A new glycosylation method. Part 2: Study of carbohydrate elongation onto the gold nanoparticles in a colloidal phase

Article abstract of DOI:10.1016/j.tet.2007.01.011

A new reaction for carbohydrate elongation for synthesis of oligosaccharide using gold colloidal nanoparticles (GCNPs) has been developed. The gold core in this colloidal phase synthesis was prepared by a reduction of tetrachloroauric acid with 30,31-dithia-3,6,9,12,15,18,43,46,49,52,55,58-dodecaoxa-1,60-hexacontanediol. The presented alkanethionyl oligomeric ethylene glycol worked as a stabilizer of GCNPs and as a linker in chemical elongations of carbohydrates. This colloidal phase synthesis has several advantages such as (1) remnants of reagents and glycosyl donors in each reaction could be easily removed by ultrafiltration or gel filtration column chromatography, (2) further purifications are not required, and (3) the reactions can be monitored by MALDI-TOF MS directly without any pretreatment. In fact, we have successfully synthesized lactose derivative on GCNPs and will report these results in this paper.

Full text of DOI:10.1016/j.tet.2007.01.011

Tetrahedron 63 (2007) 2418–2425
A new glycosylation method. Part 2: Study of carbohydrate
elongation onto the gold nanoparticles in a colloidal phase*
b
b
a,b,
Hiroki Shimizu,a, Masahiro Sakamoto, Noriko Nagahori and Shin-Ichiro Nishimura
*
*
aDrug-Seeds Discovery Research Laboratory, Hokkaido Center, National Institute of Advanced Industrial Science and Technology
(AIST), Sapporo 062-8517, Hokkaido, Japan
bDivision of Advanced Chemical Biology, Graduate School of Advanced Life Science, Frontier Research Center for Post-Genomic
Science and Technology, Hokkaido University, Sapporo 011-0021, Hokkaido, Japan
Received 14 November 2006; revised 25 December 2006; accepted 9 January 2007
Available online 12 January 2007
Abstract—A new reaction for carbohydrate elongation for synthesis of oligosaccharide using gold colloidal nanoparticles (GCNPs) has been
developed. The gold core in this colloidal phase synthesis was prepared by a reduction of tetrachloroauric acid with 30,31-dithia-
3,6,9,12,15,18,43,46,49,52,55,58-dodecaoxa-1,60-hexacontanediol. The presented alkanethionyl oligomeric ethylene glycol worked as a sta-
bilizer of GCNPs and as a linker in chemical elongations of carbohydrates. This colloidal phase synthesis has several advantages such as (1)
remnants of reagents and glycosyl donors in each reaction could be easily removed by ultrafiltration or gel filtration column chromatography,
(2) further purifications are not required, and (3) the reactions can be monitored by MALDI-TOF MS directly without any pretreatment. In
fact, we have successfully synthesized lactose derivative on GCNPs and will report these results in this paper.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
syntheses to carbohydrate chemistry is that creating carbo-
hydrate diversity, namely glycosylation, requires regioselec-
tivity among hydroxyl groups and stereoselectivity on an
anomeric position. On the other hand, polymer or solid sup-
ported syntheses have several disadvantages, for example,
monitoring the reactions is difficult, the reactivity often
decreases compared to that in solution phase, and so on.
Since the great advent of solid-phase peptide synthesis by
Merrifield,1 preparation of diverse peptides in large amounts
became easier, and this established a universal method that
has deeply contributed to the revelation of the biological
roles of peptides.2 Although carbohydrates have been recog-
nized as key compounds in a wide range of biological func-
tions, many of their molecular based ‘cause and effect’ roles
remain in a black box. To open this box, we believe that es-
tablishing methods to prepare a variety of oligosaccharides
in large amounts is important. We have, therefore, already
reported on the development of an automated oligosaccha-
ride synthesizer3 and new synthesis approaches using micro-
wave irradiation for glycopeptide2,4 and oligosaccharides.5
Recently, gold colloidal nanoparticles (GCNPs), the first
related chemical research reported by Faraday,7 became
widely used in chemical and biological research.8 Biologi-
cally, for example, they have been used as biosensing mate-
rials for proteins and biological ligands,9–11 and also applied
in a carbohydrate mediated interaction research field,
namely carbohydrate–proteins (i.e., lectin)12,13 and even in
weak carbohydrate–carbohydrate binding systems.14 Mean-
while in chemical applications, GCNPs have been used in the
synthesis of monolayer-protected cluster molecules,15 NMR
study,16 direct observation of ligand–protein binding by
time-of-flight mass spectrometry (MALDI-TOF MS),17,18
enzymatic oligosaccharide synthesis,18 peptide synthesis,19
and so on.
Polymer or solid supported oligosaccharide syntheses are, of
course, typical methods used to make oligosaccharide syn-
thesis easy.6 Although there are many reports that have
shown many advantages of these methods, it is still hard to
say that universal oligosaccharide synthesis is finally estab-
lished. The difficulty in applying polymer or solid supported
*
Among these reports, we have especially focused on the use
of GCNPs in mass spectrometry. Recent reports have shown
that the self-assembled monolayer (SAM) of a thiol com-
pound chemisorbed onto a gold surface could be ionized di-
rectly with cleavage of the Au–S bond by laser irradiation.17
This means that if chemical reactions are carried out on the
For Part 1: see Ref. 5.
Keywords: Nanoparticle; Carbohydrate; Glycosylation; Colloidal phase;
Oligosaccharide.
* Correspondingauthors. Tel.:+81118578497;fax:+81118578435(H.S.);
tel.: +81 11 857 8472; fax: +81 11 706 9042 (S.-I.N.); e-mail addresses:
hiroki.shimizu@aist.go.jp; shin@glyco.sci.hokudai.ac.jp; tiger.nishimura@
aist.go.jp
0040–4020/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2007.01.011

H. Shimizu et al. / Tetrahedron 63 (2007) 2418–2425
2419
compound conjugated through Au–S bond to GCNPs, we
can monitor the reactions by MALDI-TOF MS. In addition,
if we can avoid reduction of the chemical reactivity by car-
rying out reactions in the colloidal phase, we can overcome
some of the disadvantages in polymer or solid supported
syntheses as mentioned above.
the GCNPs with undecyl tetra(ethylene glycol) were pre-
cipitated when Lewis acid (TMSOTf or BF3OEt2) was
added to a solution of the GCNPs in organic solvents,
e.g. acetonitrile, dichloromethane. Finally hexa(ethylene
glycol) was used in our study for carbohydrate elongation
and deprotection.
