Indium-mediated stereoselective synthesis of truncated, 6- and 7-carbon sialic acids

Article abstract of DOI:10.1055/s-2002-35564

Several 6- and 7-carbon sialic acid derivatives were synthesized, without tedious protecting group manipulations, in high overall yields. A key step of the synthesis was the chain extension of suitable α-amino aldehyde derivatives by an indium-mediated addition of ethyl 2-(bromomethyl)acrylate. Under acidic reaction conditions, the corresponding extended enoates were obtained with high trans-stereoselectivity. Ozonolysis furnished the desired 4-acylamino-substituted hexulosonic and heptulosonic acids in free form for biochemical studies.

Full text of DOI:10.1055/s-2002-35564

2104
LETTER
Indium-Mediated Stereoselective Synthesis of Truncated, 6- and 7-Carbon
Sialic Acids
s
tereoselective synthes
a
is of trunca
t
ted sia
l
h
ic acids ias Warwel, Wolf-Dieter Fessner*
Darmstadt University of Technology, Department of Organic Chemistry and Biochemistry, Petersenstr. 22, 64287 Darmstadt, Germany
Fax +49(6151)166636; E-mail: fessner@tu-darmstadt.de
Received 25 September 2002
Abstract: Several 6- and 7-carbon sialic acid derivatives were syn-
thesized, without tedious protecting group manipulations, in high
overall yields. A key step of the synthesis was the chain extension
of suitable α-amino aldehyde derivatives by an indium-mediated
addition of ethyl 2-(bromomethyl)acrylate. Under acidic reaction
conditions, the corresponding extended enoates were obtained with
high trans-stereoselectivity. Ozonolysis furnished the desired 4-
acylamino-substituted hexulosonic and heptulosonic acids in free
form for biochemical studies.
Key words: allylations, Barbier-type reactions, indium α-amino
aldehydes, sialic acids
Scheme 1 Chemical and enzymatic routes for synthesis of N-acetyl-
neuraminic acid.
Sialic acids, having an 8- or 9-carbon backbone, are main-
ly found as terminal components of cell surface glycopro-
gram to develop new routes towards neo-oligosaccharides
teins, glycolipids and oligosaccharides1 that are of
decisive importance in many physiological and patholog-
ical processes.2 For an improved understanding of the
structure-activity relationship in sialooligosaccharide
binding motifs and their metabolism (specifically con-
cerning sialyltransferases and sialosidases) there is a need
for a synthetic access to modified and lower homologous,
non-natural sialic acid derivatives.
containing modified sialic acids8 we became interested in
an efficient synthetic access to truncated sialic acids that
would allow for the introduction of structural diversity by
the N-acyl group. Herein, we report a new synthetic strat-
egy that is based on the preparation of α-amino aldehydes
from allylamine precursors, followed by an indium-medi-
ated chain-extension with ethyl 2-(bromomethyl)acrylate.
This methodology is known to be compatible with a broad
variety of functional groups and thus obviates the need for
tedious protective group manipulations.
A number of enzymatic3 and nonenzymatic4 syntheses of
sialic acids have been developed, especially for the most
important N-acetylneuraminic acid (Neu5Ac, 1,
Scheme 1). In the case of truncated sialic acids having a
6- or 7-carbon backbone, the conventional enzymatic ap-
proach fails because C3 or C4 aldehyde precursors are gen-
erally not, or only very poor, substrates for Neu5Ac
aldolase.5 Therefore, the truncated C7-sialic acid had usu-
ally not been prepared by synthesis, but by an oxidative
chain degradation of natural sialooligosaccharides (e.g.,
by periodate oxidation/borohydride reduction/hydroly-
sis).6
α-Amino aldehydes are notorious for their high chemical
reactivity and configurational lability.9 Some α-amino
propionaldehyde derivatives, which would be suitable for
our approach, had previously been synthesized starting
from the corresponding amino acids.7,9a According to our
experience, however, ozonolysis of olefinic precursors is
a preferable, clean and practical alternative to prepare
pure α-amino aldehydes.10 This procedure renders further
purification unnecessary, so that the unstable compounds
can be used immediately for further conversions. This is
particularly useful for compounds that carry further polar
functions as required in this study, such as in 2-amino-2-
deoxyglyceraldehyde (4) or 2-amino-2-deoxyerythrose
(7) derivatives.
