Convergent synthesis of N-linked glycopeptides on a solid support

Article abstract of DOI:10.1021/ja9740071

Solid-supported synthesis can be conducted to produce a variety of glycopeptides in good overall yields. The carbohydrates are formed by the glycal assembly method. The polymer-bound construct terminates in a glycal. The terminal double bond can be functi

Full text of DOI:10.1021/ja9740071

J. Am. Chem. Soc. 1998, 120, 3915-3927
3915
Convergent Synthesis of N-Linked Glycopeptides on a Solid Support
J. Y. Roberge,‡ X. Beebe,‡ and S. J. Danishefsky‡,§
Contribution from Laboratory for Bioorganic Chemistry, Sloan-Kettering Institute for Cancer Research,
1275 York AVenue, Box 106, New York, New York 10021 and Department of Chemistry, Columbia
UniVersity, New York, New York 10027
ReceiVed NoVember 24, 1997
Abstract: Solid-supported synthesis can be conducted to produce a variety of glycopeptides in good overall
yields. The carbohydrates are formed by the glycal assembly method. The polymer-bound construct terminates
in a glycal. The terminal double bond can be functionalized to provide a C2-N-acetyl glucosamine linkage
with an amino group in the anomeric position. The latter can be coupled, in a convergent manner, to the
γ-carboxyl group of an aspartyl residue on a preformed peptide. Iodosulfonamidation of the polymer-bound
glucal to the N-acetyl glucosamine using anthracenesulfonamide was crucial for the success of the solid-phase
synthesis. This general method was employed in the formation of a variety of glycopeptides.
Introduction
blocking groups which are appropriate to both the carbohydrate
and peptidal domains.7
The automated solid-phase synthesis of oligonucleotides and
polypeptides has had a major impact on the fields of molecular
biology and pharmacology by providing access to needed
molecular tools. Among these invaluable synthetic products
are oligonucleotide primers for application of the polymerase
chain reaction (PCR),1 labeled probes for gene detection,
synthetic mutagenic agents to alter DNA sequences, and
potential therapeutic agents.2 Correspondingly, solid-phase
peptide synthesis is widely used for drug lead discovery, for
the preparation of synthetic vaccines, and for peptide bioscreen-
ing3,4 including the generation of combinatorial peptide librar-
ies.5,6 Similarly, it is likely that synthetic oligosaccharides and
glycopeptides could find considerable application (vide infra).
The surge of interest in glycoproteins8,9 arises from heightened
awareness of their importance in diverse biochemical processes
including cell growth regulation, binding of pathogens to cells,10
and intercellular communication and metastasis.11 Glycopro-
teins serve as cell differentiation markers and assist in protein
folding and transport, possibly by providing protection against
proteolysis.12
Improved isolation techniques and structural elucidation
methods13 have revealed significant levels of microheterogeneity
in naturally produced glycopeptides.14 Single eukaryotic cell
lines often produce many glycoforms of any given protein
sequence. For instance, erythropoietin (EPO), a clinically useful
red blood cell stimulant for anemia, is glycosylated by more
than 13 known types of oligosaccharide chains when expressed
in Chinese hamster ovary cells (CHO).15 The efficacy of
erythropoietin is heavily dependent on the type and extent of
glycosylation.16
The naturally occurring N-asparagine glycopeptides are â-N-
linked to an N-acetylglucosamine segment of a chitobiose, which
is part of the high mannose antennary structure (Figure 1). This
pentasaccharide core carbohydrate is the medium to which
virtually all N-asparagine glycoproteins are attached. The
specific role of N-linked proteins is not well understood.
However, the necessity of such arrays has been demonstrated
in that mice which lack the N-acetylglucosamine transferase
gene die at midgestation.17 It is possible that the glycosylation
However, the synthesis of oligosaccharides and glycopeptides
is intrinsically a more challenging problem than is the case with
oligonucleotides or oligopeptides. First, there is greater struc-
tural diversity in the potential linkage sites for joining the
saccharide units. Further complicating the goal of oligosac-
charide synthesis is the need to control the stereochemistry of
each anomeric linkage. The laboratory synthesis of glycopep-
tides faces additional difficulties stemming from the chemical
sensitivity of the bonds connecting the first amino acid to the
carbohydrate domain. Needless to say, any attempt to bring to
bear the advantages of polymer-bound synthesis protocols on
the glycopeptide problem would be dependent on the use of
‡ Sloan-Kettering.
§ Columbia University.
(7) Meldal, M. Curr. Opin. Struct. Biol. 1994, 4, 710.
(8) Feizi, T.; Bundle, D. Curr. Opin. Struct. Biol. 1994, 4, 673.
(9) Bill, R. M.; Flitsch, S. L. Chem., Biol. 1996, 3, 145-149.
(10) Bahl, O. P. An introduction to glycoproteins; Bahl, O. P., Ed.; Marcel
Dekker: New York, 1992; p 1.
(11) Kobata, A. Acc. Chem. Res. 1993, 26, 319.
(12) Opdenakker, G.; Rudd, P. M.; Ponting, C. P.; Dwek, R. A. FASEB
1993, 7, 1330.
(13) Dell, A.; Khoo, K.-H. Curr. Opin. Struct. Biol. 1993, 3, 687.
(14) Dwek, R. A.; Edge, C. J.; Harvey, D. J.; Wormald, M. R.; Parekh,
R. B. Annu. ReV. Biochem. 1993, 62, 65.
(15) Lee, Y. C.; Lee, R. T. Neoglycoconjugates: Preparation and
Applications; Lee, Y. C.; Lee, R. T., Eds.; Academic Press Inc.: London,
1994.
(1) McPherson, M. J.; Quirke, P.; Taylor, G. R. PCR A Practical
Approach; McPherson, M. J., Quirke, P., Taylor, G. R., Eds.; IRL Press at
Oxford University Press: Oxford, 1991; p 253.
(2) Blackburn, G. M.; Gait, M. J. Nucleic Acids in Chemistry and Biology;
Blackburn, G. M., Gait, M. J., Eds.; IRL Press at Oxford University Press:
Oxford, 1990; p 446.
(3) Bray, A. M.; Jhingran, A. G.; Valerio, R. M.; Maeji, N. J. J. Org.
Chem. 1994, 59, 2197.
(4) Jung, G.; Beck-Sickinger, A. G. Angew. Chem., Int. Ed. Engl. 1992,
31, 367.
(5) Gallop, M. A.; Barrett, R. W.; Dower, W. J.; Fodor, S. P. A.; Gordon,
E. M. J. Med. Chem. 1994, 37, 1233.
(6) Nestler, H. P.; Bartlett, P. A.; Still, W. C. J. Org. Chem. 1994, 59,
4723.
(16) Watson, E.; Bhide, A.; van Halbeek, H. Glycobiology 1994, 4, 227.
S0002-7863(97)04007-9 CCC: $15.00 © 1998 American Chemical Society
Published on Web 04/08/1998

3916 J. Am. Chem. Soc., Vol. 120, No. 16, 1998
Roberge et al.
Figure 1. Complex type biantennary structure.
has a conformational effect on the protein which influences the
rate or even energetics of protein folding.18 Indeed, NMR
studies on the glycopeptide 48 synthesized herein has shown
that glycosylation has a definite effect on limiting the confor-
mational possibilities of the peptide.19
Elucidation of the biological relevance of particular glyco-
protein oligosaccharide chains would benefit from isolation of
homogeneous entities. Currently, glycoprotein heterogeneity
renders this process particularly labor intensive. Some relief
may be in sight in that particular cell lines can be selected to
produce more homogeneous glycoproteins for structure-activity
relationship studies.20 However, the problem of isolation of pure
glycoproteins from natural sources remains daunting.
Fortunately, receptors normally recognize only a small
fraction of a given macromolecular glycoconjugate. Conse-
quently, synthesis of smaller but well-defined putative glyco-
peptide ligands could emerge as functionally competitive with
isolation as a source of critical structural information.15 Indeed,
prior to the work described herein, important progress in
glycopeptide synthesis pioneered by Kunz and others had
matured to the point where synthetic access to homogeneous
target systems could be achieved both in solution and in the
solid phase.7,21-29 An important advance in this area was
achieved by Cohen-Anisfeld and Lansbury who described a
highly convergent solution-based coupling of selected already
available saccharides with peptides.27
assembly method. We also demonstrate the capacity to extend
the peptide domain while the synthetic ensemble is polymer-
bound. The method we are describing here allows for fashion-
ing the carbohydrate domain of choice and also benefits from
the advantages associated with solid-phase synthesis in the all
critical carbohydrate-peptide coupling step.
Synthetic Strategy
The most successful approach to solid phase N-linked
glycopeptide synthesis which is currently practiced in several
variations in other laboratories is summarized in Scheme 1A.
A peptide segment bearing a terminal amine equivalent residue
is built on a solid support (see 2 f 3 f 4). A solution-based
glycosylated asparagine is assembled (see structure 7) by
coupling an adequately protected aspartic acid 6 with 1â-
glucosylamine 5. The latter is prepared using Koenig’s Knorr
chemistry or via anomeric amination of a naturally occurring
anomeric saccharide.23,24 The R-amino function of structure 7
is most often protected as its 9-fluorenylmethoxycarbonyl
(Fmoc) derivative and the R-carboxyl as a pentafluorophenyl
(Pfp) ester. Coupling of the two domains affords 8. Depro-
tection of the R-amino linkage generates 9. The next Fmoc
protected amino acid 1 is coupled to 9. After retrieval from
the support and deprotection of the resultant blocking groups,
target system 10 is obtained.
By contrast, we propose an alternate method which is
illustrated under Scheme 1B.30 An oligosaccharide terminating
in a glycal is constructed on the solid support (see structure
11). As shown earlier, 11 can be an extended linear structure,
or can contain branching as desired.31-35 Through chemistry
which we were projecting (vide infra Scheme 2) 11 would be
converted to the solid-phase bound 12, bearing a terminal 2-N-
acetyl-1â-aminoglucosamine residue (GlcNAc). Peptide 13
would be readily assembled through standard solution-phase
peptide synthesis methodology or by a solid-phase assembly
Herein, we report a new and efficient solid-phase synthesis
of N-linked glycopeptides based on our solid phase glycal
(17) Ioffe, E.; Stanley, P. Proc. Natl. Acad. Sci. U.S.A. 1996, 91, 728-
732.
(18) Imperiali, B.; Rickert, K. W. Proc. Natl. Acad. Sci. U.S.A. 1995,
92, 97-101.
(19) Live, D. H.; Kumar, R. A.; Beebe, X.; Danishefsky, S. J. Proc.
Natl. Acad. Sci. U.S.A. 1996, 93, 12759-12761.
(20) Lehrman, M. A.; Yucheng, Z. US Patent No. US 5272070A, 1993.
(21) Kunz, H. Pure Appl. Chem. 1993, 65, 1223.
(22) Meldal, M.; Bock, K. Glycoconjugates J. 1994, 11, 59.
(23) Meldal, M. Glycopeptides synthesis; Meldal, M., Ed.; Academic
Press Inc.: London, 1994; p 145.
(30) Roberge, J. Y.; Beebe X.; Danishefsky, S. J. Science 1995, 269,
202.
(24) Danishefsky, S. J.; Roberge, J. Y. Syntheses of oligosaccharides
and glycopeptides on insoluble and soluble supports. In Glycopeptides and
Related Compounds; Large D. G., Warren, C. D., Eds.; Marcel Dekker:
New York, 1997; p 245.
(25) Paulsen, H. Angew. Chem., Int. Ed. Engl. 1990, 29, 823.
(26) Kunz, H. Angew. Chem., Int. Ed. Engl. 1987, 26, 294.
(27) Cohen-Anisfeld, S. T.; Lansbury, P. T., Jr. J. Am. Chem. Soc. 1993,
115, 10531.
(31) Randolph, J. T.; McClure, K. F.; Danishefsky, S. J. J. Am. Chem.
Soc. 1995, 117, 5712.
(32) Randolph, J. T.; Danishefsky, S. J. Angew. Chem., Int. Ed. Engl.
1994, 33, 1470.
(33) Danishefsky, S. J.; Randolph, J. T.; Roberge, J. Y.; McClure, K.
F.; Ruggeri, R. B. Schering Lecture Series; Schering: Berlin, Germany,
1995, Vol. 26.
(34) Danishefsky, S. J.; Randolph, J. T.; Roberge, J. Y.; McClure, K.
F.; Ruggeri, R. B. Polym. Prepr. 1994, 35, 977.
(35) Danishefsky, S. J.; McClure, K. F.; Randolph, J. T.; Ruggeri, R. B.
Science 1993, 260, 1307.
(28) Anisfeld, S. T.; Lansbury, P. T., Jr. J. Org. Chem. 1990, 55, 5560.
(29) Vetter, D. T., D.; Singh, S. K.; Gallop, M. A. Angew. Chem., Int.
Ed. Engl. 1995, 34, 60- 63.

