Identification of an acetyl disulfide derivative in the synthesis of thiosialosides

Article abstract of DOI:10.1016/j.carres.2009.10.017

The first report of the formation of an acetyl disulfide sialoside during the synthesis of thioglycosides is described. This compound is a by-product in the synthesis of the 2-thioacetyl sialoside commonly used in thioglycoside preparation. Our investigations into the identification of this novel disulfide are described.

Full text of DOI:10.1016/j.carres.2009.10.017

Carbohydrate Research 345 (2010) 160–162
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Note
Identification of an acetyl disulfide derivative in the synthesis of thiosialosides
*
Goreti Ribeiro Morais, Ines Felix Oliveira, Andrew J. Humphrey, Robert A. Falconer
Institute of Cancer Therapeutics, School of Life Sciences, University of Bradford, Bradford BD7 1DP, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
The first report of the formation of an acetyl disulfide sialoside during the synthesis of thioglycosides is
described. This compound is a by-product in the synthesis of the 2-thioacetyl sialoside commonly used in
thioglycoside preparation. Our investigations into the identification of this novel disulfide are described.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 9 September 2009
Received in revised form 12 October 2009
Accepted 20 October 2009
Available online 22 October 2009
Keywords:
Thiosialoside
Sialoside
Sialic acid
Disulfide
Thioacetates are commonly employed as a means by which to
introduce sulfur into organic molecules, on route to thiols or disul-
fides. They are typically synthesised by the reaction of alkyl halides
with potassium thioacetate.1 Our interest in thiols arises from
work in the area of thioglycoside synthesis,2 and our experience
with disulfides.3 Glycosyl disulfides have received considerable
interest recently due to their utility as glycosyl donors.4 The obser-
vations described here result from our efforts to synthesise
thiosialosides.
N-Acetylneuraminic acid (sialic acid, Neu5Ac) is commonly
found on the surface of mammalian cells and is of particular
importance in cellular recognition processes, cell adhesion and dis-
ease states.5,6 There is therefore great interest in the synthesis of
sialosides. Furthermore, due to stronger chemical and enzymatic
stability than their O-glycosyl counterparts, sialosides bearing a
sulfur atom at the anomeric carbon have been extensively studied
as glycosyl mimics7 or as synthetic intermediates.8
group, who additionally suggested that this by-product was often
not detected.18
In our efforts to synthesise both alkyl and aryl thiosialosides via
this route, formation of a third product (compound X) was ob-
served, in addition to glycal 3 (Scheme 1). The synthesis of thiosi-
alosides is routinely accomplished via the established method of
simultaneous S-acetate de-protection and alkylation,19 as exempli-
fied in the case of ethyl thiosialoside 420 (Scheme 2). In our expe-
rience, removal of 3 from reaction mixtures is easier following this
alkylation step. To our surprise, the third product X also appeared
to be affected during this reaction, yielding Y (Scheme 2). The reac-
tion outlined in Scheme 1 was therefore re-visited in an attempt to
isolate and identify unknown by-product X.
Preparative HPLC allowed separation of the components of this
reaction mixture. NMR and low resolution mass spectrometry con-
firmed the presence of compounds 2 and 3. Whilst NMR analysis
proved inconclusive at this stage, compound X exhibited a mass
Typically, the key step in the synthesis of thiosialosides is the
aforementioned conversion of 2-chlorosialoside 1 into 2-thioace-
tylsialoside 2 by reaction with potassium thioacetate (Scheme
1).9 This reaction is known to commonly result in the formation
of an unsaturated by-product (per-O-acetyl Neu5Ac2en1Me, 3), a
protected form of Neu5Ac2en, which is of considerable interest
in its own right, both chemically10 and biologically.11–17 Glycal 3
is notoriously difficult to separate from desired compound 2 and
other thiosialosides, often necessitating complex purification pro-
cedures and/or HPLC. This was first noted by the von Itzstein
AcO
Cl
AcO
OAc
O
COOMe
OAc
AcHN
OAc
1
KSAc
AcO
AcO
COOMe
AcO
+
AcO
OAc
O
COOMe
+
O
SAc
X (10-60%)
AcHN
AcHN
OAc
OAc
3 (10%)
2 (30-80%)
* Corresponding author. Tel.: +44 1274 235842; fax: +44 1274 233234.
