A highly convergent approach for the synthesis of disaccharide repeating units of peptidoglycan

Article abstract of DOI:10.1016/S0040-4039(02)01753-7

A highly convergent strategy has been developed for the synthesis of the two possible saccharide-repeating units of peptidoglycan using TBDMS and PMB ethers as orthogonal protecting groups and a two-directional glycosylation strategy for the efficient ass

Full text of DOI:10.1016/S0040-4039(02)01753-7

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 7805–7807
A highly convergent approach for the synthesis of disaccharide
repeating units of peptidoglycan
Abhijit Roy Chowdhury, Aloysius Siriwardena and Geert-Jan Boons*
Complex Carbohydrate Research Center, University of Georgia, 220 Riverbend Road, Athens, GA 30602-4712, USA
Received 14 August 2002; accepted 21 August 2002
Abstract—A highly convergent strategy has been developed for the synthesis of the two possible saccharide-repeating units of
peptidoglycan using TBDMS and PMB ethers as orthogonal protecting groups and a two-directional glycosylation strategy for
the efficient assembly of the spacer containing disaccharides. © 2002 Elsevier Science Ltd. All rights reserved.
Septic shock resulting from gram-positive bacterial
infections has increased progressively over the last two
decades and currently causes one third of all cases of
sepsis.1,2 Gram-positive sepsis is often linked to a sys-
temic inflammatory response to peptidoglycan (PGN)
in the blood of infected patients. A major structural
component of the cell wall of gram-positive bacteria is
a large branched glycopeptide polymer (MW>125 000),
TNFa), which in turn cause the deleterious effects of
gram-positive sepsis. The first step in the production of
these mediators is the binding of PGN to the cluster
differentiation antigen CD14 on mononuclear phago-
cytes. The Toll-like receptor 2, a transmembrane recep-
tor protein, is also involved in the transmission of the
PGN signal to the interior of the cell.3,4
PGN that consists of alternating b(1-4)-linked
NAc and N-acetylmuramic acid (MurNAc, the 3-O-
lactyl ether of -GlcNAc) residues (Fig. 1) to which is
D-Glc-
Data from our laboratory indicate that activation of
CD14 and Toll-like receptor 2, leading to the produc-
tion of TNFa, requires a multivalent presentation of a
partial structure of PGN.5 This finding was established
with synthetic polymers functionalized with pendant
muramyl dipeptide moieties (MDP, a partial-structure
of the repeating unit of PGN, Fig. 1). This neogly-
copeptide polymer was shown to induce dose-depen-
dent production of TNFa by human monocytes,
whereas monomeric MDP had no effect. The activity of
the synthetic polyvalent MDP was, however, an order
of a magnitude smaller than that of PGN and we
conjectured that a more complete epitope might be
required for its optimal binding of CD14 and Toll like
receptor 2.
D
-
D
appended peptide chains. This backbone is inter-linked
by peptide bridges, thus forming an insoluble three-
dimensional network. PGN exerts its clinical effects by
initiating the production of multiple host-derived
inflammatory mediators (e.g. tumor necrosis factor;
As part of a program designed to clarify which struc-
tural elements of PGN are required for binding to
CD14 and Toll-like receptor 2 and thus the induction
of endogenous proinflammatory mediators, we report
here a highly convergent synthesis of the protected
PGN-repeating disaccharides 9 and 11. The lactate
moiety of these disaccharides can be coupled to pep-
tides and the artificial anomeric spacer allows conjuga-
tion of these units to an activated polymer to give
polyvalent ligands.
Figure 1. Structures of PGN and muramyldipeptide.
* Corresponding author. Tel.: +1-706-542-9161; fax: +1-706-542-4412;
e-mail: gjboons@ccrc.uga.edu
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01753-7

