Innate immune responses of synthetic lipid a derivatives of Neisseria meningitidis

Article abstract of DOI:10.1002/chem.200701165

Differences in the pattern and chemical nature of fatty acids of lipid A of Neisseria meningitides lipooligosaccharides (LOS) and Escherichia coli lipopolysaccharides (LPS) may account for differences in inflammatory properties. Furthermore, there are indications that dimeric 3-deoxy-D-mannooct- 2-ulosonic acid (KDO) moieties of LOS and LPS enhance biological activities. Heterogeneity in the structure of lipid A and possible contaminations with other inflammatory components have made it difficult to confirm these observations. To address these problems, a highly convergent approach for the synthesis of a lipid A derivative containing KDO has been developed. which relies on the ability to selectively remove or unmask in a sequential manner an isopropylidene acetal, 9-fluorenylmethoxycarbonyl (Fmoc), allyloxycarbonate (Alloc), azide, and thexyldimethylsilyl (TDS) ether. The strategy was employed for the synthesis of N. meningitidis lipid A containing KDO (3). Mouse macrophages were exposed to the synthetic compound and its parent LOS, E. coli lipid A (2), and a hybrid derivative (4) that has the asymmetrical acylation pattern of E. coli lipid A, but the shorter lipids of meningococcal lipid A. The resulting supernatants were examined for tumor necrosis factor alpha (TNF-α) and interferon beta (IFN-β) production. The lipid A derivative containing KDO was much more active than lipid A alone and just slightly less active than its parent LOS, indicating that one KDO moiety is sufficient for full activity of TNF-α and IFN-β induction. The lipid A of N. meningitidis was a significantly more potent inducer of TNF-α and IFN-β than E. coli lipid A, which is due to a number of shorter fatty acids. The compounds did not demonstrate a bias towards a MyD88- or TRIF-dependent response.

Full text of DOI:10.1002/chem.200701165

DOI: 10.1002/chem.200701165
Innate Immune Responses of Synthetic Lipid A Derivatives of
Neisseria meningitidis
Yanghui Zhang, Jidnyasa Gaekwad, Margreet A. Wolfert, and Geert-Jan Boons*[a]
Abstract: Differences in the pattern
and chemical nature of fatty acids of
lipid A of Neisseria meningitides lipoo-
ligosaccharides (LOS) and Escherichia
coli lipopolysaccharides (LPS) may ac-
count for differences in inflammatory
properties. Furthermore, there are indi-
cations that dimeric 3-deoxy-d-manno-
oct-2-ulosonic acid (KDO) moieties of
LOS and LPS enhance biological activ-
ities. Heterogeneity in the structure of
lipid A and possible contaminations
with other inflammatory components
have made it difficult to confirm these
observations. To address these prob-
lems, a highly convergent approach for
the synthesis of a lipid A derivative
containing KDOhas been developed,
which relies on the ability to selectively
remove or unmask in sequential
manner an isopropylidene acetal, 9-flu-
orenylmethoxycarbonyl (Fmoc), allyl-
coli lipid A, but the shorter lipids of
meningococcal lipid A. The resulting
supernatants were examined for tumor
necrosis factor alpha (TNF-a) and in-
terferon beta (IFN-b) production. The
lipid A derivative containing KDOwas
much more active than lipid A alone
and just slightly less active than its
parent LOS, indicating that one KDO
moiety is sufficient for full activity of
TNF-a and IFN-b induction. The lipid
A of N. meningitidis was a significantly
more potent inducer of TNF-a and
IFN-b than E. coli lipid A, which is
due to a number of shorter fatty acids.
The compounds did not demonstrate a
bias towards a MyD88- or TRIF-de-
pendent response.
a
A
yldimethylsilyl (TDS) ether. The strat-
egy was employed for the synthesis of
N. meningitidis lipid
A containing
KDO( 3). Mouse macrophages were
exposed to the synthetic compound
and its parent LOS, E. coli lipid A (2),
and a hybrid derivative (4) that has the
asymmetrical acylation pattern of E.
Keywords:
glycosylation
oligosaccharides
carbohydrates
· immunology
·
·
Introduction
Toll-like receptor 4 (TLR4), a member of the TLR family
that participates in pathogen recognition.[5,6] Immediately
distal to TLR4 activation are two intracellular cascades that
regulate signal transduction processes, gene expression, and
production of (pro)inflammatory mediators.[5,6] One of these
cascades requires a specific intracellular adaptor protein
called MyD88, while the other cascade utilizes the TRIF
adaptor protein. The MyD88-dependent pathway leads to
up-regulation of cytokines and chemokines such as TNF-a,
IL-1b, IL-6, and monocyte chemoattractant protein 1 (MCP-
1), whereas the TRIF-dependent pathway leads to the pro-
duction of IFN-b, which in turn activates the STAT-1 path-
way resulting in the production of mediators such as inter-
feron-inducible protein 10 (IP-10) and nitric oxide.[7] Al-
though the initiation of acute inflammatory responses is im-
portant for the prevention of infections, over-activation
leads to the clinical symptoms of septic shock.
