An efficient synthetic route to glycoamino acid building blocks for glycopeptide synthesis

Article abstract of DOI:10.1021/ol048342n

(Chemical Equation Presented) Chemical glycopeptide synthesis requires access to gram quantities of glycosylated amino acid building blocks. Hence, the efficiency of synthesis of such building blocks is of great importance. Here, we report a fast and high

Full text of DOI:10.1021/ol048342n

ORGANIC
LETTERS
2004
Vol. 6, No. 22
4001-4004
An Efficient Synthetic Route to
Glycoamino Acid Building Blocks for
Glycopeptide Synthesis
Mallesham Bejugam and Sabine L. Flitsch*
School of Chemistry, The UniVersity of Edinburgh, King’s Buildings,
West Mains Road, Edinburgh EH9 3JJ, United Kingdom
s.flitsch@ed.ac.uk
Received August 20, 2004
ABSTRACT
Chemical glycopeptide synthesis requires access to gram quantities of glycosylated amino acid building blocks. Hence, the efficiency of
synthesis of such building blocks is of great importance. Here, we report a fast and highly efficient synthetic route to Fmoc-protected asparaginyl
glycosides from unprotected sugars in three steps with high yields. The glycosylated amino acids were successfully incorporated into target
glycopeptides 7 and 8 by standard Fmoc solid-phase peptide synthesis.
It is estimated that more than half of all proteins carry
carbohydrate side chains, with the majority being N-glycans
linked to asparagine residues that are part of the Asn-X-Ser/
Thr tripeptide codon for N-glycosylation.1 Very little is
understood about how the glycan chains specifically modu-
late stability and activity of glycoproteins because of the
complexity and heterogeneity of glycoprotein samples iso-
lated from biological sources. Glycopeptides are therefore
important targets for in vitro synthesis,2 and with the advent
of the native chemical ligation of peptides even small- and
medium-sized glycoproteins are now targeted by chemical
synthesis.3
N-Glycans can influence glycoprotein biosynthesis, struc-
ture, and activity in many ways. One intriguing question is
that of the role of the highly conserved core pentasaccharide
of N-glycans. Although N-glycosylation is diverse across
tissues and species, the pentasaccharide core which is linked
through the reducing N-acetylglucosamine to an asparaginyl
side chain of the polypeptide is conserved in almost all higher
organisms. Recent crystallographic and NMR data has shown
that the core, especially the first two N-acetylglucosamine
residues, can have important interactions with the protein
and be responsible for stabilization of the active protein
structure.4
To address questions of the structure-activity relationship
of N-glycans, we are seeking to develop synthetic methods
that would give us good access to glycopeptides carrying
(1) Apweiler, R.; Hermjakob, H.; Sharon, N. Biochim. Biophys. Acta
1999, 1473, 4.
(2) (a) Marcaurelle, L. A.; Bertozzi, C. R. Glycobiology 2002, 12 (6),
69R-77R (b) Warren, J. D.; Miller, J. S., Keding, S. J.; Danishefsky, S. J.
J. Am. Chem. Soc. 2004, 126, 6576.
(3) (a) Dawson, P. E.; Kent, S. B. H. Annu. ReV. Biochem. 2000, 69,
923. (b) Shin, Y.; Winans, K. A.; Bakes, B. J.; Kent, S. B. H.; Ellman, J.
A.; Bertozzi, C. R. J. Am. Chem. Soc. 1999, 121, 11684. (c) Dudkin, V.
Y.; Miller, J. S.; Danishefsky, S. J. J. Am. Chem. Soc. 2004, 126, 736. (d)
Bang, D.; Chopra, N.; Kent, S. B. H. J. Am. Chem. Soc. 2004, 126, 1377.
(4) (a) Imperiali, B.; O’Connor, S. E. Curr. Opin. Chem. Biol. 1999, 3
(6), 643. (b) Petrescu, A. J.; Milac, A. L.; Petrescu, S. M.; Dwek, R. A.;
Wormald, M. R. Glycobiology. 2004, 14 (2), 103.
10.1021/ol048342n CCC: $27.50
© 2004 American Chemical Society
Published on Web 09/30/2004

