Toward synthesis of the isosteric sulfonate analogues of the at-iii binding domain of heparin

Article abstract of DOI:10.1021/ol900952d

D-Glucuronate and L-iduronate containing disaccharides related to the antithrombin-binding pentasaccharide of heparin, in which one of the sulfate esters is systematically replaced by a sodium sulfonatomethyl moiety, were synthesized. The sulfonic acid gr

Full text of DOI:10.1021/ol900952d

ORGANIC
LETTERS
2009
Vol. 11, No. 12
2619-2622
Toward Synthesis of the Isosteric
Sulfonate Analogues of the AT-III
Binding Domain of Heparin
Miha´ly Herczeg,† La´szlo´ La´za´r,† Aniko´ Borba´s,*,† Andra´s Lipta´k,† and
Sa´ndor Antus†,‡
Research Group for Carbohydrates of the Hungarian Academy of Sciences, P.O. Box
94, H-4010 Debrecen, Hungary, and Department of Organic Chemistry, UniVersity of
Debrecen, P.O. Box 20, H-4010 Debrecen, Hungary
borbasa@puma.unideb.hu
Received April 30, 2009
ABSTRACT
D-Glucuronate and L-iduronate containing disaccharides related to the antithrombin-binding pentasaccharide of heparin, in which one of the
sulfate esters is systematically replaced by a sodium sulfonatomethyl moiety, were synthesized. The sulfonic acid group was introduced by
stereoselective radical addition onto the exomethylene moiety of the appropriate glycoside derivatives, and the resulting sulfonatomethyl
glucosides were used as acceptors.
Heparin, a linear, highly sulfated glycosamino-glycan polysac-
charide is a well-known blood anticoagulant and has been
employed in medical practice since the late 1930s.1
A
characteristic pentasaccaharide domain of heparin termed
DEFGH (1, Figure 1) binds and activates the coagulation
inhibitor antithrombin III (AT-III) through an allosteric
mechanism, the activated AT-III then inhibits both the
procoagulant serine proteinase factor Xa and another coagu-
lation enzyme, thrombin, by forming an inactive ternary
complex.2,3 Fondaparinux (2), a closely related synthetic
counterpart of the AT-III binding pentasaccharide, does not
inhibit thrombin and exhibits only factor Xa inhibitory
activity via binding to AT-III. However, because of superior
antithrombotic activity it became the first of a new class of
antithrombotic agents (Arixtra).4 Replacement of the N-
sulfate groups by O-sulfates has subsequently led to non-
Figure 1. Structure of the antithrombin binding domain of heparin
(1), synthetic antithrombotic agent fondaparinux (2), and the non-
glycosaminoglycan analogue idraparinux (3).
† Hungarian Academy of Sciences.
‡ University of Debrecen.
(1) Casu, B. AdV. Carbohydr. Chem. Biochem. 1985, 43, 51.
(2) Thunberg, L.; Backstro¨m, G.; Lindahl, U. Carbohydr. Res. 1982,
glycosaminoglycan analogues, such as idraparinux (3),
having simplified structure much easier to synthesize and
possessing increased anticoagulant activity and longer dura-
100, 393
(3) Jin, L.; Abrahams, J. P.; Skinner, R.; Petitou, M.; Pike, R. N. Proc.
Natl. Acad. Sci. U.S.A. 1997, 94, 14683
.
.
10.1021/ol900952d CCC: $40.75
Published on Web 05/27/2009
 2009 American Chemical Society