Here we will report some of our efforts to open the door to
new oligosaccharide synthesis methods. We have chosen a
lactose derivative as a target of the demonstration of this
colloidal phase synthesis, because it required two sets of
alternating steps between the glycosylation and selective
deprotection and because the use of lactose on GCNPs has
already been reported.14
O
Br
+
HO
OH
5
8
a
O
HO
6
8
1
b
6
2. Results and discussion
2.1. Strategy
O
S
HO
9
2
2
Figure 1 shows our strategy for this research. The linker be-
tween the core gold and sugar functioned also as a stabilizer
of the GCNPs. Glycosylation and selective deprotection of
hydroxyl protecting groups were performed in order on these
GCNPs. In each step, reagents could be easily washed out by
ultrafiltration. The reactions could be detected directly by
MALDI-TOF MS.
c
O
Au
S
HO
6
9
3
Scheme 1. Preparation of gold nanoparticles, 3. Reagents and conditions:
(a) NaH, 100 ꢀC, 18 h (67%); (b) (1) AcSH, AIBN, toluene, 70 ꢀC, 20 h;
(2) NaOMe, MeOH, rt, 18 h (83%); (c) KAuCl4, NaBH4, MeOH/H2O
(3/2), rt, 2 h.
2.2. Preparation of the GCNPs
Compound 120 was prepared by a coupling of hexaethylene
glycol and 11-bromo-1-undecene with sodium hydride in
67% yield. Radical reaction with thioacetic acid followed
by saponification gave disulfide 2 in 83% yield. Reduction
of tetrachloroauric acid with sodium borohydride in a mixed
solvent of methanol and water was performed in the pres-
ence of 2.16 Depending on the mixed solvent ratio, we got
two differently characterized GCNPs (Fig. 2). When the re-
action was carried out in 15/1 MeOH/water, GCNPs turned
red, transmission electron microscopy (TEM) showed that
the majority of GCNPs were well-separated as single parti-
cles with diameters ranging from 5 to 8 nm (Fig. 3), and di-
sulfide 2 was mainly observed by MALDI-TOF MS analysis.
On the other hand, when the reaction was carried out in 3/2
MeOH/water, the GCNPs’ color became brown, the size of
the GCNPs was less than 5 nm, and the monosulfide, which
was a thiol of undecyl hexa(ethylene glycol), was mainly ob-
served by MALDI-TOF MS analysis. In this study, we have
used the latter GCNPs as they make MS analysis and scien-
tific discussion easier.
First, as in Scheme 1, we synthesized a long alkanethiol
chain having a hydrophilic oligomeric ethylene glycol,
which functioned as a stabilizer of GCNPs and also as
a linker to connect synthesized oligosaccharides and
GCNPs.18 Yamada et al. showed peptide syntheses on gold
nanoparticles with a simple linker such as undecyl alkane
and got good yields,19 but in our study, glycosylation re-
quired dissolving in organic solvents and deprotection to
yield oligosaccharides would proceed in water or alcohols.
In addition, for our further aim of using them for biological
research, i.e. binding assay, it was necessary to dissolve in
water. So we had decided to introduce ethylene glycols as
part of the linkers. Glycosylation did not occur onto the
GCNPs with undecyl di(ethylene glycol) as a linker, and
Linker
Au
HO
S
1) Glycosylation
2) Deprotection
Oligosaccharide
2.3. Preparation of monosaccharide derivatives
Linker
Au
S
As donors of glycosylations onto GCNPs, we have designed
the glucose derivative 7 and galactose derivative 10 (Scheme
2). After levulinoylation of 4,21 deprotection of the allyl
group on the anomer gave hemiacetal 6. Treatment with tri-
chloroacetonitrile and DBU (1,8-diazabicyclo[5.4.0]undec-
7-ene) afforded the trichloroacetoimidate donor 7 in 87%
yield. Similarly, for the galactosyl moiety, levulinoylation
and deprotection of the p-methoxyphenyl group followed
by treatment with trichloroacetonitrile and DBU gave the
glycosyl donor 10.
MALDI-TOF MS
Oligosaccharide
Linker
SH
or
Oligosaccharide
Linker
Linker
S
S
Oligosaccharide
Figure 1. Overview of synthesis on gold nanoparticles and detection by
MALDI-TOF MS.

2420
H. Shimizu et al. / Tetrahedron 63 (2007) 2418–2425
OBz
OBz
O
O
a
c
b
HO
LevO
BzO
BzO
BzO
BzO
OAll
OAll
4
5
OBz
O
OBz
O
LevO
BzO
LevO
BzO
NH
CCl
BzO
BzO
OH
O
3
7
6
HO
LevO
BzO
size: 5 - 8 nm
size: <5 nm
MeOH / H2O (3:2)
were used as a solvent
OBz
O
OBz
O
d
MeOH / H2O (15:1)
were used as a solvent
to prepare 3.
OMP
OMP
BzO
BzO
8
BzO
to prepare 3.
9
LevO
BzO
m/z
m/z
OBz
O
e
monomer
dimer
NH
BzO
O
CCl
3
monomer
dimer
10
Scheme 2. Preparation of monosaccharide synthetic moieties. Reagents
and conditions: (a) LevOH, DCC, DMAP, CH2Cl2, 0 ꢀC–rt, 15 h (99%);
(b) Ir{(COD)[PCH3Ph2]2}PF6, THF, rt, 2 h, then 1 M HCl, AcOH, 70 ꢀC,
2 h (93%); (c) CCl3CN, DBU, CH2Cl2, rt, 2 h (87%); (d) LevOH, DCC,
DMAP, CH2Cl2, 0 ꢀC–rt, 15 h (89%); (e) (1) CAN, toluene/CH3CN/H2O
(4/3/3), 0 ꢀC–rt, 4.5 h; (2) CCl3CN, DBU, CH2Cl2, rt, 1.5 h (80%).
