides.
For the capping (and tagging) of the primary hydroxy
groups, the sterically demanding R-azidoisobutyric or the
(heptadecafluorodecyl)diisopropylsilyl groups were em-
ployed. Furthermore, Ito and co-workers11b successfully
synthesized a tetrasaccharide employing an attractive
capping procedure based on the use of chloroacetyl
moieties. We would like to introduce in this paper a novel
capping reagent that fulfills the following demands: (i)
the capping reagent must readily react with all types of
hydroxy groups; (ii) the linkage formed between the
capping reagent and the hydroxy group must lead to
nonbasic/nonnucleophilic and nonacidic caps that are
inert to all reactions all along the oligosaccharide con-
struction; (iii) the caps must be compatible with the
* Corresponding author.
(1) Varki, A. Glycobiology 1993, 3, 97.
(2) Dwek, R. A. Chem. Rev. 1996, 96, 683.
(3) Sharon, N.; Lis, H. In GlycosciencessStatus and Perspectives;
Gabius, J ., Gagius, S., Eds.; Chapman and Hall: Weinheim, 1997; p
133.
(4) (a) Schmidt, R. R. Angew. Chem. 1986, 98, 213; Angew. Chem.,
Int. Ed. Engl. 1986, 25, 212. (b) Schmidt, R. R.; Kinzy, W. Adv.
Carbohydr. Chem. Biochem. 1994, 50, 21. (c) Danishefsky, S. J .;
Bilodeau, M. T. Angew. Chem. 1996, 108, 1482; Angew. Chem., Int.
Ed. Engl. 1996, 35, 1380. (d) Garegg, P. J . Adv. Carbohydr. Chem.
Biochem. 1996, 52, 179. (e) Demchenko, A.; Stauch, G.-J . Synlett 1997,
818.
(5) For a recent review, see: (a) Seeberger, P. H.; Haase, W.-C.
Chem. Rev. 2000, 100, 4349-4393. For additional relevant literature,
see: (b) Zhu, T.; Boons, G.-J . Chem. Eur. J . 2001, 7, 2382. (c) Zhu, T.;
Boons, G.-J . J . Am. Chem. Soc. 2000, 122, 10222. (d) Nicolaou, K. C.;
Watanabe, N.; Li, J .; Winssinger, N. Angew. Chem. 1998, 110, 1636;
Angew. Chem., Int. Ed. 1998, 37, 1559. (e) Nicolaou, K. C.; Winssinger,
N.; Pastor, J .; DeRoose, F. J . Am. Chem. Soc. 1997, 119, 449. (f)
Rademann, J .; Geyer, A.; Schmidt, R. R. Angew. Chem. 1998, 110, 1309;
Angew. Chem., Int. Ed. 1998, 37, 1241. (g) Rademann, J .; Schmidt, R.
R. J . Org. Chem. 1997, 62, 3650.
(8) (a) Wu, X.; Grathwohl, M.; Schmidt, R. R. Angew. Chem. 2002,
114, 4664; Angew. Chem., Int. Ed. 2002, 41, 4489. For other benzyl-
type linkers, see: (b) Ito, Y.; Kanie, O.; Ogawa, T. Angew. Chem. 1996,
108, 2691-2693; Angew. Chem., Int. Ed. Engl. 1996, 35, 2510-2512.
(c) Shimizu, H.; Ito, Y.; Kanie, O.; Ogawa, T. Bioorg. Med. Chem. Lett.
1996, 6, 2841-2846. (d) Mehta, S.; Whitfield, D. M. Tetrahedron Lett.
1998, 39, 5907-5910. (e) Cross, G. G.; Whitfield, D. M. Synlett 1998,
487-488. (f) Fukase, K.; Nakai, Y.; Egusa, K.; Porco, J . A., J r.;
Kusumoto, S. Synlett 1999, 1074-1078.
(9) Kates, S. A.; Alberico, F.; Solid-Phase Synthesis: A Pratical
Guide; Marcel Dekker: New York, 2000.
(10) Caruthers, M. H. Science 1985, 230, 281.