ConVergent Synthesis of N-Linked Glycopeptides
J. Am. Chem. Soc., Vol. 120, No. 16, 1998 3917
Scheme 1
Scheme 2. Preparation of Polymer-Bound
2-N-Acetyl-1-â-glucosylamine Via the Azasulfonamidation
Reaction
provide 15. In addition, targeted deprotection of the C-terminus
and coupling to peptide containing free N-terminus would allow
for elongation of the peptide chain while the glycopeptide 14
is still bound to the solid support.
The advantages, in principle, of convergence and the stream-
lining of protection requirements inherent in our proposal are
obvious on inspection. However, to implement the method, a
sound protocol to convert a polymer-bound glycal of the type
11 to 12 which bears the â-disposed anomeric amine would be
necessary. Fortunately, a simple sequence was devised as
shown schematically in Scheme 2. Crucial to the success of
the method was the use of the anthracenesulfonamide36 in the
azasulfonamidation sequence.37 Thus, the addition step (see
retrieval sequence. Coupling of 12 + 13 should lead to solid-
phase bound 14. Retrieval and global deprotection would
(36) Robinson, A. J.; Wyatt, P. B. Tetrahedron 1993, 49, 11329.
(37) McDonald, F. E.; Danishefsky, S. J. J. Org. Chem. 1992, 57, 7001.

3918 J. Am. Chem. Soc., Vol. 120, No. 16, 1998
Scheme 3. Solution Synthesis of the Trisaccharide
Roberge et al.
structure 16), the azide-induced rearrangement (see structure
17) and the fashioning of the solid-phase bound GlcNAc residue
bearing a 1â-amino function (see structure 12) all occur
smoothly as will be demonstrated.
protected galactoses would also provide a polar linker arm from
the polystyrene used in the polymer-supported synthesis. The
linkage to the polymer support would correspond to the 6-O-
TIPS position of the starting galactal 18.
In practice, the known disaccharide 18 was extended with
the necessary glucal in the usual way. Following epoxidation
with dimethyldioxirane, subsequent Lewis acid mediated ep-
oxide opening with the glycosyl acceptor 3,4-di-O-benzyl glucal
and acetylation, the trisaccharide 21 was in hand. The terminal
glucal was to be transformed into the â-N-linked glucosamine
glycopeptide.
Iodosulfonamidation of 21 with iodonium bis(sym-collidine)
perchlorate38 and anthracene sulfonamide (22) gave the trans-
diaxial iodosulfonamide 23 in good yield. Upon treatment of
this compound with the highly soluble tetrabutylammonium
azide39 the iodosulfonamide linkage underwent the expected
C1R f C2R rearrangement with capture at C1 to give the â
anomeric azide (see compound 24). Acetylation of the nitrogen
atom of the sulfonamide allowed for subsequent deprotection
of the anthracene sulfonyl group using thiophenol and Hunig’s
base35 (see 25 f 26). These extremely mild conditions gave
the stable trisaccharide N-acetylglucosamine 26 with the ano-
meric azide intact. Reduction of the azide with aluminum
amalgam provided the anomeric amine 27. This amine was
unstable to silica gel and was used immediately in the peptide
coupling reaction.
Results and Discussion
We first explored the projected glycopeptide synthesis in
solution phase in order to demonstrate the feasibility and
selectivity of each projected step in the reaction sequence. The
use of anthracenesulfonamide in the iodosulfonamidation reac-
tion was pivotal in the synthesis of the â-N-linked glycopeptides.
Other sulfonamides such as benzensulfonamide, p-toluene-
sulfonamide, and â-(trimethylsilyl)ethanesulfonamide were em-
ployed but resulted in a variety of problems including poor
solubility and a low yield in the subsequent rearrangement (vide
infra). Most disabling to the use of these sulfonamides was
our inability to achieve late stage deprotection without con-
comitant destruction of the glycopeptide. The anthracene-
sulfonamide group proved to be unique in that it could be
discharged under a variety of mild conditions which we
envisioned to be compatible with polymer-supported synthesis
and glycopeptide stability. These expectations were realized.
The required reagent is also highly soluble in THF. This solvent
is optimal for the iodosulfonamidation reaction and also a good
polystyrene swelling solvent. Thus, the use of the anthracene-
based agent results in a more efficient and complete iodosul-
fonamidation reaction.
Synthesis of the Peptides. The peptide sequence dictated
by nature for the â-N-linked glycopeptides requires a specific
AsnXSer/Thr where the asparagine is the site of glycosylation
and X corresponds to any amino acid except proline. In the
cellular machinery, glycosylation takes place on the internal
The trisaccharide with the necessary glucosamine anomeric
amine was synthesized using the glycal assembly method as
shown in Scheme 3. The trisaccharide 21, prepared as shown
in Scheme 3 was chosen to test the feasibility of convergent
glycopeptide synthesis because of the excellent â glycosylation
selectivity of the galactal carbonate derivatives and the ease of
placement of the desired terminal glucal moiety. The carbonate-
(38) Lemieux, R. U.; Levine, S. Can. J. Chem. 1964, 42, 1473.
(39) Bra¨ndstro¨m, A.; Lam, B.; Palmertz, A. Acta Chemica Scand. B 1974,
28, 699.

