Chemoenzymatic synthesis of sulfated O-linked oligosaccharides: Epitopes for MECA-79

Article abstract of DOI:10.1016/S0040-4039(02)01795-1

Sulfated oligosaccharides I, II and III, representatives of those found on L-selectin counterreceptors, have been efficiently prepared employing a combination of chemical and enzymatic synthetic methods.

Full text of DOI:10.1016/S0040-4039(02)01795-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 7743–7747
Chemoenzymatic synthesis of sulfated O-linked oligosaccharides:
epitopes for MECA-79
Frederic Be´lot,* David Rabuka, Minoru Fukuda and Ole Hindsgaul
Glycobiology Program, The Burnham Institute, 10901 North Torrey Pines Road, La Jolla, CA 92037, USA
Received 16 July 2002; accepted 27 August 2002
Abstract—Sulfated oligosaccharides I, II and III, representatives of those found on L-selectin counterreceptors, have been
efficiently prepared employing a combination of chemical and enzymatic synthetic methods. © 2002 Elsevier Science Ltd. All
rights reserved.
1. Introduction
which lead us to consider the use of sialyltransferase
and fucosyltransferase as an alternative way to synthe-
size the desired hexasaccharide. Therefore a chemo-
enzymatic strategy was envisaged to deliver the target
oligosaccharides in a more efficient way than the con-
ventional strategy.
Sulfated oligosaccharides are implicated in a number of
biological roles, including development, differentiation,
and homeostasis. Those present on L-selectin counter-
receptors play a critical role in lymphocyte homing and
recirculation.1 L-Selectin counterreceptors on periph-
eral lymph node high endothelial venules in humans
and rodents include a heterogeneous group of glyco-
proteins termed peripheral node addressins (PNAd)
which are recognized by the MECA-79 antibody.2 The
MECA-79 antibody can inhibit lymphocyte adhesion in
vivo, indicating that the MECA-79 epitope binds to the
functional L-selectin ligand.2 The MECA-79 epitope
had been suggested to be a sulfated O-linked oligosac-
charide.3,4 Only recently has the minimum epitope been
identified as a sulfated extended core 1 mucin type
We planned that the tetrasaccharide I would be the key
intermediate of our synthesis as it could be prepared
chemically and then also used as a substrate for the
enzymatic reactions leading to the target oligosaccha-
rides II and III. I could be prepared through the
coupling of two disaccharides, and the sulfation at the
6-O-GlcNAc position was accomplished by employing
a benzyl ether as a temporary protecting group (Fig. 2).
O-glycan, b-
D
-Gal-(1“4)-b-
D
-GlcNAc6SO3Na-(1“3)-
b- -Gal-(1“3)-a-
D
D
-GalNAc-(1“R) which is a back-
bone structure for PNAd that is detected by
MECA-79.5 We report herein the chemical and enzy-
matic synthesis of novel sulfated O-linked oligosaccha-
rides I–III recognized by MECA-79 of which I was
found to be the minimum required epitope (Fig. 1).
Our synthetic scheme was substantially modified from
previous work dedicated to the synthesis of such sul-
fated oligosaccharides.6 During the last decade, a num-
ber of glycosyltransferases have become available7
* Corresponding author. Present address: Unite´ de Chimie
Organique, Institut Pasteur, URA CNRS 2128, 25–28 Rue du Dr
Roux, 75 724 Paris Cedex 15, France. Fax: (+33)(0)1 45 68 84 04;
e-mail: frederic.belot@voila.fr
Figure 1. Target oligosaccharides.
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01795-1

7744
F. Be´lot et al. / Tetrahedron Letters 43 (2002) 7743–7747
Figure 2. Key features of the chemoenzymatic approach to I, II and III.
2. Synthesis of the acceptor 9
residue were masked by a benzylidene acetal with TFA
and benzaldehyde. The acceptor 4 was obtained in an
overall yield of 69% (Scheme 1).
Two major difficulties were envisioned in the synthesis
of the acceptor 9. The starting material GalNAc is very
costly and therefore the desired GalNAc would be
synthesized conveniently and cost effectively through
readily available GlcNAc. The other major difficulty in
the synthesis would be the Gal(b1“3)GalNAc linkage.