(6) Bodzanszky, M. Peptide Chemistry: A Pratical Textbook, 2nd ed.;
Springer-Verlag: Berlin, 1993.
(7) J ung, G.; Beck-Sickinger, A. G. Angew. Chem. 1992, 104, 375;
Angew. Chem., Int. Ed. Engl. 1992, 31, 367.
(11) (a) Palmacci, E. R.; Hewitt, M. C.; Seeberger, P. H. Angew.
Chem. 2001, 113, 4565; Angew. Chem., Int. Ed. 2001, 40, 4433. (b)
Ando, H.; Manabe, S.; Nakahara, Y.; Ito, Y. Angew. Chem. 2001, 113,
4861; Angew. Chem., Int. Ed. 2001, 40, 4725.
10.1021/jo0354239 CCC: $27.50 © 2004 American Chemical Society
Published on Web 02/20/2004
J . Org. Chem. 2004, 69, 1853-1857
1853

Wu and Schmidt
SCHEME 1. Ca p p in g Rea ction w ith Ben zoyl Isocya n a te
SCHEME 2. Retr osyn th esis of Ta r get Cor e
N-Glyca n Tetr a sa cch a r id e 4
permanent and temporary protecting groups used in the
synthesis. Hence, these caps have to be orthogonal to the
temporary protecting group pattern.
Resu lts a n d Discu ssion
The outlined prerequisites for a capping reagent led
us, for instance, to investigate O-benzyl,12 O-(4-methoxy-
benzyl),13 and O-phthalimidomethyl trichloroacetimi-
dates;14 however, incomplete reaction with hydroxy groups
of low reactivity was observed. Reaction, for instance,
with benzoyl chloride in the presence of a base affected
the stability of Fmoc groups. Therefore, highly reactive
isocyanates should be a reasonable choice. Trichloroacetyl
isocyanate15 could satisfy all the requirements for a good
capping reagent, yet the N-trichloroacetylcarbamoyl cap
is not stable under the weakly basic conditions essential
for the cleavage such as phenoxyacetyl (PA) protecting
groups. However, benzoyl isocyanate16 (Scheme 1, 1)
satisfied all conditions for a good capping reagent: (i) 1
reacts readily and quantitatively with sterically less
accessible hydroxy groups under neutral conditions, as
shown for the reaction with 4-O-unprotected glucosamine
derivative 2 that gave 4-O-(N-benzoylcarbamoyl)-protect-
ed 3 within 5 min in quantitative yield. (ii) The N-
benzoylurethane moiety formed is nonacidic/nonnucleo-
philic and nonbasic, and (iii) it is not affected by the
cleavage of Fmoc and PA temporary protecting groups
under weakly basic conditions. The versatility of this
capping reagent is demonstrated in the successful solid-
phase oligosaccharide synthesis (SPOS) of a tetrasaccha-
ride of the glycan core of N-glycopeptides and of a
trisaccharide containing the Galili antigen.
SP OS of a Tetr a sa cch a r id e of th e N-Glyca n Cor e.
Recently, we introduced a highly efficient solid-phase
synthesis of a branched N-glycan hexasaccharide con-
taining the core structure.8 The novel linker and the
temporary protecting group pattern containing different
types of ester linkages permitted highly chemoselective
reactions and release of the product from the polymer
support with a benzyl aglycon moiety that can be
removed by hydrogenolysis. O-Fmoc-protected O-glycosyl
trichloroacetimidates17 as glycosyl donors permitted also
the desired stereocontrol at the anomeric center.4 This
concept was also applied to the synthesis of core N-glycan
tetrasaccharide 4 (Scheme 2)18 in combination with 1 as
a capping reagent. Hence, the retrosynthetic analysis
leads to spacer-linker-polymer support 5 containing the
previously employed (4-hydroxymethyl-phenyl)methanol
unit19 and O-glycosyl trichloroacetimidates 6-8 as gly-
cosyl donors. Because of the lack of a highly â-selective
mannosyl donor,20,21 the required Manâ(1-4)GlcN dis-
accharide donor 7 had to be prepared independently.8,20
Building blocks 6 and 8 are readily available following
previously reported procedures.17d,22
For the synthesis of spacer-linker-polymer support
5, (4-hydroxymethyl-phenyl)methanol (Scheme 3, 9) was
monotritylated to give 10,17e which on reaction with
Merrifield resin containing benzoyl chloride groups 1117e
in pyridine in the presence of 4-(dimethylamino)pyridine
(DMAP) gave ester 12; unreactive acid chloride groups
were quenched by treatment with methanol. Detrityla-
tion of 12 with trifluoroacetic acid (TFA) in dichlo-
romethane afforded polymer 5. The loading could be
easily determined by the amount of trityl cation re-
leased: 0.146 mmol/g of dry resin led to good glycosyla-
tion results; loadings below 0.1 or above 0.2 mmol/g of
dry resin diminished the overall yields of the oligosac-
charide synthesis.
