New efficient route for synthesis of lipid A by using affinity separation

Article abstract of DOI:10.1055/s-2001-18109

New efficient synthesis of lipid A, an immunostimulating glycoconjugate of bacteria, was achieved for the construction of lipid A library by using synthesis based on affinity separation (SAS), where the compounds possessing a barbituric acid (BA)-tag are selectively and rapidly purified by interaction with an artificial receptor for BA. Glycosylation of a glycosyl acceptor possessing the BA-tag with a 4' -phosphorylated N-Troc glucosaminyl trichloroacetimidate gave the disaccharide 4' -phosphate, which was purified by the affinity separation. Successive rémoval of protective groups and introduction of acyl groups were then effected and the synthetic intermediate at each step was purified by the affinity separation. Cleavage of the tag and subsequent deprotection afforded Escherichia coli lipid A. SAS enabled the rapid preparation of lipid A, therefore, proved to be a promising method for synthesis of other complex glycoconjugates.

Full text of DOI:10.1055/s-2001-18109

LETTER
1693
New Efficient Route for Synthesis of Lipid A by using Affinity Separation
New
E
fficient Rou
o
te for Synthe
s
sisof
L
ip
h
id
A
iyuki Fukase,* San-Qi Zhang, Keiko Iseki, Masato Oikawa, Koichi Fukase, Shoichi Kusumoto
Department of Chemistry, Graduate School of Science, Osaka University, Toyonaka, Osaka 560-0043, Japan
E-mail: koichi@chem.sci.osaka-u.ac.jp
Received 3 September 2001
OH
O
O
Abstract: New efficient synthesis of lipid A, an immunostimulat-
ing glycoconjugate of bacteria, was achieved for the construction of
lipid A library by using synthesis based on affinity separation
(SAS), where the compounds possessing a barbituric acid (BA)-tag
are selectively and rapidly purified by interaction with an artificial
receptor for BA. Glycosylation of a glycosyl acceptor possessing
the BA-tag with a 4’-phosphorylated N-Troc glucosaminyl trichlo-
roacetimidate gave the disaccharide 4’-phosphate, which was puri-
fied by the affinity separation. Successive removal of protective
groups and introduction of acyl groups were then effected and the
synthetic intermediate at each step was purified by the affinity sep-
aration. Cleavage of the tag and subsequent deprotection afforded
Escherichia coli lipid A. SAS enabled the rapid preparation of lipid
A, therefore, proved to be a promising method for synthesis of other
complex glycoconjugates.
O
O
(HO)2P O
O
HO
O
O
NH
NH
O
P(OH)2
O
O
O
O
O
OH
O
O
OH
O
(C14
)
(C14
)
(C14
)
(C14
)
(C12
)
(C14
)
Figure 1 Lipid A of Escherichia coli (1)
Key words: lipid A, high throughput synthesis, affinity separation,
barbituric acid, library
moieties were then introduced step by step to the respec-
tive positions.8 Novel lipid A analogues which possess
four (R)-3-hydroxyacyl moieties of shorter chain length
were synthesized via the synthetic route and the biological
tests clearly showed the critical importance of the chain
length of the acyl moieties in lipid A for the activity. In the
previous route, both amino groups were protected with
trichloroethoxycarbonyl (Troc) group and hence the same
acyl group was introduced to both amino groups. In the
present paper, we wish to report a new efficient synthesis
of lipid A by using a solution and solid-phase hybrid syn-
thesis via a novel divergent synthetic route, where differ-
ent acyl groups can be introduced to 2, 2’, 3, and 3’
positions.
