Synthesis of the zwitterionic repeating unit of the O-antigen from Shigella sonnei and chain elongation at both ends

Article abstract of DOI:10.1021/ol502395k

Shigella sonnei O-antigen features a zwitterionic disaccharide repeat encompassing two rare monosaccharides. The synthesis of the AB repeat and of trisaccharides ABA' and B'AB, which validates chain elongation at either end, is reported. All targets were synthesized using a postglycosylation oxidation strategy in combination with imidate chemistry. Precursors to residue A were obtained from L-glucose. The AAT (B) donor and acceptor were obtained from D-glucosamine. A one-step Pd(OH)₂/C-mediated deprotection provided the propyl glycoside targets.

Full text of DOI:10.1021/ol502395k

Letter
pubs.acs.org/OrgLett
Synthesis of the Zwitterionic Repeating Unit of the O‑Antigen from
Shigella sonnei and Chain Elongation at Both Ends
Helene B. Pfister†,‡,§ and Laurence A. Mulard*,†,‡
́
̀
†Institut Pasteur, Unite
́ ́
de Chimie des Biomolecules, 28 rue du Dr Roux, 75724 Paris Cedex 15, France
‡CNRS UMR3523, Institut Pasteur, 75015 Paris, France
§Universite
́ ́
Paris Descartes Sorbonne Paris Cite, Institut Pasteur, 75015 Paris, France
S
* Supporting Information
ABSTRACT: Shigella sonnei O-antigen features a zwitterionic
disaccharide repeat encompassing two rare monosaccharides. The
synthesis of the AB repeat and of trisaccharides ABA′ and B′AB,
which validates chain elongation at either end, is reported. All targets
were synthesized using a postglycosylation oxidation strategy in
combination with imidate chemistry. Precursors to residue A were
obtained from L-glucose. The AAT (B) donor and acceptor were
obtained from D-glucosamine. A one-step Pd(OH)2/C-mediated deprotection provided the propyl glycoside targets.
Shigellae are Gram-negative enteroinvasive bacteria causing
shigellosis in humans.1 Shigellosis remains one of the five major
diarrheal diseases in children under five.2 Epidemiological data
still call for the development of a Shigella vaccine.2 Out of the 4
species of Shigella, S. sonnei is of special concern.3 It is present
in developing countries where disease is endemic and is the
prevalent species in transitional and industrialized countries.2
Naturally acquired immunity against Shigella is in part directed
against the O-specific polysaccharide (O-SP) of its lip-
opolysaccharide (LPS).4 On that basis, polysaccharide con-
jugates have received particular attention in the search for a
Shigella vaccine. Protective efficacy was demonstrated for the
most advanced detoxified LPS-based candidates.5 Conversely,
synthetic carbohydrate haptens are being explored as promising
alternatives to material of biological origin. In this context, we
have proposed the first synthetic carbohydrate-based vaccine
candidate against the prevalent Shigella serotype.6 The designed
conjugate encompasses a three-repeat segment of the O-SP
from S. flexneri 2a.7 To answer the need for a broadly protective
vaccine against endemic shigellosis, the strategy was expanded
to other S. flexneri serotypes.8 Herein, the target is the O-SP
from S. sonnei. This zwitterionic polysaccharide has a
disaccharide repeating unit made of two uncommon amino-
sugars, a 2-acetamido-2-deoxy-L-altruronic acid (residue A) and
a 2-acetamido-4-amino-2,4,6-trideoxy-D-galactopyranose (AAT,
residue B) 1,2-trans linked to one another (Figure 1).9 Besides
its zwitterionic nature, a peculiar feature of the S. sonnei O-SP is
the occurrence of three amino groups, two of which present as
acetamides, within a disaccharide repeat. We report an original
synthesis of disaccharide AB, the biological repeat from S.
sonnei O-SP, and of the frame-shifted trisaccharides ABA′ and
B′AB. In doing so, we validate the feasibility of chain
elongation of an orthogonally protected AB intermediate by
one residue at either end, paving the way to larger synthetic S.
sonnei O-SP fragments.