Neisseria meningitidis is a Gram-negative bacterium that
causes fulminant, rapidly fatal sepsis, and meningitis.[1] The
morbidity and mortality of meningococcal bacteremia has
been linked to a systematic inflammatory response to lipo-
oligosaccharides (LOS) of affected patients.[2–4] LOS, a
major component of the outer membrane of N. meningitidis,
initiates the production of multiple host-derived inflamma-
tory mediators such as tumor necrosis factor alpha (TNF-a),
interferon beta (IFN-b), interleukin 1 (IL-1), IL-6, arachi-
donic acid metabolites, and leukotrienes. The production of
these mediators is initiated by the interaction of LOS with
[a] Dr. Y. Zhang, J. Gaekwad, Dr. M. A. Wolfert, Prof. G.-J. Boons
Complex Carbohydrate Research Center
The University of Georgia, 315 Riverbend Road
Athens, GA 30602 (USA)
Meningococcal LOS lacks the repeating O-antigen of en-
teric lipopolysaccharides (LPS) but has a conserved inner-
core region composed of l-glycero-d-manno-heptosides and
Fax : (+1)706-542-4412
E-mail: gjboons@ccrc.uga.edu
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
a
dimeric 3-deoxy-d-manno-oct-2-ulosonic acid (KDO)
558
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 558 – 569

FULL PAPER
moiety linked to lipid A.[8,9] Furthermore, the lipid A of N.
meningitidis is hexa-acylated in a symmetrical fashion (1)
whereas enteric bacteria have an asymmetrically hexa-acy-
lated lipid A (2) (Figure 1).[10–13] Also, a number of the fatty
alone stimulated naïve allogeneic CD4+ cells to secrete en-
hanced levels of IFN-g relative to T-cells primed with imma-
ture dendritic cells.[18]
Several other studies have suggested that the KDO
moiety of LPS or LOS contributes to inflammatory respons-
es. For example, it has been found that salmonella lipid A is
inactive whereas the parent LPS is a potent activator of
NF-kB in a TLR4-dependent manner in a human monocytic
cell line.[19] In addition, a synthetic enteric lipid A containing
a di-KDOmoiety was a more potent elicitor of TNF- a and
IL-6 compared with its parent lipid A.[20] Furthermore, LPS
from a nitrogen-fixing symbiont, Rhizobium sin-1 is able to
significantly inhibit the E. coli LPS-dependent synthesis of
TNF-a by human monocytic cells.[21–25] A comparison of the
biological responses of synthetic and isolated lipid A deriva-
tives and R. sin-1 LPS indicated that the KDOmoieties are
important for optimal antagonistic properties. Thus, it is
probable that the cell surface receptors that recognize LPS
bind to the lipid A as well as to the KDOmoiety of LPS.
Although studies with LOS isolated from meningococci
have indicated that it possesses unique immunological prop-
erties, heterogeneity in the structure of lipid A and possible
contaminations with other inflammatory components of the
bacterial cell wall have made it difficult to confirm these re-
sults.[14,17,18] Furthermore, the acylation pattern as well as
fatty acid length differ between meningococcal and E. coli
lipid A. Hence, it has been impossible to establish which
structural feature is responsible for the unique inflammatory
properties. Fortunately, homogeneous lipid A derivatives
can be obtained by chemical synthesis,[13,16] an approach
which makes it also possible to prepare panels of different
compounds to establish which structural features are respon-
sible for particular biological properties.
We report here the preparation of a prototypical lipid A
derived from N. meningitidis (1) and a similar derivative
containing a KDOmoiety ( 3). Proinflammatory properties
of these compounds have been determined in a mouse mac-
rophage cell line and the results compared with similar data
for lipid A 2, which is derived from E. coli, and compound
4, which has the asymmetrical acylation pattern of E. coli
but fatty acids that are similar in length to those of N. meni-
gitidis lipid A. It has been found that the lipid A of N. men-
ingitidis (1) is a more potent inducer of TNF-a and IFN-b
than E. coli lipid A (2). The greater potency was attributed
to the shorter fatty acids of 1 and not to its symmetrical acy-
lation pattern. Furthermore, the KDOmoiety of 3 signifi-
cantly enhanced the potency of proinflammatory responses.
Figure 1. Chemical structures of target lipid As.
acids of N. meningitidis are shorter compared with those of
Escherichia coli. It has been suggested that the differences
in the nature and pattern of fatty acid substitution of enteric
and meningococcal lipid A affects their biological proper-
ties. For example, it has been shown that at low concentra-
tions meningococcal LOS is a potent inducer of MyD88-
and TRIF-dependent cytokines, whereas at the same pmole
concentrations E. coli LPS induced comparable levels of
TNF-a, IL-1b, and MIP-3a but significantly less IFN-b,
nitric oxide, and IP-10.[14]
It has long been thought that the inflammatory properties
of LPS and LOS reside in the lipid A moiety.[13,15,16] Howev-
er, recent studies have shown that lipid A expressed by me-
ningococci with defects in KDObiosynthesis or transfer has
significantly reduced bioactivities compared to KDO2 con-
taining meningococcal LOS.[17] Removal of the KDOmoiet-
ies of wild type LOS by mild acetic acid treatment also atte-
nuated cellular responses. Interestingly, dendritic cells stimu-
lated with KDO2-lipid A from meningococci but not lipid A
Results and Discussion
Chemical synthesis: To determine biological properties of N.
meningitidis lipid A and LOS, we have synthesized com-
pounds 1 and 3 (Figure 1) by a highly convergent approach.
Compound 1 is a prototypical lipid A derived from N. men-
ingitidis LOS, and is hexasubstituted in a symmetrical fash-
ion. Compound 3 has a structure similar to 1, except that
Chem. Eur. J. 2008, 14, 558 – 569
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
559

G.-J. Boons et al.
the C-6’ moiety of its lipid A is extended by a KDOmoiety.