The synthesis of 7 is based on a recently reported direct
â-selective mannosylation procedure20 with 1320h as a
donor having 2,3-O-alkyl-4,6-O-benzylidene protection
and 1420h as an acceptor, furnishing the desired â-disac-
charide 15 in good yield (â:R ) 4:1, 71%) (Scheme 4). For
the introduction of the temporary protecting groups, the
allyl protecting group was first removed under Pd
(12) Wessel, H.-P.; Diversen, T.; Bundle, D. R. J . Chem. Soc., Perkin
Trans. 1 1985, 2247.
(13) Seokchan, K.; Salomon, R. G. Tetrahedron Lett. 1989, 6279.
(14) Ali, I. A. I.; Abdel-Rahman, A. A.-H.; El Ashry, E. S. H.;
Schmidt, R. R. Synthesis 2003, 7, 1065.
(15) Speziale, A. J .; Smith, L. K. J . Org. Chem. 1963, 28, 1805.
(16) Koppel, I.; Koppel. J .; Leito, I.; Pihl, V. J . Chem. Soc., Perkin.
Trans. 2 1993, 655.
(17) (a) Roussel, F.; Knerr, L.; Grathwohl, M.; Schmidt, R. R. Org.
Lett. 2000, 2, 3043. (b) Roussel, F., Knerr, L.; Schmidt, R. R. Eur. J .
Org. Chem. 2001, 2067. (c) Wu, X.; Grathwohl, M.; Schmidt, R. R. Org.
Lett. 2001, 3, 747. (d) Grathwohl, M.; Schmidt, R. R. Synthesis 2001,
15, 2263. (e) Roussel, F.; Takhi, M.; Schmidt, R. R. J . Org. Chem. 2001,
66, 8540.
(18) Chiesa, M. V.; Schmidt, R. R. Eur. J . Org. Chem. 2000, 3541.
(19) Douglas, S. P.; Whitfield, D. M.; Krepinsky, J . J . J . Am. Chem.
Soc. 1995, 117, 2116.
(20) (a) Crich, D.; Sun, S. J . Org. Chem. 1996, 61, 4506-4507. (b)
Crich, D.; Sun, S. J . Org. Chem. 1997, 62, 1198-1199. (c) Crich, D.;
Sun, S. J . Am. Chem. Soc. 1997, 119, 11217-11223. (d) Crich, D.; Sun,
S. J . Am. Chem. Soc. 1998, 120, 435-436. (e) Weingart, R.; Schmidt,
R. R. Tetrahedron Lett. 2000, 41, 8753.
(21) Abdel-Rahman, A. A.-H.; J onke, S.; El Ashry, E. S. H.; Schmidt,
R. R. Angew. Chem. 2002, 114, 3100; Angew. Chem., Int. Ed. 2002,
41, 2710.
(22) Wegmann, R.; Schmidt, R. R. J . Carbohydr. Chem. 1987, 357.