All three targets were prepared in the form of propyl
glycosides 1 (AB-Pr), 2 (ABA′-Pr), and 3 (B′AB-Pr) to mimic
the 1,2-trans linkage occurring in the native polymer (Scheme
1). They were obtained from a common AB allyl glycoside
Scheme 1. Di- and Trisaccharide Targets 1−3 and Key
Common Intermediates 4−6
Figure 1. Biological repeat of the O-SP from S. sonnei (AB).9
Received: August 12, 2014
Published: September 11, 2014
© 2014 American Chemical Society
4892
dx.doi.org/10.1021/ol502395k | Org. Lett. 2014, 16, 4892−4895

Organic Letters
Letter
intermediate 4, whose amino group at position 4B is masked as
an azido moiety. The latter was selected for its small size, its
orthogonal properties to most protecting groups, and its easy
conversion to an amine upon hydrogenation. In addition,
disaccharide 4 features trichloroacetamide moieties at positions
2A and 2B to ensure appropriate anchimeric assistance during
glycosylation,10 and direct recovery of the acetamide moieties
upon final hydrodechlorination.8a,11 Disaccharide 4 was built
from the unknown AAT acceptor 5 and allyl altroside 6. The
latter featured a 4,6-O-benzylidene acetal in agreement with the
adopted postglycosidation oxidation strategy.12
Synthesis of the AAT Acceptor 5 and Donor 13. Most often
present as an α-linked residue and occasionally occurring as a β-
linked component, AAT is a constituent of a variety of
zwitterionic bacterial polysaccharides.13 It was synthesized
either as its methyl glycoside or in diverse forms compatible
with donor formation (reviewed in ref 13a) and use thereof. In
the latter case, most strategies employ 2-azido derivatives.13a,14
Only on rare occasions, when AAT was involved in 1,2-trans
glycosylation, were other N-protecting groups used at position
2.15 Noticeably, in their original synthesis of the methyl
anomeric deallylation and reaction of the resulting hemiacetal
with PTFACl.
Synthesis of the L-Altrose Donor 18 and L-Altruronate Acceptor
19 via Orthogonally Protected 6. L-Altrosamine residues are not
commonly encountered in bacterial polysaccharides, nor are
they easily available from natural sources. Besides concern
about handling S. sonnei,18,19 routes to 2-amino-2-deoxy-altrose
derivatives have been explored mostly owing to interest in
mannosamines and their derivatives, among which are
uronates.20 Therefore, the synthesis of donor 18 was adapted
from previous work.20,18 Starting from commercially available L-
glucose, Fischer-type glycosylation followed by regioselective
benzylidenation gave a 3:1 α/β mixture of allyl glycosides that
could be separated by flash chromatography (Scheme 3). The
Scheme 3. Synthesis of the Orthogonally Protected L-Altrose
Donor 18 and L-Altruronate Acceptor 19
̀
glycoside of the S. sonnei BA disaccharide, Liptak et al. used a 2-
N-trichloroacetyl thioethyl glycoside donor, itself obtained from
a 2-phtalimido AAT precursor.15b Instead, the strategy depicted
herein is based on the early stage introduction of a 2-N-
trichloroacetyl moiety10,16 and on the powerful N-phenyl-
trifluoroacetimidate glycosylation chemistry.17 Thus, starting
from allyl glycoside 7,16 obtained from glucosamine hydro-
chloride via the corresponding β-tetraacetate,18 transesterifica-
tion, regioselective 6-O-tosylation of the resulting triol, and
subsequent O-acetylation gave tosylate 8 (Scheme 2).