It was envisaged that compound 3 could be prepared from
monosaccharides 5 and 6, KDOdonor 7,[26] and fatty acids 8
and 9[27] (Scheme 1). Thus, coupling of 5 with 6 will give or-
thogonally protected disaccharide 15, which will be subject-
ed to mild acidic conditions to remove the isopropylidene
without affecting any of the other protecting groups. The C-
6’ hydroxyl group of the resulting compound can then be re-
gioselectively glycosylated with 7 followed by phosphoryla-
tion of the C-4’ hydroxyl. Next, the orthogonal protecting
groups 9-fluorenylmethoxycarbonyl (Fmoc), allyloxycarbon-
ate (Alloc), and azido will be individually removed, which
allows acylation with any lipid at C-2, C-2’, and C-3 to give
easy access to a panel compounds differing in acylation pat-
tern. At an early stage of the synthesis, the C-3’ hydroxyl of
15 was acylated with (R)-3-benzyloxy-dodecanoic acid be-
cause it was observed that the C-4’-phosphate triester can
migrate during acylation of the C-3’ hydroxyl. After comple-
tion of the acylations, the anomeric thexyldimethylsilyl
(TDS) group can be selectively cleaved and phosphorylation
of the resulting lactol followed by global deprotection
should provide target compound 3.
Glycosyl acceptor 5 and donor 6 could be prepared from
known derivatives 10[28] and 12,[29] respectively (Scheme 1).
Thus, the C-3 hydroxyl of 10 was protected by an Alloc
group by treatment with Alloc chloride in the presence of
N,N,N’,N’-tetramethylenediamine (TMEDA)[30] in CH2Cl2 to
give 11 in a yield of 96%. Regioselective reductive opening
of the benzylidene acetal of 11 proved more difficult than
anticipated. Conventional procedures such as treatment with
BH3·THF/Bu2BOTf or BH3·THF/TMSOTf[31] resulted in a
loss of the Alloc group. Fortunately, the reaction of 11 with
triethyl silane and PhBCl2 in the presence of molecular
sieves at ꢀ758C[32] gave 5 in an excellent yield of 94% as
only the C-6 hydroxyl.
The C-3 hydroxyl group of 12 was acylated with (R)-3-
benzyloxy-dodecanoic acid (8) using 1,3-dicyclohexylcarbo-
diimide (DCC) and 4-dimethylaminopyridine (DMAP) as
the activation reagents to give 13 in an excellent yield of
95%. Next, the azido function of 13 was reduced with zinc
in a mixture of acetic acid and CH2Cl2. The resulting amine
was immediately protected as an Fmoc carbamate by reac-
tion with FmocCl in the presence of diisopropylethylamine
(DIPEA) to give fully protected 14. Removal of the anome-
ric TDS ether of 14 was achieved by treatment with tetrabu-
tylammonium fluoride (TBAF) buffered with acetic acid in
THF followed by conversion of the resulting lactol into tri-
chloroacetimidate 6 (a/b ꢁ1:1) by reaction with trichloro-
AHCTREUNG
presence of molecular sieves (4) Š) in CH2Cl2 at ꢀ408C
gave disaccharide 15 in a modest yield of 42%. Surprisingly,
only the b-anomer of trichloroacetimidate 6 had been con-
sumed while the a-anomer remained intact, even when the
temperature or amount of TMSOTf was increased. An at-
tempt to prepare selectively the b-anomer of 6 using Cs2CO3
as the base resulted unexpectedly in the formation of the a-
anomer as the predominant product. Fortunately, the use of
trifluoromethanesulfonic acid (TfOH) instead of TMSOTf
could drive the glycosylation to completion with consump-
tion of both the a- and b-anomer and provided disaccharide
15 in an excellent yield of 94%. Next, the isopropylidene
group of 15 was removed with trifluoroacetic acid (TFA) in
wet CH2Cl2 to yield 16 in an almost quantitative yield.
Glycosylation of 16 with KDOdonor 7 was carried out
with the aid of BF3·OEt2 in the presence of MS 4 Š in
CH2Cl2 to afford an inseparable mixture of 17 and its b-
anomer (a/b 9:1) in a combined yield of 67% (Scheme 2).
The anomeric configuration of the KDOglycosides was es-
tablished by comparing the difference between the chemical
shift values of the C-3 methyl-
ene protons.[35] In this respect,
a
larger difference between
the chemical shift values of C-
methylene protons is ob-
3
served for a-anomers of KDO
glycosides that reside in a boat
conformation, as compared to
similar values for the corre-
sponding b-anomers. In the
case of KDOglycosides that
adopt a chair form, a large
chemical shift difference is ob-
served for b-anomers whereas
the a-anomer provides a small
difference.[36] After careful ex-
amination of the NMR spectra
of the glycosylation products,
it was found that the chemical
shift difference of the C-3
methylene protons of the
major product was 0.59 ppm
Scheme 1. a) AllocCl, TMEDA, CH2Cl2; b) PhBCl2, Et3SiH, MS 4 Š, CH2Cl2, ꢀ758C; c) 8, DCC, DMAP,
CH2Cl2; d) Zn/HOAc, CH2Cl2; then FmocCl, DIPEA, CH2Cl2; e) Bu4NF, AcOH, THF; then CCl3CN, Cs2CO3,
CH2Cl2.
560
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 558 – 569

Synthetic Lipid A Derivatives
FULL PAPER
C-3 methylene protons of the
major product had decreased
to 0.08 ppm while the corre-
sponding difference increased
to 0.39 ppm for the minor
product. Furthermore, the cou-
pling constants of the C-3
methylene protons with H-4
established that the glycosides
adopt
a chair conformation.
Thus, these data provide con-
vincing evidence that the
major product has an a-
anomeric configuration.
