Automated solid-phase synthesis of a β-(1,3)-glucan dodecasaccharide

Article abstract of DOI:10.1002/chem.201204518

β-Glucans are a group of structurally heterogeneous polysaccharides found in bacteria, fungi, algae and plants. β-(1,3)-D-Glucans have been studied in most detail due to their impact on the immune system of vertebrates. The studies into the immunomodulatory properties of these glucans are typically carried out with isolates that contain a heterogeneous mixture of polysaccharides of different chain lengths and varying degrees of branching. In order to determine the structure-activity relationship of β-(1,3)-glucans, access to homogeneous, structurally-defined samples of these oligosaccharides that are only available through chemical synthesis is required. The syntheses of β-glucans reported to date rely on the classical solution-phase approach. We describe the first automated solid-phase synthesis of a β-glucan oligosaccharide that was made possible by innovating and optimizing the linker and glycosylating agent combination. A β-(1,3)-glucan dodecasaccharide was assembled in 56 h in a stereoselective fashion with an average yield of 88 % per step. This automated approach provides means for the fast and efficient assembly of linker-functionalized mono- to dodecasaccharide β-(1,3)-glucans required for biological studies. β-Glucans are a group of structurally heterogeneous polysaccharides found in bacteria, fungi, algae and plants. The first automated solid-phase synthesis of a β-glucan oligosaccharide is reported. A β-(1,3)-glucan dodecasaccharide was assembled in 56 h in a stereoselective fashion with an average yield of 88 % per step (see scheme; Fmoc = 9-fluorenylmethoxycarbonyl, PIV = pivaloyl). Copyright

Full text of DOI:10.1002/chem.201204518

FULL PAPER
DOI: 10.1002/chem.201204518
Automated Solid-Phase Synthesis of a b-(1,3)-Glucan Dodecasaccharide
Markus W. Weishaupt, Stefan Matthies, and Peter H. Seeberger*[a]
Abstract: b-Glucans are a group of
structurally heterogeneous polysacchar-
ides found in bacteria, fungi, algae and
plants. b-(1,3)-d-Glucans have been
studied in most detail due to their
impact on the immune system of verte-
brates. The studies into the immuno-
modulatory properties of these glucans
are typically carried out with isolates
that contain a heterogeneous mixture
of polysaccharides of different chain
lengths and varying degrees of branch-
ing. In order to determine the struc-
ture–activity relationship of b-(1,3)-glu-
cans, access to homogeneous, structur-
ally-defined samples of these oligosac-
charides that are only available
through chemical synthesis is required.
The syntheses of b-glucans reported to
date rely on the classical solution-
phase approach. We describe the first
automated solid-phase synthesis of a b-
glucan oligosaccharide that was made
possible by innovating and optimizing
the linker and glycosylating agent com-
bination. A b-(1,3)-glucan dodecasac-
charide was assembled in 56 h in a ster-
eoselective fashion with an average
yield of 88% per step. This automated
approach provides means for the fast
and efficient assembly of linker-func-
tionalized mono- to dodecasaccharide
b-(1,3)-glucans required for biological
studies.
Keywords: automation · beta glu-
cans
· glycosylation · protecting
groups · solid-phase synthesis
Introduction
chains. Recognition of b-glucans and their interaction with
recombinant murine C-type lectin Dectin-1 have been
shown to be strongly dependent on the glucan structure.[6]
Therefore, studies utilizing isolated b-(1,3)-glucans can lead
to inconsistent and sometimes even contradictory results.[7]
In order to obtain reliable and reproducible results, the
studies concerning the biological roles of b-(1,3)-glucans
have to be performed with homogeneous, well-defined oli-
gosaccharide structures that are only accessible via chemical
synthesis. Several strategies relying on classical solution-
phase syntheses have been developed to produce pure b-
glucan oligosaccharides. A linear b-(1,3)-glucan pentasac-
charide was accessed using an iterative approach with glyco-
sylation yields ranging from 74 to 89%.[8] Other iterative ap-
proaches to b-(1,3)-glucan fragments report problems relat-
ed to the stereocontrol during glycosylation reactions.[9]
Consequently, the solution-phase approaches for longer b-
glucan structures typically rely on convergent synthesis strat-
egies employing di-, tetra- and pentasaccharide units. Nota-
ble examples are the synthesis of the linear hexasacchar-
ide,[10] and the syntheses of linear hexadeca- and branched
heptadecasaccharides by Tanaka and Takahashi.[11] The use
of pre-assembled units is efficient but only allows access to
glucans of certain chain lengths. Ideally, a strictly modular
assembly process would provide maximum flexibility and
minimal manual labor. Here, we describe the development
of an iterative automated solid-phase synthesis of linear b-
(1,3)-glucans that can produce oligosaccharides of any chain
length. The power of that approach is highlighted by the
synthesis of the linker-functionalized b-(1,3)-glucan dodeca-
saccharide 1.
Among the most abundant polysaccharides in bacteria,
fungi, algae and plants are the b-glucans, a heterogeneous
class of oligosaccharides composed of d-glucose. b-Glucans
play various biological roles, as components of the cell wall
or as energy storage polysaccharides. The b-(1,3)-glucans in
particular are polymers of d-glucose with a linear b-(1,3)-
glycosidic backbone. These polysaccharides can either be
linear or have side chains of d-glucose that are either b-
(1,4)- or b-(1,6)-linked to the backbone.[1] b-Glucans are rec-
ognized by the innate immune system and have been shown
to have immunomodulatory,[2] antitumor,[3] antibacterial and
antifungal properties.[4] Recent studies identified synthetic
b-glucans as the basis for development of carbohydrate vac-
cines.[5]
Most of the biological studies on b-glucans use material
isolated from natural sources. These isolates usually contain
a structurally heterogeneous mixture of b-(1,3)-glucan oligo-
saccharides of variable chain lengths and with varying side
[a] M. W. Weishaupt, S. Matthies, Prof. Dr. P. H. Seeberger
Department of Biomolecular Systems
Max Planck Institute of Colloids and Interfaces
Am Mꢀhlenberg 14, 14476 Potsdam (Germany)
Fax : (+49)30-838-59302
and
Department of Chemistry and Biochemistry
Freie Universitꢁt Berlin
Arnimallee 22, 14195 Berlin (Germany)
Fax : (+49)331-567-9302
E-mail: Seeberger@mpikg.mpg.de
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201204518.
Chem. Eur. J. 2013, 19, 12497 – 12503
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
12497

Results and Discussion
acetal was seen as a suitable protection for the C4 and C6
hydroxyl groups as it has already been successfully em-
ployed in several syntheses of b-(1,3)-glucans.[8,10–11] The
Fmoc group was chosen for temporary protection of the C3
hydroxyl for of its stability to acids, the ease of assessing the
efficiency of each glycosylation step[13] and the mild condi-
tions required for its removal after each coupling step.
Merrifield resin[14] was selected as solid support since it
has been employed successfully in the solid-phase synthesis
of carbohydrates.[15] The versatility of functionalized Merri-
field resin 13 in the synthesis of various oligosaccharides has
recently been demonstrated.[15e] The base-labile linker
allows for the simultaneous cleavage of the crude products
from the solid support and the removal of base-labile pro-
tecting groups present on the oligosaccharide. Furthermore,
it is stable to the most common activation conditions for
thioglycosides.