tion of action as compared to the natural pentasaccharide.5
Idraparinux can be regarded as a lead compound for the
development of second-generation synthetic antithrombotics.
Structure-activity relationship studies of synthetic ana-
logues of the heparin pentasaccharide revealed that four
sulfate groups and two carboxylate groups are essential for
the activation of AT-III, and an extra sulfate group at the
position O-3 of unit H improves the activity.4,6 The type of
charge is also crucial; an essential sulfate group cannot be
replaced by a phosphate, and the carboxylate groups may
not be exchanged for CH2OSO3- residues.4 Replacement of
the sulfate groups with isosteric sulfonatomethyl moieties
has not been investigated until now, although this replace-
ment may result in bioactive mimetics. It has been shown,
for example, that the isosteric phosphonate analogues7-9 of
mannose-6-phosphate binds with high affinity to the cation-
independent mannose 6-phosphate receptor. Isosteric sul-
fonate analogues of the AT-III binding pentasaccharide of
heparin may provide further information on structure-activity
relationship. Therefore, we decided to prepare analogues of
3, in which the sulfate esters are partially replaced with
sulfonatomethyl moieties. As a beginning of this program
we now describe the synthesis of three sulfonatomethyl
analogues of the EF fragment and three sulfonatomethyl
analogues of the GH fragment of compound 3.
reactions discussed for the synthesis of 7. For the preparation
of the 6-sulfonatomethylglucoside acceptors (14 and 15)
compound 1213 was oxidized, and then Wittig reaction of
the resulting 6-aldehyde afforded the 6,7-unsaturated hep-
toside 13. In the course of the addition of the radical anion
formed from NaHSO3 onto the double bond of 13, the
2-naphthylmethyl (NAP) group was also split off from
position 4, resulting in the sulfonic acid salt 14, which was
converted into the sulfonic acid ester 15 in two steps.
Synthesis of analogues of the glucuronic acid containing EF
fragment was carried out by means of coupling of the
sulfonic acid salt acceptors 6, 10, and 14 with acetobromo
glucose donor 16 to afford disaccharides 17, 18, and 19
(Scheme 2) and a subsequent oxidation into uronic acid at a
later stage of the synthesis.
Scheme 2. Glycosylation of Sulfonatomethyl Acceptors 6, 10,
and 14 with Acetobromo Glucose 16 To Obtain Fully Protected
Disaccharides 17, 18, and 19
The sulfonatomethyl glucoside acceptors in the form of
salts (6, 10, 14), as well as in the form of esters (7, 11, 15),
were synthesized as shown in Scheme 1. The 2-exomethylene
Scheme 1
.
Synthesis of Glucoside Acceptors Carrying a
Sulfonatomethyl Moiety
In the case of disaccharides 17 and 18, deacetylation
and 6′-O-triphenylmethylation followed by introduction
of the methyl groups into the secondary positions and
subsequent hydrolysis of the trityl moiety afforded com-
pounds 20 and 23, respectively. Oxidation of the primary
position of the glucose unit of 20 and 23 followed by
catalytic hydrogenation of the benzyl protecting groups
resulted in diols 21 and 24, which upon sulfation with
(4) van Boeckel, C. A. A.; Petitou, M. Angew. Chem., Int. Ed. Engl.
1993, 32, 1671.
(5) Westerduin, P.; van Boeckel, C. A. A.; Basten, J. E. M.; Broekhoven,
M. A.; Lucas, H.; Rood, A.; van der Heiden, H.; van Amsterdam, R. G. M.;
van Dinther, T. G.; Meuleman, D. G.; Visser, A.; Vogel, G. M. T.; Damm,
J. B. L.; Overklift, G. T. Bioorg. Med. Chem. 1994, 2, 1267.
(6) van Boeckel, C. A. A.; Petitou, M. Angew. Chem., Int. Ed. 2004,
32, 1671.
(7) Vidil, C.; More`re, A.; Garcia, M.; Barragan, V.; Hamdaoui, B.;
Rochefort, H.; Montero, J.-L. Eur. J. Org. Chem. 1999, 447.
(8) Berkowitz, D. B.; Maiti, G.; Charett, B. D.; Dreis, C. D.; Macdonald,
R. D. Org. Lett. 2004, 6, 4921.
derivative 410 was reacted with NaHSO3 to furnish the
desired 2-sulfonatomethyl-glucoside 5, in a stereoselective
radical addition,11 the moderate yield was a consequence of
partial hydrolysis of the benzylidene acetal under the slightly
acidic conditions. Regioselective opening of the 4,6-O-acetal
ring of 5 afforded acceptor 6 in salt form, which was
converted into the sulfonic acid ester 7 via a two-step
procedure involving liberation of the sulfonic acid with ion-
exchange resin followed by methylation with diazomethane.
The 3-sulfonatomethyl-glucoside acceptors (salt 10 and
methyl ester 11) were prepared from 812 by applying the
(9) Jeanjean, A.; Gary-Bobo, M.; Nirde´, P.; Leiris, S.; Garcia, M.;
More`re, A. Bioorg. Med. Chem. Lett. 2008, 18, 6240.
(10) Sarda, P.; Olesker, A.; Luka´cs, G. Carbohydr. Res. 1992, 229, 161.
(11) La´za´r, L.; Csa´va´s, M.; Borba´s, A.; Gye´ma´nt, Gy.; Lipta´k, A. ArkiVoc
2004, Vii, 196.
(12) Yoshimura, J.; Kawauchi, N.; Yasumori, T.; Sato, K.; Hashimoto,
H. Carbohydr. Res. 1984, 133, 255.
(13) Borba´s, A.; Szabo´, Z. B.; Szila´gyi, L.; Be´nyei, A.; Lipta´k, A.
Tetrahedron 2002, 58, 5723.
(14) Tronchet, J. M. J.; Eder, H. HelV. Chim. Acta 1978, 61, 2254.
(15) Jacquinet, J. C.; Petitou, M.; Duchaussoy, P.; Lederman, I.; Choay,
J.; Torri, G.; Sinay¨, P. Carbohydr. Res. 1984, 130, 221.
2620
Org. Lett., Vol. 11, No. 12, 2009

