De novo synthesis of a 2-acetamido-4amino-2,4,6-trideoxy-d-galactose (AAT) building block for the preparation of a bacteroides fragilis a1 polysaccharide fragment

Article abstract of DOI:10.1021/ol1003912

(Figure Presented) Zwitterionic polysaccharides (ZPSs) are potent T-cell activators that naturally occur on the cell surface of bacteria and show potential as immunostimulatory agents. An unusual, yet important component of many ZPSs is 2-acetamido-4-amino-2,4,6-trideoxy-D-galactose (AAT). AAT building block 2 was prepared via a de novo synthesis from N-Cbz-L-threonine 5. Furthermore, building block 2 was used to synthesize disaccharide 15 that constitutes a fragment of zwitterionic polysaccharide A1 (PS A1) found in Bacteroldes fragilis.

Full text of DOI:10.1021/ol1003912

ORGANIC
LETTERS
2010
Vol. 12, No. 7
1624-1627
De Novo Synthesis of a 2-Acetamido-4-
amino-2,4,6-trideoxy-D-galactose (AAT)
Building Block for the Preparation of a
Bacteroides fragilis A1 Polysaccharide
Fragment
Rajan Pragani, Pierre Stallforth, and Peter H. Seeberger*
Max-Planck-Institute of Colloids and Interfaces, Department of Biomolecular Systems,
Am Mu¨hlenberg 1, 14476 Potsdam, Germany, and Freie UniVersita¨t Berlin, Institute
for Chemistry and Biochemistry, Arnimallee 22, 14195 Berlin, Germany
peter.seeberger@mpikg.mpg.de
Received February 15, 2010
ABSTRACT
Zwitterionic polysaccharides (ZPSs) are potent T-cell activators that naturally occur on the cell surface of bacteria and show potential as
immunostimulatory agents. An unusual, yet important component of many ZPSs is 2-acetamido-4-amino-2,4,6-trideoxy-D-galactose (AAT). AAT
building block 2 was prepared via a de novo synthesis from N-Cbz-L-threonine 5. Furthermore, building block 2 was used to synthesize
disaccharide 15 that constitutes a fragment of zwitterionic polysaccharide A1 (PS A1) found in Bacteroides fragilis.
Access to synthetic carbohydrates drives our understanding
of molecular glycobiology.1 Although many monosaccharide
building blocks are readily accessed by protection of com-
mercially available sugars, monosaccharides specific to many
bacteria are only available via lenghty synthetic routes. As
a result, de novo synthesis2,3 of unusual, biologically
important monsaccharide building blocks from linear frag-
ments in turn derived from the chiral pool has emerged as
an alternative to access carbohydrates.
One particular monosaccharide, 2-acetamido-4-amino-
2,4,6-trideoxy-D-galactose (AAT), is an important component
of zwitterionic polysaccharides (ZPSs) that are found on the
cell surface of bacteria. For example, AAT is part of the
lipopolysaccharide of Shigella sonnei,4 on the capsules of
Streptococcus pneumoniae5 and Bacteroides fragilis,6 and
on the cell-wall polysaccharide of Streptococcus mitis7 and
(1) For reviews on recent advances in glycobiology, see: (a) Varki, A.
Glycobiology 1993, 3, 97. (b) Dube, D. H.; Bertozzi, C. R. Nat. ReV. Drug
DiscoVery 2005, 4, 477. (c) Stallforth, P.; Lepenies, B.; Adibekian, A.;
Seeberger, P. H. J. Med. Chem. 2009, 52, 5561.
(2) For reviews on de novo carbohydrate synthesis, see: (a) Schmidt,
R. R. Pure Appl. Chem. 1987, 59, 415. (b) Kirschning, A.; Jesberger, M.;
Scho¨ning, K.-U. Synthesis 2001, 4, 507. (c) Hemeon, I.; Bennet, A. J.
Synthesis 2007, 13, 1899. (d) Vogel, P. In The Organic Chemistry of Sugars;
Levy, D. E., Fugedi, P., Eds.; Taylor and Francis Group/CRC Press: Boca
(4) Kenne, L.; Lindberg, B.; Petersson, K.; Katzenellenbogen, E.;
Romanowska, E. Carbohydr. Res. 1980, 78, 119.