Lipopolysaccharide (LPS), also termed endotoxin, is a
cell surface amphiphile characteristic of Gram-negative
bacteria. LPS induces various cytokines to exhibit potent
toxic activity such as pyrogenicity or lethality as well as
beneficial activities such as antitumor activity.1 LPS is
composed of a hydrophilic polysaccharide portion and a
hydrophobic lipid part. The lipid part was named lipid A;
its chemical structure and the full endotoxic responsibility
were established by our chemical synthesis of Escherichia
coli-type lipid A 1 (Figure 1).2,3 Lipid A of various bacte-
rial origin were shown to be structurally closely related
and to consist of: 1) (1 6) disaccharide of two D-glu-
cosamines, 2) phosphono groups at the reducing end and
the 4-position of the non-reducing sugar and 3) long-chain
acyl groups bound at 2, 2’, 3, and 3’ positions. Many
structural variations of lipid A, however, have been re-
ported mainly with regard to the acyl moieties: their types,
numbers, and locations on the hydrophilic sugar back-
bone. It has been shown that both phosphono groups at the
1- and 4’-positions and acyl groups are crucial for the bi-
ological activities of lipid A.4–7
The solid-phase strategy has greatly contributed to recent
development in combinatorial chemistry.9 Solid-phase
synthesis of complex glycoconjugates is, however, still
immature though various solid-phase oligosaccharide
synthesis have been reported.10 Recently we reported a
hybrid methodology of solid-phase synthesis and tradi-
tional solution synthesis: the reactions were carried out in
solutions and the desired products were isolated rapidly
by the host-guest interaction between the solid-support
and the desired products which possess the host and guest
structures, respectively.11 This methodology was termed
synthesis based on affinity separation (SAS). We have
used two interactions for SAS. One is interaction of a
crown ether and polymer-supported ammonium ion11a,b
and the other is the specific molecular recognition be-
tween a barbituric acid derivative and its artificial recep-
tor, which form tight complex with 6 hydrogen bonds11c
(Figure 2). Several peptides, heterocycles, and oligosac-
charides were synthesized by SAS. In the present study,
Structurally diverse lipid A analogues having various
types of acyl groups have been required for further inves-
tigation of the biological importance of acyl groups. We
therefore established a divergent synthetic route in the
previous study: a disaccharide 4’-phosphate was first con-
structed as a common synthetic intermediate and all acyl
Synlett 2001, No. 11, 26 10 2001. Article Identifier:
1437-2096,E;2001,0,11,1693,1698,ftx,en;Y17501ST.pdf.
© Georg Thieme Verlag Stuttgart · New York
ISSN 0936-5214

1694
Y. Fukase et al.
LETTER
we applied the latter method to the synthesis of a complex We first improved the synthetic procedure for the glycosyl
glycoconjugate lipid A.
donor 2 whose 2-amino and 3-hydroxy groups were dif-
ferentially protected with Troc and allyloxycarbonyl (Al-
loc) groups, respectively. In our previous study, where the
allyl group was used for temporary protection of the 1-po-
sition of the donor moiety,8 we encountered difficulty in
selective cleavage of the 1-O-allyl group in the presence
of the 3-O-Alloc group. In the attempt to isomerize the 1-
O-allyl group to the 1-O-1-propenyl group using an iridi-
um complex ([Ir(cod)(MePh2P)2]-PF6), the rate of the
isomerization in the presence of the Alloc group was
much lower than the corresponding reaction of a substrate
without the Alloc group. The reaction had to be stopped
before completion in order to suppress the decomposition
of the product. In the present study, the Alloc group was
introduced to the 3-position (Scheme 2) after isomeriza-
tion of the allyl group of 8. The product 9 was subjected
to the reductive benzylidene-ring opening as previously
The new high throughput synthetic route employed in this
work is outlined in Scheme 1. The barbituric acid (BA)
tag was attached to the 4-position of the glycosyl acceptor
via an acylamino benzyl linker, which can be cleaved by
DDQ oxidation.12 A (1 6) disaccharide 4’-phosphate 4
was constructed as a common key synthetic intermediate
by the coupling of two monosaccharides, i.e., a glycosyl
trichloroacetimidate 2 as a donor and a glycosyl acceptor
3 having the BA-tag. All acyl moieties were then intro-
duced step by step to the respective positions. After simul-
taneous cleavage of the linker and the BA-tag, the
glycosyl phosphate was formed by the cleavage of the al-
lyl glycoside and subsequent phosphorylation of the 1-hy-
droxy group. Removal of all the benzyl-type protecting
groups by catalytic hydrogenolysis afforded the desired
lipid A 1.