Scheme 2. Synthesis of the Protected AAT Residue 12 and
Conversion into Acceptor 5 and Donor 13
α-anomer 14 was converted to the allo-epoxide 15 by means of
the 2,3-di-O-mesyl intermediate. Selective trans-diaxial opening
of epoxide 15 by sodium azide in the presence of a minimal
amount of ammonium chloride enabled the regioselective
installation of an amino precursor at C-2 providing altropyrano-
side 16 as ascertained from NMR data. Yet, partial trans-
diequatorial opening of the epoxide ring into allyl 3-azido-4,6-
O-benzylidene-3-deoxy-α-L-glucopyranoside could not be
avoided (8%, Scheme S1), adding to the scarce examples21
that do not fully conform to the Furst−Plattner rules.
̈
Benzylation of alcohol 16 provided the fully protected 17,
which was converted into its trichloroacetamide analogue 6
upon selective reduction of the azido moiety under Staudinger
conditions and trichloroacetylation of the resulting crude
amine. In the first place,22 trichloroacetamide 6 was isolated
in poor yield together with the carbodiimide dimer (Scheme
S2). The latter coeluted with 4-N,N-diethylamino-1,1,1-
trichloro-3-buten-2-one,23 arising from the oxidation of
triethylamine by trichloroacetyl chloride. Isocyanate formation
from a trichloroacetamide via a haloform reaction has
precedent.24 On that basis, we assumed that the carbodiimide
side product was likely formed by reaction of the
iminophosphorane intermediate with the corresponding
isocyanate.25 In an attempt to drive the hydrolysis of the
iminophosphorane to completion, the amount of triphenyl-
phosphine and water was increased and the reaction temper-
Conversion of the latter into the 6-iodo analogue 9, followed
by NaBH3CN-mediated reduction and acetate removal, gave
the deoxy intermediate 10. Fine-tuning of the reaction
conditions enabled the critical regioselective acetylation at O-
3 to give alcohol 11. A decisive two-step process, involving
formation of the intermediate triflate, and its nucleophilic
displacement with sodium azide provided the fully protected 4-
azido precursor 12. On the one hand, selective acetate removal
from the latter gave acceptor 5. On the other hand,
intermediate 12 was readily converted into donor 13, following
4893
dx.doi.org/10.1021/ol502395k | Org. Lett. 2014, 16, 4892−4895

Organic Letters
Letter
ature was raised. Unfortunately, these conditions led to an
imine issued from the reaction of the excess phosphine with the
newly formed trichloroacetamide 6 (Scheme S2).26 Finally,
using only a stoichiometric amount of phosphine enabled a
high yielding two-step transformation (Scheme 3). Deallylation
of the orthogonally protected 6, and reaction of the formed
hemiacetal with PTFACl, gave donor 18, equipped for
anchimeric assistance. Alternatively, the fully protected 6 was
subjected to acetal acidolysis, selective 2,2,6,6-tetramethyl-
piperidine-1-oxy (TEMPO) radical/[bis(acetoxy)iodo] ben-
zene (BAIB) oxidation27 of the primary alcohol under
conditions that left the allyl aglycon untouched,11 and
subsequent benzyl esterification to evolve into the uronate
acceptor 19.
desired AB-Pr disaccharide 1 following RP-HPLC purification.
Alternatively, disaccharide 4 was readily converted into an AB
donor. Thus, chloroacetylation of the 4A-OH from intermediate
4 furnished the fully protected 22. Anomeric deallylation of the
latter and subsequent reaction of the hemiacetal intermediate
with PTFACl gave donor 23.
