Nitroglycal concatenation: A broadly applicable and efficient approach to the synthesis of complex O-glycans

Article abstract of DOI:10.1002/1521-3773(20010716)40:14<2654::AID-ANIE2654>3.0.CO;2-L

Base-catalyzed glycosylations provide the basis for a new and general entry to the synthesis of mucin-type O-glycans. The desired α-linked 2-acetamidoglycosyl amino acids B are accessible selectively starting from glycals of type A. Fmoc = 9-fluorenylmeth

Full text of DOI:10.1002/1521-3773(20010716)40:14<2654::AID-ANIE2654>3.0.CO;2-L

COMMUNICATIONS
bonds are formed in a ªcritical rangeº of DpKa values scattered
roughly around 1.0.[12, 13] For substituted pyridine ´ PCP adducts in
CCl4 solution, an ideal DpKa for ª50% proton transferº is reported as
1.6.[13a] Since a more polar environment generally favors more ionic
forms, it must be expected that in crystals, a pyridine ´ PCP adduct with
Nitroglycal Concatenation: A Broadly
Applicable and Efficient Approach to the
Synthesis of Complex O-Glycans**
‡
Gottfried A. Winterfeld and Richard R. Schmidt*
DpKa ˆ 1.6 is formally ionic (N H ´´´ O ) and adducts with really
50% proton transfer are characterized by DpKa < 1.6. Because the
influence of the crystalline environment cannot be predicted quanti-
tatively, one cannot predict the exact hydrogen-bond geometry from a
known DpKa value and molecular geometry. In any case, the DpKa of
0.77[12] makes 4-MePy ´ PCP an interesting candidate in the search for
an O H N bond with a centered H atom (pKa ˆ 5.22 (Py), 6.03 (4-
MePy), 5.26 (PCP)).[12a]
The mucin class of glycopetides has attracted much
attention in recent years, because it subsumes numerous
structures of fundamental importance in biological processes
such as fertilization, parasitic infection, inflammation, im-
mune defense, cell-growth, and cell ± cell adhesion.[1] Syn-
thesis of the characteristic a-glycosidic linkage between
2-acetamido-2-deoxy-d-galactopyranose and the hydroxy
groups of l-serine and l-threonine, however, proved difficult.
Most syntheses of a-O-linked glycopeptides rely essentially
on the methodology introduced by Paulsen in 1978: The
glycosylations are carried out with glycosyl donors that have a
non-participating azido group at position 2 as latent amino
function as well as a leaving group at the anomeric center.[2, 3]
Enzymatic syntheses have also been reported, for example the
synthesis of Core 1 and the corresponding sialylated Core 1
structure.[4] Core structures are defined as the binding region
of the saccharides directly bound to the protein.[19] Recently,
we have shown that for the synthesis of the simplest mucin
structure, the TN antigen, Michael addition to 2-nitrogalactal
may serve as an efficient alternative approach.[5] This
fundamentally new approach has now been developed to a
comprehensive and powerful methodology that provides
highly stereoselective access to 3-O- and 6-O-branched mucin
structures.
All mucin core structures contain at the reducing end a
N-acetylgalactosamine a-glycosidically linked to l-serine or
l-threonine. Eight core structures of mucin-type glycopep-
tides have been identified to date; they bear additional
glycosyl residues at either position 6 or position 3 or at both
positions to form complex O-glycans (Scheme 1). To demon-
strate that nitroglycal concatenation is a well-suited tool for
the synthesis of all members of the mucin family we
strategically chose two target molecules from the 6-O-
branched structures (STN antigen and Core 7) and one target
molecule from the 3-O-branched structures (Core 1). These
structures are generated by reaction sequences I ± III
(Scheme 2 and 3).
[15] I. Majerz, Z. Malarski, L. Sobczyk, Chem. Phys. Lett. 1997, 274, 361 ±
364.
[16] Z. Malarski, I. Majerz, T. Lis, J. Mol. Struct. 1987, 158, 369 ± 377.
