Neoglycoprotein cancer vaccines: Synthesis of an azido derivative of GM3 and its efficient coupling to proteins through a new linker

Article abstract of DOI:10.1016/S0040-4039(02)00071-0

A GM3 derivative having an azido group on its reducing end was synthesized in nine separate steps, and it was then conjugated to proteins by a new linker. Therefore, the azido group was reduced to a free amino group, which was followed by the introduction

Full text of DOI:10.1016/S0040-4039(02)00071-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 1599–1602
Neoglycoprotein cancer vaccines: synthesis of an azido derivative
of GM3 and its efficient coupling to proteins through
a new linker
Jie Xue, Yanbin Pan and Zhongwu Guo*
Department of Chemistry, Case Western Reserve University, 10900 Euclid Avenue, Cleveland, OH 44106, USA
Received 31 October 2001; accepted 2 January 2002
Abstract—A GM3 derivative having an azido group on its reducing end was synthesized in nine separate steps, and it was then
conjugated to proteins by a new linker. Therefore, the azido group was reduced to a free amino group, which was followed by
the introduction of a 4-pentenoyl group. The carbonꢀcarbon double bond of the 4-pentenoyl group was ozonolyzed to form a
carbonyl group, and GM3 was thus linked to proteins via the carbonyl group by reductive amination. © 2002 Elsevier Science
Ltd. All rights reserved.
It was revealed more than three decades ago that
oncogenic transformation was associated with dramatic
change in cell surface carbohydrates.1 Since then,
numerous carbohydrate antigens that are either specific
or markedly overexpressed on cancer cells have been
characterized, and the structural motifs are therefore
immunotolerance problems. Nevertheless, the concept
of TACA-based therapeutic cancer vaccines has been
well established, and several synthetic glycoconjugate
vaccines are now in clinical trials.6,9,16–20
The glycoconjugates that are most commonly employed
for vaccine designs are neoglycoproteins with the car-
bohydrate antigens covalently linked to a carrier
protein,21 e.g. tetanus toxoid21 or keyhole limpet hemo-
cyanin (KLH).18,19,22,23 In regard to the preparation of
neoglycoproteins, one major concern is the coupling of
TACAs to carrier proteins. This problem is more obvi-
ous and significant if we combine it with the complexity
of carbohydrate syntheses,24 since a good coupling
method is not necessarily compatible to the strategies,
including protection and glycosylation, used in carbo-
hydrate syntheses. For instance, reductive amination is
a simple and reliable conjugation method that has been
widely used for carbohydrate–protein couplings,21 and
thus, many linkers containing carbonyl groups or func-
tionalities that can be converted to carbonyl groups are
designed for this purpose.9,16,21,25,26 However, if the
linkers are attached to the oligosaccharide reducing
ends at an early stage of the syntheses, they may make
it difficult to design a compatible protection–deprotec-
tion strategy. Introduction of linkers to finished
oligosaccharides at the final stage may circumvent this
problem, but the established linkers and attachment
methods usually require multiple transformations of the
precious oligosaccharides, such as the removal of origi-
nal protecting groups and then introduction of a new
protection that is compatible with the linker, attach-
called
tumor-associated
carbohydrate
antigens
(TACAs).2–8
TACAs are the molecular basis of a new and very
attractive antitumor development, i.e. use of cancer
vaccines to treat cancer patients.7,9–15 Many TACAs
have been exploited in this connection, and it has been
recognized that, to induce specific antitumor immune
responses in patients, TACAs have to be conjugated to
an immunostimulatory agent for overcoming
Scheme 1. Reagents and conditions: (a) selective reduction of
the azido group; (b) (CH2ꢁCH(CH2)nCO)2O; (c) oxidative
cleavage of CꢁC; (d) reductive amination using NaBH3CN.
* Corresponding author. Tel.: +1 216 368-3736; fax: +1 216 368-3006;
e-mail: zxg5@po.cwru.edu
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)00071-0

1600
J. Xue et al. / Tetrahedron Letters 43 (2002) 1599–1602
when lactose was refluxed in acetic anhydride in the
presence of sodium acetate, the expected a-anomer 7
ment of the linker via glycosylation and final deprotec-
tion. Moreover, some of the potential linkers, e.g.
unsaturated acyls, cannot be easily applied even under
this circumstance.