(5) (a) Lindberg, B.; Lindqvist, B.; Lo¨nngren, J.; Powell, D. A.
Carbohydr. Res. 1980, 78, 111. (b) Stroop, C. J. M.; Xu, Q.; Retzlaff, M.;
Abeygunawardana, C.; Bush, C. A. Carbohydr. Res. 2002, 337, 335.
(6) Baumann, H.; Tzianabos, A. O.; Brisson, J.-R.; Kasper, D. L.;
Jennings, H. J. Biochemistry 1992, 31, 4081.
Raton, FL, 2006; Chapter 13, p 629
.
(3) For selected examples of recent de novo carbohydrate synthesis, see:
(a) Northrup, A. B.; MacMillan, D. W. C. Science 2004, 305, 1752. (b)
Babu, R. S.; Zhou, M.; O’Doherty, G. A. J. Am. Chem. Soc. 2004, 126,
(7) Bergstro¨m, N.; Jansson, P.-E.; Kilian, M.; Sørensen, U. B. S. Eur.
J. Biochem. 2000, 267, 7147.
3428. (c) Enders, D.; Grondal, C. Angew. Chem., Int. Ed. 2005, 44, 1210
.
10.1021/ol1003912  2010 American Chemical Society
Published on Web 03/02/2010

S. pneumoniae.8 Zwitterionic polysaccharide A1 (PS A1 (1);
Figure 1) found in B. fragilis is the putative cause of
postoperative intra-abdominal abscesses in humans and has
been shown to activate a MHCII-mediated T-cell response
in the absence of proteins.9 As a potential immunostimulant,
PS A1 when conjugated to tumor-associated antigen Tn (D-
GalNAc) has been shown to promote a T-cell-dependent
immune response against Tn in mice.10 Due to the im-
munological properties of PS A1 and related ZPSs,9-11 the
development of efficient syntheses for AAT building blocks
in order to access homogeneous ZPS fragments for im-
munological study remains important.
equatorial azide via an azido nitration reaction of glycal 3.
The carbon skeleton of this glycal could be generated by a
Dieckmann cyclization of acetate 4 that is derived in two
steps from N-Cbz-L-threonine.
Scheme 2. Synthesis of Glycal 3
The synthesis of glycal 3 commenced with acid mediated
methyl ester formation of N-Cbz-L-threonine 5 (Scheme 2).
The resulting alcohol was acetylated to furnish acetate 4 in
95% yield over two steps. Two equivalents15 of LHMDS
were used to deprotonate the acetate and NHCbz groups to
induce a Dieckmann cyclization.16 The crude ꢀ-ketoester was
methylated with K2CO3/Me2SO4 to provide enone 6 in 73%
yield over two steps. Methoxy enone 6 was then reduced
with DIBAL in a 1,2 manner and upon acidic workup,
rearranged to an intermediate enone.17 Luche reduction18 of
the crude enone at -78 °C gave glycal 7 in 77% yield over
two steps. The free hydroxyl group was finally protected as
an acetate ester to afford glycal 3 in 90% yield.
Figure 1. B. fragilis zwitterionic polysaccharide A1 (1).
Previous syntheses12 of AAT building blocks originate
either from glucosamine or mannose and are quite lengthy.
Shorter syntheses13 provide building blocks with a participa-
tory C2-nitrogen protecting group and thus cannot be used
to form R-linked AAT disaccharides such as the one found
in PS A1. We felt that a de novo approach14 would reduce
the number of steps required to generate a properly func-
tionalized AAT building block for PS A1. Here we report a
de novo synthesis of an AAT glycosylating agent, and its
use in the assembly of a PS A1 disaccharide fragment.
At this stage, azido functionalization of glycal 3 was
attempted. Azido nitration12b,19 generated azido nitrate 8 in
67% yield as an inseparable 3.5:1 C2-diastereomeric mixture
that favored the desired equatorial azide (Table 1, entry 1).
