Synthesis of the repeating unit of the O-specific polysaccharide of Shigella sonnei and quantitation of its serologic activity

Article abstract of DOI:10.1016/S0960-894X(99)00585-5

The chemical synthesis of the zwitterionic disaccharide 2 is described that corresponds to the repeating unit of the O-specific polysaccharide (1) of the Gram-negative human pathogen Shigella sonnei. Passive hemolysis inhibition tests using a hyperimmune

Full text of DOI:10.1016/S0960-894X(99)00585-5

Bioorganic & Medicinal Chemistry Letters 10 (2000) 19±21
Synthesis of the Repeating Unit of the O-speci®c Polysaccharide
of Shigella sonnei and Quantitation of its Serologic Activity
Aniko Toth, a Adel Medgyes, a Istvan Bajza, a Andras Liptak, a,*
Gyula Batta, b Tivadar Kontrohr, c Klara Peter€y c and Vince Pozsgay d
aResearch Group for Carbohydrates of the Hungarian Academy of Sciences and Institute of Biochemistry, Lajos Kossuth University,
Debrecen, Hungary, H-4010
bResearch Group for Antibiotics of the Hungarian Academy of Sciences, Debrecen, Hungary
c
Â
Institute of Microbiology, University Medical School, Pecs, Hungary
dLaboratory of Developmental and Molecular Immunity, National Institute of Child Health and Human Development,
National Institutes of Health, Bethesda, MD, 20892, USA
Received 11 August 1999; accepted 6 October 1999
AbstractÐThe chemical synthesis of the zwitterionic disaccharide 2 is described that corresponds to the repeating unit of the O-
speci®c polysaccharide (1) of the Gram-negative human pathogen Shigella sonnei. Passive hemolysis inhibition tests using a hyper-
immune rabbit serum raised against S. sonnei showed that the serologic activity of the disaccharide 2 is nearly 2- to 3-fold higher
than those of its component monosaccharides. NMR data of 2 are in support of the proposed structure of the O-speci®c poly-
saccharide. # 1999 Elsevier Science Ltd. All rights reserved.
Shigella sonnei is a Gram-negative bacterium that can
cause diarrhea and dysentery in humans. A major cell
surface component of this bacterium is its O-speci®c
polysaccharide1,2 (O-SP) that is the outermost, sero-
determinant domain of the highly complex lipopoly-
saccharide. The O-SP is an essential virulence factor of
Shigellae including S. sonnei: only strains that have
their O-SP fully expressed are virulent. Additionally, the
O-SP is related to host immunity in that protection
against infection is a correlate of the IgG antibody level
against the O-SP.3 The O-SP of S. sonnei is a non-
immunogenic molecule (MW ꢀ25 kDa) that must be
conjugated to an immunogenic protein carrier to induce
antibodies. Clinical trials have con®rmed that protein
conjugates of the O-SP elicited IgG levels in adult
volunteers that were similar to those in patients con-
valescent from shigellosis.4 The O-SP of S. sonnei is
composed of the zwitterionic disaccharide repeating
unit 1 consisting of the rare sugars 2,4,6-trideoxy-2-
acetamino-4-amino-d-galactose and 2-acetamido-2-
deoxy-l-altruronic acid that are connected through 1,2-
trans interglycosidic linkages.1,2
Our approach to vaccine development is based on the
assumption that fragments of O-SPs mimic epitopes of
the native polymers and thus may induce poly-
saccharide-speci®c antibodies when conjugated to an
immunogenic carrier. Recent data con®rmed this
assumption and showed that albumin conjugates of
chemically synthesized extended oligosaccharide frag-
ments5,6 of the O-SP of S. dysenteriae type 1 elicited higher
levels of the homologous O-SP-speci®c IgG antibodies
than did the conjugate of the native polysaccharide.7
As a prelude to chemical synthesis of oligosaccharides
corresponding to 1 we have reported the synthesis of the
methyl glycosides of its component monosaccharides in
their natural glycosidic con®guration.8 We have also
presented evidence that in the native polysaccharide
both pyranose residues are preferentially in the 4C1
chair conformation and that the polysaccharide exists in
the zwitterionic form.9 Here we describe the ®rst synth-
esis of a disaccharide part (2) of the O-SP that to our
knowledge is also the ®rst report of a zwitterionic dis-
accharide containing free amino and carboxyl groups at
alternating monosaccharide residues. The disaccharide
2 was constructed from an appropriately protected
altruronic acid derivative (5) that was used as the
acceptor and from the activated donor 9. The common
*Corresponding author. Tel.: +36-52-316-666; fax: +36-52-512-913;
e-mail: liptaka@tigris.klte.hu
0960-894X/00/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(99)00585-5