E-mail address: r.a.falconer1@bradford.ac.uk (R.A. Falconer).
Scheme 1. Synthesis of 2-thioacetyl sialoside 2.
0008-6215/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2009.10.017

G. R. Morais et al. / Carbohydrate Research 345 (2010) 160–162
161
Table 2
AcO
AcO
COOMe
SAc
COOMe
AcO
AcO
ICH2CH3
Et2NH
OAc
OAc
HPLC analysis of thiosialosides
O
O
Retention timea (min)
S
Et
Compound
AcHN
AcHN
OAc
OAc
2 (SAc)
6.3
7.3
6.9
2 (40%)
[+ 3 (10%) + X (50%)]
4 (65%)
[+ 3 (10%) + Y(25%)]
4 (SEt)
6 (SSAc)
8 (SSEt)
8.5
Scheme 2. Synthesis of ethyl thiosialoside 4.
Y + Z
7.2, 8.5
7.2, 8.5
Co-injection: (Y + Z), 4, 8
a
HPLC analyses were performed using an Agilent Technologies 1200 system,
with diode array detection, using a C18 reverse phase column (Agilent Eclipse XDB:
4.6 ꢁ 100 mm).
32 amu higher than 2. It was therefore hypothesised that com-
pound X could either be a product of oxidation of the thioacetate
to the
a
-oxo-sulfone 521,22 or acetyl disulfide 6 (Fig. 1)—both of
which would result in a mass increase of 32. Analysis of compound
X by high resolution mass spectrometry gave the ammonium ad-
duct [M+NH4]+ at 599.1575. The respective calculated masses for
We observed that the ratio of compound 2 to by-product X var-
ied quite markedly with differing batches of KSAc, including newly
purchased material from various suppliers. Given the potential dif-
the
a-oxo-sulfone 5 and acetyl disulfide 6 were 599.1753 and
599.1575, respectively. Whilst not definitive, this strongly sug-
gested the presence of the acetyl disulfide derivative 6. This com-
pound has not been reported previously. The IR spectrum also
lacked a strong absorption signal at 1700 cmꢀ1, characteristic of
ficulty in synthesising pure acetyl disulfide 6 and a-oxo-sulfone 5
(the latter being particularly unstable21,22), we decided to investi-
gate further by subjecting purified compound X to the alkylation
conditions outlined in Scheme 2. For comparison, ethyl disulfide
8 (the expected alkylation product if X were acetyl disulfide 6)
was purposely synthesised from 2-thiosialoside 723 using method-
ology previously developed in our laboratory3 (Scheme 3). Com-
pound 8 was previously synthesised by Hummel and Hindsgaul
via a different route,24 but the characterisation was not reported.
NMR (Table 1) and HPLC analysis (Table 2) showed 8 to be identical
to compound Y. It was therefore concluded that compound X does
indeed correspond to novel disulfide 6.
a
-oxo-sulfones.21
AcO
AcO
COOMe
SO2Ac
COOMe
O
SSAc
AcO
AcO
OAc
OAc
O
AcHN
AcHN
OAc
OAc
5
6
Interestingly, alkylating pure acetyl disulfide 6 (X) resulted in a
second compound in addition to ethyl disulfide 8 (Y). On isolation
and close inspection of the NMR spectra (Table 1), this was identi-
fied as ethyl thioglycoside 4. This was further confirmed by HPLC
(Table 2). Given that thioacetate 2 was absent from the starting
material, we deduced that ethyl sulfide 4 produced in this case re-
sulted from disulfide 6 (Scheme 4).
Figure 1. Structures of
a
-oxo-sulfone 5 and acetyl disulfide 6.
AcO
AcO
COOMe
SH
COOMe
O
SSEt
AcO
AcO
OAc
OAc
DEAD
EtSH
O
AcHN
AcHN
OAc
OAc
7
8
SSAc 6 was found to be unstable to diethylamine treatment
(routinely used in such reactions), leading to disulfide cleavage.
Subsequent alkylation of the resulting 2-thiosialoside (7) gave 4.
Scheme 3. Synthesis of ethyl disulfide 8.