O
N
O
N
H
N
H
O
H
H
N
O
O
O
O
N
N
N
H
H
N
H
O
O
R
N
O
H
N
O
Figure 2 Host-guest interaction of a polymer-supported receptor with the barbituric acid tag
HO
BnO
NH
O
O
O
HN
+
OH
OAlloc
O
CCl3
O
P
O
O
O
O
O
O
O
OAllyl
O
HN
O
N
N
O
NHFmoc
NHTroc
H
H
2
3
BA
LINK
O
O
BnO
O
BnO
O
O
O
O
OAlloc
O
OH
OCOR''
OCOR
O
O
O
O
O
OAllyl
NHFmoc
O
OAllyl
O
O
P
P
NHCOR''
NHTroc
NHCOR'
LINK
BA
LINK
BA
4
5
O
O
BnO
O
BnO
O
O
OCOR''
O
OCOR''
O
O
O
P
OCOR
OCOR
OBn
OBn
1
O
O
O
O
O
HO
NHCOR'
OAllyl
NHCOR''
HO
NHCOR'
O
O
P
P
NHCOR''
6
7
Scheme 1 Strategy for the high throughput synthesis of lipid A by using the affinity separation
Synlett 2001, No. 11, 1693–1698 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
New Efficient Route for Synthesis of Lipid A
1695
reported to give the 6-O-benzylated 10 in a good yield The acceptor 20 tagged with BA moiety was prepared as
with high regioselectivity.13 Phosphitylation on the free 4- shown in Scheme 3 via the 4-O-p-azidobenzyl ether 17,
hydroxy group followed by oxidation14 gave the phos- which was prepared from a 4,6-O-p-nitrobenzylidene de-
phate 11 in 99% yield. The 1-propenyl group was smooth- rivative 14. We first attempted to introduce the p-ni-
ly deprotected with iodine and water to give the 1- trobenzyl group to the 6-position of N-Fmoc glucosamine
hydroxy product 12 in 94% yield. Compound 12 was then allyl glycoside 13 (Fmoc = 9-fluorenylmethoxycarbonyl)
converted as usual to the trichloroacetimidate 2 in a good by a new one-pot reductive benzylation using benzalde-
yield.
hydes, TMS2O, TMSOTf, and Et3SiH.8 Unexpectedly, the
p-nitrobenzylidenated product 14 was obtained in a good
yield by the reaction using p-nitrobenzaldehyde. Com-
pound 14 was not formed in the absence of Et3SiH. The 3-
hydroxy group in 14 was p-methoxybenzylated by using
p-methoxybenzyl (MPM) trichloroacetimidate to give 15.
Since nitrobenzylidene moiety is electron-deficient, re-
ductive opening of the nitrobenzylidene ring did not pro-
ceed under attempted several reaction conditions. We then
transferred the p-nitrobenzylidene 15 to p-azidoben-
zylidene 16 via reduction of the nitro group to an amino
group with Zn–Cu and acetylacetone followed by nucleo-
philic aromatic substitution of diazonium salts by an azido
group. The azidobenzylidene derivative 16 was also pre-
pared by direct p-azidobenzylidenation of 13 by the use of
p-azidobenzaldehyde, TMS2O, TMSOTf, and Et3SiH fol-
lowed by introduction of the MPM group. The p-azido-
benzylidene 16 was then subjected to reductive opening
with BH3 Me2NH and BF3 Et2O in CH2Cl2. The reaction
proceeded smoothly to form the desired 4-O-p-azidoben-
zylated glucosamine derivative 17, regioselectively. The
azido group of 17 was reduced by Zn and HOAc while the
resulting amino group was coupled with glutaric anhy-
O
O
O
O
c
a, b
Ph
Ph
OH
OAlloc
83%
98 %
O
OAllyl
NHTroc
O
O
NHTroc
8
9
BnO
BnO
O
d, e
O
OAlloc
O
OAlloc
O
99%
HO
O
P
O
O
O
NHTroc
NHTroc
10
11
BnO
O
O
f
g
OAlloc
2
OH
O
94%
98%
α/β = 3/1
P
O
O
NHTroc
12
Scheme
2
Synthesis
of
the
glycosyl
donor.
a)
[Ir(cod)(MePh2P)2]PF6, H2, THF; b) AllocCl, pyridine, CH2Cl2; c)
Me2NH BH3, BF3 Et2O, CH3CN; d) N,N-Diethyl-1,5-dihydro-3H-
2,4,3-benzodioxaphosphepin-3-amine, 1H-tetrazole, CH2Cl2; e)
mCPBA, -20 °C; f) I2, water, THF; g) CCl3CN, Cs2CO3, CH2Cl2
e, b
40%
HO
O
O
O
O
O
O
O
a
b
c, d
O2N
O2N
N3
OH
OH
OMPM
OMPM
96%
quant.