As described previously,8 4-methoxyphenyl 3-O-ally-b-
D
-galactopyranoside 5 was prepared from a commer-
cially available galactose penta-acetate. After
benzoylation of 5 (BzCl, pyridine, 80%), anomeric posi-
tion of 6 was deblocked (CAN, toluene, acetonitrile,
water) and subsequently converted to a trichloroacetim-
idate 7 (CCl3CN, DBU, DCM) in 83% overall yield
(two steps) (Scheme 2).
The GalNAc derivative 3 was prepared from com-
pound 19 by a literature procedure10 using intramolecu-
lar SN2 displacement of the 4-O-triflate by the
3-O-pivaloyl group. The correct galacto structure was
1
deduced from the H NMR spectra where H-4 signals
Use of benzoyl ester as neighboring group at the posi-
tion 2 was preferred to acetyl group because, in our
hands, the desired 1,2-trans glycosidic linkage was
preferably obtained with the former (details not
presented).
resonated at 5.30 ppm (J3,4=3.2 Hz, J4,5<1.0 Hz) for 3a
and 5.32 ppm (J3,4=3.1 Hz, J4,5<1.0 Hz) for 3b. Both
compounds 3a and 3b were O-depivaloylated using
Zemplen conditions, and the 4 and 6 positions of the
Scheme 1. Reagents and conditions: (a) PivCl, Py, DCM, 67% (two steps from 2-deoxy-2-acetamido-glucopyrannose); (b) Tf2O,
Py, DCE then H2O, 90°C; (c) MeONa, MeOH, 80% (two steps); (d) PhCHO, TFA, 76%.

F. Be´lot et al. / Tetrahedron Letters 43 (2002) 7743–7747
7745
Scheme 2. Reagents and conditions: (a) BzCl, Py, 80%; (b) CAN, Tol, CH3CN, H2O; (c) CCl3CN, DBU, DCM, 83% (two steps).
Coupling of 4 and 7 was accomplished in slightly
harsher conditions (DCE, TfOH cat., molecular sieves,
55°C)15 that afforded the Gal-GalNAc disaccharide 8 in
good and reproducible yields of 69% (1H NMR data:
Ref. 14). These unusual coupling conditions helped us
to avoid orthoester as a side product. These conditions
were also used for two other glycosylations in our goal
to furnish the synthesis of tetrasaccharide I.
b(1“4) substitution.14 The anomeric position was
deblocked by TFA/water and was subsequently con-
verted to the imidate 14 (74%, two steps) (Scheme 4).
4. Synthesis of the sulfotetrasaccharide I and its
unsulfated analogue I%
Glycosylation between 14 and 9 used the previously
mentioned conditions,15 which gave the fully protected
tetrasaccharide 1514 in a satisfactory yield (50%). The
phthalimido group was replaced by acetamido group to
afford 16 (72%) (Scheme 5).
Removal of the benzylidene group of 8 (AcOH/water)
followed by treatment with Ac2O, pyridine (81% two
steps), afforded the ester protected intermediate, which
was then treated with Pd(Cl)2 in MeOH to afford the
disaccharide acceptor 9 (80%) (Scheme 3).
Treatment with Pd/C and H2 selectively cleaved the
benzyl ether to give 17 and sulfation was carried out
using Me3N·SO3 in DMF to give the 6-O-sulfated
tetrasaccharide 18 in 76% overall yield. Final saponifi-
cation (MeOH, MeONa) yielded quantitatively the
target sulfated oligosaccharide I (Scheme 6). In addi-
tion, saponification of 17 led quantitatively to the ana-
logue non-sulfated tetrasaccharide I%.18
3. Synthesis of the donor 14
According to our synthetic strategy, a benzyl group was
introduced at the 6 position of the GlcNAc residue to
temporarily mask the position to be sulfated, and the
desired GlcNAc derivative 11 was readily obtained
from the known 2-(trimethylsilyl)ethyl 3-O-acetyl-4,6-
O-benzylidene-2-deoxy-2-phthalimido-b-D-glucopyran-
oside 1011 by reductive opening of the benzylidene
group, by treatment with Et3SiH, and BF3·Et2O in
DCM.12 Under the normal conditions we found that
the glycosylation of 11 and 1213 gave a very poor yield;
however, when new conditions were employed15 disac-
charide 13 was obtained in a good yield of 85%. The 1H
NMR spectrum confirmed unambiguously the correct
5. Enzymatic sialylation and fucosylation of
sulfonato-pentasaccharide II and
sulfonato-hexasaccharide III
We planned to use enzymes for introducing key sialic
acid and fucose residues to the sulfated hexasaccharide.