3
(H-1: l 5.68, 4.47; J1,2 8.2, 7.7 Hz) was obtained as the
only product that was purified by recrystallization from
toluene–hexane. Treating 7 with azidotrimethylsilane
(N3TMS) and boron trifluoride diethyl etherate
(BF3·OEt2) in methylene dichloride afforded the glyco-
As a part of our campaign to develop new cancer
vaccines, we are interested in finding new and efficient
methods for TACA synthesis and oligosaccharide-
protein conjugation. We describe herein a new coupling
strategy that introduces the linker to TACAs after their
syntheses are completed and an efficient method to
prepare the GM3 derivative that can be used for this
new strategy.
3
syl azide 8 (H-1: l 4.63, 4.48; J1,2 8.8, 7.4 Hz) in an
excellent yield (93%). Then, deacetylation of 8 under
basic condition, followed by 2,2-dimethoxypropane and
acetic anhydride treatments, gave the acetonide 9. Gly-
cosyl acceptor 10 was finally obtained on the removal
of the isopropylidene group in 9 under acidic condition
in THF aqueous solution.
Our general design is shown in Scheme 1. First, depro-
tected TACAs with an azido group on their reducing
ends (1) will be prepared. As an azido group is stable to
most conditions involved in carbohydrate synthesis, it
may be used as a protecting group of the anomeric
center in the preparation of TACAs. On the other
hand, the azido group can be very easily and selectively
reduced to give the glycosyl amine 2 that will facilitate
the attachment of an acyl group. Thus, after deprotec-
tion of carbohydrates and azido reduction, an acyl
linker can be introduced. Even if benzyl is employed as
the permanent protection of the oligosaccharides, after
reductive debenzylation and concomitant azido reduc-
tion, an acyl group can still be selectively attached to
the amino group, which was shown in the syntheses of
N-linked glycopeptides.27 Therefore, introduction of an
unsaturated acyl group to the free amino group in 2
will afford the glycosyl amide 3. Then, the car-
bonꢀcarbon double bond in 3 can be ozonolyzed to
give an aldehyde 4 which will be finally coupled to the
free amino groups of a carrier protein (5) by well-estab-
lished reductive amination and produce the glycoconju-
gate vaccine 6.
The sialosyl donor 12 was prepared in an excellent
overall yield (76%) following a reported procedure36
(Scheme 3). When sialic acid was treated by an acidic
resin, amblyst-H+, in methanol and then by acetic
anhydride in pyridine, a fully protected derivative of
sialic acid (11) was obtained. Compound 11 was trans-
formed to the glycosyl donor 12 by reaction with
thiophenol using BF3·OEt2 as promoter.
D-lactose
a
OAc
OAc
O
OAc
O
OAc
OAc
O
OAc
O
b
O
AcO
N3
O
AcO
AcO
AcO
OAc
OAc
OAc
OAc
8
c,d,e
OAc
7
Me
Me
OH
HO
O
OAc
O
OAc
O
OAc
OAc
O
f
O
O
O
AcO
O
N3
N3
AcO
OAc
OAc
OAc
OAc
10
9
Scheme 2. Reagents and conditions: (a) Ac2O, NaOAc, reflux,
6 h, 88%; (b) N3TMS, BF3·OEt2, CH2Cl2, 0°C to rt,
overnight, 93%; (c) NaOMe, MeOH, rt, 4 h, quant.; (d)
Me2C(OMe)2, TsOH, DMF, 40°C, 2 days, 92%; (e) Ac2O,
Pyr., rt, overnight, 80%; (f) HCl, THF, H2O, rt, overnight,
40%.
To test the design, we chose GM3 and its conjugates as
the synthetic targets with 4-pentenoyl as the linker.