Oligosaccharide
Oligosaccharide
Linker
Linker
Oligosaccharide
Linker
Figure 2. Physical characteristics of gold nanoparticles. When the gold
nanoparticle was prepared in MeOH/H2O (15/1), particle size was 5–8 nm
and was observed mainly as a dimer of oligosaccharide by MALDI-TOF
MS, while the gold nanoparticle was prepared in MeOH/H2O (3/2), particle
size was smaller than 5 nm and was observed mainly as a monomer of
oligosaccharide by MALDI-TOF MS.
ion of 11, this result of only 25% of hydroxyl groups being
glycosylated might be reasonable. In the case of the size of
GCNPs assumed as 4 nm, the number of linkers onto these
GCNPs could be calculated as 326 molecules.22 The surface
area of one GCNP (f¼4 nm) was around 50 nm2, thus occu-
pation area for one linker onto GCNPs was around 0.15 nm2.
This indicated that the linker existed every 0.4 nm. If the car-
bohydrate elongation could not occur to the linker existing
side by side with the already glycosylated linker due to the
3
a
OBz
O
Au
S
S
O
LevO
BzO
O
O
9
6
11
OBz
b
OBz
O
O
Au
HO
9
6
BzO
OBz
12
c
LevO
BzO
OBz
O
Figure 3. TEM image of GCNP prepared in MeOH/water (15/1).
OBz
O
O
Au
S
O
O
9
6
BzO
OBz
OBz
13
2.4. Carbohydrate elongation onto GCNP 3
The first glycosylation onto GCNPs 3 was performed with
the glycosyl donor 7 and the promoter trimethylsilyl trifluoro-
methanesulfonate (Scheme 3). MALDI-TOF MS for glyco-
sylated GCNPs gave ion signals, some of which were at m/z
491 and at m/z 1063 for the sodium adducts of the linker thiol
itself and the glycosylated linker, respectively. The ratio of
their intensity was around 3:1 (Fig. 4a). Although it was
hard to discuss the glycosylation yield based on these inten-
sities because their ionization efficiency would be different
and the ion signal at m/z 491 might contain the fragment
d
HO
HO
OH
O
OH
O
O
Au
S
O
O
9
6
HO
OH
OH
14
Scheme 3. Carbohydrate elongation onto gold nanoparticle. Reagents and
conditions: (a) (1) 7, TMSOTf, CH2Cl2, CH3CN, 0 ꢀC–rt, 3 h; (2) Ac2O,
Py, rt, 12 h; (b) H2NNH2$H2O, AcOH, THF, MeOH, 0 ꢀC–rt, 3 h; (c) 10,
TMSOTf, CH3CN, 0 ꢀC–rt, 3 h; (d) 0.5 M NaOMe in MeOH, rt, 12 h.

H. Shimizu et al. / Tetrahedron 63 (2007) 2418–2425
2421
Intens. [a.u]
X104
HO
OBz
SH
a)
O
6
9
O
LevO
O
491
BzO
SH
9
O
6
OBz
2
1063
11 before capping
11 after capping
Acetyl
Capping
+Ac
533
4
2
b)
c)
d)
Deprotection
of Lev group
Compound 12
4
2
-Lev
965
LevO
BzO
OBz
?
O
O
OBz
2
Galactosylation
Compound 13
?
?
2
1
(1185)
1538
Deprotections
e)
Compound 14
491
observed as dimers
653
600
815
400
800
1000
HO
1200 1400 1600
1800
SH
m/z
OH
OH
O
O
O
HO
O
HO
O
OH
9
6
OH
OH
O
OH
HO
HO
HO
O
O
O
or
SH
HO
O
6
SH
9
O
6
OH
9
OH
Figure 4. Tendency for MALDI-TOF MS along chemical reactions onto GCNPs: (a) 11 before acetyl capping; (b) 11 after acetyl capping; (c) 12 obtained by
deprotection of the levulinoyl group of 11; (d) 13 obtained by galactosylation of 12; (e) 14 after deprotection of 13.
steric hindrance of benzoyl protected carbohydrates, theo-
retically only 25% of the linkers could be glycosylated.
very well as there was no signal at m/z 965. Meanwhile, on
the other hand, we could see signals of a monosaccharide
with the linker having a levulinoyl group at m/z 1063,
although we have determined that there was no levulinoy-
lated monosaccharide in the previous step as shown in
Figure 4c. It might be that either the levulinoylated mono-
saccharide was not detected in Figure 4c due to a lesser ion-
ization efficiency than other compounds, or some of the
acetyl capped linker had been deacylated in the delevulinoy-
lation stage and subsequently galactosylated. We also ob-
served disaccharide at m/z 1185 in Figure 4d. This might
be the result of donor byproduct in the galactosylation step
as 1–1 linked trehalose-type, which indicated that washing
of the gold nanoparticle 13 was not enough although it
was washed carefully several times. To confirm which com-
pounds were actually synthesized onto the GCNPs, depro-
tection of hydroxyl acetate was carried out against 13 and
the MALDI-TOF MS result is shown in Figures 4e and 5.
We have observed not only the designed supposed lactose
derivatives 14, but also oligosaccharide moieties, namely
monosulfides and disulfides of disaccharide (supposed lac-
tose), monosaccharides (glucose or galactose), and the orig-
inal linker itself. (N.B. In this discussion, we have not
considered the low m/z area (< around 700) in MALDI-
TOF MS, unless particular assignable numbers were seen.)