[17] The observation that the H atom is on time-average closer to the O
atom at room temperature, and closer to the N atom at 20 K could
have a number of reasons: 1) The substance might be dimorphic, with
one crystal form having a different kind of hydrogen bond than the
other. 2) There might be gradual proton migration upon cooling; this
is the effect that was actually found in the second experiment. 3) The
H atom might be disordered over two positions, one closer to the O
atom and the other closer to the N atom, and relative occupancies
change with temperature; similar effects are known, for example, with
the O H ´´´ O hydrogen bond in benzoic acid: C. C. Wilson, N.
Shankland, A. J. Florence, J. Chem. Soc. Faraday Trans. 1996, 92,
5051 ± 5057. 4) The difference between the structures at room temper-
ature and at 20 K might be nothing more than an artifact caused by
experimental inaccuracies.
[18] a) C. C. Wilson in Neutron Scattering Data Analysis (Ed.: M. W.
Johnson), Adam Hilger, Bristol, 1990, chap. 2; b) C. C. Wilson, J. Mol.
Struct. 1997, 405, 207 ± 217.
[19] H.-B. Bürgi, J. D. Dunitz, Acc. Chem. Res. 1983, 16, 153 ± 161.
[20] Data point 2 is from 2,4-Me2Py ´ PCP at 20 K; crystal structure to be
published elsewhere. Data point 3 is from 2-MePy ´ PCP at 30 K: T.
Steiner, C. C. Wilson, I. Majerz, Chem. Commun. 2000, 1231 ± 1232.
[21] T. Steiner, W. Saenger, Acta Crystallogr. Sect. B 1994, 50, 348 ± 357.
[22] S. N. Smirnov, N. S. Golubev, G. S. Denisov, H. Benedict, P. Schah-
Mohammedi, H.-H. Limbach, J. Am. Chem. Soc. 1996, 118, 4094 ±
4101.
[23] In 4-MePy ´ PCP, the hydrogen bond is geometrically centered at
about 90 K. This H position is singular in geometric terms, but not the
one where the O H and H N bonds have equal ªbond ordersº s ˆ 1ꢁ2.
Free O H and N H covalent bonds have different lengths (e.g. 0.957
and 1.012 Š in H2O and H3N vapor, respectively), and also O H and
N H bonds with s ˆ 1ꢁ2 should have different X H distances. Even the
most recent parametrization of distance/valence relations is very
inaccurate for strong O H N bonds, because no experimental data
from that region could be used,[11] but suggests that an O H bond with
s ˆ 1ꢁ2 is 0.060 Š shorter than an N H bond with s ˆ 1ꢁ2. According to
Table 1, this situation occurs in 4-MePy ´ PCP between 125 and 150 K.
For reaction sequences I and II as well as for the synthesis
of TN building blocks, nitroglycal 2 is the key intermediate of
our synthesis. Nitroglycal 2 can be obtained in 84% yield from
protected galactal 1[6] by using a two-step, one-pot procedure
involving addition of acetyl nitrate to the glycal functionality
and subsequent elimination of acetic acid (Scheme 2).[5]
[*] Prof. Dr. R. R. Schmidt, Dr. G. A. Winterfeld
Fachbereich Chemie, Universität Konstanz, Fach M 725
78457 Konstanz (Germany)
Fax : (‡49)7531-883135
E-mail: Richard.Schmidt@uni-konstanz.de
[**] This work was supported by the Deutsche Forschungsgemeinschaft
and the European Community (grant no. FAIR-CT 97-3142). G.A.W.
gratefully acknowledges a RIKEN/Studienstiftung des Deutschen
Volkes fellowship. We are grateful to Dr. A. Geyer for his help in the
structural assignments by NMR experiments and to Dr. K.-H. Jung for
his help in the preparation of the manuscript.
2654
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4014-2654 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2001, 40, No. 14

COMMUNICATIONS
Protected serine 3a and threonine 3b add to this glycosyl
donor 2 in virtually quantitative yield and under complete
stereocontrol to form the desired a-linked 2-nitroglycosides
4a and 4b, respectively. The base-catalyzed glycosylation,
which requires 0.1 equivalents of potassium tert-butoxide, is
performed at room temperature and reaches completion
within 10 min in the serine case and within 120 min in the
threonine case. Compared to our previous observations with a
perbenzylated nitrogalactal,[5] glycosyl donor 2 bearing a
sterically demanding protective group at position 6 shows
significantly enhanced reactivity and stereocontrol.