GM3 is a TACA that is abundant on melanoma, breast
cancer and other tumors.2,6,7 GM3 and its conjugates
are the focus of many biological and synthetic stud-
ies.2,6,7,28–33 Moreover, GM3 contains a sialosyl residue
that often creates problems to the coupling between
carbohydrates and proteins. Thus, a strategy developed
on GM3 may be of general significance. 4-Pentenoyl
was employed as the linker due to its proper chain
length. We have studied the coupling of GM3 to two
proteins, KLH and human serum albumin (HSA), by
the new linker. KLH is a favorite carrier protein for
cancer vaccines due to its excellent immunological
properties22,23 and rich lysine residues for the attach-
ment of TACAs. For instance, Danishefsky and his
coworkers9,18,34,35 used KLH in several cancer vaccines
that are now in clinical trials. On the other hand, HSA
conjugates are widely accepted as the coating antigens
in immunological evaluations of vaccines, though HSA
is not often used as a carrier protein.
OAc
OAc
OAc
OAc
COOMe
SPh
COOMe
OAc
AcO
AcO
b
a
sialic acid
O
O
AcHN
AcHN
AcO
AcO
11
12
c
OAc
OAc
MeOOC
OH
OAc
O
OAc
O
AcO
AcHN
O
O
AcO
O
N3
OAc
AcO
13
OAc
d
OH
HOOC
O
OH
OH
O
OH
OH
HO
HO
O
O
HO
O
AcHN
N3
OH
OH
14
Scheme 3. Reagents and conditions: (a) i. Amblyst-H+,
MeOH, 45–50°C, 6 h; ii. Ac2O, Pyr., rt, overnight, 88%; (b)
,
PhSH, BF3·OEt2, MS 4 A, CH2Cl2, 0°C to rt, overnight, 86%;
The synthesis of the azido derivative of GM3 started
from the modification of lactose (Scheme 2). Thus,
,
(c) 10, NIS, TfOH, MS 3 A, MeCN, −35°C, 2 days, 26%; (d)
1.0 M NaOH, MeOH–H2O (1:1), quant.

J. Xue et al. / Tetrahedron Letters 43 (2002) 1599–1602
1601
hyde. Thus, the product was dissolved in acetone and
water (1:9) and treated with trifluoroacetic acid at room
temperature to afford the hydrated aldehyde 17 (H: l
5.15). Coupling reactions between 17 and proteins were
realized in a 0.1 M aqueous NaHCO3 buffer (pH
7.6–8.0).39 Thus, a solution of 17, NaBH3CN and the
protein, HSA or KLH, in buffer was kept at 37°C in
the dark for 3 days with occasional shaking. The reac-
tion mixture was then dialyzed against distilled water
(with frequent change of water) and finally lyophilized
to afford 18 and 19, respectively.
The glycosylation of 10 by 12 was achieved by
Hasegawa method,29,37,38 i.e. sialosylation in acetonitrile
with glycosyl sulfide as donor and N-iodosuccinimide
(NIS)–triflic acid (TfOH) as promoter. As expected, the
desired trisaccharide 13 (H-1 of lactose: l 4.61, 4.54;
3
3J1,2 8.5, 7.5 Hz; H-3’s of sialic acid: l 2.76, J 12.5, 4.5
3
Hz; l 1.83; J 12.5, 12.0 Hz) was obtained as a major
product (26% isolated yield), and the b-anomer (H-3s
3
3
of sialic acid: l 2.46, J 12.5, 4.5 Hz; l 1.81; J 12.5,
11.5 Hz) as a minor one (8%). The anomeric isomers
were separated by column chromatography, and their
configurations were derived by comparing their 1H
NMR data to that of the similar structures
reported.29,30,37,38 Finally, deprotection of 13 using 1.0
M NaOH in water and methanol (1:1) afforded the
azido derivative of GM3 (14, H-1: l 4.78, 4.54; 3J1,2 8.5,
The carbohydrate loading (Wcarbohydrate/Wconjugate
×
100%) of glycoconjugates 18 and 19 was determined by
examining their sialic acid contents using well-estab-
lished resorcinol method.40 It turned out that 18 and 19
contained 13.4 and 15.9% of carbohydrate, respectively.
They are in the ideal loading range for glycoconjugate
vaccines.
3
8.0 Hz; H-3’s of sialic acid: l 2.77, J 12.5, 4.5 Hz; l
3
1.84; J 12.0, 12.0 Hz; Ac: l 2.04).
The procedure to link GM3 to proteins is shown in
Scheme 4. The azido group was reduced to a free amino
group in a H2 atmosphere using 10% Pd/C as catalyst.