In addition as shown in Figure 5, all the disulfide signals
Acetyl capping for these non-reacted hydroxyl groups would
be necessary to avoid unfavorable byproducts in further syn-
thesis steps and to reveal whether the fragment ion’s signals
have been obtained at m/z 491 or not. Although small signals
of disulfides were observed in Figure 4b at m/z 1041 and at
m/z 1572 (not marked) as the sodium adducts of the dimer of
the acetylated linkers and the conjugated compound with
glycosylated and acetylated linkers, respectively, the sodium
adducts of monosulfides, the acetylated linker thiol and the
glycosylated linker have been observed at m/z 533 and at
m/z 1063, respectively. The ratio of their intensities was
around 4.5:1, and we could not find the linker thiol itself
at m/z 491. Selective deprotection of the levulinoyl group
in 11 proceeded well, which was proved by MALDI-TOF
MS of 12 as shown in Figure 4c. The second glycosylation,
introducing the galactose moiety, which resulted in the sup-
posed lactose derivative also succeeded, but we should point
out a few things at this stage. The resulting disaccharide
would be a lactose due to the use of regioselective hydroxyl
protected monosaccharides and well-known neighboring ef-
fect for glycosylation, but unfortunately only mass spectral
data were available. MALDI-TOF MS for 13, as shown in
Figure 4d, revealed that the second glycosylation proceeded

2422
H. Shimizu et al. / Tetrahedron 63 (2007) 2418–2425
Figure 5. Details of Figure 4e, which is the MALDI-TOF MS of GCNPs 14. Mass signals were assigned.
came from the combination of the linker with an oligosaccha-
ride and the one without. This result supported the assump-
tion that the carbohydrate elongation could not occur to the
linker existing side by side with the already glycosylated
linker on the GCNPs, because it is natural to think that cou-
pling the closest linkers with each other to form the disulfides
in the ionization stage for the MALDI-TOF MS was favored.
and used as purchased without further purification. Reac-
tions in the solution phase were monitored by TLC, which
was performed with 0.25 mm precoated silica gel 60F254
on glass from Merck (Darmstadt, Germany). Compounds
were detected by dipping the TLC plates in an ethanolic so-
lution of sulfuric acid (5% v/v) and heating. Silica gel N60
(40–50 nm) from Kanto Chemical (Tokyo, Japan) was used
for silica gel chromatography. Melting points were deter-
mined on an ASONE ATM-01 and are uncorrected. Optical
rotations were recorded on a Perkin Elmer Polarimeter 343
at room temperature (approximately 18–23 ꢀC). All NMR
data are reported in parts per million downfield shift from
tetramethylsilane. 1H NMR spectra were routinely recorded
at 400 MHz on a BRUKER AVANCE 400 spectrometer at
300 K and chemical shifts are expressed relative to that of
the residual proton in the NMR solvents (d 7.26 ppm for
CHCl3). 13C NMR spectra were routinely recorded at
100 MHz on a BRUKER AVANCE 400 spectrometer at
300 K and chemical shifts are expressed relative to that of
the deuterated solvents (d 77.0 ppm for CDCl3). High-reso-
lution FAB-mass data were recorded by the in-house mass
spectrometry service at the Center for Instrumental Analy-
sis, Hokkaido University. MALDI-TOF MS analyses were
performed in reflector mode with the Ultraflex instrument
(Bruker, Germany) with 2,5-dihydroxybenzoic acid (DHB).
Ions generated by a pulsed UV laser beam (nitrogen laser,
l¼337 nm, 5 Hz) were accelerated to a kinetic energy of
23.5 kV. Ultrafiltration was performed by Centriplus YM-
50 (MWCO¼50,000) and prewashed once with proper
solvents.
3. Conclusions
In conclusion, we have succeeded in demonstrating carbo-
hydrate elongations on GCNPs, which has some advantages
over solid-phase syntheses, namely the direct monitoring of
the chemical reaction by MALDI-TOF MS and the fact that
the resulting GCNPs can be applied directly to biological
carbohydrate research. There was good reactivity onto
GCNPs as we have discussed for the galactosylation in
Figure 4d, but it is hard to conclude at this point that the gly-
cosylation reactivity of the colloidal phase is better than that
of solid phase. Although we successfully obtained these
fruitful results to create a new approach for oligosaccharide
synthesis, many issues remain to be solved including de-
termination of the suitable density of glycosylation points
(linkers) onto GCNPs for the synthesis of oligosaccharides,
establishment of a quantitative monitoring system by
MALDI-MS, application of large scale synthesis with intro-
duction of a cleavage point to yield the oligosaccharide
itself, preparation of more complicated oligosaccharides,
and so on. We will report these in due course.
4.2. Preparation of the GCNPs
4. Experimental
4.1. General
4.2.1. 3,6,9,12,15,18-Hexaoxa-28-nonacosen-1-ol (1).20
To stirred hexaethylene glycol (11.5 mL, 46.0 mmol)
were added sodium hydride (460 mg, 60% oil disper-
sion, 11.5 mmol) and 11-bromo-1-undecene (2.50 mL,
11.5 mmol) successively. The reaction mixture was stirred
overnight at 100 ꢀC, cooled to room temperature, and poured
All reagents were purchased from Wako Pure Chemical
Industries Ltd. (Osaka, Japan), Tokyo Kasei Kogyo Co. (To-
kyo, Japan) or Aldrich Chemical Co. (Milwaukee, WI, USA)

H. Shimizu et al. / Tetrahedron 63 (2007) 2418–2425
2423
into water. The aqueous layer was extracted four times with
chloroform. Combined extracts were dried over MgSO4 and
concentrated in vacuo. The resulting residue was purified by
column chromatography [silica gel: EtOAc/EtOH/water (25/
5.0, 2.3 Hz, H-5), 4.25 (1H, ddt, J¼13.1, 5.2, 1.5 Hz,
CH2]CHCH2–), 4.07 (1H, ddt, J¼13.1, 6.1, 1.5 Hz,
CH2]CHCH2–), 2.66–2.37 (4H, m, H3CCOCH2CH2CO–
O–), 2.00 (3H, s, H3CCOCH2CH2COO–); 13C NMR
(100 MHz, CDCl3): d 205.8 (H3CCOCH2CH2COO–),
171.5 (H3CCOCH2CH2COO–), 166.2 (C6H5CO–),
1
2/1)] to give 1 (3.37 g, 67%) as a colorless oil; H NMR
(400 MHz, CDCl3): d 5.75 (1H, ddt, J¼17.1, 10.2, 6.8 Hz,
CH2]CHCH2–), 4.94 (1H, dd, J¼17.1, 2.0 Hz, CH2]
CHCH2–), 4.87 (1H, dd, J¼10.2, 2.0 Hz, CH2]CHCH2–),
3.68–3.51 (24H, m, –O(CH2CH2O)6H), 3.39 (2H, t,
J¼6.8 Hz, –CH2O(CH2CH2O)6H), 2.89 (1H, br s, –OH),
1.99 (2H, q, J¼6.8 Hz, CH2]CHCH2–), 1.52 (2H, qui,
J¼6.8 Hz, C9H17CH2CH2O–), 1.34–1.18 (12H, m,
C3H5C6H12C2H4O–).
