Chemical Synthesis of a Glycopeptide Derived from Skp1 for Probing Protein Specific Glycosylation

Article abstract of DOI:10.1002/chem.201501598

Skp1 is a cytoplasmic and nuclear protein, best known as an adaptor of the SCF family of E3-ubiquitin ligases that label proteins for their degradation. Skp1 in Dictyostelium is posttranslationally modified on a specific hydroxyproline (Hyp) residue by a pentasaccharide, which consists of a Fucα1,2-Galβ-1,3-GlcNAcα core, decorated with two α-linked Gal residues. A glycopeptide derived form Skp1 was prepared to characterize the α-galactosyltransferase (AgtA) that mediates the addition of the α-Gal moieties, and to develop antibodies suitable for tracking the trisaccharide isoform of Skp1 in cells. A strategy was developed for the synthesis of the core trisaccharide-Hyp based on the use of 2-naphthylmethyl (Nap) ethers as permanent protecting groups to allow late stage installation of the Hyp moiety. Tuning of glycosyl donor and acceptor reactivities was critical for achieving high yields and anomeric selectivities of glycosylations. The trisaccharide-Hyp moiety was employed for the preparation of the glycopeptide using microwave-assisted solid phase peptide synthesis. Enzyme kinetic studies revealed that trisaccharide-Hyp and trisaccharide-peptide are poorly recognized by AgtA, indicating the importance of context provided by the native Skp1 protein for engagement with the active site. The trisaccharide-peptide was a potent immunogen capable of generating a rabbit antiserum that was highly selective toward the trisaccharide isoform of full-length Skp1. Antibody tracking: A glycopeptide containing a trisaccharide-hydroxyproline moiety was synthesized to characterize the substrate requirements of α-galactosyltransferase (AgtA) that mediates the addition of the α-Gal residues to the glycan of the glycoprotein Skp1 and to develop antibodies suitable for tracking the trisaccharide isoform of Skp1 in cells.

Full text of DOI:10.1002/chem.201501598

DOI: 10.1002/chem.201501598
Full Paper
&
Glycopeptide Synthesis
Chemical Synthesis of a Glycopeptide Derived from Skp1 for
Probing Protein Specific Glycosylation
Zoeisha S. Chinoy,[a] Christopher M. Schafer,[b, c] Christopher M. West,[b, d] and Geert-
Jan Boons*[a]
Abstract: Skp1 is a cytoplasmic and nuclear protein, best
known as an adaptor of the SCF family of E3-ubiquitin ligas-
es that label proteins for their degradation. Skp1 in Dictyos-
telium is posttranslationally modified on a specific hydroxy-
proline (Hyp) residue by a pentasaccharide, which consists
of a Fuca1,2-Galb-1,3-GlcNAca core, decorated with two a-
linked Gal residues. A glycopeptide derived form Skp1 was
prepared to characterize the a-galactosyltransferase (AgtA)
that mediates the addition of the a-Gal moieties, and to de-
velop antibodies suitable for tracking the trisaccharide iso-
form of Skp1 in cells. A strategy was developed for the syn-
thesis of the core trisaccharide-Hyp based on the use of 2-
naphthylmethyl (Nap) ethers as permanent protecting
groups to allow late stage installation of the Hyp moiety.
Tuning of glycosyl donor and acceptor reactivities was criti-
cal for achieving high yields and anomeric selectivities of
glycosylations. The trisaccharide-Hyp moiety was employed
for the preparation of the glycopeptide using microwave-as-
sisted solid phase peptide synthesis. Enzyme kinetic studies
revealed that trisaccharide-Hyp and trisaccharide-peptide are
poorly recognized by AgtA, indicating the importance of
context provided by the native Skp1 protein for engage-
ment with the active site. The trisaccharide-peptide was
a potent immunogen capable of generating a rabbit antise-
rum that was highly selective toward the trisaccharide iso-
form of full-length Skp1.
Introduction
Glycosylation of Skp1 in the social amoeba Dictyostelium
was established by metabolic incorporation of [3H]Fuc,[3] and
confirmed by compositional analysis of the protein showing
the presence of N-acetyl-glucosamine (GlcNAc), fucose (Fuc),
and galactose (Gal).[4] Mass spectrometry based-sequencing
studies[4b] have shown that Skp1 is modified by a pentasacchar-
ide attached to a hydroxyproline (Hyp) residue at position 143.
Based on exoglycosidase digestions, the core trisaccharide has
the structure of the type 1 blood group H antigen and is modi-
fied by two additional a-linked Gal residues.
Skp1 is a small, highly conserved eukaryotic protein of 160–
190 amino acids that is proposed to play multiple cellular func-
tions. Skp1 is frequently complexed with the F-box domain of
F-box proteins (FBP), and the resulting heterodimers are often
incorporated into the SCF (Skp1, cullin-1, F-box protein, Roc1/
Rbx1/Hrt1) family of E3-ubiquitin ligases that is expressed uni-
versally in the cytoplasmic and nuclear compartments of eu-
karyotes.[1] Considering the multitude of FBPs, the E3SCF-Ub li-
gases are expected to target hundreds of proteins for polyubi-
quitination and subsequent degradation by the 26S-protea-
some.[2]
Assembly of the glycan of Skp1 requires the prior oxidation
of proline-143 by a cytoplasmic non-heme FeII-dependent diox-
ygenase - the prolyl 4-hydroxylase PhyA, to give a Hyp residue.
Modification of Hyp with a galactoside or arabinoside is
common in plants and certain algae[5] but, due to the localiza-
tion of the enzymes in the secretory compartment, these gly-
coproteins are secreted.[6] In Dictyostelium, the Skp1 Hyp143 is
modified by aGlcNAc mediated by the enzyme N-acetylgluco-
saminyltransferase-1 (Gnt1). Addition of the second and third
sugars of the Skp1 glycan is mediated by PgtA, which has dual
b3GalT/a2FucT activity resulting in the formation of the blood
group H type 1 antigen Fuca1,2Galb1,3GlcNAca-Hyp143. Two
a-galactoside residues are added by the enzyme AgtA, which
synthesizes the following pentasaccharide: Gala1,?Gala1,3Fu-
ca1,2Galb1,3GlcNAca-Hyp143. The linkage of the terminal
aGal residue is yet to be determined.
[a] Dr. Z. S. Chinoy, Prof. Dr. G.-J. Boons
Complex Carbohydrate Research Center and Department of Chemistry
University of Georgia, 315 Riverbend Road, Athens, GA 30602 (USA)
E-mail: gjboons@ccrc.uga.edu
[b] Dr. C. M. Schafer, Prof. Dr. C. M. West
Department of Biochemistry and Molecular Biology and Oklahoma Center
for Medical Glycobiology, University of Oklahoma Health Sciences Center
Oklahoma City, OK 73104 (USA)
[c] Dr. C. M. Schafer
Current address: Oklahoma Medical Research Foundation
825 N.E. 13th Street, Oklahoma City, OK 73104 (USA)
[d] Prof. Dr. C. M. West
Current address: Dept. of Biochemistry and Molecular Biology
University of Georgia, 120 Green Street, Athens, GA 30602 (USA)
In Dictyostelium, PhyA has important roles in O2-dependent
development. Genetically-induced changes in PhyA expression
levels cause inverse changes in the level of O2-required for cul-
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201501598.
Chem. Eur. J. 2015, 21, 11779 – 11787
11779
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
mination, the developmental process in which the multicellular
slug converts to a fruiting body.[7] Conversely, genetically-in-
duced changes in Skp1 expression levels cause a corresponding
change in the level of O2 required for culmination, in a way
that depends on the presence of Pro143.[8] Glycosyltransferase
mutants exhibit requirements for intermediate levels of O2,
with disruption of AgtA causing a particularly disruptive
effect,[9] implying a unique regulatory role for the trisaccharide
form of Skp1 or additional functions of AgtA. Although AgtA
can catalyze the addition of the first of the two aGal residues
to aFuc and Fuca1,2Galb- acceptors, evidence indicates that
the acceptor trisaccharide must reside in the context of full-
length Skp1 for efficient addition of the first aGal
and is absolutely required for addition of the second
aGal. This depends at least in part on second site rec-
ognition of Skp1 by the WD40-repeat domain in
AgtA.[9b] Further studies are needed to investigate
the role of the local peptide environment on confor-
mation of the glycan that may also be an important
determinant for glycan recognition. In vitro and in
vivo studies stress the importance of the glycan in
promoting interactions with F-box proteins,[10] which
may represent the mechanism by which hydroxyl-
ation and glycosylation of Skp1 mediate or modulate
O2-sensing during Dictyostelium development.
