Synthesis of novel sialylmimetics as biological probes

Article abstract of DOI:10.1016/S0960-894X(01)00276-1

Glycomimetics are increasingly being recognised as powerful tools in the search for novel compounds that possess useful biological properties. This paper describes our preliminary efforts towards the development of novel mimetics of sialic acid thioglycosides. These sialylmimetics are readily prepared and have been shown, in some instances, to have biological properties similar to sialic acid thioglycosides. Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(01)00276-1

Bioorganic& Medicinal Chemistry Letters 11 (2001) 1587–1590
Synthesis of Novel Sialylmimetics as Biological Probes
Susan J. Bradley,b Ashmath Fazli,b Milton J. Kiefela,b and Mark von Itzsteina,b,
*
aCentre for Biomolecular Science and Drug Discovery, Griffith University (Gold Coast Campus),
PMB 50 Gold Coast Mail Centre, Queensland, 9726, Australia
bDepartment of Medicinal Chemistry, Monash University (Parkville Campus), 381 Royal Parade, Parkville, Victoria, 3052, Australia
Received 20 December 2000; accepted 11 April 2001
Abstract—Glycomimetics are increasingly being recognised as powerful tools in the search for novel compounds that possess useful
biological properties. This paper describes our preliminary efforts towards the development of novel mimetics of sialic acid thio-
glycosides. These sialylmimetics are readily prepared and have been shown, in some instances, to have biological properties similar
to sialic acid thioglycosides. # 2001 Elsevier Science Ltd. All rights reserved.
Sialicacids are involved in a number of important bio-
logical processes including cell–cell communication and
cell–microbe interactions.1À4 The importance of sialic
acids in these processes, especially with respect to
human disease states, has led to interest in the synthesis
of natural and modified sialicacids 5À7 both as probes of
sialic acid-recognising proteins, and as potential glyco-
pharmaceuticals. The complexity of sialic acid chem-
istry, particularly in the areas of functional group
manipulations and glycosidation reactions,5,7 together
with the inherently poor pharmacological properties of
carbohydrate-based compounds has resulted in recent
interest in the development of mimetics of sialic
acids.8À12 These sialylmimetics are designed to retain the
structural features essential for interaction with a parti-
cular protein, but are structurally simpler compounds
with potentially improved pharmacological profiles.
Some excellent examples of the success of sialylmimetics
as potential pharmaceuticals include work directed
towards the development of mimetics of sialyl Lewis x
(1), particularly those designed to inhibit E-selectin.
This has resulted in several compounds (e.g., 2), that
exhibit comparable affinity for E-selectin when com-
pared to sialyl Lewis x.8À10,13 It is pertinent to note that,
in addition to significant backbone alterations, the sia-
lylmimetic 2 contains a simple carboxylate group in
place of the entire sialic acid moiety present in sialyl
Lewis x. Considerable effort has also been directed
towards the development of sialylmimetics as inhibitors
of sialidases.14À17 This work has focused on mimetics of
2,3-didehydrosialic acids and includes the carbocyclic
derivative 3,14 a potent inhibitor of influenza virus sia-
lidase, and mimetics based on uronic acids,15À17 such as
compounds of the general structure 4.
As part of our continued interest in sialic acid chemistry
and biochemistry, we have been interested in further
developing methods that provide ready access into a
range of related sialylmimetics that would serve as
probes for a number of sialic acid-recognising proteins.
We have also had a long term interest in a-(2,6)-linked
thiosialosides typified by 5, both as metabolically stable
biological
probes
for
sialic
acid-recognising
proteins,18À20 and as potential inhibitors of rotaviral
infection.21 Our interest in the development of novel
sialylmimetics prompted us to investigate the synthesis
of sialylmimetics of the general structure 6, which can
be considered as mimetics of thiosialosides like 5.