On the other hand, the length of an aldehyde precursor
poses, in principle, no limitation to the most notable
chemical route, which is the indium-mediated chain ex-
tension of unprotected aldoses leading to common C8 and
C9 sialic acids.4 To our knowledge however, only three
synthetic routes have been explored yet for the prepara-
tion of lower sialic acids that do not start from 1 or its con-
jugates, but using strategies that involve tedious
protection-deprotection schemes.7 In the context of a pro-
To obtain suitable olefinic precursors to the desired N-
protected 2-amino-2-deoxyglyceraldehyde derivatives
4a/b (Scheme 2), cinnamaldehyde (2) was converted to
(E)-2-azido-4-phenyl-3-buten-1-ol according to a litera-
ture procedure,10 and the azide was reduced with tri-
phenylphosphine. N-Acylation provided the stable
precursors 3a/b. The acetyl and phenylacetyl groups were
chosen for N-acyl modification, because the former would
Synlett 2002, No. 12, 02 12 2002. Article Identifier:
1437-2096,E;2002,0,12,2104,2106,ftx,en;G28502ST.pdf.
© Georg Thieme Verlag Stuttgart · New York
ISSN 0936-5214

LETTER
Stereoselective Synthesis of Truncated Sialic Acids
2105
match the natural substitution pattern of neuraminic acid, for the In-mediated methacrylate addition to the corre-
and the latter would provide a convenient opportunity for sponding fully protected α-amino aldehyde.7b Recently,
further modification at a final product stage because phe- we had reported that the diastereoselectivity of indium-
nylacetyl amides can be hydrolyzed selectively under mediated addition reactions to carbohydrates can be sig-
mild conditions using a penicillin G acylase.11 Ozonolysis nificantly enhanced by employing aqueous acidic reaction
of the styrene moiety in 3a/b followed by reductive work- conditions.13 Indeed, we observed that the polar, short-
up smoothly provided the free aldehydes 4a/b in aqueous chain aldehydes 4 and 7, when exposed to the indium re-
solution, which were used immediately for the next step.
agent system in an acidic ethanol–water mixture, also
showed a significantly improved threo-diastereoselectivi-
ty (8a–d, consistent threo:erythro ratio = 83:17), which
strongly supports a pronounced stereodiscriminating
chelating effect upon protonation of intermediates en
route to the transition state.14 Because of their proclivity
for polymerization, the acrylate adducts 8a–d were usual-
ly not isolated but ozonized directly. The resultant α-ke-
toester intermediates spontaneously cyclized to give the
truncated sialic acid esters as a mixture of stereoisomers,
from which the desired major threo-diastereomers 9a–d
could be readily isolated by chromatography. The corre-
sponding sialic acids 10a–d were finally obtained by alka-
line hydrolysis in 55–70% overall yield (based on the
olefinic precursors 3 or 6; 4 steps).12
Scheme 2 Reagents and conditions: (a) Me3S+I–, NaH, DMSO–
THF; (b) NaN3, aq acetone; (c) PPh3, aq THF; (d) Ac2O, pyridine or
PhAcCl, Et3N, MeOH; (e) O3, then Me2S; (f) VO(acac)2, t-BuOOH;
(g) [(NMe2)2CNH2]+N3 .
–
For the higher homologous precursor to give a C7 sialic
acid analog, a syn-arrangement of functional groups was
required. The synthesis was similarly designed to start
from 5-phenyl-2,4-pentadien-1-ol (5, Scheme 2) and in-
volved allyl-selective mono-epoxidation using tert-butyl
hydroperoxide in the presence of vanadate, followed by
epoxide-opening in situ using tetramethylguanidinium
azide to afford the corresponding anti-configured azido
compound in 40% overall yield. Reduction of the azide
group and selective N-acylation of the resulting amino al-
cohol produced stable aminodiols 6a/b. Subsequent ozo-
nolysis smoothly generated the desired 2-amino-2-
deoxyerythrose derivatives 7a/b which were used directly
in the ensuing conversion. In comparison with other meth-
ods for the synthesis of N-protected α-amino alde-
hydes,12,9a the use of an olefinic precursor indeed proved
advantageous in that it offered short reaction sequences
and higher overall yields.