HO
R3O
OR6
O
R6
+
Core 6:
Core 7:
GlcNAc β (1-6)
GalNAc α (1-6)
Neu5Ac α (2-6)
NH2
AcHN
O
O
STN-Antigen:
R
O
TN Antigen
R3
R3, R6
Core 1: Gal β (1-3)
Core 2:
Core 4:
GlcNAc β (1-6)
Gal β (1-3)
Core 3: GlcNAc β (1-3)
Core 5: GalNAc α (1-3)
Core 8: Gal α (1-3)
GlcNAc β (1-6)
GlcNAc β (1-3)
The peracetylated TN antigens 10a and 10b are accessible
from glycosides 4a and 4b in 81% yield after removal of silyl
protective groups, reduction of the nitro group with Raney
Scheme 1. Core structures of mucin-type O-glycans.
BnO
NHBoc
OTBDPS
O
HO
OtBu
BnO
a, b
BnO
R
O
OTBDPS
1
O
3a: R = H
3b: R = Me
BnO
2 (84%) NO2
reaction sequence I
acid-catalyzed 6-O elongation
reaction sequence II
reiterative, base-catalyzed 6-O elongation
KOtBu,
toluene
3a, 3b
BnO
OH
BnO
OH
BnO
BnO
OTBDPS
O
c-f
c
O
O
BnO
BnO
NHBoc
NHBoc
OtBu
NHBoc
OtBu
AcHN
O2N
O2N
O
OtBu
O
O
O
R
O
R
O
5a: R = H
5b: R = Me
6a (76%)
(97%)
(97%)
4a: R = H
4b: R = Me
(97%)
(97%)
OAc
BnO
OTBDPS
O
AcO
OP(OEt)2
KOtBu,
toluene
TMSOTf, EtCN,
-45°C
BnO
O
c - e, g, e
CO2Me
AcHN
AcO
NO2
7
AcO
2
OAc
AcO
CO2Me
O
BnO
BnO
AcO
OAc
OTBDPS
O
O
AcHN
AcO
O
BnO
AcO
BnO
AcO
BnO
NHBoc
OtBu
X
AcHN
O
X
O
O
O
BnO
NHBoc
NHBoc
OtBu
AcHN
R
O
O
OtBu
O
X = NO2 (47%)
8a: R = H,
R
O
O
8b: R = Me,
(65%)
11a: X = NO2
X = NO2
(70%)
(82%)
d, e
9a: R = H,
9b: R = Me,
X = NHAc
10a: R = H,
10b: R = Me,
iv, v
(75%)
(81%)
(81%)
12a: X = NHAc
X = NHAc (85%)
g, e, h, i
h, i
c, g, e, h, i
OAc
AcO
AcO
OAc
CO2Me
AcO
AcO
OAc
O
O
O
O
AcO
AcO
AcHN
AcO
AcO
AcO
NHFmoc
OH
AcO
AcHN
O
AcHN
O
O
O
NHFmoc
OH
AcO
NHFmoc
OH
AcHN
R
O
O
AcHN
O
14a: R = H
14b: R = Me
(90%)
(88%)
15a (68%)
O
13b (98%)
Me
O
Scheme 2. Synthesis of TN antigen, STN antigen, and Core 7 building blocks for glycopeptide synthesis; a) HNO3, Ac2O; b) Et3N, CH2Cl2; c) TBAF, AcOH,
THF; d) Raney-Ni T4 (Pt), H2, EtOH; e) Ac2O, pyridine; f) NaOMe, MeOH; g) Pd/C, H2, AcOH, MeOH; h) TFA, CH2Cl2; i) Fmoc-ONSu, NaHCO3,
MeCN, H2O. Ac ˆ acetyl, Bnˆ benzyl, Boc ˆ tert-butoxycarbonyl, Fmoc ˆ fluoren-9-ylmethoxycarbonyl, Su ˆ succinimidyl, TBDPS ˆ tert-butyldiphenylsilyl,
TFA ˆ trifluoroacetic acid.