*Corresponding author at first address. Fax: +61-7-5552-8098;
e-mail: m.vonitzstein@mailbox.gu.edu.au
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00276-1

1588
S. J. Bradley et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1587–1590
commercially available, which facilitates the ease of
introduction of the variable ‘R’ group. Comparison of
the two routes towards sialylmimetics like 6 depends on
the relative ease of preparation of the respective thiol-
acetyl derivatives 8 and 10, and the efficiency of the
subsequent coupling reactions. To determine the most
effective and flexible approach towards sialylmimetics of
the general structure 6, we have examined both routes
towards such compounds.
As can be seen from the proposed sialylmimetic 6, the
entire sialic acid portion has been replaced by a car-
boxylate functionality and a variable group ‘R’. Our
intention in opting to prepare sialylmimetics of the
general structure 6 was to investigate the effects of
replacing the entire sialic acid moiety with a group that
maintains the charge present in thiosialosides. In addi-
tion, the variable ‘R’ group in 6 would allow us the
opportunity to explore the effects of hydrophobicity,
hydrophilicity or steric bulk, on a range of sialic acid-
recognising proteins. Importantly, the carboxylate
functionality of 6 is in the same relative position as the
sialic acid carboxylate group in thiosialosides like 5.
Others have shown that replacement of an entire sialic
acid moiety with a carboxylate group can lead to com-
pounds that have improved (or in some cases compar-
able) affinity for particular proteins.8
As depicted in Scheme 1 the synthesis of sialylmimetics
via ‘route-a’ involves coupling between a C-6 activated
carbohydrate derivative (i.e., 7) and the a-thiolacetyl
ester derivative 8. We have previously described the
synthesis of the 6-bromo-6-deoxy glucoside derivative
11 from methyl b-d-glucopyranoside.18,21 The appro-
priate coupling partner was prepared by treatment of
commercially available ethyl 2-bromopropionate with
potassium thioacetate (2.5 equiv) in acetone at room
temperature for 18 h, to give the desired a-thiolacetyl
derivative 12 in 82% yield after distillation (Kugelrohr
bp 70–75 ꢀC @ 5 mmHg). Following our earlier studies
towards the synthesis of thiosialosides,18,21 a mixture of
the 6-bromo-6-deoxy glucoside derivative 11 and the
thiolacetyl ester derivative 12 was treated with Et2NH in
N,N-dimethylformamide at room temperature under
N2. Coupling was found to be extremely slow, with the
desired sialylmimetic 13 obtained in a moderate 62%
yield after 2 days.
From a retrosyntheticviewpoint, there are two distinct
pathways towards sialylmimetics of the general struc-
ture 6 (Scheme 1). The first approach (‘route-a’) involves
coupling between a C-6 activated carbohydrate like 7
and the a-thiolacetyl carboxylate derivative 8. The
advantage of this route is that it is analogous to the
approach that we have previously used to prepare thio-
sialosides like 5,18,21 and as such we already have access
to the requisite C-6 activated carbohydrates 7. How-
ever, a potential drawback of ‘route-a’ is that every time
we wish to alter the variable ‘R’ group in sialylmimetics
like 6, we would have to prepare the requisite a-thiol-
acetyl ester derivative 8.
Identification of the product from the coupling reaction
between 11 and 12 was evident from examination of the
1H NMR spectrum that showed characteristic signals
for the H-6 protons at d 2.72 and d 2.89 indicating they
are adjacent to sulfur. The remainder of the resonances in
the 1H NMR spec trum of13 are consistent with the struc-
ture as shown. Attempts at improving the efficiency of the
coupling between 11 and 12 by raising the reaction tem-
perature to 45 ꢀC resulted in extensive decomposition of
starting materials. Furthermore, attempted coupling
between the a-thiolacetyl ester derivative 12 and the 6-O-
trifluoromethanesulfonyl galactoside derivative 1421 resul-
ted in a complex reaction product containing none of the
desired galactose-based sialylmimetic 15. Interestingly,
and contrary to our previous studies towards the synthesis
of thiosialosides,21 exposure of the 6-bromo-6-deoxy-
galactoside derivative 16 to 12 in N,N-DMF containing
Et2NH at 25 ꢀC smoothly furnished 15 in 71% yield.
