Direct access to new β-d-galactofuranoconjugates: Application to the synthesis of galactofuranosyl-l-cysteine and l-serine

Article abstract of DOI:10.1016/j.tetlet.2011.01.001

Galactofuranose post-translational modifications, although quite rare, were detected in some biomolecules produced by parasites. While hexopyranosides were already linked to various peptides and proteins, few hexofuranosides have been artificially conjugated to amino acids. We thus report herein a robust glycosylation methodology to obtain S-alkyl, O-serine and S-cysteine-β-d- galactofuranosides starting from readily available galactofuranose donors. O-Acetyl, thioimidoyl and acetimidoyl donors were compared in terms of yields and selectivity when reacted with mercaptans, l-cysteine and l-serine. Acetimidates turned out to be the best notably for amino acids glycosylation.

Full text of DOI:10.1016/j.tetlet.2011.01.001

Tetrahedron Letters 52 (2011) 1121–1123
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Direct access to new b-D-galactofuranoconjugates: application to the
synthesis of galactofuranosyl-L-cysteine and L-serine
⇑
⇑
Dancho Danalev , Laurent Legentil , Richard Daniellou, Caroline Nugier-Chauvin, Vincent Ferrières
Ecole Nationale Supérieure de Chimie de Rennes, CNRS, UMR 6226, Avenue du Général Leclerc, CS 50837, 35708 Rennes Cedex 7, France
Université européenne de Bretagne, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Galactofuranose post-translational modifications, although quite rare, were detected in some biomole-
cules produced by parasites. While hexopyranosides were already linked to various peptides and pro-
teins, few hexofuranosides have been artificially conjugated to amino acids. We thus report herein a
Received 26 October 2010
Revised 13 December 2010
Accepted 4 January 2011
Available online 7 January 2011
robust glycosylation methodology to obtain S-alkyl, O-serine and S-cysteine-b-
D-galactofuranosides start-
ing from readily available galactofuranose donors. O-Acetyl, thioimidoyl and acetimidoyl donors were
compared in terms of yields and selectivity when reacted with mercaptans,
timidates turned out to be the best notably for amino acids glycosylation.
L-cysteine and L-serine. Ace-
Keywords:
Carbohydrates
Glycosylation
Ó 2011 Elsevier Ltd. All rights reserved.
Thiofuranosides
Glycofuranosyl amino acids
1. Introduction
to a heterogeneous mixture of glycoforms that differ in the nature
and site of glycosylation. This phenomenon added difficulties to
identify the exact effect of b-galactofuranosyl motif on the activity.
Consequently, definition of a straightforward strategy for regiose-
lective glycosylation on the native proteins is of primordial
importance.
Proper timing glycosylation and assembly strategy are vital
issue for production of glycopeptides and glycoproteins. They have
been extensively reviewed in the previous years.10–14 These tech-
niques implied the use of pre-established glycosylated amino acids
as building-blocks. As far as we are aware, only stereoselective syn-
theses of pentofuranosyl amino acids were reported and involved
arabinosyl or ribosyl glycosylation to serine or asparagine.15–17 In
this general context, we thus identified and synthesized a family
b-D-Galactofuranosyl moiety forms an original motif in nature.
Completely unknown in mammalian systems,1 it has been found
in various bacteria, fungi and parasites as part of complex glyco-
conjugates.2,3 These latter were involved as cell-wall constituents
or were incorporated into proteins by post-translational modifica-
tions. Post-translational glycosylations add particular properties to
biomolecules, favour stability against proteolysis, and in certain
cases increase the biological activities of the protein itself.4,5 For
example, b-galactofuranosyl residues are integrated to the O-
linked glycan of Aspergillus niger glucoamylase,6 the invertase
produced by Aspergillus nidulans7 or the N-linked glycans of
a
-galactosidase A from A. niger.8 This latter microorganism is an
important fungi involved in the industrial production of many sub-
stances for food production and preservation notably. In particular,
of b-
simple sugar probes for further protein ligation (Fig. 1). Our targets
were designed around b- -Galf connected to either a functional-
ized thioalkyl arm (1a to 1d) or the side chain of -cysteine (1e)
D-galactofuranosyl (b-D-Galf) conjugates that can be used as
a-galactosidase A catalytically trims
a
-linked galactose from oligo-
D
saccharides. It was suggested that the presence of terminal
b-galactofuranosyl residues on the N-linked glycans stopped fur-
ther elongation of the oligomer. However, the biological signifi-
cance of this monosaccharide has yet to be fully elucidated and
its associated antigenicity has to be fully understood. Recently
our group showed that oligofuranoses possessed some adjuvant
properties.9 Moreover, expression of glycoproteins leads in general
L
or serine (1f). Interestingly, thioglycosidic bonds in 1a to 1e are
R
1a
1b
SBn
SCH2CO2Et
R
AcO
O
1c
1d
1e
1f
S(CH2)2CO2Me
S(CH2)3Cl
SCH2CH(NHAc)CO2Et
AcO
AcO
OAc
⇑
Corresponding author. Tel.: +33 223 238 140; fax: +33 223 238 046 (L.L.).