,
Scheme 3. Reagents and conditions: (a) cat. TfOH, DCE, 4 A molecular sieves, 55°C, 69%; (b) AcOH/H2O, 90°C; (c) Ac2O, Py,
81% (two steps); (d) Pd(Cl)2, MeOH, 80%.
,
Scheme 4. Reagents and conditions: (a) Et3SiH, BF3·Et2O, DCM, 66%; (b) cat. TfOH, DCE, 4 A molecular sieves, 55°C, 85%; (c)
TFA, DCM, then CCl3CN, DBU, DCM, 74% (two steps).

7746
F. Be´lot et al. / Tetrahedron Letters 43 (2002) 7743–7747
,
Scheme 5. Reagents and conditions: (a) cat. TfOH, DCE, 4 A molecular sieves, 55°C, 50%; (b) ethylene diamine, t-BuOH, 90°C,
12 h, then Ac2O, MeOH, Et3N, 72% (two steps); (c) Pd/C, H2, EtOH, 80%.
Scheme 6. Reagents and conditions: (a) (CH3)3N·SO3, DMF, 95%; (b) MeOH, NaOMe, quant.
Scheme 7. Reagents and conditions: (a) SAT(N), CMP-NANA, MOPS (pH 7.4), Triton S4, BSA, H2O, CIP, 55%; (b) FucT(V),
GDP-Fuc, MnCl2, DTT, Tris–HCl (pH 7.4), quant.
The sulfation is thought to be the final step in the
proposed pathway of 6-O-sulfated Sialyl Lewisx.16,17
We were therefore interested in determining whether
sialyltransferase and/or fucosylfransferase could take
the sulfated oligosaccharide as a substrate. With tetra-
saccharide I in hand, selective sialylation was achieved

F. Be´lot et al. / Tetrahedron Letters 43 (2002) 7743–7747
7747
J
J
2,NH=7.6 Hz, GalNAc N-H), 5.48 (dd, 1H, J1,2=7.8 Hz,
2,3=10.0 Hz, Gal H-2), 5.39 (s, 1H, PhCH), 5.10 (m, 2H,
in a single step using a2,3-sialyl transferase and CMP-
NANA. The reaction did not go to completion, how-
ever, ion exchange chromatography enabled us to
isolate the desired pentasaccharide II in 55% yield. For
fucosylation, incubation of II with FucT-V and GDP-
Fuc gave the sulfonato-hexasaccharide III in a nearly
quantitative yield.18,19 Both enzymes demonstrated a
tolerance for sulfated acceptor substrates as the enzymes
in the biosynthetic route for the MECA-79 epitope
(Scheme 7).