natural N-glycan core structures and a number of specific
analogues.
Table 1. Optimization of Microwave-Assisted Kochetkov
During previous work on the synthesis of glycopeptides,
two strategies have been employed. One approach introduces
the carbohydrate as part of a glycoamino acid building block
during solid-phase synthesis of the polypeptide chain.5
Alternatively, the carbohydrate can be attached to a selec-
tively deprotected aspartic acid residue once the polypeptide
has been formed using the method developed by Lansbury.6
By a similar route, unnatural glycopeptides have also been
generated using selective alkylation of free cysteine residues.7
Amination
solvent
T (°C)
time (min)
product
dimera (%)
MeOH
CH3CN
DMF
DMSO
DMSO
DMSO
DMSO
40
40
40
40
50
60
40
90
90
90
90
90
90
45
no product
no product
trace
80-90
60-70
ND
ND
ND
5
15
20
5
Both of the routes to peptides bearing N-glycans generally
use â-glycosylamines as key intermediates,8 with subsequent
acylation by a suitably protected amino acid or a polypeptide
side chain depending on the approach taken. Such glycosyl-
amines are accessible by two routes: from suitably protected
glycosyl azides requiring multiple protection and deprotection
steps (generally five steps) or by direct amination (one step)
using the Kochetkov reaction.8 The latter method starts with
a fully deprotected reducing sugar which is treated with 40-
50 times excess ammonium bicarbonate for 6 days.8 The
longer reaction times and difficulties in removing ammonium
bicarbonate effectively by evaporation (a process that takes
several days) have been a major drawback of the Kochetkov
amination reaction.9,10
50-60
60-70
a ND (not determined). Dimerization of the amine increased with
temperature from 40 to 60 °C.
spectrometry. Such a dimer is formed by further condensation
of 2a to starting sugar. The formation of this dimer in up to
10% yield has also been observed in thermal reactions8b and
has been shown not to interfere with subsequent acylations.
The formation of dimer was significantly increased at higher
temperatures such as 50 and 60 °C (Table 1), and reaction
temperatures were therefore generally kept at 40 °C.
The optimized reaction conditions were used successfully
to afford glycosylamines 2a-f from mono-, di-, and tri-
saccharides 1a-f in excellent yields (Table 2). In all cases,
Prompted by reports that imine formation can be acceler-
ated by microwave irradiation,11 we decided to study the
Kochetkov reaction using microwave irradiation. Our aim
was to develop methodology that would overcome the
substantial practical drawbacks of this key reaction for
glycopeptide synthesis, in particular, shorten reaction times
and reduce the amount of bicarbonate needed.
Initially, the reaction conditions were optimized for
N-acetyl glucosamine (1a), which is conserved as the
reducing monosaccharide unit of the pentasaccharide core
(Table 1). Of the various solvents tested, only DMSO resulted
in good yields of 2a (Table 1, compare the first four entries).
Gratifyingly, the reaction was found to be efficient with only
5-fold excess (w/w) of ammonium carbonate over sugar
compared to the 40-50-fold excess needed under thermal
conditions.8 Reactions appeared to be complete after 90 min
of microwave irradiation (10 W), while maintaining the
vessel temperature at 40 °C and maximum pressure at 250
psi.
Table 2. Microwave-Assisted Kochetkov Amination
substrate
R1
R
product (yield, %)a
1a
1b
1c
1d
1e
1f
H
NHAc
OH
OH
NHAc
OH
OH
2a (87)
2b (86)
2c (75)
2d (35)
2e (75)
2f (70)
Glc (â1-
Gal (â1-
GlcNAc (â1-
Glc(R1-4)Glc (R1-
Glc (R1-
In addition to the product 2a, a small amount of side
product, diglycosylamine, was also observed by mass
a Yields were calculated from 1H NMR spectra in D2O and are based
on the integration of the anomeric proton of crude glycosylamine and
substrate.
(5) (a) Wagner, M.; Dziadek, S.; Kunz, H. Chem. Eur. J. 2003, 9 (24),
6018. (b) Meinjohanns, E.; Meldal, M.; Paulsen, H.; Dwek, R. A.; Bock,
K. J. Chem. Soc., Perkin Trans. 1998, 1, 549.
(6) Cohen-Anisfeld, S. T.; Lansbury, P. T., Jr. J. Am. Chem. Soc. 1993,
115, 10531.
(7) Macmillan: D.; Daines, A. M.; Bayrhuber, M.; Flitsch, S. L. Org.
Lett. 2002, 4, 1467.
(8) (a) Likhosherstov, L.; Novikova, O.; Derveitskaja, V. A.; Kochetkov,
N. K. Carbohydr. Res. 1986, 146, C1-C5. (b) Vetter, D.; Gallop, M. A.
Bioconjugate Chem. 1995, 6, 316.
(9) Lubineau, A.; Auge, J.; Drouillat, B. Carbohydr. Res. 1995, 266,
211.
the small excess of ammonium carbonate and DMSO was
easily removed by freeze-drying the reaction mixture over-
night to yield a colorless hygroscopic solid, which could used
for further acylation studies without any purification. Selec-
tive formation of the â-glycosylamine was demonstrated by
NMR.
(10) Tokuda, Y.; Takahashi, Y.; Matoishi, K.; Ito, Y.; Sugai, T. Synlett
2002, 1, 57.
(11) Perreux, L.; Loupy, A. Tetrahedron 2001, 57, 9199.
Formation of glycosylamines was also attempted starting
from peracetylated chitobiose and lactose to determine if
4002
Org. Lett., Vol. 6, No. 22, 2004