E-mail address: laurent.legentil@ensc-rennes.fr (L. Legentil).
Present address: University of Chemical Technology and Metallurgy, Department
of Organic Chemistry, Sofia 1756, Bulgaria, 8 Blvd. Kliment Ohridski, Bulgaria. Tel.:
+359 2 81 63 418, fax: +359 2 868 54 88.
OCH2CH(NHCbz)CO2Me
Figure 1. Targeted b-
D-galactofuranosides.
0040-4039/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2011.01.001

1122
D. Danalev et al. / Tetrahedron Letters 52 (2011) 1121–1123
Table 1
known to be more stable against acidic and enzymatic hydrolysis
compared to their O-counterpart.18
Access to alkyl 1-thio-b-D-galactofuranosides 1a–d
SR2
Grafting of serine onto different glycosides was widely de-
scribed in particular in the hexopyranose series.19,20 However, di-
rect O-glycosylation of the side chain of the protected serine
usually necessitates activation of the carbohydrate donor such as
glycosyl bromide or thiophenyl glycoside.17 As for cysteine, it is
one of the least abundant amino acids in proteins and was popular-
ized as a good linker for glycopeptide synthesis by different
groups.21 Its glycosylation on pyranoses followed two main path-
ways; anomeric nucleophilic thiolate anions generated from
pyranosyl isothiouronium could substitute iodo-alanine22 in good
to excellent yields while Koenigs–Knorr reaction allowed moderate
access to thioglycosidic bond.23 Considering the literature, there is,
therefore, a need to develop a new methodology for amino acids
glycosylation in particular in the furanose series. Indeed, no exam-
ple describes direct glycosylation of cysteine or serine on galactof-
uranoses. Consequently we thought to evaluate the potential of
acetate, thioimidate and trichloroacetimidate as furanosyl donors
for direct glycosylation with alkyl mercaptans and amino acid
acceptors (Scheme 1).
AcO
AcO
O
R2SH
O
R1
AcO
AcO
AcO
BF3.OEt2
CH2Cl2
OAc
OAc
AcO
Donor (R1)
2 ( ,b-OAc)
Entry
T (°C)
Product
R2
Yield (%)
1
2
3
4
5
6
7
8
a
0
20
0
20
0
20
0
20
1a
1a
1b
1b
1c
1c
1d
1d
SBn
SBn
81
53
69
63
63
74
83
52
3 (b-SPyrim)
2 ( ,b-OAc)
3 (b-SPyrim)
2 ( ,b-OAc)
3 (b-SPyrim)
2 ( ,b-OAc)
3 (b-SPyrim)
a
SCH2CO2Et
SCH2CO2Et
S(CH2)2CO2Me
S(CH2)2CO2Me
S(CH2)3Cl
a
a
S(CH2)3Cl
room temperature with complete diastereoselectivity. However,
glycosylation of benzyl mercaptan under previous conditions pro-
vided 1a in a much lower yield of 53% (entry 2). Similar conclu-
sions were reached when our protocol was extended to various
thiol nucleophiles; 2-mercaptoacetic acid ethyl ester (Mpe-OEt,
entries 3 and 4), 3-mercaptopropanoic acid methyl ester (Mpa-
OMe, entries 5 and 6) and 3-chloropropane mercaptan (entries 7
and 8). These functionalized thioalkyl derivatives could serve as
arm linkers for further ligation to peptides or proteins.27 Further-
more, replacement of the natural amino acid cysteine with Mpa
in peptide or proteins often leads to significant increase of their
biological activities.28 Once again, in general, per-O-acetylated do-
nor 2 turned out to be better in terms of yields and low tempera-
ture was required to access pure 1,2-trans diastereoisomers 1b to
1d.
