A simplified procedure for gram-scale production of sialylglycopeptide (SGP) from egg yolks and subsequent semi-synthesis of Man3GlcNAc oxazoline

Article abstract of DOI:10.1016/j.carres.2014.07.013

Heterogeneity of glycan structures in native glycoconjugates always hampers precise studies on carbohydrate-involved biological functions. To construct homogeneous glycoconjugates from natural resource of homogeneous glycans is therefore a practical approach to solve this problem. We report here an optimized procedure for gram-scale production of sialylglycopeptide (SGP) containing a disialyl biantennary complex-type N-glycan from egg yolks. Our new procedure simplified the extraction process by treating the egg yolk powder with 40% acetone, avoiding massive emulsification, high-speed centrifugation, and sophisticated chromatography in reported methods. Subsequent semi-synthesis of the N-glycan core Man₃GlcNAc oxazoline from SGP was accomplished for the first-time via glyco-trimming and successive oxazoline formation. This efficient semi-synthesis provides an alternative to the pure chemical approach that involves multi-step total synthesis and facilitates the application of endo-glycosidase-enabled chemoenzymatic synthesis of various homogeneous glycoconjugates.

Full text of DOI:10.1016/j.carres.2014.07.013

Carbohydrate Research 396 (2014) 62–69
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
A simplified procedure for gram-scale production
of sialylglycopeptide (SGP) from egg yolks and subsequent
semi-synthesis of Man3GlcNAc oxazoline
b,c,
Bingyang Sun a,b, Wenzheng Bao b,c, Xiaobo Tian b, Mingjing Li b,c, Hong Liu b, Jinhua Dong a, , Wei Huang
⇑
⇑
a Key Laboratory of Structure-Based Drug Design and Discovery, Ministry of Education, Shenyang Pharmaceutical University, Shenyang 110016, China
b CAS Key Laboratory of Receptor Research, Shanghai Institute of Materia Medica, Chinese Academy of Sciences, 555 Zuchongzhi Road, Pudong, Shanghai 201203, China
c iHuman Institute, ShanghaiTech University, Shanghai 201210, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 16 June 2014
Received in revised form 16 July 2014
Accepted 18 July 2014
Available online 27 July 2014
Heterogeneity of glycan structures in native glycoconjugates always hampers precise studies on carbohy-
drate-involved biological functions. To construct homogeneous glycoconjugates from natural resource of
homogeneous glycans is therefore a practical approach to solve this problem. We report here an opti-
mized procedure for gram-scale production of sialylglycopeptide (SGP) containing a disialyl biantennary
complex-type N-glycan from egg yolks. Our new procedure simplified the extraction process by treating
the egg yolk powder with 40% acetone, avoiding massive emulsification, high-speed centrifugation, and
sophisticated chromatography in reported methods. Subsequent semi-synthesis of the N-glycan core
Man3GlcNAc oxazoline from SGP was accomplished for the first-time via glyco-trimming and successive
oxazoline formation. This efficient semi-synthesis provides an alternative to the pure chemical approach
that involves multi-step total synthesis and facilitates the application of endo-glycosidase-enabled che-
moenzymatic synthesis of various homogeneous glycoconjugates.
Keywords:
Egg yolk
Sialylglycopeptide
Man3GlcNAc oxazoline
Chemoenzymatic transglycosylation
Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction
Yamamoto et al.17–19 reported the transglycosylation activity of
endo-b-N-acetylglucosaminidase from Mucor hiemalis (Endo-M)
Glycosylation as a major post-translational modification of
protein is involved in many important biological processes such
as protein folding and quality control,1–4 cell adhesion and migra-
tion,5,6 viral entry and pathogenesis,7,8 immune response and reg-
ulation,9,10 etc. More than 50% of mammalian proteins and about
70% of clinical protein drugs are glycoproteins.11 It has been tre-
mendously demonstrated that glycan substructures may signifi-
cantly affect functions of glycoproteins. However, heterogeneity
of glycan structures in natural glycoproteins hampers the
functional studies on precise structure–function relationship. As
a solution to this problem, an efficient chemoenzymatic method
for homogeneous glycoprotein synthesis has emerged, which takes
advantage of the transglycosylation activity of endo-b-N-acetylglu-
cosaminidases,11–16 a class of endoglycosidases that hydrolyze the
glycosidic bond in the N,N0-diacetylchitobiose motif of N-glycans,
in the presence of glycan donor substrates.
using the egg-yolk sialylglycopeptide (SGP, Fig. 1)20,21 as the donor
substrate. The Endo-M enzyme cleaves the biantennary disialyl
complex-type N-glycan from SGP and transfers it onto another
GlcNAc-containing peptide/protein forming a new homogeneous
glycopeptide/glycoprotein though in
a relatively low yield.
Thereafter, Shoda et al.22 firstly discovered the synthetic glycan
oxazoline, which is the transit-state form of endo-glycosidase-
catalyzed N-glycan hydrolysis, could also serve as the transglyco-
sylation donor substrate and dramatically improve the efficiency.
Wang’s and Fairbanks’ groups synthesized a series of N-glycan
oxazolines23–32 including the core tetrasaccharide Man3GlcNAc
oxazoline (Fig. 1) as the substrates of Endo-A from Arthrobacter
protophormiae33 that promoted excellent transglycosylation yields
of resulting homogeneous glycopeptides and glycoproteins.
This chemoenzymatic transglycosylation approach presents a
powerful tool for efficient synthesis of diverse glycoconjugates
as summarized in Figure 1. Wang and his co-workers have made
significant contributions to this method and successfully
expanded its application in synthesis of important therapeutic
⇑
Corresponding authors. Tel.: +86 24 23986402; fax: +86 24 23904249 (J.D.);
tel.: +86 21 20231000x2517; fax: +86 21 50807088 (W.H.).