Results and Discussion
The target glycopeptide 1 was synthesized by first preparing
trisaccharide 5, which was modified by an appropriately pro-
tected Hyp residue to give, after a number of protecting group
manipulations, glycosylated amino acid 6, which was used in
solid-phase peptide synthesis (Figure 1). Late stage installation
of the Hyp moiety was attractive because this amino acid can
adopt cis/trans configurations complicating NMR analysis. It
was anticipated that trisaccharide 5 could readily be prepared
from monosaccharide building blocks 2, 3 and 4. The non-par-
ticipating azido moiety of 5 would allow the installation of the
Studies on O2-dependent hydroxylation of Skp1 in
cells, relative stability of Skp1 isoforms in cells, and
competitive interactions of Skp1 isoforms with bind-
ing partners, have all been made possible by iso-
form-specific antibodies that differentiate unmodi-
fied, hydroxylated, and GlcNAcylated forms of Skp1
from each other and all other isoforms.[10–11] We rea-
Figure 1. Strategy for the chemical synthesis of a glycopeptide derived from Skp1. L is
appropriate leaving group.
soned that availability of an antibody specific for the
trisaccharide form would be similarly exploitable for
related studies and to investigate the contingency of
Skp1 glycosylation in response to stress and during develop-
ment. The failure of available anti-blood group (type 1) anti-
bodies to react with Skp1 necessitates the generation of new
Abs.
Hyp moiety as an a-glycoside. Furthermore, Nap ethers were
selected as permanent protecting groups because they can be
readily removed by oxidation with 2,3-dichloro-5,6-dicyano-1,4-
benzoquinone (DDQ)[12] to give, after standard O-acetylation,
acetyl esters that are desirable for glycopeptide synthesis. The
more commonly employed benzyl ethers could not be em-
ployed because their removal by catalytic hydrogenation
would result in reduction of the azido moiety. Also, possible
late stage benzyl ether removal after installation of the Hyp
residue did not offer an attractive approach because their re-
moval is not compatible with the presence of an Fmoc pro-
tecting group.
Thus, a TMSOTf mediated glycosylation[13] of trichloroaceti-
midate 2a with acceptor 3 gave disaccharide 7a in a modest
yield of 32%, surprisingly as a mixture of anomers (Table 1,
entry 1). The low yield of coupling product was due to rear-
rangement of the trichloroacetimidate to give a trichloro-N-
acetamide adduct. Furthermore, despite the presence of a par-
ticipating acetyl ester at C-2 of the glycosyl donor, a mixture of
anomers was obtained. The use of acetonitrile as a participating
solvent[14] only marginally improved the yield and anomeric se-
lectivity of the glycosylation (entry 2). It is known that the use
We report here the chemical synthesis of a glycopeptide de-
rived from Skp1 that is modified by a blood group H type 1 tri-
saccharide, which was conjugated to KLH and employed for
antibody production in rabbits. A synthetic strategy was devel-
oped in which a trisaccharide was synthesized that was modi-
fied by Hyp to give, after a number of chemical manipulations,
a glycosylated amino acid that could be employed for glyco-
peptide synthesis. 2-Naphthylmethyl (Nap) ethers were em-
ployed as permanent protecting groups to allow late stage in-
stallation of the Hyp moiety. It was found that tuning glycosyl
donor and acceptor reactivities was critical for achieving high
yields and anomeric selectivities of the glycosylations. The suit-
ability of the novel glycopeptide for probing AgtA acceptor
substrate preference and to induce an isoform specific anti-
body is presented.
Chem. Eur. J. 2015, 21, 11779 – 11787
www.chemeurj.org
11780
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
of N-phenyl-trifluoroacetimidates[15] as glycosyl donors can sup-
press rearrangement, and indeed the use of glycosyl donor 2b
improved the yield of the glycosylation, however, disaccharide
7a was still obtained as a mixture of anomers (entries 3 and 4).
An N-iodosuccinimide (NIS)/trifluoromethanesulfonic acid
(TfOH) mediated coupling[16] of 2c with 3 in a mixture of aceto-
nitrile and dichloromethane at À308C gave the required disac-
charide 7a in an acceptable yield of 71% mainly as the b-
anomer (entries 5 and 6). The yield and anomeric selectivity
could be further improved by employing glycosyl donors 2d
or 2e, which have at C-2 a 2,5-difluorobenzoyl[17] or Lev[18]
esters, respectively to give disaccharides 7b and 7c (entries 7–
9). In each case, the use of acetonitrile as a cosolvent improved
the b-anomeric selectivity of the glycosylation. Probably, aceto-
nitrile stabilizes an acyloxonium ion intermediate thereby pro-
moting 1,2-trans anomeric selectivity. Alternatively, it may pro-
mote the formation of an a-nitrilium ion which upon displace-
ment by a sugar alcohol will give a b-glycoside. Disaccharides
7a-c were converted to glycosyl acceptor 8 by removal of the
acetyl or dFBz protecting groups of 7a and 7b, respectively by
using ZemplØn conditions, whereas the Lev ester of 7c was
cleaved using hydrazine acetate.
Table 2. Optimization fucosylation.
Entry Solvent
T
Activator
Yield [%] (Isolated) a/b (Isolated)
1
2
3
4
5
6
DCM/Et2O RT
IDCT
IDCT
NIS, TMSOTf 63
46
56
5:1
8.7:1
5:1
diox/tol
diox/tol
DCM
RT
08C
À308C NIS, TMSOTf 78
À308C NIS, TfOH 74
À108C NIS, TMSOTf 94
3.4:1
3:1
3.7:1
Et2O
Et2O
with DDQ has not been reported. Hydrolysis of naphthylidene
acetal of 5 using 10% TFA in DCM, subsequent acetylation in
acetic anhydride and pyridine, followed by DDQ oxidation in
a mixture of DCM and water gave the required compound in
low yield. Byproducts that were identified included a derivative
having a 4,6-O-naphthylidene acetal on the galactoside residue
and compounds having a naphthoate ester at C-6 or C-4 of
the galactoside moiety. The formation of these byproducts can
be rationalized by a mechanism in which the removal of the
first Nap ether of the galactoside moiety (C-4 or C-6) is fol-
lowed by oxidation of the second Nap ether to give an inter-
mediate carbocation that despite the presence of water can be
trapped by the neighboring alcohol to give a naphthylidene
acetal. The latter derivative can be further oxidized by DDQ to
form the naphthoate ester.
Next, attention was focused on the fucosylation of glycosyl
acceptor 8 using donor 4 to give trisaccharide 5 (Table 2). Vari-
ous promoters and solvent systems were examined, and it was
found that the most favorable anomeric selectivity was ach-
ieved when iodonium dicollidine triflate (IDCT)[19] was used as
the promoter in a mixture of dioxane in toluene[20] (Table 2,
entry 2). The best yield was, however, accomplished using NIS/
TMSOTf as the activator in diethyl ether at À108C (Table 2,
entry 6). The use of a corresponding trichloroacetimidate
donor activated by TMSOTf in dichloromethane at À308C gave
the trisaccharide 5 in a low yield of 11 % (data not shown).
Although Nap ethers are commonly employed as temporary
protecting groups,[12] their use for permanent alcohol protec-
tion is less well established and to the best of our knowledge
the removal of several of these functionalities by oxidation
These unexpected results led us to investigate an alternative
approach for the deprotection of the Nap ethers. Hydrogenoly-
sis using Pd/C could not be employed as the azido functionali-
ty, which would be required for the installation of an a-linked
Hyp moiety, would also be reduced under these conditions.
Lewis acids such as FeCl3 have found utility for the deprotec-
tion of benzyl ethers.[21] Hence, attention was turned to the
use of FeCl3 for the removal of the Nap ethers (Scheme 1).
Thus, trisaccharide 9 was subjected to anhydrous FeCl3 fol-
lowed by acetylation with acetic anhydride and pyridine,
which gratifyingly gave compound 10 (yield of 43%) in which
the azide functionality had remained intact and all the Nap
ethers cleaved together with the anomeric TDS protecting
group. The anomeric acetyl ester of 10 was selectively cleaved
with hydrazine acetate and the resulting lactol converted into
trichloroacetimidate 11 by treatment with trichloroacetonitrile
in the presence of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU). A
triflic acid promoted glycosylation of 11 with benzyl protected
N-a-(9-fluorenylmethyloxycarbonyl)-l-trans-4-hydroxyproline
gave compound 12 in a yield of 71% mainly as the a-anomer.
Reduction of the azido moiety of 12 was accomplished using
thioacetic acid[22] to provide 13 which was subjected to hydro-
genolysis to give 6, which is appropriately protected for use in
glycopeptide synthesis.
Table 1. Optimizing galactosylation.