The alternative approach (‘route-b’) involves introduction
of the thiolacetyl group onto the carbohydrate portion of
the mimeticbefore coupling with an a-halo-ester derivative
(e.g., 9). Requisite C-6 thiolacetylated carbohydrates of
the general structure 10 would be directly available from
C-6 activated carbohydrates like 7. The potential
advantage of ‘route-b’ is that several a-halo-esters are
For the alternative approach (‘route-b’ in Scheme 1) to be
a viable strategy towards sialylmimetics of the general
Scheme 1. Retrosyntheticapproach towards sialylmimetics.

S. J. Bradley et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1587–1590
1589
structure 6, we required an efficient synthesis of C-6
thiolacetyl derivatives like 10. Accordingly, treatment of
the previously described18,21 6-bromo-6-deoxy deriva-
tives 11 and 16 with potassium thioacetate furnished the
corresponding 6-thiolacetyl derivatives 17 and 18,
respectively, in >90% yield.
Although the sialylmimetics 13, 15, and 21–24 are mix-
tures of diastereomers, we felt they would serve as useful
biological probes for proteins involved in sialic acid
recognition processes. As part of our continued interest
in the development of a-(2,6)-linked thiosialosides as
potential inhibitors of rotavirus,21 we undertook a pre-
liminary investigation of the rotavirus inhibitory activ-
ity of some of the sialylmimetics described above.
Although there is some conjecture as to if cell-surface
sialic acid residues are essential for recognition by rota-
virus, the current consensus is that most strains of
rotavirus require sialic acid for the recognition pro-
cess.25 Despite reports suggesting that partial O-acet-
ylation in the glycerol side chain of sialic acid glycosides
results in compounds with improved inhibition of some
animal strains of rotavirus,21,26 it remains to be proven
if sialylmimetics like those described above can be
recognised by rotavirus. As part of our preliminary
investigations towards this end, treatment of the sia-
lylmimetics 13 and 15 with dilute NaOH (1.0 M) gave
the deprotected sialylmimetics 25 and 26, in ꢁ70%
yield after HPLC purification.
Although this approach had provided access to the
desired compounds, the 6-bromo-6-deoxy derivatives 11
and 16 had themselves been derived from the corre-
sponding 6-O-p-toluenesulfonyl derivatives 19 and
20.18,21 We therefore decided a more efficient route to
the C-6 thiolacetyl derivatives 17 and 18 would be
directly from the 6-O-p-toluenesulfonyl derivatives 19
and 20. Towards this end, exposure of the 6-O-p-tolue-
nesulfonyl derivatives 19 and 20 to potassium thioace-
tate in acetone at reflux22 afforded the C-6 thiolacetyl
derivatives 17 and 18, respectively, in ꢁ80% yield after
chromatography (Scheme 2). The yield for the intro-
duction of the thiolacetyl group is slightly lower for
tosylate displacement versus bromide displacement.
However, the overall sequence from commercially
available material is more efficient, since one step (bro-
mide displacement of the 6-O-p-toluenesulfonyl group
to give the 6-bromo-6-deoxy derivative) has been elimi-
nated and the overall chemical yield is comparable.
Analysis of the sialylmimetics 25 and 26 as inhibitors of
rotaviral infection involved the use of a standard in
vitro neutralisation assay as we have described pre-
viously.21 Neither sialylmimetics 25 nor 26 showed any
inhibition against the NCDV (bovine) or the Wa
(human) strains of rotavirus (IC50 >25 mM for 25 and
26 against both strains).27 Previously we have reported
that the thiosialoside 5 shows some inhibition (IC50
6.25 mM21) against the NCDV (bovine) strain of rota-
virus. It remains to be determined why the sialylmim-
etics 25 and 26 do not inhibit rotavirus. It may be that,
unlike the interaction of sialyl Lewis x mimetics such as
2 with selectins, the rotavirus recognition/adhesion pro-
cess requires interaction with sialic acid functional
groups in addition to the carboxylate moiety. The
incorporation of hydrophilic groups at ‘R’ in sialyl-
mimetics like 6 may prove interesting in this regard.