E-mail addresses: dongjh66@hotmail.com (J. Dong), huangwei@simm.ac.cn
(W. Huang).
glycoproteins,34,35 glycopeptides36–38
,
novel glyco-clusters,39
glyco-natural products,40 etc. (Fig. 1.) Recently, two papers have
http://dx.doi.org/10.1016/j.carres.2014.07.013
0008-6215/Ó 2014 Elsevier Ltd. All rights reserved.

B. Sun et al. / Carbohydrate Research 396 (2014) 62–69
63
OH
HO
HO
HOOC
O
HO
HO
O
O
O
O
O
HO
HO
AcHN
AcHN
OH
HO
HO
HO
HO
O
HO
O
H
N
OH
O
HO
O
ref. 20, 21
O
O
O
HO
O
HO
O
HO
NHAc
NHAc
HO
HO
optimized procedure
in current article
O
HO
O
O
O
HO
H
H
N
OH
HOOC
HO
HO
HO
O
Chemical
O
H2
N
N
O
O
O
N
H
N
H
N
H
CO2H
O
HO
HO
HO
AcHN
OH
AcHN
O
O
HO
total synthesis
egg yolk
NH2
H2
N
Sialylglycopeptide, SGP
ref. 23, 28
first semi-synthesis
in current article
OH
HO
O
HO
HO
HO
O
OH
O
ref. 37
O
HO
O
HO
O
O
HO
HO
HO
N
O
OH
Glyco-clusters
Glyco-dendrimers
Man3GlcNAc oxazoline (1)
Glyco-liposome
ref. 38
ref. 11, 32
ref. 34, 35, 36
OH
O
HO
HO
HO
PhCOO
O
OH
O
OH
O
HO
O
O
HO
O
O
O
O
O
HO
HO
HO
HO
HO
HO
NHAc
OH
O
OH
Glyco-natural products
Glycoproteins
Glycopeptides
Figure 1. Preparation of N-glycan core Man3GlcNAc oxazoline and its application in chemoenzymatic synthesis of various glycoconjugates.
respectively reported the Fc-specific glycan remodeling of the
monoclonal antibody drugs by Endo-S from Streptococcus
pyogenes using Man3GlcNAc oxazoline as the homogeneous
glycan donor.11,34
for separation of phenol-extracting solution from precipitates
limits the large-scale production. Sugawara et al. reported in a
patent43 using water to extract defatted egg yolks and treating
the supernatant with alcohols or other organic solvents to precip-
itate SGP. But this method encounters the problems of emulsifica-
tion, high-speed centrifugation, and complicated purification
procedures as well. We developed a newly optimized procedure
for gram-scale SGP preparation by reducing emulsification and
averting centrifugation. This simplified method provides efficient
scale-up of SGP extraction for potential industrial interest
and facilitates expeditious semi-synthesis of N-glycan core
Man3GlcNAc oxazoline for chemoenzymatic preparation of various
glycoconjugates.
Man3GlcNAc oxazoline as the core N-glycan oxazoline has been
employed in chemoenzymatic synthesis of various glycoconjugates
(Fig. 1) and its chemical total synthesis has been reported by
Wang’s and Fairbanks’ groups, respectively.23,28 The total synthesis
consists of over 30 synthetic steps with sophisticated process of
regio-selective protection and deprotection, as well as stereo-
selective glycosylation. The difficulty of long-step total synthesis
in poor economic efficiency hinders the application of chemoenzy-
matic synthesis of glycoconjugates using the N-glycan oxazoline
substrates. As an alternative option, semi-synthesis from natural
homogeneous N-glycan resources was committed to prepare
high-mannose type and biantennary complex-type N-glycan oxaz-
olines in good yields.41,42 Here, we report for the first time the
expeditious semi-synthesis of core Man3GlcNAc oxazoline from
the egg-yolk sialylglycopeptide (SGP) via enzymatic glyco-
trimming and successive oxazoline formation reaction in a water
solution.
SGP was firstly isolated by Seko et al. from egg yolks via phenol-
extraction, centrifugal separation, ethylacetate wash, and compre-
hensive size-exclusion chromatography.20 Zou et al. simplified the
later-stage purification procedure using porous graphite carbon
(PGC) extraction instead of reiterant gel-filtration operation.21
However, the problem of emulsification in these methods caused
a lot of trouble during all the processes before gel-filtration. Addi-
tionally, the requirement of long-time high-speed centrifugation
2. Results and discussion
2.1. Optimized procedure for gram-scale production of SGP
In Seko’s approach, 9% phenol aqueous solution was employed
to precipitate the unrelated proteins and extract the SGP from
egg yolks. While, the tremendous emulsification occurred in the
mixture probably is due to the blending of lipids, soluble proteins,
organic solvent, and water. The extracts were highly emulsified so
that separation of precipitates and supernatant was very difficult
and incomplete even under high-speed (10,000g) centrifugation.
A similar trouble also occurred when ethyl acetate was added to
wash the supernatant after centrifugation. Thus, we sought to opti-
mize the procedure to reduce or even avoid the emulsification for
large-scale process.

64
B. Sun et al. / Carbohydrate Research 396 (2014) 62–69
Table 1
SGP content determination
Firstly, we lyophilized the egg yolks and washed the resulted
yolk powder with diethyl ether to remove lipids that may lead to
emulsification dramatically. Next, for SGP extraction, we chose
acetone–water solution which generates much less emulsion than
phenol–water or acetonitrile–water solutions. To search the right
ratio of acetone–water solution for extraction, we used HPLC to
scout the SGP content in yolk extracts from 80%, 70%, 60%, 40% ace-
tone, and 9% phenol (Fig. 2).