ConVergent Synthesis of N-Linked Glycopeptides
Scheme 4. Synthesis of the Peptide
J. Am. Chem. Soc., Vol. 120, No. 16, 1998 3919
surface of the rough endoplasmic reticulum of the golgi
apparatus where N-glycosylation enzymes recognize the serine
or threonine and glycosylates at the asparagine one amino acid
away. The glycopeptide is then transported to the surface of
the cell by vesicles that break off of the golgi apparatus.
For our model purposes, relevant tripeptides and pentapeptides
were synthesized using standard solution-phase peptide synthesis
as described in Scheme 4. The p-methoxybenzyl ester of
aspartic acid (see compound 28) was N-protected with a Troc
(1,1,1-trichloroethyl carbamate) group. The free carboxyl was
then coupled with the tosylate salt of leucine using standard
peptide-coupling conditions. The allyl ester of the leucine was
then cleaved by Pd° and the resulting acid 32 was coupled to
the allyl esters of serine (33) and threonine (34) to give
tripeptides 35 and 36, respectively. Removal of the p-
methoxybenzyl blocking group from the aspartic acid was
accomplished by the action of neat TFA to give the tripeptides
37 and 38. It proved to be crucial to remove all traces of TFA
from these peptides to provide complete coupling with the
valuable anomeric amine trisaccharide. The pentapeptides were
prepared from the tripeptides by removal of the Troc group with
zinc dust and subsequent coupling with the commercially
available dipeptide CbzAlaLeu. Removal of the p-methoxy-
benzyl group as described provided the free aspartic acid
residues 40 and 41 to be coupled to the anomeric amine of the
trisaccharide.
As will be demonstrated (vide infra), orthogonal protecting
groups of the type pioneered by Kunz21,26 proved to be
compatible with glycopeptide synthesis and also allowed for
extension of the peptide part of the glycopeptide from the
C-terminus while it is still bound to the polymer support.
The agenda next called for fully convergent coupling of
amine-terminating saccharide 27, with our synthetic peptides.
We started with the goal of coupling tripeptide 38 and
pentapeptide 40 each of which contains a single differentiated
free γ-aspartamine type carbohydrate. Standard peptide-
coupling conditions under the action of IIDQ40 using these free
aspartic acid residues afforded glycopeptides 42 and 43 in
reasonable yield (Scheme 5). However, purification of these
glycopeptides by column chromatography proved to be tedious
partly due to the precipitation of the unreacted peptide on the
column. The silyl protecting group (and envisioned model for
the eventual point of attachment to the polymer-support) was
removed by HF‚pyridine41 to give the glycopeptide alcohols 44
and 45. Subsequent deprotection of the C-terminal allyl group
of each glycopeptide with Pd° and dimethylbarbituric acid42 gave
the acids 46 and 47. The Troc group of each glycopeptide was
removed with Zn°,41 the benzyl protecting groups were removed
by catalytic hydrogenation,43,44 and the acetates and carbonates
were cleaved by potassium cyanide in methanol,45,46 to give
the fully deprotected trisaccharide tripeptide 48 and trisaccharide
pentapeptide 49. These fully deprotected glycopeptides pro-
vided among the largest synthetic glycopeptides available by
synthetic means.
Polymer-Supported Synthesis of Trisaccharide Tripeptide
and Trisaccharide Pentapeptide. While we did not seek to
optimize the solution-phase synthesis of glycopeptides, the
successful realization of the basic plan was certainly of great
value as we began to evaluate extensions to solid-phase
synthesis. The polymer-supported assembly of the trisaccharide
glycopeptides followed the general logic of the solution-phase
assembly as shown in Scheme 6. Epoxidation of the known30,31
polymer-bound disaccharide glycal 50 with 2,2-dimethyldiox-
irane and subsequent glucosylation with 3,4-di-O-benzyl glucal
gave the polymer-bound trisaccharide 51 which was acetylated
to provide 52. Iodosulfonamidation of this system with iodo-
nium bis(sym-collidine) perchlorate and anthracenesulfonamide
(23) proceeded, as expected, to give 53. Azide-induced
sulfonamide rearrangement with the highly soluble tetra-n-
(41) Matsuura, S.; Niu, C.-H.; Cohen, J. N. J. Chem. Soc., Chem.
Commun. 1976, 451.
(42) Kunz, H.; Ma¨rz, J. Synlett 1992, 591.
(43) Jones, D. A., Jr. Tetrahedron Lett. 1977, 33, 2853.
(44) Schlatter, J. M.; Mazur, R. H.; Goodmonson, O. Tetrahedron Lett.
1977, 18, 2851.
(45) Paulsen, H. Syntheses of complex oligosaccharide sequences of
glycoconjugates; Paulsen, H., Ed.; GBF: Hamburg, Germany, 1987; Vol.
8, p 115.
(40) Kiso, Y.; Yajima, H. J. Chem. Soc., Chem. Commun. 1972, 942.
(46) Mori, K.; Sasaki, M. Tetrahedron Lett. 1979, 20, 1329.

3920 J. Am. Chem. Soc., Vol. 120, No. 16, 1998
Roberge et al.
Scheme 5. Solution Coupling of the Peptides and Protecting Group Removal
Scheme 6. Solid-Phase Glycal Assembly and Anomeric Amine Synthesis
butylammonium azide followed by presumed N-acetylation
provided the anomeric azide 54. The sulfonamide could now
be removed either stepwise with thiophenol and Hunig’s base,
or inundation with 1,3-propanedithiol and Hunig’s base. Each
of these conditions effected concomitant reduction of the azide
to give polymer-bound N-acetyl glucosamine 55 with the
anomeric amine in place for coupling with a peptide. The
propanedithiol reduction of the anomeric amine is particularly
well suited for the polymer-supported work due to the ease of
workup compared with solution-phase synthesis.
In solution phase the convergent coupling of carbohydrates
and mature peptides is difficult. Separation of the product from
the unreacted components and byproducts is not a trivial matter.
This process is greatly simplified when conducted with carbo-
hydrate bound to a solid support. Most of the excess peptide
is recovered by washing the polymer and readily purified by

ConVergent Synthesis of N-Linked Glycopeptides
Scheme 7. Solid-Phase Glycopeptide Synthesis
J. Am. Chem. Soc., Vol. 120, No. 16, 1998 3921
Scheme 8
chromatography. Small amounts of peptide are lost due to
cyclization of activated aspartate. The coupling of the polymer-
bound trisaccharide 55 with the peptides shown in Scheme 7
proceeded smoothly. The peptides were usually used in 2-3
equiv excess to the polymer-bound amine. Retrieval from the
polymer-support with HF‚pyridine gave the freed glycopeptides
44, 45, 60, and 61 in 20-35% overall yield based on the initial
loading of polymer-bound galactal carbonate. These compounds
had previously been obtained from the solution-phase work and
fully deprotected.
Extension of the Peptide Portion while Bound to the
Polymer-Support. Another feature of our method is that the
presence of orthogonal protecting groups on the C- and
N-termini of the peptide provides the opportunity to extend the
peptide chain while it is still bound to the solid support, either
one amino acid at a time or, convergently, with larger fragments.
Alternatively, after removal from the support, the freed peptide
terminus may provide suitable functionality for linking to a
carrier molecule to generate other glycoconjugates.47 This
potentiality has been reduced to practice. Thus the peptide
portion of the polymer-bound glycopeptide 58 (Scheme 8) was
extended. The C-terminus was deprotected as shown to give
the acid, 62. Solid support bound 62 was then coupled to
tripeptide Asp(OPMB)LeuThr(OBn)OAll with a free N-terminus
to give polymer-bound trisaccharide octapeptide 64. Retrieval
from the solid support afforded trisaccharide octapeptide 65 in
an 18% overall yield from polymer-bound galactal carbonate.
There is the possibility of racemization associated with the use
of IIDQ in this peptide coupling. Racemized product was not
In practice, protected trisaccharide-pentapeptide 45 was
obtained in 37% overall yield based on the initial loading of
galactal carbonate. Thus, the aVerage yield for each of the 10
steps of the sequence to this compound was 91%. Chromatog-
raphy on a short column of reverse phase silica (C-18) was
sufficient to obtain this compound in pure form. This rather
straightforward purification capability arises from the previously
described “self-policing”34,35 feature of the solid-phase glycal
assembly method and underscores the efficiency in the conver-
sion of the terminal glycal to the terminal glucosylamine.
(47) Unverzagt, C.; Kunz, H. Bioorg., Med. Chem. 1993, 3, 197.

3922 J. Am. Chem. Soc., Vol. 120, No. 16, 1998
Roberge et al.
Scheme 9
Scheme 10. Solid-Phase Lactosamine-Pentapeptide Synthesis
observed in the crude 1H NMR of the final glycopeptide cleaved
from the solid support.
Similarly, the polymer-bound lactal derivative 75 could be
transformed into a lactosamine-linked glycopeptide where C6
of the glucosamine residue is protected as its benzyl ether as
shown in Scheme 10 in a 24% overall yield. This methodology
had proven compatible with more difficult glycosylation strate-
gies.
Attempts to form unnatural glycopeptides from the blood
group antigen carbohydrates by fucosylation at the glycal stage
of 75 proved difficult. However, this problem was alleviated
by first performing the iodosulfonamidation on the disaccharide
75 with the 2-position free to give iodosulfonamide (Scheme
11). Subsequent azide-induced rearrangement (see compound
83) allowed for fucosylation in the 2-position of the galactose
moiety to give the polymer-bound H-type II blood group
Disaccharide Pentapeptides and a Branched Trisaccharide
Pentapeptide. The methodology described above has proven
to be effective for other glycosylation sequences. Thus the
polymer-bound disaccharide 66 (Scheme 9) with a â-1-3
glycosylation pattern could be acetylated to give disaccharide
67. Subsequent iodosulfonamiation (see system 68), azide
rearrangement (see system 69), sulfonamide acetylation, and
reduction provided the polymer-bound anomeric amine 70. The
latter was coupled to the previously described pentapeptide 41
and yielding, upon removal from the polymer-support, the
disaccharide pentapeptide 72 in 20% overall yield based on the
initial loading of polymer-bound galactal carbonate.