Alternatively, the sialylmimetics 25 and 26 are only
monosaccharides and therefore may not carry sufficient
non-sialic acid carbohydrate information for recogni-
tion by the rotavirus adhesion protein. It is worth not-
ing that the E-selectin inhibitor, sialylmimetic 2, does
incorporate additional carbohydrate information by
virtue of the fucose moiety.
Scheme 2. Synthesis of C-6 thiolacetyl derivatives. Reagents and con-
ditions: (a) TsCl (1.1 equiv), pyridine, 0 ꢀC, 2.5 h; (b) Ac2O, pyridine,
15 h; (c) KSAc (2.5 equiv), acetone, 55 ꢀC, 48 h.
With the requisite C-6 thiolacetyl derivatives 17 and 18
readily available, our attention turned to their coupling
to commercially available a-bromo-ester derivatives.
Attempts at coupling the C-6 thiolacetyl glucoside deri-
vative 17 and ethyl 2-bromopropionate, utilising our
Et2NH mediated coupling procedure,18 resulted in only
a modest yield (ꢁ50%) of the desired sialylmimetic 13
even after extended reaction times (2 days) at elevated
temperature (45 ꢀC). Fortunately, hydrazine acetate
mediated coupling23,24 between the C-6 thiolacetate
derivatives 17 and 18 and ethyl 2-bromopropionate
afforded the sialylmimetics 13 and 15, respectively, in
high yield. In the same way, the sialylmimetics 21–24
were prepared by coupling the C-6 thiolacetyl derivatives
17 and 18 with the corresponding commercially available
a-bromo-esters. In all cases the purified products were
obtained in yields above 80% after chromatography.
In summary, we have described a simple and efficient
synthesis of mimetics of a-(2,6)-linked thiosialosides.
The method described provides maximum flexibility in
terms of the introduction of alternative functionality in
the unit that replaces the sialic acid moiety, and utilises
readily accessible starting materials. Although we are
yet to demonstrate the usefulness of such sialylmimetics
as probes for sialic acid-recognising proteins, we are
currently investigating sialylmimetics with different

1590
S. J. Bradley et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1587–1590
functionality at the variable ‘R’ group, as well as alter-
native mono- and disaccharide templates. Results from
the analysis of further sialylmimetics as inhibitors of
rotaviral infection, and as probes for sialic acid-recog-
nising proteins, will be published in due course.
13. Thoma, G.; Magnani, J. L.; Patton, J. T. Bioorg. Med.
Chem. Lett. 2001, 11, 923.
14. Kim, C. U.; Lew, W.; Williams, M. A.; Liu, H.; Zhang,
L.; Swaminathan, S.; Bischofberger, N.; Chen, M. S.; Mendel,
D. B.; Tai, C. Y.; Laver, W. G.; Stevens, R. C. J. Am. Chem.
Soc. 1997, 119, 681.
15. Florio, P.; Thomson, R. J.; Alafaci, A.; Abo, S.; von Itz-
stein, M. Bioorg. Med. Chem. Lett. 1999, 9, 2065.
16. Florio, P.; Thomson, R. J.; von Itzstein, M. Carbohydr.
Res. 2000, 328, 445.
Acknowledgements
Associate Professor Ian Holmes and Ms. Clare Jolly
(University of Melbourne) are thanked for their assis-
tance with the rotavirus assays. Helpful discussions with
Dr. Robin Thomson are greatly appreciated. The ARC
is gratefully acknowledged for the provision of funding,
as is Monash University for the award of a Scholarship
to SJB.
17. Smith, P. W.; Robinson, J. E.; Evans, D. N.; Sollis, S. L.;
Howes, P. D.; Trivech, N.; Bethell, R. C. Bioorg. Med. Chem.
Lett. 1999, 9, 601.
18. Bennett, S.; von Itzstein, M.; Kiefel, M. J. Carbohydr. Res.
1994, 259, 293.
19. Wilson, J. C.; Kiefel, M. J.; Angus, D. I.; von Itzstein, M.
Org. Lett. 1999, 1, 443.