deacetylation and amination could be achieved at the same
time. However, under our reaction conditions only 1-O-
deacetylated glycosides together with trace amounts of
glycosylamines were observed.
Scheme 1. Synthesis of Cbz-Protected Glycoamino Acid 6
The crude glycosylamines 2a-e were then used for the
preparation of glycoamino acid building blocks (Table 3).
Table 3. Synthesis of Fmoc-Protected Glycoamino Acids from
Crude Glycosylamines
Thus, the glycosylated asparagine building blocks 4a and
4b were incorporated into target glycopeptides 7 and 8 using
Fmoc-based solid-phase peptide synthesis on the Wang resin
(Scheme 2).
Scheme 2. Fmoc-SPPS of Glycopeptide 7 and 8
substrate
R1
R
product (yield, %)b
2a
2b
2c
2d
2e
H
NHAc
OH
OH
NHAc
OH
4a (89)
4b (86)
4c (79)
4d (74)
4e (81)
Glc (â1-
Gal (â1-
GlcNAc (â1-
Glc(R1-4) Glc (R1-
b Overall yields from 2a-e.
Coupling of the amines (2a-e) with N-R-Fmoc-protected
L-aspartic acid R-tert-butyl ester using HOBt/HBTU as
coupling agent and subsequent acidic deprotection resulted
in the corresponding Fmoc-protected glyco amino acids 4a-e
in good overall yields. Optimized yields were obtained with
only 1.2-fold excess of amino acid over crude amine, which
shows that most of the ammonium carbonate had been
successfully removed during workup from the previous
reaction mixture.
Alternatively, the Cbz-protected glycoamino acid 6 was
prepared via acylation of 2a with the â-acyl fluoride of N-R-
Cbz protected L-aspartic acid R-isobutyl ester, followed by
enzymatic hydrolysis of the isobutyl ester with papain
(Scheme 1).10
One of the future aims of our studies is the chemical
synthesis of the measles binding domains of the CD46
complement cofactor in humans. Structural studies have
shown that the core saccharides (in particular, the first two
GlcNAc residues) interact strongly with the polypeptide
backbone, and functional studies have highlighted that
glycosylation is essential for measles virus recognition.12
Our aim is the total synthesis of such receptor modules
bearing defined natural and unnatural carbohydrate chains
using native chemical ligation. Such synthesis requires access
to glycosylated peptides such as 7 and 8.
Reaction conditions for each coupling step were 4-fold
excess of amino acid with TBTU, HOBt, and DIPEA in DMF
for 4 h, followed by Fmoc deprotection using 20% piperidine
in DMF for 1 h. All amino acid coupling steps were single
couplings except for coupling of the glycoamino acids 4a
and 4b, which were incorporated into peptide using double
coupling before Fmoc deprotection to ensure completion.
The crude glycopeptides were cleaved from the resin using
a mixture of TFA (94%)/H2O (2.5%)/EDT (2.5%)/TIS (1%)
for 4 h, in good overall yields (67% for 7 and 54% for 8
based on a loading of 0.56 mmol/gram of resin). Analysis
by LC-MS(ESI) confirmed their structure.
(12) Casasnovas, J. M.; Larvie, M.; Stehle, T. EMBO J. 1999, 18, 2911.
Org. Lett., Vol. 6, No. 22, 2004
4003

In summary, we have developed a short (three-step)
synthetic route from unprotected sugar to glycosylated
asparagine building blocks for glycopeptide synthesis by
using microwave-assisted Kochetkov amination as a key
reaction. This route can be applied to mono- and oligo-
saccharides and yields glycosylated asparagine building
blocks in a short period of time with minimal workup. The
overall yields of the microwave-assisted aminations appear
similar to that reported for the thermal conversion and
suggest that this methodology will be very useful for other
applications such as the preparation of glycoconjugates for
subsequent immobilization.8b The application of the gly-
coamino acid building blocks in glycopeptide synthesis of 7
and 8, which are key intermediates for the synthesis of
glycoproteins by native chemical ligation, was successfully
demonstrated.
Acknowledgment. We thank Drs. Ashley Causton,
Alison Daines, and Beatrice Maltman for assistance and
helpful discussions and John Miller and Robert Smith for
analytical support. We thank ORS, BBSRC, Wellcome Trust,
and University of Edinburgh for financial support.
Supporting Information Available: Experimental pro-
cedures and characterization data [1H NMR, 13C NMR,
HRMS (FAB)] for all new compounds and LC-MS data
for target glycopeptide 7 and glycopeptide 8. This material
is available free of charge via the Internet at http://pubs.acs.org.
OL048342N
4004
Org. Lett., Vol. 6, No. 22, 2004