Angew. Chem. Int. Ed. 2001, 40, No. 14
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4014-2655 $ 17.50+.50/0
2655

COMMUNICATIONS
O
acceptor 6a is glycosylated making repeated use of
nitrogalactal 2. Again the glycosylation affords
stereoselectively the a-glycoside 11a. The second
glycosylation cycle is completed by reduction of the
nitro group and N-acetylation of the resulting
amine to afford 12a. The synthesis of building
block 15a[13] is achieved after exchange of all
protective groups on the carbohydrate moiety of
12a for acetyl groups, removal of both Boc and tert-
butyl groups, and installation of the Fmoc protec-
tive group.
Having demonstrated the efficiency of nitroglycal
concatenation for the synthesis of unsubtituted and
6-O-substituted O-glycans by elongation of a pre-
formed TN structure, we devised a Core 1 synthesis
to open the methodology towards 3-O-branched
structures and to the use of disaccharidic glycals in
the nitration/Michael addition protocol. We started
from known disaccharide glycal 16,[14] which was
converted to the per-O-benzylated derivative 17
(Scheme 3, reaction sequence III). Glycal 17 under-
goes addition of acetyl nitrate and elimination of
acetic acid as described for 1[5] to afford the
corresponding nitroglycal 18. This Michael acceptor
was glycosylated under standard conditions with 3a
and 3b to give the corresponding a-glycosides 19a
and 19b. Reduction of the nitro groups proceeded
similarly to all previous examples and afforded the
acetamido glycosides 20a and 20b after N-acetyla-
tion. Benzyl protective groups were exchanged for
acetyl protective groups, the protective groups on
the amino acid moiety were removed, and finally
BnO
BnO
HO
BnO
O
O
OBn
O
OTBDPS
OTBDPS
O
O
O
a – c
O
O
BnO
OH
OBn
X
16
17: X = H
18: X = NO2
(67%)
(88%)
d, e
KOtBu,
toluene
3a, 3b
BnO
reaction sequence III
BnO
O
OBn
O
base-catalyzed glycosylation with
a disaccharidic nitroglycal
BnO
BnO
O
NHBoc
OtBu
X
OBn
O
R
O
19a: R = H,
19b: R = Me,
(91%)
(93%)
X = NO2
X = NO2
f,
g
20a: R = H, X = NHAc
20b: R = Me,X = NHAc
(87%)
(81%)
h, g, i, j
AcO
AcO
AcO
AcO
O
OAc
O
O
NHFmoc
OH
AcHN
OAc
O
Me
O
21b: R = Me
(85%)
Scheme 3. Synthesis of a Core 1 building block for glycopeptide synthesis; a) NaOMe,
MeOH; b) TBAF, THF; c) BnBr, NaH, DMF; d) Ac2O, HNO3; e) Et3N, CH2Cl2;
f) Raney-Ni T4 (Pt), H2, EtOH; g) Ac2O, pyridine; h) Pd/C, H2, AcOH, MeOH; i) TFA,
CH2Cl2; j) Fmoc-ONSu, NaHCO3, MeCN, H2O.
nickel T4/H2,[5] N-acetylation, and protective group manipu-
lations (Scheme 2). Known[7] building blocks for glycopeptide
synthesis 14a and 14b[8] are obtained after cleavage of Boc
and tert-butyl protective groups on the amino acid moiety and
installation of the Fmoc protective group.
the Fmoc protective group was introduced to give the
known[7b,g, 15] target building block 21b.[16]
In conclusion, the syntheses of TN and STN antigens as well
as Core 7 and Core 1 structures have been achieved by using
Michael addition reactions to nitroglycals as the key reaction.
Nitroglycal concatenation has been applied reiteratively and
combined with either anomeric leaving-group-based glycosy-
lations[17] or the glycal assembly method,[18] demonstrating its
versatility. The readily available starting nitroglycals as well as
the stereochemically defined outcome of all base-catalyzed
glycosylations shown here highlight this new methodology.
The STN antigen 13b displays an a-linked N-acetylneu-
raminic acid unit at the 6-O position of the TN structure. To
arrive at a suitable glycosyl acceptor for the acid-catalyzed
sialylation reaction, silyl protection at position 6 was removed
by using a buffered tetrabutylammonium fluoride (TBAF)
solution in THF (Scheme 2, reaction sequence I).[9] Nitro-
glycosides 5a and 5b can be applied to a sialylation reaction
with neuraminic acid phosphite 7[10] under Lewis acid
catalysis. At 458C in propionitrile the reaction affords a-
linked sialosides 8a and 8b. The nitro group of these
disaccharides can then be reduced to the amine and N-acety-
lated to afford derivatives 9a and 9b. Removal of the benzyl
protective groups, peracetylation, removal of the tert-butyl
protective group and exchange of the Boc for the Fmoc
protective group affords the known STN building block
13b[11, 12] suitable for glycopeptide synthesis.