After removal of the catalyst through filtration, 4-pen-
tenoic anhydride was added, and the mixture was
stirred at room temperature overnight. When ninhydrin
test showed no remaining free amine, the solution was
condensed to dryness. The residual product was dis-
solved in water and washed carefully with ethyl ester
and methylene dichloride. The water solution was
lyophilized to give 15 as a white powder (Pent: l 5.85,
Therefore, we describe herein a convergent synthesis of
the azido derivative of GM3, as well as its convenient
coupling to proteins by a new linker. This new coupling
strategy was proved to be very efficient, and it may also
be utilized to prepare other useful neoglycoproteins.
The immunological properties of GM3-KLH conjugate
18 as a cancer vaccine is now under investigation, and
the results will be reported elsewhere.
Acknowledgements
3
5.13, 5.07, 2.53, 2.47, 2.40, 2.39; NH: 5.24, J 2.5 Hz;
H-1: l 4.57; 3J 8.0, 2.5 Hz). The product was practically
pure (1H NMR). Then, ozone was bubbled into a
solution of 15 in methanol at −78°C until it turned to
blue, when dimethyl sulfide was added. The mixture
was warmed up slowly to room temperature and finally
concentrated under vacuum. The reaction offered a
dimethyl acetal 16 (OMe: l 3.35) instead of the alde-
We thank Mr. Jim Faulk for the MS measurements.
This project was supported by American Cancer Soci-
ety (IRG-91-022-06), CWRU Ireland Cancer Center
and
a Research Innovation Award (RI0663) of
Research Corporation.
References
1. Hakomori, S.; Murakami, W. T. Proc. Natl. Acad. Sci.
USA 1968, 59, 254–261.
2. Hakomori, S.; Zhang, Y. Chem. Biol. 1997, 4, 97–104.
3. Hakomori, S. Curr. Opin. Immunol. 1991, 3, 646–653.
4. Fukuda, M. Cancer Res. 1996, 56, 2237–2244.
5. Nilsson, O. APMIS Suppl. 1992, 100, 149–161.
6. Ragupathi, G. Cancer Immunol. Immunother. 1998, 46,
82–87.
7. Livingston, P. O. Curr. Opin. Immunol. 1992, 4, 624–629.
8. Reisfeld, R. A.; Cheresh, D. A. H. Adv. Immunol. 1987,
40, 323–377.
9. Danishefsky, S. J.; Allen, J. R. Angew. Chem., Int. Ed.
Engl. 2000, 39, 837–863.
10. Linehan, D. C.; Goedegebuure, P. S.; Eberlein, T. J.
Annals Surg. Oncol. 1996, 3, 219–228.
11. Ben-Efraim, S. Tumor Biol. 1999, 20, 1–24.
12. Chen, C.-H.; Wu, T.-C. J. Biomed. Sci. 1998, 5, 231–252.
13. Hellstrom, K. E.; Gladstone, P.; Hellstrom, I. Mol. Med.
Today 1997, 286–290.
14. Toyokuni, T.; Singhal, A. K. Chem. Soc. Rev. 1995,
231–242.
Scheme 4. Reagents and conditions: (a) i. 10% Pd/C, H2,
MeOH, rt, 4 h; ii. 4-pentenoic anhydride, MeOH, rt, 12 h,
>95%; (b) O3, MeOH, −78°C to rt, ca. 2–3 h, quant.; (c) TFA,
acetone–H2O (1:9), rt, overnight, quant.; (d) HSA or KLH,
NaBH3CN, 0.1 M NaHCO3 aq. buffer, 37°C, 3 days.

1602
J. Xue et al. / Tetrahedron Letters 43 (2002) 1599–1602
27. (a) Wen, S.; Guo, Z. Org. Lett. 2001, 3, 3773–3776; (b)
Guo, Z.; Nakahara, Y.; Nakahara, Y.; Ogawa, T. Angew.
Chem., Int. Ed. Engl. 1997, 36, 1464–1467 and references
cited therein.
28. Earle, M. A.; Manku, S.; Hultin, P. G.; Li, H.; Palcic, M.
M. Carbohydr. Res. 1997, 301, 1–4.
15. Reichel, F.; Ashton, P.; Boons, G. J. Chem. Commun.
1997, 2087–2088.