Table 1
1H NMR data
Chemical shift (ppm), multiplicity
Coupling constantsa (Hz)
2 (SAc)
4 (SEt)
6 (SSAc)
8 (SSEt)
Y + Zb
H-9b
H-9a
4.37, dd
2.4, 12.4
4.02, dd
5.8, 12.4
5.20–5.22, m
5.34, dd
2.2, 6.7
4.29, dd
2.5, 12.5
4.10, dd
5.1, 12.5
5.35–5.38, m
5.30, dd
2.0, 8.4
4.34, d
12.1
4.12, dd
4.7, 12.1
5.29–5.33, m
5.29–5.33, m
4.36, dd
2.4, 12.4
4.11, dd
5.0, 12.4
5.26–5.30, m
5.26–5.30, m
H-8
H-7
H-6
4.65, dd
2.2, 10.8
4.05–5.01, m
4.87–4.91, m
2.60, dd
4.6, 12.9
2.09, dd
9.2, 12.9
5.16, d
3.83, dd
2.0, 10.8
4.00–4.06, m
4.82–4.87, m
2.70, dd
4.6, 12.7
1.97, dd
12.0, 12.7
5.24, d
3.86, d
9.4
3.96, dd
1.3, 10.6
4.00–4.05, m
4.85–4.90, m
2.68, dd
4.8, 12.7
2.25, dd
12.0, 12.7
5.24, d
H-5
H-4
H-3eq
4.00–4.05, m
4.83–4.88, m
2.81, dd
4.4, 12.2
2.07, dd
9.4, 12.2
5.21, d
2.68, dd; 2.70, dd
4.8, 12.7; 4.6, 12.7
2.25, dd; 1.97, dd
12.0, 12.7; 12.0, 12.7
H-3ax
NH
9.9
10.8
9.9
NAc
SAc
CH2
CH3
1.88, s
2.27, s
1.89, s
1.88, s
2.46, s
1.90, s
2.52, dq; 2.75, dq
1.18, dd
2.75–2.85, m
1.30, t
2.52, dq; 2.75–2.85, m
1.18, dd; 1.30, t
a
NMR spectra were recorded using a JEOL ECA-600 (600 MHz) spectrometer at room temperature in CDCl3. Chemical shifts are reported in ppm downfield relative to
Me4Si. The additional signals of O-acetates and the methyl ester are omitted.
b
Significant peaks noted only.

162
G. R. Morais et al. / Carbohydrate Research 345 (2010) 160–162
AcO
(dd, 1H, H-3ax, J3ax,4 12.0 Hz, J3ax,3eq 12.7 Hz), 2.68 (dd, 1H, H-3eq,
AcO
COOMe
R
COOMe
SSAc
AcO
ICH2CH3
AcO
OAc
OAc
J3eq,4 4.8 Hz, J3ax,3eq 12.7 Hz), 2.75–2.85 (m, 2H), 3.80 (s, 3H,
COOMe), 3.96 (dd, 1H, H-6, J6,7 1.3 Hz, J5,6 10.6 Hz), 4.00–4.05 (m,
1H, H-5), 4.11 (dd, 1H, H-9a, J8,9a 5.0 Hz, J9a,9b 12.4 Hz), 4.36 (dd,
1H, H-9b, J8,9b 2.4 Hz, J9a,9b 12.4 Hz), 4.85–4.90 (m, 1H, H-4), 5.24
(d, 1H, NH, J5,NH 9.9 Hz), 5.26–5.30 (m, 2H, H-7 and H-8); 13C
NMR (CDCl3) d 13.69, 20.81, 20.88, 20.92, 21.19, 23.30, 33.78,
37.10, 49.57, 53.15, 62.12, 67.41, 69.51, 69.72, 88.93, 74.77,
168.11, 170.10, 170.25, 170.78, 171.09, 171.12; ES+ MS
C22H33O2NS2 (567.63) m/z (%) 590 [M+Na]+ (100); HRMS (ES+)
found 585.1783, calcd for C22H37O12N2S2 585.1782 [M+NH4]+.
O
O
AcHN
AcHN
Et2NH
OAc
OAc
6
4 R = SEt (50%)
8 R = SSEt (50%)
Scheme 4. Alkylation of acetyl disulfide 6.
The mechanism by which acetyl disulfide 6 is formed in the
reaction outlined in Scheme 1 is not certain. It was hypothesised,
however, that oxidation of KSAc produces diacetyl disulfide, which
further reacts to yield AcSSꢀ and AcSAc.
