Preparation of core 2 type tetrasaccharide carrying decapeptide by benzyl protection-based solid-phase synthesis strategy

Article abstract of DOI:10.1016/S0040-4039(02)01947-0

β-D-Gal-(1→4)-β-D-GlcNAc-[β-D-Gal-(1→3)] -α-D-GalNAc-(1→3)-L-Ser/Thr building blocks for solid-phase synthesis of glycopeptide were stereoselectively synthesized in a benzyl-protected form. The key glycosylation reaction to form β-D-GlcNAc linkage was established by the use of protected N-trichloroacetyl-D-lactosaminyl fluoride. Usefulness of the building block was demonstrated by solid-phase synthesis of a segment of human leukosialin. Benzyl protecting group was efficiently removed by 'low acidity TfOH' conditions.

Full text of DOI:10.1016/S0040-4039(02)01947-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 8395–8399
Preparation of core 2 type tetrasaccharide carrying decapeptide
by benzyl protection-based solid-phase synthesis strategy
Yutaka Takano, Motoki Habiro, Masaomi Someya, Hironobu Hojo and Yoshiaki Nakahara*
Institute of Glycotechnology, Department of Applied Biochemistry, Tokai University, Kitakaname 1117, Hiratsuka,
Kanagawa 259-1292, Japan
Received 5 August 2002; revised 4 September 2002; accepted 6 September 2002
Abstract—b-
of glycopeptide were stereoselectively synthesized in a benzyl-protected form. The key glycosylation reaction to form b-
linkage was established by the use of protected N-trichloroacetyl- -lactosaminyl fluoride. Usefulness of the building block was
D
-Gal-(1“4)-b-
D
-GlcNAc-[b-
D
-Gal-(1“3)]-a-
D
-GalNAc-(1“3)-
L
-Ser/Thr building blocks for solid-phase synthesis
D
-GlcNAc
D
demonstrated by solid-phase synthesis of a segment of human leukosialin. Benzyl protecting group was efficiently removed by ‘low
acidity TfOH’ conditions. © 2002 Elsevier Science Ltd. All rights reserved.
It has been recognized that N- and O-linked glycans
present in secreted and cell surface glycoproteins con-
tribute to a variety of properties of the glycoprotein
such as stability against proteolysis, immunogenicity,
signals for cell adhesion, and conformation.1 However,
detailed functions of those glycans have remained to
elucidate because of micro-hetrogeneity in oligosaccha-
ride structures and low availability of the homogeneous
samples from natural sources. On the other hand,
recent advancement in solid-phase technology have
made possible rapid synthesis of the glycopeptides bear-
ing structurally defined oligosaccharides. As part of
projects on the synthesis of complex glycopeptides
directed toward structure–activity relationship studies,
the benzyl protection-based solid-phase synthesis of
glycopeptides have been extensively investigated within
this group.2 We report here an efficient synthesis of
amide group and the conditions being compatible with
the presence of base-labile 9-fluorenylmethoxycarbonyl,
allyl ester, and NeuAc-Gal-(1“4)-lactone moieties.
However, the crucial coupling reaction involving
trichloroacetimidate protocol afforded a hardly separa-
ble mixture of stereoisomeric hexasaccharides (a/b=1/
3), and thus obtained core 2 hexasaccharyl threonine
was insufficient in amount to test the feasibility as
building block for solid-phase synthesis. We therefore
sought to develop an alternative 2-N-masking group to
preclude the cumbersomeness arising from the
stereoisomers.