165.79
(C6H5CO–),
165.77
(C6H5CO–),
118.0
(CH2]CHCH2–), 95.2 (C-1), 71.8 (C-2), 70.6 (C-3), 69.0
(CH2]CHCH2–), 68.9 (C-4), 67.9 (C-5), 62.6 (C-6), 37.9
(H3CCOCH2CH2COO–), 29.4 (H3CCOCH2CH2COO–),
27.9 (H3CCOCH2CH2COO–); FAB HRMS m/z for
C35H35O11 (M+H+) calcd 631.2179, found 631.2191.
4.3.2. 2,3,6-Tri-O-benzoyl-4-O-levulinoyl-a,b-D-gluco-
pyranose (6). A solution of Ir{(COD)[PCH3Ph2]2}PF6
(156 mg, 184 mmol) in THF (3 mL) was stirred under hydro-
gen atmosphere for 5 min and the reaction mixture turned
clear. After degassing and changing to nitrogen atmosphere,
to the mixture was added 5 (2.32 g, 3.68 mmol) in THF
(10 mL). The reaction mixture was stirred for 2 h at room
temperature, and then were added acetic acid (10 mL) and
1 M HCl (2 mL). The reaction mixture was then stirred for
2 h at 70 ꢀC, cooled to room temperature, and concentrated
in vacuo with toluene. The residue was dissolved in EtOAc,
washed with satd NaHCO3 aq twice and brine, dried over
MgSO4, and concentrated in vacuo. The resulting residue
was purified by column chromatography [silica gel: n-hex-
ane/EtOAc (2/1, 1/1)] to give 6 (2.02 g, 93%, a/b¼ca. 5/1
1H NMR (400 MHz, CDCl3): d 8.13–7.34 (15H, m, a- and
b-Aromatic-H), 6.07 (1H, t, J¼10.0 Hz, a-H-3), 5.77 (1H,
t, J¼9.7 Hz, b-H-3), 5.70 (1H, t, J¼3.7 Hz, a-H-1), 5.49
(1H, t, J¼10.0 Hz, a-H-4), 5.44 (1H, t, J¼9.7 Hz, b-H-4),
5.29 (1H, dd, J¼9.7, 8.0 Hz, b-H-2), 5.23 (1H, dd, J¼10.0,
3.7 Hz, a-H-2), 5.00 (1H, t, J¼8.0 Hz, b-H-1), 4.64 (1H,
dd, J¼12.3, 2.3 Hz, a-H-6), 4.64 (1H, dd, J¼12.3, 2.3 Hz,
b-H-6), 4.54 (1H, ddd, J¼10.0, 4.9, 2.3 Hz, a-H-5), 4.47
(1H, dd, J¼12.3, 4.9 Hz, b-H-6), 4.43 (1H, dd, J¼12.3,
4.9 Hz, a-H-6), 4.20 (1H, d, J¼8.0 Hz, b-OH), 4.40 (1H,
ddd, J¼9.7, 4.9, 2.3 Hz, b-H-5), 3.84 (1H, d, J¼3.7 Hz,
a-OH), 2.64–2.34 (4Hꢁ2, m, a- and b-H3CCOCH2CH2–),
1.98 (3H, s, a-H3CCO–), 1.97 (3H, s, b-H3CCO–);
13C NMR (100 MHz, CDCl3): d 206.0 (a-H3CCO–), 205.9
(b-H3CCO–), 171.5 (a-H3CCOCH2CH2COO–), 171.3
(b-H3CCOCH2CH2COO–), 166.7 (b-C6H5CO–), 166.4
(a-C6H5CO–), 166.3 (b-C6H5CO–), 165.9 (a- and b-
C6H5CO–), 165.8 (a-C6H5CO–), 96.0 (b-C-1), 90.4 (a-C-
1), 74.2 (b-C-2), 72.5 (b-C-3 or b-C-5), 72.4 (b-C-3 or
b-C-5), 72.2 (a-C-2), 70.3 (a-C-3), 68.8 (a- and b-C-4),
67.7 (a-C-5), 62.7 (b-C-6), 62.5 (a-C-6), 37.9 (a- and b-
H3CCOCH2CH2–), 29.4 (a- and b-H3CCOCH2CH2COO–),
27.9 (a- and b-H3CCO–); FAB HRMS m/z for C32H31O11
(M+H+) calcd 591.1866, found 591.1876.
4.2.2. 30,31-Dithia-3,6,9,12,15,18,43,46,49,52,55,58-do-
decaoxa-1,60-hexacontanediol (2). To a stirred solution
of 1 (2.78 g, 6.40 mmol) in toluene (20 mL) were added
thioacetic acid (4.76 mL, 64.0 mmol) and AIBN (2,20-
azobisisobutyronitrile) (525 mg, 3.20 mmol) successively.
The reaction mixture was stirred for 20 h at 70 ꢀC, cooled
to room temperature, and concentrated in vacuo. The result-
ing residue was roughly purified by column chromatography
[silica gel: EtOAc/EtOH/water (25/2/1)]. To the product in
methanol (10 mL) was added sodium methoxide (173 mg,
3.20 mmol). The reaction mixture was stirred overnight at
room temperature, neutralized with Dowex, and concen-
trated in vacuo. The resulting residue was purified by column
chromatography [silica gel: EtOAc/EtOH/water (10/2/1)] to
give 2 (2.49 g, 83%) as a colorless oil; 1H NMR (400 MHz,
CDCl3): d 3.74–3.56 (48H, m, –O(CH2CH2O)6Hꢁ2),
3.44 (4H, t, J¼7.0 Hz, –CH2O(CH2CH2O)6Hꢁ2), 2.71
(2H, br s, –O(CH2CH2O)6Hꢁ2), 2.68 (4H, t, J¼7.3 Hz,
–SCH2–ꢁ2), 1.67 (4H, qui, J¼7.3 Hz, –SCH2CH2–ꢁ2),
1.57 (4H, qui, J¼7.0 Hz, C9H17CH2CH2O–ꢁ2), 1.39–1.27
(28H, m, –SC2H4C7H14–ꢁ2); 13C NMR (100 MHz,
CDCl3): d 72.6, 71.5, 70.63, 70.59, 70.5, 70.43, 70.35,
70.1, 61.8 (–CH2OH), 39.2 (–SCH2–), 29.65, 29.58, 29.54,
29.50, 29.2, 28.5, 26.1; FAB HRMS m/z for C46H95O14S2
(M+H+) calcd 935.6163, found 935.6171.