Scheme 1. Retrosynthetic Analysis of AAT Building Block 2
(12) (a) Hermans, J. P. G.; Elie, C. J. J.; van der Marel, G. A.; van
Boom, J. H. J. Carbohydr. Chem. 1987, 6, 451. (b) Smid, P.; Jo¨rning,
W. P. A.; van Duuren, A. M. G.; Boons, G. J. P. H.; van der Marel, G. A.;
van Boom, J. H. J. Carbohydr. Chem. 1992, 11, 849. (c) van den Bos,
L. J.; Boltje, T. J.; Provoost, T.; Mazurek, J.; Overkleeft, H. S.; van der
Marel, G. A. Tetrahedron Lett. 2007, 48, 2697.
(13) (a) Liav, A.; Jacobson, I.; Sheinblatt, M.; Sharon, N. Carbohydr.
Res. 1978, 66, 95. (b) Lo¨nn, H.; Lo¨nngren, J. Carbohydr. Res. 1984, 132,
39. (c) Medgyes, A.; Farkas, E.; Lipta´k, A.; Pozsgay, V. Tetrahedron 1997,
53, 4159. (d) Liang, H; Grindley, T. B. J. Carbohydr. Chem. 2004, 23, 71.
(e) Cai, Y.; Ling, C.-C.; Bundle, D. R. J. Org. Chem. 2009, 74, 580.
(14) For recent de novo carbohydrate syntheses from our laboratory,
see: (a) Timmer, M. S. M.; Adibekian, A.; Seeberger, P. H. Angew. Chem.,
Int. Ed. 2005, 44, 7605. (b) Adibekian, A.; Bindscha¨dler, P.; Timmer,
M. S. M.; Noti, C.; Schu¨tzenmeister, N.; Seeberger, P. H. Chem.sEur. J.
2007, 13, 4510. (c) Stallforth, P.; Adibekian, A.; Seeberger, P. H. Org.
Lett. 2008, 10, 1573. (d) Adibekian, A.; Timmer, M. S. M.; Stallforth, P.;
van Rijn, J.; Werz, D.; Seeberger, P. H. Chem. Commun. 2008, 3549.
(15) Ge, P.; Kirk, K. L. J. Org. Chem. 1996, 61, 8671.
In the retrosynthetic analysis of AAT building block 2
(Scheme 1), we envisioned the installation of the C2-
(8) Karlsson, C.; Jansson, P.-E.; Sørensen, U. B. S. Eur. J. Biochem.
1999, 265, 1091, and references cited within.
(9) For selected reviews and studies on zwitterionic polysaccharides,
see: (a) Tzianabos, A.; Wang, J. Y.; Kasper, D. L. Carbohydr. Res. 2003,
338, 2531. (b) Mazmanian, S. K.; Kasper, D. L. Nat. ReV. Immunol. 2006,
6, 849. (c) Mazmanian, S. K.; Round, J. L.; Kasper, D. L. Nature 2008,
453, 620.
(16) Ren, F.; Hogan, P. C.; Anderson, A. J.; Myers, A. G. Org. Lett.
2007, 9, 1923.
(10) De Silva, R. A.; Wang, Q.; Chidley, T.; Appulage, D. K.; Andreana,
P. R. J. Am. Chem. Soc. 2009, 131, 9622.
(17) Kocienski, P.; Narquizian, R.; Raubo, P.; Smith, C.; Farrugia, L. J.;
Muir, K.; Boyle, F. T. J. Chem. Soc., Perkin Trans. 1 2000, 2357.
(18) Luche, J.-L. J. Am. Chem. Soc. 1978, 100, 2226.
(11) For a review on cell-wall glycopolymers, see: Weidenmaier, C.;
Peschel, A. Nat. ReV. Microbiol. 2008, 6, 276
.
(19) Lemieux, R. U.; Ratcliffe, R. M. Can. J. Chem. 1979, 57, 1244.