Phosphilation of 17 with
N,N-diethyl-1,5-dihydro-2,3,4-
benzodioxaphosphepin-3-
amine in the presence of 1H-
tetrazole followed by in situ
oxidation with m-chloroper-
oxybenzoic acid (mCPBA)[37]
gave 18. Fortunately, at this
stage of the synthesis the a-
and b-anomer could be sepa-
rated by silica gel column chro-
matography. Interestingly, the
chemical shift difference of the
methylene protons of com-
pound 18 and that of its b-
anomer were both very small,
an observation that may be
due to the deshielding effects
of the C-3 methylene protons
by the aromatic group of the
phosphate diester. As a matter
of fact, it has been noted that
the empirical rule described
above does not apply to all
subsequent KDOderivatives
protected as 4,5-isopropylidene
acetals. To provide additional
information, the isopropyli-
dene acetal of 18 and its b-
Scheme 2. a) TfOH, CH2Cl2, ꢀ508C; b) TFA, H2O , CHCl2; c) 7, BF3·Et2O , CHCl2, MS 4 Š, 08C; d) N,N-di-
2
2
ethyl-1,5-dihydro-3H-2,3,4-benzodioxaphosphepin-3-amine, 1H-tetrazole, CH2Cl2; then mCPBA, ꢀ208C;
e) DBU, CH2Cl2; then 9, DCC, CH2Cl2; f) [Pd
A
CH2Cl2; g) Zn/HOAc, CH2Cl2; then 9, DCC, CH2Cl2; h) TFA, H2O , CHCl2; then tetrabenzyl diphosphate,
anomer were removed. NMR
data showed that the chemical
2
LiN
A
shift difference of the C-3
while the corresponding difference was 0.12 ppm for the
minor product. Furthermore, the KDOglycosides adopted
boat conformations due to protection of the C-4,5 diol as an
isopropylidene acetal, which was confirmed by the coupling
methylene protons was 0.10 while the corresponding differ-
ence increased to 0.37 for the a-anomer, an observation that
provided an additional piece of evidence to convincingly
show that the major product is an a-anomer, because when
the isopropylidene acetal was removed, the KDOexisted as
a chair form. The chair form was demonstrated by the large
coupling constant (12.0 Hz) between H-4 and H-3ax.
Having the advanced trisaccharide 18 in hand, attention
was focused on the selective acylation of relevant hydroxyls
and amines. Thus, the Fmoc protecting group of 18 was re-
constants of the C-3 methylene protons with H-4 (JH3a,H4
=
J
H3e,H4 =4.8 Hz). Thus, the data imply that the major product
has an a-anomeric configuration. To provide additional evi-
dence of the assigned anomeric configurations, the isopropy-
lidene acetal of glycosylation product 16 was removed by
treatment with TFA in CH2Cl2. NMR data of the resulting
compound showed that the chemical shift difference of the
moved using 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) in
A
Chem. Eur. J. 2008, 14, 558 – 569
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
561

G.-J. Boons et al.
CH2Cl2 and the resulting amino group acylated with (R)-3-
dodecanoyl-tetradecanoic acid (9) using DCC as the activat-
ing agent to give compound 19. Removal of the Alloc group
phate in the presence of lithium bis(trimethyl)silylamide in
THF at ꢀ788C[39] to give 27. Global deprotection of 27 by
catalytic hydrogenolysis gave the requisite lipid A 1.
of 19 was easily accomplished by treatment with [Pd(PPh3)4]
G
in the presence of BuNH2 and HCOOH,[38] and subsequent
acylation of the resulting hydroxyl with 8 using DCC and
DMAP as activating agent afforded 20. Next, reduction of
the azido function of 20 with zinc in a mixture of acetic acid
and CH2Cl2 followed by acylation of the resulting amine
with 9 using standard conditions furnished fully acylated 21.
Next, the isopropylidene acetal and anomeric TDS of 21
Biological evaluation: Based on the results of recent stud-
ies[5,6] it is clear that LPS-induced cellular activation through
TLR4 is complex as many signaling elements are involved.
However, it appears that there are two distinct initiation
points in the signaling process, one being a specific intracel-
lular adaptor protein called MyD88 and the other an adap-
tor protein called TRIF, which operates independently of
MyD88. It is well established that TNF-a secretion is a pro-
totypical measure for activation of the MyD88-dependent
pathway, whereas secretion of IFN-b is commonly used as
an indicator of TRIF-dependent cellular activation.
There are some indications that structurally different lipid
A derivatives can differentially utilize signal transduction
pathways leading to complex patterns of proinflammatory
responses. For example, it has been suggested that meningo-
coccal LOS is a potent inducer of MyD88- and TRIF-depen-
dent cytokines, whereas at the same pmol concentrations E.
coli LPS induced comparable levels of MyD88 derived cyto-
kines but significantly less TRIF-associated cytokines.[14,17]
In addition, there are indications that the KDOmoieties of
meningococcal LOS are required for optimal biological
properties.[18]
were removed by treatment with TFA/H2O3:2 in CH Cl2
2
and the anomeric hydroxyl of the resulting compound was
regioselectively phosphorylated using tetrabenzyl diphos-
phate in the presence of lithium bis(trimethyl)silylamide in
THF at ꢀ788C[39] to give anomeric phosphate 22 as only the
a-anomer. Finally, global deprotection of 22 could easily be
accomplished by catalytic hydrogenolysis over Pd-black to
give target product 3.
