The direct formation of glycosyl thiols from reducing sugars allows one-pot protein glycoconjugation

Article abstract of DOI:10.1002/anie.200600685

(Figure Presented) Sweet and easy: A one-pot method consisting of direct thionation (1) followed by thiol-mediated chemoselective ligation (2) can be used for site-selective protein glycosylation. This procedure, which uses the Lawesson reagent, has been shown to be fully compatible with unprotected sugars, the products of which can be directly used in a selenenylsulfide protein glycosylation strategy.

Full text of DOI:10.1002/anie.200600685

Angewandte
Chemie
Glycoproteins
DOI: 10.1002/anie.200600685
The Direct Formation of Glycosyl Thiols from
Reducing Sugars Allows One-Pot Protein
Glycoconjugation**
Gonçalo J. L. Bernardes, David P. Gamblin, and
Benjamin G. Davis*
Glycoconjugates have become essential tools for the inves-
tigation of many biological processes,[1] and it is now well
established that protein and lipid-bound carbohydrate units
play essential roles in cell signaling regulation,[2] cellular
differentiation,[3] and immune response.[4] In recent years,
glycosyl thiols have become useful building blocks for the
synthesis of certain glycoconjugates that may be considered to
be analogues of glycopeptides and glycoproteins.[5–8] Their use
has allowed specific glycosylation of peptides to form S-linked
glycopeptides through alkylation[6] or conjugate-addition
strategies.[7] Moreover, we have demonstrated that a combi-
[*] G. J. L. Bernardes, Dr. D. P. Gamblin, Dr. B. G. Davis
Chemistry Research Laboratory
Department of Chemistry
University of Oxford
Mansfield Road, Oxford, OX13TA (UK)
Fax: (+44)1865-275-652
E-mail: Ben.Davis@chem.ox.ac.uk
[**] We gratefully acknowledge the Fundaç¼o para a CiÞncia e a
Tecnologia, Portugal for the award of a doctoral fellowship (G.J.L.B.,
grant BD/16997/2004), and the EPRSC (D.P.G.). We also thank
Katie J. Doores and Dr. Graeme Horne for helpful discussions, and
Alister French and Norberto Reis for graphics.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2006, 45, 4007 –4011
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4007

Communications
nation of site-directed mutagenesis followed by thiol-medi-
ated chemoselective ligation can be used for site-selective
protein glycosylation. In this two-step strategy (Glyco–SeS),[8]
a cysteine residue is incorporated into the desired position
within the protein backbone; the free thiol in the side chain of
this cysteine is subsequently converted into a selenenylsulfide
following exposure to phenylselenenyl bromide. This acti-
vated protein, upon the addition of a glycosyl thiol, is
converted directly into the corresponding homogenous gly-
coprotein.
Given this value of such glycosyl thiols and their other
potential uses, for example, as precursors[9] to widely used
thioglycosides[10] and glycosyldisulfide[11] donors, it is all the
more surprising that methods for producing these compounds
are somewhat protracted and that until now no ready and
general direct method for their preparation exists. Further-
more, thioglycosides and their resulting glycoconjugates often
demonstrate increased chemo- and enzymatic stability and
are tolerated by most biological systems.[12] Indeed, gold salts
of thioaldoses[12b] are used in the treatment of rheumatoid
arthritis and have recently been identified as potential
blockers of transformed growth of lung-cancer cells.[12c]
The most frequently employed method for the synthesis of
glycosyl thiols 1 involves the treatment of glycosyl halides 2
with a sulphur nucleophile (either thiourea[13] or potassium
thioacetate) in acetone,[14] followed by mild hydrolysis
(Scheme 1).[15,16] Other methods include the use of anomeric
Scheme 2. One-pot protein glycosylation with reducing sugars isolated
from natural sources.
Scheme 3. Two possible mechanisms for the direct formation of
glycosyl thiols from reducing sugars through path A) open chain or
path B) an oxacarbenium ion intermediate.
The mechanism of Lawessonꢀs reagent[21] (LR) suggests
that it might service either pathway, acting potentially as both
an oxaphilic electrophile and a sulphur source. This reagent
has been extensively used for the efficient conversion of a
wide variety of carbonyl functions into their corresponding
thiocarbonyl functionalities.[22] Moreover, an earlier report
had highlighted a rare single use of LR in the conversion of a
benzylic alcohol into the corresponding thiol,[23] thereby
suggesting potential utility in SN1 or SN1-like pathways.
Therefore, owing to the enhanced reactivity of the anomeric
hydroxy group in SN1-like processes, it was thought that this
Scheme 1. Current approaches for the synthesis of glycosyl thiols.