Entry Donor Solvent
Activator Yield [%] (isolated) b/a (isolated)
1
2
3
4
5
6
7
8
9
2a
2a
2b
2b
2c
2c
2d
2d
2e
DCM
DCM/CH3CN TMSOTf
DCM TMSOTf
DCM/CH3CN TMSOTf
DCM
DCM/CH3CN NIS, TfOH 71
DCM NIS, TfOH 45
TMSOTf
32
43
77
67
2:1
3.7:1
2:1
4:1
NIS, TfOH 34
3.6:1
5.6:1
2:1
7.5:1
8.6:1
DCM/CH3CN NIS, TfOH 90
DCM/CH3CN NIS, TfOH 85
An alternative approach for the synthesis of 6 was explored
to avoid the loss of expensive trisaccharide during the depro-
Chem. Eur. J. 2015, 21, 11779 – 11787
www.chemeurj.org
11781
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
charide that could be more read-
ily deprotected but also greatly
facilitated the galactosylation
and fucosylation reactions. Spe-
cifically, the lower reactivity of
14 led to an improvement in
glycosylation yield and anomeric
selectivity. Furthermore, fucosy-
lation of the resulting acceptor
(16), which has also a lower re-
activity, also led to higher yield-
ing and stereoselective glycosy-
lation.
Glycopeptide 1 was synthe-
sized by microwave-assisted
solid phase glycopeptide synthe-
sis[23] using Rink Amide AM LL
resin (Scheme 3). The first eight
amino acids were introduced
using an HBTU-mediated HOBt
ester activation protocol (!19).
The glycosylated amino acid 6
was introduced manually using
microwave-assisted solid phase
peptide synthesis to give 20.
The resin was returned to the
automated peptide synthesizer
to further elongate the glyco-
peptide. Cleavage from the resin
was accomplished using 94%
TFA, 2.5% H2O, 2.5% EDT, and
1% TIPS to afforded glycopep-
tide 22 which was treated with
NaOMe buffered with guanidi-
Scheme 1. a) 10% TFA in DCM; b) Ac2O, pyridine; c) FeCl3, DCM; d) Ac2O, pyridine; e) H2NNHOAc, DMF; f) CCl3CN,
DBU, DCM; g) 9-fluorenylmethyloxycarbonyl-l-trans-4-hydroxyproline, TfOH, Et2O, 08C, 15 min; h) AcSH, pyridine;
i) 10 % Pd/C, DMF, H2.
Scheme 2. a) NIS, TfOH, DCM/CH3CN, À308C, 30 min; b) H2NNH·OAc, EtOH/toluene; c) 4, NIS, TMSOTf, Et2O,
À108C, 20 min; d) 10% TFA in DCM; e) Ac2O, pyridine; f) DDQ, DCM/H2O.
tection of the Nap ethers. It was envisaged that the use of gal-
actosyl donor 14, which has permanent acetyl esters at C-3, 4
and 6 and a temporary Lev ester at C-2, would pro-
vide a trisaccharide that can be more readily convert-
ed into target trisaccharide 6 (Scheme 2). Thus, a gly-
cosylation of 3 with 14 in the presence of NIS and
TfOH in a mixture of DCM and acetonitrile gave dis-
accharide 15 in a yield of 81% as only the b-anomer.
The Lev ester of 15 was removed using hydrazine
acetate to give glycosyl acceptor 16 which was cou-
pled with fucosyl donor 4 in diethyl ether at À108C
using NIS/TMSOTf as the activator system, to give the
trisaccharide 17 in a 94% yield as only the a-anomer.
Subsequent treatment of 17 with 10% TFA in DCM
to remove the naphthylidene acetal, acetic anhydride
and pyridine to acetylate the resulting alcohols, DDQ
to oxidatively remove the Nap ether, and finally
acetic anhydride and pyridine for conversion of the
resulting alcohol into acetyl ester, gave trisaccharide
18 in a much improved yield of 86% over four steps.
A remarkable observation was that replacement of
the ether protecting groups of galactosyl donor 2e
ne·HCl to remove the acetyl esters,[24] to give after purification
by C18 column chromatography, the target glycopeptide 1.
with acetyl esters (14) did not only lead to a trisac- Scheme 3. Assembly of Skp1 derived glycopeptide.
Chem. Eur. J. 2015, 21, 11779 – 11787
www.chemeurj.org 11782
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Substrate activity of the trisaccharide peptides toward AgtA
to modification by AgtA, or otherwise renders the glycan un-
recognizable to the AgtA active site.
Previous studies showed that the order of preference of au-
thentic and model acceptor substrates for AgtA is: FGGN-Skp1
> FG-pNP > F-pNP > FGGn-pNP, with little effect when the
pNP aglycon is substituted with Bn or octyl.[9b] The substrate
activities of FGGn-Hyp 23 (see Supporting Information for
structure and details) and FGGn-peptide 1 were analyzed for
ranking in this activity series, which might provide information
regarding the importance of nearby amino acids for enzyme
recognition and catalysis. Four methods of product detection
were developed, to help avoid misinterpretation that could
occur if substrate glycosylation interfered with detection.
Under the conditions tested, activity toward FGGn-Hyp 23 was
<10% that of FGGn-pNP (Figure 2B), and 1 exhibited even
lower activity (Figure 2D–F). In addition, a three-fold excess of
23 did not inhibit the reaction with F-pNP (Figure 2C). Further
descriptions and analysis of the results are included in the sup-
plement. The new acceptor activity hierarchy is: FGGN-Skp1 >
FG-pNP > F-pNP > FGGn-pNP > FGGn-Hyp > FGGn-peptide.
The inhibitory nature of the peptide and Hyp, relative to pNP,
suggests that the local amino acids contribute to a conforma-
tion of the glycan that renders the 3-OH of Fuc less accessible
Development of Skp1 specific antibodies
Glycoform-specific antibodies (Abs) are desirable for studying
regulation of Skp1 glycosylation. To develop an Ab specific for
the trisaccharide isoform of Skp1, FGGn-peptide 1 was conju-
gated to maleimide activated keyhole limpet hemocyanin
(mcKLH) to give a conjugate carrying 641 glycan residues and
used to immunize two rabbits. Ab specificity was examined by
Western blotting using a panel of mutant cells uniquely ex-
pressing six of the seven isoforms of Skp1 generated by the
Skp1 modification pathway. As shown in Figure 3A, pAb
UOK104 exhibited a high degree of selectivity for trisaccharide
(FGGn) modified Skp1, in the absence of affinity selection, and
minor detection of pentasaccharide (GGFGGn) modified Skp1.
pAb UOK105 exhibited similar, but weaker reactivity to all Skp1
isoforms (not shown).
To test the basis for specific recognition of FGGn-Skp1, pAb
UOK104 was tested for binding of a BSA-conjugate modified
by lacto-N-fucopentaose-I (LNFP-I, Fuca1,2Galb1,3GlcNAcb1,3-
Galb1,4Glc-) which is terminated by the same H-type trisac-
charide as FGGn-Skp1 except the
proximal sugar GlcNAc is in a b-
linkage. As a negative control
a BSA conjugate of lacto-N-fuco-
pentaose-III (LNFP-III, Galb1,3(Fu-
ca1,3)GlcNAcb1,3Galb1,4Glc-)
which has the same composition
but different arrangement of the
monosaccharides, was com-
pared. Binding of UOK104 was
not detected to either LNFP-I or
LNFP-III (Figure 3B). As a positive
control, LNFP-I was specifically
recognized by ab3355, a mAb
that
recognizes
Fuca1,2-
Galb1,3GlcNAcb1-, validating the
assay.[25] Recognition of enzymat-
ically assembled FGGn-Skp1 by
an antibody raised against the
synthetic glycopeptide supports
the linkage assignments used to
design the glycopeptide. Further
studies are needed to determine
whether the epitope(s) recog-
nized by UOK104 is a peptide
Figure 2. Acceptor activities of synthetic AgtA substrates. A) Comparison of AgtA activity towards F-pNP and
FGGn-pNP. Reactions were prepared with 10 mm UDP-[3H]Gal (2500 mCi per mmol) and His6 AgtA, and incubated for conformation imposed by the
60 min. Incorporation was determined using method 1. B) Activity towards FGGn-pNP and FGGn-Hyp 23 were de-
presence of the glycan, or con-
termined in the presence of 5 mm UDP-[3H]Gal (10000 mCi/mmol) and of 0.6”His6 AgtA (relative to panel A) for
sists jointly of both glycan and
60 min. Incorporation was determined using method 2. C) The effect of 23 on the 60 min reaction of 0.06”Hi-
s6 AgtA with 0.5 mm F-pNP in the presence of 5 mm UDP-[3H]Gal (5000 mCi per mmol), using method 2. D) Activity
towards FGGn-pNP and FGGn-peptide 1 was analyzed using 10 mm UDP-[3H]Gal (2500 mCi per mmol) and 1”Hi-
s6 AgtA for 180 min. Incorporation was determined using method 1. E) Similar to D, except that 120 min reactions
containing 10 mm UDP-[3H]Gal (5000 mCi/mmol) and 1.5”His6 AgtA were analyzed using method 3. F) Comparison
of activities towards 1 coupled to KLH and FGGn-Skp1. Reactions were conducted using 5 mm UDP-[3H]Gal
(10000 mCi per mmol), the indicated concentrations of FGGn-peptide (0.66 mgmLÀ1 KLH-conjugate) or FGGn-Skp1,
and 1” or 0.003”His6 AgtA, respectively. Incorporation was determined using method 4. Error bars indicate ÆSD.
peptide determinants.