High coupling efficiencies for each glycosylation step are re-
quired to synthesize b-(1,3)-glucan oligosaccharides such as
dodecasaccharide 1 on solid support (Figure 1). Control of
Figure 1. Retrosynthetic analysis of dodecasaccharide 1.
Building block 8 was synthesized in seven steps starting
from b-d-glucose pentaacetate 4 (Scheme 1). Intermediate 5
was prepared according to previously reported proce-
dures.[12,16] Esterification of the C2 hydroxyl group with piv-
aloyl chloride afforded thioglycoside 6. Removal of the TBS
silyl ether using a buffered solution of TBAF·3H2O and gla-
cial acetic acid gave alcohol 7 that was treated with FmocCl
to afford thioglycoside building block 8. Functionalized Mer-
rifield resin 13 was prepared according to previously descri-
bed procedures.[15e]
The solid phase assembly of b-(1,3)-glucans on functional-
ized solid support 13 was performed using three equivalents
of thioglycoside 8 in three repetitions per glycosylation cycle
(Scheme 2). The glycosylations were carried out on the au-
tomated synthesizer[15e] by suspending functionalized resin
13 and three equivalents of glycosylating agent 8 in di-
chloromethane at ꢀ408C before a solution of NIS and
TfOH in dichloromethane/acetonitrile (1:1) was added drop-
wise via a syringe pump. The reaction mixture was agitated
using an argon flow while maintaining the temperature for
five minutes. Then the temperature was increased to ꢀ108C
and the mixture was agitated for another 40 min before
draining the reaction vessel using argon pressure. This pro-
cedure was executed three times, followed by washing steps
using different solvents. Following each glycosylation cycle,
the temporary Fmoc protecting group was removed using a
20% (v/v) solution of piperidine in DMF. The reaction
vessel was drained and the resin was washed. These steps
were repeated until oligosaccharides of the desired chain
lengths were assembled (Scheme 2). The crude product was
cleaved from the solid support by saponification with
sodium methoxide in a mixture of methanol and dichloro-
methane.
the anomeric configuration of the newly formed glycosidic
bonds is an imperative in order to avoid the formation of
complex mixtures of products. The design of the building
block including protecting group pattern and anomeric leav-
ing group, the choice of linker and the type of solid support
were made mindful of these requirements (Scheme 1).
Thioglucoside 8 was envisioned as a reliable building
block to produce b-configured glucans. The anomeric leav-
ing group was chosen since it has already been used for the
introduction of b-glucosidic linkages,[12] while the pivaloyl
ester in the C2 position was introduced to ensure selective
formation of b-glycosidic linkages. The 4,6-O-benzylidene
b-(1,3)-Glucan trisaccharide 14 served as the first synthet-
ic target and a context for the optimization of reaction con-
ditions (Scheme 2). The automated elongation cycles were
followed by saponification. LC-MS analysis indicated that
the main fraction of the crude material contains desired tri-
saccharide 14 (Figure 2). The contents of minor fractions
were identified as various resin/linker degradation side
Scheme 1. Synthesis of glycosylating agents 2 and 7. a) 2-Methyl-5-tert-
butylthiophenol, BF3·OEt2, CH2Cl2, 94%; b) NaOMe, MeOH, quant.;
c) benzaldehyde dimethyl acetal, CSA, acetonitrile, 87%; d) TBSCl, imi-
dazole, DMF, 08C, 69%; e) PivCl, DMAP, pyridine, 79%;
f) TBAF·3H2O, AcOH, DMF, 97%; g) FmocCl, pyridine, CH2Cl2, 97%;
h) BH3·THF, TMSOTf, CH2Cl2, 97%; i) benzyl bromide, NaH, DMF,
78%; j) TBAF·3H2O, AcOH, DMF, 85%; k) FmocCl, pyridine, CH2Cl2,
91%; l) dibutyl phosphate, NIS, TfOH, 4 ꢃ MS, CH2Cl2, 71%.
12498
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 12497 – 12503

Automated Carbohydrate Synthesis
FULL PAPER
It was suspected that the combination of protecting
groups on thioglycoside 8 was partly responsible for difficul-
ties in obtaining structures longer than trisaccharides. Previ-
ously, Ensley et al. reported conformational distortions in
the hexopyranose ring of glucose residues in protected b-
(1,3)-glucan structures.[10] Those glucose residues were
equipped with a bulky ester protecting group on C2, similar
to the pivaloyl group, in combination with a 4,6-O-benzyli-
dene acetal. It is conceivable that a conformational distor-
tion in combination with steric hindrance caused by the pro-
tecting groups on glycosylating agent 8 prevents chain elon-
gation past the trisaccharide stage.
In order to test this hypothesis, building block 7 was
modified by replacing the rigid 4,6-O-benzylidene acetal
with benzyl ether protecting groups on C4 and C6. The
benzyl ether groups allow for more conformational flexibili-
ty, increase the reactivity of the glycosylating agent[17] and
are stable to both the acidic glycosylation and the basic de-
protection conditions. Furthermore, the anomeric leaving
group of thioglycoside 8 was replaced by the more reactive
dibutylphosphate group, leading to glycosylating agent 2
(Figure 1, Scheme 1).[18] In our hands, functionalized resin 13
was not compatible with the conditions used for glycosyl
phosphate activation. Consequently, further optimization
relied on functionalized Merrifield resin 16 with a photola-
bile linker (3) as the solid support.[18b] Building block 2 was
synthesized from thioglycoside intermediate 6 in five steps,
or from starting material 4 in ten steps, respectively. Regio-
selective opening of the benzylidene acetal yielded mono-
saccharide 9, and subsequent benzylation of the resulting
free C6 hydroxyl group led to compound 10. Removal of
the silyl ether at C3 gave free alcohol 11 that was trans-
formed into Fmoc carbonate 12. Conversion of the thiogly-
coside to a glycosyl phosphate afforded final building block
2 in 42% yield over five steps (Scheme 1). Photolabile
linker 3 was synthesized and coupled to the Merrifield resin
to obtain linker-functionalized solid support 16 (Sche-
me 3).[18b]
Glycosylations on functionalized solid support 16 utilized
three equivalents of glycosyl phosphate 2 in three repeats
for each glycosylation cycle (Scheme 3). For each glycosyla-
tion, three equivalents of TMSOTf were added dropwise via
a syringe pump to the reaction vessel containing nucleophile
and three equivalents of glycosylating agent 2 at ꢀ158C.
The temperature was maintained for 45 min, then raised to
08C and kept at 08C for 15 min. Following each glycosyla-
tion cycle, the temporary Fmoc protecting group was re-
moved using a solution of piperidine in DMF (20% (v/v)).