1
by H NMR) as a yellow oil; [a]2D0 +108.1 (c 1.0, CHCl3);
4.3. Preparation of monosaccharide derivatives
4.3.1. Allyl 2,3,6-tri-O-benzoyl-4-O-levulinoyl-a-D-gluco-
pyranoside (5). To a stirred solution of allyl 2,3,6-tri-O-
benzoyl-a-D-glucopyranoside (4)21 (1.97 g, 3.69 mmol),
DMAP (4-(dimethylamino)pyridine) (90.2 mg, 738 mmol),
and DCC (N,N0-dicyclohexylcarbodiimide) (914 mg,
4.43 mmol) in dichloromethane (10 mL) was added levulinic
acid (454 mL, 4.43 mmol) dropwise at 0 ꢀC and the reaction
mixture was stirred for 15 h as it was allowed to slowly warm
to room temperature. The mixture was then diluted with
chloroform, washed with brine, dried over MgSO4, and con-
centrated in vacuo. The resulting residue was purified by
column chromatography [silica gel: n-hexane/EtOAc (1/1)]
to give 5 (2.32 g, 99%) as a colorless oil; [a]2D0 +150.7 (c
1.1, CHCl3); 1H NMR (400 MHz, CDCl3): d 8.11–7.35
(15H, m, Aromatic-H), 6.03 (1H, t, J¼10.0 Hz, H-3), 5.84
(1H, dddd, J¼17.2, 10.4, 6.1, 5.2 Hz, CH2]CHCH2–),
5.44 (1H, t, J¼10.0 Hz, H-4), 5.33 (1H, d, J¼3.7 Hz, H-1),
5.27 (1H, dq, J¼17.2, 1.5 Hz, CH2]CHCH2–), 5.25 (1H,
dd, J¼10.0, 3.7 Hz, H-2), 5.13 (1H, dq, J¼10.4, 1.5 Hz,
CH2]CHCH2–), 4.59 (1H, dd, J¼12.2, 2.3 Hz, H-6), 4.47
(1H, dd, J¼12.2, 5.0 Hz, H-6), 4.34 (1H, ddd, J¼10.0,
4.3.3. 2,3,6-Tri-O-benzoyl-4-O-levulinoyl-a-D-glucopyr-
anosyl 2,2,2-trichloroacetimidate (7). To a stirred solution
of 6 (258 mg, 437 mmol) in dichloromethane (4 mL) were
added trichloroacetonitrile (438 mL, 3.45 mmol) and DBU
(13 mL, 87.4 mmol). The reaction mixture was stirred for
2 h at room temperature and directly purified by column
chromatography [silica gel: n-hexane/EtOAc (5/2, 1/1)] to
give 7 (281 mg, 87%) as a yellow oil; [a]2D0 +9.9 (c 1.0,
1
CHCl3); H NMR (400 MHz, CDCl3): d 8.61 (1H, s, NH),

2424
H. Shimizu et al. / Tetrahedron 63 (2007) 2418–2425
8.09–7.34 (15H, m, Aromatic-H), 6.78 (1H, d, J¼3.7 Hz,
H-1), 6.11 (1H, t, J¼10.0 Hz, H-3), 5.57 (1H, t, J¼
10.0 Hz, H-4), 5.53 (1H, dd, J¼10.0, 3.7 Hz, H-2), 4.65–
4.61 (1H, m, H-6), 4.54–4.48 (2H, m, H-5 and H-6), 2.66–
2.38 (4H, m, H3CCOCH2CH2–), 1.99 (3H, s, H3CCO–);
13C NMR (100 MHz, CDCl3): d 205.6 (H3CCO–), 171.5
(H3CCOCH2CH2COO–), 166.1 (C6H5CO–), 165.7
(C6H5CO–), 165.4 (C6H5CO–), 160.5 (–OC(]NH)CCl3),
93.1 (C-1), 90.7 (–OC(]NH)CCl3), 70.64 (C-2 or C-5),
70.63 (C-2 or C-5), 70.3 (C-3), 68.0 (C-4), 62.1 (C-6),
37.9 (H3CCOCH2CH2–), 29.4 (H3CCOCH2CH2–), 27.9
(H3CCO–); FAB HRMS m/z for C34H30Cl3NNaO11
(M+Na+) calcd 756.0782, found 756.0771.
m, Aromatic-H), 6.83 (1H, d, J¼3.6 Hz, H-1), 5.96
(1H, dd, J¼9.5, 3.2 Hz, H-3), 5.90 (1H, dd, J¼3.2, 1.0 Hz,
H-4), 5.83 (1H, dd, J¼9.5, 3.6 Hz, H-2), 4.75 (1H,
ddd, J¼6.9, 6.3, 1.0 Hz, H-5), 4.54 (1H, dd, J¼11.4,
6.9 Hz, H-6), 4.42 (1H, dd, J¼11.4, 6.3 Hz, H-6), 2.77–
2.69 (4H, m, H3CCOCH2CH2–), 2.10 (3H, s, H3CCO–);
13C NMR (100 MHz, CDCl3):
d 205.7 (H3CCO–),
171.8 (H3CCOCH2CH2COO–), 165.9 (C6H5CO–), 165.6
(C6H5CO–), 165.5 (C6H5CO–), 160.6 (–OC(]NH)CCl3),
93.8 (C-1), 90.8 (–OC(]NH)CCl3), 69.6 (C-5), 68.2
(C-3), 68.1 (C-4), 67.7 (C-2), 61.9 (C-6), 37.8
(H3CCOCH2CH2–), 29.7 (H3CCOCH2CH2–), 27.8
(H3CCO–); FAB HRMS m/z for C34H30Cl3NNaO11
(M+Na+) calcd 756.0782, found 756.0797.