Although N-phthaloyl group has been widely used in
order to synthesize 1,2-trans glycoside of
D-glu-
cosamine,5 numerous other N-protecting groups such
as N-tetrachlorophthaloyl,6,7 N-dichlorophthaloyl,8 N-
trichloroethoxycarbonyl,9 N,N-dithiasuccinoyl,10 N-
allyloxycarbonyl,11 N-trichloroacetyl,12 N-pentenoyl,13
core 2 O-linked tetrasaccharide {b-
D
-Gal-(1“4)-b-
D-
-Ser/
GlcNAc-[b- -Gal-(1“3)]-a- -GalNAc-(1“3)-
D
D
L
Thr} building blocks and solid-phase synthesis of a
glycopeptide modeled after the leukosialin (CD 43)
segment (215–224) of activated T-lymphocytes.3
N-dimethylmaleoyl,14
2,5-dimethylpyrrole,15
N,N-
diacetyl,16 N-thioglycoloyl,17 N-[1,3-dimethyl-2,4,6(1H,
3H,5H)-trioxopyrimidin-5-ylidene]methyl,18 have been
also applied to the stereocontrol. Amongst them N-
trichloroacetyl group meets our requirements since it
was reported to be convertible into acetyl group with-
out basic conditions. In contrast, most of other N-pro-
tecting groups essentially require base treatment for
their cleavage before transformation into acetamide.
Recently, we have disclosed the first total synthesis of a
core 2 disialylated glycohexaosyl threonine building
block in a benzyl-protected form. The synthesis was
designed so as to condense two trisaccharide segments
in a convergent manner.4 2-Azido-2-deoxy-
D-glucose
derivative was employed as the precursor of GlcNAc
residue, based on ready conversion of azide to acet-
We attempted a new synthesis of core 2 tetrasaccharide
by using N-trichloroacetyl group.19 Since suitably pro-
tected Gal-b-(1“3)-GalN3-Ser/Thr 3 and 4, the key
* Corresponding author.
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01947-0

8396
Y. Takano et al. / Tetrahedron Letters 43 (2002) 8395–8399
disaccharide components, were readily available by
hydrolysis of the benzylidene group of known 12a and
2,2f respectively, the lactosamine counterpart was syn-
thesized as follows (Scheme 1). AgOTf-mediated glyco-
sylation of t-butyldiphenylsilyl 2-azido-3,6-di-O-benzyl-
two steps). Highly stereoselective generation of the
a-fluoride was evidenced by characteristic signal of the
anomeric proton.21
With Cp2ZrCl2–AgClO4 employed as promoter,22 cou-
pling reaction of glycosyl donor 17 and acceptors 3/4
(1.1 equiv.) in CH2Cl2 at −15°C proceeded smoothly
(<2 h) to give tetrasaccharides 18 and 19 as the sole
product in 72 and 70% yield, respectively. As antici-
pated, both coupling reactions exclusively afforded b-
glycoside, whose anomeric configuration was assignable
2-deoxy-b-
D
-glucopyranoside 64b with 2,3,4,6-tetra-O-
acetyl-a- -galactopyranosyl bromide 5 afforded disac-
D
charide 8 in 80% yield. Deacetylation (NaOMe/MeOH)
was followed by benzylidene acetal formation
(dimethoxytoluene, p-TsOH, CH3CN) to give diol 10,
which was benzylated (NaH, BnBr, DMF) to produce
12 in 79% yield in three steps. Azide reduction with
powdered Zn and AcOH in CH2Cl2 proceeded to ex-
clusively give 2-amino derivative 14. Synthesis of 14
was alternatively performed starting from t-butyl-
diphenylsilyl 3,6-di-O-benzyl-2-deoxy-2-phthalimido-b-
1
from the JCH coupling constant.21 The promoter
23
Sn(OTf)2 was also effective for this coupling reaction.
Then trichloroacetamide group was transformed into
acetamide under reductive conditions. It is to be noted
that tributyltin hydride reduction recommended for this
transformation12 is inapplicable to these particular
compounds 18 and 19 due to the presence of labile allyl
ester group. Dechlorination was effected by treatment
with excess (>50 equiv.) powdered Zn and AcOH in
CH2Cl2 at room temperature though rather slow to
complete (3 d). By this treatment, reduction of azide to
amine was simultaneously promoted. The products
were acetylated with Ac2O in MeOH to afford 20 and
D
-glucopyranoside 720 through glycosylation with 5 (9:
82%), deacetylation (100%), benzylidenation (11: 97%),
benzylation (13: 66%), and dephthaloylation with
ethylenediamine in n-BuOH (98%). Amine 14 was then
treated with trichloroacetyl chloride in pyridine to
afford trichloroacetamide 15 in 87% yield. Cleavage of
the silyl ether with n-Bu4NF and AcOH in THF gave
hemiacetal 16, which was converted into glycosyl
fluoride 17 by treatment with Et2NSF3 in THF (84% in
Scheme 1.