Scheme 3 Reagents and conditions: (a) In, EtOH, aq HCl; (b) O3,
then Me2S, and chromatographic separation (9a 69%, 9b 67%, 9c
65%, 9d 79%); (c) aq LiOH (10a 93%, 10b 90%, 10c 84%, 10d 88%).
The synthetic approach described here offers a convenient
access to truncated sialic acids, with flexibility to intro-
duce a broad structural variation in the N-acyl moiety.
Such sialic acids will be of utility for further detailed stud-
ies into the substrate tolerance of enzymes involved in the
metabolism of sialosides, and they offer new avenues into
the preparation of neo-sialooligosaccharides that are of
importance for biological studies.15 It ought to be men-
tioned that the procedures described above yield racemic
products, but that enantiomerically pure materials may be
obtained by simple modifications of synthetic methods for
their asymmeric variants (e.g., asymmetric epoxidation),
or by kinetic resolution during the enzymatic activation to
the nucleotide sugars.8
The -amino aldehydes 4a/b and 7a/b generated in situ
were used immediately for In-mediated Barbier-type re-
actions with ethyl 2-(bromomethyl)acrylate (Scheme 3).12
Initially, under neutral reaction conditions only unsatis-
factory selectivity was observed (e.g., threo:erythro ratio
for 8a/c = 60:40); very poor stereoselectivity, under neu-
tral reaction conditions, has also been reported recently
Acknowledgment
This work was supported by the Fonds der Chemischen Industrie.
Synlett 2002, No. 12, 2104–2106 ISSN 0936-5214 © Thieme Stuttgart · New York

2106
M. Warwel, W.-D. Fessner
LETTER
(10) Hung, R. R.; Straub, J. A.; Whitesides, G. M. J. Org. Chem.
1991, 56, 3849.
References
(1) (a) Angata, T.; Varki, A. Chem. Rev. 2002, 102, 439.
(b) Roy, R.; Laferrière, C. A.; Pon, R. A. Methods Enzymol.
1994, 247, 351. (c) Corfield, A. P.; Schauer, R. In Sialic
Acids, Chemistry, Metabolism and Function; Schauer, R.,
Ed.; Springer Verlag: New York, 1982, 5. (d) Vliegenthart,
J. F. G.; Mamerling, J. P. In Sialic Acids, Chemistry,
Metabolism and Function; Schauer, R., Ed.; Springer
Verlag: New York, 1982, 59.
(2) (a) Vimr, E.; Lichtensteiger, C. Trends Microbiol. 2002, 10,
254. (b) Kiefel, M. J.; von Itzstein, M. Chem. Rev. 2002,
102, 471. (c) Jaeken, J.; Matthijs, G. Annu. Rev. Genomics
Hum. Genet. 2001, 2, 129. (d) Schauer, R. Glycoconjugate
J. 2000, 17, 485. (e) Crocker, P. R.; Varki, A. Trends
Immunol. 2001, 22, 337. (f) Sillanaukee, P.; Ponnio, M.;
Jaaskelainen, I. P. Eur. J. Clin. Invest. 1999, 29, 413.
(g) Rosenberg, S. Biology of the Sialic Acids; Plenum Press:
New York, 1995.
(3) (a) For a review, see: Fessner, W.-D.; Walter, C. Top. Curr.
Chem. 1996, 184, 97. (b) Augé, C.; David, S.; Gautheron,
C.; Veyrières, A. Tetrahedron Lett. 1985, 26, 2439.
(c) Simon, E. S.; Bednarski, M. D.; Whitesides, G. M. J. Am.
Chem. Soc. 1988, 110, 7159. (d) Schrell, A.; Whitesides, G.
M. Liebigs Ann. Chem. 1990, 1111. (e) Lin, C.-H.; Sugai,
T.; Halcomb, R. L.; Ichikawa, Y.; Wong, C. H. J. Am. Chem.
Soc. 1992, 114, 10138. (f) Mahmoudian, M.; Noble, D.;
Drake, C. S.; Middleton, R. F.; Montgomery, D. S.; Piercey,
J. E.; Ramlakhan, D.; Todd, M.; Dawson, M. J. Enzyme
Microb. Technol. 1997, 20, 393.
