Synthesis of a new transition-state analog of the sialyl donor. Inhibition of sialyltransferases

Article abstract of DOI:10.1016/S0040-4039(01)00204-0

A new class of glycosyltransferase inhibitor has been designed and synthesized. The designed inhibitors 3a/3b provide conformational mimicry of the transition state in sialyltransfer reactions. The key synthetic steps involve a Meinwald rearrangement and a palladium-catalyzed carbonylation reaction. The results of kinetic studies show that 3a/3b exhibit significant inhibition on both 2,3- and 2,6-sialytransferases.

Full text of DOI:10.1016/S0040-4039(01)00204-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 2451–2453
Synthesis of a new transition-state analog of the sialyl donor.
Inhibition of sialyltransferases
Hongbin Sun, Jingsong Yang, Katie E. Amaral and Benjamin A. Horenstein*
Department of Chemistry, University of Florida, Gainesville, FL 32611-7200, USA
Received 22 December 2000; accepted 2 February 2001
Abstract—A new class of glycosyltransferase inhibitor has been designed and synthesized. The designed inhibitors 3a/3b provide
conformational mimicry of the transition state in sialyltransfer reactions. The key synthetic steps involve a Meinwald rearrange-
ment and a palladium-catalyzed carbonylation reaction. The results of kinetic studies show that 3a/3b exhibit significant inhibition
on both 2,3- and 2,6-sialytransferases. © 2001 Published by Elsevier Science Ltd.
Sialic acids play an important role in a variety of
biological processes.1 They are usually attached to the
terminal positions of glycoproteins, glycolipids and
oligosaccharides. Of more than 100 different sialic
acids, N-acetylneuraminic acid (NeuAc) is the most
abundant one.2 The sugar-nucleotide CMP-NeuAc 1
(Fig. 1) is the key substrate for biosynthesis of sialy-
lated glycoconjugates in which 1 is transferred by sialyl-
transferases to an acceptor hydroxyl group in a variety
of substrates including polysialic acids, glycoproteins
and gangliosides.3 These glycosylation patterns consti-
tute important binding determinants for various cell–
as characterized by kinetic isotope effects which
strongly support an oxocarbenium ion-like transition
state structure 2 for both solvolysis and the sialyltrans-
ferase catalyzed reactions.6 According to this model,
positive charge is accumulated at C-1% and O-1% and the
carboxylate group approaches coplanarity with the
oxocarbenium ion plane. A common strategy often
used to design inhibitors of glycosidases and/or glyco-
syltransferases is to mimic the electronic charge distri-
bution of a proposed transition state.7 On the other
hand, we noted that in addition to the positive charge
and the coplanarity developed in 2, the increased dis-
tance between the leaving CMP group and C-1% might
be another useful factor for inhibitor design. As a test
for this distinct concept, we designed compound 3a, in
which the conjugated carboxylate group mimics the
oxocarbenium ion coplanarity in transition state 2, and
moreover, the CMP group attached to the bicyclic
structure is held at a distance to mimic a late transition
state for bond cleavage. Here we wish to report the
cell interactions
which include masking of
trypanosomal immunogenicity, viral infection and repli-
cation, and cell adhesion.4 Therefore, development of
potent and selective inhibitors of sialyltransferases may
be useful in a variety of biochemical applications. How-
ever, only a few potent inhibitors of sialyltransferases
have so far been developed.5 We have previously
reported the solvolytic behavior and enzymology of 1,
Figure 1.
Keywords: transition states; inhibitors; glycosyltransferases; sialytransferases; synthesis.
* Corresponding author. Tel.: (352) 392-9859; fax: (352) 846-2095; e-mail: horen@chem.ufl.edu
0040-4039/01/$ - see front matter © 2001 Published by Elsevier Science Ltd.
PII: S0040-4039(01)00204-0