Conclusion
Reports dealing with the chemi-
cal synthesis of glycopeptides
Chem. Eur. J. 2015, 21, 11779 – 11787
www.chemeurj.org
11783
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
by pAb UOK104 as there was no
recognition of LNFP I-BSA, which
contains the same Skp1 glycan
attached to BSA. However, the
possibility that UOK104 recog-
nizes the glycan in a conforma-
tion induced by the Skp1 pep-
tide cannot be excluded. pAb
UOK104 is expected to be useful
for studying the regulation of
glycosylation in Dictyostelium,
and the selective detection of
FGGn-Skp1 in mixtures with
other isoforms.
Figure 3. Specificity of anti-59/KLH UOK104 antibody. A) Western blot analysis of a panel of Dictyostelium mutants
that accumulate the indicated Skp1 glycoforms. Panels were probed with either pAb UOK104 or UOK77 (pan-spe-
cific for all Skp1 isoforms) at 1:1000 dilution. Note: An irregularity during placement of the gel on the blot mem-
brane resulted in distortion of relative mobilities of the Skp1 isoforms. B) The glycan specificity of UOK104 was
tested against BSA conjugates of LNFP-I and LNFP-III, which were adsorbed in solutions containing the indicated
pmol of glycan equivalents in 50 mL to wells of a 96-well ELISA plate. Coated wells were probed with UOK104
(1:500) or mAb ab3355 (1:10), with specificity for the blood group H type 1 antigen present in LNFP-I/BSA.
Experimental Section
General procedure for trichloroa-
cetimidate glycosylations: Ac-
ceptor 3 and donor (2a or 2b)
were dissolved in an appropriate
solvent (DCM or a 1:1 mixture of DCM/CH3CN), and molecular
sieves (4 Š) were added. The mixture was stirred for 1 h at room
temperature under an atmosphere of argon. The mixture was
cooled to À308C and TMSOTf (0.2 equiv) was added. The reaction
mixture was left stirring at À308C for 15 min, after which it was di-
luted with DCM, filtered and the filtrate washed with a saturated
solution of NaHCO3 and water, dried (MgSO4), filtered and the fil-
trate was concentrated in vacuo. The resulting residue was purified
by silica gel column chromatography (hexanes/EtOAc) to afford
the target compound.
that are glycosylated at Hyp deal mainly with compounds
having simple monosaccharide additions.[26] The preparation of
glycopeptides having a complex glycan moiety linked to this
amino acid has received even less attention, and to the best of
our knowledge efforts have only been focused on the prepara-
tion of compounds derived from plants, which have very differ-
ent oligosaccharide compositions compared to the glycopepti-
des described in this study. The properly protected trisacchar-
ide-Hyp 6 was synthesized in excellent overall yield using
a linear glycosylation strategy combined with the selection of
appropriate protecting groups. The use of Nap ethers as a per-
manent protecting group made it possible to install a hydroxy-
proline moiety at a late stage of synthesis. However, it was crit-
ical to avoid Nap ethers that were 1,2-cis to an alcohol due to
unexpected byproduct formation. Furthermore, it was found
that the application of galactosyl donor of low reactivity
having acetyl esters at C-3, C-4 and C-6 and a temporary Lev
ester at the C-2, gave markedly better yields and anomeric se-
lectivities in glycosylations compared to the use of an ether
protected and more reactive donor. Surprisingly, removal of
the Lev ester of the resulting glycosylation product also pro-
vided a superior acceptor. The target glycopeptide was synthe-
sized using automated microwave-assisted solid-phase peptide
synthesis using Rink Amide AM LL resin. The in vitro substrate
dependence of AgtA revealed that galactose was transferred
to trisaccharide-Hyp 23, however it was a poor substrate and
the extended glycopeptide 1 was even less active. These re-
sults are consistent with folding of the acceptor hydroxyl of
Fuc into a trisaccharide-dependent glycan conformation that is
not recognized by AgtA.[27] The underlying hydroxyproline or
glycopeptide may stabilize this unfavorable conformation lead-
ing to even lower activity levels towards these substrates.
The glycopeptide proved to be an excellent immunogen
when conjugated to the carrier protein KLH, for inducing
a highly glycoform-specific antiserum toward FGGn-Skp1. The
Skp1 peptide appears to play an important role in recognition
General procedure for thioglycoside glycosylations: Acceptor 3
and donor (2c, 2d, 2e, or 14) were dissolved in an appropriate sol-
vent (DCM or a 1:1 mixture of DCM/CH3CN). Molecular sieves (4 Š)
were added and the mixture was stirred for 1 h at room tempera-
ture under an atmosphere of argon. NIS was added and the result-
ing solution was cooled to À308C, followed by the addition of
TfOH. The reaction mixture was stirred in the dark at À308C for
15 min. The reaction mixture was diluted with DCM, filtered into
a solution of sodium thiosulfate and stirred until the solution
turned colorless. The organic layer was washed with a saturated
solution of NaHCO3 and water, dried (MgSO4), filtered and the fil-
trate was concentrated in vacuo. The resulting residue was purified
by silica gel column chromatography (hexanes/EtOAc) to afford
the disaccharides 7a–7c and 15.
General methods for automated microwave-assisted solid-phase
peptide synthesis (MW-SPPS): (Glyco)peptides were synthesized
by established protocols on a CEM Liberty Automated Microwave
Peptide Synthesizer equipped with a UV detector using N-a-Fmoc-
protected amino acids and 2-(1H-benzotriazole-1-yl)-1,1,3,3-tetra-
methyluronium hexafluorophosphate (HBTU)/1-hydroxybenzotria-
zole (HOBt) as the activating reagents. The compounds were pre-
pared on a Rink Amide AM LL resin using the following amino acid
building blocks: Fmoc-Ile-OH, Fmoc-Gln(Trt)- OH, Fmoc-Glu(OtBu)-
OH, Fmoc-Hyp(tBu)-OH, Fmoc-Thr(tBu)-OH, Fmoc-Phe-OH, Fmoc-
Asp(OtBu)-OH, Fmoc-Asn(Trt)-OH, Fmoc-Arg(Pbf)-OH and Fmoc-
Lys(Boc)-OH. Double couplings were employed for all amino acids.
Deprotection of the N-a-Fmoc was achieved using 20% 4-methyl
piperidine in DMF.
Dimethylthexylsilyl [3,4,6-tri-O-acetyl-b-d-galactopyranosyl]-(1!
3)-4,6-O-(2-napthylidene)-2-deoxy-azido-b-d-glucopyranoside
Chem. Eur. J. 2015, 21, 11779 – 11787
www.chemeurj.org
11784
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
(16): To a solution of compound 15 (1.91 g, 1.637 mmol) in EtOH
(80 mL) and toluene (40 mL), hydrazine acetate (251 mg,
2.78 mmol) was added and the resulting reaction mixture was left
stirring at room temperature for 1 h. The reaction mixture was con-
centrated under reduced pressure and the resulting residue was
purified by silica gel column chromatography (hexanes/EtOAc 2:1,
v:v) to give 16 (1.64 g, 94%) as an amorphous white solid.[15]
1H NMR (CDCl3, 300 MHz): d = 0.00 (t, J=3.9 Hz, 6H, CH3-Si-CH3),
0.71 (s, 12H, TDS-(CH3)2C-C(CH3)2), 1.46 (dd, J=13.7, 6.8 Hz, 1H,
TDS-CH), 1.59 (s, 3H, COCH3), 1.80–1.81 (m, 3H, COCH3), 1.90 (s,
3H, COCH3), 2.87 (s, 1H, OH), 3.22–3.29 (m, 2H, GlcN H-5, GlcN H-
2), 3.56–3.76 (m, 6H, Gal H-5, GlcN H-4, GlcN H-3, Gal H-2, GlcN H-
6a, Gal H-6a), 3.90 (dd, J=11.1, 7.5 Hz, 1H, Gal H-6a), 4.14 (dd, J=
10.5, 5.0 Hz, 1H, GlcN H-6a), 4.39 (d, J=7.8 Hz, 1H, Gal H-1), 4.46
(d, J=7.6 Hz, 1H, GlcN H-1), 4.71 (dd, J=10.3, 3.4 Hz, 1H, Gal H-3),
5.12 (d, J=3.3 Hz, 1H, Gal H-4), 5.51 (s, 1H, Naphthylidene-H),
7.27–7.30 (m, 2H, 2”aromatic CH), 7.36 (dd, J=8.5, 1.5 Hz, 1H, aro-
matic CH), 7.62–7.67 (m, 3H, 3”aromatic CH), 7.72 (s, 1H, aromatic
CH); 13C NMR (75 MHz; CDCl3): d = 0.00, 1.03, 21.54, 21.65, 22.95,
23.08, 23.57, 23.77, 23.87, 27.97, 37.00, 64.37, 69.62, 70.11, 71.05,
71.71, 72.90, 74.39, 75.38, 82.80, 83.00, 100.80, 104.68, 107.03,
126.58, 128.53, 129.46, 129.73, 130.91, 131.44, 131.49, 135.97,
136.85, 137.31, 173.27, 173.32, 173.42; MALDI: m/z: calcd for
C37H51N3NaO13Si: 796.3089; found 796.3134 [M+Na]+.