the SO3·pyridine complex furnished the target compounds
22 and 25 with high overall yields (Scheme 3). Disac-
the donor 32 to afford the desired R-linked iduronic acid
containing disaccharides 33, 34, and 35 (Scheme 5). Syn-
Scheme 5. Glycosylation of Methylsulfonatomethyl Acceptors 7,
Scheme 3
. Synthesis of 2-, 3-, and 6-Sulfonatomethyl
11, and 15 with Donor 32 To Obtain Fully Protected
Disaccharides 33, 34, and 35
Analogues of the EF Disaccharide
thesis of the target GH analogues involving introduction of
two methyl groups into the iduronic acid unit and two sulfate
ester functions into the glucose unit was attempted as
depicted in Scheme 6, to furnish the 3- and 6-sulfonatomethyl
disaccharides 39 and 41 with good yields.
charide 19 was also deacetylated; however, subsequent
tritylation was unsuccessful. Therefore, the tetrahydroxy
compound was converted into 26 by selective oxidation.
Methylation of compound 26 carried a risk of ꢀ-elimina-
tion of the base-sensitive glucuronic residue. However, it
proceeded smoothly, and no elimination could be ob-
served. Removal of the benzyl groups by catalytic hydro-
genation afforded compound 27 quantitatively. Subsequent
sulfation of the diol 27 gave the third analogue 28 of the
EF fragment (Scheme 3). For the synthesis of the GH
analogues, L-iduronic acid trichloroacetimidate 32 was
chosen as the donor prepared from 2914 by a two-step
epimerization of C-515 (30), followed by deisopropylide-
nation, acetylation (31), selective deacetylation of the
anomeric position, and imidate formation (Scheme 4).
In the case of the 2-sulfonatomethyl derivative, however,
upon methylation an inseparable mixture of the desired
Scheme 6
.
Synthesis of Sulfonatomethyl Analogues (39 and 41)
of the GH Disaccharide
Scheme 4. Synthesis of the L-Iduronic Acid Donor
product 36 and the byproduct 37 (as a result of ꢀ-elimination
of the base-sensitive iduronic acid residue) were formed in
a ratio of 3:1.
To overcome this difficulty, compound 33 was deacety-
lated, and the resulting diol was methylated with diaz-
omethane in the presence of silica gel as catalyst16 to give
42 with high yield. The sulfonic ester was converted into
salt with sodium iodide, and subsequent treatment of
the crude product with sodium hydroxide solution gave
the disodium salt, which was then debenzylated, and the
resulting diol was sulfated to afford the desired analogue
43 (Scheme 7).
The sulfonatomethyl glucoside acceptors were then gly-
cosylated in form of the methyl esters 7, 11, and 15 with
(16) Nakata, T.; Nagao, S.; Mori, N.; Oishi, T. Tetrahedron Lett. 1985,
26, 6461.
Org. Lett., Vol. 11, No. 12, 2009
2621

duced by NaHSO3 addition onto the exomethylene moiety
of the appropriate glycoside derivatives. The glycosylation
reactions were carried out either with sulfonic acid salt or
with sulfonic acid ester acceptors. The carboxylic function
of the uronic acid unit was formed on a disaccharide level
in the case of the EF analogues but on a monosaccharide
level in the case of the GH analogues. Methylation reactions
were carried out both under basic and acidic conditions.
Synthesis of tri- and pentasaccharide analogues applying the
presented methods is underway.
Scheme 7
.
Synthesis of the 2-Sulfonatomethyl Analogue of the
GH Disaccharide
Acknowledgment. This work was supported by the
Hungarian Research Fund (OTKA NK 48798 to A.L. and
K 62802 to A.B.)
Supporting Information Available: Experimental pro-
cedures, characterization data and selected NMR spectra.
This material is available free of charge via the Internet at
http://pubs.acs.org
In conclusion, we synthesized three sulfonatomethyl
analogues of the EF fragment and three sulfonatomethyl
analogues of the GH fragment of the nonglycosaminoglycan
blood-anticoagulant 3. The sulfonic acid group was intro-
OL900952D
2622
Org. Lett., Vol. 11, No. 12, 2009