Lipid A derivative 1 could easily be prepared by starting
from previously reported[40] compound 23 (Scheme 3). Thus,
the azido function of 23 was reduced with activated Zn in a
mixture of acetic acid and CH2Cl2 and the amine of the re-
sulting compound was reacted with (R)-2-dodecanoyloxyte-
tradecanoic acid (9) in the presence of DCC to give 24. The
removal of the Alloc protecting group of 24 could easily be
accomplished by treatment with [Pd
A
To address these issues, we have examined the well-de-
fined compounds 1–4 and E. coli LPS for the ability to ini-
tiate production of TNF-a and IFN-b. Compounds 1 and
2[40] are prototypical lipid As derived from N. meningitidis
LOS and E. coli LPS, respectively. Both compounds are
hexa-acylated but differ in the nature and substitution pat-
tern of fatty acids. Compound 3 has a structure similar to 1,
yl group of the resulting compound 25 was acylated with
(R)-2-benzyloxytetradecanoic acid (8) using DCC and
DMAP as activating agents to afford fully acylated 26. The
anomeric tert-butyldimethylsilyl (TBS) ether of 26 was re-
moved by treatment with HF in pyridine and the resulting
anomeric hydroxyl phosphorylated using tetrabenzyl diphos-
Scheme 3. a) Zn/HOAc, CH2Cl2; then 9, DCC, CH2Cl2; b) [Pd
ACHTREUNG
benzyl diphosphate, LiN
ACHTREUNG
562
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 558 – 569

Synthetic Lipid A Derivatives
FULL PAPER
except that the C-6’ moiety of its lipid A is extended by a
KDOmoiety. Compound 4[40] is a hybrid derivative that has
the asymmetrical acylation pattern of E. coli lipid A but the
shorter lipids of meningococcal lipid.
Mouse macrophages (RAW 264.7 gNO(ꢀ) cells) were ex-
posed over a wide range of concentrations to compounds 1–
4 and meningococcal LOS. After 5.5 h, the supernatants
were harvested and examined for mouse TNF-a and IFN-b
using a commercial and in-house developed capture ELISA,
respectively. Potencies (EC50, concentration producing 50%
activity) and efficacies (maximal level of production) were
determined by fitting the dose-response curves to a logistic
equation using PRISM software.
As can be seen in Figure 2, synthetic N. meningitidis lipid
A (1) is a significantly more potent inducer of TNF-a and
IFN-b than E. coli lipid A (2). Furthermore, the difference
in EC50 values of the hybrid derivative 4 and N. meningitidis
lipid A (1) is very small (Table 1). Thus, the data suggest
that the shorter fatty acids of N. meningitidis lipid A are pri-
marily responsible for its greater biological activity.
The EC50 values of the KDOcontaining derivative 3 and
meningococcal LOS are very similar whereas these values
are significantly smaller than those of lipid A 1. These re-
sults demonstrate the importance of the core region of LOS
for biological activity and that one KDOmoiety is sufficient
for full activity of TNF-a and IFN-b induction by LOS.
Finally, a comparison of the EC50 values of TNF-a and
IFN-b for each compound indicated that the values for
TNF-a are slightly smaller than those of IFN-b (2–6 fold),
indicating a somewhat higher potency for TNF-a produc-
tion. No significant differences were observed between effi-
cacies of secreted TNF-a and IFN-b. Thus, it appears that
for the compounds tested there is no clear bias towards a
MyD88- or TRIF-dependent response. It may be possible
that previously observed bias may be due to contaminants
or, alternatively, due to minor lipid A components having
unique acylation substitutions or patterns.
Conclusion
A convergent approach for the synthesis of a lipid A deriva-
tive containing KDOhas been developed, which allows for
the convenient synthesis of a panel of analogues differing in
fatty acid acylation patterns and degree of phosphorylation.
The new synthetic approach relies on the ability to selective-
ly remove or unmask in a sequential manner an isopropyli-
dene acetal, a Fmoc carbamate, an Alloc carbonate, azide,
and TDS ether. The strategy was employed for the synthesis
of N. meningitidis lipid A containing KDO( 3). The com-
pound was tested for cytokine production along with the
synthetic N. meningitidis lipid A (1), its parent LOS, E. coli
lipid A (2), and a hybrid derivative (4) that has the asym-
metrical acylation pattern of E. coli lipid A but the shorter
lipids of meningococcal lipid A. Examination of potencies
and efficacies of TNF-a and IFN-b production showed that
the lipid A derivative containing KDOwas much more
active than lipid A alone and just slightly less active than its
parent LOS, indicating that one KDO moiety is sufficient
for full activity of TNF-a and IFN-b production. It still
needs to be established whether the increase in activity is
due to specific interactions with relevant cell surface recep-
tors or due to alterations in pharmacokinetic properties. It
has also been found that the lipid A of N. meningitidis is a
significantly more potent inducer of TNF-a and IFN-b than
E. coli lipid A, which is attributed to a number of shorter
fatty acids. For each compound, the values for TNF-a were
only slightly smaller than those for IFN-b whereas no signif-
icant differences were observed between the efficacies.
Thus, the compounds tested do not demonstrate a clear bias
towards a MyD88- or TRIF-dependent response.
Table 1. EC50 values[a] [nm] of N. meningitidis LOS and lipid A deriva-
tives 1–4.
TNF-a
(0.058–0.13)
(1.3–1.9)
(16–28)
(0.074–0.12)
(2.5–6.7)
IFN-b
(0.085–0.15)
(3.6–4.5)
(105–147)
(0.17–0.49)
(12–23)
N. meningitidis LOS
lipid A 1
lipid A 2
lipid A 3
lipid A 4
0.085
1.5
21
0.092
4.1
0.11
4.0
124
0.20
16
G
ACHTREUNG
G
ACHTREUNG
U
ACHTREUNG
[a] Values of EC50 are reported as best-fit values and as minimum-maxi-
mum range (best-fit value ꢂ std. error).
Figure 2. TNF-a and IFN-b production by murine macrophages after
stimulation with LOS and lipid A derivatives. Murine RAW gNO(ꢀ)
cells were incubated for 5.5 h with increasing concentrations of N. menin-
gitidis LOS or lipid A derivatives 1–4 as indicated. a) TNF-a and b) IFN-
b in cell supernatants were measured using ELISAs.