Extraction condition
SGP content in extracts
(mg/1 g yolk powder)
Total weight of extracts
(mg/1 g yolk powder)
70% Acetone
60% Acetone
50% Acetone
40% Acetone
30% Acetone
20% Acetone
9% Phenol
0
10.7
13.2
15.5
17.6
28.6
37.9
55.4
0.38
0.62
1.02
1.13
1.08
0.71
There were no detectable SGP but other unidentified substances
in yolk extracts by 80% or 70% acetone (Fig. 2a and b). When
increasing the water portion in extracting solvent, SGP was pulled
out from the yolk cake by 60% and 40% acetone as shown in
Figure 2c and d (the SGP peak was marked with a star symbol).
Compared with extract by 9% phenol (Fig. 2f), there are much less
unrelated peaks in the HPLC profiles of 40% acetone and much less
emulsification during the handling as well. We further investigated
extracts by 50%, 30%, and 20% acetone and quantified the SGP con-
tents in all extracts by HPLC (Table 1). The extraction rates of SGP in
40%, 30%, and 20% acetone indicated the best results attaining 1.02,
1.13, and 1.08 mg per gram egg yolk lyophilized powder, against
0.71 mg SGP per gram yolk powder by 9% phenol. In the end, we
chose 40% acetone as the optimal extraction condition because of
these reasons: (1) the SGP content in total extract weight by 40%
acetone was about 5.8% which was the most efficient in all extrac-
tion conditions (Table 1); (2) SGP in extracts by 60% and 50% ace-
tone were much less (Fig. 2 and Table 1); (3) emulsification
gradually occurred in extraction procedure by 30% and 20% acetone
but not by 40% acetone. To test the completion of SGP extraction by
40% acetone, we performed second-round extraction and deter-
mined the HPLC profile (Fig. 2e) indicating no more detectable
SGP left in the precipitates after first extraction.
After extracted by 40% acetone, the residue yolk cake could be
simply filtered through a layer of celite, averting emulsification
trouble and high-speed centrifugal separation. Thereafter, we tried
the purification of SGP via three different approaches, respectively:
gel filtration with a Sephadex G-25 column, preparative HPLC iso-
lation with a C18 column, and solid-phase extraction with an
active carbon column. All these approaches could successfully iso-
late the pure SGP in similar yields. The procedure of active carbon
extraction was most convenient with highest separation capability
and less costly, while HPLC and gel filtration have limited sample
loading and often required second-round purification especially
2000000
450000
(a)
(e)
1800000
400000
350000
300000
250000
200000
150000
100000
50000
1600000
1400000
1200000
1000000
800000
600000
400000
200000
0
0
2500000
2400000
(b)
(f)
2000000
1900000
1400000
900000
400000
1500000
1000000
500000
*
0
-100000
600000
500000
400000
300000
200000
100000
0
2250000
(c)
(g)
1750000
1250000
750000
250000
*
-250000
900000
-100000
0
5
10
15
20
25
30
35
40
Retention Time (min)
120
100
80
60
40
20
0
(d)
800000
955.7290
#
(h)
700000
600000
500000
400000
300000
200000
1433.0929
*
100000
0
-100000
0
5
10
15
20
25
30
35
40
500
1000
1500
2000
2500
3000
Retention Time (min)
m/z
Figure 2. HPLC profiles of egg yolk extracts by different extracting solvents and ESI-HRMS of purified SGP. (a) extracts by 80% acetone; (b) 70% acetone; (c) 60% acetone; (d)
40% acetone; (e) second extracts by 40% acetone; (f) 9% phenol; (g) HPLC profile of purified SGP; (h) ESI-HRMS data of purified SGP. The SGP peak in HPLC profiles is marked
#
with ⁄; the HRMS reference is marked with in the MS data.

B. Sun et al. / Carbohydrate Research 396 (2014) 62–69
65
Figure 3. 1H NMR spectrum of egg-yolk SGP.
2.2. Efficient semi-synthesis of N-glycan core Man3GlcNAc
oxazoline
300 egg yolks (3.3 L)
With plenty amount of SGP in hand, we sought to develop a
new semi-synthetic approach to prepare the N-glycan core
Man3GlcNAc oxazoline which is widely applied in chemoenzy-
matic synthesis of various homogeneous glycoconjugates. Our
strategy was to truncate the non-reducing end saccharide moieties
by specific glycosidase enzymes to give the Man3GlcNAc2-peptide
(2). Then, the core Man3GlcNAc glycan was cleaved from 2 by
Endo-A and transformed to the oxazoline following Shoda’s
approach44 using DMC (2-chloro-1,3-dimethyl-imidazolinium
chloride) in presence of Et3N (Scheme 2).
(i) added 1.5 L water, stirred at rt for 1h
(ii)lyophilization
Yolk powder (2.1 Kg)
(i) Et2O wash (6 L x 2), 70% acetone wash (6 L)
(ii) 40% acetone extraction (3 L x 2)
(iii) drying of extracted solution
exo-Glycosidases were widely used to trim non-reducing end
carbohydrate moieties of N-glycans for diverse glycan struc-
tures.45–47 Firstly, we treated the SGP with neuraminidase to cleave
the two sialic acids and monitored by HPLC and ESI-HRMS which
clearly showed the complete removal of sialic acid (Fig. S1). Then
b1,4galactosidase was added to trim exposed galactose moieties
on the non-reducing end. After confirmation on the galactose
digestion, N-acetyl-glucosaminidase was introduced for truncation
on non-reducing end GlcNAc residues with b1,2-linkage to the
mannoses, giving Man3GlcNAc2-peptide (2) after being isolated
by preparative HPLC. All the processes of enzymatic glyco-hydroly-
sis were controlled step-by-step under HPLC and MS determina-
tion (Fig. S1). Since these three enzymes maintain good activity
in the same phosphate buffer (pH 6.0), we performed the one-
pot enzymatic hydrolysis by adding all three enzymes in the solu-
tion and successfully attained the desired product 2 in a good yield.