68%
HO
OAllyl
NHFmoc
O
OAllyl
NHFmoc
O
OAllyl
O
OAllyl
NHFmoc
NHFmoc
15
16
14
13
O
O
HN
O
HO
HO
O
O
O
g, h
84%
f
19
HN
OMPM
OMPM
O
NH2
O
O
71%
O
OAllyl
NHFmoc
O
OAllyl
i
N3
N
H
HO
NHFmoc
87%
18
17
HO
O
O
HN
HN
OR
O
O
O
O
O
OAllyl
O
N
N
H
O
NHFmoc
H
20 (R = MPM)
3 (R = H)
j
88%
Scheme 3 Synthesis of the glycosyl acceptor having the BA-tag. a) 4-Nitrobenzaldehyde, TMS2O, TMSOTf, Et3SiH, THF; b) 4-Methoxy-
phenylmethyl trichloroacetimidate, Sn(OTf)2, THF; c) Zn–Cu, acetylacetone, THF; d) NaNO2, HCl, NaN3, water, THF; e) 4-Azidobenzalde-
hyde, TMS2O, TMSOTf, Et3SiH, THF; f) Me2NH BH3, BF3 Et2O, CH2Cl2; g) Zn, HOAc; h) glutaric anhydride, CH2Cl2; i) DIC, CH2Cl2,
affinity separation; j) BF3 Et2O, CH2Cl2, affinity separation.
Synlett 2001, No. 11, 1693–1698 ISSN 0936-5214 © Thieme Stuttgart · New York

1696
Y. Fukase et al.
LETTER
dride. The product 18 was then coupled with BA-tag 19
by using diisopropylcarbodiimide (DIC). The desired BA-
tagged product 20 was purified by the affinity separation
as follows. The reaction mixture in CH2Cl2 was directly
applied to a resin column filled with CH2Cl2. The tagged
20 was retained in the column, whereas all untagged com-
pounds such as excess 18, DIC, and N,N’-diisopropylurea
were washed out from the column simply by elution with
CH2Cl2. Elution of 20 with CH2Cl2–methanol (1:1) and
concentration of the eluate gave purified 20, whose MPM
group was then removed with BF3 Et2O in CH2Cl2 to give
the glycosyl acceptor 3.15
O
BnO
O
O
O
OH
OAlloc
a
O
O
+
2
3
O
OAllyl
O
P
88%
NHFmoc
NHTroc
LINK
BA
4
O
BnO
O
O
O
OCOR''
b
OAlloc
O
O
O
OAllyl
NHFmoc
O
P
90%
NHTroc
LINK
BA
BA
BA
BA
21
Glycosylation reaction of the acceptor 3 with the imidate
2 was carried out in CH2Cl2 at –20 °C by the use of TM-
SOTf as a catalyst to give the desired -disaccharide in
88% yield after affinity separation (Scheme 4). Acyl
groups were then introduced to the disaccharide step by
step. Acylation of the 3-position of 4 with (R)-3-(4-triflu-
oromethylbenzyloxy)tetradecanoic acid16,17 proceeded
smoothly by using DIC and DMAP to give 21. Cleavage
of 2’-N-Troc group with Zn–Cu couple was followed by
N-acylation with (R)-3-(dodecanoyloxy)tetradecanoic
acid3b,16 and DIC. The third acyl group was introduced to
the 2-amino group after cleavage of the Fmoc group fol-
lowed by N-acylation using DIC. The 3’-O-Alloc group
was then removed with Pd(PPh3)4, HCO2H, and buty-
lamine. Final acylation to the 3’-position of 23 with (R)-
3-(tetradecanoyloxy)tetradecanoic acid3b,16 was carried
out by using DIC and DMAP to give fully acylated disac-
charide 24. After each reaction step, the product was rap-
idly separated from the reaction mixture by the affinity
separation. Though with less affinity than the BA-tag,
DMAP was unexpectedly also retained to the column of
BA-receptor so that the fractions of BA-tagged products
21 and 24 were inevitably contaminated with DMAP even
after affinity separation. Successive silica-gel short col-
umn chromatography was, therefore, used for the com-
plete removal of DMAP.