Y. Takano et al. / Tetrahedron Letters 43 (2002) 8395–8399
8397
21 in 91 and 96% yield, respectively. Deallylation with
Pd(PPh3)4 and dimedone in THF gave rise to the
suitably protected building blocks 22 and 23.
a C18 column to evaluate efficiency of the solid-phase
synthesis. Fig. 1 exhibits the chromatogram of the
product, in which peaks 1–6 are peptide derivatives.
These compounds separated by HPLC were character-
ized by MALDI TOF MS. Peak 1 corresponds to the
unreacted octapeptide, whereas peak 2 represents
desired 28. The accompanying small peaks 3 and 4 are
the glycopeptides lacking a Gal residue. Mass spectral
data of peak 5 and 6 are coincident with those of the
glycopeptide lacking lactosamine moiety and the pep-
tide missing glycothreonine unit, respectively.
In order to demonstrate usefulness of the core 2 build-
ing block for solid-phase synthesis, a decapeptide seg-
ment (215–224) 24 of human T-lymphocyte
glycoprotein leukosialin, which carried a single glycan
linked to one of the putatively glycosylated threonine
residues, was chosen as a model. It is noteworthy that
synthesis of the N-terminal heptadecapeptide possess-
ing core 1 trisaccharide [a-
(1“3)-a- -GalNAc-(1“3)-
recently.24
D
-Neu5Ac-(2“3)-b-
D-Gal-
D
L-Thr] was reported
Similarly, fully deprotected glycopeptide 24 was
released from resin 27 through N-deprotection with
20% piperidine followed by treatment with reagent K
Solid-phase synthesis was commenced with commercial
Fmoc Sieber amide resin (0.25 mmol/g). According to
Fmoc protocol, the eight amino acids were manually
condensed using Fmoc amino acid (4 equiv.), DCC,
HOBt, and DIEA in NMP for 1 h (Scheme 2).
Removal of Fmoc group was performed with 20%
piperidine/NMP. Capping procedure with Ac2O was
employed after each coupling reaction. t-Bu (Ser/Thr)
and Tr (Asn) groups were employed for side-chain
protection. Then octapeptide-bound resin 25 was
allowed to react with 23 (2 equiv.) in the presence of the
same coupling reagents overnight. Finally Fmoc proline
was condensed by double coupling procedure to com-
plete the target sequence.
The glycopeptide thus synthesized was cleaved from
resin 27 by treatment with reagent K (aq. TFA, thioan-
isole, 1,2-ethanedithiol, phenol). The crude product pre-
cipitated from ether was debenzylated by using the
conditions of low acidity trifluoromethanesulfonic acid
(10% TfOH, 30% Me2S, 8% m-cresol, 2% 1,2-
ethanedithiol, 50% TFA) at −15 to −5°C.25 The result-
ing mixture was analyzed by reversed-phase HPLC with
Figure 1.
Scheme 2.

8398
Y. Takano et al. / Tetrahedron Letters 43 (2002) 8395–8399
References
1. Varki, A. Glycobiology 1993, 3, 97–130.
2. (a) Nakahara, Y.; Nakahara, Y.; Ogawa, T. Carbohydr.
Res. 1996, 292, 71–78; (b) Guo, Z.-W.; Nakahara, Y.;
Nakahara, Y.; Ogawa, T. Angew. Chem., Int. Ed. Engl.
1997, 36, 1464–1466; (c) Nakahara, Y.; Nakahara, Y.;
Ito, Y.; Ogawa, T. Tetrahedron Lett. 1997, 38, 7211–7214;
(d) Guo, Z.-W.; Nakahara, Y.; Nakahara, Y.; Ogawa, T.