16. (a) Slovin, S. F.; Scher, H. I. Semin. Oncol. 1999, 26,
448–454; (b) Slovin, S. F.; Ragupathi, G.; Adluri, S.;
Ungers, G.; Terry, K.; Kim, S.; Spassova, M.; Born-
mann, W. G.; Fazzari, M.; Dantis, L.; Olkiewicz, K.;
Lloyd, K. O.; Livingston, P. O.; Danishefsky, S. J.; Scher,
H. I. Proc. Natl. Acad. USA. 1999, 96, 5710–5715.
17. Tjoa, B. A.; Simmons, S. J.; Elgamal, A.; Rogers, M.;
Ragde, H.; Kenny, G. M.; Troychak, M. J.; Boynton, A.
L.; Murphy, G. P. Prostate 1999, 40, 125–129.
18. MacLean, G. D.; Miles, D. W.; Rubens, R. D.; Reddish,
M. A.; Longenecker, B. M. J. Immunother. Emphasis
Tumor Immunol. 1996, 19, 309–316.
19. Longenecker, B. M.; Reddish, M.; Koganty, R.;
MacLean, G. D. Ann. NY Acad. Sci. 1993, 690, 276–291.
20. Stipp, D. Fortune 1998, 138, 168–176.
21. Lee, Y. C.; Lee, R. T. Neoglycoconjugate: Preparation
and Applications; Academic Press: San Diego, 1994.
22. Helling, F.; Shang, A.; Calves, M.; Zhang, S.; Ren, S.;
Yu, R. K.; Oettgen, H. F.; Linvingston, P. O. Cancer
Res. 1994, 54, 197–205.
23. Livingston, P. O. Immunol. Rev. 1995, 145, 147–156.
24. Hanessian, S. Preparative Carbohydrate Chemistry; New
York: Marcel Dekker, Inc, 1997 and references cited
therein.
25. Allen, J. R.; Allen, J. G.; Zhang, X.; Williams, L. J.;
Zatorski, A.; Ragupathi, G.; Linvingston, P. O.; Dan-
ishefsky, S. J. Chem. Eur. J. 2000, 1366–1375.
26. Allen, J. R.; Danishefsky, S. J. J. Am. Chem. Soc. 1999,
121, 10875–10882.
29. Murase, T.; Ishida, H.; Kiso, M.; Hasegawa, A. Carbo-
hydr. Res. 1989, 188, 71–80.
30. Fujita, S.; Numata, M.; Sugimot, M.; Tomita, K.;
Ogawa, T. Carbohydr. Res. 1992, 228, 347–370.
31. Lonn, H.; Stenvall, K. Tetrahedron Lett. 1992, 33, 115–
116.
32. Eisele, T.; Schmidt, R. R. Liebigs Ann. 1997, 865–872.
33. Nagao, Y.; Nekado, T.; Ikeda, K.; Achiwa, K. Chem.
Pharm. Bull. 1995, 43, 1536–1642.
34. Ragupathi, G.; Park, T. Y.; Zhang, S.; Kim, I. J.;
Graber, L.; Aluri, S.; Lloyd, K. O.; Danishefsky, S. J.;
Livingston, P. O. Angew. Chem., Int. Ed. Engl. 1997, 36,
125–128.
35. Kuduk, S. D.; Schwarz, J. B.; Chen, X. T.; Glunz, P. W.;
Sames, D.; Ragupathi, G.; Livingston, P. O.; Danishef-
sky, S. J. J. Am. Chem. Soc. 1998, 120, 12474–12485.
36. Hasegawa, A.; Murase, T.; Ogawa, M.; Ishida, H.; Kiso,
M. J. Carbohydr. Chem. 1990, 9, 415–428.
37. Ishida, H.; Ohta, Y.; Tsukada, T.; Kiso, M.; Hasegawa,
A. Carbohydr. Res. 1993, 246, 75–88.
38. Murase, T.; Ishida, H.; Kiso, M.; Hasegawa, A. Carbo-
hydr. Res. 1988, c1–c4, 184.
39. Pon, R.; Lussier, M.; Yang, Q.; Jennings, H. J. J. Exp.
Med. 1997, 185, 1929–1938.
40. Svennerholm, L. Biochim. Biophys. Acta 1957, 24, 604–
611.