1854 J . Org. Chem., Vol. 69, No. 6, 2004

Solid-Phase Synthesis of Complex Oligosaccharides
presence of dibutylborane triflate24 afforded the desired
6-O-unprotected intermediate, which on treatment with
FmocCl in pyridine gave disaccharide 16 with the two
orthogonal temporary protecting groups. Removal of the
O-silyl group with HF‚pyridine and then reaction with
trichloroacetonitrile in the presence of sodium hydride
as a base furnished the desired glycosyl donor 7 without
affecting the temporary protecting groups; even the base-
sensitive Fmoc group was retained.
SCHEME 3. P r ep a r a tion of Sp a cer -Lin k er a n d
Atta ch m en t to Resin
With the required building blocks at our disposal, the
multistep synthesis of tetrasaccharide 4 was initiated
(Scheme 5). Linker-loaded polystyrene resin 5 was gly-
cosylated with 3 equiv of O-glucosaminyl trichloroace-
timidate 6 under trimethylsilyl triflate (TMSOTf) acti-
vation (0.3 equiv) to furnish the corresponding support-
bound monosaccharide 17 (Scheme 5).
The glycosylation also in the next steps was carried
out twice at -40 °C to ensure high-yielding reaction. The
Fmoc protecting group was selectively removed by treat-
ment of 17 with triethylamine in dichloromethane af-
fording monosaccharide acceptor resin 18. Next, an
elongation of the immobilized monosaccharide 18 was
performed by reaction with 3 equiv of Manâ(1f4)GlcN
disaccharide donor 7 upon activation with TMSOTf (0.35
equiv), thus leading to resin-bound trisaccharide 19;
repetition of the glycosylation reaction twice ensures a
high yield. Because the C-2 phthalimido protecting group
reduces the nucleophilicity of the 4-hydroxy group of
glucosamine, a capping reaction was performed to block
unconsumed acceptor resin 18. Hence, reaction with
highly reactive benzoyl isocyanate in dichloromethane at
room temperature afforded 4-O-protected resin 20 with-
out affecting the Fmoc protecting group. Removal of the
Fmoc group by treatment of resin 19 with Et3N followed
by washing with CH2Cl2/THF (1:1) and then glycosylation
with donor 8 in CH2Cl2 at -20 °C in the presence of
TMSOTf (0.3 equiv) as a catalyst led to the desired
tetrasaccharide resin 21. The supernatant of an analyti-
cal cleavage reaction from 2 to 4 mg of resin 21 can be
analyzed spectroscopically by MALDI-TOF MS, thus
proving the success of the capping strategy (Figure 1 in
Supporting Information). Two signals corresponding to
the MNa+ ion of 22 and 23 as the only detectable peaks
[23 (MNa+: 1879.9); 22 (MNa+: 631.1, 778.2)] are
obtained. Preparative cleavage under the conditions of
analytical cleavage and then O-acetylation with acetic
anhydride furnished after HPLC pure target molecule 4
as ascertained by NMR data (1H NMR: δ 4.72 (d, J 1b,2b
) 7.4 Hz), 4.67 (s, J 1c,2c < 1.0 Hz), 4.79 (s, J 1d,2d < 1.0
Hz)). The total yield was 39% over eight steps, which is
an average yield of 86% per step. In comparison with
previous results8a capping does, as expected, lead to the
same overall yield; however, the number of byproducts
is reduced.
SCHEME 4. Syn th esis of th e Requ ir ed
Ma n â(1f4)GlcN Disa cch a r id e Don or 7
SP OS of a Tr isa cch a r id e Con t a in in g t h e Ga lili
An tigen . A natural IgG antibody, directed against the
GalR(1-3)Gal moiety, the Galili antigen, was found to
be present in large amounts in human serum,25 which
greatly affects organ transplantation. Typically, this
disaccharide is â(1-4)-linked to glucosamine; therefore,
catalysis with tert-butyl alcohol as the nucleophile23 and
then phenoxyacetylation was performed. Reductive open-
ing of the 4,6-O-benzylidene group with borane in the
(23) Bieg, T.; Szeja, W. J . Carbohydr. Chem. 1985, 4, 441.
(24) Liang, L.; Chan, T.-H. Tetrahedron Lett. 1998, 39, 355.
(25) Galili, U.; Clark, M. R.; Shohet, S. B.; Buehler, J .; Macher, B.