Experimental Section
General synthetic methods: Column chromatography was performed on
silica gel 60 (EM Science, 70–230 mesh). Reactions were monitored by
Chem. Eur. J. 2008, 14, 558 – 569
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
563

G.-J. Boons et al.
thin-layer chromatography (TLC) on Kieselgel 60 F254 (EM Science),
and compounds were detected by examination under UV light and by
charring with 10% sulfuric acid in MeOH. Solvents were removed under
reduced pressure at < 408C. CH2Cl2 was distilled from NaH and stored
over molecular sieves (3) Š). Tetrahydrofuran (THF) was distilled from
sodium directly prior to application. MeOH was dried by refluxing with
magnesium methoxide and then was distilled and stored under argon.
Pyridine was dried by heating under refluxing over CaH2 and then dis-
tilled and stored over molecular sieves (3) Š). Molecular sieves (3 and
4 Š) used for reactions, were crushed and activated in vacuo at 3908C
during 8 h and then for 2–3 h at 3908C directly prior to application. Opti-
cal rotations were measured using a Jasco model P-1020 polarimeter.
1H NMR and 13C NMR spectra were recorded on Varian spectrometers
(models Inova500 and Inova600) equipped with Sun workstations.
1H NMR spectra were recorded in CDCl3 and referenced to residual
CHCl3 at 7.24 ppm, and 13C NMR spectra were referenced to the central
peak of CDCl3 at 77.0 ppm. Assignments were made by standard gCOSY
and gHSQC. High resolution mass spectra were obtained on a Bruker
model Ultraflex MALDI-TOF mass spectrometer. Signals marked with a
subscript L belong to the biantennary lipids, whereas signals marked with
a subscript L’ belong to their side chain. Signals marked with a subscript
S symbol belong to the monoantennary lipids.
9.3 Hz, H-3), 4.70–2.59 (m, 5H, H-1, CH2Ph, OCH2CH=CH2), 3.84 (dd,
1H,
J5,6a =2.4, J6a,6b =11.7 Hz, H-6a), 3.71 (dd, 1H, J5,6b =4.2, J6a,6b =
11.7 Hz, H-6b), 3.65 (t, 1H, J3,4 =J4,5 =9.6 Hz, H-4), 3.42–3.37 (m, 1H, H-
5), 3.34 (dd, 1H, J1,2 =7.2, J2,3 =10.5 Hz, H-2), 1.71–1.62 [m, 1H, CH-
A
G
N
0.17 ppm (s, 3H, SiCH3); 13C NMR (75 MHz, CDCl3): d = 154.24 (C=
O), 137.26–127.89 (m, aromatic), 131.17 (OCH2CH=CH2), 119.13
(OCH2CH=CH2), 96.81 (C-1), 78.52 (C-3), 75.29 (C-4), 74.99 (C-5), 74.66
(OCH2CH=CH2), 68.82 (CH2Ph), 66.49 (C-2), 61.49 (C-6), 33.73 [SiC-
A
G
G
N
A
G
z: calcd for C25H39N3O7SiNa: 544.2455; found: 544.2548 [M+Na]+.
Dimethylthexylsilyl 2-azido-3-O-[(R)-3-benzyloxy-dodecanoyl]-2-deoxy-
4,6-O-isopropylidene-b-d-glucopyranoside (13): A reaction mixture of
(R)-3-benzyloxy-dodecanoic acid (8; 970 mg, 3.17 mmol) and DCC
(949 mg, 4.60 mmol) in CH2Cl2 (10 mL) was stirred at room temperature
for 10 min, and then compound 12 (1.15 g, 2.88 mmol) and DMAP
(35 mg, 0.29 mmol) were added. The reaction mixture was stirred at
room temperature for 10 h, after which the solids were removed by filtra-
tion, and the residue washed with CH2Cl2 (2”1 mL). The combined fil-
trates were concentrated in vacuo, and the residue was purified by silica
gel column chromatography (hexane/ethyl acetate 15:1) to yield 13 as a
syrup (1.82 g, 94%). Rf =0.55 (hexane/ethyl acetate 8:1); [a]2D5 =ꢀ10.08
Synthesis of KDOdonor 7 and compound 1 and 1H and 13C NMR spec-
tra are given in the Supporting Information.