These program modules were performed twelve times, and
the reactions were monitored by Fmoc quantification. By
collecting the individual solutions resulting from cleavage of
each of the Fmoc protecting groups and measuring the ab-
sorption of those solutions at 301 nm, the yield of each gly-
cosylation cycle was calculated (see Table S1 in the Support-
ing Information). While such calculations do not provide
exact coupling yields for each cycle, they are a useful esti-
mate of the glycosylation efficiencies and deliver a real time
Scheme 2. General scheme for the automated solid-phase synthesis of b-
(1,3)-glucans using thioglycoside 8 and functionalized Merrifield resin 13.
a) 3ꢄ3 equiv 7, NIS, TfOH, CH2Cl2/dioxane, ꢀ408C (5 min) ! ꢀ108C
(40 min); b) piperidine (20% v/v), DMF, RT, 15 min (3ꢄ); c) NaOMe,
MeOH, CH2Cl2.
products, while deletion sequences could not be detected.
Encouraged by these methodological advances, our atten-
tion turned to longer oligosaccharide sequences. Attempts
to synthesize linear tetrasaccharide 15 (Scheme 2) or longer
glucans, however, were not crowned by success. In all cases
the longest structure that was detected by ESI-MS was a tri-
saccharide. The HPLC chromatograms of the crude products
matched that for the synthesis of trisaccharide 14 (Figure 2)
and, consequently, this approach was abandoned.
Figure 2. Automated syntheses of linear b-(1,3)-glucan trisaccharide 14
(lower trace) and tetrasaccharide 15 (upper trace): HPLC chromato-
grams of the crude products. Nucleosil C4, acetonitrile/water 5% (5 min)
to 95% (40 min), 254 nm.
Chem. Eur. J. 2013, 19, 12497 – 12503
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12499

P. H. Seeberger et al.
parable to previously published solution phase syntheses of
b-(1,3)-glucan oligosaccharides.[8,10–11] In contrast to the solu-
tion-phase approach, the automated solid phase assembly of
dodecasaccharide 17 including release from the solid sup-
port was finished within 56 h and required only one purifica-
tion step. The iterative approach, as opposed to the conver-
gent strategies, facilitates the synthesis of b-(1,3)-glucan
structures of any length from monosaccharide to dodecasac-
charide with only one monosaccharide building block.
Dodecasaccharide 17 was deprotected in two steps
(Scheme 4). First, the pivaloyl esters were removed with
sodium methoxide to produce deacylated dodecasaccharide
18. Finally, all remaining protecting groups were removed
by hydrogenation over Pd(OH)2 to afford fully deprotected
b-glucan dodecasaccharide 1. The structure of 1 was con-
firmed by MALDI-TOF analysis and comparison of the
1H NMR spectrum with isolated curdlan, a naturally occur-
ring linear b-1,3-glucan.[19]
Scheme 3. Automated solid-phase synthesis of b-glucan dodecasaccharide
17. a) 3ꢄ3 equiv 2, TMSOTf, CH2Cl2, ꢀ158C (45 min) ! 08C (15 min);
b) piperidine (20% v/v), DMF, RT, 15 min (3ꢄ); c) hn, CH2Cl2.
feedback on potential problems. The mean calculated glyco-
sylation yield over 12 glycosylation cycles was 89%.
After completion of the automated synthesis, the crude
product was cleaved from the solid support by UV irradia-
tion of the resin in a flow reactor.[18b] The crude mixture was
analyzed by HPLC and MALDI-TOF indicating that the
main fraction (Figure 3) contains desired dodecasaccharide
17. Isolation of this HPLC fraction and analysis by NMR es-
tablished that the structure of this material corresponds to
dodecasaccharide 17. The JC1ꢀH1 coupling constants obtained
by HSQC measurements confirmed the correct stereochem-
istry of the desired 1,2-trans-glycosidic linkages (JC1ꢀH1
161 Hz).
=
Scheme 4. Global deprotection of 17 to yield dodecasaccharide 1.
a) NaOMe, MeOH, CH2Cl2, 56%; b) Pd(OH)2, H2, THF, H2O, AcOH,
52%.
Conclusion
In summary, an automated approach to the synthesis of b-
(1,3)-glucans on Merrifield resin was developed. In the con-
text of this synthesis, the glycosylating agent was optimized,
as were the linker and the reaction conditions. Glycosyl
phosphate building block 2 and photolabile-linker-function-
alized Merrifield resin 16 proved to be a suitable combina-
tion for the automated stereocontrolled synthesis of dodeca-
saccharide 17 with an average yield of 88% per step. Pro-
tected glucan 17 was deprotected in two steps to obtain the
target linker-equipped b-(1,3)-glucan dodecasaccharide (1).
This method allows for the rapid solid phase assembly of
well-defined b-(1,3)-glucans of the desired length and sup-
plies the material suitable for further SAR studies on b-
(1,3)-glucan oligosaccharides.
Figure 3. Automated synthesis of dodecasaccharide 17: HPLC chromato-
gram of the crude product. YMC-Pack Diol-300, H/EtOAc, 10% (5 min)
to 40% (70 min), 254 nm.
Dodecasaccharide 17 was obtained in 4.6% overall yield
over 25 steps corresponding to an efficiency of 88.4% per
step. If both Fmoc deprotection and photolytic cleavage
from the solid support were to proceed quantitatively, the
mean glycosylation yield were 77.4%. These yields are com-
12500
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 12497 – 12503

Automated Carbohydrate Synthesis
FULL PAPER
(101 MHz, CDCl3): d=176.8, 154.6, 149.8, 143.4, 143.2, 141.4, 141.3,
137.0, 136.8, 131.9, 130.2, 130.0, 129.3, 128.4, 128.0, 128.0, 127.4, 127.4,
126.4, 125.6, 125.4, 125.3, 120.1, 120.1, 101.8, 87.5, 78.4, 77.3, 70.6, 70.3,
68.7, 46.7, 39.0, 34.6, 31.5, 27.2, 20.5 ppm; IR (thin film): n˜ = 2964, 2868,
Experimental Section
General methods: See Supporting Information.
1752, 1255, 1098, 734 cmꢀ1
; HRMS: m/z: calcd for C44H48O8SNa
(2-Methyl-5-tert-butylphenyl)-4,6-O-benzylidene-3-O-tert-butyldimethyl-
silyl-2-O-pivaloyl-1-thio-b-d-glucopyranoside (6): Thioglycoside 5[14]
(7.80 g, 14.32 mmol) was dissolved in anhydrous pyridine (72 mL) under
argon, and DMAP (350 mg, 2.86 mmol) was added. The solution was
cooled to 08C and pivaloyl chloride (4.4 mL, 35.8 mmol) was added drop-
wise. The reaction was heated to 808C and stirred overnight. When TLC
analysis indicated complete consumption of the starting material (TLC:
cyclohexane/ethyl acetate 9:1), the mixture was diluted with CH2Cl2,
washed with 1m HCl (aq) and sat. NaHCO3 (aq) the organic layer was
dried over MgSO4 and filtered. The solvent was evaporated in vacuo and
the crude material was purified by flash column chromatography (silica
gel, cyclohexane/ethyl acetate) to afford 6 (7.09 g, 79%) as a white foam.