In contrast to the standard 1C4 conformation adopted by allyl
glucoside 14, altrose derivatives have been shown to occupy all
1C4, OS2, and 4C1 conformations in solution.28 When present in
the form of a benzylidene acetal as in monosaccharide 16 and
disaccharide 20, the α-linked altrose residue is locked in a
distorted conformation (Table 1, entry 1). The 1,3-transdiaxial
substitution pattern and the axial substitutions at C-1, C-2, and
C-3 probably account for important energy constraints
Synthesis of Disaccharide AB-Pr (1), Trisaccharide ABA′-Pr
(2), and BAB′-Pr (3) via the AB Intermediate 4. TMSOTf-
mediated coupling of the AAT acceptor 5 and the altrosaminyl
donor 18 proceeded smoothly at −30 °C to furnish building
block 20 (Scheme 4). Benzylidene cleavage provided diol 21,
1
destabilizing a standard C4 chair. Instead, removal of the 4,6-
acetal and subsequent oxidation drive the altropyranose ring
into a significant conformational change as seen in the benzyl
altruronate 4 (Table 1, entry 2).
3
Table 1. Selected JH,H Coupling Constants (Hz) for the
Altrose Residue in Compounds 1−4, 16, 20, and 25
Scheme 4. Synthesis of AB-Pr (1) and of the AB Donor (23)
entry
compd
3J1,2/3J2,3
compd
3J1,2/3J2,3
1
2
16
4
0.9/3.0
3.3/5.4
20
25
1
<0.5/2.7
7.5/10.4
8.3/10.6
8.5/10.8
A: 8.4/10.7
A′: 8.1/10.3
3
2
3
With disaccharide AB both in the form of acceptor 4 and
donor 23 in hand, the next step consisted of assessing the
feasibility of chain elongation at either end. Toward this aim,
the uronate acceptor 19 was reacted with donor 23 under
conventional glycosylation conditions (Scheme 5). Quenching
the reaction with a minimal amount of Et3N permitted the
isolation of the desired coupling product 24 in a good 69%
yield. However, whereas epimerization at C-5A was not
observed, formation of the 4A:5A-unsaturated trisaccharide
uronate, likely occurring during basic treatment,29 could not
be avoided (Scheme S3). Alternatively, trisaccharide 25 was
obtained from the TMSOTf-mediated coupling of acceptor 4
with the AAT donor 13. Interestingly, as a result of the
glycosylation at O-4, the altruronate residue A switched from a
4
distorted conformation in acceptor 4 to a more standard C1
conformation in trisaccharide 25 (Table 1, entry 2). Altogether,
data strongly suggest that the conformation adopted by the
altruronate residues in S. sonnei protected oligosaccharides is
highly dependent on their substitution pattern.
Trisaccharides ABA′-Pr (2) and B′AB-Pr (3) were isolated
following RP-HPLC purification of the material obtained by
which was subjected to chemo- and regioselective oxidation at
C-6 using the TEMPO/BAIB system as described above.
Subsequent conversion of the newly generated carboxyl group
into a benzyl ester gave the AB disaccharide acceptor 4.
Pd(OH)2/C-mediated allyl reduction, benzyl hydrogenolysis,
and concomitant hydrodechlorination at two sites gave the
Scheme 5. Synthesis of ABA′-Pr (2) and B′AB-Pr (3)
4894
dx.doi.org/10.1021/ol502395k | Org. Lett. 2014, 16, 4892−4895

Organic Letters
Letter
(12) van den Bos, L. J.; Codee, J. D. C.; Litjens, R. E. J. N.; Dinkelaar,
J.; Overkleeft, H. S.; van der Marel, G. A. Eur. J. Org. Chem. 2007,
3963.
appropriate ester cleavage in the fully protected 24 and 25,
respectively, and subsequent Pd(OH)2/C-mediated removal
3
and/or reduction of all remaining protecting groups. JH,H
(13) (a) Christina, A. E.; Blas Ferrando, V. M.; de Bordes, F.; Spruit,
NMR data of the propyl glycosides 1−3 indicated that the
altropyranose residue exists preferentially in the 4C1 con-
formation as in the S. sonnei O-SP30 (Table 1, entry 3). This
trend seemed less pronounced for residue A′ than for residue A
in ABA′-Pr (2). As a whole, the available NMR data call for
further chain elongation to reach conformational mimics of the
natural O-SP.30
The first synthesis of the biological repeat of the zwitterionic
O-SP of S. sonnei and of two frame-shifted trisaccharides was
reported. Use of the postglycosylation oxidation strategy gave
the AB building block 4, designed and validated for chain
elongation at both ends.