The N-glycan core Man3GlcNAc was then released from glyco-
peptide 2 catalyzed by glycosidase Endo-A in a pH 6.0 phosphate
buffer. The high-resolution mass spectrum indicated m/z value of
730.2378 and 863.4834 in excellent agreement with theoretical
mass of the release glycan [M+Na] 730.2382 and GlcNAc-peptide
[M+H] 863.4838. After lyophilization, the obtained Man3GlcNAc
in situ was directly transformed to its oxazoline form in a water
solution following the reference. The reaction was performed in
D2O in the presence of DMC and Et3N. 1H NMR monitoring of the
reaction showed H1 proton of GlcNAc as a doublet peak at
5.95 ppm with a coupling constant of 7.2 Hz that is the character-
istic signal of GlcNAc anomeric proton in the oxazoline form. The
integration of a full proton for this peak suggested the completion
of oxazoline formation. These data implicated the efficient one-pot
oxazoline formation in spite of the presence of buffer, peptide, and
enzyme remained in the reaction. The oxazoline (1) was purified
by solid-phase extraction via a PGC cartridge and the proton
NMR spectrum as shown in Figure 4 is identical to the reference.
This semi-synthesis provided an efficient approach for rapid
Yolk extract powder (36.7 g)
active carbon/celite (2:1) column,
eluted by 25% MeCN
SGP (1.9 g)
Scheme 1. Optimal procedure for gram-scale SGP production.
for overloaded cases. The purified SGP was characterized by HPLC,
HRMS, and NMR. The HPLC profile showed a single symmetry peak
(Fig. 2g) suggesting a good purity. The ESI-HRMS spectrum (Fig. 2h)
indicated m/z values of 1433.0929 and 955.7290 which are highly
accordant with calculated SGP 2-charge mass 1433.0933 and
3-charge mass 955.7308. The NMR spectrum (Fig. 3) is also
identical to the reference data.
The optimal procedure for gram-scale SGP production is sum-
marized in Scheme 1. Three hundred egg yolks were stirred with
water and freeze dried to yolk powder. The powder was washed
successively with ethyl ether and 70% acetone after thorough
mixing. Thereafter, 40% acetone was added to the residue and
mixed rigorously, then the solution was filtered through celite.
The filtrate was concentrated and lyophilized. The resulted extract
powder was dissolved in water and subject to solid-phase extrac-
tion with an active carbon/celite (2:1) column. The eluting solution
by 25% MeCN gives the pure SGP in a yield of 0.9 g/kg yolk powder.
In the references, the SGP yields were reported as 8 mg/yolk20 or
680 mg/100 yolks,21 however, it is too rough to compare the SGP
content in each egg yolk because of the individual differences in
egg batches. For example, the volume of 115 egg yolk was 1.9 L
in the reference,20 but only about 1.3 L in our case suggesting dif-
ferences in size of eggs used. Thus we normalized the yield of SGP
in per gram or kilogram egg yolk powder.

66
B. Sun et al. / Carbohydrate Research 396 (2014) 62–69
OH
HO
O
HO
HO
HO
O
H
N
OH
O
HO
O
O
O
Neuraminidase
O
HO
O
HO
HO
O
b1,4galactosidase
NHAc
NHAc
HO
HO
HO
N-acetyl-glucosaminidase
SGP
O
OH
HO
O
O
O
H
H
N
H2N
N
N
H
N
H
N
H
CO2H
O
O
NH2
H2N
Man3GlcNAc2-peptide (2)
Endo-A
OH
OH
HO
HO
HO
HO
O
O
HO
HO
DMC/Et3N
HO
O
OH
O
HO
O
OH
O
O
O
HO
O
HO
O
OH
NHAc
HO
O
HO
O
O
HO
HO
HO
HO
HO
HO
N
O
OH
O
OH
Man3GlcNAc in situ
Man3GlcNAc oxazoline (1)
Scheme 2. Semi-synthesis of Man3GlcNAc oxazoline from SGP.
preparation of the core N-glycan oxazoline compared with long-
step total synthesis.
expeditiously catalyze the transglycosylation of Man3GlcNAc
oxazoline to GlcNAc-RNase forming a new homogeneous glyco-
form of RNase. We performed the one-pot glycan remodeling by
mixing native Man5–9GlcNAc2-RNase, Endo-A, and Man3GlcNAc
oxazoline in a pH7.0 phosphate buffer. After 1 h incubation at
37 °C, Man3GlcNAc2-RNase was given as expected (see Supporting
information).
Endo-A has diverse substrate specificity in acceptor structures
including GlcNAc-containing proteins, peptides, natural products,
glycoclusters, etc. We synthesized an O-GlcNAc-amino acid to test
the chemoenzymatic transglycosylation of Man3GlcNAc oxazoline
to an O-GlcNAc moiety. The Fmoc-Ser(O-GlcNAc)-OH (4), prepared
2.3. Chemoenzymatic transglycosylation with Man3GlcNAc
oxazoline
Using the Man3GlcNAc oxazoline prepared by our new method
as the donor substrate, we repeated the reported experiment25 on
chemoenzymatic glycosylation of ribonuclease (RNase) B, a glyco-
protein with catalytic activity for RNA degradation. N-Glycans of
native RNase B is a mixture of Man5–9GlcNAc2 which could be
rapidly cleaved by Endo-A giving GlcNAc-RNase. Endo-A could also
Figure 4. 1H NMR spectrum of Man3GlcNAc oxazoline.