4.3.4. p-Methoxyphenyl 2,3,6-tri-O-benzoyl-4-O-levuli-
noyl-b-D-galactopyranoside (9). To a solution of p-
methoxyphenyl 2,3,6-tri-O-benzoyl-b-D-galactopyranoside
(8)23 (1.83 g, 3.06 mmol), DMAP (74.8 mg, 612 mmol),
and DCC (758 mg, 3.67 mmol) in dichloromethane
(10 mL) was added levulinic acid (376 mL, 3.67 mmol) drop-
wise at 0 ꢀC. The reaction mixture was stirred for 15 h as it
was allowed to slowly warm to room temperature. The reac-
tion mixture was then diluted with chloroform, washed with
satd NaHCO3 aq and brine, dried over MgSO4, and concen-
trated in vacuo. The resulting residue was purified by column
chromatography [silica gel: n-hexane/EtOAc (1/1)] to give 9
(2.13 g, 89%) as a white powder; [a]2D0 +47.2 (c 1.0, CHCl3);
1H NMR (400 MHz, CDCl3): d 8.06–7.37 (15H, m, Aro-
matic-H), 6.94 (2H, d, J¼9.1 Hz, Aromatic-H), 6.66 (2H,
d, J¼9.1 Hz, Aromatic-H), 5.93 (1H, dd, J¼10.4, 8.0 Hz,
H-2), 5.79 (1H, d, J¼3.4 Hz, H-4), 5.52 (1H, dd, J¼10.4,
3.4 Hz, H-3), 5.17 (1H, d, J¼8.0 Hz, H-1), 4.64 (1H, dd, J¼
11.4, 7.6 Hz, H-6), 4.49 (1H, dd, J¼11.4, 5.7 Hz, H-6), 4.32
(1H, dd, J¼7.6, 5.7 Hz, H-5), 3.71 (3H, s, CH3O–), 2.80–
2.69 (4H, m, H3CCOCH2CH2–), 2.12 (3H, s, H3CCO–);
13C NMR (100 MHz, CDCl3): d 205.9 (H3CCO–), 171.8
4.4. Carbohydrate elongation onto GCNP 3
4.4.1. Gold nanoparticle (3). To a solution of 2 in water
(2.00 mL, 10 mM) were added MeOH (3 mL), potassium
tetrachloroaurate(III) in water (35 mL, 270 mM), and sodium
borohydride in water (200 mL, 100 mM) with vigorous stir-
ring. The reaction mixture was stirred vigorously for 2 h at
room temperature and passed through the centrifugal UF
unit (3000g, 20 min) with methanol four times to yield sup-
posed 3.
4.4.2. Glycosylations and deprotections onto gold nano-
particle (Scheme 3). After co-evaporation with toluene, to
3 (2.79 mmol of calculated amount based on hydroxyl
groups) in dichloromethane (3 mL) and acetonitrile (1 mL)
were added 7 (42.5 mg, 57.8 mmol) and TMSOTf (10 mL,
55.3 mmol) at 0 ꢀC under nitrogen atmosphere. The reaction
mixture was stirred for 3 h as it was allowed to slowly warm
to room temperature, diluted with acetonitrile, and passed
through the centrifugal UF unit (3000g, 20 min) with aceto-
nitrile four times to yield supposed 11.
(H3CCOCH2CH2COO–),
166.0
(C6H5CO–),
165.6
(C6H5CO–), 165.3 (C6H5CO–), 155.8 (Aromatic-C), 151.1
(Aromatic-C), 119.0 (Aromatic-C), 114.5 (Aromatic-C),
101.2 (C-1), 71.7 (C-3), 71.5 (C-5), 69.5 (C-2), 67.6 (C-4),
62.0 (C-6), 55.6 (CH3O–), 37.9 (H3CCOCH2CH2–), 29.7
(H3CCOCH2CH2–), 27.9 (H3CCO–); FAB HRMS m/z for
C39H37O12 (M+H+) calcd 697.2285, found 697.2292.
For acetyl capping of the remaining hydroxyl groups, to the
product were added acetic anhydride (1 mL) and pyridine
(2 mL). The reaction mixture was stirred for 12 h at room
temperature, concentrated in vacuo, dissolved in mixed
solvent, acetonitrile/methanol (1/1 v/v), and passed through
the centrifugal UF unit (3000g, 20 min) with acetonitrile/
methanol (1/1) four times.
4.3.5. 2,3-Di-O-benzoyl-4-O-levulinoyl-a-D-galactopyr-
anosyl 2,2,2-trichloroacetimidate (10). To a solution of 9
(1.90 g, 2.73 mmol) in toluene/CH3CN/water (20 mL, 4/3/3
v/v/v) was added CAN (cerium(IV) ammonium nitrate)
(13.1 g, 24.0 mmol) at 0 ꢀC and the reaction mixture was
stirred for 4.5 h as it was allowed to slowly warm to room
temperature. The reaction mixture was then diluted with
EtOAc, washed with water, satd NaHCO3 aq twice and
brine, dried over MgSO4, and concentrated in vacuo. The
resulting residue was roughly purified by column chromato-
graphy [silica gel: toluene/EtOAc (4/1, 1/1)] to give crude
supposed hemiacetal. The product (1.20 g) was dissolved
in dichloromethane, and to the solution were added trichlo-
roacetonitrile (2.80 mL, 20.3 mmol) and DBU (62 mL,
406 mmol). The reaction mixture was stirred for 1.5 h at
room temperature and directly purified by column chromato-
graphy [silica gel: toluene/EtOAc (5/2)] to give 10 (1.35 g,
80%) as a yellow oil; [a]2D0 +74.1 (c 1.0, CHCl3); 1H NMR
(400 MHz, CDCl3): d 8.60 (1H, s, NH), 8.01–7.34 (15H,
To a solution of acetyl capped 11 in THF/methanol (2.5 mL,
1/1 v/v) were added acetic acid (40 mL) and hydrazine
monohydrate (20 mL) at 0 ꢀC. The reaction mixture was
stirred for 3 h as it was allowed to slowly warm to room tem-
perature, added acetic acid (1 mL), concentrated in vacuo,
dissolved in mixed solvent, acetonitrile/methanol (1/1 v/v),
and passed through the centrifugal UF unit (3000g,
20 min) with acetonitrile/methanol (1/1) four times to yield
supposed 12.