ConVergent Synthesis of N-Linked Glycopeptides
J. Am. Chem. Soc., Vol. 120, No. 16, 1998 3923
Scheme 11. Solid-Phase Synthesis of a Branched Glycopeptide
determinant trisaccharide derivative 84. Acetylation, reduction,
peptide coupling led to 85. Retrieval of this construct and
removal from the polymer support provided the branched
trisaccharide pentapeptide 86 in a 10% overall yield. The
lactosamine pentapeptide 81 was also recovered from the
polymer support in a 6% overall yield. This indicated a
fucosylation yield, in the context of the solid-phase bound
glycopeptide, of ca. 60%. It is possible that the HF‚pyridine
deprotection also cleaved the acid labile fucosyl residue, thus
lowering the yield of the desired glycopeptide. This would give
a product similar to 81 but lacking the 2-O-acetyl group on the
galactose. This type of compound was not detected in the crude
1H NMR of the glycopeptide obtained from the solid support.
as well as interpolation of the suitable â-linked mannose in turn
branched with R-mannose residues.31,51-53 The logic by which
these linkages can be implemented from glycals has been
described in previous publications from our laboratory.54 The
translation of these findings to solid methodology represents a
not unsubstantial developmental goal.
Experimental Section
General Methods. Melting points are not corrected. Infrared
spectra were recorded on a Perkin-Elmer 1600 series FTIR. 1H NMR
spectra were obtained on a Bruker AMX-400 NMR (400 MHz) and
are reported in parts per million (δ) relative to SiMe4 (0.00 ppm) as an
internal reference, with coupling constants (J) reported in hertz. 13C
NMR spectra were obtained at 100 MHz and are reported in δ relative
to CDCl3 (77.00 ppm) as an internal reference, with coupling constants
(J) reported in hertz. High-resolution mass spectra were recorded on
a JOEL JMS-DX-303 HF mass spectrometer. Optical rotations were
recorded on a Jasco DIP-370 polarimeter using a 1 dm cell at the
reported temperatures and concentrations.
Chemicals used were reagent grade and used as supplied except
where noted. Pyridine, benzene, and dichloromethane (CH2Cl2) were
distilled from calcium hydride under N2. Tetrahydrofuran (THF) was
distilled from sodium/benzophenone ketyl under N2. Analytical thin-
layer chromatography was performed on E. Merck silica gel 60 F254
plates (0.25 mm) and E. Merck HPTLC RP-18 WF254s plates 0.20 mm.
Compounds were visualized by dipping the plates in a cerium sulfate-
ammonium molybdate solution followed by heating. Liquid column
chromatography was performed using forced flow of the indicated
solvent on Sigma H-Type silica gel (10-40 µm) for normal phase and
EM Science LiChroprep RP-18 (15-25 µm) for reverse phase.
Trisaccharide 20. To a cooled (0 °C) solution of the disaccharide
glycal 18 (1.282 g, 2.48 mmol) in 25 mL of CH2Cl2 was added 74 mL
of dimethyldioxirane (≈0.08 M) in acetone. After 30 min, the starting
glycal was not visible by TLC and the solvent was removed under a
stream of N2 and then under high vacuum for 5 h. To a solution of
the resulting epoxide in 10 mL of THF was added a solution of 3,4-
dibenzyl glucal 19 (1.303 g, 3.99 mmol) in 10 mL of THF. The mixture
was cooled to - 78 °C, and ZnCl2 (3.7 mL of 1 M in Et2O, 3.7 mmol)
was added. The reaction was allowed to warm to room temperature
overnight and was quenched with NaHCO3 (30 mL satd aq), the aqueous
layer was extracted with EtOAc (3 × 70 mL), and combined organics
Summary
In considering the future of approaches to glycopeptides by
chemical synthesis, it is also well to recall the excellent method
of Wong which involves enzymatically mediated elaboration
of a solid-phase-bound glycosylated peptide construct using
glycosyl transferases to append unprotected nucleoside phos-
phate activated monosaccharides in the elongation phase.48 The
Wong method also partakes of a most admirable feature in that
retrieval from the solid support is mediated with controllable
action of protease.
Obviously, there is merit in either the chemo- and or
chemoenzymatic methodologies. Clearly, the relief from pro-
tecting groups in enzymatically mediated chemistry is a massive
advantage. Both methods allow for the use of unnatural amino
acids and non-amino acids. We note, however, that the method
described herein is, in principle, totally general in that it does
not presuppose the existence of the transferases and the
availability of nucleoside activated hexoses. It will accom-
modate the inclusion of unnatural (artificial) sugars in the
scheme. Such building blocks are available from the Lewis
acid-catalyzed diene-aldehyde cyclocondensation reaction.49,50
Clearly, all workable approaches, whether purely chemical or
chemoenzymatic, are complementary and serve to foster the
advent of carefully designed, fully synthetic glycopeptides.
In terms of our own program, the next milestone must involve
elongation of the terminal GlcNac residue to a chitobiose moiety
(51) Liu, K. K.-C.; Danishefsky, S. J. J. Org. Chem. 1994, 59, 1892.
(52) Griffith, D. A.; Danishefsky, S. J. J. Am. Chem. Soc. 1990, 112,
(48) Schuster, M.; Wang, P.; Paulson, J. C.; Wong, C.-H. J. Am. Chem.
Soc. 1994, 116, 1135.
(49) Danishefsky, S. J. Chemtracts: Org. Chem. 1989, 2, 273.
(50) Berkowitz, D. B.; Danishefsky, S. J.; Schulte, G. K. J. Am. Chem.
Soc. 1992, 114, 4518.
5811.
(53) Griffith, D. A.; Danishefsky, S. J. J. Am. Chem. Soc. 1991, 113,
5863.
(54) Danishefsky, S. J.; Bilodeau, M. T. Angew. Chem., Int. Ed. Engl.
1996, 35, 1381.