B. Sun et al. / Carbohydrate Research 396 (2014) 62–69
67
following the reported method (see Supporting information), was
incubated with oxazoline (1) and Endo-A. The LC–MS data showed
clearly the efficient glycosylation giving new peak of
Fmoc-Ser(O-Man3GlcNAc2)-OH confirmed by the HRMS data (see
Supporting information). In biomedical proteomic studies, the
attachment of Man3GlcNAc moiety could be useful for enrichment
of O-GlcNAc glycopeptides by mannose-binding lectins. A related
research is undergoing.
4.3. Preparation of egg yolk powder
a
Fresh 300 egg yolks were separated from egg white carefully.
Water (1.5 L) was added and the mixture was stirred at rt for 1 h.
The residue was subject to a freeze drier for lyophilization until
complete dryness, giving the yellow egg yolk powder 2.1 kg.
4.4. Extraction condition optimization on SGP isolation from
egg yolk powder
3. Conclusion
To develop an optimal isolation method for SGP production, we
tried various extraction conditions and quantified the SGP contents
in all extracts by HPLC.
An optimized procedure for the gram-scale production of
SGP from egg yolks was successfully achieved that expedites
the subsequent semi-synthesis of N-glycan core Man3GlcNAc
oxazoline compared with long-step total synthesis. This approach
provides the rapid preparation of homogeneous N-glycan
substrates from natural resource and facilitates the following
chemoenzymatic synthesis of homogeneous glycoprotein and
other glycoconjugates.
4.4.1. Extraction by 9% phenol
Yolk powder (100 g) was mixed with 9% aqueous phenol solu-
tion (300 mL) and the mixture was vigorously stirred for 2 h with
an ice bath. The residue in massive emulsion was centrifuged at
10,000g for 30 min. The less emulsive supernatant was separated
from the precipitates and subject to second-round centrifugation.
The final supernatant was washed with ethyl acetate
(300 mL ꢀ 3). The aqueous layer was lyophilized to give a pale yel-
low powder (5.54 g). The resulted yolk extract powder was subject
to analytic HPLC for SGP content determination.
4. Experimental section
4.1. General
Chemical reagents and solvents were purchased from Sinop-
harm Chemical Reagent Co. (Shanghai, China) and used without
further purification. HPLC grade acetonitrile was purchased from
Meryer (Shanghai, China). Neuraminidase and b-1,4-galactosidase
were purchased from New England Biolabs (Ipswich, MA).
b-N-Acetylglucosaminidase and ribonuclease B was purchased
from Sigma–aldrich (Shanghai, China). endo-b-N-Acetylglucosa-
minidase from Arthrobacter protophormiae (Endo-A) was purchased
from Accendatech Co. (Tianjin, China). The unit definitions of
enzyme activities were following the description from manufac-
turers. Porous graphite carbon (PGC) cartridges (Hypercarb) were
purchased from Thermo Fisher Scientific (Beijing, China). Fresh
egg yolks were obtained from hen’s eggs purchased from local
market. TLC staining for carbohydrate samples was performed by
dipping the plates into 10% H2SO4 in methanol and drying with a
heat gun. Nuclear magnetic resonance (NMR) spectra were mea-
sured on a Varian-MERCURY Plus-400 instrument. The chemical
shifts were assigned in ppm and the coupling constants in Hz.
ESI-MS spectra were measured on an Agilent 6110 single quadru-
pole spectrometer. ESI-HRMS spectra were measured on an Agilent
6230 LC–TOF MS spectrometer. The deconvolution MS data were
generated by MagTran 1.02.48
4.4.2. Extraction by various acetone–water solutions
Yolk powder (100 g) was mixed with ethyl ether (300 mL) and
stirred vigorously at rt for 30 min. The ether layer was removed
by filtration and the solid was mixed with additional ethyl ether
(300 mL) and the processes of wash and filtration were repeated.
The pellet residue was respectively subject to 80%, 70%, 60%, 50%,
40%, 30%, and 20% acetone (150 mL twice) extraction. The extract
solution was filtered and lyophilized to give the yolk extract pow-
ders under different conditions. The SGP content of each extract
powder was measured by HPLC and the results are summarized
in Figure 2 and Table 1.
4.5. Purification of SGP from yolk extract powder from 40%
acetone
4.5.1. Gel-filtration with a Sephadex G-25 column
Yolk extract powder from 40% acetone (1 g) was dissolved in
5 mL water and subject to a Sephadex G-25 column, eluted by
0.1 M AcOH. The fractions containing pure SGP were combined
and lyophilized. The fractions containing SGP and other impurities
were subject to second-round gel-filtration purification. All pure
SGP was combined and lyophilized to give a white powder
(48 mg). The HPLC, NMR, and MS data indicated the purity of
SGP was more than 95%.
4.2. High-performance liquid chromatography (HPLC)
Analytical RP-HPLC was performed on two instruments with
different columns. Method A was performed on an Angilent 1260
4.5.2. Preparative HPLC with a C18 column
Yolk extract powder from 40% acetone (1 g) was dissolved in
5 mL water and subject to the preparative HPLC system, eluted
by an isocratic water solution containing 0.1% TFA for 10 min then
a linear gradient of 0–10% acetonitrile containing 0.1% TFA for
additional 20 min. The fractions containing pure SGP were com-
bined and lyophilized. The fractions containing SGP and other
impurities were subject to second-round HPLC purification. All
pure SGP was combined and lyophilized to give a white powder
(51 mg). The HPLC, NMR, and MS data indicated the purity of
SGP was more than 95%.