deoxy-azido-b-d-glucopyranoside (18): To a cooled solution of tri-
saccharide 17 (2.79 g, 2.44 mmol) in a mixture of DCM (100.8 mL)
and water (10 mL), trifluoroacetic acid (11.2 mL) was added. The re-
action mixture was stirred for 1 h after which it was neutralized
with triethylamine, concentrated under reduced pressure and the
residue azeotropically dried with toluene (3”15 mL). The resulting
residue was dissolved in DCM (50 mL), pyridine (6 mL) and acetic
anhydride (6 mL) and the resulting mixture was left stirring for
18 h. The reaction mixture was cooled (08C) and quenched with
MeOH, concentrated in vacuo and azeotropically dried with tolu-
ene (4”20 mL) to give the product (~2.96 g), which was used
without further purification. To a solution of the crude product (~
2.96 g, 2.715 mmol) in a mixture of DCM (66 mL) and water
(6.6 mL), 2,3-dichloro-5,6-dicyano-p-benzoquinone (DDQ) (1.2 g,
5.43 mmol) was added and the reaction mixture was left stirring
vigorously for 80 min. The reaction mixture was concentrated in
vacuo. The resulting residue was dissolved in DCM (50 mL), acetic
anhydride (20 mL) and pyridine (20 mL) and the mixture left stir-
ring for 24 h. The reaction mixture was cooled (08C) and quenched
with MeOH, concentrated in vacuo and diluted with EtOAc,
washed with a saturated aqueous solution of NaHCO3 and water.
The organic layer was dried (MgSO4), filtered, and the filtrate was
concentrated in vacuo. The resulting residue was purified by silica
gel column chromatography (hexanes/EtOAc 1:1, v:v) to afford the
1
product 18 (2.08 g, 86% over 4 steps). H NMR (CDCl3, 500 MHz): d
Dimethylthexylsilyl [3,4-di-O-acetyl-2-O-(2-methylnaphthyl)-b-l-
fucopyranosyl]-(1!2)-[3,4,6-tri-O-acetyl-b-d-galactopyranosyl]-
(1!3)-4,6-O-(2-napthylidene)-2-deoxy-azido-b-d-glucopyrano-
side (17): To a solution of donor 4 (2.24 g, 5.168 mmol) and accept-
or 16 (2 g, 2.584 mmol) in anhydrous Et2O (52 mL), molecular
sieves (4 Š) were added and the resulting solution was left stirring
for 1 h at room temperature under an atmosphere of argon. NIS
(1.45 g, 6.46 mmol) was added, and the resulting solution was
cooled to À108C, followed by the addition of a TMSOTf (46.8 mL).
The reaction mixture was stirred in the dark at À108C for 20 min
after which it was diluted with DCM and filtered into a solution of
sodium thiosulfate and the solution was stirred until it turned col-
orless. The organic layer was extracted and washed with a saturated
solution of NaHCO3 and water, dried (MgSO4), filtered and the fil-
trate was concentrated in vacuo. The resulting residue was purified
by silica gel column chromatography (hexanes/EtOAc 2:1, v:v) to
= 0.00 (s, 6H, CH3-Si-CH3), 0.68 (s, 12H, TDS-(CH3)2C-C(CH3)2), 1.02
(d, J=6.3 Hz, 3H, Fuc CH3), 1.47 (t, J=6.5 Hz, 1H, TDS-CH), 1.75
(dd, J=6.0, 0.9 Hz, 6H, 2”COCH3), 1.78 (s, 3H, COCH3), 1.84 (d, J=
0.8 Hz, 9H, 3”COCH3), 1.88 (d, J=0.9 Hz, 3H, COCH3), 1.95 (d, J=
0.9 Hz, 3H, COCH3), 3.07 (dd, J=9.5, 8.3 Hz, 1H, GlcN H-2), 3.40
(dd, J=9.5, 4.3 Hz, 1H, GlcN H-5), 3.46 (t, J=9.8 Hz, 1H, GlcN H-3),
3.59 (t, J=8.7 Hz, 1H, Gal H-2), 3.63 (t, J=6.8 Hz, 1H, Gal H-5), 3.81
(dd, J=10.6, 8.0 Hz, 1H, Gal H-6a), 3.93–3.97 (m, 3H, GlcN H-6a,
GlcN H-6b, Gal H-6b), 4.32 (dd, J=15.8, 7.1 Hz, 2H, Fuc H-5, GlcN
H-1), 4.60 (t, J=9.7 Hz, 2H, Gal H-1, GlcN H-4), 4.79–4.82 (m, 2H,
Fuc H2, Gal H-3), 5.07 (d, J=3.3 Hz, 1H, Gal H-4), 5.12 (dt, J=8.6,
3.9 Hz, 3H, Fuc H-3, Fuc H-4, Fuc H-1); 13C NMR assigned from
HSQC (125 MHz, CDCl3): d = 0.33 (”2), 18.59, 21.25, 22.91, 23.57 (”
8), 36.85, 63.75 (”2), 65.74, 67.74, 70.06, 70.73, 71.06, 71.39 (”2),
73.38, 74.05, 74.71, 75.04, 76.37, 79.03, 98.95, 100.28, 103.26;
MALDI: m/z: calcd for C42H65N3NaO22Si: 1014.3727; found:
1014.1285 [M+Na]+.
1
afford the trisaccharide 17 (2.79 g, 94%) as the a anomer. H NMR
(CDCl3, 500 MHz): d = 0.14 (d, J=13.88 Hz, 6H, CH3-Si-CH3), 0.70 (s,
12H, TDS-(CH3)2C-C(CH3)2), 0.94 (d, J=6.4 Hz, 3H, Fuc CH3), 1.47 (t,
J=6.7 Hz, 1H, TDS-CH), 1.61 (dd, J=11.3, 5.4 Hz, 6H, 2”COCH3),
1.74 (d, J=8.5 Hz, 6H, 2”COCH3), 1.89 (s, 3H, COCH3), 3.23–3.28
(m, 3H, Gal H-5, GlcN H-2, GlcN H-5), 3.58–3.76 (m, 5H, GlcN H-3,
GlcN H-6a, Fuc H-2, GlcN H-4, Gal H-6a), 3.80–3.86 (m, 2H, Gal H-
6b, Gal H-2), 4.10 (dd, J=10.5, 4.8 Hz, 1H, GlcN H-6b), 4.43–4.55
(m, 4H, Fuc H-5, CHHNap, CHHNap, GlcN H-1), 4.72 (d, J=7.9 Hz,
1H, Gal H-1), 4.83 (dd, J=9.7, 3.2 Hz, 1H, Gal H-3), 5.00 (d, J=
3.0 Hz, 1H, Gal H-4), 5.08 (d, J=2.9 Hz, 1H, GlcN H-4), 5.17–5.21 (m,
2H, Fuc H-1, Fuc H-3), 5.46 (s, 1H, Naphthylidene-H), 7.16 (d, J=
8.3 Hz, 1H, aromatic CH), 7.24–7.31 (m, 4H, 4”aromatic CH), 7.40
(d, J=8.5 Hz, 1H, aromatic CH), 7.49 (s, 1H, aromatic CH), 7.56–
7.70 (m, 7H, 7”aromatic CH). MALDI: 13C NMR assigned from
HSQC (125 MHz, CDCl3): d = 0.00, 18.59, 20.58, 22.24, 22.90 (”3),
36.18, 63.07, 67.72, 69.38, 70.38, 71.04 (”2), 72.04, 72.37, 74.03 (”
2), 75.35, 76.02, 76.35, 79.67, 82.33, 98.92, 99.59, 102.91, 104.57,
126.15, 127.81, 128.47 (”4), 128.80 (”2), 130.13 (”2), 130.46 (”3);
MALDI: m/z: calcd for C58H73N3NaO19Si: 1166.4505; found:
1166.6457 [M+Na]+.