Â
A. Toth et al. / Bioorg. Med. Chem. Lett. 10 (2000) 19±21
20
feature of both intermediates is the presence of a tri-
chloroacetyl protecting group at their respective amino
function. This protecting group has previously been
shown to be suitable for N-protection without adverse
outcomes during glycosylation reactions.10 It was also
demonstrated to have good trans-directing e€ect and to
provide the substrates with acceptable solubility prop-
erties. The synthetic steps to the key donor and acceptor
intermediates are described next.
yield. Next, a trichloroacetyl group was installed at the
free amino group by exposure of 7 to trichloroacetyl
chloride in the presence of triethylamine (!8). The
synthesis of the glycosyl donor 9 was completed by
acetylation of 8 with pyridine/acetic anhydride, in 94%
overall yield for two steps.
Having readied the donor (9) and the acceptor moieties
(5), the fully protected disaccharide 10 was prepared by
NIS/TfOH-promoted activation11,12 of the thioglycoside
9 in the presence of 0.55 equimolar amount of the
acceptor 5 in 85% yield (Scheme 3). Replacement of the
N-trichloroacetyl groups with the acetyl groups was
previously described through Bu3SnH-mediated reduc-
tive cleavage of the chlorine atoms.10 This approach
failed to give satisfactory results when applied to 10.
Fortunately, treatment of 10 with NaOH in MeOH
excellently removed the N-trichloroacetyl groups to
a€ord an intermediate that was converted to the N-
acetamido derivative 11 by conventional acetylation, in
72% overall yield for two steps. As the ®nal phase in
the synthetic sequence, compound 11 was subjected to
The altruronic acid residue 5 (Scheme 1). The synthetic
route started from the glycoside 3 that was available
from an earlier work.8 In the ®rst step the azido group
in compound 3 was reduced to an amino group with H2
over Pd/C to a€ord 4. Hydrogenolytic cleavage of the
O-benzyl group in this step was prevented by the use of
one equivalent of pyridine. Subsequent treatment of 4 with
trichloroacetyl chloride a€orded 5 in 56% overall yield.
The trideoxy-galactose residue 9 (Scheme 2). Starting
compound was thioglycoside8 6 that was converted to
the amine 7 by treatment with ethylenediamine, in 94%
Scheme 1. Reagents and conditions: (a) H2/Pd-C, MeOH, C5H5N, 25 ꢁC, 2 h 81%; (b) 1.1. equiv of Cl3CCOCl, Et3N, CH2Cl2, 0 ꢁC, 1 h 70%.
Scheme 2. Reagents and conditions: (a) NH2CH2CH2NH2 (excess), EtOH, re¯ux, 2 h, 94%; (b) 1.2 equiv of Cl3CCOCl, Et3N, CH2Cl2, 0 ꢁC, 20 min;
(c) C5H5N, Ac2O, 25 ꢁC, 2 h, 94% for two steps.
Scheme 3. Reagents and conditions; (a) 1.8 equiv of 9, 2.3 equiv of NIS, TfOH (cat), CH2Cl2, 30 ꢁC, 1.5 h, 85%; (b) aq NaOH, MeOH, 25 ꢁC,
2.5 days; (c) Ac2O, MeOH, 0 ꢁC, 50 min, 72% for two steps; (d) H2/Pd(OH)2-C, EtOH, AcOH 25 ꢁC, 4 days, 51%.