20. Ciccotosto, S.; Kiefel, M. J.; Abo, S.; Stewart, W.; Quelch,
K.; von Itzstein, M. Glycoconjugate J. 1998, 15, 663.
21. Kiefel, M. J.; Beisner, B.; Bennett, S.; Holmes, I. D.; von
Itzstein, M. J. Med. Chem. 1996, 39, 1314.
References and Notes
22. Robina, I.; Go
Fuentes, J. Synth. Commun. 1998, 28, 2379.
23. Park, W. K. C.; Meunier, S. J.; Zanini, D.; Roy, R. Car-
bohydr. Lett. 1995, 1, 179.
˜
mez-Bujedo, S.; Fernandez-Bolanos, J. G.;
´ ´
1. Schauer, R.; Kamerling, J. P. In Glycoproteins II; Mon-
treuil, J., Vliengenthart, J. F. G., Schachter, H., Eds.; Elsevier:
Amsterdam, 1997; pp 243–402.
2. Biology of the Sialic Acids; Rosenberg, A., Ed.; Plenum:
New York, 1995.
3. Varki, A. Glycobiology 1993, 3, 97.
4. Crocker, P. R.; Feizi, T. Curr. Opin. Struct. Biol. 1996, 6,
679.
5. Zbiral, E. In Carbohydrates—Synthetic Methods and
Applications in Medicinal Chemistry; Ogura, H., Hase-
gawa, A., Suami, T., Eds.; VCH: Weinheim, 1992; pp
304–339.
24. General procedure for hydrazine acetate mediated cou-
plings: A solution of the C-6 thiolacetyl derivative (1.0 mmol)
in dry DMF (10 mL) was degassed for 20 min by bubbling N2
ꢂ
into the solution. H2NNH2 AcOH (1.15 mmol) was added and
the solution stirred for 30 min before the addition of the a-
bromo-ester (1.1 mmol) and Et3N (1.15 mmol). After 4 h
under N2 the mixture was diluted with EtOAc(20 mL) and
washed with dil HCl (1 M, 20 mL), H2O (20 mL), dried
(Na2SO4), and concentrated.
6. von Itzstein, M.; Thomson, R. J. Top. Curr. Chem. 1997,
186, 119.
7. von Itzstein, M.; Kiefel, M. J. In Carbohydrates in Drug
Design; Witczak, Z. J., Nieforth, K. A., Eds.; Marcel Dekker:
New York, 1997; pp 39–82.
25. See for example: Guo, C.-T.; Nakagomi, O.; Mochizuki,
M.; Ishida, H.; Kiso, M.; Ohta, Y.; Suzuki, T.; Miyamoto, D.;
Hidari, K. I.-P. J.; Suzuki, Y. J. Biochem. 1999, 126, 683.
Delorme, C.; Brussow, H.; Sidoti, J.; Roche, N.; Karlsson, K.-
A.; Neeser, J.-R.; Teneberg, S. J. Virol. 2001, 75, 2276.
26. Willoughby, R. E.; Yolken, R. H. J. Infect. Dis. 1990, 161,
116.
8. Sears, P.; Wong, C.-H. Angew. Chem., Int. Ed. 1999, 38,
2300.
9. Kretzschmar, G. Tetrahedron 1998, 54, 3765.
10. Huwe, C. M.; Woltering, T. J.; Jiricek, J.; Weitz-Schmidt,
G.; Wong, C.-H. Bioorg. Med. Chem. 1999, 7, 773.
11. Roy, R. In Carbohydrates in Drug Design; Witczak, Z. J.,
Nieforth, K. A., Eds.; Marcel Dekker: New York, 1997; pp
83–135.
27. Each sialylmimetic was incubated with a rotavirus strain
prior to incubation on MA104 cells. After incubation, virus
neutralisation was determined using indirect immuno-
fluorescent staining. Results are expressed as the concentration
of sialylmimetic at which 50% infection of control infected
monolayers occurred (IC50).
12. Barchi, J. J. Curr. Pharm. Des. 2000, 6, 485.