7806
A. Roy Chowdhury et al. / Tetrahedron Letters 43 (2002) 7805–7807
Contemporary oligosaccharide synthesis often makes
use of orthogonally protected saccharide building
blocks that can be assembled into complex structures
using efficient glycosylation strategies.6,7 It was
expected that the disaccharide 6, protected with orthog-
onally TBDMS and PMB ethers, would be a conve-
nient intermediate for the synthesis of both 9 and 11
(Scheme 1). The key disaccharide 6 could easily be
prepared from monosaccharide building blocks 2 and 5,
which could themselves be prepared from precursor 1.8
The spacer-containing saccharide 6 could be assembled
by a two-directional glycosylation strategy9,10 involving
coupling of the partially protected glycosyl donor 4
with 3-azido propanol to give 5, which in the next step
could be used as an acceptor in glycosylation with 2.
the benzylidene acetal function to give the glycosyl
donor 4 in a yield of 72%. NIS/TMSOTf-mediated
coupling14 of partially protected glycosyl donor 4 with
the highly reactive acceptor 3-azido-propanol gave 5 in
good yield. Due to the low reactivity of the C-4
hydroxyl of 4, no self-condensation was observed.
Compound 5 could immediately be used as a glycosyl
acceptor in a NIS/TMSOTf-mediated coupling with
glycosyl donor 2 to give disaccharide 6 in a yield of
62%.
In an earlier attempt to obtain an orthogonally pro-
tected chitibiose derivative, we employed the combina-
tion of allyl- and p-methoxybenzyl ethers. However,
NIS/TMSOTf mediated coupling of 5 with a derivative
of 2 that had an allyl in lieu of the TBDMS ether gave
substantial quantities of a disaccharide that had its allyl
moiety iodinated. Several other promoters were investi-
gated but each was incompatible with the allyl ether.
The C-3 hydroxyl of key intermediate 1 was converted
into a TBDMS ether by reaction with TBDMSOTf in
presence of 2,6-lutidine to give glycosyl donor 2 in a
yield of 96%.11 Conventional silylation conditions such
as TBDMSCl and imidazole resulted only in recovery
of starting material.12 Compound 1 was also protected
with a PMB group using NaH, PMBCl in THF to give
3 in a good yield of 84%. Treatment of 3 with
Removal of the phthalimido functions in disaccharide 6
using ethylene diamine in ethanol15,16 followed by
acetylation of the revealed amino groups with acetic
anhydride in methanol gave derivative 7. n-Tetra-
13
BH3·Me3N, and AlCl3 led to regioselective opening of
butylammonium fluoride (TBAF) treatment of 7
Scheme 1. Reagents and conditions. (i) TBDMSOTf, 2,6-lutidine, DCM; (ii) PMBCl, NaH, TBAI, THF, D; (iii) Me3N·BH3, AlCl3,
THF; (iv) HO(CH2)3N3, NIS, TMSOTf; (v) NIS, TMSOTf; (vi) H2N(CH2)2NH2, EtOH then Ac2O, MeOH; (vii) Bu4NF, THF;
(viii) NaH, S-2-chloropropionic acid, dioxane; (ix) DDQ, DCM, H2O.