OAll), 5.06 (d, 1H, J1,2=3.3 Hz GalNAc H-1), 5.04 (d,
1H, Gal H-1), 4.60 (dd, 1H, J5,6a=7.3 Hz, J6a,6b=11.5
Hz, Gal H-6a), 4.51 (m, 1H, J2,3=11.3 Hz, GalNAc H-2),
4.42 (dd, 1H, J5,6b=5.1 Hz, Gal H-6b), 4.40 (dd, 1H,
J
3,4=3.0 Hz, J4,5<1.0 Hz, GalNAc H-4), 4.20 (m, 1H, Gal
H-5), 4.15–3.70 (m, 6H, GalNAc H-6a, H-6b, H-3, Gal
H-3, OAll), 3.60–3.30 (m, 3H, GalNAc H-5, O-Octyl),
1.60 (s, 3H, Ac). 13: 5.90 (dd, 1H, J3,4=3.4 Hz, J4,5<1 Hz,
Gal H-4), 5.80 (dd, 1H, J2,3=10.8 Hz, J3,4=8.9 Hz,
GlcNAc H-3), 5.68 (dd, 1H, J1,2=7.9 Hz, J2,3=10.4 Hz,
Gal H-2), 5.40 (dd, 1H, Gal H-3), 5.32 (d, 1H, J1,2=8.5
Hz, GlcNAc H-1), 4.80 (d, 1H, Gal H-1), 4.85 and 4.40
(2d, 2H, PhCH2), 4.43 (m, 2H, Gal H-6a H-6b), 4.25 (dd,
1H, GlcNAc H-2), 4.17 (t, 1H, J4,5=10.5 Hz, GlcNAc
H-4), 4.06 (m, 1H, Gal H-5), 3.90 (m, 1H, OSE), 3.71 (dd,
1H, GlcNAc H-6a), 3.60–3.50 (m, 2H, GlcNAc H-5
H-6b), 3.42 (m, 1H, OSE), 1.92 (s, 3H, OAc), 0.80 (m, 2H,
OSE) and −0.1 (s, 9H, OSE). 15: 5.82 (dd, 1H, J3,4=3.2
Hz, J4,5<1 Hz, Gal H-4), 5.80 (dd, 1H, J3,4=3.0 Hz,
J4,5<1.0 Hz, Gal H-4), 5.60–5.30 (m, 6H, GlcNAc H-3,
GalNAc H-4 and N-H, Gal 2H-2 H-3), 5.02, 4.80 and
4.65 (3d, 3H, J1,2=8.2 Hz, J1,2=7.9 Hz, J1,2=8.0 Hz, Gal
2H-1, GlcNAc H-1), 4.76 (d, 1H, J1,2=3.3 Hz, GalNAc
H-1), 4.54 and 4.22 (2d, 2H, PhCH2), 4.46–4.24 (m, 7H,
GlcNAc H-2 H-4, GalNAc H-2, Gal 4×H-6), 4.10–3.50
(m, 9H, Gal 2×H-5 H-3, GlcNAc 2×H-6, H-5, GalNAc
H-5, 2xH-6), 1.97, 1.93, 1.70, 1.60 (4s, 12H, Ac).
6. Conclusion
Sulfated oligosaccharides were efficiently synthesized
using chemoenzymatic synthetic scheme with the
enzymes utilizing the sulfated substrates. In addition,
new glycosylations conditions were tested with success
when standard conditions were unsuitable, and it was
demonstrated that controlled heating of glycosylation
reaction mixtures was a way to activate acceptors with-
out degradation.
Acknowledgements
This work was supported by the NIH grant
P01CA71932.
15. Standard procedure: A mixture of donor (case 1: 7, 1.7
equiv.; case 2: 12, 2 equiv.; case 3: 14, 1.5 equiv.),
acceptor (case 1: 4, 1 equiv.; case 2: 11, 1 equiv.; case 3:
References
,
1. Rosen, S. D. Am. J. Pathol. 1999, 155, 1013–1020.
2. Steeter, P. R.; Rouse, B. T.; Butcher, E. C. J. Cell. Biol.
1988, 107, 1853–1862.
3. Hemmerich, S.; Butcher, E. C.; Rosen, S. D. J. Exp. Med.
1994, 180, 2219–2226.
4. Hiraoka, N.; Petryniak, B.; Nakayama, J.; Tsuboi, S.;
Susuki, M.; Yeh, J. C.; Izawa, D.; Tanaka, T.; Lowe, J.
B.; Fukuda, M. Immunity 1999, 11, 79–89.
5. Yeh, J. C.; Hiraoka, N.; Petryniak, B.; Nakayama, J.;
Ellies, L. G.; Rabuka, D.; Hindsgaul, O.; Marth, J. D.;
Lowe, J. B.; Fukuda, M. Cell 2001, 105, 957–969.
6. Misra, A. K.; Ding, Y.; Lowe, J. B.; Hindsgaul, O.
Bioorg. Med. Chem. Lett. 2000, 10, 1505–1509.
7. Koeller, K. M.; Wong, C. H. Chem. Rev. 2000, 100,
4465–4494.