1
(c=1.0, CHCl3); H NMR (300 MHz, CDCl3): d = 7.34–7.26 (m, 5H, ar-
omatic), 4.91 (t, 1H, J2,3 =J3,4 =9.6 Hz, H-3), 4.62 (d, 1H, J1,2 =7.5 Hz,
H-1), 4.60 (d, 1H, J=11.4 Hz, CHHPh), 4.48 (d, 1H, J=11.4 Hz,
Dimethylthexylsilyl
deoxy-b-d-glucopyranoside
3-O-allyloxycarbonyl-2-azido-4,6-O-benzylidene-2-
(11): Allyl chloroformate (366 mL,
CHHPh), 3.89–3.83 (m, 2H, H-6a, H-3S), 3.74 (t, 1H,
J5,6b =J6a,6b =
3.44 mmol) was added dropwise to a cooled (08C) solution of compound
10 (1.25 g, 2.87 mmol) and N,N,N’,N’-tetramethylethylenediamine
(TMEDA) (281 mL, 1.87 mmol) in CH2Cl2 (10 mL). The reaction mixture
was stirred at room temperature for 3 h, and then diluted with CH2Cl2
(20 mL) and washed with saturated aqueous NaHCO3 (2”20 mL) and
brine (2”20 mL). The organic phase was dried (MgSO4), filtered, and
the filtrate was concentrated in vacuo. The residue was purified by silica
gel column chromatography (hexane/ethyl acetate 30:1) to give 11 as a
10.5 Hz, H-6b), 3.64 (t, 1H, J3,4 =J4,5 =9.6 Hz, H-4), 3.34–3.23 (m, 2H, H-
2, H-5), 2.71 (dd, 1H, J2Sa,2Sb =15.0, J2Sa,3S =6.3 Hz, H-2Sa), 2.51 (dd, 1H,
J
2Sa,2Sb =15.0, J2Sb,3S =6.0 Hz, H-2Sb), 1.69–1.47 [m, 3H, H-4S, CH
1.38 (s, 3H, CH3 of isopropylidene), 1.28 (s, 3H, CH3 of isopropylidene),
1.24 [brs, 14H, H-(5S-11S)], 0.89–0.84 [m, 15H, H-12S, SiC(CH3)2CH-
(CH3)2], 0.19 (s, 3H, SiCH3), 0.18 ppm (s, 3H, SiCH3); 13C NMR
(75 MHz, CDCl3): d = 170.56 (C=O), 138.50–127.50 (m, aromatic), 99.67
[C(CH3)2 of isopropylidene], 97.43 (C-1), 75.70 (C-3S), 71.63 (C-4), 71.46,
71.41 (C-3, CH2Ph), 67.46, 67.37 (C-2, C-5), 61.93 (C-6), 39.77 (C-2S),
34.50–14.07 [m, SiC(CH3)2CH(CH3)2, CH3 of isopropylidene, C-(4S-12S)],
A
ACHTREUNG
AHCTREUNG
AHCTREUNG
colorless oil (1.43 g, 96%). Rf =0.60 (hexane/ethyl acetate 5:1); [a]D25
=
ꢀ34.98 (c=1.0, CHCl3); 1H NMR (500 MHz, CDCl3): d = 7.42–7.32 (m,
5H, aromatic), 5.98–5.86 (m, 1H, OCH2CH=CH2), 5.47 (s, 1H, >CHPh),
5.33 (d, 1H, J=17.0 Hz, OCH2CH=CHH), 5.22 (d, 1H, J=11.0 Hz,
A
ACHTREUNG
ꢀ2.25 (SiCH3), ꢀ3.32 ppm (SiCH3); HR MS: m/z: calcd for
C36H61N3O7SiNa: 698.4176; found: 698.3518 [M+Na]+.
OCH2CH=CHH), 4.87 (t, 1H, J2,3 =J3,4 =10.0 Hz, H-3), 4.69 (d, 1H, J1,2
7.5 Hz, H-1), 4.64 (d, 2H, J=6.0 Hz, OCH2CH=CH2), 4.28 (dd, 1H,
5,6a =5.0, J6a,6b =10.0 Hz, H-6a), 3.76 (dd, 1H, J5,6b =J6a,6b =10.5 Hz, H-
6b), 3.67 (d, 1H, J3,4 =J4,5 =9.0 Hz, H-4), 3.48–3.40 (m, 2H, H-2, H-5),
1.68–1.63 [m, 1H, CH(CH3)2], 0.89–0.87 [m, 12H, SiC(CH3)2CH(CH3)2],
0.19 (s, 3H, SiCH3), 0.18 ppm (s, 3H, SiCH3); 13C NMR (75 MHz,
CDCl3): d 154.15 (C=O), 136.75–126.17 (m, aromatic, OCH2CH=
=
Dimethylthexylsilyl 3-O-[(R)-3-benzyloxy-dodecanoyl]-2-deoxy-2-(9-fluo-
renylmethoxycarbonyl)-4,6-O-isopropylidene-b-d-glucopyranoside (14):
A suspension of compound 13 (1.82 g, 2.70 mmol) and zinc (<10 micron,
1.75 g, 27.0 mmol) in a mixture of acetic acid (300 mL) and CH2Cl2
(15 mL) was stirred at room temperature for 5 h, after which it was dilut-
ed with ethyl acetate (40 mL). The solids were removed by filtration, and
the residue was washed with ethyl acetate (2”3 mL). The combined fil-
trates were washed with saturated aqueous NaHCO3 (2”30 mL) and
brine (2”20 mL). The organic phase was dried (MgSO4), filtered, and
the filtrate was concentrated in vacuo to afford the crude amine as a pale
yellow oil. The resulting amine was dissolved in CH2Cl2 (15 mL), and
then FmocCl (767 mg, 2.97 mmol) and DIPEA (517 mL, 2.97 mmol) were
added. The reaction mixture was stirred at room temperature for 2 h,
after which it was diluted with CH2Cl2 (20 mL) and washed with brine
(2”30 mL). The organic phase was dried (MgSO4) and concentrated in
vacuo. The residue was purified by silica gel column chromatography
(hexane/ethyl acetate 10:1) to yield 14 as a colorless syrup (2.02 g, 86%,
two steps). Rf =0.55 (hexane/ethyl acetate 5:1); [a]2D5 =ꢀ2.88 (c=1.0,
J
A
N
ACHTREUNG
=
CH2), 119.07 (OCH2CH=CH2), 101.54 (>CHPh), 97.56 (C-1), 78.57 (C-
4), 75.35 (C-3), 68.93 (OCH2CH=CH2), 68.57 (C-6), 67.12 (C-2), 66.32
(C-5), 33.82 [SiC
A
(CH3)2], 24.78 [SiC
G
N
19.76, 18.46, 18.36 [SiC
(SiCH3); HR MS: m/z: calcd for C25H37N3O7SiNa: 542.2298; found:
542.2475 [M+Na]+.