Rf =0.55 (cyclohexane/ethyl acetate 9:1); [a]2D0 = ꢀ33.24 (c=1.0, CHCl3);
1H NMR (400 MHz, CDCl3): d=7.56 (d, J=2.0 Hz, 1H), 7.50–7.46 (m,
2H), 7.39–7.35 (m, 3H), 7.20 (dd, J=8.0, 2.0 Hz, 1H), 7.11 (d, J=8.0 Hz,
1H), 5.52 (s, 1H), 5.14 (dd, J=10.2, 8.3 Hz, 1H), 4.79 (d, J=10.2 Hz,
1H), 4.32 (dd, J=10.5, 4.9 Hz, 1H), 3.99 (t, J=8.6 Hz, 1H), 3.82 (t, J=
10.2 Hz, 1H), 3.63 (t, J=9.2 Hz, 1H), 3.52 (td, J=9.8, 5.0 Hz, 1H), 2.33
(s, 3H), 1.31 (s, 9H), 1.28 (s, 9H), 0.80 (s, 9H), 0.01 (s, 3H), ꢀ0.05 ppm
(s, 3H); 13C NMR (101 MHz, CDCl3): d=176.9, 149.7, 137.1, 135.9,
133.1, 130.0, 129.3, 128.3, 128.0, 126.5, 124.7, 102.3, 87.1, 81.7, 74.4, 73.1,
70.3, 68.8, 39.1, 34.7, 31.5, 27.6, 26.0, 20.3, 18.3, ꢀ3.6, ꢀ4.6 ppm; IR (thin
film): n˜ = 2961, 2929, 2860, 1741, 1480, 1385, 1249, 1133, 1100, 859, 839,
759 cmꢀ1; HRMS: m/z: calcd for C35H52O6SSiNa [M+Na]+: 651.3146;
found: 651.3168.
[M+Na]+: 759.2962; found: 759.2973.
(2-Methyl-5-tert-butylphenyl)-4-O-benzyl-3-O-tert-butyldimethylsilyl-2-
O-pivaloyl-1-thio-b-d-glucopyranoside (9): Thioglycoside
6
(2.0 g,
3.18 mmol) was coevaporated with toluene three times and subsequently
dissolved in CH2Cl2 (32 mL) under argon. BH3·THF (19.1 mL,
85.94 mmol) was added, and the solution was cooled to 08C. TMSOTf
(290 mL, 1.59 mmol) was added dropwise, and the solution was warmed
to room temperature and stirred overnight. After completion (TLC: cy-
clohexane/ethyl acetate 9:1), the reaction was cooled to 08C, quenched
by the dropwise addition of methanol and neutralized by the dropwise
addition of triethylamine. The volatiles were removed in vacuo and the
residue was dissolved in CH2Cl2 and washed with water. The organic
layer was dried over MgSO4 and filtered. The solvent was evaporated in
vacuo and the crude material was purified by flash column chromatogra-
phy (silica gel, cyclohexane/ethyl acetate) to afford 9 (1.94 g, 97%) as a
white foam. Rf =0.30 (cyclohexane/ethyl acetate 9:1); [a]2D0 = ꢀ13.40 (c=
1.0, CHCl3); 1H NMR (400 MHz, CDCl3): d=7.48 (d, J=2.1 Hz, 1H),
7.37–7.26 (m, 5H), 7.19 (dd, J=7.9, 2.0 Hz, 1H), 7.11 (d, J=8.1 Hz, 1H),
5.09 (dd, J=10.0, 8.4 Hz, 1H), 4.84 (d, J=11.7 Hz, 1H), 4.74 (d, J=
10.0 Hz, 1H), 4.65 (d, J=11.7 Hz, 1H), 3.92 (t, J=8.4 Hz, 1H), 3.83 (dd,
J=12.0, 1.9 Hz, 1H), 3.67 (dd, J=12.0, 4.1 Hz, 1H), 3.59 (dd, J=9.5,
8.6 Hz, 1H), 3.44 (ddd, J=9.5, 4.5, 2.6 Hz, 1H), 2.33 (s, 3H), 1.30 (s,
9H), 1.29 (s, 9H), 0.89 (s, 9H), 0.08 (s, 3H), 0.06 ppm (s, 3H); 13C NMR
(101 MHz, CDCl3): d=177.4, 149.8, 138.1, 135.7, 133.1, 130.1, 128.5,
127.8, 127.5, 127. 5, 124.6, 86.1, 79.4, 78.8, 76.3, 74.8, 73.3, 62.2, 39.3, 34.7,
31.5, 27.8, 26.0, 20.3, 18.1, ꢀ3.5, ꢀ4.0 ppm; IR (thin film): n˜ = 3493, 2959,
2929, 2856, 1738, 1480, 1395, 1260, 1157, 1127, 1075, 1037, 838, 779, 733,
697 cmꢀ1; HRMS: m/z: calcd for C35H54O6SSiNa [M+Na]+: 653.3303;
found: 653.3325.
(2-Methyl-5-tert-butylphenyl)-4,6-O-benzylidene-2-O-pivaloyl-1-thio-b-d-
glucopyranoside (7): Thioglycoside 6 (15.6 g, 24.8 mmol) was dissolved in
DMF (125 mL), and the solution was cooled to 08C. A mixture of
TBAF·3H2O (31.4 g, 100.0 mmol) and glacial acetic acid (8.5 mL,
149.0 mmol) in DMF (125 mL) was added dropwise, and the solution was
warmed to 358C and stirred overnight. Upon completion (TLC: cyclo-
hexane/ethyl acetate 7:1), the reaction was cooled to room temperature,
diluted with Et2O, and washed with 1m HCl (aq) and sat. NaHCO3 (aq).
The organic layer was dried over MgSO4 and filtered. The solvent was
evaporated in vacuo and the crude material was purified by flash column
chromatography (silica gel, cyclohexane/ethyl acetate) to afford 7 (12.4 g,
(2-Methyl-5-tert-butylphenyl)-4,6-di-O-benzyl-3-O-tert-butyldimethylsil-
yl-2-O-pivaloyl-1-thio-b-d-glucopyranoside (10): Thioglycoside 9 (1.9 g,
3.01 mmol) was dissolved in DMF (14 mL) under argon. Sodium hydride
(130 mg, 5.42 mmol) was added, and the suspension was cooled to 08C.