9, 1 equiv.) and 4 A powdered molecular sieves in anhy-
drous 1,2-dichloroethane was stirred for 1 h at rt under
dry Ar. Triflic acid was added dropwise and the mixture
was stirred for 1 h 30 min at 55°C. After cooling to rt,
Et3N was added, and the mixture was filtered and concen-
trated. The residue was eluted from a column of silica gel
to provide the desired compound (case 1: 8, 69%; case 2:
13, 85%; case 3: 15, 50%).
16. Bowman, K. G.; Cook, B. N.; de Graffenried, C. L.;
Bertozzi, C. R. Biochemistry 2001, 40, 5382–5391.
17. Koeller, K. M.; Smith, M. E. B.; Wong, C.-H. J. Am.
Chem. Soc. 2000, 122, 742–743.
18. Partial NMR and MS data for final products: I%: 1H NMR
(D2O): l: 4.89 (d, 1H, J1,2=3.6 Hz, H-1GalNAc), 4.72 (d,
1H, J1,2=8.3 Hz, H-1GlcNAc), 4.47 (d, 1H, J1,2=7.3 Hz,
H-1Gal), 4.44 (d, 1H, J1,2=8.0 Hz, H-1Gal), 2.04 and 2.03
(2s, 6H, 2 NHAc); 13C NMR: 105.55, 103.69, 103.48,
97.80 (4 C-1); HRMS calcd for C36H64N2NaO21 (M+Na+):
883.389, found 883.388. I: 1H NMR (D2O): l: 4.88 (d,
1H, J1,2=3.6 Hz, H-1GalNAc), 4.75 (d, 1H, J1,2=8.2 Hz,
H-1GlcNAc), 4.53 (d, 1H, J1,2=7.5 Hz, H-1Gal), 4.48 (d, 1H,
8. Bazin, H. G.; Du, Y.; Polat, T.; Linhardt, R. J. J. Org.
Chem. 1999, 64, 7254–7259.
9. Aguilera, B.; Romero-Ramirez, L.; Abad-Rodriguez, J.;
Corrales, G.; Nieto-Sampedro, M.; Fernandez-May-
oralas, A. J. Med. Chem. 1998, 41, 4599–4606.
10. (a) Lay, L.; Nicotra, F.; Panza, L.; Russo, G. Helv. Chim.
Acta 1994, 77, 509–514; (b) Be´lot, F.; Jacquinet, J.-C.
Carbohydr. Res. 2000, 325, 93–106.
11. Kartha, K. P. R.; Field, R. A. Tetrahedron 1997, 34,
11753–11766.
12. Debenham, S. D.; Toone, E. J. Tetrahedron: Asymmetry
2000, 11, 385–387.
J
1,2=7.9 Hz, H-1Gal), 2.05 and 2.02 (2s, 6H, 2 NHAc); 13
C
NMR: 105.52, 103.62, 103.44, 97.65 (4 C-1); HRMS calcd
for C36H63N2Na2O24S (M+Na+): 985.328, found 985.325.
III: 1H NMR (D2O): l: 5.10 (d, 1H, J1,2=3.9 Hz, H-1Fuc),
4.72 (d, 1H, J1,2=3.5 Hz, H-1GalNAc), 4.59 (d, 1H, J1,2
=
8.0 Hz, H-1GlcNAc), 4.42 (m, 2H, 2 H-1Gal), 2.72 (dd, 1H,
H-3e), 2.02–1.98 (3s, 9H, 3 NHAc), 1.78 (t, 1H, H-3a),
13. Rio, S.; Beau, J.-M.; Jacquinet, J.-C. Carbohydr. Res.
1991, 219, 71–90.
14. Partial H NMR (CDCl3): 8: 5.85 (dd, 1H, J3,4=3.1 Hz,
1.18 (d, 3H,
J5,6=6.5 Hz, H-6Fuc); MS calcd for
C53H90N3NaO36S (M+): 1399.49, found 1399.4.
19. Sialyltransferase: ST3GalIII, Calbiochem (566218) and
Fucosyltransferase: FucT-V, Calbiochem (344320).
1
J4,5<1.0 Hz, Gal H-4), 5.60 (m, 1H, OAll), 5.52 (d, 1H,