Acknowledgements
It is possible that chlorosialoside 1 may undergo nucleophilic
attack by the acetyl disulfide anion, to yield SSAc 6 directly. Prep-
aration of AcSSAc25 and subsequent reaction with chlorosialoside 1
proved this not to be the case. However, on repeating this reaction
in the presence of KSAc, SSAc 6 was obtained almost exclusively.
Interestingly, treatment of purified thioacetate 2 with a batch of
KSAc that had promoted extensive disulfide formation also led to
the formation of SSAc 6, suggesting that the source of 6 is actually
thioacetate 2. This would require either reaction of AcSSAc with
thioacetate 2, via a radical-based mechanism, or by reaction with
AcSAc. Auto-oxidation of thioacetic acid by air and light has also
been observed26 and may be significant. Further studies are re-
quired to understand this reaction further.
All three compounds (i.e., 2, 3 and 6) co-elute on thin layer
chromatography plates and during flash chromatography. Prepara-
tive HPLC is required for effective separation. SSAc 6 has since been
synthesised and purified from thiosialoside 7 using a similar meth-
odology employed in the synthesis of ethyl disulfide 8, and further
confirmed the identity of compound X.
This work was supported by the EPSRC (G.R.M.), the Association
for International Cancer Research (A.J.H.) and Yorkshire Cancer
Research (R.A.F.). The authors thank Andrew Healey for running
low resolution mass spectra and the EPSRC National Mass
Spectrometry Service Centre, University of Wales, Swansea, for
high resolution accurate mass measurements. Dr. Kamyar Afarin-
kia, Dr. Klaus Pors and Dr. Helen Sheldrake are thanked for useful
discussions.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.carres.2009.10.017.
References
1. Procter, D. J. J. Chem. Soc., Perkin Trans. 1 2000, 835–871.
2. (a) Morais, G. R.; Humphrey, A. J.; Falconer, R. A. Carbohydr. Res. 2009, 344, 1039–
1045; (b) Morais, G. R.; Humphrey, A. J.; Falconer, R. A. Tetrahedron 2008, 64,
7426–7431; (c) Falconer, R. A.; Toth, I. Bioorg. Med. Chem. 2007, 15, 7012–7020;
(d) Falconer, R. A.; Jablonkai, I.; Toth, I. Tetrahedron Lett. 1999, 40, 8663–8666.
3. Morais, G. R.; Falconer, R. A. Tetrahedron Lett. 2007, 48, 7637–7641.
4. Davis, B. G.; Ward, S. J.; Rendle, P. M. Chem. Commun. 2001, 189–190.
5. Varki, A. Glycobiology 1992, 2, 25–40.
In summary, unexpected observations during the synthesis of
thiosialosides are described. The synthesis of thiosialosides using
KSAc leads to disulfide by-product formation, which is extremely
variable and dependent on the commercial source of KSAc.
6. Zhang, P.; Zuccolo, A. J.; Li, W.; Zheng, R. B.; Ling, C. C. Chem. Commun. 2009,
4233–4235.
7. Witczak, Z. J. Curr. Med. Chem. 1999, 6, 165–178.
1. Experimental
8. Boysen, M. M. K. Chem. Eur. J. 2007, 13, 8649–8659.
9. Warner, T. G.; Lee, L. A. Carbohydr. Res. 1988, 176, 211–218.
10. Morais, G. R.; Oliveira, R. S.; Falconer, R. A. Tetrahedron Lett. 2009, 50, 1642–1644.
11. Meindl, P.; Tuppy, H. Monatsh. Chem. 1969, 1295–1306.
12. Schreiner, E.; Zbiral, E.; Kleineidam, R. G.; Schauer, R. Liebigs Ann. Chem. 1991,
129–134.
13. Zbiral, E.; Brandstetter, H. H.; Christian, R.; Schauer, R. Liebigs Ann. Chem. 1987,
781–786.
14. von Itzstein, M.; Wu, W. Y.; Kok, G. B.; Pegg, M. S.; Dyason, J. C.; Jin, B.; Phan, T.
V.; Smythe, M. L.; White, H. F.; Oliver, S. W.; Colman, P. M.; Varghese, J. N.;
Ryan, D. M.; Woods, J. M.; Bethell, R. C.; Hotham, V. J.; Cameron, J. M.; Penn, C.