2452
H. Sun et al. / Tetrahedron Letters 42 (2001) 2451–2453
synthesis and inhibitory activities of compounds 3a and
3b (diastereoisomer of 3a).
of N-acetyl-cytidine11b with 4,4%-dimethoxytrityl chlo-
ride in pyridine, followed by acetylation of the 2%,3%-
hydroxyl groups with acetic anhydride in pyridine and
deprotection of the 5%-hydroxyl group with 90%
aqueous acetic acid, gave the required triacetate in 41%
overall yield. The coupling of 9 with the cytidine 5%-
phosphoramidite in the presence of tetrazole, followed
by oxidation with tert-butyl hydroperoxide and base-
catalyzed removal of the cyanoethyl group gave a
diastereomeric mixture of 10a and 10b. Saponification
of 10a/10b with sodium methoxide in methanol–water
afforded a mixture of 3a and 3b (1:1 by NMR), which
was purified by anion-exchange HPLC (NH4HCO3–
MeOH–H2O buffer as eluent).12 Our efforts to separate
the diastereoisomers 3a and 3b were unsuccessful, so
the mixture of 3a/3b was used for the inhibition studies.
We envisioned that the bicyclic structure of 3a could be
established by Meinwald rearrangement of norbornadi-
ene 4.8 With this in mind, we initiated a synthetic
sequence, as shown in Scheme 1. Meinwald rearrange-
ment of norbornadiene 4 with m-chloroperoxybenzoic
acid and subsequent reduction with sodium borohy-
dride gave racemic 6-endo-hydroxymethyl bicyclo-
[3.1.0]hex-2-ene. Protection using tert-butyldimethyl-
silyl chloride or tert-butyldiphenylsilyl chloride in the
presence of imidazole proceeded in 85–90% yields.
Hydroboration–oxidation furnished the desired alcohol
5a and its regioisomer 11a in high yields (5a:11a:5:2,
by NMR spectroscopy).9 The ratio of diastereomers
produced by hydroboration was insensitive to the O-
protecting group. Stereoisomers 5a/11a or 5b/11b could
be separated by careful column chromatography. Oxi-
dation of 5a or 5b with chromium trioxide-pyridine
gave ketone 6a (78%) or ketone 6b (86%), respectively.
Alternatively, a mixture of 5b/11b was used for the
oxidation reaction to give the desired ketone 6b, which
could be easily separated from its regioisomer by flash
column chromatography. Reaction of ketone 6b with
trifluoromethanesulfonic anhydride using 2,6-di-tert-
butyl-4-methylpyridine as a base afforded vinyl triflate
7 in 69% yield. Palladium-catalyzed carbonylation of 7
in methanol proceeded smoothly to form the unsatu-
rated ester 8 in 80% yield.10 Deprotection of 8 with
tetrabutylammonium fluoride (TBAF) gave the key
intermediate 9 in 85% yield.
The inhibitory effects of 3a/3b on rat 2,3- and 2,6-sialy-
transferases were investigated with the use of radiola-
beled [9-3H] CMP-NeuAc as a donor substrate. The
acceptor sugars employed were lactose (for 2,3-sialy-
transferase) and lacNac (for 2,6-sialytransferase). The
results show that 3a/3b are competitive inhibitors of
both 2,3- and 2,6-sialyltransferases, and the Kis were
estimated to be 10 and 20 mM for 2,3- and 2,6-sialy-
transferases, respectively. These results provide further
impetus for development of new analogs that incorpo-
rate additional features of the enzyme-bound transition
state.
In summary, we report the design and synthesis of a
new class of sialyltransferase inhibitors which mimic
transition state geometry. The key synthetic steps
involve a Meinwald rearrangement and a palladium-
catalyzed carbonylation reaction. Despite lacking some
of the structural features (e.g. C6–C9 glycerol side
chain) of the natural sialyltransferase substrate, CMP-
NeuAc, the designed inhibitors 3a/3b showed signifi-
cant inhibition on both 2,3- and 2,6-sialytransferases.
Cytidine-2%,3%-di-O-acetyl-4-N-acetyl-5%-(N,N-diisopro-
pyl-2-cyanoethyl)-phosphoramidite was prepared from
the triacetyl cytidine11a by the literature method.11c We
found that a more facile synthesis of the triacetyl
cytidine could be carried out with the following one-pot
sequence. Selective protection of the 5%-hydroxyl group
Scheme 1. Reagents and conditions: (a) (i) m-CPBA; (ii) NaBH4; (iii) TBDMSCl or TBDPSCl, imidazole. (b) (i) BH3·SMe2; (ii)
H2O2, NaOH, 90–95%. (c) CrO3, pyridine, 78% for 6a; 86% for 6b. (d) Tf2O, 2,6-di-tert-butyl-4-methylpyridine, 69%. (e)
Pd(OAc)2, PPh3, Et3N, MeOH, CO, DMF, rt, 80%. (f) TBAF, THF, 85%. (g) (i) Cytidine-2%,3%-di-O-acetyl-4-N-acetyl-5%-(N,N-
diisopropyl-2-cyanoethyl)-phosphoramidite, 1H-tetrazole, CH2Cl2; (ii) t-BuOOH; (iii) Et3N, CH2Cl2, 51% for three steps. (h) (i)
NaOMe, MeOH–H2O; (ii) HPLC (MonoQ, NH4HCO3–MeOH–H2O); (iii) desalting with Amberlite IR-120, 30%.