Benzyl bromide (1.1 mL, 9.04 mmol) was added dropwise. The solution
was warmed to room temperature and stirred for 3 h. Upon completion
(TLC: cyclohexane/ethyl acetate 3:1), the reaction was cooled to 08C
and quenched by the dropwise addition of methanol, diluted with ether
and washed with water, NH4Cl (aq) and water. The organic layer was
dried over MgSO4 and filtered. The solvent was evaporated in vacuo, and
the crude material was purified by flash column chromatography (silica
gel, cyclohexane/ethyl acetate) to afford 10 (1.70 g, 78%) as a white
foam. Rf =0.67 (cyclohexane/ethyl acetate 3:1); [a]2D0 = ꢀ7.78 (c=1.0,
CHCl3); 1H NMR (400 MHz, CDCl3): d=7.58 (d, J=2.0 Hz, 1H), 7.38–
7.19 (m, 10H), 7.17 (dd, J=7.9, 2.0 Hz, 1H), 7.09 (d, J=8.1 Hz, 1H),
5.11 (dd, J=10.0, 8.3 Hz, 1H), 4.79 (d, J=11.6 Hz, 1H), 4.70 (d, J=
10.0 Hz, 1H), 4.63 (d, J=11.6 Hz, 1H), 4.55 (d, J=12.3 Hz, 1H), 4.50 (d,
J=12.2 Hz, 1H), 3.89 (t, J=8.3 Hz, 1H), 3.72–3.68 (m, 2H), 3.63 (dd, J=
9.4, 8.4 Hz, 1H), 3.54 (dt, J=9.5, 3.2 Hz, 1H), 2.34 (s, 3H), 1.30–1.27 (m,
18H), 0.89 (s, 9H), 0.07 (s, 3H), 0.03 ppm (s, 3H); 13C NMR (101 MHz,
CDCl3): d=177.4, 149.7, 138.3, 138.2, 135.7, 133.8, 129.8, 128.5, 128.4,
128.0, 127.7, 127.7, 127.6, 127.3, 124.2, 86.6, 79.2, 79.0, 76.4, 74.6, 73.6,
73.3, 69.1, 39.2, 34.7, 31.5, 27.8, 26.1, 20.4, 18.1, ꢀ3.5, ꢀ4.0 ppm; IR (thin
film): n˜ = 2958, 2929, 2858, 1737, 1455, 1362, 1259, 1128, 1096, 882, 838,
778, 734, 697 cmꢀ1; HRMS: m/z: calcd for C42H60O6SSiNa [M+Na]+:
743.3772; found: 743.3789.
97%) as a white foam. Rf =0.05 (cyclohexane/ethyl acetate 7:1); [a]D20
=
ꢀ30.66 (c=1.0, CHCl3). 1H NMR (400 MHz, CDCl3): d=7.59 (d, J=
1.7 Hz, 1H), 7.48 (dd, J=6.4, 2.7 Hz, 2H), 7.37 (dd, J=4.9, 1.5 Hz, 3H),
7.24 (dd, J=8.1, 1.8 Hz, 1H), 7.14 (d, J=8.0 Hz, 1H), 5.57 (s, 1H), 5.08–
4.97 (m, 1H), 4.78 (d, J=10.1 Hz, 1H), 4.35 (dd, J=10.5, 5.0 Hz, 1H),
3.95 (t, J=9.0 Hz, 1H), 3.84 (t, J=10.3 Hz, 1H), 3.65 (t, J=9.3 Hz, 1H),
3.50 (td, J=9.7, 5.0 Hz, 1H), 2.36 (s, 3H), 1.34–1.27 ppm (m, 18H);
13C NMR (101 MHz, CDCl3): d=177.9, 149.7, 137.0, 136.8, 132.2, 130.1,
129.7, 129.5, 128.5, 126.4, 125.3, 102.0, 87.1, 80.8, 74.1, 72.6, 70.3, 68.7,
39.1, 34.6, 31.5, 27.3, 20.5 ppm; IR (thin film): n˜ = 3473, 2964, 2868, 1738,
1150, 1098, 996, 760 cmꢀ1; HRMS: m/z: calcd for C29H46O6SNa [M+Na]+:
537.2281; found: 537.2260.
(2-Methyl-5-tert-butylphenyl)-4,6-O-benzylidene-3-O-fluorenylmethoxy-
carbonyl-2-O-pivaloyl-1-thio-b-d-glucopyranoside (8): Thioglycoside
7
(2.4 g, 4.74 mmol) and FmocCl (1.8 g, 7.10 mmol) were dissolved in
CH2Cl2 (26 mL) under argon, and pyridine (770 mL, 9.48 mmol) was
added dropwise. The reaction was stirred overnight. Upon completion
(TLC: cyclohexane/ethyl acetate 3:1), the reaction mixture was diluted
with CH2Cl2 and washed with 1m HCl (aq) and sat. NaHCO3 (aq). The
organic layer was dried over MgSO4 and filtered. The solvent was evapo-
rated in vacuo and the crude material was purified by flash column chro-
matography (silica gel, cyclohexane/ethyl acetate) to afford 8 (3.4 g,
(2-Methyl-5-tert-butylphenyl)-4,6-di-O-benzyl-2-O-pivaloyl-1-thio-b-d-
glucopyranoside (11): Thioglycoside 10 (1.7 g, 2.36 mmol) was dissolved
in DMF (12 mL), and the solution was cooled to 08C. A mixture of
TBAF·3H2O (4.46 g, 14.15 mmol) and glacial acetic acid (1.1 mL,
18.86 mmol) in DMF (12 mL) was added dropwise, and the solution was
warmed to 358C and stirred overnight. Upon completion (TLC: cyclo-
hexane/ethyl acetate 3:1), the reaction was cooled to room temperature,
diluted with Et2O, and washed with 1m HCl (aq) and sat. NaHCO3 (aq).
97%) as a white foam. Rf =0.63 (cyclohexane/ethyl acetate 3:1); [a]D20
=
1
ꢀ2.53 (c=1.0, CHCl3); H NMR (400 MHz, CDCl3): d=7.75 (dd, J=7.6,
0.6 Hz, 2H), 7.61–7.58 (m, 2H), 7.55 (dd, J=7.6, 0.7 Hz, 1H), 7.45 (dd,
J=6.8, 3.0 Hz, 2H), 7.42–7.28 (m, 6H), 7.28–7.18 (m, 2H), 7.15 (d, J=
8.0 Hz, 1H), 5.57 (s, 1H), 5.32–5.19 (m, 2H), 4.85 (d, J=9.8 Hz, 1H),
4.44–4.33 (m, 2H), 4.31–4.18 (m, 2H), 3.93–3.82 (m, 2H), 3.61 (td, J=
9.8, 5.0 Hz, 1H), 2.37 (s, 3H), 1.32 (s, 9H), 1.19 ppm (s, 9H); 13C NMR
Chem. Eur. J. 2013, 19, 12497 – 12503
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12501

P. H. Seeberger et al.
The organic layer was dried over MgSO4 and filtered. The solvent was
evaporated in vacuo and the crude material was purified by flash column
chromatography (silica gel, cyclohexane/ethyl acetate) to afford 11
(1.22 g, 85%) as a white foam. Rf =0.52 (cyclohexane/ethyl acetate 3:1);
Solution B: Activator (3ꢀ3 equiv per glycosylation cycle): TMSOTf
(19.2 mL, 106 mmol).
Solution C: Fmoc deprotection: piperidine (20% v/v) in DMF.
Module A: Swelling of resin prior to synthesis: The reaction vessel was
charged with functionalized resin, and CH2Cl2 (2 mL) was added. The
resin was swelled in CH2Cl2 for 2 h. At the beginning of the synthesis,
the reaction vessel was drained by argon pressure.