DCM=dichloromethane.
acetates under Lewis acidic conditions[17] and the Birch
reduction of anomeric thiobenzyl.[18] Defaye et al. have
described the preparation of glucosyl thiols from d-glucose
typically in yields of < 30% by bubbling hydrogen sulfide into
a solution of the deprotected sugar in hydrogen fluoride.[19]
However, this procedure requires special precautions and
generates a complex mixture of products.
We considered that a direct method of thiol formation in
combination with Glyco–SeS[8] would allow a one-pot protein
glycosylation method that could utilize sugars directly iso-
lated from natural sources (Scheme 2).[20] With the goal of
finding a more efficient strategy for the direct synthesis of
glycosyl thiols, we speculated that a reagent that operated
through a concerted Lewis acid activation and displacement
might allow chemo- and regioselective thionation of C1 in the
presence of other, perhaps unprotected, functionalities.
However, to date, no such reagent or method exists. Such a
C1-selective thionation might be considered mechanistically
in two ways as shown in Scheme 3.
Table 1: Optimization of direct thionation reaction conditions.
Entry
EquivLR
T [8C]
t [h]
Solvent[a]
Yield [%]
1A
1B
1C
1D
1E
1F
1G
1H
1I
2
RT
80
80
80
reflux
RT
80
80
reflux
RT
48
5
2.5
2
2
48
5
2.5
2
48
5
dioxane
dioxane
dioxane
dioxane
dioxane
toluene
toluene
toluene
toluene
no reaction
56
83
58
0.6
1.2
2
1.2
2
0.6
1.2
1.2
1.2
0.6
1.2
84
no reaction
54
70
75
14
49
62
1J
1K
1L
acetonitrile
acetonitrile
acetonitrile
80
80
2.5
[a] Reactions conducted in anhydrous solvents (5 mL for a 200 mg scale
reaction) and under an atmosphere of argon.Bn=benzyl.
4008 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 4007 –4011

Angewandte
Chemie
procedure might proceed in an analogous fashion with sugar
substrates. This approach appeared attractive for several
reasons: 1) allowing direct formation of the anomeric thio-
sugars in one step from the corresponding anomeric alcohols;
2) application to differently protected and to unprotected
sugars and 3) the possibility for direct site-selective glycosy-
lation of proteins with sugars isolated from natural sources in
just two steps and in one pot (Scheme 2).
tions were optimized with respect to solvent system, temper-
ature, reaction time, and equivalents of LR (Table 1). It was
observed that prolonged reaction times and large excesses of
LR were detrimental, resulting in the generation of numerous
side products. Optimum conditions used anhydrous dioxane
at 808C or above over a period of 2–3 h with 1.2 equivalents of
LR affording 6 in a yield of 83–84% (Table 1, entries C and
E). Although yields in toluene were nearly comparable
(Table 1, entry I), the use of more-polar dioxane opened up
the exciting possibility of using deprotected sugars, which are
not soluble in toluene or acetonitrile.
Initial investigations (Table 1) used reducing sugar 5 as a
test substrate. Encouragingly,
a first attempt (Table 1,
entry B) performed in anhydrous toluene at 808C by using
0.6 equivalents of LR for 5 h directly afforded thiogalacto-
pyranose 6 in 54% yield as a single a-anomer. Fortunately, 1-
thiohexoses do not mutarotate under basic or neutral
conditions.[24] Following this initial success, reaction condi-
Next we explored the scope and limitations of this LR-
mediated protocol (Table 2), demonstrating that this proce-
dure is quite general and is applicable for the preparation of a
variety of differently protected 1-thiosugars, including for the
Table 2: Direct formation of glycosyl thiols using Lawesson’s reagent.[a]
[c]
Entry
1
Substrate[b]
T [8C]
t
EquivLR
Glycosyl thiol
Yield
see Table 1
82%
a/b 2:1
2
3
4
5
7
9
80
80
80
80
80
2.5 h
2 h
1.2
1.2
1.2
1.2
1.2
8
80%
a anomer only
10
12
14
16
78%
a anomer only
11
13
1 h
82%
a/b 1:5
3 h
85%
a anomer only
6
7
15
17
3 h
3 h
80
110
110
1.2
1.5
2
no product formed
–
32% (63%)
b anomer only
41% (68%)
b anomer only
43% (65%)
b anomer only
47% (69%)
b anomer only
24 h
72 h
18
20
8
19
80
2.5 h
24 h
1.2
1.5
110
75%
a/b 2:5
9
10
11
12
21
23
25
27
80
80
80
80
2.5 h
1
22
24
26
28
77%
a/b 2:1
1 h
1
85%
a anomer only
45 min
4.5 h
1
82%
b anomer only
1.2
74%
3:1
13
29
80
4 h
1.2
30
[a] Reactions conducted in anhydrous dioxane under an atmosphere of argon; see the Supporting Information for details. [b] All substrates were
synthesized according to literature procedures; see the Supporting Information for details [c] Parentheses indicate yield based on recovered starting
material. TBDMS=dimethyl-1,1-dimethylethyl-silyl, Phth=phthalimidyl.