2. Results and discussion
First a model study for direct glycosylation between the avail-
able per-O-acetylated-D
-galactofuranose 224 and benzyl mercaptan
was implemented (Table 1). Best yields occurred in a non-protic
and non-participating solvent such as dichloromethane. The nature
of the Lewis acid seemed to influence greatly the outcome of the
reaction. Weak one as copper(II) triflate failed to initiate the
reaction while the strongest trimethylsilyltriflate resulted in the
formation of one side product namely 2,3,5,6-tetra-O-acetyl galact-
ose. However, boron trifluoride diethyl etherate presents the right
acidity balance to afford the aimed benzyl 1-thio-galactofurano-
side 1a as a pure 1,2-trans anomer in 81% yield (Table 1, entry 1).
To ensure best yields and diastereoselectivity, the reaction
proceeded at 0 °C. Indeed, when performed at room temperature,
the resulting reaction mixture was polluted with some 1,2-cis ano-
mer. b-Configuration was obtained thanks to the presence of the
participating acetate group at position 2 and was confirmed by a
small NMR coupling constant between anomeric proton H-1 and
H-2 (J1,2 = 2 Hz).25
Direct ligation either on N-terminal or side chain of serine, cys-
teine or glycine provided the simplest access to glycopeptides.
Consequently, we have adapted our methodology for direct glyco-
sylation of serine and cysteine derivatives (Table 2). Surprisingly,
our protocol could not be reproduced for the synthesis of S-galac-
tofuranosyl cysteine 1e. Indeed while N-acetyl-L-cysteine methyl
ester was not intrinsically chemically different from the other thiol
acceptors, ligation to per-O-acetylated donor 2 appeared to be
unsuccessful (entry 1). Good yields were however obtained with
thioimidate 3 but the reaction turned out to be sluggish (entry
3). To overcome these difficulties, a third donor, namely galactofur-
anosyl trichloroacetimidate 4 was used.29 It was obtained accord-
ing to a two-step procedure starting from 2 (Scheme 2). Firstly,
the anomeric position was carefully deacetylated with hydrazine
acetate30 in dimethylformamide. It has never been described be-
To extend the study, another well documented donor 326 was
also used (Table 1). We chose a pyrimidazoyl aglycon as it led to
the best yields for glycosylation with various alcohols when
acetate protected. It had also the advantage to react readily at
X
AcO
CO2Me
NHP
O
Table 2
AcO
Access to S-galactofuranosyl cysteine and O-galactofuranosyl serine 1e–f
OAc
1e 1f
AcO
X = O or S
CO2Me
HX
,
AcO
O
NHR2
R2
X
AcO
AcO
O
CO2Me
O
R1
P : Protecting Group
R2HN
AcO
AcO
Promoter
CH2Cl2
AcO
AcO
OAc
OAc
OAc
AcO
Entry Donor (R1)
AcO
R2
OAc
SPyrim
OC(NH)CCl3
Conditions (T, °C) Product (X) R2
Yield (%)
S
R1
AcO
O
2
3
4
n
1
2
3
4
5
6
2 (
2 (
a
a
,b-OAc)
,b-OAc)
BF3ꢀOEt2 (0)
BF3ꢀOEt2 (0)
BF3ꢀOEt2 (20)
BF3ꢀOEt2 (20)
Cu(OTf)2 (20)
TMSOTf (0)
1e (S)
1f (O)
1e (S)
1f (O)
1f (O)
1e (S)
Ac
Cbz 65
Ac
Cbz
Cbz 48
Ac 80
—
AcO
AcO
3 (b-SPyrim)
3 (b-SPyrim)
3 (b-SPyrim)
4 [b-
OC(NH)CCl3]
4 [b-
79
—
OAc
1a to 1d
1<n<3
R1 = Bn, CO2Me, CO2Et or Cl
7
TMSOTf (0)
1f (O)
Cbz 75
OC(NH)CCl3]
Scheme 1. Proposed retrosynthesis for b-D-galactofuranosides 1a–f.

D. Danalev et al. / Tetrahedron Letters 52 (2011) 1121–1123
1123
like also to thank National Research Fund of Bulgaria for a financial
support (contract DOO2-296).
AcO
AcO
N2H4.AcOH
OAc
O
O
OH
AcO
AcO
AcO
DMF
OAc
OAc
Supplementary data
AcO
38%
5
2
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.tetlet.2011.01.001.
CCl3
NH
CCl3CN
DBU
CH2Cl2
O
AcO
O
References and notes
AcO
AcO
OAc
1. de Lederkremer, R. M.; Colli, W. Glycobiology 1995, 5, 547–552.