[2,3,4-Tri-O-acetyl-b-l-fucopyranosyl]-(1!2)-[3,4,6-tri-O-acetyl-b-
d-galactopyranosyl]-(1!3)-4,6-di-O-acetyl-2-deoxy-azido-b-d-
glucopyranosyl-2,2,2-trichloroacetimidate
(11):
HF/pyridine
(16.6 mL) was added drop wise to a cooled (08C) solution of the
trisaccharide 18 (2.08 g, 2.1 mmol) in pyridine (40 mL) in an HDPE
container. The reaction mixture was stirred at 08C for 30 min and
then left stirring for an additional 3 h at room temperature. The re-
action mixture was diluted with DCM, cooled (08C) and a solution
of NaHCO3 was added and left stirring for 40 min. The organic
layer was washed with a saturated aqueous solution of NaHCO3
and water. The organic layer was dried (MgSO4), filtered, and the
filtrate was concentrated in vacuo. The resulting residue was puri-
fied by silica gel column chromatography (hexanes/EtOAc 1:2, v:v)
to afford the hemiacetal (1.69 g, 95%) as an amorphous white
solid. Trichloroacetonitrile (0.62 mL, 6.061 mmol) and DBU (44.1 mL,
0.242 mmol) were added sequentially to a solution of the hemiace-
tal (1.03 g, 1.21 mmol) in DCM (22 mL) and the reaction mixture
was stirred under an atmosphere of Ar for 18 h. The reaction mix-
ture was concentrated in vacuo and the resulting residue was puri-
fied by silica gel column chromatography (hexanes/EtOAc 1:1, v:v)
to afford donor 11 (860 mg, 72%) as an amorphous white solid.
Dimethylthexylsilyl [2,3,4-tri-O-acetyl-b-l-fucopyranosyl]-(1!2)-
[3,4,6-tri-O-acetyl-b-d-galactopyranosyl]-(1!3)-4,6-di-O-acetyl-2-
Chem. Eur. J. 2015, 21, 11779 – 11787
www.chemeurj.org
11785
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1H NMR (CDCl3, 500 MHz): d = 1.17 (t, J=6.9 Hz, 3H, Fuc CH3), 1.93
(s, 9H, 3”COCH3), 1.99 (d, J=11.3 Hz, 6H, 2”COCH3), 2.04 (d, J=
4.4 Hz, 6H, 2”COCH3), 2.09 (d, J=6.5 Hz, 3H, COCH3), 3.62 (dd, J=
10.4, 3.5 Hz, 1H, GlcN H-2), 3.80 (dd, J=9.8, 7.7 Hz, 1H, Gal H-2),
3.86 (t, J=6.6 Hz, 1H, Gal H-5), 3.98–4.19 (m, 6H, Gal H-6a, GlcN H-
6a, GlcN H-6b, GlcN H-5, Gal H-6b, GlcN H-3), 4.47 (q, J=6.5 Hz,
1H, Fuc H-5), 4.69 (d, J=7.6 Hz, 1H, Gal H-1), 4.92–5.00 (m, 3H,
Fuc H-2, Gal H-3, GlcN H-4), 5.24 (d, J=3.4 Hz, 1H, Gal H-4), 5.27–
5.29 (m, 2H, Fuc H-4, Fuc H-3), 5.34 (d, J=3.8 Hz, 1H, Fuc H-1),
6.48 (d, J=3.4 Hz, 1H, GlcN H-1), 8.80 (s, 1H, NH); 13C NMR as-
signed from HSQC (125 MHz, CDCl3): d = 15.39, 20.70 (”8), 61.21,
61.55, 61.88 (”2), 62.87, 64.87, 67.52 (”2), 67.85, 68.52, 70.51,
70.84, 71.17, 71.51, 73.50, 74.49, 94.41, 95.75, 100.39; MALDI: m/z:
calcd for C36H47Cl3N4NaO22: 1015.1645; found: 1015.1096 [M+Na]+.
glucopyranosyl]-proline benzyl ester (13): Thiol acetic acid
(10 mL) was added to a solution of compound 12 (470 mg,
0.3686 mmol) in pyridine (10 mL) and the reaction mixture was
stirred at room temperature for 24 h. The reaction mixture was di-
luted with EtOAc, washed with a saturated solution of NaHCO3 and
a solution of CuSO4, dried (MgSO4), filtered and the filtrate was
concentrated in vacuo. The resulting residue was purified by silica
gel column chromatography (hexanes/EtOAc 3:7, v:v) to afford 13
(360 mg, 75%). 1H NMR (CDCl3, 500 MHz): d = 1.21–1.26 (m, 3H,
Fuc CH3), 1.97–2.14 (m, 27H, 9”COCH3), 2.20–2.25 (m, 1H, HyPro-
bHa), 2.57 (d, J=0.6 Hz, 1H, HyPro-bHb), 3.65 (d, J=10.7 Hz, 0.5H,
1
HyPro-dHa = ), 3.71–3.75 (m, 2.5H, Gal H-2, HyPro-dHb, HyPro-
2
1
dHa = ), 3.82 (t, J=6.6 Hz, 1H, Gal H-5), 3.92 (q, J=8.8 Hz, 2H,
2
1
GlcNAc H-3, GlcNAc H-5), 3.99 (t, J=8.9 Hz, 1.5H, Fmoc-eH = , Gal H-
2
6a), 4.11 (d, J=11.9 Hz, 1H, GlcNAc H-6a), 4.17–4.26 (m, 3.5H, Gal
N-a-(9-Fluorenylmethyloxycarbonyl)-l-trans-4-O-[[2,3,4-tri-O-
acetyl-b-l-fucopyranosyl]-(1!2)-[3,4,6-tri-O-acetyl-b-d-galacto-
pyranosyl]-(1!3)-4,6-di-O-acetyl-2-deoxy-azido-a-d-glucopyra-
nosyl]-proline benzyl ester (12): To a solution of the donor 11
(1.08 g, 1.0846 mmol) and acceptor N-a-(9-fluorenylmethyloxycar-
bonyl)-l-trans-4-hydroxyproline (370 mg, 0.834 mmol) in Et2O
(24 mL), molecular sieves (4 Š) were added and the setup left to
stir for 1 h at room temperature under an atmosphere of argon.
The reaction flask was then cooled to 08C and TfOH (7.4 mL,
0.0834, as a 10:1 dilution in Et2O) was added and the reaction left
to stir at 08C for 15 min. The reaction mixture was diluted with
DCM, filtered and the filtrate washed with a saturated solution of
NaHCO3 and water, dried (MgSO4), filtered and the filtrate was con-
centrated in vacuo. The resulting residue was purified by silica gel
column chromatography (hexanes/EtOAc 1:1, v:v) to afford the
product 12 (757 mg, 71%) as a separable 5:1 a/b mixture respecti-
vely.1H NMR (C6H6, 500 MHz): d = 1.34–1.37 (m, 3H, Fuc CH3), 1.42
(t, J=4.6 Hz, 3H, COCH3), 1.56 (s, 3H, COCH3), 1.61 (d, J=2.3 Hz,
3H, COCH3), 1.71–1.75 (m, 6H, 2”COCH3), 1.79–1.94 (m, 10H, 3”
COCH3, HyPro-bHa), 2.04–2.07 (m, 1H, HyPro-bHb), 3.07–3.17 (m,
1H, Gal H-5), 3.22–3.26 (m, 1H, GlcN H-2), 3.51 (t, J=5.8 Hz, 0.5H,
HyPro-dHa), 3.57–3.61 (m, 1H, HyPro-dHb), 3.74–3.79 (m, 1H, Gal
H-6a), 3.86–3.92 (m, 3H, Gal H-6a, HyPro-gH, HyPro-dHa, Fmoc-
1
H-6b, GlcNAc H-6b, Fmoc CHH, Fmoc-eH = ), 4.33–4.37 (m, 1.5H,
2
1
1
GlcNAc H-2, HyPro-gH = ), 4.43–4.52 (m, 3.5H, Gal H-1, HyPro-gH = ,
2
2
1
1
Fmoc CH = H, Fuc H-5, HyPro-aH = ), 4.62 (d, J=7.4 Hz, 1H, HyPro-
2
2
1
1
aH = , Fmoc CH = H), 4.79 (d, J=3.2 Hz, 1H, GlcNAc H-1), 4.86–4.87
2
2
(m, 1H, GlcNAc H-4), 4.94 (td, J=9.8, 3.1 Hz, 2H, Gal H-3, Fuc H-2),
1
5.07 (d, J=12.1 Hz, 0.5H, BnCH = H), 5.14 (dd, J=21.6, 8.8 Hz, 1H,
2
1
BnCHH), 5.24 (dd, J=14.1, 10.5 Hz, 2.5H, BnCH = H, Fuc H-1, Gal H-
2
4), 5.35 (s, 1H, Fuc H-4), 5.47 (dd, J=10.8, 3.3 Hz, 1H, Fuc H-3),
7.31 (t, J=6.9 Hz, 6H, 6”aromatic CH), 7.41–7.57 (m, 5H, 5”aro-
matic CH), 7.78 (d, J=6.7 Hz, 2H, 2”aromatic CH); 13C NMR
(125 MHz; C6D6): d = 14.05, 14.20, 15.36, 19.62, 19.92, 20.10, 20.20,
20.39, 20.49, 20.56, 22.95, 23.06, 23.10, 29.65, 29.95, 30.04, 32.17,
36.14, 36.22, 37.23, 47.30, 52.56, 55.57, 58.21, 59.92, 60.64, 60.89,
62.41, 62.60, 65.61, 66.98, 67.59, 67.80, 67.88, 67.91, 69.23, 69.95,
70.15, 72.46, 73.33, 73.82, 74.01, 77.30, 96.81, 97.62, 99.02, 100.85,
120.18, 125.05, 125.16, 125.19, 125.31, 125.49, 127.21, 127.32,
128.61, 135.69, 135.87, 141.48, 141.60, 143.77, 143.96, 144.60,
154.98, 168.97, 169.63, 169.85, 169.93, 170.00, 170.12, 170.22,
170.29, 170.54, 171.75; MALDI: m/z: calcd for C63H74N2NaO27:
1313.4377; found: 1313.5216 [M+Na]+.