3924 J. Am. Chem. Soc., Vol. 120, No. 16, 1998
Roberge et al.
were washed with brine and dried (Na2SO4). The solvent was removed
in vacuo and the crude material purified using flash chromatography
(5:4:1 EtOAc:hexanes:CH2Cl2) to give 20 (1.782 g, 84%) as a white
foam: 1H NMR (400 MHz, CDCl3) δ 7.34 (10H, m), 6.40 (1H, d, J )
6.1 Hz), 4.89 (1H, dd, J ) 2.9, 6.1 Hz), 4.86-4.83 (2H, m), 4.68-
4.63 (4H, m), 4.58-4.54 (2H, m), 4.49 (1H, dd, J ) 4.6, 7.8 Hz), 4.38
(1H, d, J ) 7.6 Hz), 4.18-3.86 (10H, m), 3.82-3.69 (3H, m), 3.55
(1H, dd, J ) 4.5, 9.5 Hz), 3.22 (1H, d, J ) 4.3 Hz), 2.85 (1H, d, J )
3.1 Hz), 1.06 (21H, m).
azide (294 mg, 1.240 mmol) in one portion. The reaction was stirred
at room temperature for 40 min and then filtered through silica gel,
the solvent removed in vacuo, and the crude material purified by column
chromatography using 10-50 µm silica gel and 4:1:5 EtOAc:CH2Cl2:
hexanes to give 434 mg (85%) of the anomeric azide trisaccharide 24:
[R]23D -26.4 (CH2Cl2, c ) 0.975); FTIR (NaCl, neat) 3308, 2942, 2117,
1814, 1757, 1454, 1371, 1331, 1224, 1162, 1147, 1085 cm-1; 1H NMR
(400 MHz, CDCl3) δ 9.19 (2H, d, J ) 9.3 Hz), 8.57 (1H, s), 7.98 (2H,
d, J ) 8.2 Hz), 7.60 (2H, dd, J ) 9.3, 6.6 Hz), 7.49 (2H, dd, J ) 8.2,
6.6 Hz), 7.31-7.25 (6H, m), 7.19 (2H, m), 7.08 (2H, m), 5.38 (1H, d,
J ) 7.4 Hz), 5.03 (2H, m), 4.92 (1H, t apparent, J ) 3.5 Hz), 4.85
(2H, m), 4.70-4.61 (4H, m), 4.47 (3H, m), 4.34 (1H, d, J ) 7.7 Hz),
3.98-3.83 (6H, m), 3.56-3.42 (6H, m), 2.12 (3H, s), 2.03 (3H, s),
1.09 (m, 21H); 13C NMR (100 MHz, CDCl3) δ 168.8, 153.9, 153.3,
137.5, 137.4, 135.4, 131.2, 129.9, 129.5, 128.8, 128.5, 128.4, 127.9,
127.6, 125.2, 124.9, 97.8, 96.5, 88.8, 80.3, 74.1, 73.9, 72.99, 72.7, 71.9,
71.4, 69.5, 68.9, 68.0, 67.7, 67.2, 67.1, 61.6, 57.8, 20.7, 20.6, 17.9,
11.8; HRMS (FAB) calcd for C61H72O20N4SiSNa (M + Na+) 1263.4130,
found 1263.4150.
Trisaccharide 21. To a solution of the trisaccharide glycal 20 (1.274
g, 1.48 mmol) in 20 mL of THF was added 1.40 mL (14.82 mmol) of
acetic anhydride followed by 2.0 mL (14.82 mmol) of collidine and a
catalytic amount of DMAP. The reaction was stirred at room
temperature for 2.5 h and then quenched with NaHCO3 (30 mL satd
aq), the aqueous layer was extracted with EtOAc (3 × 50 mL), and
combined organics were washed with CuSO4 (2 × 20 mL satd aq) and
brine (2 × 30 mL) and dried (Na2SO4). The solvent was removed in
vacuo and the crude material purified using flash chromatography
(gradient 3:7 to 4:6 EtOAc:hexanes) to give 21 (1.152 g, 82%) as a
white foam: [R]23D -38.8 (CH2Cl2, c ) 0.71); FTIR (NaCl, neat) 2943,
1814, 1756, 1648, 1454, 1371, 1225, 1172, 1066 cm-1; 1H NMR (400
MHz, CDCl3) δ 7.29 (10H, m), 6.39 (1H, d,J ) 6.2 Hz), 5.08-5.03
(3H, m), 4.93 (1H, m), 4.89-4.85 (2H,m), 4.82-4.79 (2H, m), 4.76-
4.72 (2H, m), 4.66 (1H, d, J ) 11.5 Hz), 4.61 (1H, d, J ) 11.8 Hz),
4.53 (1H, d, J ) 11.8 Hz), 4.14-3.89 (7H, m), 3.86 (1H, d, J ) 9.2
Hz), 3.81 (1H, dd, J ) 6.2, 11.2 Hz), 3.74 (1H, dd, J ) 5.7, 7.8 Hz),
3.65 (1H, ddd, J ) 5.0, 5.5, 10.5 Hz), 2.13 (3H, s), 2.09 (3H, s), 1.08
(21H, m); 13C NMR (100 MHz, CDCl3) δ 168.9, 168.7, 154.0, 153.2,
144.5, 138.4, 138.2, 128.4, 127.8, 127.7, 127.6, 99.6, 98.6, 96.6, 75.9,
74.7, 74.0, 73.7, 73.1, 73.0, 72.0, 71.5, 70.2, 69.5, 69.4, 68.9, 67.8,
67.2, 67.1, 61.7, 20.7, 17.9, 11.8; HRMS (FAB) calcd for C47H62O18-
SiNa (M + Na+) 965.3604, found 965.3613.
Trisaccharide 25. To a solution of the azide 24 (368 mg, 0.296
mmol) in 20 mL of THF was added 0.70 mL (7.40 mmol) of acetic
anhydride followed by 0.7266 g (5.92 mmol) of DMAP. The reaction
was protected from light and stirred at room temperature for 1 h and
then quenched with NaHCO3 (50 mL satd aq), the aqueous layer was
extracted with EtOAc (3 × 70 mL), and combined organics were
washed with NaHCO3 (50 mL satd aq), NH4Cl (2 × 50 mL satd aq),
and brine (50 mL) and dried (Na2SO4). The solvent was removed in
vacuo and the crude material purified using column chromatography
with 10-50 µm silica gel and a gradient of 3.5:1:4.5 EtOAc:CH2Cl2:
hexanes to give 343 mg (90%) of 25 as a yellow solid: [R]23 +27.3
D
(CH2Cl2, c ) 1.02); FTIR (NaCl, neat) 2942, 2118, 1815, 1758, 1701,
1
1370, 1224, 1165, 1086 cm-1; H NMR (400 MHz, CDCl3, rotamers
Trisaccharide 23. To a cooled (-10 °C ice-acetone bath) solution
of the trisaccharide glycal 21 (532 mg, 0.564 mmol), 9-anthracene-
sulfonamide 22 (507 mg, 1.971 mmol), and 0.887 g of powdered 3 Å
molecular sieves in 50 mL of THF was added freshly prepared iodonium
bis(sym-collidine) perchlorate (544 mg, 1.16 mmol) in one portion, and
the mixture was gently stirred for 10 min at this temperature. The
yellow suspension was warmed to 0 °C and stirred for an additional
45 min before being cooled to -10 °C. A cold (-10 °C) solution of
ascorbic acid (2.75 g) in THF:water (10:1, 60 mL) was added. The
reaction was filtered, and the filtrate was diluted with 35 mL of H2O
and the aqueous layer extracted with EtOAc (2 × 100 mL). The
combined organics were washed with satd CuSO4(aq) (2 × 50 mL)
and brine (2 × 50 mL) and dried (Na2SO4). The solvent was removed
in vacuo, and the crude material was purified by column chromatog-
raphy with 10-50 µm silica gel (3:1:6 EtOAc:CH2Cl2:hexanes) to give
at room temperature) δ 9.40 (1H, d, J ) 9.4 Hz), 9.29 (1H, d, J ) 9.4
Hz), 8.81 (0.5H, s), 8.73 (0.5H, s), 8.07 (2H, d, J ) 8.4 Hz), 8.03 (2H,
d, J ) 8.4 Hz), 7.71 (2H, m), 7.54 (2H, m), 7.37-7.17 (10H, m), 6.65
(1H, d, J ) 9.0 Hz), 5.48 (0.5H, d, J ) 8.1 Hz), 5.36 (0.5H, d, J )
10.5 Hz), 5.13-4.87 (6H, m), 4.78-4.71 (4H, m), 4.57 (0.5H, d, J )
11.1 Hz), 4.54 (0.5H, d, J ) 11.3 Hz), 4.40 (1H, m), 4.33-3.58 (8H,
m), 3.45 (1H, m), 2.17 (1.5H, s), 2.15 (1.5H, s), 2.13 (1.5H, s), 2.12
(1.5H, s), 2.07 (1.5H, s), 2.05 (1.5H, s), 1.06 (21H, m); 13C NMR (100
MHz, CDCl3, rotamers at room temperature) δ 174.2, 172.4, 169.0,
168.8, 168.7, 154.1, 153.3, 138.2, 137.9, 137.1, 131.7, 131.3, 131.2,
129.6, 129.5, 129.2, 128.6, 128.4, 128.3, 128.1, 127.9, 127.7, 127.5,
125.9, 125.7, 125.6, 125.3, 98.1, 96.7, 96.3, 86.8, 86.6, 80.6, 80.3, 79.9,
78.8, 75.3, 74.7, 73.4, 72.8, 72.4, 72.0, 71.8, 71.6, 71.5, 69.6, 69.5,
69.3, 68.5, 67.8, 67.6, 67.4, 66.3, 65.2, 63.1, 61.6, 26.6, 24.1, 20.7,
17.9, 11.8.
(619 mg, 83%) the iodo sulfonamide 23: [R]23 -28.9 (CH2Cl2, c )
D
Trisaccharide 26. To a cooled (0 °C) solution of the trisaccharide
25 (282 mg, 0.220 mmol) in 7 mL of THF was added 89 µL (0.87
mmol) of thiophenol followed by stepwise addition of 72 µL (0.41
mmol) diisopropylethylamine. The reaction was stirred for 1 h, and
then 1 mL of CH3OH was added and the reaction slowly lightened in
color. The reaction was stirred 30 min more and then quenched with
a solution of 15 mL of NH4Cl combined with 6 drops of 1 N HCl and
diluted with 50 mL of EtOAc. The organic layer was removed and
the aqueous layer extracted with EtOAc (2 × 50 mL). The combined
organics were washed with H2O (15 mL), NaHCO3 (15 mL of satd
aq), and brine and dried (Na2SO4). The crude material was filtered
through silica gel and the solvent removed in vacuo. The product was
purified by column chromatography with 10-50 µm silica gel using a
gradient of 9:1 to 6:4 CH2Cl2:EtOAc to give 208 mg (91%) of the
1.04); FTIR (NaCl, neat) 3280, 2942, 1808, 1756, 1623, 1454, 1371,
1
1336, 1223, 1167, 1068 cm-1; H NMR (400 MHz, CDCl3) δ 9.31
(2H, d, J ) 9.3 Hz), 8.72 (1H, s), 8.06 (2H, d, J ) 8.4 Hz), 7.71 (2H,
dd, J ) 9.3, 6.6 Hz), 7.55 (2H, dd, J ) 8.4, 6.6 Hz), 7.32-7.24 (8H,
m), 7.16 (2H, m), 6.29 (1H, m), 5.15 (1H, t, J ) 73.6 Hz), 5.09 (2H,
m), 4.97 (1H, d, J ) 8.6 Hz), 4.86 (2H, d, J ) 14.6 Hz), 4.76-4.69
(3H, m), 4.62 (2H, m), 4.39 (1H, d, J ) 11.5 Hz), 4.26 (1H, d, J )
11.5 Hz), 4.08-3.96 (4H, m), 3.93-3.86 (2H, m), 3.78 (2H, m), 3.47
(1H, dd, J ) 5.7, 9.7 Hz), 3.37 (1H, m), 3.17 (2H, m), 2.17 (3H, s),
2.11 (3H, s), 1.10 (m, 21H); 13C NMR (100 MHz, CDCl3) δ 168.8,
168.8, 154, 137.9, 137.3, 136.0, 131.2, 130.4, 129.4, 129.2, 128.3, 127.9,
127.8, 127.7, 127.6, 125.5, 125.0, 96.7, 96.5, 83.7, 74.7, 74.2, 73.7,
72.6, 72.2, 71.9, 71.4, 70.8, 69.4, 68.2, 67.4, 67.2, 67.1, 61.6, 31.9,
20.8, 20.7, 17.9, 11.8.
desired N-acetyl azide trisaccharide 26: [R]23 -30.6 (CH2Cl2, c )
D
Trisaccharide 24. To a solution of the iodosulfonamide 23 (548
mg, 0.413 mmol) in 25 mL of THF was added tetrabutylammoium
6.16); FTIR (NaCl, neat) 3293, 2943, 2115, 1813, 1757, 1665, 1545,
1
1455, 1371, 1225, 1087 cm-1; H NMR (400 MHz, CDCl3) δ 7.31
(10H, m), 5.90 (1H, m), 5.06 (2H,m), 5.02 (1H, t apparent, J ) 3.5
Hz), 4.94 (1H, m), 4.88-4.81 (4H, m), 4.79 (1H, d, J ) 3.5 Hz), 4.73,
(2H, m), 4.64 (1H, d, J ) 11.5 Hz), 4.61 (1H, d, J ) 11.5 Hz), 4.08-
3.85 (7H, m), 3.70-3.57 (4H, m), 3.51 (1H, t apparent, J ) 8.4 Hz),
2.13 (3H, s), 2.10 (3H, s), 1.85 (3H, s), 1.08 (21H, m); 13C NMR (100
MHz, CDCl3) δ 170.5, 168.9, 168.8, 153.9, 153.3, 138.2, 137.8, 128.4,
128.2, 128.0, 127.8, 127.7, 98.0, 96.8, 87.7, 80.6, 78.0, 76.3, 74.5, 73.2,
(55) Iodonium bis(sym-collidine) perchlorate: Under nitrogen, iodine
(775 mg, 3 mmol) was added to a suspension of silver bis(sym-collidine)
perchlorate and sym-collidine (45 µL, 0.34 mmol) in 12 mL of chloroform
(dried over activated basic alumina). The mixture was vigorously stirred
for 15 min and the bright yellow suspension was filtered over flame-dried
Celite under dry nitrogen. Ether (10 mL) was added and the faintly green
precipitate was filtered over sintered glass under dried nitrogen, washed
with ether and dried under reduced pressure. Yield ∼1 g.