Angew. Chem. Int. Ed. 2006, 45, 4007 –4011
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4009

Communications
Table 3: Synthesis of glycosyl thiols from unprotected sugars.
direct conversion of sugars fully protected with benzyl and
methyl ether (78–85%). Partially O-benzyl-protected sugars
with an O-acetyl function at C-2 were also converted to the
corresponding glycosyl thiols in good yields (82–85%). Lower
yields were observed on deactivated, so called disarmed,[25]
systems such as acetylated sugars (41–43%); however, the
starting material was largely recovered and could be recycled
(yields based on recovered starting material 63–69%). Nitro-
gen functionality at C-2 in the form of phthalimide (Table 2,
entry 12), and base and acid-labile protecting groups, such as
silyl groups and acetonides (Table 2, entry 13), remained
stable under the reaction conditions, thereby expanding the
functional-group tolerance significantly.
Finally, application of this methodology to unprotected
sugars was investigated (Table 3). The optimized conditions
were applied to unprotected sugars, and after 48 h the
formation of the desired thiol product and corresponding
disulfide species was observed. Chromatographic separation
of these from the LR decomposition products proved
cumbersome and was exacerbated by further disulfide
formation.[26] Two strategies were adopted to aid in isolation.
First, an acetylation–deacetylation protocol afforded the
Entry Substrate
product
Yield A[a]
Yield B[b]
61%
a/b 1:9
71%
a/b 1:6
1
2
3
Glc
31a
31b
31c
48%
63%
Man
Gal
a anomer a anomer
58%
a/b 1:8
70%
a/b 1:4
[a] Protocol A: 1) LR (1.5 equiv), anhydrous dioxane, 1108C. 2) Ac2O,
pyridine. 3) NaOMe, MeOH. [b] Protocol B: 1) LR (1.5 equiv), anhydrous
dioxane, 1108C. 2) Bu3P, MeOH/CHCl3; see the Supporting Information
for full details.
corresponding glycosyl thiols 31a–c in fair yields (Table 3).
Alternatively, treatment with tributylphosphine (solution in
Table 4: Two-step strategy for direct protein conjugation from deprotected reducing sugars through LR and Glyco–SeS.[8]
Sugar
Thiol product[a]
Protein product[b]
Conversion [%][c] ESI-MS Found
(calcd)
Glc
32a >95
27066 (27064)
27064 (27064)
27064 (27064)
27234 (27225)
Man
32b >95
32c >95
32d >95
Gal
Lactose (Galb1,4Glc)
Maltopentaose
(Glca1,4)4Glc
32e >95
27722 (27711)
Galabiose (Gaa1,4Gal)
(Xyla1,3Glc)
32 f >95
32g >95
27224 (27225)
27194 (27194)
[a] Typically, LR (1.2 equiv) was added to a solution of the deprotected sugar in anhydrous dioxane and left to stir at 1108C for 48 h; see the Supporting
Information for more details. [b] Typically, crude thiol (20–50 equiv), in water, was added to preactivated SBLS156C-SePh in CHES (70 mm), MES
(5 mm), CaCl2 (2 mm); pH 9.5. After 30 min at RT, the reaction was analyzed by LC–MS; [c] Conversion determined by ESI-MS.
4010 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 4007 –4011

Angewandte
Chemie
[6] a) S. Knapp, D. S. Myers, J. Org. Chem. 2002, 67, 2995 – 2999;
CHCl3/MeOH) reduced the resulting disulfides allowing the
direct isolation of pure unprotected glycosyl thiols 31a–c in
63–71% yields (Table 3). These results importantly demon-
strated application to unprotected sugars providing a method
to prepare glycosyl thiols suitable for protein glycosylation.
Following this key result, it was envisaged that our newly
developed thionation procedure would offer a direct route
from free sugars to glycoproteins by using our protein
glycosylation strategy (Scheme 2). A variety of representa-
tive, free sugars (Table 4), including those isolated from N-
(galabiose Gala1,4Gal)[27] and O-linked (Xyla1,3Glc) glyco-
proteins,[28] were treated with LR after the optimized
conditions.[29] The subsequent addition of these sugars to a
selenenylsulfide-activated single-cysteine mutant protein[8]
(subtilisin Bacillus lentus, SBLS156C), gave complete con-
version to the corresponding glycoproteins (Table 4).