Org. Lett., Vol. 12, No. 7, 2010
1625

creating this glycosidic linkage. Specifically, a complex C4-
OH galactosamine nucleophile performed poorly in glyco-
sylations with an AAT thioglycoside (activated with Ph2SO/
Tf2O or NIS/TMSOTf) and lactol (activated with Ph2SO/
Tf2O). Thus, a variety of galactosamine nucleophiles (12-14)
were prepared from known alcohol 1124 (which was prepared
in eight steps from D-galactose; see Supporting Information).
In particular, alcohol 11 was first protected with either a Lev,
Fmoc, or Nap protecting group (Scheme 4). Then, the
benzylidene acetal of each intermediate was reduced with
TFA/TES to afford alcohols 12-14 in good yields.
Table 1. Azido Functionalization of Glycal 3
entry
conditions
product yield, % C2 dra
1
2
CAN, NaN3, CH3CN, -20 °C
PhI(OAc)2, TMSN3, Ph2Se2,
CH2Cl2, -25 °C
8
67
68
3.5:1b
1:1b
9
a Ratios determined by 1H NMR. b Diastereomers could not be separated
by silica gel chromatography; major isomer shown.
Scheme 4. Synthesis of Galactosamine Nucleophiles 12-14
Surprisingly, the azido selenation20 of glycal 3 furnished
selenide 9 with poor stereoselectivity (1:1 dr), regarding the
C2-position (Table 1, entry 2). This result stands in stark
contrast to the high equatorial azide selectivity for the azido
selenation of the structurally related di-O-acetyl fucal.21
Two AAT building blocks were synthesized from nitrate
8 (Scheme 3). Glycosyl trichloroacetimidate 10 was prepared
by selective hydrolysis of the anomeric nitrate22 followed
by K2CO3 mediated imidate formation. Glycosyl N-phenyl
trifluoroacetimidate 2 was also prepared since recent studies23
have shown that the N-phenyl trifluoroimidate leaving group
performs well for armed, deoxyhexose glycosylating agents.
Similar to the preparation of trichloroacetimidate 10, the
anomeric nitrate in 8 was removed followed by Cs2CO3
mediated imidate formation to afford 2.
Glycosylations between AAT building blocks (2 and 10)
and nucleophiles (12-14) were then studied by employing
standard activation conditions. Unfortunately, alcohol 12 did
not react with trichloroacetimidate 10 to form desired
disaccharide 15 (Table 2, entries 1 and 2). Although 10 has
been used in a successful R-selective glycosylation with a
C4-OH galactosamine nucleophile,12b the galactosamine was
derivatized to place the hydroxyl group in an equatorial
position, which rendered it more nucleophilic.
Scheme 3. Synthesis of AAT Building Blocks 2 and 10
It was hoped that the attenuated reactivity of N-phenyl
trifluoroacetimidate 2 compared to trichloroacetimidate 10
would slow the rate of building block decomposition and
allow for a successful glycosylation reaction. Initial attempts
to activate N-phenyl trifluoroacetimidate 2 with catalytic
TMSOTf or BF3·OEt2 in the presence of alcohol 12 at -78
°C (and by slowly warming to 0 °C) were indeed able to
furnish desired disaccharide 15 (Table 2, entries 3 and 4).
Unfortunately, both reactions did not proceed to completion,
even though N-phenyl trifluoroacetimidate 2 was completely
consumed in both cases, and the selectivity was poor with
use of TMSOTf. Conducting the reaction at 0 °C improved
the R-selectivity of the TMSOTf-catalyzed glycosylation
(Table 2, entry 5). Further improvement of the R-selectivity
was attempted by using a 4:1 Et2O/CH2Cl2 solvent mixture
(Table 2, entry 6). Unfortunately, under these conditions,
With building blocks 2 and 10 in hand, we wanted to test
their ability to effect the glycosylation of C4-OH galac-
tosamine nucleophiles in order to prepare a disaccharide
motif found in the ZPSs of S. mitis, S. pneumoniae, and B.
fragilis. This glycosylation was considered to be potentially
problematic due to the low nucleophilicity of the axial C4-
OH of galactosamine and previously described difficulties12c
(20) Mironov, Y. V.; Sherman, A. A.; Nifantiev, N. E. Tetrahedron Lett.