O
BnO
O
O
OCOR''
O
c, d
OAlloc
O
O
O
O
O
OAllyl
NHFmoc
O
88%
P
NHCOR'
LINK
22
O
BnO
O
O
OCOR''
O
e, f
OAlloc
O
O
84%
OAllyl
NHCOR''
O
P
NHCOR'
LINK
23
O
BnO
O
O
OCOR''
O
g, h
OCOR
O
O
OAllyl
NHCOR''
O
P
80%
NHCOR'
LINK
24
O
O
HN
O
=
O
O
BA
LINK
O
O
HN
N
H
N
H
RCO = (R)-3-(tetradecanoyloxy)tetradecanoyl
R'CO = (R)-3-(dodecanoyloxy)tetradecanoyl
R''CO = (R)-3-(4-trifluoromethylbenzyloxy)tetradecanoyl
The p-acylaminobenzyl linker was not smoothly cleaved
by conventional catalytic hydrogenolysis. The linker moi-
ety should hence be removed before 1-O-phosphoryla-
tion, since the glycosyl phosphate moiety is chemically
unstable and might be damaged during the cleavage of the
linker. We previously reported that p-acylaminobenzyl
ethers were selectively cleaved by DDQ oxidation in a
rate comparable to the cleavage of a MPM group.12 Treat-
ment of 4 with DDQ (1.2 equivalents) was thus examined
as a model reaction in CH2Cl2–H2O (10:1) at room tem-
perature for 33 hours. The desired 4-O-free product 25
was obtained only in 19% yield, whereas undesired 6’-O-
debenzylated compound 26 was formed in 26% yield by
over-oxidation of 25 (Table 1, entry 1). By contrast, treat-
ment of 4 with 10% TFA in CH2Cl2 afforded the desired
compound 25 in 59% yield after purification by prepara-
tive silica-gel TLC (entry 2).
Scheme 4 Glycosylation and acylation to form the basic structure of
lipid A. a) TMSOTf, MS4A, CH2Cl2; b) (R)-3-(4-trifluoromethylben-
zyloxy)tetradecanoic acid, DIC, DMAP, CH2Cl2, affinity separation
then silica-gel chromatography; c) Zn–Cu, HOAc; d) (R)-3-(dodeca-
noyloxy)tetradecanoic acid, DIC, CH2Cl2, affinity separation; e)
DBU, CH2Cl2; f) (R)-3-(4-trifluoromethylbenzyloxy)tetradecanoic
acid, DIC, CH2Cl2, affinity separation; g) Pd(PPh3)4, HCO2H, n-
BuNH2, THF; h) (R)-3-(tetradecanoyloxy)tetradecanoic acid, DIC,
DMAP, affinity separation then silica-gel chromatography.
follows. Byproducts possessing the BA moiety cannot be
removed by the affinity separation. All byproducts accu-
mulated during multistep of SAS were removed after
cleavage of the tag. In particular, a considerable amount
of the undesired N-2,2-dichloroethoxycarbonylated com-
pound was formed at the cleavage step of N-Troc group.
Compound 6 was purified thoroughly by silica-gel col-
umn chromatography and by gel-permeation chromatog-
raphy using recycling HPLC apparatus. The 1-O-allyl
Cleavage of the BA tag moiety in 24 was thus carried out
in 20% TFA in CHCl3 to give the desired 6 in 15% yield
(Scheme 5). The low yield of this step can be explained as
Synlett 2001, No. 11, 1693–1698 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
New Efficient Route for Synthesis of Lipid A
1697
Table 1 Cleavage of the p-Acylaminobenzyl Linker
O
BnO
O
O
O
RO
O
OH
OAlloc
O
O
O
O
O
O
OAllyl
O
P
O
O
OH
OAlloc
O
O
HN
HN
NHFmoc
NHTroc
O
HO
OAllyl
O
P
O
O
O
NHFmoc
NHTroc
O
N
H
N
H
25 (R = Bn)
26 (R = H)
4
Entry
Reagents
Time
Yield (%)
25
26
1
2
DDQ (1.2 equiv)
in CH2Cl2–MeOH (10:1)
33 h
33 h
19
26
10% TFA in CH2Cl2
59
N.D.
group of 6 was then removed as usual to yield 1-hydroxy In summary, we established a new efficient synthetic
disaccharide 27. Finally, the 1-O- -phosphorylation was route to lipid A by using the affinity separation protocol.