Bioorg. Med. Chem. 1997, 5, 1917–1924; (e) Nakahara,
Y.; Nakahara, Y.; Ito, Y.; Ogawa, T. Carbohydr. Res.
1998, 309, 287–296; (f) Nakamura, K.; Ishii, A.; Ito, Y.;
Nakahara, Y. Tetrahedron 1999, 55, 11253–11266; (g)
Ando, S.; Nakahara, Y.; Ito, Y.; Ogawa, T.; Nakahara,
Y. Carbohydr. Res. 2000, 329, 773–780; (h) Hojo, H.;
Watabe, J.; Nakahara, Y.; Nakahara, Y.; Ito, Y.;
Nabeshima, K.; Toole, B. P. Tetrahedron Lett. 2001, 42,
3001–3004; (i) Ishii, A.; Hojo, H.; Nakahara, Y.; Ito, Y.;
Nakahara, Y. Biosci. Biotechnol. Biochem. 2002, 66, 225–
232.
3. Fukuda, M.; Tsuboi, S. Biochim. Biophys. Acta 1999,
1455, 205–217.
Figure 2.
4. (a) Singh, L.; Nakahara, Y.; Ito, Y.; Nakahara, Y.
Tetrahedron Lett. 1999, 40, 3769–3772; (b) Singh, L.;
Nakahara, Y.; Ito, Y.; Nakahara, Y. Carbohydr. Res.
2000, 325, 132–142.
5. Lemiux, R. U.; Takeda, M. T.; Chung, B. Y. In Synthetic
Methods for Carbohydrates; El Khadem, H. S., Ed.; ACS
Symposium Series 39; American Chemical Society: Wash-
ington, DC, 1976; pp. 90–115.
6. Debenham, J. S.; Madsen, R.; Roberts, C.; Fraser-Reid,
B. J. Am. Chem. Soc. 1995, 117, 3302–3303.
7. Castro-Palomino, J. C.; Schmidt, R. R. Tetrahedron Lett.
1995, 36, 5343–5346.
8. Lergenmu¨ller, M.; Ito, Y.; Ogawa, T. Tetrahedron 1998,
54, 1381–1394.
(82.5% TFA, 5% thioanisole, 5% phenol, 5% H2O, 2.5%
1,2-ethanedithiol). The resulting partially protected gly-
copeptide was submitted to debenzylation with the low
acidity TfOH mixture. Fig. 2 shows HPLC profile of
the product. Glycopeptide 24 was eluted as peak 2,
while the unreacted octapeptide and the glycopeptide
missing a Gal residue are included in peak 1 and peak
3, respectively. The desired glycopeptide was isolated by
HPLC, and characterized by amino acid analysis as
well as by mass spectrometry. The overall yield esti-
mated was 27% based on the amino acid analysis.
9. Qiu, D.; Koganty, R. R. Tetrahedron Lett. 1997, 38,
45–48.
In conclusion, core 2 O-linked tetrasaccharide building
blocks 22 and 23 were synthesized via stereocontrolled
glycosylation by using N-trichloroacetyl-D-lactosaminyl
10. Meinjohanns, E.; Meldal, M.; Paulsen, H.; Bock, K. J.
Chem. Soc., Perkin Trans. 1 1995, 405–415.
11. Vargas-Berenguel, A.; Meldal, M.; Paulsen, H.; Bock, K.
J. Chem. Soc., Perkin Trans. 1 1994, 2615–2619.
12. Blatter, G.; Beau, J.-M.; Jacquinet, J.-C. Carbohydr. Res.
1994, 260, 189–201.
glycosyl donor 17. The benyl-protected tetrasaccharide
was used for the Fmoc-based solid-phase synthesis of
glycopeptide 24, representing leukosialin (215–224).
Removal of the benzyl protecting group was efficiently
accomplished under the low acidity TfOH conditions. It
was observed that the conditions also led to truncation
of the glycan chains in part. Further studies aimed at
expanding this benzyl-protection approach to the syn-
thesis of glycopeptides bearing complex N-linked gly-
cans are currently underway.
13. Madsen, R.; Roberts, C.; Fraser-Reid, B. J. Org. Chem.
1995, 60, 7920–7926.