2004, 45, 9107.
(21) Bedini, E.; Esposito, D.; Parrilli, M. Synlett 2006, 6, 825.
(22) Gauffeny, F.; Marra, A.; Shun, L. K. S.; Sinay¨, P.; Tabeur, C.
Carbohydr. Res. 1991, 219, 237.
(23) Comegna, D.; Bedini, E.; Di Nola, A.; Iadonisi, A.; Parrilli, M.
Carbohydr. Res. 2007, 342, 1021.
(24) Grundler, G.; Schmidt, R. R. Liebigs Ann. Chem. 1984, 1826.
Org. Lett., Vol. 12, No. 7, 2010
1626

Table 2. Glycosylations between AAT Building Blocks 2 and 10 and Galactosamine Nucleophiles 12-14
building
entry
blocka
nucleophilea
catalysta
TMSOTf
BF3·OEt2 CH2Cl2
TMSOTf CH2Cl2
BF3·OEt2 CH2Cl2
TMSOTf
TMSOTf
TMSOTf
TMSOTf
TMSOTf
solvent
CH2Cl2
temp, °C 4 Å MS product conversion,b
%
R:ꢀ selectivityc
1
2
3
4
5
6
7
8
9
10
10
2
2
2
2
2
2
2
12
12
12
12
12
12
12
13
14
-78 to 0
-78 to 0
-78 to 0
-78 to 0
0
0
0
0
0
yes
yes
yes
yes
yes
yes
no
15
15
15
15
15
15
15
16
17
0
0
55
30
57
0
>95 (74)
>95 (81)
>95 (59)
1:2.5
6.5:1
7:1
CH2Cl2
Et2O/CH2Cl2 (4:1)
CH2Cl2
CH2Cl2
CH2Cl2
5:1
5.5:1
5:1
no
no
a Reactions were run with 1.5 equiv of building block, 1.0 equiv of nucleophile, 0.1 equiv of catalyst, and at 0.02 M concentration. b Conversion
determined by ratio of product to nucleophile in the crude H NMR; isolated yield of R-anomer in parentheses. R/ꢀ ratios determined by H NMR of the
crude product.
c
1
1
N-phenyl trifluoroacetimidate 2 could not be activated and
was recovered. Finally, the glycosylation between N-phenyl
trifluoroacetimidate 2 and nucleophile 12 was attempted
without preactivated 4 Å molecular sieves (Table 2, entry
7). Although the R-selectivity dropped slightly to 5:1,
nucleophile 12 was completely consumed in the reaction.
Desired R-linked disaccharide 15 was separated from the
ꢀ-isomer by silica gel chromatography and isolated in 74%
yield. Disaccharides 16 and 17 were prepared in analogous
fashion with similar yields and selectivites (Table 2, entries
8 and 9). For the newly formed glycosidic bond, the anomeric
proton of disaccharides 15-17 was determined by HSQC
spectroscopy, and all disaccharides displayed an anomeric
coupling constant (3JH1,H2 ) 3.6-4.0 Hz) confirming the
presence of an R-linkage.
novo synthesis of AAT building blocks 2 and 10 from
N-Cbz-L-threonine 5 and the successful use of N-phenyl
trifluoroacetimidate 2 in glycosylation reactions with galac-
tosamine nucleophiles 12-14. Further progress toward the
synthesis of the repeating subunit of PS A1 (1) and related
ZPSs, and the study of their immunological properties will
be reported in due course.
Acknowledgment. The authors gratefully thank the Max
Planck Society for funding, the Alexander von Humboldt
Foundation for a postdoctoral research fellowship (R.P.), and
the Studienstiftung des deutschen Volkes for a Ph.D.
fellowship (P.S.).
Supporting Information Available: Experimental pro-
cedures and tabulated spectroscopic data for all new com-
pounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
In summary, we have completed the synthesis of orthogo-
nally protected disaccharides 15-17 that are an essential
component of the ZPS repeating subunits of S. mitis, S.
pneumoniae, and B. fragilis. Key to this synthesis is the de
OL1003912
Org. Lett., Vol. 12, No. 7, 2010
1627