2. Peltier, P.; Euzen, R.; Daniellou, R.; Nugier-Chauvin, C.; Ferrières, V. Carbohydr.
Res. 2008, 343, 1897–1923.
3. Richards, M.; Lowary, T. ChemBioChem 2009, 10, 1920–1938.
4. Rudd, P. M.; Joao, H. C.; Coghill, E.; Fiten, P.; Saunders, M. R.; Opdenakker, G.;
Dwek, R. A. Biochemistry 1994, 33, 17–22.
88%
4
Scheme 2. Synthesis of b-D-galactofuranosyl trichloroacetimidate 4.
5. Susaki, H.; Suzuki, K.; Ikeda, M.; Yamada, H.; Watanabe, H. K. Chem. Pharm. Bull.
1998, 46, 1530–1537.
6. Wallis, G. L. F.; Hemming, F. W.; Peberdy, J. F. Biochim. Biophys. Acta 2001, 1525,
19–28.
7. Hemming, F. W.; Wallis, G. L. F.; Peberdy, J. F. Anal. Biochem. 2000, 279, 136–
141.
8. Wallis, G. L. F.; Easton, R. L.; Jolly, K.; Hemming, F. W.; Peberdy, J. F. Eur. J.
Biochem. 2001, 268, 4134–4143.
9. Chlubnova, I.; Filipp, D.; Spiwok, V.; Dvorakova, H.; Daniellou, R.; Nugier-
Chauvin, C.; Kralova, B.; Ferrières, V. Org. Biomol. Chem. 2010, 8, 2092–2102.
10. Buskas, T.; Ingale, S.; Boons, G.-J. Glycobiology 2006, 16, 113R–136R.
11. Davis, B. Angew. Chem., Int. Ed. 2009, 48, 4674–4678.
12. Davis, B. G. Chem. Rev. 2002, 102, 579–602.
13. Gamblin, D. P.; Scanlan, E. M.; Davis, B. G. Chem. Rev. 2008, 109, 131–163.
14. Taylor, C. M. Tetrahedron 1998, 54, 11317–11362.
15. Bonache, M. A.; Nuti, F.; Le Chevalier Isaad, A.; Real-Fernández, F.; Chelli, M.;
Rovero, P.; Papini, A. M. Tetrahedron Lett. 2009, 50, 4151–4153.
16. Ginisty, M.; Gravier-Pelletier, C.; Le Merrer, Y. Tetrahedron: Asymmetry 2006,
17, 142–150.
fore on furanosides as previous groups have preferentially used
thiophenyl furanosides as intermediates.
On the assumption that furanoses are more reactive than pyra-
noses, regioselectivity was ensured by limiting reaction time to 1 h.
Under these conditions, the corresponding hemiacetal was isolated
in 38% yield at best. Beyond this time, mixture was observed. Fur-
ther conversion into the desired trichloroacetimidate
4 was
achieved by reaction with trichloroacetonitrile in the presence of
1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) in 88% yield.
As expected, trichloroacetimidate 4 proved to be the best donor
and allowed us to readily glycosylate cysteine under common
TMSOTf activation in 80% yield (Table 2, entry 6).31
As for the introduction of serine, t-butyloxycarbonyl protection
turned out to be not compatible with the acidic conditions em-
ployed. Gratifyingly, N-benzylcarbamate serine methyl ester was
a good acceptor and reacted with the various donors 2 (Table 2, en-
try 2), 3 (entries 4 and 5) and 4 (entry 7). Interestingly, a weaker
Lewis acid, copper(II) triflate was required with thioimidate 3.
These results illustrated the importance of the Lewis acid nature
for glycosylation reactions.21 Optimized yields were obtained once
again with trichloroacetimidate 4 to give O-galactofuranosyl serine
in 75% yield.
17. Kurosu, M.; Li, K. J. Org. Chem. 2008, 73, 9767–9770.
18. Driguez, H. In Glycoscience Synthesis of Substrate Analogs and Mimetics; Driguez,
H., Thiem, J., Eds.; Springer: Berlin/Heidelberg, 1997; Vol. 187, pp 85–116.
19. Arsequell, G.; Valencia, G. Tetrahedron: Asymmetry 1997, 8, 2839–2876.
20. Kihlberg, J.; Aahman, J.; Walse, B.; Drakenberg, T.; Nilsson, A.; Soederberg-
Ahlm, C.; Bengtsson, B.; Olsson, H. J. Med. Chem. 1995, 38, 161–169.