A. Proc. Natl. Acad. Sci. U.S.A. 1987, 84, 1369.
J . Org. Chem, Vol. 69, No. 6, 2004 1855

Wu and Schmidt
SCHEME 5. Solid -P h a se Syn th esis of Tetr a sa cch a r id e 4 w ith Ben zoyl Isocya n a te 1 a s a Ca p p in g Rea gen t
SCHEME 6. Retr osyn th esis of Ta r get
trisaccharide 24 (Scheme 6) seemed to be another inter-
Tr isa cch a r id e 24 Con ta in in g th e Ga lr(1-3)Ga l
esting target molecule for the investigation of the utility
Ga lili An tigen
of capping reagent 1 in SPOS.
The retrosynthetic analysis follows the concept for the
synthesis of target molecule 4, thus leading to spacer-
linker-polymer support 5 and to known glycosyl donors
6, 25, and 26.17d,22 Thus, resin 18 (Schemes 5 and 7) was
common to both syntheses. Glycosylation of 18 with 3
equiv of O-galactosyl trichloroacetimidate 25 in the
presence of TMSOTf (0.35 equiv) at -40 °C furnished the
corresponding support-bound disaccharide 27 (Scheme 7),
and repetition of the glycosylation reaction twice ensures
a high yield; capping reaction with 1 was carried out in
CH2Cl2 at room temperature, affording resin 20.
Removal of the Fmoc group by treatment of resin 27
with Et3N and then washing with CH2Cl2/THF (1:1) and
1856 J . Org. Chem., Vol. 69, No. 6, 2004

Solid-Phase Synthesis of Complex Oligosaccharides
SCHEME 7. Solid -P h a se Syn th esis of Tr isa cch a r id e 24 w ith Ben zoyl Isocya n a te 1 a s a Ca p p in g Rea gen t
glycosylation with donor 26 in a mixture solvent of Et2O/
CH2Cl2 (1:4) at -20 °C in the presence of TMSOTf (0.3
equiv) as a catalyst led to the desired trisaccharide resin
28. Therefore, preparative cleavage of the product from
the resin 28 and an O-acetylation step led, after HPLC,
to pure target molecule 24 (the R/â ratio of the galacto-
sylation is 8:1) as ascertained by NMR data (1H NMR:
δ 4.54 (d, J 1b,2b ) 8.2 Hz, 1b-H), 5.03 (d, J 1c,2c ) 3.2 Hz,
1c-H), 5.11 (d, J 1a,2a ) 8.3 Hz, 1a-H)). The total yield
was 25% over eight steps, which is an average yield of
80% per step. MALDI-TOF-MS analysis of the final
product showed two signals corresponding to the MNa+
ion of 24 and 29 as practically the only detectable peaks
[ 24 (MNa+: 1489.7); 29 (MNa+: 716.3) ] (Figure 2 in
Supporting Information).
capping reagent could be performed. This capping re-
agent fulfills the demands, i.e., complete reaction with
unconsumed hydroxy groups in difficult glycosylation
reactions, inertness to all reactions along the construction
of the oligosaccharide on solid support, and compatibility
with the protecting groups and the linker system em-
ployed. Thus, accumulation of products stemming from
incomplete capping is avoided, which makes purification
of final products convenient. Application of this novel
methodology to the solid-phase synthesis of more complex
and branched oligosaccharides is currently under way.
Ack n ow led gm en t. This work was supported by the
Deutsche Forschungsgemeinschaft, the Fonds der Che-
mischen Industrie, and the European Community (Grant
No. ARPN-CT-2000-00001/GLYCOTRAIN).
Con clu sion
Su p p or tin g In for m a tion Ava ila ble: Spectral data for all
new compounds and Experimental Section. This material is
available free of charge via the Internet at http://pubs.acs.org.
In summary, efficient solid-phase synthesis of a core
N-glycan tetrasaccharide and of a trisaccharide contain-
ing the Galili antigen with benzoyl isocyanate as a
J O0354239
J . Org. Chem, Vol. 69, No. 6, 2004 1857