1
[a]2D0 = ꢀ30.64 (c=1.0, CHCl3); H NMR (400 MHz, CDCl3): d=7.61 (d,
J=2.0 Hz, 1H), 7.35–7.21 (m, 10H), 7.19 (dd, J=8.0, 2.0 Hz, 1H), 7.10
(d, J=8.0 Hz, 1H), 4.92 (dd, J=10.0, 9.1 Hz, 1H), 4.80 (d, J=11.1 Hz,
1H), 4.69–4.51 (m, 4H), 3.79 (t, J=8.9 Hz, 1H), 3.75–3.72 (m, 2H), 3.63
(t, J=9.2 Hz, 1H), 3.49 (ddd, J=9.7, 3.8, 2.5 Hz, 1H), 2.35 (s, 3H), 1.28–
1.26 ppm (m, 18H); 13C NMR (101 MHz, CDCl3): d=178.4, 149.7, 138.1,
138.1, 136.4, 133.1, 129.9, 129.1, 128.6, 128.5, 128.2, 128.1, 127.8, 124.8,
86.4, 79.0, 78.2, 77.8, 75.0, 73.7, 72.7, 69.0, 39.1, 34.6, 31.4, 27.3, 20.4 ppm;
IR (thin film): n˜ = 3489, 2962, 2868, 1737, 1480, 1454, 1363, 1280, 1155,
1092, 1045, 822, 736, 698 cmꢀ1; HRMS: m/z: calcd for C36H46O6SNa
[M+Na]+: 629.2907; found: 629.2956.
Module B: Washing before glycosylation: The resin was washed with
THF (6ꢄ) and CH2Cl2 (6ꢄ) at room temperature.
Module C: Glycosylation: Solution A (1 mL) was added to the reaction
vessel, and the suspension was cooled to ꢀ158C while the resin was
being agitated by an argon flow. Upon reaching the set temperature, sol-
ution B (1 mL) was added to the reaction vessel. The temperature was
maintained for 45 min and then raised to 08C for 15 min. The reaction
vessel was then drained by argon pressure and the resin was washed with
CH2Cl2 (6ꢄ).
(2-Methyl-5-tert-butylphenyl)-4,6-di-O-benzyl-3-O-fluorenylmethoxycar-
bonyl-2-O-pivaloyl-1-thio-b-d-glucopyranoside (12): Thioglycoside 11
(1.20 g, 1.98 mmol) and FmocCl (770 mg, 2.97 mmol) were dissolved in
CH2Cl2 (11 mL) under argon, and pyridine (320 mL, 3.96 mmol) was
added dropwise. The reaction was stirred overnight. Upon completion
(TLC: cyclohexane/ethyl acetate 3:1), the reaction mixture was diluted
with CH2Cl2 and washed with 1m HCl (aq) and sat. NaHCO3 (aq). The
organic layer was dried over MgSO4 and filtered. The solvent was evapo-
rated in vacuo and the crude material was purified by flash column chro-
matography (silica gel, cyclohexane/ethyl acetate) to afford 12 (1.5 g,
Module D: Washing after glycosylation cycle: The temperature was raised
to 258C, and the resin was washed with THF (6ꢄ) and CH2Cl2 (6ꢄ).
Module E: Fmoc deprotection: The resin was washed with DMF (3ꢄ).
Solution C (2 mL) was added to the reaction vessel, and the suspension
was agitated by an argon flow for 15 min. The reaction vessel was then
drained and the solvents were transferred to a fraction collector via
argon pressure.
Module F: Washing after Fmoc deprotection: The temperature was set to
258C, and the resin was washed with DMF (6ꢄ), THF (6ꢄ), AcOH in
CH2Cl2 (6ꢄ), and CH2Cl2 (6ꢄ).
91%) as a white foam. Rf =0.66 (cyclohexane/ethyl acetate 3:1); [a]D20
=
+4.68 (c=1.0, CHCl3); 1H NMR (400 MHz, CDCl3): d=7.73 (d, J=
7.3 Hz, 2H), 7.65–7.61 (m, 1H), 7.58 (d, J=7.5 Hz, 2H), 7.41–7.17 (m,
13H), 7.16–7.08 (m, 3H), 5.21–5.14 (m, 2H), 4.76–4.68 (m, 1H), 4.66–
4.50 (m, 4H), 4.29 (d, J=7.5 Hz, 2H), 4.20–4.14 (m, 1H), 3.93–3.84 (m,
1H), 3.76–3.70 (m, 2H), 3.59–3.52 (m, 1H), 2.36 (s, 3H), 1.28 (s, 9H),
1.18 ppm (s, 9H); 13C NMR (101 MHz,CDCl3): d=176.8, 154.7, 149.7,
143.3, 143.3, 141.3, 141.3, 138.0, 137.7, 136.6, 132.8, 129.9, 129.5, 128.5,
128.4, 128.0, 127.9, 127.8, 127.4, 127.3, 125.3, 125.0, 120.1, 120.1, 86.9,
80.8, 79.1, 75.9, 74.9, 73.7, 70.4, 70.0, 68.6, 46.7, 38.9, 34.6, 31.4, 27.1,
20.4 ppm; IR (thin film): n˜ = 2961, 2909, 2868, 1752, 1452, 1255, 1133,
1093, 968, 738, 698 cmꢀ1; HRMS: m/z: calcd for C51H56O8SNa [M+Na]+:
851.3588; found: 851.3616.
Dodecasaccharide 17 (Scheme 4): The reaction vessel of the synthesizer
was charged with functionalized Merrifield resin 16 (25 mg, loading:
0.47 mmolgꢀ1). Program module A was executed once. Then twelve
cycles of the program described in Table 1 were executed by the synthe-
sizer. The resin was removed from the reaction vessel and swelled in
CH2Cl2. The suspension was irradiated with UV light by delivering the
suspension via syringe pump through a FEP tubing (inner diameter:
0.03 inch; volume: 12 mL) wrapped around a UV light source (medium
pressure Hg lamp with arc lengths of 27.9 cm and power of 450 W, sur-
rounded by a Pyrex UV filter with 50% transmittance at 305 nm). The
resin was slowly injected from a disposable syringe (2 mL) and pushed
through the tubing with 15 mL CH2Cl2 (flow rate: 300 mLminꢀ1). The
tubing was washed with 15 mL CH2Cl2/MeOH (1:1 v/v, flow rate:
300 mLminꢀ1 for 8 mL and 4 mLminꢀ1 for 7 mL), and finally with 15 mL
MeOH (flow rate: 4 mLminꢀ1). The suspension leaving the reactor was
directed into a filter where the resin was filtered off and washed with
CH2Cl2/MeOH (1:1 v/v), MeOH and CH2Cl2, while the filtrate was col-
lected. The tubing was re-equilibrated with 15 mL CH2Cl2using a flow
rate of 4 mLminꢀ1. The entire procedure was repeated three times.[18b] (A
detailed description of the experimental procedure and setup of the pho-
tocleavage can be found on page S20 of the Supporting Information of
ref. [18b], as well as in the Supporting Information of this manuscript.