In summary, we have established that Lawessonꢀs reagent
may be used in a direct and general manner for the
preparation of glycosyl thiols from the corresponding anome-
ric lactols/reducing sugars. Notably, this procedure has also
been shown to be fully compatible with unprotected sugars,
the products of which can be directly used in our selenenyl-
sulfide-mediated protein glycosylation strategy.[8] The result is
a one-pot method for direct protein glycosylation. Further
investigations into the use of LR are currently being explored
by this laboratory.
b) X. Zhu, K. Pachamuthu, R. R. Schmidt, J. Org. Chem. 2003,
68, 5641 – 5651; c) X. Zhu, R. R. Schmidt, Chem. Eur. J. 2004, 10,
875 – 887; d) D. A. Thayer, N. Y. Henry, M. C. Galan, C. H.
Wong, Angew. Chem. 2005, 117, 4672 – 4675; Angew. Chem. Int.
Ed. 2005, 44, 4596 – 4599.
[7] a) Y. Zhu, W. A. van der Donk, Org. Lett. 2001, 3, 1189 – 1192;
b) Y. Zhu, M. D. Gieselman, H. Zhou, O. Averin, W. A. van der
Wonk, Org. Biomol. Chem. 2003, 1, 3304– 3315.
[8] D. P. Gamblin, P. Garnier, S. van Kasteren, N. J. Oldham, A. J.
Fairbanks, B. G. Davis, Angew. Chem. 2004, 116, 846 – 851;
Angew. Chem. Int. Ed. 2004, 43, 828 – 833.
[9] M. M. Ponpipom, W. K. Hagmann, L. A. OꢀGrady, J. J. Jackson,
D. D. Wood, H. J. Zweerink, J. Med. Chem. 1990, 33, 861 – 867.
[10] P. J. Garegg, Adv. Carbohydr. Chem. Biochem. 1997, 52, 179 –
205.
[11] a) B. G. Davis, S. J. Ward, P. M. Rendle, Chem. Commun. 2001,
189 – 190; b) E. J. Grayson, S. J. Ward, A. L. Hall, P. M. Rendle,
D. P. Gamblin, A. S. Batsanov, B. G. Davis, J. Org. Chem. 2005,
70, 9740 – 9754.
[12] a) B. Capon, Chem. Rev. 1969, 69, 407 – 498; b) S. T. Crooke,
R. M. Snyder, T. R. Butt, D. J. Ecker, H. S. Allaudeen, B. Monia,
C. K. Mirabelli, Biochem. Pharmacol. 1986, 35, 3423 – 3431;
c) M. Stallings-Mann, L. Jamieson, R. P. Regala, C. Weems, N. R.
Murray, A. P. Fields, Cancer Res. 2006, 66, 1767 – 1774.
[13] a) Y. C. Lee, C. P. Stowell, M. J. Krantz, Biochemistry 1976, 15,
3956 – 3963; b) Y. C. Lee, C. P. Stowell, Methods Enzymol. 1982,
83, 278 – 288.
[14] J. M. MacDougall, X. D. Zhang, W. E. Polgar, T. V. Khroyan, L.
Toll, J. R. Cashman, J. Med. Chem. 2004, 47, 5809 – 5815.
[15] a) B. D. Johnston, B. M. Pinto, J. Org. Chem. 2000, 65, 4607 –
4617; b) O. B. Wallace, D. M. Springer, Tetrahedron Lett. 1998,
39, 2693 – 2694.
Received: February 21, 2006
Published online: May 4, 2006
[16] W. Koenigs, E. Knorr, Ber. Dtsch. Chem. Ges. 1901, 34, 957 – 981.
[17] F. M. Ibatulin, K. A. Shabalin, J. V. Jꢁnis, A. G. Shavva, Tetrahe-
dron Lett. 1998, 39, 7961 – 7964.
[18] S. A. Holick, L. Anderson, Carbohydr. Res. 1974, 34, 208 – 213.
[19] In one instance a yield of 67% was reported for this method.
However the use of HFand H2S requires special precautions and
is not mild or convenient; see: J. Defaye, A. Gadelle, C.
Pedersen, Carbohydr. Res. 1991, 217, 51 – 58.
Keywords: carbohydrates · glycoproteins · glycosyl thiols ·
lawesson reagent · protein modifications
.