́
W. A.; Overkleeft, H. S.; Codee, J. D. C.; van der Marel, G. A.
Carbohydr. Res. 2012, 356, 282. (b) Emmadi, M.; Kulkarni, S. S. Nat.
Prod. Rep. 2014, 31, 870.
(14) (a) Schumann, B.; Pragani, R.; Anish, C.; Pereira, C. L.;
Seeberger, P. H. Chem. Sci. 2014, 5, 1992. (b) Emmadi, M.; Kulkarni,
S. S. Nat. Protocols 2013, 8, 1870. (c) Pedersen, C. M.; Figueroa-Perez,
I.; Boruwa, J.; Lindner, B.; Ulmer, A. J.; Zahringer, U.; Schmidt, R. R.
Chem.Eur. J. 2010, 16, 12627. (d) Wu, X.; Cui, L.; Lipinski, T.;
Bundle, D. R. Chem.Eur. J. 2010, 16, 3476. (e) Iynkkaran, I.; Bundle,
D. R. Carbohydr. Res. 2010, 345, 2323.
(15) (a) Pedersen, C. M.; Figueroa-Perez, I.; Lindner, B.; Ulmer, A.
J.; Zaehringer, U.; Schmidt, R. R. Angew. Chem., Int. Ed. 2010, 49,
2585. (b) Toth, A.; Medgyes, A.; Bajza, I.; Liptak, A.; Batta, G.;
Kontrohr, T.; Peterffy, K.; Pozsgay, V. Bioorg. Med. Chem. Lett. 2000,
10, 19.
(16) Boutet, J.; Kim, T. H.; Guerreiro, C.; Mulard, L. A. Tetrahedron
Lett. 2008, 49, 5339.
(17) Yu, B.; Sun, J. Chem. Commun. 2010, 46, 4668.
(18) Medgyes, A.; Farkas, E.; Liptak, A.; Pozsgay, V. Tetrahedron
1997, 53, 4159.
ASSOCIATED CONTENT
* Supporting Information
■
S
Schemes S1−S4, abbreviations, experimental protocols, 1H and
13C NMR spectra for new compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
(19) Medgyes, A.; Bajza, I.; Farkas, E.; Pozsgay, V.; Liptak, A. J.
Carbohydr. Chem. 2000, 19, 285.
(20) Walvoort, M. T.; Moggre, G. J.; Lodder, G.; Overkleeft, H. S.;
Codee, J. D.; van der Marel, G. A. J. Org. Chem. 2011, 76, 7301.
(21) Wiggins, L. F. J. Chem. Soc. 1944, 0, 522.
(22) Torres-Sanchez, M. I.; Zaccaria, C.; Buzzi, B.; Miglio, G.;
Lombardi, G.; Polito, L.; Russo, G.; Lay, L. Chem.Eur. J. 2007, 13,
6623.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: laurence.mulard@pasteur.fr.
Notes
The authors declare no competing financial interest.
(23) Fraser, R. R.; Swingle, R. B. Tetrahedron 1969, 23, 3469.
(24) Braverman, S.; Cherkinsky, M.; Kedrova, L.; Reiselman, A.
Tetrahedron Lett. 1999, 40, 3235.
(25) (a) Tsuge, O.; Kanemasa, S.; Matsuda, K. J. Org. Chem. 1984,
́ ́
49, 2688. (b) Aguilar-Moncayo, M.; Díaz-Perez, P.; García Fernandez,
J. M.; Ortiz Mellet, C.; García-Moreno, M. I. Tetrahedron 2012, 68,
ACKNOWLEDGMENTS
■
The authors thank C. Guerreiro (UCB) for the HPLC analyses
and F. Bonhomme (CNRS UMR 3523) for the HRMS spectra.