UHPLC system with a C18 column (1.8
l
m, 3.0 ꢀ 100 mm) at
32 °C. The column was eluted with an isocratic 2% acetonitrile con-
taining 0.1% TFA in first 5 min then a linear gradient of 2–5%
acetonitrile containing 0.1% TFA in additional 10 min at a flow
rate of 1 mL/min. Method
ChuangXinTongHeng LC3000 (analytic) instrument with a C18 col-
umn (5
B was performed on a Beijing
l
m, 4.6 ꢀ 150 mm) at 40 °C. The column was eluted with a
linear gradient of 2–5% acetonitrile containing 0.1% TFA in first
15 min then 5–90% acetonitrile in additional 15 min at a flow rate
of 1 mL/min. Preparative HPLC was performed on
ChuangXinTongHeng LC3000 (preparative) instrument with a pre-
parative C18 column (10
a Beijing
4.5.3. Solid-phase extraction with an active carbon/celite
column
Yolk extract powder from 40% acetone (1 g) was dissolved in
5 mL water and subject to an active carbon/celite (2:1) column
l
m, 20 ꢀ 300 mm). The column was
eluted with a suitable gradient of aqueous acetonitrile containing
0.1% TFA at a flow rate of 10 mL/min.

68
B. Sun et al. / Carbohydrate Research 396 (2014) 62–69
(2 ꢀ 10 cm). The column was washed by water containing 0.1% TFA
(50 mL), 5% acetonitrile containing 0.1% TFA (50 mL), and 10%
MeCN containing 0.1% TFA (50 mL). Then, it was eluted by 25% ace-
tonitrile containing 0.1% TFA (50 mL). The elution fractions were
J = 7.2 Hz, 1H, H1 of GlcNAc), 4.94 (s, 1H, H1 of a-Man), 4.80 (s,
1H, H1 of a-Man0), 4.59 (s, 1H, H1 of b-Man), 4.23 (m, 1H, H3 of
GlcNAc), 4.05 (m, 1H, H2 of GlcNAc), 4.01 (d, J = 2.8 Hz, 1H, H2 of
b-Man), 3.91 (m, 1H, H2 of a-Man), 3.86 (m, 1H, H2 of a-Man0),
3.25 (m, 1H, H5 of GlcNAc), 1.92 (d, J = 1.6 Hz, 3H, Ac).
combined and lyophilized to give SGP as
a white powder
(52 mg). The HPLC, NMR, and MS data indicated the purity of
SGP was more than 95%.
4.9. Chemoenzymatic synthesis of Man3GlcNAc2-RNase by one-
pot glycan remodeling
4.6. Optimized procedure for gram-scale SGP production
Chemoenzymatic synthesis of Man3GlcNAc2-RNase was per-
formed following the reported approach summarized below.
Yolk powder (2.1 kg) was washed thoroughly with ethyl ether
(6 L ꢀ 2) and 70% acetone (6 L). The residue was extracted by 40%
acetone (3 L ꢀ 2) and filtered. The filtrate was concentrated and
subject to an active carbon/celite (2:1) column purification, giving
SGP as a white powder (1.9 g, yield 0.9 g/kg yolk powder). Analytic
HPLC: tR = 13.7 min (Method A); tR = 11.3 min (Method B); ESI-
HRMS: calcd for C112H189N15O70 [M+2H]2+ 1433.0923, found
1433.0929, [M+3H]3+ 955.7308, found 955.7290; 1H NMR
(D2O,400 MHz): d 4.96 (s, 1H, H1 of Man-4), 4.86 (d, J = 10.0 Hz,
1H, H1 of GlcNAc-1), 4.77 (s, 1H, H1 of Man-40), 4.60 (s, 1H, H1
of Man-3), 4.50 (m, 1H), 4.44 (m, 3H, H1 of GlcNAc-2, 5, and 50),
4.27 (m, 2H, H1 of Gal-6 and 60), 2.83 (m, 4H, Lys-CH2), 2.68 (dd,
J = 5.2, 16.0 Hz, 1H, Asn-betaH), 2.54 (dd, J = 7.2, 16.0 Hz, 1H,
Asn-betaH), 2.49 (m, 2H, H3eq of NeuAc), 1.90–1.83 (m, 18H, Ac
ꢀ6), 1.73 (m, 4H, Lys-CH2), 1.62–1.51 (m, 6H, Lys-CH2, H3ex of
NeuAc), 1.27 (m, 4H, Lys-CH2), 1.20 (d, J = 7.6 Hz, 3H, Ala-CH3),
1.01 (d, J = 6.4 Hz, 3H, Thr-CH3), 0.78 (d, J = 6.0 Hz, 6H, Val-CH3).
Native RNase
(400 g, 580 nmol) in a phosphate buffer (100 mM, pH 7.0,
15 L) were incubated at 37 °C with Endo-A (0.1 U) for 1 h.
LC–MS data indicated the transglycosylation product
B (1 mg, 67 nmol) and Man3GlcNAc oxazoline
l
l
Man3GlcNAc2-RNase was obtained in about 80% yield. ESI-MS:
calcd M = 14576.25, found [M+7H]7+ 2083.0709, [M+8H]8+
1822.6968, [M+9H]9+ 1620.7407, [M+10H]10+ 1458.3588,
[M+11H]11+ 1325.9682, [M+12H]12+ 1215.3837, [M+13H]13+
1122.2093; deconvolution MS 14575.73.
4.10. Chemoenzymatic synthesis of Fmoc-Ser(O-Man3GlcNAc2)-
OH
Fmoc-Ser(O-GlcNAc)-OH (prepared following reported method
as described in Supporting information, 26.5
Man3GlcNAc oxazoline (207 g, 300 nmol) in a phosphate buffer
(100 mM, pH 7.0, 28 L) were incubated at 37 °C with Endo-A
lg, 50 nmol),
l
l
4.7. Synthesis of Man3GlcNAc2-peptide (2) by enzymatic
digestion
(0.2 U) for 1 h. LC–MS data indicated the transglycosylation prod-
uct Fmoc-Ser(O-Man3GlcNAc2)-OH was obtained in about 85%
yield. ESI-MS: calcd
1220.4353.