carried out via 1-O-lithiation and subsequent treatment Lipid A analogues can be synthesized within a few weeks
with tetrabenzyl diphosphate in THF at –78 °C.8,18 The by the SAS method. In our previous works, several
protected 1,4’-bisphosphate 7 thus obtained was purified months have been required to obtain the same lipid A by
by silica-gel column chromatography. Subsequently, all the conventional solution phase synthetic route. The
the protecting groups of 7 were removed by hydrogenoly- present study enabled us rapid and efficient construction
sis (10 kgcm–2 of H2) with Pd–black in one-step. The of lipid A libraries possessing various kinds of acyl moi-
product obtained in 75% yield was identical with E. coli eties for elucidation of their structure-activity relation-
lipid A 1 that we synthesized previously.3,7c
ships.
O
BnO
Acknowledgement
O
O
a
OCOR''
The present work was financially supported in part by „Research for
the Future“ Program No. 97L00502 from the Japan Society for the
Promotion of Science, by Special Coordination Funds, and Grants-
in-Aid for Scientific Research No. 09044086 and No. 12640571
from the Ministry of Education, Culture, Sports, Science and Tech-
nology, Japan.
OCOR
O
24
O
O
15%
HO
NHCOR'
OAllyl
NHCOR''
O
P
6
O
BnO
O
O
OCOR''
O
b, c
OCOR
OH
O
O
References and Notes
96%
HO
O
P
NHCOR''
NHCOR'
(1) (a) Rietschel, E. T.; Kirikae, T.; Schade, F. U.; Mamat, U.;
Schmidt, G.; Loppnow, H.; Ulmer, A. J.; Zähringer, U.;
Seydel, U.; Padova, F. D.; Schreier, M.; Brade, H. FASEB J.
1994, 8, 217. (b) Rietschel, E. T.; Brade, H. Sci. Am. 1992,
267 (August), 26.
(2) Westphal, O.; Lüderitz, O. Angew. Chem. 1954, 66, 407.
(3) (a) Imoto, M.; Yoshimura, H.; Yamamoto, M.; Sakaguchi,
N.; Kusumoto, S.; Shiba, T. Tetrahedron Lett. 1985, 26,
1545. (b) Imoto, M.; Yoshimura, H.; Shimamoto, T.;
Sakaguchi, N.; Kusumoto, S.; Shiba, T. Bull. Chem. Soc.
Jpn. 1987, 60, 2205.
27
O
BnO
O
O
O
d
OCOR''
O
P
OCOR
OBn
OBn
O
O
90%
HO
NHCOR'
O
O
P
NHCOR''
7
e
(4) Rietschel, E. T.; Kirikae, T.; Schade, F. U.; Ulmer, A. J.;
Holst, O.; Brade, H.; Schmidt, G.; Mamat, U.; Grimmecke,
H.-D.; Kusumoto, S.; Zäringer, U. Immunobiol. 1993, 187,
169.
1
75%
negative ESI/TOF-MS
[m/z 1797.12 [M-H]−]
(5) Holst, O. Angew. Chem., Int. Ed. Engl. 1995, 34, 2000.
(6) Rietschel, E. T.; Brade, H.; Holst, O.; Brade, L.; Müller-
Loennies, S.; Mamat, U.; Zähringer, U.; Beckmann, F.;
Seydel, U.; Brandenburg, K.; Ulmer, A. J.; Mattern, T.;
Heine, H.; Schletter, J.; Loppnow, H.; Schönbeck, U.; Flad,
Scheme 5 Removal of the BA-tag and completion of the synthesis.
a) 20% TFA in CHCl3; b) [Ir(cod)(MePh2P)2]PF6, H2, THF; c) I2, wa-
ter, THF; d) tetrabenzyl diphosphate, LiN(TMS)2, THF; e) H2 (10
kgcm–2), Pd–black, THF.
Synlett 2001, No. 11, 1693–1698 ISSN 0936-5214 © Thieme Stuttgart · New York

1698
Y. Fukase et al.
LETTER
(11) (a) Zhang, S.-Q.; Fukase, K.; Kusumoto, S. Tetrahedron
H.-D.; Hauschildt, S.; Schade, U. F.; Padova, F. D.;
Kusumoto, S.; Schumann, R. R. Current Topics in
Microbiology and Immunology, Vol. 216, Pathology of
Septic Shock; Rietschel, E. T.; Wagner, H., Eds.; Springer-
Verlag: Berlin Heidelberg, 1996, Chap. Bacterial Endotoxin,
Chemical Constitution, Biological Recognition, Host
Response, and Immunological Detoxification, 39.
