Synthesis of iduronic acid building blocks for the modular assembly of glycosaminoglycans

Article abstract of DOI:10.1021/jo0340760

The modular synthesis of glycosaminoglycans requires straightforward methods for the production of large quantities of protected uronic acid building blocks. In particular, the preparation of fully differentiated iduronic acids has proven particularly cha

Full text of DOI:10.1021/jo0340760

iduronic acid building blocks used in the synthesis of
large heparin structures have been reported, but remain
lengthy and involve several steps that produce multiple
products.3 Methods for the completely selective conver-
sion of 5-aldopentoses to iduronic acid derivatives8 and
for the selective silylation of the anomeric hydroxyl of
iduronic acids9 have been described, but have not been
incorporated into the synthesis of iduronic acid monosac-
charide building blocks. Here we report a short syn-
thetic route to iduronic acid building blocks through
the conversion of diacetone glucose to the key inter-
mediate methyl 3-O-benzyl-1,2-O-isopropylidene-R-L-
idopyranosiduronate 6. This iduronic acid derivative can
serve as a glycosyl acceptor or can be readily converted
to fully differentiated iduronic acid trichloroacetimidate
glycosyl donors. The synthesis described incorporates a
number of past chemistries through improved and sim-
plified procedures, and requires only a few purification
steps.
Commercially available diacetone glucose 1 was trans-
formed to diol 2 through benzylation and selective acetal
cleavage (Scheme 1).5 Treatment of 2 with aqueous
sodium periodate adsorbed onto silica yielded aldehyde
310,11 that was used without purification. Reaction of 3
with freshly prepared trithiophenylmethylithium af-
forded L-idose-configured thioortho ester in high yield
with no D-glucose product detectable.8 The reaction
products were treated directly with CuCl2/CuO to effect
the cleavage of the thioortho ester to the furanose methyl
ester 4, along with small amounts of the corresponding
phenylthioester. Stirring the crude product mixture with
K2CO3 in methanol converted this byproduct to the
desired methyl ester 4. Removal of the isopropylidene
group from furanose 4 by reaction with 90% TFA (aq)
yielded the crystalline 3-O-benzyl iduronic methyl ester
5 in its pyranose form.12
Syn th esis of Id u r on ic Acid Bu ild in g Block s
for th e Mod u la r Assem bly of
Glycosa m in oglyca n s
Gregory J . S. Lohman, Diana K. Hunt,
J ens A. Ho¨germeier, and Peter H. Seeberger*,†
Massachusetts Institute of Technology, 77 Massachusetts
Avenue 18-292, Cambridge, Massachusetts 02139
seeberg@mit.edu
Received J anuary 22, 2003
Abstr a ct: The modular synthesis of glycosaminoglycans
requires straightforward methods for the production of large
quantities of protected uronic acid building blocks. In
particular, the preparation of fully differentiated iduronic
acids has proven particularly challenging. An efficient route
to methyl 3-O-benzyl-1,2-O-isopropylidene-R-L-idopyranosi-
duronate 6 from diacetone glucose in nine steps and 36%
overall yield is described. Idopyranosiduronate 6 is useful
as a glycosyl acceptor and as an intermediate that may be
further elaborated into iduronic acid trichloroacetimidate
glycosyl donors for the assembly of glycosaminoglycan
structures as illustrated here.
L-Iduronic acid is found naturally as a component of
the glycosaminoglycans heparin, heparan, and dermatan.
While these biopolymers are known to have diverse
biological function, the structure-activity relationships
remain poorly understood.1,2 Access to defined glycosami-
noglycan sequences through a general, modular synthesis
would be a significant asset to the biochemical studies
of these compounds.
Glycosaminoglycan synthesis requires large quantities
of differentially protected iduronic acid molecules and
necessitates concise and efficient methods for the produc-
tion of iduronic acid synthons. Since iduronic acid itself
is not commercially available, syntheses of iduronic acid
derivatives from a variety of starting materials, including
idose,3 glucose,4,5 glycals,6 and glucuronic acid,7 have been
developed. While L-idose is very costly, rendering it an
undesirable starting material, syntheses with other
starting materials require the inversion of the C-5
stereocenter on a D-gluco sugar. Few methods reported
for this inversion have realized full selectivity for the
desired configuration. Syntheses from idose to procure
Installation of a 1,2-isopropylidene onto 5 locked the
1
sugar in the C4 pyranose form and afforded key inter-
mediate 6.12 Use of highly reactive 2-methoxypropene13,14
to form the desired isopropylidene acetal prevented
opening of the sugar ring and thus the trapping of
thermodynamically more stable furanose isomers. Partial
hydrolysis of the crude product mixture with acidic resin
in methanol produced the desired product 6 in 48% yield
from 4, along with recovered 5 (36%). The synthesis of 6
from diacetone glucose was achieved in nine steps and
36% overall yield (56% assuming complete resubmission
of 5). Only three chromatographic steps were required
in this synthetic sequence; additionally, the procedures
for the transformation of 3 to 4 and 5 to 6 have been
† Present address: Laboratorium fu¨r Organische Chemie, ETH
Ho¨nggerberg/HCI F 315, Wolfgang-Pauli-Strasse 10, CH-8093 Zu¨rich,
Switzerland.
(1) Bernfield, M.; Go¨tte, M.; Park, P. W.; Reizes, O.; Fitzgerald,
M. L.; Lincecum, J .; Zako, M. Annu. Rev. Biochem. 1999, 68, 729-
777.
(2) Capilia, I.; Linhardt, R. J . Angew. Chem., Int. Ed. 2002, 41, 390-
412.
(3) Tabeur, C.; Machetto, F.; Mallet, J .; Duchaussoy, P.; Petitou, M.;
Sinay¨, P. Carbohydr. Res. 1996, 281, 253-276.
(4) Rochepeau-J obron, L.; J aquinet, J . C. Carbohydr. Res. 1997, 303,
395-406.
(8) Lubineau, A.; Gavard, O.; Alais, J .; Bonnaffe´, D. Tetrahedron
Lett. 2000, 41, 307-311.
(9) Ojeda, R.; de Paz, J . L.; Mart´ın-Lomas, M.; Lassaletta, J . M.
Synlett 1999, 8, 1316-1318.
(10) Ste¸powska, H.; Zamojski, A. Tetrahedron 1999, 55, 5519-
5538.
(11) Krajewski, J . W.; Gluzinski, P.; Pakulski, Z.; Zamojski, A.;
Mishnev, M.; Kemme, A. Carbohydr. Res. 1994, 252, 97-105.
(12) Orgueira, H. A.; Bartolozzi, A.; Schell, P.; Seeberger, P. H.
Angew. Chem., Int. Ed. 2002, 41, 2128-2131.
(13) Gelas, J .; Horton, D. Heterocycles 1981, 16, 1587.
(14) Wolfrom, M. L.; Diwadkar, A. B.; Gelas, J .; Horton, D. Carbo-
hydr. Res. 1974, 35, 87.
(5) Orgueira, H. A.; Bartolozzi, A. B.; Schell, P.; Litjens, R. E. J . N.;
Palmacci, E. R.; Seeberger, P. H. Chem. Eur. J . 2003, 9, 140-169.
(6) Schell, P.; Orgueira, H. A.; Roehrig, S.; Seeberger, P. H.
Tetrahedron Lett. 2001, 42, 3811-3814.
(7) Medakovic´, D. Carbohydr. Res. 1994, 253, 299-300.
10.1021/jo0340760 CCC: $25.00 © 2003 American Chemical Society
Published on Web 08/28/2003
J . Org. Chem. 2003, 68, 7559-7561
7559