Experimental Section
2: Concentrated nitric acid (24 mL, 0.38 mol) was added dropwise to acetic
anhydride (240 mL) at 108C under constant stirring. The external temper-
ature was further lowered to 108C to keep the internal temperature in the
range of 10 ± 208C during the addition. Once the addition was complete, the
solution was cooled further to 508C upon which a precipitate formed.
Then a solution of galactal 1 (30 g, 0.053 mol) in acetic anhydride (120 mL)
was added over a period of 10 ± 15 min, and the mixture stirred at this
temperature for 30 min. After the reaction mixture had been allowed to
warm to 228C, it became clear. The reaction mixture was poured into iced
water (500 mL), brine (250 mL) was added, and the aqueous layer
extracted with diethyl ether (3 Â 300 mL). The combined organic extracts
were dried over sodium sulfate and the solvents removed by coevaporation
with toluene. The crude intermediate 2-nitrogalactopyranose was dissolved
in dichloromethane (50 mL) and slowly added to an ice-cold, stirred
solution of triethylamine (22 mL, 0.159 mol) in dichloromethane (50 mL).
After complete addition the cooling bath was removed and stirring
Core 7 bears a second a-linked galactosamine moiety at
position 6 of the TN structure. This second unit may be
installed by using a reiterative Michael-addition approach
(Scheme 2, reaction sequence II). The 6-position of nitro-
glycoside 4a is unmasked as mentioned before and the nitro
group reduced to the amine and N-acetylated. Glycosyl
2656
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4014-2656 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2001, 40, No. 14

COMMUNICATIONS
continued for 20 min. The organic phase was washed with 2n HCl solution
and dried over sodium sulfate. Removal of the volatiles and column
chromatographic purification (toluene/ethyl acetate 98:2) of the residue
reported: 14a: [a]2D5 ˆ ‡88.0 (c ˆ 0.3, CHCl3), ref. [7a]: [a]2D0 ˆ ‡89.9
(c ˆ 1, CHCl3), ref. [7d]: [a]2D2 ˆ ‡82.3 (c ˆ 1, CHCl3), ref. [7e]: [a]D
ˆ
‡87.5 (c ˆ 2, CHCl3), ref. [7f]: [a]2D3 ˆ ‡74.4 (c ˆ 1, CHCl3); 14b:
[a]2D5 ˆ ‡61.5 (c ˆ 0.2, CHCl3), ref. [7a]: [a]2D0 ˆ ‡65.0 (c ˆ 1.45,
furnished 2 as a light yellow oil (27 g, 84%). [a]D25
ˆ
7.5 (c ˆ 12, CHCl3).
1H NMR (600 MHz, CDCl3, 258C, TMS): d ˆ 7.77 (s, 1H; 1-H), 7.61 ± 7.60
CHCl3), ref. [7b]: [a]D20 ˆ ‡90 (c ˆ 0.4 ± 0.7, CHCl3), ref. [7c]: [a]2D5
ˆ
(m, 3H; arom. H), 7.41 ± 7.40 (m, 2H; arom. H), 7.34 ± 7.14 (m, 15H; arom.
‡64 (c ˆ 1.5, CHCl3), ref. [7d]: [a]2D2 ˆ ‡75.8 (c ˆ 1, CHCl3), Ref. [7e]:
[a]D ˆ ‡59 (c ˆ 0.5, CHCl3), ref. [7f]: [a]2D3 ˆ ‡72 (c ˆ 0.72, CHCl3),
ref. [7g]: [a]2D3 ˆ ‡63.7 (c ˆ 1.0, CHCl3), ref. [7 h]: [a]2D3 ˆ ‡90.6 (c ˆ
1.55, CHCl3).