(4) (a) For a review, see: Li, C.-J.; Chan, T.-H. Tetrahedron
1999, 55, 11149. (b) Li, C.-J.; Chan, T.-H. J. Chem. Soc.,
Chem. Commun. 1992, 747. (c) Kim, E.; Gordon, D. M.;
Schmid, W.; Whitesides, G. M. J. Org. Chem. 1993, 58,
5500. (d) Gordon, D. M.; Whitesides, G. M. J. Org. Chem.
1993, 58, 7937. (e) Li, C. J.; Lee, M. C.; Wei, Z. Y.; Chan,
T. H. Can. J. Chem. 1994, 72, 1181. (f) Chan, T.-H.; Lee,
M.-C. J. Org. Chem. 1995, 60, 4228. (g) Choi, S.-K.; Lee,
S.; Whitesides, G. M. J. Org. Chem. 1996, 61, 8739.
(h) Banwell, M. G.; Savi, C. D.; Watson, K. Chem. Commun.
1998, 1189.
(11) Waldmann, H.; Sebastian, D. Chem. Rev. 1994, 94, 911.
(12) Typical reaction conditions for the indium-mediated
allylation: To a solution of the aminoaldehyde (1.0 mmol)
and ethyl 2-(bromomethyl)-acrylate16 (386 mg; 2.0 mmol) in
a mixture made from 12.5 mL EtOH and 2.5 mL 0.1 M HCl
was added indium powder (230 mg; 2.0 mmol) at r.t. The
suspension was vigorously stirred until TLC indicated
complete conversion of the starting material, and was then
filtered through a pad of Celite. Water was added (20 mL),
and the resulting mixture was concentrated to 20 mL under
vacuum followed by extraction with EtOAc (3 × 30 mL).
The combined organic layers were washed with brine
(1 × 20 mL) and water (1 × 20 mL), dried over Na2SO4,
filtered, and concentrated under vacuum. Without further
purification, the remaining crude colorless solid was taken
up in MeOH and treated with ozone at –78 °C. Ozonide
reduction (Me2S, r.t.) followed by flash chromatography
provided compounds 9a–d as colorless syrups, which were
hydrolyzed by treatment with 2 equivalents of aq LiOH to
furnish stereoisomerically pure truncated sialic acid
derivatives 10a–d.
Spectroscopic data: 10a: 1H NMR (300 MHz, D2O):
δ = 3.82 (m, 1 H, 4-H), 3.67 (m, 2 H, 6-H), 3.53 (m, 1 H,
5-H), 2.15 (dd, 1 H, J3eq,3ax = 13.1 Hz, J3eq, 4 = 4.9 Hz, 3eq-H),
1.90 (s, 3 H, CH3), 1.69 (dd, 1 H, J3ax,3eq = 13.1 Hz,
J3ax,4 = 11.2 Hz, 3ax-H); 13C NMR (75 MHz, D2O):
= 177.23 (C-1), 175.07 (C=O), 97.75 (C-2), 68.27 (C-4),
63.67 (C-6), 54.36 (C-5), 41.33 (C-3), 24.40 (CH3); ESI-MS:
m/z = 218 ([M–H]-, 100). 10b: 1H NMR (300 MHz, D2O):
δ = 7.26–7.38 (m, 5 H, Har), 3.58 (s, 2 H, CH2), 3.48–3.96
(m, 4 H, 4-,5-,6-H), 2.14 (dd, 1 H, J3eq,3ax = 13.0 Hz,
J
3eq,4 = 4.8 Hz, 3eq-H), 1.88 (dd, 1 H, J3ax,3eq = 13.0 Hz,
J3ax,4 = 11.1 Hz, 3ax-H); 13C NMR (75 MHz, D2O):
= 178.27 (C=O), 177.98 (C-1), 138.07 (Ci), 131.64,
131.29, 129.63 (CHo,m,p), 99.24 (C-2), 69.26 (C-4), 64.19 (C-
6), 55.42 (C-5), 45.28 (CH2), 42.50 (C-3); ESI-MS: m/
z = 318 ([M + Na]+, 100), 300(40). 10c: 1H NMR (300 MHz,
D2O): δ = 3.65–3.93 (5 H, m, 4-,5-,6-,7-H), 2.32 (1 H, dd,
J
3eq,3ax = 13.0 Hz, J3eq,4 = 5.0 Hz, 3eq-H), 2.04 (3 H, s, CH3),
1.87 (1 H, dd, J3ax,3eq = 13.0 Hz, J3ax,4 = 11.6 Hz, 3ax-H); 13
C
(5) (a) Suttajit, M.; Urban, C.; McLean, R. L. J. Biol. Chem.