ConVergent Synthesis of N-Linked Glycopeptides
J. Am. Chem. Soc., Vol. 120, No. 16, 1998 3925
72.9, 72.0, 71.7, 69.6, 69.1, 68.3, 67.7, 67.3, 61.5, 55.4, 23.3, 20.7,
20.6, 17.8, 11.9.
dd, J ) 7.0, 16.0 Hz), 2.09 (3H, s), 2.06 (3H, s), 1.85 (3H, s), 1.64
(6H, m), 1.35 (3H, d, J ) 7.0 Hz), 1.20 (3H, d, J ) 6.0 Hz), 1.09
(21H, m), 0.93 (6H, d, J ) 6.0 Hz), 0.88 (6H, d, J ) 5.0 Hz); 13C
NMR (CD3OD) δ 175.4, 174.6, 174.1, 173.7, 172.3, 171,1, 170.6, 170.5,
158.1, 155.8, 155.2, 139.4, 139.2, 139.1, 137.2, 132.8, 129.3, 129.2,
128.9, 128.8, 128.7, 128.6, 128.5, 119.1, 998, 98.6, 84.3, 79.5, 79.1,
77.2, 76.4, 75.9, 75.6, 75.4, 75.2, 74.1, 73.9, 71.8, 71.2, 70.9, 69.1,
68.6, 67.7, 66.9, 63.0, 58.0, 55.3, 53.2, 53.1, 51.9, 50.9, 41.7, 41.2,
37.7, 25.6, 25.5, 23.5, 23.2, 22.1, 21.8, 20.9, 18.4, 18.2, 16.5, 12.8;
HRMS (FAB) calcd for C90H124N7O29Si 1794.8216, found 1794.8200.
Trisaccharide 27. To a solution of the trisaccharide azide 26 (93.5
mg, 0.090 mmol) in 1.7 mL of THF and 0.7 mL of 2-propanol was
added freshly prepared aluminum amalgam (from 48 mg, 1.79 mmol
Al and HgCl2 (10% aq)) followed by 130 µL (7.17 mmol) of H2O. A
precipitate formed immediately upon addition of the H2O. The reaction
was stirred at room temperature for 16 h, and no more starting material
was identified by TLC. The crude material was filtered through Celite
and the solvent removed in vacuo to give 82.2 mg (90%) of crude 27
which was used immediately: 1H NMR (400 MHz, CDCl3) δ 7.31
(10H, m), 5.19 (1H, d, J ) 7.7 Hz), 5.09-5.04 (2H, m), 4.95 (1H, m),
4.90 (2H, dd, J ) 1.7, 8.5 Hz), 4.87-4.80 (3H, m), 4.75-4.73 (2H,m),
4.66-4.59 (2H, m), 4.08-3.85 (7H, m), 3.68-3.45 (6H, m), 2.14 (3H,
s), 2.12 (3H, s), 1.84 (3H, s), 1.57 (2H, br s), 1.08 (21H, m).
Glycopetide 42. To a solution of the trisaccharide amine 27 (22
mg, 0.022 mmol) and the tripeptide 38 (TrocAspLeuThr(OBn)OAll,
31 mg, 0.048 mmol) in 1 mL of dichloromethane was added IIDQ (13
mL, 0.043 mmol). The mixture was stirred at room temperature for
24 h. The mixture was poured into 10 mL of 0.05 N hydrochloric
acid and was extracted with 3 × 20 mL of ethyl acetate. The combined
organic fractions were washed with brine, dried with sodium sulfate,
and concentrated. The product was purified by column chromatography
with fine silica gel using a mixture of 3:2:5 acetone:CH2Cl2:hexane to
give 25.4 mg of slightly impure trisaccharide tripeptide. The mixture
was purified by column chromatography with fine silica gel using a
gradient of 97.5:2.5 to 95:5 CH2Cl2/methanol to give 21.4 mg (60%)
of the desired product 42: [R]23D -9.4° (c 1.15, CH2Cl2); FTIR (NaCl,
neat) 3286, 2948, 1813, 1753, 1719, 1644, 1536, 1371, 1224, 1078
cm-1; 1H NMR (400 MHz, CDCl3) δ 7.31 (16H, m), 7.07 (1H, d, J )
7.7 Hz), 6.63 (1H, d, J ) 8.2 Hz), 6.56 (1H, d, J ) 9 Hz), 5.82 (1H,
m),5.29 (1H, d, J ) 15 Hz), 5.22 (1H, d, J ) 10 Hz), 5.15-4.30 (20H,
m), 4.08-3.50 (13H, m), 2.72-2.60 (2H, m), 2.13 (3H, s), 2.10 (3H,
s), 1.76 (3H, s), 1.55 (2H, m), 1.52 (1H, m), 1.40-1.00 (27H, m) 0.99-
0.80 (6H, m); 13C NMR (100 MHz, CDCl3) δ 172.15, 172.00, 171.56,
170.11, 169.97, 168.97, 168.84, 154.41, 154.17, 153.28, 137.99, 137.79,
131.52, 128.79, 128.68, 128.53, 128.36, 128.30, 128.00, 127.94, 127.80,
127.78, 118.98, 98.29, 96.53, 95.27, 80.54, 79.63, 78.17, 77.20, 76.30,
74.99, 74.66, 74.29, 74.19, 74.11, 73.37, 71.97, 71.36, 70.71, 69.74,
69.31, 67.29, 66.99, 66.10, 61.62, 56.66, 54.01, 52.08, 51.34, 40.91,
37,15, 29.67, 29.24, 24.57, 23.19, 22.89, 22.07, 20.79, 20.73, 17.91,
17.87, 16.19, 16.19, 11.78; HRMS calcd for C76H103Cl3N5O27Si
1650.5674, found 1650.5646.
Glycopeptide 46. Dimethylbarbituric acid (17 mg, 0.11 mmol) and
tetrakis(triphenylphosphine)palladium (2 mg, 0.0018 mmol) were
successively added to a solution of alcohol 44 (27.2 mg, 0.018 mmol)
in 2.5 mL of THF, and the dark orange mixture was stirred at room
temperature for 2 h. The solvent was removed under reduced pressure,
and the residue was dissolved in a minimum amount of MeOH (∼0.3
mL) and applied over reverse phase silica [3 g, C-18, 1:1 H2O (0.1%
TFA):MeOH f MeOH] to give 46 (29 mg). The product was
contaminated with byproducts of the tetrakis(triphenylphosphine)-
palladium and was used for the next step.
Glycopeptide 43. To a solution of the pentapeptide 41 (80.3 mg,
0.0789 mmol) in 2 mL of CH2Cl2 was added the trisaccharide 27 (132.8
mg, 0.1455 mmol) in 3 mL of CH2Cl2 via Teflon cannula. To this
mixture was added IIDQ (25 µL, 0.0789 mmol), and the reaction was
stirred at room temperature for 10.5 h. The reaction was quenched
with a mixture of 20 mL of water and 4 mL of 0.1 N HCl, and the
aqueous phase was extracted with 6 × 30 mL of ethyl acetate. The
organics were washed with brine (30 mL) and dried (Na2SO4). The
solvent was removed in vacuo and the crude glycopeptide purified by
flash chromatography using 95:5 CH2Cl2:CH3OH to give 83.6 mg (55%)
of the desired glycopeptide 43: [R]24D -27.0° (c 2.29, CH2Cl2); FTIR
(neat) 3284, 2942, 1812, 1754, 1636, 1539, 1371, 1224, 1073 cm-1;
1H NMR (400 MHz, CD3OD) δ 7.30 (20H, m), 5.86, (1H, ddd, J )
5.5, 10.5, 16.5 Hz), 5.30 (1H, dd, J ) 1.2, 16.5 Hz), 5.19 (1H, d, J )
1.2, 10.5 Hz), 5.06-5.13 (3H, m), 5.01-5.03 (2H, m), 4.84-4.95 (6H,
m), 4.73-4.79 (4H, m), 4.65-4.70 (3H, m), 4.51-4.62 (5H, m), 4.47-
0.39 (2H, m), 4.30-4.34 (1H, m), 4.17-3.86 (12H, m), 3.69-3.65
(4H, m), 3.51-3.47 (2H, m), 2.72 (1H, d, J ) 5.0, 16.0 Hz), 2.64 (1H,
Glycopeptide 44. To a cold (-10 °C to -15 °C) solution of the
fully protected trisaccharide tripeptide 42 (19.1 mg, 0.012 mmol) and
anisole (2.5 µL, 0.023 mmol) in CH2Cl2 (3 mL) was added HF‚pyridine
(16.5 µL, 0.58 mmol). The mixture was stirred for 50 min at -10 °C
to -15 °C and then water (5 mL) was added and the mixture was
warmed to room temperature. The mixture was extracted with ethyl
acetate (3 × 30 mL) and washed a basic mixture of brine and saturated
bicarbonate (2 × 5 mL), and the organic alyers were dried with sodium
sulfate. After filtration and concentration, 44 was obtained as a crude
amorphous solid, essentially pure by proton NMR. The mixture was
purified by column chromatography with fine silica gel using a gradient
of 97.5:2.5 to 95:5 CH2Cl2 methanol to give 14.6 mg (84%) of the
desired product. This product still had traces of impurities and was
used directly in the next deprotection step.
Glycopeptide 48. The acid 46 (26.2 mg, 0.018 mmol) was dissolved
in a mixture of 2.5 mL of MeOH and 0.6 mL of acetic acid, and zinc
dust (Aldrich, 30 mg, 0.45 mmol) was added. The suspension was
stirred at room temperature for 15 h. The gray mixture was filtered
over Celite, the solids were washed successively with ethanol (15 mL)
and MeOH (15 mL), and the solvents were combined and evaporated.
The residue was applied on reverse phase silica [3 g, C-18, 7:3 H2O
(0.1% TFA):MeOH f 9:1] to give the TFA salt (20.7 mg, ∼83%). Rf
) 0.23 [7:3 H2O (0.1% TFA):MeOH]; FTIR (KBr pellet); 3306 (OH),
2928, 1804, 1752, 1670 (CdO), 1534, 1372, 1083; HRMS (FAB) calcd
for C61H78N5O25 1280.4990, found 1280.5000. The product was still
slightly impure by 1H NMR and was used for the next step. The
material from the previous reaction (20.7 mg, 0.014 mmol) was
dissolved in 2 mL of MeOH and 2 mL of acetic acid and palladium-
(II) acetate (12 mg, 0.053 mmol) was added. The yellow mixture was
kept under a positive pressure of hydrogen (balloon) for 17 h with
efficient stirring. At the end of this period, the black suspension was
sonicated for 5 min and filtered over Celite, and the solids were washed
with MeOH and acetic acid. The solvents were removed under reduced
pressure. The completion of the reaction was monitored by observing
1
the disappearance of the aromatic signals in the H NMR spectra of
the crude product in D2O. The solvent was removed under reduced
pressure, and the residue was dissolved 4 mL of MeOH. Potassium
cyanide (2.6 mg, 0.04 mmol) was added, and the pH was observed to
be between 7 and 8 (moist pH paper). After 1 h, additional potassium
cyanide was added (1 mg, 0.015 mmol), and the mixture was stirred at
pH ∼8 for 9 h. A 100 µL volume of acetic acid was added, most of
the solvents were removed under a flow of nitrogen in an efficient
fume hood, and the residue was kept under vacuum for 30 min. The
glassy solid was applied on reverse phase silica [1 g, C-18, H2O (0.1%
TFA) f 9:1 H2O (0.1% TFA):MeOH] to give 48 as its TFA salt (10
mg, 65% overall yield from 46). Rf ) 0.55 [H2O (0.1% TFA)]; [R]23
D
-0.6° (c 0.48, H2O); FTIR (KBr pellet); 3424 (OH), 2928, 1653 (CdO),
1552, 1388, 1077 cm-1; 1H NMR (400 MHz, D2O) δ 5.10 (d, 1H, J )
10 Hz, H1 GlcNAc), 4.46 (d, 1H, J ) 9 Hz, H1 Gal), 4.44 (d, 1H, J
) 8 Hz, H1 Gal), 4.39 (dd, 1H, J ) 9 Hz, J ) 4 Hz, HR Asn), 4.30-
4.22 (m, 2H, Hâ Thr, HR Leu), 4.10-4.03 (m, 1H), 4.00-3.50 (m,
18H), 3.05 (dd, 1H, J ) 18 Hz, J ) 4 Hz, Hâ Asn), 2.90 (dd, 1H, J
) 18 Hz, J ) 9 Hz, Hâ Asn), 2.01 (s, 3H, NHAc), 1.75-1.63 (m, 3H,
Hâ, Hγ Leu), 1.17 (d 3H, J ) 6 Hz, Me Thr), 0.95 (d 3H, J ) 6 Hz,
Me Leu), 0.91 (d 3H, J ) 6 Hz, Me Leu); 13C (D2O) δ 177.40, 176.21,
175.21, 172.57, 169.96, 164.90 (2JCF ) 37 Hz), 115.50 (1JCF ) 286
Hz), 104.71, 104.63, 79.72, 77.95, 76.58, 75.61, 75.33, 74.11, 73.91,
72.15, 72.08, 70.81, 70.39, 70.10, 70.06, 70.02, 69.37, 62.39, 61.46,
55.35, 54.51, 50.88, 41.05, 36.77, 25.75, 23.52 (2C), 22.12, 20.60;
HRMS (FAB, Na+) m/z 874.3821 (MH+), calcd for C34H60N5O21
874.3780.