SCHEME 1
SCHEME 2
SCHEME 4
SCHEME 3
by acetylation of the crude material, afforded 10a and
10b (Scheme 3). These anomers were separated and
exposed to benzylamine to cleave the anomeric acetate
selectively. Since 10a and 10b react at different rates
under these conditions, better yields were obtained by
treating these isomers separately. The resultant lactol
was transformed to glycosyl trichloroacetimidate 11 by
reaction with trichloroacetonitrile and DBU.
The second route to differentially protected iduronic
acid derivatives from 6 was the selective formation of
anomeric silyl ethers.9 After protection of the C-4 hy-
droxyl, the isopropylidenes of 8 and 9 were cleaved with
aqueous TFA (Scheme 4). Both 8 and 9 were exposed to
a silyl chloride at low temperature to selectively pro-
tect the C-1 hydroxyl and produce 12 and 15, respec-
tively. The steric bulk of the C-4 protecting group re-
quired the adjustment of reaction conditions to achieve
good yield and selectivity. The larger benzyl group
allowed use of tert-butyldimethylsilyl chloride (TBS-Cl)
as the silylating agent. The smaller levulinate ester re-
quired lower temperatures and the bulkier thexyldi-
methylsilyl chloride (TDS-Cl) to achieve good selec-
tivity.
simplified from those previously reported and the yields
improved.8,12
L-iduronic acid derivative 6 may serve as a glycosyl
acceptor in the synthesis of disaccharide modules,5 or
may be converted into a variety of iduronic acid trichlo-
roacetimidate glycosyl donors. The introduction of dif-
ferent protecting groups, including silyl ethers, esters,
and alkyl ethers, on the C-4 hydroxyl can be readily
achieved.5 Silylation can be effected through the use of
silyl triflates, esterification through acid anhydrides, and
DMAP. Alkylation poses more of a challenge, as the C-5
stereocenter of 6 may epimerize in the presence of strong
base, but was accomplished with silver oxide and the
corresponding alkyl bromide (Scheme 2).
Differential protection of the C-1 and C-2 hydroxyls
was achieved through two routes. One option capitalized
on the faster rate of cleavage of anomeric acetates over
other ester protecting groups.15 Treatment of 7 with
aqueous TFA to effect isopropylidene cleavage, followed
Further elaboration of 12 by introduction of a levuli-
nate ester on the C-2 hydroxyl afforded 13. Cleavage of
the anomeric silyl ether yielded lactol, which was con-
(15) Ernst, B.; Hart, G. W.; Sinay¨, P. Carbohydrates in Chemis-
try and Biology; Wiley-VCH: Weinheim, Germany, 2000; Vol. 1,
Chapter 2.
7560 J . Org. Chem., Vol. 68, No. 19, 2003