A. Roy Chowdhury et al. / Tetrahedron Letters 43 (2002) 7805–7807
7807
resulted in clean removal of the TBDMS ether to give
8. Several conditions were explored to install a -lac-
plas, M.; Moore, J. N.; Boons, G.-J. J. Am. Chem. Soc.
2001, 123, 8145–8146.
6. Boons, G.-J. Tetrahedron 1996, 52, 1095–1121.
7. Wong, C.-H.; Ye, X.; Zhang, Z. J. Am. Chem. Soc. 1998,
120, 7137–7138.
8. Lo¨nn, H. Carbohydr. Res. 1985, 139, 105–114.
9. Zhu, T.; Boons, G.-J. Angew. Chem., Int. Ed. 1999, 38,
1629–1632.
10. Zhu, T.; Boons, G.-J. Tetrahedron Lett. 1998, 39, 2187–
2190.
D
tate moiety at C-3% of 8 but the best results were
obtained with NaH and S-2-chloropropionic acid in
1,4-dioxan to give 9 in a yield of 76%.17,18 Alternatively,
disaccharide 7 could serve in the synthesis of the target
compound 11 by removal of the PMB ether using DDQ
in dichloromethane/water19 to give 10 which was
treated with NaH and S-2-chloropropionic acid.
Recently, considerable attention has been focussed on
the development of efficient procedures for the synthe-
sis of the repeating unit of PGN.20–22 A major synthetic
obstacle was lactonization of muramic acid derivatives
that have a free C-4 hydroxyl. In one approach,22 this
11. Corey, E. J.; Cho, H.; Rucker, C.; Hua, D. H. Tetra-
hedron Lett. 1981, 22, 3455–3458.
12. Corey, E. J.; Venkateswarlu, A. J. Am. Chem. Soc. 1972,
94, 6190–6191.
13. Monica, E. K.; Garegg, P. J.; Hultberg, H.; Oscarson, S.
J. Org. Chem. 1983, 2, 305–311.
14. Veeneman, G. H.; Van Leeuwen, S. H.; Van Boom, J. H.
problem was resolved by first installing L-alanine, the
first amino acid residue of the muramyl peptide chain,
followed by conversion into a glycosyl acceptor for the
synthesis of a disaccharide. We addressed this problem
by first synthesizing a disaccharide and thereafter incor-
porating the lactate arm. This approach has the advan-
tage that one disaccharide can be used for the synthesis
of both possible repeating units of PGN (MurNAc-Glc-
NAc and GlcNAc-MurNAc). Alternative synthetic
approaches have employed 2-amino-2-deoxyglucopy-
ranosyl donors protected by a N-trichloroethoxycar-
bonyl (Troc) group. The rationale for the use of a
N-Troc group is that the corresponding glycosyl donors
are 40 times more reactive than a corresponding phthal-
imido protected derivative23 and moreover that it
directs a glycosylation to give exclusively a 1,2-trans
linked product by neighboring group participation.
However, we found that glycosyl donors that were
protected with phthalimido groups led to high yielding
glycosylations. More importantly, the phthalimido
group is compatible with the rather strong basic condi-
tions used for the incorporation of the lactate moiety
whereas the Troc would not survive these conditions.
Tetrahedron Lett. 1990, 31, 1331–1334.
15. Kanie, O.; Crawley, S. C.; Palcic, M. M.; Hindsgaul, O.
Carbohydr. Res. 1993, 243, 139–164.
16. Debenham, J. S.; Debenham, S. D.; Fraser-Reid, B.
Bioorg. Med. Chem. 1996, 4, 1909–1918.
17. Hasegawa, A.; Kaneda, Y.; Yasuhiko, G.; Nishibori, K.;
Kiso, M. Carbohydr. Res. 1981, 94, 143–163.
18. All new compounds gave satisfactory NMR spectro-
scopic, mass spectroscopic data. Selected data for com-
pound 9: Colorless syrup; Rf 0.36 (MeOH: DCM, 8:92,
v/v) MALDI-TOF m/z=900.6 [M+Na]+, 916.4 [M+K]+.
1
Selected signals in H NMR (CDCl3, 500 MHz) l 5.5 (s,
1H, PhCH), 4.2 (d, 3H, H-1, J=7.79 Hz), 2.1 (s, 3H,
NHCOCH3), 2.0 (s, 3H, NHCOCH3), 1.8 (m, 2H,
OCH2CH2CH2N3),
1.4
(d,
3H,
J=6.79
Hz,
(CH(CH3)COOH). 13C NMR (CDCl3, 75 MHz) l 101.3
(PhCH), 100.9, 100.7 (C-1, C-1%), 29.0 (OCH2CH2-
CH2N3), 23.5 (NHCOCH3), 22.7 (NHCOCH3), 18.7
(CH(CH3)COOH). Compound 11: Colorless syrup; Rf
0.21 (MeOH: DCM, 8:92, v/v) MALDI-TOF m/z=895.5
[M+Na]+. 1H NMR (CD3OD, 500 MHz) l 5.3 (s, 1H,
PhCH), 5.0 (dd, 2H, J=11.7 Hz, H-1, H-1%), 1.9 (s, 3H,
NHCOCH3), 1.9 (s, 3H, NHCOCH3), 1.7 (m, 2H,
OCH2CH2CH2N3), 1.1 (d, 3H, J=6.83 Hz, (CH(CH3)-
COOH). 13C NMR (CD3OD, 75 MHz) l 101.1 (C-1,
C-1%), 29.2 (CH(CH3)COOH), 23.8 (NHCOCH3), 22.0
(OCH2CH2CH2N3), 21.8 (NHCOCH3).
Acknowledgements
This research was supported by the NIH (GM61761)
and the American Heart Association (0051229B).
19. Horita, K.; Yoshioka, T.; Tanaka, T.; Oikawa, Y.; Yone-
mitsu, O. Tetrahedron 1986, 42, 3021–3028.
20. Inamura, S.; Fukase, K.; Kusumoto, S. Tetrahedron Lett.
2001, 42, 7613–7616.
21. VanNieuwenhze, M. S.; Mauldin, S. C.; Zia-Ebrahimi,
M.; Aikins, J. A.; Blaszczak, L. C. J. Am. Chem. Soc.
2001, 123, 6983–6988.
22. Saha, S. L.; Nieuwenhze, S. V.; Hornback, W. J.; Aikins,
J. A.; Blaszczak, L. C. Org. Lett. 2001, 3, 3575–3577.
23. Zhang, Z.; Ollmann, I. R.; Ye, X.; Wischnat, R.; Baasov,
T.; Wong, C.-H. J. Am. Chem. Soc. 1999, 121, 734–753.
References
1. Bone, R. C. Arch. Intern. Med. 1994, 154, 26–34.
2. Nogare, A. R. Am. J. Med. Sci. 1994, 302, 50–65.
3. Aderem, A.; Ulevitch, R. J. Nature 2000, 406, 782–786.
4. Mantovani, A.; Muzio, M. Microbes Infection 2000, 2,
251–255.
5. Siriwardena, A.; Jorgensen, M.; Wolfert, M.; Vanden-