3926 J. Am. Chem. Soc., Vol. 120, No. 16, 1998
Roberge et al.
Glycopeptide 45. To a cooled (-10 °C) solution of glycopeptide
43 (17.3 mg, 0.0090 mmol) and anisole (2 µL, 0.018 mmol) in 2 mL
of CH2Cl2 was added 0.2 mL of HF‚pyridine, and the reaction was
stirred for 2 h. The reaction was quenched with a cooled (-10 °C)
mixture of 5 mL of 31.5 mM phosphate buffer and 10 mL of ethyl
acetate. The aqueous layer was extracted with 4 × 20 mL of ethyl
acetate, the combined organics were washed with pH 8 brine (9 mL of
brine and 1 mL of satd bicarbonate) and dried (Na2SO4), and the solvent
removed in vacuo. The resulting white powder was purified by RP-
18 column chromatography, eluting with a gradient of 7:3 to 9:1 CH3-
OH:H2O to give 9.6 mg (65%) of glycopeptide 45 as a white powder:
3.52 (11H, m), 2.90 (1H, dd, J ) 5.5, 16.5 Hz), 2.77 (dd, J ) 7.5,
16.5 Hz), 2.03 (3H, s), 1.70-1.56 (6H, m), 1.39 (3H, d, J ) 7.0 Hz),
1.20 (3H, d, J ) 6.5 Hz), 0.96 (6H, d, J ) 6.0 Hz), 0.90 (6H, d, J )
6.0 Hz); HRMS (FAB) calcd for C43H75N7O23Na 1080.4792, found
1080.4800.
Trisaccharide 51. In a polymer synthesis flask, the polymer-bound
disaccharide glycal 50 (981 mg, ∼0.77 mmol) was suspended in 15
mL of CH2Cl2 for 1 h at room temperature. The mixture was cooled
to 0 °C, 3,3-dimethyldioxirane (30 mL in acetone, ∼2.4 mmol) was
added, and the mixture was gently stirred for 1 h 20 min. The solvents
were filtered, and the resin was kept under reduced pressure for 7 h. A
solution of 3,4-dibenzyl glucal (dried by azeotropic distillation with
benzene, 1.67 g, 5.12 mmol) in 10 mL of THF was added at 0 °C to
the polymer-bound epoxide and a solution of zinc chloride (1.53 mL,
1 M in ether, 1.53 mmol). The mixture was gently stirred at room
temperature for 16 h. The solvents were filtered, and the resin was
successively washed with THF (4 × 50 mL) and CH2Cl2 (3 × 50 mL)
and was kept under vacuum overnight to give 1.23 g of polymer-bound
trisaccharide 51.
Trisaccharide 52. In a polymer synthesis flask, the polymer-bound
trisaccharide glycal 51 (1.23 g, ∼0.76 mmol) was suspended in 20 mL
of THF for 1 h at room temperature. Acetic anhydride (1.44 mL, 15.2
mmol) and sym-collidine (2.0 mL, 15.2 mmol) were successively added,
and the mixture was gently stirred for 12 h. The solvents were filtered,
and the resin was successively washed with THF (4 × 50 mL) and
CH2Cl2 (3 × 50 mL) and was kept under vacuum overnight to give
1.22 g of polymer-bound trisaccharide 52.
Trisaccharide 53. In a polymer synthesis flask, the polymer-bound
trisaccharide glycal 52 (559 mg, ∼0.212 mmol) and 9-anthracene-
sulfonamide (381 mg, 1.5 mmol) were suspended in 15 mL of THF
for 1 h at room temperature. The mixture was cooled to -10 °C (ice-
acetone bath) and freshly prepared iodonium bis(sym-collidine) per-
chlorate (261 mg, 1.1 mmol) was added in one portion, and the mixture
was gently stirred for 10 min at this temperature. The yellow
suspension was warmed to 0 °C and stirred for an additional 1 h before
being cooled to -10 °C. A cold (-10 °C) solution of ascorbic acid (2
g) in THF:water (10:1, 40 mL) was added, turning the solution almost
colorless while the polymer remained bright yellow. After 5 min at
-10 °C, the mixture was warmed to room temperature (∼1 h). The
solvents were filtered, and the resin was successively washed with THF
(4 × 50 mL) and CH2Cl2 (3 × 50 mL) and was kept under vacuum
overnight to give 643 mg of polymer-bound trisaccharide 53. This
resin fluoresced blue at 365 nm.
Trisaccharide 54. In a polymer synthesis flask, the polymer-bound
iodo sulfonamide 53 (553 mg, ∼0.18 mmol) was suspended in 12 mL
of THF for 1 h at room temperature. Tetrabutylammonium azide (261
mg, 1.1 mmol) was added, and the mixture was gently stirred for 5 h.
The solvent was filtered, and the resin was successively washed with
THF (4 × 50 mL) and CH2Cl2 (3 × 50 mL) and was kept under vacuum
overnight to give 520 mg of polymer-bound trisaccharide with the
anomeric azide and 2-sulfonamide. This resin also fluoresced blue at
365 nm. In a polymer synthesis flask, the polymer-bound azide
sulfonamide (469 mg, ∼0.16 mmol) was suspended in 15 mL of THF
for 1 h at room temperature. Acetic anhydride (752 µL, 8 mmol) and
4-(N,N-dimethylamino)pyridine (779 mg, 6.4 mmol) were successively
added, and the mixture was gently stirred for 3 h. The solvent was
filtered, and the resin was successively washed with THF (3 × 50 mL)
and CH2Cl2 (3 × 50 mL) and was kept under vacuum overnight to
give 480 mg of polymer-bound trisaccharide 54. This resin fluoresced
green at 365 nm.
[R]24 -14.2° (c 1.38, CH2Cl2); FTIR (neat) 3289, 2956, 1807, 1750,
D
1637, 1533, 1454, 1371, 1229, 1171, 1073 cm-1; 1H NMR (400 MHz,
CD3OD) δ 7.31 (20H, m), 5.88, (1H, ddd, J ) 5.5, 10.5, 16.5 Hz),
5.31 (1H, dd, J ) 1.0, 16.5 Hz), 5.19 (1H, d, J ) 1.2, 10.5 Hz), 5.14-
5.00 (5H, m), 4.96 (1H, t, J ) 4.0 Hz), 4.92-4.89 (6H, m), 4.81-4.68
(7H, m), 4.66-4.57 (7H, m), 4.49-4.41 (2H, m), 4.34 (1H, d, J )
6.0, 9.0 Hz), 4.19-4.00 (7H, m), 3.92 (1H, t, J ) 10.0 Hz), 3.82-
3.67 (6H, m), 3.52 (2H, d, J ) 5.0 Hz), 2.76 (1H, d, J ) 5.0, 16.0 Hz),
2.64 (1H, dd, J ) 7.0, 16.0 Hz), 2.09 (3H, s), 2.06 (3H, s), 1.85 (3H,
s), 1.74-1.60 (6H, m), 1.35 (3H, d, J ) 7.0 Hz), 1.22 (3H, d, J ) 6.5
Hz), 0.94 (6H, d, J ) 6.0 Hz), 0.89 (6H, d, J ) 6.0 Hz); 13C NMR
(100 MHz, CD3OD) δ 175.6, 174.9, 174.3, 173.6, 172.5, 172.4, 171.2,
170.7, 168.8, 161.8, 155.6, 155.4, 139.7, 139.4, 139.3, 133.1, 129.4,
129.3, 128.9, 128.8, 128.7, 128.6, 119.1, 102.0, 99.2, 84.4, 79.7, 79.1,
76.6, 76.1, 75.8, 75.6, 75.4, 73.3, 71.9, 71.2, 70.6, 69.8, 68.6, 68.4,
68.3, 67.7, 67.0, 61.8, 58.2, 55.5, 53.3, 53.2, 52.1, 51.1, 41.9, 41.4,
37.8, 25.8, 25.6, 23.5, 23.2, 22.1, 21.9, 21.2, 20.8, 18.1, 16.5; HRMS
(FAB) calcd for C81H103N7O29 1638.6881, found 1638.6840.
Glycopeptide 47. Dimethylbarbituric acid (2.1 mg, 0.013 mmol)
and tetrakis(triphenylphosphine)palladium (1.0 mg, 0.0009 mmol) were
successively added to a solution of alcohol 45 (3.6 mg, 0.0022 mmol)
in 2.0 mL of THF, and the dark orange mixture was stirred at room
temperature for 5 h. The solvent was removed under reduced pressure,
and the residue was dissolved in a minimum amount of MeOH (∼0.3
mL) and applied over reverse phase silica, C-18, 7:3 H2O (0.1% TFA):
MeOH f MeOH] to give trisaccharide pentapeptide 47 (3.5 mg,
100%): [R]24D -14.3° (c 0.46, CH2Cl2); FTIR (neat) 3298, 1812, 1754,
1651, 1514, 1454, 1371, 1221, 1168, 1069 cm-1; 1H NMR (400 MHz,
CD3OD) δ 8.25 (1H, d, J ) 9.0 Hz), 8.16 (1H, d, J ) 8.0 Hz), 8.07
(1H, d, J ) 7.5 Hz), 7.91-7.87 (2H, m), 7.56-7.51 (2H, m), 7.31
(20H, m), 5.13-4.99 (4H, m), 4.96 (1H, t, J ) 3.5 Hz), 4.91-4.89
(5H, m), 4.79-4.69 (5H, m), 4.66-4.60 (3H, m), 4.57-4.54 (1H, m),
4.50-4.45 (2H, m), 4.31-4.38 (1H, m), 4.21-4.00 (6H, m), 3.92 (1H,
t, J ) 10.0 Hz), 3.86-3.64 (6H, m), 3.52 (2H, d, J ) 5.5 Hz), 2.75
(1H, dd, J ) 5.0, 16.0 Hz), 2.67 (1H, dd, J ) 9.0, 16.0 Hz), 2.09 (3H,
s), 2.05 (3H, s), 1.87 (3H, s), 1.73-1.57 (6H, m), 1.35 (3H, d, J ) 7.0
Hz), 1.21 (3H, d, J ) 6.5 Hz), 0.94 (6H, d, J ) 6.0 Hz), 0.89 (6H, d,
J ) 6.0 Hz); 13C NMR (CD3OD) δ 199.2, 175.8, 174.9, 174.0, 173.4,
172.7, 172.6, 170.9, 155.8, 155.6, 139.9, 139.6, 129.5, 129.4, 129.1,
129.0, 128.9, 128.8, 128.7, 100.1, 99.5, 84.8, 79.9, 79.3, 77.4, 76.9,
76.1, 75.8, 75.6, 75.2, 75.0, 72.3, 72.2, 71.8, 71.4, 70.1, 68.9, 68.6,
67.8, 62.0, 58.1, 55.6, 53.4, 52.2, 51.3, 41.9, 41.6, 37.9, 25.9, 25.7,
23.6, 23.2, 22.1, 21.9, 20.8, 18.2, 16.7.
Glycopeptide 49. In a sealed tube, a mixture of Pd(OAc)2 and
glycopeptide 47 (4.7 mg, 0.0027 mmol) in 4 mL of CH3OH and 2
drops of acetic acid was charged and evacuated with 4 × H2 and filled
to a pressure of 60 psi and stirred overnight. The reaction was then
sonicated and filtered through Celite, and the solvent was removed in
vacuo to give a crude white powder, the 1H NMR of which showed no
more benzyl groups in the aromatic region. The debenzylated
glycopeptide was redissolved in 1 mL of CH3OH, and 50 µL of 0.1 M
KCN/CH3OH was added. The reaction was pH 8-9 by pH paper. It
was stirred at room temperature for 4 h and then quenched with 2 drops
of trifluoroacetic acid, and the solvent was removed under a stream of
N2. The crude material was purified by RP-18 column chromatography,
eluting with a gradient of 1:9 to 9:1 CH3OH:H2O to give 1.2 mg (0.0010
mmol, 37%) of the glycopeptide 49 as a white powder: FTIR (neat)
3326, 1671, 1538, 1204 cm-1; 1H NMR (400 MHz, D2O) δ 5.08 (1H,
d, J ) 9.8 Hz), 4.75 (1H, m), 4.49 (1H, d, J ) 7.8 Hz), 4.45 (1H, d,
J ) 7.9 Hz), 4.40-4.24 (3H, m), 4.07 (1H, dd, J ) 3.0, 9.5 Hz), 3.98-
Trisaccharide 55. In a polymer synthesis flask, the polymer-bound
trisaccharide azide sulfonamide 54 (51 mg, ∼0.018 mmol) was
suspended in 5 mL of DMF for 1 h at room temperature. 1,3-
Propanedithiol (105 µL, 1.04 mmol) and N,N-diisopropyl-N-ethylamine
(109 µL, 0.63 mmol) were added, and the suspension was gently stirred
at room temperature for 6 h. The resin was successively washed with
DMF (3 mL) and THF (2 × 4 mL), and an aliquot of the resin was
examined by IR (KBr pellet). The reaction was incomplete as was
indicated by the presence of an azide stretch at 2115 cm-1. The resin
was suspended in 2 mL of DMF, 1,3-propanedithiol (220 µL, 2.19
mmol) and N,N-diisopropyl-N-ethylamine (220 µL, 1.26 mmol) were