N-a-(9-Fluorenylmethyloxycarbonyl)-l-trans-4-O-[[2,3,4-tri-O-
acetyl-b-l-fucopyranosyl]-(1!2)-[3,4,6-tri-O-acetyl-b-d-galacto-
pyranosyl]-(1!3)-4,6-di-O-acetyl-2-deoxy-2-(N-acetamido)-a-d-
glucopyranosyl]-proline (6): 10% Pd on activated carbon (30 mg)
was added to a solution of 13 (151 mg, 0.117 mmol) in DMF
(3.8 mL), and the mixture stirred for 20 min at room temperature.
The argon was replaced with H2, and the reaction mixture was
stirred for 2.5 h. The reaction mixture was filtered through celite,
and the filtrate was concentrated in vacuo. The resulting residue
was purified by silica gel column chromatography (CHCl3/MeOH/
AcOH, 99:2:0.2, v:v:v) to afford compound 6 (119 mg, 85%) as an
1
eH = ), 3.94–3.98 (m, 1H, GlcN H-5), 4.07 (t, J=6.9 Hz, 0.5H, Fmoc-
2
1
eH = ), 4.12 (td, J=6.8, 3.2 Hz, 2H, Gal H-2, GlcN H-6a), 4.22–4.38 (m,
2
4H, GlcN H-6b, GlcN H-3, Fmoc-CH2), 4.51 (t, J=3.5 Hz, 1H, HyPro-
1
1
1
aH = , GlcN H-1 = ), 4.56 (d, J=3.5 Hz, 0.5H, GlcN H-1 = ), 4.62–4.69
2
2
2
1
(m, 1.5H, Gal H-1, HyPro-aH = ), 4.85–4.90 (m, 1.5H, Fuc H-5, Bn
2
1
1
CHH = ), 4.96–5.05 (m, 1.5H, Bn-CHH = ), 5.07–5.12 (m, 1H, GlcN H-4),
2
2
5.20 (dt, J=9.4, 4.4 Hz, 1H, Gal H-3), 5.28 (dd, J=10.1, 3.3 Hz, 1H,
Gal H-4), 5.43 (ddd, J=10.6, 6.1, 4.2 Hz, 1H, Fuc H-2), 5.73 (d, J=
3.2 Hz, 1H, Fuc H-4), 5.87 (d, J=3.84 Hz, 1H, Fuc H-1), 5.94 (ddd,
J=23.4, 11.0, 3.2 Hz, 1H, Fuc H-3), 6.93–7.05 (m, 3H, 3”aromatic
CH), 7.15–7.28 (m, 6H, 6”aromatic CH), 7.42–7.60 (m, 4H, 4”aro-
matic CH); 13C NMR (125 MHz; C6D6): d = 15.58, 19.53, 19.87, 19.97,
20.14, 20.17, 20.31, 20.35, 20.39, 20.48, 35.91, 37.02, 47.48, 47.54,
51.80, 52.15, 58.00, 58.42, 60.44, 62.38, 62.83, 65.18, 65.23, 66.85,
66.91, 67.03, 67.62, 68.26, 68.45, 68.50, 69.01, 69.05, 69.33, 70.57,
71.68, 71.76, 71.87, 73.98, 74.02, 74.63, 75.08, 76.12, 78.13, 95.93,
97.29, 97.95, 100.79, 101.05, 120.07, 120.17, 120.19, 125.27, 125.39,
127.16, 127.23, 127.29, 127.34, 127.74, 128.33, 128.42, 128.50,
128.59, 128.62, 135.83, 136.01, 141.58, 141.61, 141.71, 144.27,
144.35, 144.42, 144.55, 154.25, 154.66, 169.10, 169.33, 169.71,
169.77, 169.81, 169.87, 170.03, 170.07, 170.25, 170.29, 170.61,
171.88, 171.98; MALDI: m/z: calcd for C61H70N4NaO26: 1297.4176;
found: 1297.4746 [M+Na]+.
1
amorphous white solid. H NMR (CDCl3, 500 MHz): d = 1.21 (s, 3H,
Fuc CH3), 1.99–2.11 (m, 27H, 9”COCH3), 2.26–2.53 (m, 2H, HyPro-
1
bHa, HyPro-bHb), 3.60 (d, J=10.8 Hz, 0.5H, HyPro-dHa = ), 3.72 (dt,
2
1
J=17.3, 8.2 Hz, 2.5H, HyPro-dHa = , HyPro-dHb, Gal H-2), 3.83 (d, J=
2
5.9 Hz, 1H, Gal H-5), 3.97 (dd, J=19.6, 8.9 Hz, 3H, GlcNAc H-3,
GlcNAc H-5, Gal H-6a), 4.11–4.61 (m, 11 H, GlcNAc H-6a, GlcNAc H-
6b, Gal H-6b, Fmoc-eH, Fmoc CHH, GlcNAc H-2, HyPro-gH, HyPro-
aH, Fuc H-5, Gal H-1, Fmoc CHH), 4.80 (d, J=3.3 Hz, 1H, GlcNAc H-
1), 4.85–4.89 (m, 1H, GlcNAc H-4), 4.92–4.97 (m, 2H, Gal H-3, Fuc
H-2), 5.25 (t, J=4.0 Hz, 2H, Gal H-4, Fuc H-1), 5.34 (s, 1H, Fuc H-4),
5.45 (dd, J=10.8, 3.0 Hz, 1H, Fuc H-3), 5.67 (d, J=7.2 Hz, 1H, OH),
7.31 (t, J=7.1 Hz, 2H, 2”aromatic CH), 7.37–7.41 (m, 2H, 2”aro-
matic CH), 7.56 (d, J=6.2 Hz, 2H, 2”aromatic CH), 7.77 (d, J=
7.0 Hz, 2H, 2”aromatic CH); 13C NMR assigned from HSQC
(125 MHz, CDCl3): d = 15.40, 20.72 (”8), 22.71, 35.65 (”2), 37.31 (”
2), 47.27 (”2), 51.59, 51.92, 52.25, 57.56, 58.23, 60.22, 60.55 (”2),
62.54 (”2), 65.20, 66.86, 67.19, 67.86, 68.19 (”3), 68.85 (”2), 69.18,
N-a-(9-Fluorenylmethyloxycarbonyl)-l-trans-4-O-[[2,3,4-tri-O-
acetyl-b-l-fucopyranosyl]-(1!2)-[3,4,6-tri-O-acetyl-b-d-galacto-
pyranosyl]-(1!3)-4,6-di-O-acetyl-2-deoxy-2-(N-acetamido)-a-d-
Chem. Eur. J. 2015, 21, 11779 – 11787
www.chemeurj.org
11786
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[2] J. R. Skaar, J. K. Pagan, M. Pagano, Nat. Rev. Mol. Cell Biol. 2013, 14, 369–
381.
[3] B. Gonzalez-Yanes, J. M. Cicero, R. D. Brown, Jr., C. M. West, J. Biol. Chem.
1992, 267, 9595–9605.
69.85, 71.84, 72.84, 73.50, 73.83, 76.16, 76.82, 96.41, 97.73, 100.72,
120.31, 124.95, 127.28, 127.94; MALDI: m/z: calcd for
C56H68N2NaO27: 1223.3907; found: 1223.4296 [M+Na]+.
Glycopeptide 1: The glycopeptide was synthesized by SPPS on
a Rink Amide AM resin (147 mg, 0.05 mmol) as described above.