verted to trichloroacetimidate 14. Similarly, product 15
was converted to its pivaloyl ester 16 and acetate ester
17. The anomeric thexyl ethers were cleaved with
HF‚pyridine, and the lactols converted to the correspond-
ing trichloroacetimidates 18 and 19.
Ack n ow led gm en t. Financial support from the Na-
tional Institute of Health (HL-64799 and HL-62598), the
Research Corporation (Research Innovation Award to
P.H.S.), CaPCURE (Research Awards to P.H.S.), MIT-
DuPont Alliance (Predoctoral Fellowship for G.J .S.L.),
and the American Society for Engineering Education
(National Defense Science and Engineering Graduate
Fellowship for G.J .S.L.) is gratefully acknowledged.
Funding for the MIT-DCIF Varian 500-MHz NMR was
provided by NSF (Award CHE-9808061). Funding for
the MIT-DCIF Varian 300-MHz NMR was provided by
NSF (Award DBI-9729592). We thank Dr. Hernan
Orgueira and Emma R. Palmacci for helpful discussions.
The synthesis of compound 6 in 36% overall yield in
nine steps from diacetone glucose is significantly shorter
than the previous route,12 allowing for the rapid produc-
tion of multigram quantities of iduronic acid building
blocks. Key intermediate 6 may be elaborated through
selective anomeric hydroxyl silylation or selective acetate
cleavage to produce a variety of iduronic acid derivatives,
including the glycosyl donors 11, 14, 18, and 19, using
chemistry readily adaptable to other protecting group
schemes. The chemistry presented offers for the first time
a general, high-yielding, and easily scaleable route to
iduronic acid synthons for glycosaminoglycan assembly,
addressing a long-standing obstacle in the convenient
production of these compounds.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures and characterization data for all compounds not
previously reported. This material is available free of charge
via the Internet at http://pubs.acs.org.
J O0340760
J . Org. Chem, Vol. 68, No. 19, 2003 7561