3
2
H), 4.78 (d, J3,4 ˆ 3.5 Hz, 1H; 3-H), 4.75 (d, J ˆ 10.9 Hz, 1H; benzyl. H),
2
4.68 (d, J ˆ 10.9 Hz, 1H; benzyl. H), 4.59 ± 4.57 (m, 2H; 5-H, benzyl. H),
4.51 (d, 2J ˆ 12.0 Hz, benzyl. H), 4.22 ± 4.14 (m, 2H; 6-H, H6'), 3.81 (t,
3J4,3 ˆ 4.6, 3J4,5 ˆ 4.6 Hz, 1H; 4-H), 1.05 (s, 9H; C4H9); MS (FAB): m/z: 610
[9] S. Hanessian, P. Lavallee, Can. J. Chem. 1975, 53, 2975 ± 2979; S.
Hanessian, P. Lavallee, Can. J. Chem. 1977, 55, 562 ± 565.
[10] T. J. Martin, R. R. Schmidt, Tetrahedron Lett. 1992, 33, 6123 ± 6126;
T. J. Martin, R. Brescello, A. Toepfer, R. R. Schmidt, Glycoconjugate
J. 1993, 10, 16 ± 25.
‡
‡
[M‡H] , 632 [M‡Na] .
4b: Nitrogalactal 2 (13 g, 21.3 mmol) and 3b (7.1 g, 25.6 mmol) were dried
under high vacuum and dissolved in dry toluene (250 mL) under argon.
Then potassium tert-butoxide solution (2.1 mL of a 1m solution in THF)
was added and stirring continued for 120 min. Acetic acid (2 mL) was used
to acidify the reaction mixture and all solvents were removed under
reduced pressure. The residue was purified by column chromatography
(toluene/ethyl acetate 20:1) to furnish 4b as a colorless oil (18.3 g, 97%).
[a]2D5 ˆ ‡53.3 (c ˆ 5, CHCl3); 1H NMR (600 MHz, CDCl3, 258C, TMS):
d ˆ 7.61 ± 7.60 (m, 4H; arom. H), 7.39 ± 7.18 (m, 16H; arom. H), 5.32 (d,
[11] a) B. Liebe, H. Kunz, Angew. Chem. 1997, 109, 629 ± 631; Angew.
Chem. Int. Ed. Engl. 1997, 36, 618 ± 621; b) B. Liebe, H. Kunz, Helv.
Chim. Acta 1997, 80, 1473 ± 1482; c) J. B. Schwarz, S. D. Kuduk, X.-T.
Chen, D. Sames, P. W. Glunz, S. J. Danishefsky, J. Am. Chem. Soc.
1999, 121, 2662 ± 2673.
[12] 13b: [a]2D5 ˆ ‡35.5 (c ˆ 2.5, CHCl3), ref. [11b]: [a]2D2 ˆ ‡28.9 (c ˆ 1.0,
MeOH), ref. [11c]: [a]2D3 ˆ ‡36.7 (c ˆ 1.04, CHCl3).
3
3
3J1,2 ˆ 4.4 Hz, 1H; 1-H), 4.96 (d, JNH,a ˆ 9.7 Hz, 1H; NH), 4.93 (dd, J2,1
ˆ
[13] 15a: [a]2D5 ˆ ‡88.8 (c ˆ 1, CHCl3); 1H NMR (600 MHz, CDCl3): d ˆ
7.89 (d, J ˆ 7.5 Hz, 2H; arom. H), 7.71 (t, J ˆ 7.7 Hz, 2H; arom. H)
7.40 ± 7.30 (m, 4H; arom. H), 5.45 (d, 3J4,3 ˆ 3.2 Hz, 1H; 4a-H), 5.33 (d,
3J4,3 ˆ 2.6 Hz, 1H; 4b-H), 5.21 ± 5.17 (m, 2H; 3a-H, 3b-H) 4.88 (d,
3J1,2 ˆ 3.3 Hz, 1H; 1a-H), 4.77 (d, 3J1,2 ˆ 3.5 Hz, 1H; 1b-H), 4.55 ± 4.42