[1] a) A. Varki, Glycobiology 1993, 3, 97 – 130; b) R. A. Dwek,
Chem. Rev. 1996, 96, 683 – 720; c) P. M. Rudd, T. Elliot, P.
Cresswell, I. A. Wilson, R. A. Dwek, Science 2001, 291, 2370 –
2376; d) B. G. Davis, Chem. Rev. 2002, 102, 579 – 601; e) J. B.
Lowe, J. D. Marth, Annu. Rev. Biochem. 2003, 72, 643 – 691;
f) K. J. Doores, D. P. Gamblin, B. G. Davis, Chem. Eur. J. 2006,
12, 656 – 665.
[2] a) A. Helenius, M. Aebi, Science 2001, 291, 2364– 2369; b) R.
Kannagi, Curr. Opin. Struct. Biol. 2002, 12, 599 – 608.
[3] K. Nishiwaki, Y. Kubota, Y. Chigira, S. K. Roy, M. Suzuki, M.
Schvarzstein, Y. Jigami, N. Hisamoto, K. Matsumoto, Nat. Cell
Biol. 2004, 6, 31 – 37.
[20] J. B. L. Damm, J. P. Kamerling, G. W. K. van Dedem, J. F. G.
Vliegenthart, Glycoconjugate J. 1987, 4, 129 – 144.
[21] a) B. S. Pedersen, S. Scheibye, W. H. Nilson, S. -O. Lawesson,
Bull. Soc. Chim. Belg. 1978, 87, 223 – 228; b) T. B. Rauchfuss,
G. A. Zank, Tetrahedron Lett. 1986, 27, 3445 – 3448.
[22] a) M. P. Cava, M. I. Levinson, Tetrahedron 1985, 41, 5061 – 5087;
b) M. Jesberger, T. P. Davis, L. Barner, Synthesis 2003, 1929 –
1958.
[23] a) T. Nishio, J. Chem. Soc. Chem. Commun. 1989, 205 – 206; b) T.
Nishio, J. Chem. Soc. Perkin Trans. 1 1993, 1113 – 1117.
[24] W. Schneider, H. Leonhardt, Ber. Dtsch. Chem. Ges. 1929, 62,
1384– 1389.
[25] D. R. Mootoo, P. Konradsson, U. Udodong, B. Fraser-Reid, J.
Am. Chem. Soc. 1988, 110, 5583 – 5584.
[26] Reaction of crude product with DMF and the use of activated
carbon in an attempt to remove the excess of Lawessonꢀs reagent
failed.
[4] J. B. Lowe, Cell 2001, 104, 809 – 812.
[5] a) B. G. Davis, R. C. Lloyd, J. B. Jones, J. Org. Chem. 1998, 63,
9614– 9615; b) W. M. Macindoe, A. H. van Oijen, G.-J. Boons,
Chem. Commun. 1998, 847 – 848; c) B. G. Davis, M. A. T.
Maughan, M. P. Green, A. Ullman, J. B. Jones, Tetrahedron:
Asymmetry 2000, 11, 245 – 262; d) B. G. Davis, Chem. Commun.
2001, 351 – 352; e) D. P. Gamblin, P. Garnier, S. J. Ward, N. J.
Oldham, A. J. Fairbanks, B. G. Davis, Org. Biomol. Chem. 2003,
1, 3642 – 3644; f) G. M. Watt, J. Lund, M. Levens, V. S. K. Kolli,
R. Jefferis, G.-J. Boons, Chem. Biol. 2003, 10, 807 – 814; g) P. M.
Rendle, A. Seger, J. Rodrigues, N. J. Oldham, R. R. Bott, J. B.
Jones, M. M. Cowan, B. G. Davis, J. Am. Chem. Soc. 2004, 126,
4750 – 4751; h) T. Hotchkiss, H. B. Kramer, K. J. Doores, D. P.
Gamblin, N. J. Oldham, B. G. Davis, Chem. Commun. 2005,
4264 – 4266.
[27] N. Susuki, K. H. Khoo, C. M. Chen, H. C. Chen, Y. C. Lee, J.
Biol. Chem. 2003, 278, 46293 – 46246–46306.
[28] a) H. Nishimura, S. I. Kawabata, W. Kisiel, S. Hase, T. Itenaka, T.
Takao, Y. Shimonishi, S. Iwanaga, J. Biol. Chem. 1989, 264,
20320 – 20325; b) K. Fukase, S. Hase, T. Itenaka, S. Kusumoto,
Bull. Chem. Soc. Jpn. 1992, 65, 436 – 445.
[29] The crude thiol aldoses could be extracted after the reaction
following partition between DCM and water.
Angew. Chem. Int. Ed. 2006, 45, 4007 –4011
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4011