After co-evaporation of 12 with toluene, to 12 in acetonitrile
(2 mL) were added 10 (48.5 mg, 66.0 mmol) and TMSOTf
(10 mL, 55.3 mmol) at 0 ꢀC under nitrogen atmosphere.
The reaction mixture was stirred for 3 h as it was allowed
to slowly warm to room temperature, neutralized with tri-
ethylamine (100 mL), diluted with acetonitrile, and passed
through the centrifugal UF unit (3000g, 20 min) with aceto-
nitrile four times to yield supposed 13.

H. Shimizu et al. / Tetrahedron 63 (2007) 2418–2425
2425
6. An example reported by an author: Shimizu, H.; Ito, Y.; Kanie,
O.; Ogawa, T. Bioorg. Med. Chem. Lett. 1996, 6, 2841–2846.
7. Faraday, M. Philos. Trans. R. Soc. London 1857, 147, 302–305.
8. Reviewed in: Daniel, M.-C.; Astruc, D. Chem. Rev. 2004, 104,
293–346.
A solution of 13 in 0.5 M sodium methoxide in methanol
(1.5 mL) was stirred for 12 h at room temperature. To the
reaction mixture was added acetic acid (1 mL) and passed
through the centrifugal UF unit (3000g 20 min) with metha-
nol four times to yield supposed 14.
9. Ahn-Yoon, S.; DeCory, T. R.; Baeumner, A. J.; Durst, R. A.
Anal. Chem. 2003, 75, 2256–2261.
10. Pavlov, V.; Xiao, Y.; Shlyahovsky, B.; Willner, I. J. Am. Chem.
Soc. 2004, 126, 11768–11769.
Acknowledgements
11. Perez, J. M.; Simeone, F. J.; Saeki, Y.; Josephson, L.;
Weissleder, R. J. Am. Chem. Soc. 2003, 125, 10192–10193.
12. Otsuka, H.; Akiyama, Y.; Nagasaki, Y.; Kataoka, K. J. Am.
Chem. Soc. 2001, 123, 8226–8230.
The authors thank Dr. M. Kurogochi, our colleague in Hok-
kaido University, for his support in measuring MALDI-TOF
MS, and members of Center for Instrumental Analysis, Hok-
kaido University, for measuring high-resolution FABMS.
13. Hone, D. C.; Haines, A. H.; Russell, D. A. Langmuir 2003, 19,
7141–7144.
14. de la Fuente, J. M.; Barrientos, A. G.; Rojas, T. C.; Rojo, J.;
References and notes
ꢀ
ꢀ
Can˜ada, J.; Fernandez, A.; Penades, S. Angew. Chem., Int.
Ed. 2001, 40, 2257–2261.
1. Merrifield, R. B. J. Am. Chem. Soc. 1963, 85, 2149–2154.
2. An example reported by authors: Matsushita, T.; Hinou, H.;
Kurogochi, M.; Shimizu, H.; Nishimura, S.-I. Org. Lett.
2005, 7, 877–880.
15. Templeton, A. C.; Wuelfing, W. P.; Murray, R. W. Acc. Chem.
Res. 2000, 33, 2256–2261.
ꢀ
16. Barrientos, A. G.; de la Fuente, J. M.; Rojas, T. C.; Fernandez,
ꢀ
A.; Penades, S. Chem.—Eur. J. 2003, 9, 1909–1921.
3. (a) Hinou, H.; Fumoto, M.; Ohta, T.; Matsushita, T.; Yamada,
K.; Iwata, K.; Toda, A.; Nishiguchi, S.; Ito, T.; Takimoto, A.;
Kondo, H.; Shimizu, H.; Nakagawa, H.; Deguchi, K.;
Nakahara, Y.; Inazu, T.; Nishimura, S.-I. Glycoconjugate J.
2005, 22, L093; Proceeding of XVIIIth International Sympo-
sium on Glycoconjugate, Florence, Italy, Sept 2005; (b)
Hinou, H.; Fumoto, M.; Yamada, K.; Iwata, K.; Ohta, T.;
Toda, A.; Nishiguchi, S.; Matsushita, T.; Ito, T.; Takimoto,
A.; Kondo, H.; Shimizu, H.; Nakagawa, H.; Deguchi, K.;
Nakahara, Y.; Inazu, S.; Nishimura, S.-I. Abstracts of Papers,
PACIFICHEM 2005, Hawaii, U.S.A., Dec 15–20, 2005;
MEDI 381.
4. Matsushita, T.; Hinou, H.; Fumoto, M.; Kurogochi, M.;
Fujitani, N.; Shimizu, H.; Nishimura, S.-I. J. Org. Chem.
2006, 71, 3051–3063.
5. Yoshimura, Y.; Shimizu, H.; Hinou, H.; Nishimura, S.-I.
Tetrahedron Lett. 2005, 46, 4701–4705.
17. (a) Mouradian, S.; Nelson, C. M.; Smith, L. M. J. Am. Chem.
Soc. 1996, 118, 8639–8645; (b) Su, J.; Mrksich, M. Angew.
Chem., Int. Ed. 2002, 41, 4715–4718.
18. Nagahori, N.; Nishimura, S.-I. Chem.—Eur. J. 2006, 12, 6478–
6485.
19. Yamada, H.; Kajita, K.; Honda, A. Abstracts of Papers, 22nd
Conference on Combinatorial Chemistry (JCCF22), Japan,
O-4, April 2006.
20. Pale-Grosdemange, C.; Simon, E. S.; Prime, K. L.; Whitesides,
G. M. J. Am. Chem. Soc. 1991, 113, 12–20.
21. Zhu, Y.; Kong, F. Carbohydr. Res. 2000, 329, 199–205.
22. Hostetler, M. J.; Wingate, J. E.; Zhong, C.-J.; Harris, J. E.;
Vachet, R. W.; Clark, M. R.; Londono, J. D.; Green, S. J.;
Stokes, J. J.; Wignall, G. D.; Glish, G. L.; Porter, M. D.;
Evans, N. D.; Murray, R. W. Langmuir 1998, 14, 17–30.
23. Ohlsson, J.; Magnusson, G. Carbohydr. Res. 2000, 329, 49–55.