C52H73N3O30,
[M+H]+ 1220.4357, found
SGP (500 mg, 0.17 mmol) in a phosphate buffer (50 mM, pH 6.0,
3.5 mL) was incubated with neuraminidase (1500 U) at 32 °C until
LC–MS showed the desialylation was complete giving the asialo-
Acknowledgements
glycopeptide (ESI-HMRS: calcd for
C
90H155N13O54 [M+2H]2+
1141.9970, found 1141.9993; [M+3H]3+ 761.6672, found
761.6669). To the reaction solution was added b-1,4-galactosidase
(800 U) and the mixture was incubated at 32 °C until LC–MS
showed the degalactosylation was complete giving the
This work was supported by the National Natural Science
Foundation of China (NNSFC, No. 21372238), National
1000-Young-Talent Program of China, and Shanghai Pujiang Talent
Program (No. 13PJ1410200). W.H. thanks Prof. Lai-Xi Wang in
University of Maryland School of Medicine for his kind mentoring.
We thank the members in Huang’s group for helpful discussion and
technical support.
GlcNAc2Man3GlcNAc2-peptide
(ESI-HMRS:
calcd
for
C78H135N13O44 [M+2H]2+ 979.9441, found 979.9430; [M+3H]3+
653.6320, found 653.6310). The pH of the reaction was adjusted
to 5.0 and b-N-acetylglucosaminidase (5 U) was added. The mix-
ture was incubated at 25 °C until LC–MS showed the GlcNAc
removal was complete. The residue was subject to preparative
HPLC purification to give the Man3GlcNAc2-peptide (2) as a white
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.carres.2014.
07.013.
powder (195 mg, 74%). ESI-HMRS: calcd for
C62H109N11O34
[M+H]+ 1552.7217, found 1552.7210; [M+2H]2+ 776.8648, found
776.8639.
References
4.8. Synthesis of Man3GlcNAc oxazoline (1)
1. Spiro, R. G. J. Biol. Chem. 2000, 275(46), 35657–35660.
2. Helenius, A.; Aebi, M. Science 2001, 291(5512), 2364–2369.
3. Aebi, M.; Bernasconi, R.; Clerc, S.; Molinari, M. Trends Biochem. Sci. 2010, 35(2),
74–82.
4. Amin, M. N.; Huang, W.; Mizanur, R. M.; Wang, L. X. J. Am. Chem. Soc. 2011,
133(36), 14404–14417.
5. Casey, R. C.; Oegema, T. R., Jr.; Skubitz, K. M.; Pambuccian, S. E.; Grindle, S. M.;
Skubitz, A. P. Clin. Exp. Metastasis 2003, 20(2), 143–152.
6. Guo, H. B.; Lee, I.; Kamar, M.; Akiyama, S. K.; Pierce, M. Cancer Res. 2002, 62(23),
6837–6845.
Man3GlcNAc2-peptide (195 mg, 0.125 mmol) in a phosphate
buffer (50 mM, pH 6.0, 2 mL) was incubated with Endo-A (10 U)
at 37 °C until LC–MS showed the complete cleavage, releasing
Man3GlcNAc (ESI-HMRS: calcd for C26H45NO21 [M+H]+ 708.2562,
found 708.2555, [M+Na]+ 730.2382, found 730.2378) and
GlcNAc-peptide (ESI-HMRS: calcd for
C
36H66N10O14 [M+H]+
863.4838, found 863.4834). The residue was treated with DMC
(350 mg, 2.07 mmol) and Et3N (0.85 mL, 6.1 mmol) and incubated
at 0 °C for 1 h. The reaction solution was diluted by 1 mM aqueous
NaOH solution and subject to solid-phase extraction with PGC car-
tridge (Hypercarb). After being washed with 5% and 10% MeCN
containing 1 mM NaOH, the cartridge was eluted with 25% MeCN
containing 1 mM NaOH to give the product Man3GlcNAc oxazoline
as a white solid (82 mg, 95%). 1H NMR (D2O, 400 MHz): d 5.95 (d,
7. Vigerust, D. J.; Shepherd, V. L. Trends Microbiol. 2007, 15(5), 211–218.
8. Skehel, J. J.; Wiley, D. C. Annu. Rev. Biochem. 2000, 69, 531–569.
9. Rudd, P. M.; Elliott, T.; Cresswell, P.; Wilson, I. A.; Dwek, R. A. Science 2001,
291(5512), 2370–2376.
10. Lowe, J. B. Cell 2001, 104(6), 809–812.
11. Goodfellow, J. J.; Baruah, K.; Yamamoto, K.; Bonomelli, C.; Krishna, B.; Harvey,
D. J.; Crispin, M.; Scanlan, C. N.; Davis, B. G. J. Am. Chem. Soc. 2012, 134(19),
8030–8033.
12. Wang, L. X. Carbohydr. Res. 2008, 343(10–11), 1509–1522.
13. Wang, L. X.; Huang, W. Curr. Opin. Chem. Biol. 2009, 13(5–6), 592–600.

B. Sun et al. / Carbohydrate Research 396 (2014) 62–69
69
14. Yamamoto, K.; Fujimori, K.; Haneda, K.; Mizuno, M.; Inazu, T.; Kumagai, H.
Carbohydr. Res. 1997, 305(3–4), 415–422.
15. Gamblin, D. P.; Scanlan, E. M.; Davis, B. G. Chem. Rev. 2009, 109(1), 131–163.
16. Yamamoto, K. Biotechnol. Lett. 2013, 35(11), 1733–1743.
17. Yamamoto, K.; Kadowaki, S.; Watanabe, J.; Kumagai, H. Biochem. Biophys. Res.
Commun. 1994, 203(1), 244–252.