1971, 246, 810. (b) Kim, M.-J.; Hennen, W. J.; Sweers, H.
M.; Wong, C.-H. J. Am. Chem. Soc. 1988, 110, 6481.
(c) Fitz, W.; Schwark, J. R.; Wong, C.-H. J. Org. Chem.
1995, 60, 3663.
(6) (a) Kuhn, R.; Gauhe, A. Chem. Ber. 1965, 98, 395.
(b) McLean, R. L.; Suttajit, M.; Beidler, J.; Winzler, R. J. J.
Biol. Chem. 1971, 246, 803. (c) Peters, B. P.; Aronson, N.
N. Carbohydr. Res. 1976, 47, 345. (d) Reuter, G.; Schauer,
R.; Szeiki, C.; Kamerling, J. P.; Vliegenthart, J. F. G.
Glycoconjugate J. 1989, 6, 35.
NMR (75 MHz, D2O): = 177.77 (C-1), 175.1 (C=O), 98.17
(C-2), 75.80 (C-6), 69.39 (C-4), 63.80 (C-7), 55.22 (C-5),
41.91 (C-3), 25.03 (CH3); ESI-MS: m/z = 248 ([M – H]-,
100), 230(41). 10d: 1H NMR (300 MHz, D2O): = 7.28–
7.40 (5 H, m, Har), 3.63 (2 H, s, CH2), 3.50–3.91 (5 H, m, 4-
,5-,6-,7-H), 2.33 (1 H, dd, J3eq,3ax = 12.9 Hz, J3eq,4 = 4.9 Hz,
3eq-H), 1.84 (1 H, dd, J3ax,3eq = 12.9 Hz, J3ax,4 = 11.8 Hz,
3ax-H); 13C NMR (75 MHz, D2O): δ = 178.16 (C=O), 175.66
(C-1), 137.72 (Ci), 131.92, 131.78, 130.19 (Co,m,p), 97.81 (C-
2), 75.70 (C-6), 68.94 (C-4), 63.86 (C-7), 55.16
(7) Synthesis of derivatives of 10a: (a) from D-
glucuronolactone:Vlahov, I. R.; Vlahova, P. I.; Schmidt, R.
R. Tetrahedron Lett. 1991, 32, 7025. (b) From D-serine:
Chappell, M. D.; Halcomb, R. L. Org. Lett. 2000, 2, 2003.
(c) From L-ascorbic acid: Banwell, M.; De Savi, C.;
Hockless, D.; Watson, K. Chem. Commun. 1998, 645.
(8) Knorst, M.; Fessner, W.-D. Adv. Synth. Catal. 2001, 343,
698.
(9) (a) Jurczak, J.; Golebiowski, A. Chem. Rev. 1989, 89, 149.
(b) Reetz, M. T. Angew. Chem., Int. Ed. Engl. 1991, 30,
1531. (c) Steurer, S.; Podlech, J. Eur. J. Org. Chem. 1999,
1551.
(C-5), 45.10 (CH2), 41.79 (C-3); ESI-MS: m/z = 420 ([M +
Na + 4 H2O]+, 100), 402 ([M + Na + 3 H2O]+, 95)
(13) Warwel, M.; Fessner, W.-D. Synlett 2000, 6, 865.
(14) The origin of the pronounced selectivity at low pH has not
been thoroughly investigated yet but it may be speculated
that protonation of the electrophile will lead to a tighter
transition state for the addition. For an in-depth discussion,
see: Paquette, L. A.; Mitzel, T. M. J. Am. Chem. Soc. 1996,
118, 1931.
(15) Lemieux, G. A.; Bertozzi, C. R. Chem. Biol. 2001, 8, 265.
(16) Villieras, J.; Rambaud, M. Org. Synth. 1988, 66, 220.
Synlett 2002, No. 12, 2104–2106 ISSN 0936-5214 © Thieme Stuttgart · New York