14. Aly, M. R. E.; Castro-Palomino, J. C.; Ibrahim, E. I.;
El-Ashry, E. H.; Schmidt, R. R. Eur. J. Org. Chem. 1998,
2305–2316.
15. Bowers, S. G.; Coe, D. M.; Boons, G.-J. J. Org. Chem.
1998, 63, 4570–4571.
16. Castro-Palomino, J. C.; Schmidt, R. R. Tetrahedron Lett.
1995, 36, 6871–6874.
17. Castro-Palomino, J. C.; Schmidt, R. R. Tetrahedron Lett.
Acknowledgements
2000, 41, 629–632.
18. Singh, L.; Seifert, J. Tetrahedron Lett. 2001, 42, 3133–
3136.
19. Watabe, J.; Singh, L.; Nakahara, Y.; Ito, Y.; Hojo, H.;
Nakahara, Y. Biosci. Biotechnol. Biochem. 2002, 66,
1904–1914.
20. Takatani, M.; Nakama, T.; Kubo, K.; Manabe, S.;
Nakahara, Y.; Ito, Y.; Nakahara, Y. Glycoconjugate J.
2000, 17, 361–375.
This work was supported by Grant-in-Aid for Scientific
Research from the Ministry of Education, Culture,
Sports, Science, and Technology of Japan [(B) No.
13460051] and in part by CREST program of Japan
Science and Technology Corporation. The authors also
acknowledge Tokai University for a Grant-in-Aid for
High-Technology Research.

Y. Takano et al. / Tetrahedron Letters 43 (2002) 8395–8399
8399
21. Selected physical data. Compound 19: lH 5.77 (dd, 1H,
J=2.5, 53.7 Hz, H-1a); MALDI-TOF MS calcd for
C49H49Cl3FO10: 958.22 (M+Na)+. Found: 958.03. Com-
pound 20: lC 98.0 (1JCH=170.7 Hz, GalN3-C1), 99.4
(1JCH=157.5 Hz, GlcNTCA-C1), 102.7 (1JCH=160.9 Hz,
Gal-C1), 103.9 (1JCH=155.9 Hz, Gal-C1); MALDI-TOF
MS calcd for C110H112Cl3N5O24: 2014.47 (M+Na)+.
Found: 2014.42. Compound 21: lC 99.6 (1JCH=172.5 Hz,
GalN3-C1), 99.8 (1JCH=159.2 Hz, GlcNTCA-C1), 102.6
(1JCH=160.0 Hz, Gal-C1), 103.8 (1JCH=162.5 Hz, Gal-
C1); MALDI-TOF MS calcd for C111H114Cl3N5O24:
2028.68 (M+Na)+. Found: 2028.41. Compound 22: lH
1.81 and 1.56 (2s, 2×3H, AcNH); MALDI-TOF MS
calcd for C112H119N3O25: 1928.80 (M+Na)+. Found:
1928.53. Compound 23: lH 1.82 and 1.66 (2s, 2×3H,
AcNH); MALDI-TOF MS calcd for C113H121N3O25:
1942.82 (M+Na)+. Found: 1943.17. Compound 28:
MALDI-TOF MS calcd for C82H124N12O39: 1951.81 (M+
Na)+. Found: 1951.31. Compound 24: MALDI-TOF MS
calcd for C67H114N14O37: 1707.76 (M+H)+, 1729.74 (M+
Na)+. Found: 1707.89, 1729.65.
22. Suzuki, K.; Maeta, T.; Matsumoto, T.; Tsuchihashi, G.
Tetrahedron Lett. 1988, 29, 3567–3570.
23. Ito, Y.; Ogawa, T. Tetrahedron Lett. 1987, 28, 6221–6224.
24. Be´zay, N.; Dudziak, G.; Liese, A.; Kunz, H. Angew.
Chem., Int. Ed. 2001, 40, 2292–2295.
25. Tam, J. P.; Heath, W. F.; Merrifield, R. B. J. Am. Chem.
Soc. 1986, 108, 5242–5251.