Lett. 1999, 40, 7479. (b) Zhang, S.-Q.; Fukase, K.;
Kusumoto, S. Peptide Science 1999; N. Fujii: The Japanese
Peptide Society, 2000, 151. (c) Zhang, S.-Q.; Fukase, K.;
Izumi, M.; Fukase, Y.; Kusumoto, S. Synlett 2001, 590.
(12) Acylaminobenzyl linkages are stable under Lewis acidic
conditions for glycosylation or cleavage of a MPM group but
can be cleaved by DDQ oxidation: (a) Fukase, K.; Nakai,
Y.; Egusa, K.; Porco, J. A.; Kusumoto, S. Synlett 1999,
1074. (b) Fukase, K.; Egusa, K.; Nakai, Y.; Kusumoto, S.
Mol. Divers. 1997, 2, 182. (c) Fukase, K.; Tanaka, H.; Torii,
S.; Kusumoto, S. Tetrahedron Lett. 1990, 31, 389.
(13) Oikawa, M.; Liu, W.-C.; Nakai, Y.; Koshida, S.; Fukase, K.;
Kusumoto, S. Synlett 1996, 1179.
(7) Since replacement of the glycosyl phosphate by other acidic
functional groups did not have significant effect on the
biological activity, the anionic structure rather than
phosphono groups at the 1- and 4’-positions in lipid A should
be important for the biological activity: (a) Ulmer, A. J.;
Heine, H.; Feist, W.; Kusumoto, S.; Kusama, T.; Brade, H.;
Schade, U.; Rietschel, E. T.; Flad, H.-D. Infect. Immun.
1992, 60, 3309. (b) Kusama, T.; Soga, T.; Ono, Y.;
Kumazawa, E.; Shioya, E.; Nakayama, K.; Uoto, K.; Osada,
Y. Chem. Pharm. Bull. 1991, 39, 3244. (c) Liu, W.-C.;
Oikawa, M.; Fukase, K.; Suda, Y.; Kusumoto, S. Bull.
Chem. Soc. Jpn. 1999, 72, 1377.
(14) Watanabe, Y.; Komada, Y.; Ebisuya, K.; Ozaki, S.
Tetrahedron Lett. 1990, 31, 255.
(15) Glycosylation of the glycosyl acceptor 20 with glycosyl
trichloroacetimidate 2 gave lower yield than the
glycosylation of 3 with 2.
(8) Fukase, K.; Fukase, Y.; Oikawa, M.; Liu, W.-C.; Suda, Y.;
Kusumoto, S. Tetrahedron 1998, 54, 4033.
(16) (a) Oikawa, M.; Kusumoto, S. Tetrahedron: Asymmetry
1995, 6, 961. (b) Noyori, R.; Ohkuma, T.; Kitamura, M.;
Takaya, H.; Sayo, N.; Kumobayashi, H.; Akutagawa, S. J.
Am. Chem. Soc. 1987, 109, 5856.
(17) Sakai, Y.; Oikawa, M.; Yoshizaki, H.; Ogawa, T.; Suda, Y.;
Fukase, K.; Kusumoto, S. Tetrahedron Lett. 2000, 41, 6843.
(18) Inage, M.; Chaki, H.; Kusumoto, S.; Shiba, T. Chem. Lett.
1982, 1281.
(9) (a) Thompson, L. A.; Ellman, J. A. Chem. Rev. 1996, 96,
555. (b) Dolle, R. E.; Nelson, K. H. Jr. J. Comb. Chem. 1999,
1, 235. (c) Dolle, R. E. J. Comb. Chem. 2000, 2, 383.
(10) For recent review, see: Seeberger, P. H.; Haase, W.-C.
Chem. Rev. 2000, 100, 4349.
There is an erratum or addendum to this paper.
Please check www.thieme.de/chemistry/synthesis/index.html.
Synlett 2001, No. 11, 1693–1698 ISSN 0936-5214 © Thieme Stuttgart · New York