32. Fairbanks, A. J. C. R. Chim. 2011, 14, 44–58.
33. Takegawa, K.; Yamabe, K.; Fujita, K.; Tabuchi, M.; Mita, M.; Izu, H.; Watanabe,
A.; Asada, Y.; Sano, M.; Kondo, A.; Kato, I.; Iwahara, S. Arch. Biochem. Biophys.
1997, 338(1), 22–28.
34. Huang, W.; Giddens, J.; Fan, S. Q.; Toonstra, C.; Wang, L. X. J. Am. Chem. Soc.
2012, 134(29), 12308–12318.
18. Haneda, K.; Inazu, T.; Yamamoto, K.; Kumagai, H.; Nakahara, Y.; Kobata, A.
Carbohydr. Res. 1996, 292, 61–70.
35. Zou, G.; Ochiai, H.; Huang, W.; Yang, Q.; Li, C.; Wang, L. X. J. Am. Chem. Soc.
2011, 133(46), 18975–18991.
19. Akaike, E.; Tsutsumida, M.; Osumi, K.; Fujita, M.; Yamanoi, T.; Yamamoto, K.;
Fujita, K. Carbohydr. Res. 2004, 339(3), 719–722.
36. Li, H.; Li, B.; Song, H.; Breydo, L.; Baskakov, I. V.; Wang, L. X. J. Org. Chem. 2005,
70(24), 9990–9996.
20. Seko, A.; Koketsu, M.; Nishizono, M.; Enoki, Y.; Ibrahim, H. R.; Juneja, L. R.; Kim,
M.; Yamamoto, T. Biochim. Biophys. Acta 1997, 1335(1–2), 23–32.
21. Zou, Y.; Wu, Z.; Chen, L.; Liu, X.; Gu, G.; Xue, M.; Wang, P.; Chen, M. J.
Carbohydr. Chem. 2012, 31, 436–446.
22. Fujita, M.; Shoda, S.; Haneda, K.; Inazu, T.; Takegawa, K.; Yamamoto, K. Biochim.
Biophys. Acta 2001, 1528(1), 9–14.
37. Huang, W.; Zhang, X.; Ju, T.; Cummings, R. D.; Wang, L. X. Org. Biomol. Chem.
2010, 8(22), 5224–5233.
38. Huang, W.; Groothuys, S.; Heredia, A.; Kuijpers, B. H.; Rutjes, F. P.; van Delft, F.
L.; Wang, L. X. ChemBioChem 2009, 10(7), 1234–1242.
39. Huang, W.; Wang, D.; Yamada, M.; Wang, L. X. J. Am. Chem. Soc. 2009, 131(49),
17963–17971.
23. Li, B.; Zeng, Y.; Hauser, S.; Song, H.; Wang, L. X. J. Am. Chem. Soc. 2005, 127(27),
9692–9693.
40. Huang, W.; Ochiai, H.; Zhang, X.; Wang, L. X. Carbohydr. Res. 2008, 343(17),
2903–2913.
24. Zeng, Y.; Wang, J.; Li, B.; Hauser, S.; Li, H.; Wang, L. X. Chemistry 2006, 12(12),
3355–3364.
41. Umekawa, M.; Huang, W.; Li, B.; Fujita, K.; Ashida, H.; Wang, L. X.; Yamamoto,
K. J. Biol. Chem. 2008, 283(8), 4469–4479.
25. Li, B.; Song, H.; Hauser, S.; Wang, L. X. Org. Lett. 2006, 8(14), 3081–3084.
26. Huang, W.; Li, C.; Li, B.; Umekawa, M.; Yamamoto, K.; Zhang, X.; Wang, L. X. J.
Am. Chem. Soc. 2009, 131(6), 2214–2223.
42. Umekawa, M.; Higashiyama, T.; Koga, Y.; Tanaka, T.; Noguchi, M.; Kobayashi,
A.; Shoda, S.; Huang, W.; Wang, L. X.; Ashida, H.; Yamamoto, K. Biochim.
Biophys. Acta 2010, 1800(11), 1203–1209.
27. Ochiai, H.; Huang, W.; Wang, L. X. J. Am. Chem. Soc. 2008, 130(41), 13790–
13803.
28. Rising, T. W.; Claridge, T. D.; Davies, N.; Gamblin, D. P.; Moir, J. W.; Fairbanks, A.
J. Carbohydr. Res. 2006, 341(10), 1574–1596.
29. Rising, T. W.; Heidecke, C. D.; Moir, J. W.; Ling, Z.; Fairbanks, A. J. Chem. Eur. J.
2008, 14(21), 6444–6464.
43. Sugawara, S.; Osumi, K., WO2012027868 2012.
44. Noguchi, M.; Tanaka, T.; Gyakushi, H.; Kobayashi, A.; Shoda, S. J. Org. Chem.
2009, 74(5), 2210–2212.
45. Rice, K. G.; Wu, P.; Brand, L.; Lee, Y. C. Biochemistry 1993, 32(28), 7264–7270.
46. Lin, C. H.; Shimazaki, M.; Wong, C. H.; Koketsu, M.; Juneja, L. R.; Kim, M. Bioorg.
Med. Chem. 1995, 3(12), 1625–1630.
30. Parsons, T. B.; Moir, J. W.; Fairbanks, A. J. Org. Biomol. Chem. 2009, 7, 3128–
3140.
47. Kajihara, Y.; Suzuki, Y.; Yamamoto, N.; Sasaki, K.; Sakakibara, T.; Juneja, L. R.
Chemistry 2004, 10(4), 971–985.
31. Parsons, T. B.; Patel, M. K.; Boraston, A. B.; Vocadlo, D. J.; Fairbanks, A. J. Org.
Biomol. Chem. 2010, 8(8), 1861–1869.
48. Zhang, Z.; Marshall, A. G. J. Am. Soc. Mass Spectrom. 1998, 9(3), 225–233.