21. Pachamuthu, K.; Schmidt, R. R. Chem. Rev. 2006, 106, 160–187.
22. Knapp, S.; Myers, D. S. J. Org. Chem. 2002, 67, 2995–2999.
23. Baran, E.; Drabarek, S. Pol. J. Chem. 1978, 52, 941.
24. Ferrières, V.; Gelin, M.; Boulch, R.; Toupet, L.; Plusquellec, D. Carbohydr. Res.
1998, 314, 79–83.
25. NMR data for all new compounds are given in Supplementary data.
26. Euzen, R.; Ferrières, V.; Plusquellec, D. Carbohydr. Res. 2006, 341, 2759–2768.
27. Morales-Sanfrutos, J.; Lopez-Jaramillo, J.; Ortega-Munoz, M.; Megia-Fernandez,
A.; Perez-Balderas, F.; Hernandez-Mateo, F.; Santoyo-Gonzalez, F. Org. Biomol.
Chem. 2010, 8, 667–675.
3. Conclusion
28. Vezenkov, L.; Mladenova-Orlinova, L. D. Bolgarskoj akad. nauk 1990, 43, 61–63.
29. Completo, G. C.; Lowary, T. L. J. Org. Chem. 2008, 73, 4513–4525.
30. de la Fuente, J. M.; Penadés, S. Tetrahedron: Asymmetry 2002, 13, 1879–1888.
To summarize, new galactofuranosyl amino acids building
blocks were synthesized starting from readily available per-
O-acetylated and pyrimidoyl donors. It included galactofuranosyl
serine and a small library of alkyl 1-thio-galactofuranosides.
The incorporation of cysteine proved to be more challenging and
forced us to use the more reactive trichloroacetimidate donor.
Further studies to conjugate hexofuranosyl amino acids and
analogues onto more complex biomolecules are presently under
investigation.
31. Typical procedure for S-furanosylation of cysteine:
trichloroacetimidate donor 4 (40 mg, 81 mol), N-acetyl-
ester (11 mg, 62 mol) and 4 Å molecular sieves in CH2Cl2 (2 mL) were stirred
at RT for 10 min. Then after cooling to 0 °C, trimethylsilyltriflate (4 L,
20 mol) was added and the reaction was monitored by TLC (cyclohexane/
A
mixture of
l
L-cysteine methyl
l
l
l
AcOEt 1:1). After 1 h stirring, no starting material remained. So the mixture
was neutralized with Et3N (0.5 mL), filtered through a small pad of Celite and
evaporated under vacuum. The crude oil obtained was purified by column
chromatography on silica gel (CH2Cl2/MeOH 99:1) to give 1e (25 mg, 79%) as
colourless oil. ½a D20
ꢁ
= ꢂ9 (c 1, CH2Cl2). 1H NMR (400 MHz, CDCl3): d 6.44 (d, 1H,
J = 7.4 Hz, NH), 5.36 (td, 1H, J = 7.2, 4.3 Hz, H-5), 5.30 (d, 1H, J = 2.0 Hz, H-1),
5.05–5.01 (m, 2H, H-2, H-3), 4.88 (td, 1H, J = 7.4, 5.0 Hz, CHCH2S), 4.30 (dd, 1H,
J = 12.0, 4.3 Hz, H-6), 4.35–4.32 (m, 1H, H-4), 4.18 (dd, 1H, J = 12.0, 7.2 Hz, H-
60), 3.77 (s, 3H, CO2CH3), 3.18 (dd, 1H, J = 14.1, 5.0 Hz, CH2S), 3.04 (dd, 1H,
J = 14.1, 5.0 Hz, CH2S), 2.14, 2.11, 2.08, 2.06, 2.04 (5s, 15H, COCH3). 13C NMR
(100 MHz, CDCl3): d 170.9, 170.6, 170.03, 169.99, 169.85, 169.8 (CO), 88.6 (C-
1), 81.8 (C-2), 79.8 (C-4), 76.4 (C-3), 69.1 (C-5), 62.5 (C-6), 52.8 (CO2CH3), 51.9
(CHCH2S), 33.2 (CHCH2S), 23.0 (NHCOCH3), 20.8, 20.7, 20.6 (COCH3). HRMS
(ESI): calcd for C20H29O12NSNa [M+Na]+ 530.1308, found 530.1401.
Acknowledgements
We are grateful to CNRS and ‘ministère de la recherche’ for the
financial support, to ‘Centre Regional des Mesures Physiques de
l’Ouest, Université de Rennes 1’ for the Mass spectra and the
‘Agence Universitaire de la Francophonie’ (AUF) for financial
supports through an Intra-European fellowship for D.D. We would