R. Nature 1993, 363, 418–423.
1.1. Characterisation of compounds 6 and 8
1.1.1. Methyl 2-(acetylsulfanyl)-5-acetamido-4,7,8,9-tetra-O-
acetyl-2,3,5-trideoxy-2-thio-
pyranosonate (6)
a-D-glycero-D-galacto-2-nonulo
½
a 2D0 +155.0 1.0 (c 0.85, CH2Cl2); 1H NMR (CDCl3) d 1.88 (s, 3H,
ꢂ
NAc), 2.02, 2.04, 2.12 (3s, 12H, 4OAc), 2.07 (dd, 1H, H-3ax, J3ax,4
9.4 Hz, J3ax,3eq 12.2 Hz), 2.46 (s, 3H, SAc), 2.81 (dd, 1H, H-3eq,
J3eq,4 4.4 Hz, J3ax,3eq 12.2 Hz), 3.77 (s, 3H, COOMe), 3.86 (d, 1H,
H-6, J5,6 9.4 Hz), 4.00–4.05 (m, 1H, H-5), 4.12 (dd, 1H, H-9a, J8,9a
4.7 Hz, J9a,9b 12.1 Hz), 4.34 (d, 1H, H-9b, J9a,9b 12.1 Hz), 4.83–4.88
(m, 1H, H-4), 5.21 (d, 1H, NH, J5,NH 10.8 Hz), 5.31 (br s, 2H, H-7,
and H-8); 13C NMR (CDCl3) d 20.87 (2C), 20.90, 21.24, 23.20,
28.81, 37.07, 49.45, 53.50, 62.01, 67.23, 69.04, 69.57, 74.44,
84.72, 166.85, 170.00, 170.18, 170.80, 170.89, 192.51; ES+ MS
C22H31NO13S2 (581.14) m/z (%) 582.2 [M+H]+ (30), 604.2 [M+Na]+
(100); HRMS (ES+) found 599.1575, calcd for C22H35O13N2S2
599.1575 [M+NH4]+.
15. Miller, C. A.; Wang, P.; Flashner, M. Biochem. Biophys. Res. Commun. 1978, 83,
1479–1487.
16. Chong, A. K. J.; Pegg, M. S.; Taylor, N. R.; von Itzstein, M. Eur. J. Biochem. 1992,
207, 335–343.
17. Smith, P. W.; Starkey, I. D.; Howes, P. D.; Sollis, S. L.; Keeling, S. P.; Cherry, P. C.;
von Itzstein, M.; Wu, W. Y.; Jin, B. Eur. J. Med. Chem. 1996, 31, 143–150.
18. Groves, D. R.; Bradley, S. J.; Rose, F. J.; Kiefel, M. J.; von Itzstein, M.
Glycoconjugate J. 1999, 16, 13–17.
19. Bennett, S.; von Itzstein, M.; Kiefel, M. J. Carbohydr. Res. 1994, 259, 293–299.
20. Lubineau, A.; Legallic, J. J. Carbohydr. Chem. 1991, 10, 263–268.
21. Schank, K.; Werner, F. Liebigs Ann. Chem. 1979, 1977–1991.
22. Schank, K.; Frisch, A.; Zwanenburg, B. J. Org. Chem. 1983, 48, 4580–4582.
23. Cohen, S. B.; Halcomb, R. L. J. Org. Chem. 2000, 65, 6145–6152.
24. Hummel, G.; Hindsgaul, O. Angew. Chem., Int. Ed. 1999, 38, 1782–1784.
25. Bonner, W. A. J. Am. Chem. Soc. 1950, 72, 4270–4271.
26. Gautron, J. Histochemistry 1982, 76, 469–478.
1.1.2. Methyl 2-(ethylsulfanyl)-5-acetamido-4,7,8,9-tetra-O-ace
tyl-2,3,5-trideoxy-2-thio-
nosonate (8)
a-D-glycero-D-galacto-2-nonulopyra-
½
a 2D0 +33.0 1.0 (c 0.63, CH2Cl2); 1H NMR (CDCl3) d 1.30 (t, 3H, J
ꢂ
7.4 Hz), 1.90 (s, 3H, NAc), 2.02, 2.03, 2.10, 2.14 (4s, 12H, 4OAc), 2.25