(m, 4H; b-H, b'-H, 2a-H), 4.35 (t, 3J5,6 ˆ 6.6, 3J5,6' ˆ 5.6 Hz, 1H; 5a-H),
4.1, 3J2,3 ˆ 10.6 Hz, 1H; 2-H), 4.83 (d, 2J ˆ 11.0 Hz, 1H; benzyl. H), 4.77 (d,
2J ˆ 11.0 Hz, 1H; benzyl. H), 4.50 (d, 2J ˆ 11.1 Hz, 1H; benzyl. H), 4.43 (dd,
3J3,2 ˆ 10.6, J3,4 ˆ 2.9 Hz, 1H; 3-H), 4.24 ± 4.23 (brd, 1H; b-H), 4.06 ± 4.05
3
3
3
(m, 2H; a-H, 4-H), 3.88 (brt, J5,6 ˆ 6.8, J5,6' ˆ 6.8 Hz, 1H; 5-H), 3.74 (dd,
3J6,5 ˆ 7.6, 2J6,6' ˆ 10.3 Hz, 1H; 6-H), 3.68 (dd, 3J6',5 ˆ 5.9, 2J6',6 ˆ 10.0 Hz, 1H;
6'-H), 1.49, 1.45 (2s, 18H; 2C4H9), 1.04 (s, 9H; C4H9). MS (FAB): m/z: 907
3
2
4.28 ± 4.23 (m, 3H; 5b-H, a-H, Fmoc-CH), 4.02 (dd, J6,5 ˆ 6.0, J6,6'
ˆ
‡
[M‡Na] .
3
11.2 Hz, 1H; 6b-H), 3.95(s, 2H; Fmoc-CH2), 3.83 (dd, J6',5 ˆ 7.0,
2J6',6 ˆ 11.1 Hz, 1H; 6b'-H), 3.77 (t, 3J6,6' ˆ 9.6 Hz, 1H; 6a-H), 3.34 (dd,
Received: January 8, 2001 [Z16382]
3J6',5 ˆ 5.0, J6',6 ˆ 9.9 Hz, 1H; 6a'-H), 2.14 ± 1.86 (7s, 21H; 5OAc,
2
2NHAc); 13C NMR (150.8 MHz, CDCl3): d ˆ 174.0 ± 172.0 (8C),
145.3 ± 121.0 (12C), 100.0 (1a-C), 99.1 (1b-C), 71.5 (Fmoc-CH2), 70.1
(3a-C), 69.8 (3b-C), 69.3 (4a-C), 68.8 (2-C, 5a-C, 4b-C), 68.0 (5b-C),
67.4 (6a-C), 63.0 (6b-C), 57.7 (Fmoc-CH), 49.0 (2a-C), 48.7 (2b-C), 48.4
[1] a) A. Varki, Glycobiology 1993, 3, 97 ± 130; b) I. Brockhausen,
Biochem. Biophys. Acta 1999, 1473, 67 ± 95.
[2] H. Paulsen, W. Stenzel, Chem. Ber. 1978, 111, 2334 ± 2347; H. Paulsen,
W. Stenzel, Chem. Ber. 1978, 111, 2348 ± 2357; H. Paulsen, C. Kolar, W.
Stenzel, Chem. Ber. 1978, 111, 2358 ± 2369; H. Paulsen, C. Kolar,
Chem. Ber. 1979, 112, 3190 ± 3202.
‡
(a-C), 22.8, 22.7, 20.7 (2C), 20.5; MS (FAB): m/z: 966 [M‡Na] , 988
‡
[M‡2Na H] .
[14] M. T. Bilodeau, T. K. Park, S. Hu, J. T. Randolph, S. J. Danishefsky,
P. O. Livingston, S. Zhang, J. Am. Chem. Soc. 1995, 117, 7840 ±
7841.
[15] N. Mathieux, H. Paulsen, M. Meldal, K. Bock, J. Chem. Soc. Perkin
Trans. 1 1997, 2359 ± 2368.
[3] a) B. Ferrari, A. A. Pavia, Carbohydr. Res. 1980, 79, C1 ± C7; b) H.
Paulsen, J.-P. Hölck, Carbohydr. Res. 1982, 109, 89 ± 107; c) G.
Grundler, R. R. Schmidt, Liebigs Ann. 1984, 1826 ± 1847; d) W. Kinzy,
R. R. Schmidt, Carbohydr. Res. 1987, 164, 265 ± 276; e) H. Paulsen, W.
Rauwald, U. Weichert, Liebigs Ann. 1988, 75 ± 86; f) W. Kinzy, R. R.