This protocol was followed exactly.) The resulting solution was concen-
trated in vacuo. The crude product was purified by HPLC (YMC-Pack
Diol-300, hexane/ethyl acetate, 10% ethyl acetate for 5 min, to 40%
ethyl acetate in 70 min) to obtain compound 17 as a colorless oil (2.9 mg,
4.6%). 1H NMR (600 MHz, CDCl3): d=7.35–7.13 (m, 125H), 5.07 (s,
2H), 4.98–4.70 (m, 25H), 4.64–4.55 (m, 10H), 4.55–4.37 (m, 36H), 4.22–
4.08 (m, 12H), 3.78–3.26 (m, 53H), 3.17–3.07 (m, 2H), 1.51–1.40 (m,
6H), 1.30–1.00 ppm (m, 108H); 13C NMR (176 MHz, CDCl3): d=179.3,
176.7, 176.7, 176.5, 176.2, 156.5, 138.7, 138.6, 138.5, 138.4, 138.3, 138.1,
136.8, 128.7, 128.6, 128.6, 128.5, 128.5, 128.4, 128.3, 128.3, 128.2, 127.9,
127.7, 127.6, 127.6, 101.0, 99.2, 99.0, 78.9, 77.9, 76.5, 76.0, 75.8, 75.5, 75.2,
75.1, 74.9, 74.6, 74.5, 74.4, 74.1, 73.6, 73.5, 73.4, 69.5, 69.3, 69.3, 68.9, 68.8,
66.7, 41.0, 39.1, 38.8, 38.7, 32.1, 29.9, 29.3, 27.4, 27.4, 27.3, 27.2, 23.4,
22.9 ppm; HSQC (700 MHz, CDCl3): d=101.0 (JC1,H1 =161 Hz, C-1), 99.2
(JC1,H1 =161 Hz, C-1), 99.0 ppm (JC1,H1 =161 Hz, C-1); MALDI-TOF MS:
m/z: calcd for C313H379NO75Na [M+Na]+: 5377.587; found: 5377.612.
Dibutyl 4,6-di-O-benzyl-3-O-fluorenylmethoxycarbonyl-2-O-pivaloyl-b-d-
glucopyranoside phosphate (2): Thioglycoside 12 (96 mg, 0.116 mmol), di-
butyl phosphate (69 mL, 0.347 mmol) and freshly activated 4 ꢃ molecular
sieves were suspended in CH2Cl2 (1.3 mL) under argon. The suspension
was stirred at room temperature for 20 min, and then cooled to 08C. NIS
(34 mg, 0.151 mmol) was added, and the suspension was stirred for anoth-
er 20 min at 08C. TfOH (1 mL, 12 mmol) was added, and the reaction was
stirred at 08C for 1 h. Upon completion (TLC: cyclohexane/ethyl acetate
2:1), the reaction was diluted with CH2Cl2, filtered over celite and
washed with sat. NaHCO3 (aq) and sat. NaS2O3 (aq). The organic layer
was dried over MgSO4 and filtered. The solvent was evaporated in vacuo
and the crude material was purified by flash column chromatography
(silica gel, cyclohexane/ethyl acetate) to afford 2 (71 mg, 71%) as a color-
less oil. Rf =0.35 (cyclohexane/ethyl acetate 2:1); [a]2D0 = +22.09 (c=1.0,
CHCl3); 1H NMR (400 MHz, CDCl3): d=7.79–7.73 (m, 1H), 7.60 (dd,
J=17.2, 7.2 Hz, 1H), 7.43–7.21 (m, 3H), 7.16 (dd, J=6.7, 2.8 Hz, 1H),
6.25 (d, J=3.8 Hz, 1H), 5.15–5.09 (m, 1H), 4.74–4.45 (m, 1H), 4.39 (dd,
J=10.4, 7.5 Hz, 1H), 4.33–4.23 (m, 1H), 4.04 (q, J=6.6 Hz, 1H), 3.97–
3.90 (m, 1H), 3.87–3.73 (m, 1H), 3.67 (dd, J=10.9, 1.5 Hz, 1H), 1.72–
1.63 (m, 1H), 1.42 (dd, J=15.0, 7.5 Hz, 1H), 1.26 (s, 1H), 1.13 (s, 1H),
0.94 ppm (t, J=7.4 Hz, 1H); 13C NMR (101 MHz, CDCl3): d=176.8,
156.1, 143.4, 143.2, 141.4, 137.9, 137.7, 128.6, 128.6, 128.2, 128.1, 128.0,
127.4, 125.3, 125.2, 120.2, 91.8, 80.6, 75.4, 74.8, 73.8, 73.0, 70.8, 70.5, 67.9,
67.6, 67.5, 46.8, 32.4, 32.3, 27.3, 18.8, 13.7 ppm; 31P NMR (243 MHz,
CDCl3): d=ꢀ2.9 ppm; IR (thin film): n˜ = 2961, 2873, 1752, 1452, 1256,
1028, 739 cmꢀ1
;
HRMS: m/z: calcd for C48H59O12PNa [M+Na]+:
881.3636; found: 881.3629.
Automated solid-phase synthesis
Dodecasaccharide 18 (Scheme 4): Compound 17 (2.2 mg, 0.41 mmol) was
dissolved in CH2Cl2/MeOH (1:1, 0.5 mL). Sodium methoxide (25 mg,
0.46 mmol) was added, and the solution was stirred for two weeks. The
Solution A: Building block (3ꢀ3 equiv per glycosylation cycle): Com-
pound 2 (91 mg, 106 mmol) in CH2Cl2 (3 mL).
12502
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 12497 – 12503

Automated Carbohydrate Synthesis
FULL PAPER
Table 1. Program for one glycosylation/deprotection cycle using modules
B, C and D–F, executed 12 times to obtain dodecasaccharide 1.
doucheva, Pharmazie 2005, 60, 42–48; c) A. C. Pelizon, R. Kaneno,
A. Soares, D. A. Meira, A. Sartori, Physiol. Res. 2005, 54, 557–564.
[5] R. Adamo, M. Tontini, G. Brogioni, M. R. Romano, G. Costantini,
E. Danieli, D. Proietti, F. Berti, P. Costantino, J. Carbohydr. Chem.
2011, 30, 249–280.
[6] a) G. D. Brown, S. Gordon, Nature 2001, 413, 36–37; b) E. L.
Adams, P. J. Rice, B. Graves, H. E. Ensley, H. Yu, G. D. Brown, S.
Gordon, M. A. Monteiro, E. Papp-Szabo, D. W. Lowman, T. D.
Power, M. F. Wempe, D. L. Williams, J. Pharmacol. Exp. Ther. 2008,
325, 115–123.
Program Module
Description
Iterations
B
C
D
E
F
wash
glycosylation
wash
Fmoc deprotection
wash
1
3
1
3
1
[7] a) B. H. Falch, T. Espevik, L. Ryan, B. T. Stokke, Carbohydr. Res.
2000, 329, 587–596; b) K. Kataoka, T. Muta, S. Yamazaki, K. Take-
shige, J. Biol. Chem. 2002, 277, 36825–36831; c) A. Mueller, J.
Raptis, P. J. Rice, J. H. Kalbfleisch, R. D. Stout, H. E. Ensley, W.
Browder, D. L. Williams, Glycobiology 2000, 10, 339–346.
[8] F. Jamois, V. Ferriꢅres, J.-P. Guꢆgan, J.-C. Yvin, D. Plusquellec, V.