Â
A. Toth et al. / Bioorg. Med. Chem. Lett. 10 (2000) 19±21
21
Table 1. Inhibition of hemolysis of Shuigella sonnei lipopolysacchar-
ide-sensitized erthrocytes by the lipopolysaccharide-related antigens 2,
13 and 14a
of Shigella sonnei and presented spectroscopic and
immunochemical support for the structure of the native
polysaccharide. Current work is directed at the synthesis
and immunochemical study of the frame-shifted, alter-
native disaccharide repeating unit and higher-membered
saccharides corresponding to the O-speci®c poly-
saccharide of S. sonnei for their eventual use as vaccine-
components against enteric diseases.
Antigen
IC50 (mM)b
2
13
14
3.9
6.9
9.8
aIC50: 50% inhibition concentraton.
bEstimated range of error: ꢂ10%.
Acknowledgements
This work was supported by an International Reseach
Scholar's award from the Howard Hughes Medical
Institute and by the Mizutani Foundation for Gly-
coscience. Generous support from Bolyai Postdoctoral
Scholarship to I. Bajza is also acknowledged.
hydrogenolytic cleavage of the O-benzyl groups that
simultaneously reduced the azido group to a€ord the
targeted disaccharide 2.13
1
A comparison of the H and 13C NMR chemical shifts
for compounds 1,9 and 2 shows an excellent agreement
of the corresponding resonances that lends further sup-
port for the absolute con®guration of the altruronic
residue as being l and for the trideoxy-galactose residue
as being d.14 For a comparison, we have also synthe-
sized the diastereomeric disaccharide 12 that contains
the enantiomeric, d altruronic acid moiety.15 For com-
pound 12 the chemical shift of the C-5 carbon atom of
the uronic acid residue is 70.5 ppm, distinctly di€erent
from the corresponding resonances in 1 and 2 (78.4 and
76.9, respectively). Minor di€erences in the chemical
shifts for residues A in 1 and 2 are likely due to con-
formational di€erences and indicate that the dis-
accharide can only partially express conformational
features of polysaccharide 1.
References and Notes
1. Kontrohr, T. Carbohydr. Res. 1977, 58, 498.
2. Kenne, L.; Lindberg, B.; Petersson, K.; Katzenellenbogen,
E.; Romanowska, E. Carbohydr. Res. 1980, 78, 119.
3. Robbins, J. B.; Schneerson, R. J. Inf. Dis. 1990, 161, 821.
4. Ashkenazi, S.; Passwell, J. H.; Harlev, E.; Miron, D.;
Dagan, R.; Farzan, N.; Ramon, R.; Majadly, F.; Bryla, D. A.;
Karpas, A. B.; Robbins, J. B.; Schneerson, R. J. Infect. Dis.
1999, 179, 1565 and references therein.
5. Pozsgay, V. Angew. Chem., Int. Ed. 1998, 37, 138.
6. Pozsgay, V. J. Org. Chem. 1998, 63, 5983.
7. Pozsgay, V.; Chu, C.; Pannell, L.; Wolfe, J.; Robbins, J. B.;
Schneerson, R. Proc. Natl. Acad. Sci. USA 1999, 96, 5194.
8. Medgyes, A.; Farkas, E.; Liptak, A.; Pozsgay, V. Tetra-
hedron 1997, 53, 4159.
9. Batta, Gy.; Liptak, A.; Schneerson, R.; Pozsgay, V. Carbo-
hydr. Res. 1997, 305, 93.
10. Blatter, G.; Jacquinet, J. C. Carbohydr. Res. 1996, 288, 109.
11. Konradsson, P.; Udodong, U. E.; Fraser-Reid, B. Tetra-
hedron Lett. 1990, 31, 4313.
The antigenicity of the disaccharide 2 was assayed
by the passive hemolysis inhibition test as described
by Kontrohr and Peter€y.16 The method is based on
complement-mediated hemolysis of lipopolysaccharide-
sensitized erythrocytes after anti-LPS antibody binding.
Using sheep erythrocytes, guinea pig complement,
phase I lipopolysaccharide of S. sonnei,1 and hyper-
immune polyclonal rabbit serum raised against S. son-
nei, the concentration of the disaccharide 2 and those
of the component monosaccharides 13 and 14 were
determined that were necessary for 50% inhibition of
hemolysis. The results (Table 1) show that the dis-
accharide 2 is a better inhibitor than either one of its
monosaccharide components of which the altruronic
acid derivative 13 is superior. Less than 20% inhibition
could be observed with the unnatural disaccharide 12 up
to 15 mM concentration and no inhibition was seen
with unrelated saccharides.
12. Veeneman, G. H.; van Leeuwen, S. H.; van Boom, J. H.
Tetrahedron Lett. 1990, 31, 1331.
13. All new compounds had satisfactory analytical, 1H,13C
1
NMR, and mass spectral data. For 2 H NMR(D2O): d 4.73
(H-1B), 4.70 (H-1A), 4.55 (H-5A), 4.42 (H-4A), 3.90 (H-3B),
3.88 (H-2A), 3.88 (H-5B), 3.81 (H-2B), 3.70 (H-3A), 1.31 (H-
6B); 13C NMR: d 102.9 (C-1B), 100.3 (C-1A), 77.9 (C-4A), 76.9
(C-5A), 69.3 (C-3B), 69.2 (C-3A, C-5B), 54.6 (C-4B), 52.8 (C-
2B), 52.1 (C-2A), 16.3 (C-6B).
14. We had no speci®c reason to question the reported struc-
ture of the O-SP of S. sonnei. However, in the light of numer-
ous structural revisions of bacterial polysaccharides it is
always advisable to obtain independent structural proof for
these materials before undertaking a synthetic project towards
their extended portions.
15. Medgyes, A.; Bajza, I.; Farkas, E.; Pozsgay, V.; Liptak, A.
J. Carbohydr. Chem. 1999, submitted.
In summary, we have synthesized for the ®rst time the
complete repeating unit of the O-speci®c polysaccharide
16. Kontrohr, T.; Peter€y, K. J. Immunol. Methods 1976, 13,
271.