H. Sun et al. / Tetrahedron Letters 42 (2001) 2451–2453
2453
We expect that analogs of 3a/3b will exhibit more
potent and selective inhibition on sialyltransferases, and
may be extended to other glycosyltransferases.
6. (a) Horenstein, B. A.; Bruner, M. J. Am. Chem. Soc.
1996, 118, 10371; (b) Horenstein, B. A. J. Am. Chem.
Soc. 1997, 119, 1101; (c) Bruner, M.; Horenstein, B. A.
Biochemistry 1998, 37, 289; (d) Horenstein, B. A.; Bruner,
M. Biochemistry 2000, 39, 2261.
Acknowledgements
7. (a) Sinnott, M. L. Chem. Rev. 1990, 90, 1171; (b)
McCarter, J. D.; Withers, S. G. Curr. Opin. Struct. Biol.
1994, 4, 885.
We wish to thank the University of Florida and
National Science Foundation (CAREER Award MCB-
9501866 to B.A.H.) for this work.
8. Gilbert, J. C.; Smith, K. R. J. Org. Chem. 1976, 41, 3883.
9. No attempt was made to determine the stereochemistry
of the hydroxyl group relative to the cyclopropane ring,
but it was assumed that the borane attack occurred from
the less hindered side and no change in configuration
took place during the peroxide oxidation.
References
10. Cacchi, S.; Morera, E.; Ortar, G. Tetrahedron Lett. 1985,
26, 1109.
1. (a) Rosenberg, A. Biology of Sialic Acids; Plenum: New
York, 1995; (b) Reutter, W.; Stasche, R.; Stehling, P.;
Baum, O. In Glycosciences; Gabius, H. J.; Gabius, S.,
Eds.; Chapman & Hall: Weinheim, 1997; pp. 245–259.
2. Gottschalk, A. Nature 1951, 167, 845.
3. (a) Harduin-Lepers, A.; Recchi, M. A.; Delannoy, P.
Glycobiology 1995, 5, 741; (b) Varki, A. Glycobiology
1992, 2, 25; (c) Reglero, A.; Rodriguez-Aparicio, L. B.;
Lueng, J. M. Int. J. Biochem. 1993, 11, 1517; (d) Hayes,
B. K.; Varki, A. J. Biol. Chem. 1993, 268, 16155.
4. (a) Lowe, J. B. In Molecular Glycobiology; Fucuda, M.;
Hindsgaul, O., Eds.; Oxford University Press: New York,
1994; pp. 163–205; (b) Powell, L. D.; Varki, A. J. Biol.
Chem. 1995, 270, 14243; (c) Crocker, P. R.; Kelm, S.;
Hartnell, A.; Freeman, S.; Nath, D.; Vinson, M.; Muck-
low, S. Biochem. Soc. Trans. 1996, 24, 150.
11. (a) Zielinski, W. S.; Smrt, J.; Beranek, J. Coll. Czech.
Chem. Commun. 1974, 39, 3560; (b) Watanabe, K. A.;
Fox, J. J. Angew. Chem., Int. Ed. 1966, 6, 579; (c)
Kajihara, Y.; Ebata, T.; Koseki, K.; Kodama, H.; Mat-
sushita, H.; Hashimoto, H. J. Org. Chem. 1995, 60, 5732.
1
12. Selected data for 9: H NMR (300 MHz, CD3Cl) l 6.75
(m, 1H), 3.71 (s, 3H), 3.53 (m, 1H), 3.36 (m, 1H), 2.76
(m, 1H), 2.45 (m, 1H), 2.20 (m, 1H), 1.89 (m, 1H), 1.48
(m, 1H), 1.38 (bs, 1H); 13C NMR (75 MHz, CD3Cl) l
164.87, 141.61, 135.68, 56.31, 51.39, 31.14, 28.88, 24.10,
21.38; HRMS calcd for C9H13O3 (M++H): 169.0865;
1
found: 169.090. For 3a/3b: H NMR (300 MHz, D2O) l
7.75 (d, J=7.62 Hz, 1H), 6.29 (bs, 1H), 5.88 (d, J=7.62
Hz, 1H), 5.75 (m, 1H), 4.45 (m, 3H), 3.93 (m, 1H), 3.83
(m, 1H), 3.58 (m, 1H), 3.44 (m, 1H), 2.45 (m, 1H),
2.20–2.18 (m, 2H), 1.69 (m, 1H), 1.28 (m, 1H); 13C NMR
(75 MHz, D2O) l 181.87, 173.91, 166.47, 158.12, 142.10,
141.61, 137.61, 96.82, 89.63, 89.52, 83.01, 74.46, 69.70,
64.46, 61.68, 31.96, 28.83, 23.65, 21.01; HRMS calcd for
C17H21N3O10 P (M+−2Na++H): 458.0965; found: 458.087.
5. (a) Schaub, C.; Muller, B.; Schmidt, R. R. Glycoconjugate
J. 1998, 15, 345; (b) Muller, B.; Schaub, C.; Schimdt, R.
R. Angew. Chem., Int. Ed. 1998, 37, 2893; (c) Amann, F.;
Schaub, C.; Muller, B.; Schmidt, R. R. Chem. Eur. J.
1998, 4, 1106.
.
.