Schmidt, Carbohydr. Res. 1989, 193, 33 ± 47; g) Y. Nakahara, H. Iijima,
S. Sibayama, T. Ogawa, Tetrahedron Lett. 1990, 31, 6897 ± 6900;
h) X.-T. Chen, D. Sames, S. J. Danishefsky, J. Am. Chem. Soc. 1998,
120, 7760 ± 7769; i) M. Elofsson, L. A. Salvador, I. Kihlberg, Tetrahe-
dron 1997, 53, 369 ± 390; j) O. Seitz, ChemBioChem 2000, 1, 214 ± 246;
k) H. Herzner, T. Reipen, M. Schultz, H. Kunz, Chem. Ber. 2000, 100,
4495 ± 4537.
[16] 21b: The 1H NMR data are in excellent agreement with reported data
in ref. [7b]. Again differences in the optical rotation values were
found: [a]2D5 ˆ ‡25.7 (c ˆ 2, CHCl3), ref. [7b]: [a]2D1 ˆ ‡67 (c ˆ 0.4 ± 0.7,
CHCl3), ref. [7g]: [a]2D3 ˆ ‡69.2 (c ˆ 1, CHCl3).
[17] R. R. Schmidt, Angew. Chem. 1986, 98, 213 ± 236; Angew. Chem. Int.
Ed. Engl. 1986, 25, 212 ± 235; R. R. Schmidt, W. Kinzy, Adv.
Carbohydr. Chem. Biochem. 1994, 50, 21 ± 213.
[18] S. J. Danishefsky, M. T. Bilodeau, Angew. Chem. 1996, 108, 1482 ±
1522; Angew. Chem. Int. Ed. Engl. 1996, 35, 1380 ± 1419.
[19] H. Paulsen, Angew. Chem. 1990, 102, 851 ± 867; Angew. Chem. Int. Ed.
Engl. 1990, 29, 823 ± 839.
Á
[4] a) S. Bay, O. Berthie-Vergnes, V. Biberovic, D. Cantacuzene, Carbo-
Â
hydr. Res. 1997, 303, 25 ± 31; b) G. Dudziak, N. Bezay, T. Schwientek,
H. Clausen, H. Kunz, A. Liese, Tetrahedron, 2000, 56, 5865 ± 5869.
[5] G. A. Winterfeld, Y. Ito, T. Ogawa, R. R. Schmidt, Eur. J. Org. Chem.
1999, 1167 ± 1171.
[6] C. M. Timmers, S. C. M. Wigchert, M. A. Leeuwenburgh, G. A.
van der Marel, J. H. van Boom, Eur. J. Org. Chem. 1998, 91 ± 97.
[7] a) H. Paulsen, K. Adermann, Liebigs Ann. Chem. 1989, 739 ± 750;
b) B. Lüning, T. Norberg, I. Tejbrant, Glycoconjugate J. 1989, 6, 5 ± 19;
c) L. Biondi, F. Filira, M. Gobbo, B. Scolaro, R. Rocchi, F. Cavaggion,
Int. J. Pept. Protein Res. 1991, 37, 112 ± 121; d) M. Gommer, H. Kunz,
Â
Synlett 1991, 593 ± 595; e) L. Szabo, I. Ramza, C. Langdon, R. Polt,
Carbohydr. Res. 1995, 6, 5 ± 19; f) T. Vuljanic, K.-E. Bergquist, H.
Clausen, S. Roy, J. Kihlberg, Tetrahedron 1996, 52, 7983 ± 8000; g) M.
Lenck, H. Kunz, J. Prakt. Chem. 1997, 322 ± 334; h) S. D. Kuduk, J. B.
Schwarz, X.-T. Chen, P. W. Glunz, D. Sames, G. Ragupathi, P. O.
Livingston, S. I. Danishefsky, J. Am. Chem. Soc. 1998, 120, 12474 ±
12845.
[8] The NMR data of 14a, b are in excellent agreement with the reported
data in ref. [7]. However, varying optical rotation values were
Angew. Chem. Int. Ed. 2001, 40, No. 14
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4014-2657 $ 17.50+.50/0
2657