Vetvicka, Glycobiology 2004, 15, 393–407.
[9] a) H. Yu, D. L. Williams, H. E. Ensley, Tetrahedron Lett. 2005, 46,
3417–3421; b) Y. Zeng, J. Ning, F. Kong, Tetrahedron Lett. 2002, 43,
3729–3733; c) Y. Zeng, J. Ning, F. Z. Kong, Carbohydr. Res. 2003,
338, 307–311.
[10] K. F. Mo, H. Q. Li, J. T. Mague, H. E. Ensley, Carbohydr. Res. 2009,
344, 439–447.
[11] H. Tanaka, T. Kawai, Y. Adachi, N. Ohno, T. Takahashi, Chem.
Commun. 2010, 46, 8249–8251.
[12] M. Collot, J. Savreux, J. M. Mallet, Tetrahedron 2008, 64, 1523–
1535.
[13] F. R. Carrel, P. H. Seeberger, J. Org. Chem. 2008, 73, 2058–2065.
[14] R. B. Merrifield, J. Am. Chem. Soc. 1963, 85, 2149–2154.
[15] a) O. J. Plante, E. R. Palmacci, P. H. Seeberger, Science 2001, 291,
1523–1527; b) D. B. Werz, B. Castagner, P. H. Seeberger, J. Am.
Chem. Soc. 2007, 129, 2770–2771; c) J. D. C. Codꢆe, L. Krçck, B.
Castagner, P. H. Seeberger, Chem. Eur. J. 2008, 14, 3987–3994;
d) X. Liu, R. Wada, S. Boonyarattanakalin, B. Castagner, P. H. See-
berger, Chem. Commun. 2008, 3510–3512; e) L. Krçck, D. Esposito,
B. Castagner, C.-C. Wang, P. Bindschꢁdler, P. H. Seeberger, Chem.
Sci. 2012, 3, 1617–1622; f) M. T. C. Walvoort, H. van den Elst, O. J.
Plante, L. Krçck, P. H. Seeberger, H. S. Overkleeft, G. A. van der
Marel, J. D. C. Codꢆe, Angew. Chem. 2012, 124, 4469–4472; Angew.
Chem. Int. Ed. 2012, 51, 4393–4396.
reaction mixture was neutralized with Amberlite IR-120 (H+), and fil-
tered. The volatiles of the filtrate were removed in vacuo to afford com-
pound 18 (1 mg, 56%) as a white solid. H NMR (600 MHz, CDCl3): d=
7.33–7.13 (m, 125H), 5.21–5.01 (m, 14H), 4.83–4.13 (m, 48H), 3.93–3.26
(m, 74H), 3.16–3.06 (m, 2H), 1.68–1.60 (m, 4H), 1.50–1.41 ppm (m, 2H);
MALDI-TOF MS: m/z: calcd for C253H283NO63Na [M+H]+: 4367.893;
found: 4369.862.
1
Dodecasaccharide 1 (Scheme 4): Compound 18 (200 mg, 0.046 mmol) was
dissolved in THF/H2O (300 mL), and Pd(OH)2 (0.5 mg, 20 wt.%) was
added. The atmosphere was exchanged with H2, and the reaction was stir-
red at RT for five days. The reaction mixture was filtered and concentrat-
ed in vacuo. The residue was purified by reversed-phase column chroma-
tography (Waters Sep-Pak SPEVac C18, acetonitrile/water, 0 to 50%)
and lyophilized to afford 1 (50 mg, 52%) as a white solid. MALDI-TOF
MS: m/z: calcd for C77H131NO61Na [M+Na]+: 2070.723; found: 2070.774.
The correct structure of 1 was confirmed by comparison of the 1H NMR
1
spectrum of 1 with a H NMR spectrum of isolated Curdlan (linear b-1,3-
glucan from Agrobacterium biobar).[19]
Acknowledgements
We thank the Max Planck Society and the BMBF (Das Taschentuchla-
bor: Impulszentrum fꢀr integrierte Bioanalyse (IZIB), Fçrderkennzei-
chen 03IS2201G) for financial support. We also thank Dr. Steffen Eller
for assistance with the automated oligosaccharide synthesizer, Dr. An-
dreas Schꢁfer for NMR measurements and Ms. Janine Stuwe for
MALDI-TOF measurements. Furthermore, we would like to thank Mr.
Sebastian Gçtze, Dr. Mattan Hurevich and Dr. Ivan Vilotijevic for criti-
cally editing this manuscript.
[16] a) J. S. S. Rountree, P. V. Murphy, Org. Lett. 2009, 11, 871–874;
b) K. C. Nicolaou, N. Winssinger, J. Pastor, F. DeRoose, J. Am.
Chem. Soc. 1997, 119, 449–450.
[17] B. Fraser-Reid, J. C. Lopez, Reactivity Tuning in Oligosaccharide As-
sembly, Vol. 301 (Eds.: B. Fraser-Reid, J. C. Lopez), Springer,
Berlin, 2011, p. 1–29.
[1] a) B. A. Stone, A. E. Clarke, Chemistry and biology of (1!3)-b-glu-
cans, La Trobe University Press, 1992; b) A. E. Clarke, B. A. Stone,
[18] a) E. R. Palmacci, O. J. Plante, P. H. Seeberger, Eur. J. Org. Chem.
2002, 595–606; b) S. Eller, M. Collot, J. Yin, H.-S. Hahm, P. H. See-
berger, Angew. Chem. 2013, 125, 5970–5973; Angew. Chem. Int. Ed.
2013, 52, 5858–5861.
[19] C. Bromuro, M. Romano, P. Chiani, F. Berti, M. Tontini, D. Proietti,
E. Mori, A. Torosantucci, P. Costantino, R. Rappuoli, A. Cassone,
Vaccine 2010, 28, 2615–2623.
´
Biochim. Biophys. Acta 1960, 44, 161–163; c) D. B. Zekovic, S.
´
´
Kwiatkowski, M. M. Vrvic, D. Jakovljevic, C. A. Moran, Crit. Rev.
Biotechnol. 2005, 25, 205–230.
[2] P. Kougias, D. Wei, P. J. Rice, H. E. Ensley, J. Kalbfleisch, D. L. Wil-
liams, I. W. Browder, Infect. Immun. 2001, 69, 3933–3938.
[3] a) G. Chihara, Eos-Riv. Immunol.1984, 4, 85–96; b) Y. Y. Maeda, G.
Chihara, Nature 1971, 229, 634; c) Y. Y. Maeda, S. T. Watanabe, C.
Chihara, M. Rokutanda, Cancer Res. 1988, 48, 671–675.
[4] a) N. Markova, V. Kussovski, L. Drandarska, S. Nikolaeva, N. Geor-
gieva, T. Radoucheva, Int. Immunopharmacol. 2003, 3, 1557–1562;
b) N. Markova, L. Michailova, V. Kussovski, M. Jourdanova, T. Ra-
Received: December 19, 2012
Revised: May 27, 2013
Published online: August 16, 2013
Chem. Eur. J. 2013, 19, 12497 – 12503
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12503