This work has received funding from the MENESR, France
(ENS Cachan, PhD fellowship to H.P.) and the European
Commission Seventh Framework Program (FP7/2007−2013)
under Grant Agreement No. 261472-STOPENTERICS.
681.
(26) (a) Speziale, A. J.; Smith, L. R. J. Am. Chem. Soc. 1962, 84, 1868.
(b) Huchel, U.; Tiwari, P.; Schmidt, R. R. J. Carbohydr. Chem. 2010,
29, 61.
́
(27) van den Bos, L. J.; Codee, J. D. C.; van der Toorn, J. C.; Boltje,
REFERENCES
(1) Niyogi, S. K. J. Microbiol. 2005, 43, 133.
■
T. J.; van Boom, J. H.; Overkleeft, H. S.; van der Marel, G. A. Org. Lett.
2004, 6, 2165.
(2) Barry, E. M.; Pasetti, M. F.; Sztein, M. B.; Fasano, A.; Kotloff, K.
L.; Levine, M. M. Nat. Rev. Gastroenterol. Hepatol. 2013, 10, 245.
(3) Qu, F.; Bao, C.; Chen, S.; Cui, E.; Guo, T.; Wang, H.; Zhang, J.;
Tang, Y. W.; Mao, Y. Diagn. Microbiol. Infect. Dis. 2012, 74, 166.
(4) Cohen, D.; Green, M. S.; Block, C.; Slepon, R.; Ofek, I. J. Clin.
Microbiol. 1991, 29, 386.
(5) Passwell, J. H.; Ashkenzi, S.; Banet-Levi, Y.; Ramon-Saraf, R.;
Farzam, N.; Lerner-Geva, L.; Even-Nir, H.; Yerushalmi, B.; Chu, C.;
Shiloach, J.; Robbins, J. B.; Schneerson, R. Vaccine 2010, 28, 2231.
(6) Phalipon, A.; Tanguy, M.; Grandjean, C.; Guerreiro, C.; Belot, F.;
Cohen, D.; Sansonetti, P. J.; Mulard, L. A. J. Immunol. 2009, 182, 2241.
(7) Belot, F.; Guerreiro, C.; Baleux, F.; Mulard, L. A. Chem.Eur. J.
2005, 11, 1625.
(28) Ronnols, J.; Burkhardt, A.; Cumpstey, I.; Widmalm, G. Eur. J.
̈
Org. Chem. 2012, 2012, 74.
(29) Bhatt, B.; Thomson, R. J.; von Itzstein, M. Bioorg. Med. Chem.
Lett. 2012, 22, 7623.
(30) Batta, G.; Liptak, A.; Schneerson, R.; Pozsgay, V. Carbohydr. Res.
1997, 305, 93.
(8) (a) Boutet, J.; Guerreiro, C.; Mulard, L. A. J. Org. Chem. 2009, 74,
2651. (b) Chassagne, P.; Fontana, C.; Guerreiro, C.; Gauthier, C.;
Phalipon, A.; Widmalm, G.; Mulard, L. A. Eur. J. Org. Chem. 2013,
2013, 4085. (c) Hargreaves, J.; Le Guen, Y.; Guerreiro, C.; Descroix,
K.; Mulard, L. A. Org. Biomol. Chem. 2014, DOI: 10.1039/
c4ob01200c.
(9) Kenne, L.; Lindberg, B.; Petersson, K.; Katzenellenbogen, E.;
Romanowska, E. Carbohydr. Res. 1980, 78, 119.
(10) Blatter, G.; Beau, J. M.; Jacquinet, J. C. Carbohydr. Res. 1994,
260, 189.
(11) Chassagne, P.; Raibaut, L.; Guerreiro, C.; Mulard, L. A.
Tetrahedron 2013, 69, 10337.
4895
dx.doi.org/10.1021/ol502395k | Org. Lett. 2014, 16, 4892−4895