ConVergent Synthesis of N-Linked Glycopeptides
J. Am. Chem. Soc., Vol. 120, No. 16, 1998 3927
added, and the suspension was gently stirred at room temperature for
12 h. The resin was successively washed with DMF (3 mL), THF (2
× 5 mL), and CH2Cl2 (2 × 5 mL). The resin was dried under reduced
pressure to give 51 mg of solid 55. This resin did not substantially
fluoresce at 365 nm.
Glycopeptide 59. In a polymer synthesis flask, the polymer-bound
trisaccharide 55 (120.9 mg, ∼0.014 mmol) was swollen in 4 mL of
CH2Cl2 for 1 h at room temperature. The pentapeptide CbzAlaLeuAs-
pLeuThr(OBn)OAll 41 (96.4 mg, 0.106 mmol) and IIDQ (Aldrich, 30
µL, 0.101 mmol) were successively added. After 25 h, the solvent
was filtered, and the resin was washed with CH2Cl2 (5 × 5 mL), THF
(1 × 5 mL), and CH2Cl2 (3 × 5 mL), and the resin was dried under
reduced pressure to give 133.3 mg of resin 59.
added, and the dark orange mixture was stirred at room temperature.
After 14 h, the solvent was filtered, the resin was washed with CH2Cl2
(4 × 10 mL), and the resin was dried under reduced pressure to give
113.5 mg of resin 62.
Trisaccharide 83. To the polymer-supported azide 82 (102 mg,
∼0.05 mmol) swollen in 1.0 mL of THF was added tri-O-benzyl fucosyl
fluoride (269 mg, 0.511 mmol) as a 0.5 M solution in THF. To this
cooled (-10 °C) mixture was added 2,6-di-tert-butylpyridine (46 µL,
0.204 mmol) followed by stannous(II)trifluoromethanesulfonate (170
mg, 0.40 mmol). The reaction was allowed to warm to room-
temperature overnight and then diluted with CH2Cl2 and filtered, and
the polymer-support was washed with CH2Cl2 (3 × 20 mL), DME (3
× 20 mL), and CH2Cl2 (3 × 20 mL) and dried under vacuum to give
130.3 mg of polymer-bound trisaccharide 83.
Glycopeptide 45. In a Teflon flask, the polymer-bound trisaccharide
pentapeptide 59 (94.3 mg, ∼0.0273 mmol) and anisole (8.5 µL, 0.78
mmol) were suspended in 5 mL of CH2Cl2 for 1 h at room temperature.
The mixture was cooled to -10 °C (acetone/ice), and HF‚pyridine (∼50
µL, 1.7 mmol) was slowly added. After 2 h, H2O (5 mL) was added
at -10 °C, and the mixture was extracted with ethyl acetate (4 × 30
mL). The combined organic layers were washed twice with a mixture
of brine (10 mL) and saturated NaHCO3 (1-2 mL, pH 7). The organic
extracts were dried over sodium sulfate, filtered over a medium porosity
glass filter, and concentrated. The residue was purified over RP-18
silica, eluting with 7:3 MeOH:H2O (0.1% TFA) f 9:1 to give
trisaccharide pentapeptide 45 (17.9 mg, ∼37%).
Acknowledgment. We dedicate this paper to the pioneering
accomplishments of Professor Horst Kunz in the field of
glycoconjugate synthesis. This research was supported by NIH
Grant No. AI 16943. We also acknowledge a NSERC (Canada)
postdoctoral fellowship to J.Y.R. and a NIH Postdoctoral
Training Grant T32 CA62948 to X.B.
Supporting Information Available: Experimental descrip-
tions for 29, 31, 32, 34-41, 56-58, 60, 61, 63-65, 67-73,
76-82, 84-86 (20 pages). See any current masthead page for
ordering and Internet access instructions. See any current
masthead page for ordering and Web access instructions.
Glycopeptide 62. In a polymer synthesis flask, the polymer-bound
trisaccharide pentapeptide 58 (125.4 mg, ∼0.030 mmol) was swollen
in 6 mL of THF for 1 h at room temperature. To this mixture,
dimethylbarbituric acid (140 mg, 0.90 mmol) and tetrakis(triph-
enylphosphine)palladium (17 mg, 0.015 mmol) were successively
JA9740071