After automated synthesis of the first 8 amino acids, the glycosylat-
ed amino acid 6 (90 mg, 0.075 mmol, 1.5) coupling was carried out
manually on a CEM Discover SPS Microwave Synthesizer using
HATU (30 mg, 0.075 mmol) and HOAt (10 mg, 0.075 mmol) as cou-
pling reagents in the presence of DIPEA (35 mL, 0.2 mmol) in DMF
(1 mL). The mixture was left at room temperature for 18 h, fol-
lowed by microwave-irradiated coupling reaction, which was moni-
tored by Kaiser test and was complete after 10 min. The resin was
then returned to the synthesizer and peptide elongation was con-
tinued by automated synthesis as described above. The resin was
rinsed with DCM (6”5 mL) and MeOH (6”5 mL) and dried under
vacuum overnight. The resin was placed in a bubbler and treated
with 10 mL of 94% TFA, 2.5% H2O, 2.5% EDT, 1% TIPS) for 2 h with
occasional agitation. The resin was filtered off, and washed with
DCM (2”8 mL). The filtrate was concentrated to 1/5 of its volume
and the glycopeptide was precipitated out using ice-cold diethyl
ether and recovered by centrifugation at 3000 rpm for 15 min. The
glycopeptide was dissolved in 50% aqueous acetonitrile and
lyophilized. The obtained residue was dissolved in 5 mL of the
stock solution (477 mg guanidine·HCl, 23 mg Na, 45 mL MeOH,
5 mL DCM) and left stirring at room temperature for 18 h. The re-
action mixture was neutralized with AcOH and concentrated under
reduced pressure. The glycopeptide 1 was purified by C18 re-
versed-phase HPLC and lyophilized. ESI: m/z: calcd for
[C106H170KN25NaO45S]2+: 2607.1027 [1304.055]2+; found: as doubly
charged ion 1304.0514.
[4] a) E. Kozarov, H. van der Wel, M. Field, M. Gritzali, R. D. Brown, Jr., C. M.
West, J. Biol. Chem. 1995, 270, 3022–3030; b) P. Teng-umnuay, H. R.
Morris, A. Dell, M. Panico, T. Paxton, C. M. West, J. Biol. Chem. 1998, 273,
18242–18249.
[5] A. Ishiwata, S. Kaeothip, Y. Takeda, Y. Ito, Angew. Chem. Int. Ed. 2014, 53,
9812–9816; Angew. Chem. 2014, 126, 9970–9974.
[6] C. M. West, Z. A. Wang, H. van der Wel, Biochim. Biophys. Acta Gen. Subj.
2010, 1800, 160–171.
[7] C. M. West, H. van der Wel, Z. A. Wang, Development 2007, 134, 3349–
3358.
[8] Z. A. Wang, D. Singh, H. van der Wel, C. M. West, Dev. Biol. 2011, 349,
283–295.
[9] a) D. Zhang, H. van der Wel, J. M. Johnson, C. M. West, J. Biol. Chem.
2012, 287, 2006–2016; b) C. M. Schafer, M. O. Sheikh, D. Zhang, C. M.
West, J. Biol. Chem. 2014, 289, 9076–9088.
[10] a) M. O. Sheikh, C. M. Schafer, J. T. Powell, K. K. Rodgers, B. H. Mooers,
C. M. West, Biochemistry 2014, 53, 1657–1669; b) M. O. Sheikh, Y. Xu, H.
van der Wel, P. Walden, S. D. Hartson, C. M. West, Mol. Cell. Proteomics
2015, 14, 66–80.
[11] Y. Xu, Z. A. Wang, R. S. Green, C. M. West, BMC Dev. Biol. 2012, 12, 31.
[12] a) M. J. Gaunt, J. Q. Yu, J. B. Spencer, J. Org. Chem. 1998, 63, 4172–4173;
b) J. Xia, S. A. Abbas, R. D. Locke, C. F. Piskorz, J. L. Alderfer, K. L. Matta,
Tetrahedron Lett. 2000, 41, 169–173.
[13] R. R. Schmidt, W. Kinzy, Adv. Carbohydr. Chem. Biochem. 1994, 50, 21–
123.
[14] a) J. R. Pougny, P. Sinay, Tetrahedron Lett. 1976, 17, 4073–4076; b) A. J.
Ratcliffe, B. Fraser-Reid, J. Chem. Soc. Perkin Trans. 1 1990, 747–750;
c) R. R. Schmidt, M. Behrendt, A. Toepfer, Synlett 1990, 694–696.
[15] a) B. Yu, H. C. Tao, Tetrahedron Lett. 2001, 42, 2405–2407; b) S. Cai, B.
Yu, Org. Lett. 2003, 5, 3827–3830.
[16] G. H. Veeneman, S. H. van Leeuwen, J. H. van Boom, Tetrahedron Lett.
1990, 31, 1331–1334.
[17] P. Sjçlin, J. Kihlberg, J. Org. Chem. 2001, 66, 2957–2965.
[18] J. H. van Boom, P. M. J. Burgers, Tetrahedron Lett. 1976, 17, 4875–4878.
[19] G. H. Veeneman, S. H. van Leeuwen, H. Zuurmond, J. H. van Boom, J.
Carbohydr. Chem. 1990, 9, 783–796.
[20] A. Demchenko, T. Stauch, G. J. Boons, Synlett 1997, 818–820.
[21] R. Rodebaugh, J. S. Debenham, B. Fraser-Reid, Tetrahedron Lett. 1996,
37, 5477–5478.
[22] R. R. Kale, C. M. McGannon, C. Fuller-Schaefer, D. M. Hatch, M. J. Flagler,
S. D. Gamage, A. A. Weiss, S. S. Iyer, Angew. Chem. Int. Ed. 2008, 47,
1265–1268; Angew. Chem. 2008, 120, 1285–1288.
Conjugation of glycopeptide 1 to mcKLH: 0.1m sodium phos-
phate buffer pH 8.0 containing 0.15m NaCl and 5 mm EDTA
(200 mL) was added to a solution of Imject maleimide activated
mcKLH (2.3 mg) in 0.1m sodium phosphate buffer pH 7.2 contain-
ing 0.15m NaCl (200 mL). This mixture (200 mL) was then added to
glycopeptide 1 and the setup was left to incubate at room temper-
ature for 18 h. Purification was performed via spin filtration
through a 3 kDa filter and washed with 0.1m sodium phosphate
buffer pH 7.2 containing 0.15m NaCl (500 mL), six times. The KLH
conjugate was taken up in 0.1m sodium phosphate buffer pH 7.2
containing 0.15m NaCl (1 mL). This gave a conjugate with 641 resi-
dues per KLH molecule as determined by quantitative monosac-
charide analysis by HPAEC/PAD and Bio-Rad protein concentration
test.
[23] T. Matsushita, H. Hinou, M. Kurogochi, H. Shimizu, S. I. Nishimura, Org.
Lett. 2005, 7, 877–880.
[24] U. Ellervik, G. Magnusson, Tetrahedron Lett. 1997, 38, 1627–1628.
[25] Y. Yu, S. Mishra, X. Song, Y. Lasanajak, K. C. Bradley, M. M. Tappert, G. M.
Air, D. A. Steinhauer, S. Halder, S. Cotmore, P. Tattersall, M. Agbandje-
McKenna, R. D. Cummings, D. F. Smith, J. Biol. Chem. 2012, 287, 44784–
44799.
Acknowledgements
[26] a) N. W. Owens, J. Stetefeld, E. Lattova, F. Schweizer, J. Am. Chem. Soc.
2010, 132, 5036–5042; b) L. Corcilius, G. Santhakumar, R. S. Stone, C. J.
Capicciotti, S. Joseph, J. M. Matthews, R. N. Ben, R. J. Payne, Bioorg. Med.
Chem. 2013, 21, 3569–3581; c) S. Kaeothip, A. Ishiwata, Y. Ito, Org.
Biomol. Chem. 2013, 11, 5892–5907; d) H. Shinohara, Y. Matsubayashi,
Plant Cell Physiol. 2013, 54, 369–374; e) N. Xie, C. M. Taylor, J. Org.
Chem. 2014, 79, 7459–7467.
This research was supported by NIH/NIGMS Grant Nos. R01-
GM084383 (C.M.W., G.J.B.) and R01-GM037539 (C.M.W.).
Keywords: antibodies
·
carbohydrate
·
glycopeptide
·
[27] C. V. Karunaratne, T. K. Weldeghiorghis, C. M. West, C. M. Taylor, J. Am.
Chem. Soc. 2014, 136, 15170–15175.
glycosylation · posttranslational modification
[1] A. R. Willems, M. Schwab, M. Tyers, Biochim. Biophys. Acta Mol. Cell Res.
2004, 1695, 133–170.
Received: April 23, 2015
Published online on July 14, 2015
Chem. Eur. J. 2015, 21, 11779 – 11787
www.chemeurj.org
11787
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim