Efficient synthesis of non-natural ganglioside (pseudo-GM3) and fluorescent labelled lysoGM3 on the basis of polymer-assisted enzymatic strategy

Article abstract of DOI:10.1039/a809729a

The versatility of polymer-assisted enzymatic synthesis of non-natural and biologically significant glycolipid derivatives was demonstrated by constructing pseudo-ganglioside GM3 1 having the trisaccharide sequence Neu5Acα-(2→6)Galβ(1→4)Glc and a fluoresc

Full text of DOI:10.1039/a809729a

Efficient synthesis of non-natural ganglioside (pseudo-GM3) and fluorescent
labelled lysoGM3 on the basis of polymer-assisted enzymatic strategy
Kuriko Yamada, Susumu Matsumoto and Shin-Ichiro Nishimura*
Laboratory for Bio-Macromolecular Chemistry, Division of Biological Sciences, Graduate School of Science,
Hokkaido University, Sapporo 060-0810, Japan. E-mail: nishimura@polymer.sci.hokudai.ac.jp
Received (in Cambridge, UK) 14th December 1998, Accepted 12th January 1999
The versatility of polymer-assisted enzymatic synthesis of
non-natural and biologically significant glycolipid deriva-
tives was demonstrated by constructing pseudo-ganglioside
GM3 1 having the trisaccharide sequence Neu5Aca-
(2?6)Galb(1?4)Glc and a fluorescent labelled lysoGM3
2.
soluble polymer 8 prepared from 7 by the usual radical
copolymerisation with acrylamide was employed for further
sugar elongation and subsequent transglycosylation reactions,
as shown in Scheme 2.
In the previous report,6 it was demonstrated that rat
recombinant a(2?3)-(N)-sialyltransferase (Calbiochem®),
known as an enzyme catalysing sialylation reactions of
glycoprotein biosynthesis in nature, successfully transferred
sialic acid residues to the glycolipid-mimetic polymer 8. In
order to evaluate the versatility of this procedure in other
glycosyltransferases, we decided to synthesise non-natural
ganglioside isomer 1 (pseudo GM3) utilising rat liver a(2?6)-
(N)-sialyltransferase (Boehringer Mannheim). As anticipated,
this enzyme also recognised polymer 8 as a ‘glycoprotein-
mimetic polymer’ rather than as ‘glycolipid-type clusters’.
Novel glycopolymer 9 carrying trisaccharide Neu5Aca-
(2?6)Galb(1?4)Glc was obtained in quantitative yield and
this non-natural trisaccharide moiety was subsequently trans-
ferred from polymer 9 to C16 ceramide using ceramide
glycanase (CGase, Boehringer Mannheim) to afford the target
glycolipid 1† in 60% yield (11 mg) from 9 (Scheme 3). This
result shows clearly that CGase exhibits broad substrate
Cell surface glycoconjugates such as glycoproteins and glyco-
lipids are an important class of biomolecules providing a great
deal of significant information on cell–cell interactions.1 The
advent of efficient and practical methods for glycoconjugate
synthesis is urgently required. Enzymatic synthesis of carbohy-
drates with glycosyltransferases2 and glycosidases3 has been
recognised as a promising practical alternative to chemical
synthesis. Recently, we found an efficient synthetic strategy for
glycoconjugates based on novel water-soluble polymer sup-
ports with specific linkers.4–7 It was suggested that polymer 8
bearing ceramide-mimetic linkers might yield a practical
procedure for the construction of natural sphingoglycolipid
GM3.6 In addition to the synthesis of naturally occurring
gangliosides, our attention is now directed toward the applica-
bility of this strategy to the synthesis of non-natural and
biologically significant glycolipid derivatives. Here we report
efficient syntheses of pseudo-ganglioside GM3 1 having the
O
O
i
C8H17
C8H17
HO
N
HO
N
H
H
HO
OH
CO2H
NHZ
N
HO
AcHN
CPh2
3
O
O
OH
4
OH
O
HO
OH
O
O
HO
O
( )12
OAc
O
AcO
HO
OAc
OH
( )14
O
AcO
Br
HN
O
ii
OH
AcO
O
1 (Pseudo GM3)
OAc
OAc
5
HO
OH
OH
OH
OH
CO2H
OH
O
OAc
O
AcO
O
HO
O
HO
O
OAc
O
O
( )12
O
O
AcHN
AcO
C8H17
O
HO
O
N
OH
AcO
O
HN
H
OH
S
OAc
2 (Dansyl-GM3)
O
OAc
6
N
CPh2
iii, iv
NMe2
OH
OH
trisaccharide sequence Neu5Aca(2?6)Galb(1?4)Glc and a
fluorescent-labeled lysoGM3 2 via enzymatic manipulations on
the polymer support 8.
O
OH
O
HO
C8H17
O
O
N
H
HO
O
O
OH
Scheme 1 indicates an improved synthetic route to key
glycomonomer 7.6 The nucleophilicity of the hydroxy group of
glycosyl acceptor 4,† NA-octyl-N2-diphenylmethylene-l-ser-
inoamide, to the anomeric carbon of a glycosyl donor 5 was
drastically enhanced by employing a Schiff base-type protective
group8 at the amino group, and the yield of the glycoside 6†‡
was improved from 48 to 82%. Removal of the diphenyl-
methylene group by hydrogenation, subsequent coupling with
6-(N-acrylamido)hexanoic acid and O-deacetylation proceeded
smoothly and afforded compound 7 in 78% yield from 6. Water-
( )5
OH
HN
N
7
H
O
Scheme 1 Reagents and conditions : i, Pd-C, H2 gas, MeOH, 25 °C, 6 h, then
Ph2CNNH (1.0 equiv.), CSA (0.1 equiv.), CH2Cl2, 25 °C, 24 h, 82.3% from
3; ii, lactosyl bromide 5 (1.2 equiv.), 4 Å molecular sieves, AgOTf (1.2
equiv.), CH2Cl2, 0 °C, 15 h, 82.0% from 4; iii, Pd-C, H2 gas, MeOH, 25 °C,
2 h, then 6-(N-acrylamido)hexanoic acid (1.1 equiv.), 2-ethoxy-1-ethoxy-
carbonyl-1,2-dihydroquinoline (1.1 equiv.), benzene, EtOH, 25 °C, 24 h,
78.0% ; iv, NaOMe (0.4 equiv.), THF, MeOH, 25 °C, 1 h, 100 % from 6.
Chem. Commun., 1999, 507–508
507

7
selected d-sphingosine (trans-d-erythro-2-amino-4-octade-
cene-1,3-diol) as a potential candidate for the synthesis of
functional sphingoglycolipid derivatives, since the amino group
of d-sphingosine is convenient for further modification. We
were pleased to find that CGase transferred the GM3 oligo-
saccharide [Neu5Aca(2?3)Galb(1?4)Glc], to the primary
hydroxy group of d-sphingosine from the polymer 10 and
intermediate 11 was directly dansylated to afford a fluorescent-
labelled lysoGM3 2† in 50% yield (12 mg) from compound
10.
In conclusion, we have demonstrated the versatility of water-
soluble polymer supports in the enzymatic synthesis of non-
natural and biologically important glycolipid derivatives. It
should be noted that this synthetic strategy should greatly
accelerate efficient combinatorial synthesis of libraries of
glycolipids varying both in the carbohydrate and in the lipid
portions.
i
O
OR1
O
OH
O
O
OH
O
H2N
HO
O
R2O
O
N
H
n
O
OH
OH
HN
N
H
8 R1 = H , R2 = H
m
ii
HO
OH
m : n = 1 : 5
CO2H
9
R1
=
, R2
= H
HO
O
AcHN
HO
iii
HO
HO
OH
CO2H
10 R1 = H , R2
=
O
AcHN
HO
This work was partly supported by a Grant-in-Aid for
Scientific Research from the Ministry of Education, Science
and Culture, Japan (09240101) and by a grant from the Original
Industrial Technology R&D Promotion Program of the New
Energy and Industrial Technology Development Organisation
(NEDO).
Scheme 2 Reagents and conditions: i, acrylamide monomer (5.0 equiv.),
TMEDA (2.4 equiv.), ammonium persulfate (APS) (0.96 equiv.), DMSO,
H2O, 50 °C, 24 h, 92.0% from 7; ii, CMP-Neu5Ac (1.2 equiv. for lactose),
bovine serum albumin (BSA), calf intestinal alkaline phosphatase (CIAP)
(20 units), 50 mm sodium cacodylate buffer (pH 7.40), MnCl2, NaN3, a-2,6
sialyltransferase (0.1 units, 2.7 munits per 1.0 mmol acceptor), 37 °C, 48 h,
100%; iii, CMP-Neu5Ac (1.2 equiv. for lactose), BSA, CIAP (20 units), 50
mm sodium cacodylate buffer (pH 7.40), MnCl2, Triton CF-54, a-2,3
sialyltransferase (0.3 units, 8.0 munits per 1.0 mmol acceptor), 37 °C, 72 h,
100%.
Notes and references
† Selected data for 4: dH(400 MHz, CDCl3) 7.80–7.09 (m, 10 H, 2 3 Ph),
6.92 (br s, 1 H, NH), 4.31–3.81 (m, 2 H, b-H), 4.01 (br s, 1 H, a-H),
1.32–1.24 (m, 12 H, 6 3 CH2) and 0.89 (t, 3 H, J 6.7, CH3). For 6: dH(400
MHz, CDCl3) 7.60–7.03 (m, 10 H, 2 3 Ph), 5.25 (d, 1 H, J 3.4, H-4A), 5.05
(dd, 1 H, J 8.4 and 9.0, H-3), 4.84 (dd, 1 H, J 3.5 and 9.2, H-2A), 4.81 (dd,
1 H, J 3.4 and 9.0, H-3A), 4.79 (dd, 1 H, J 7.6 and 9.0, H-2), 4.38 (d, 1 H,
J 7.6, H-1), 4.32 (d, 1 H, J 7.8, H-1A), 4.29 (dd, 1 H, J 1.7 and 12.1, H-6a),
4.18 (br d, 1 H, J 6.5, a-H), 3.43 (dt, 1 H, J 1.7 and 5.2, H-5), 2.27 (s, 2 H,
NHCH2), 2.06–1.85 (each s, 21 H, OAc), 1.46–1.18 (m, 12 H, 6 3 CH2) and
0.79 (t, 3 H, J 6.7, CH3). For 1: dH(400 MHz, CDCl3) 5.71 (dt, 1 H, J 6.7
and 14.7, Cer-5), 5.46 (dd, 1 H, J 7.6 and 14.7, Cer-4), 4.24 (d, 1 H, J 7.0,
H-1A), 4.23 (d, 1 H, J 7.8, H-1), 2.78 (dd, 1 H, J 4.2 and 12.0, H-3Beq), 2.01
(s, 3 H, NAc) and 0.81 (t, 6 H, J 6.7, 2 3 CH3). For 2: dH(400 MHz, CDCl3)
8.58–7.04 (m, 6 H, dansyl), 4.35 (d, 1 H, J 7.0, H-1A), 4.20 (d, 1 H, J 7.4, H-
1), 2.73 (br d, 1 H, J 4.3, H-3Beq), 1.90 (s, 3 H, NAc) and 0.88 (t, 3 H, CH3);
lex/nm 397.5; lem/nm 460.5.
i
9
1 (Pseudo GM3)
10
ii
HO
OH
HO
OH
OH
OH
CO2H
OH
HO
O
HO
O
O
( )12
O
O
AcHN
O
OH
NH2
OH
11
iii
‡ In the previous report (ref. 6), the coupling reaction of 3 with 5 gave a
glycoside intermediate in 48% yield.
2 (Dansyl-GM3)
λex = 397.5 nm
λem = 460.5 nm
1 Y. C. Lee and R. T. Lee, Acc. Chem. Res., 1995, 28, 321.
2 O. Seiz and C.-H. Wong, J. Am. Chem. Soc., 1997, 119, 8766.
3 C. H. Tran, P. Critchley, D. H. G. Crout, C. J. Britten, S. J. Witham and
M. I. Bird, J. Chem. Soc., Perkin Trans. 1, 1998, 2295.
4 S.-I. Nishimura, K. Matsuoka and Y. C. Lee, Tetrahedron Lett., 1994, 35,
5657.
5 K. Yamada and S.-I. Nishimura, Tetrahedron Lett., 1995, 52, 9493.
6 S.-I. Nishimura and K. Yamada, J. Am. Chem. Soc., 1997, 119, 10555.
7 K. Yamada, E. Fujita and S.-I. Nishimura, Carbohydr. Res., 1997, 305,
443.
Scheme 3 Reagents and conditions : i, C16 ceramide (5.0 equiv.), 50 mm
sodium citrate buffer (pH 6.0), Triton CF-54, ceramide glycanase from
leech (0.005 units), 37 °C, 17 h, 60% from 9; ii, d-sphingosine (5.0 equiv.),
50 mm sodium citrate buffer (pH 6.0), Triton CF-54, ceramide glycanase
from leech (0.005 units), 37 °C, 17 h; iii, dansyl chloride (5.0 equiv.), Et3N
(5.0 equiv.), CHCl3, 25 °C, 2 h, 50% from 10.
specificity of transglycosylation with regard to the carbohydrate
structure of glycolipids.
Next, our interest focused on the substrate specificity of the
transglycosylation reaction carried out by CGase against
glycosyl acceptor substrates,9 as it will significantly influence
the versatility of the present technology in the combinatorial
synthesis of ‘libraries of glycolipids and their mimetics’. We
8 R. Polt, L. Szabo, J. Treiberg, Y. Li and V. Hruby, J. Am. Chem. Soc.,
1992, 114, 10249.
9 Y.-T. Li, B. Z. Carter, B. N. N. Rao, H. Schweingruber and S.-C. Li.,
J. Biol. Chem., 1991, 266, 10723.
Communication 8/09729A
508
Chem. Commun., 1999, 507–508