A
G
Dimethylthexylsilyl 3-O-allyloxycarbonyl-2-azido-4-O-benzyl-2-deoxy-b-
d-glucopyranoside (5): A suspension of 11 (1.20 g, 2.31 mmol) and molec-
ular sieve (4) Š, 300 mg) in CH2Cl2 (20 mL) was stirred at room tempera-
ture for 1 h. The mixture was cooled (ꢀ758C) and then triethylsilane
(0.55 mL, 3.47 mmol) and PhBCl2 (0.52 mL, 3.93 mmol) were added
dropwise. After stirring the reaction mixture for 1 h, it was quenched by
addition of Et3N (1 mL) and methanol (1 mL). The reaction mixture was
warmed up to room temperature and then diluted with ethyl acetate
(40 mL). The molecular sieve was removed by filtration, and the filtrate
was washed with saturated aqueous NaHCO3 (30 mL). The organic phase
was dried (MgSO4), filtered, and the filtrate was concentrated in vacuo.
The residue was purified by silica gel column chromatography (hexane/
ethyl acetate 10:1) to give 5 as a colorless oil (1.13 g, 94%). Rf =0.35
(hexane/ethyl acetate 5:1); [a]2D5 =ꢀ17.18 (c=1.0, CHCl3); 1H NMR
(300 MHz, CDCl3): d = 7.28–7.22 (m, 5H, aromatic), 5.99–5.85 (m, 1H,
OCH2CH=CH2), 5.37 (dd, 1H, J=1.2, 15.6 Hz, OCH2CH=CHH), 5.27
1
CHCl3); H NMR (300 MHz, CDCl3): d = 7.73–7.25 (m, 13H, aromatic),
5.35 (d, 1H, JNH,2 =9.0 Hz, NH), 5.27 (t, 1H, J2,3 =J3,4 =9.6 Hz, H-3), 4.81
(d, 1H, J1,2 =7.2 Hz, H-1), 4.55 (d, 1H, J=11.1 Hz, CHHPh), 4.43 (d,
1H, J=11.1 Hz, CHHPh), 4.27–4.13 (m, 3H, OCH2CH of Fmoc), 3.89–
3.69 (m, 5H, H-2, H-4, H-6a, H-6b, H-3S), 3.47 (brs, 1H,
H-5), 2.70 (dd, 1H, J2Sa,2Sb =14.7, J2Sa,3S =5.1 Hz, H-2Sa), 2.46 (dd, 1H,
J
2Sa,2Sb =14.7, J2Sb,3S =6.0 Hz, H-2Sb), 1.59–1.50 (m, 3H, H-4S, CH(CH3)),
A
1.43 (s, 3H, CH3 of isopropylidene), 1.34 (s, 3H, CH3 of isopropylidene),
1.23–1.16 [brs, 14H, H-(5S-11S)], 1.16–0.84 [m, 15H, H-12S, SiC-
A
N
(dd, 1H, J=1.2, 11.0 Hz, OCH2CH=CHH), 4.80 (dd, 1H, J2,3 =10.5, J3,4
=
ACHTREUNG
564
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 558 – 569

Synthetic Lipid A Derivatives
FULL PAPER
75.66 (C-3S), 72.08 (C-3), 71.83 (C-4), 71.15 (CH2Ph), 67.07 (C-5, OCH2
of Fmoc), 62.00 (C-6), 58.55 (C-2), 46.92 (OCH2CH of Fmoc), 39.76 (C-
and brine (2”15 mL). The organic phase was dried (MgSO4), filtered,
and the filtrate was concentrated in vacuo. The residue was purified by
silica gel column chromatography (hexane/ethyl acetate 2:1) to afford 16
as a pale yellow oil (882 mg, 98%). Rf =0.35 (hexane/ethyl acetate 1:1);
2S), 34.41–14.04 [m, SiC
N
N
12S)], ꢀ2.00 (SiCH3), ꢀ3.42 ppm (SiCH3); HR MS: m/z: calcd for
C51H73NO9SiNa: 894.4952; found: 894.4984 [M+Na]+.
[a]2D4 =ꢀ9.28 (c=1.0, CHCl3); 1H NMR (300 MHz, CD3COCD3): d
=
7.84–7.18 (m, 18H, aromatic), 6.64 (d, 1H, JNH’,2 =9.3 Hz, NH’), 5.92–5.79
(m, 1H, OCH2CH=CH2), 5.29 (dd, J=1.5, 14.4 Hz, OCH2CH=CHH),
Dimethylthexylsilyl 3-O-[(R)-3-benzyloxy-dodecanoyl]-2-deoxy-2-(9-fluo-
renylmethoxycarbonyl)-4,6-O-isopropylidene-b-d-glucopyranosyl-(1!6)-
3-O-allyloxycarbonyl-2-azido-4-O-benzyl-2-deoxy-b-d-glucopyranoside
(15): A mixture of Bu4NF (1m in THF, 5 mL) and acetic acid (800 mL)
was added dropwise to a stirred solution of 14 (1.30 g, 1.49 mmol) in
THF (15 mL). After stirring at room temperature for 36 h, the reaction
mixture was diluted with CH2Cl2 (20 mL), and then washed with saturat-
ed aqueous NaHCO3 (2”25 mL) and brine (2”25 mL). The organic
phase was dried (MgSO4), filtered, and the filtrate was concentrated in
vacuo. The residue was purified by silica gel column chromatography
(hexane/ethyl acetate 5:2) to afford a lactol as a pale yellow oil (978 mg,
90%).
5.22–5.14 (m, 2H, H-3’, O CHCH=CHH), 4.85–4.79 (m, 3H, H-1, H-1’,
2
H-3), 4.74 (d, 1H, J=11.4 Hz, CHHPh), 4.59–4.52 (m, 4H, CH2Ph,
OCH2CH=CH2), 4.42 (d, 1H, J=11.7 Hz, CHHPh), 4.21–4.10 (m, 4H,
H-6a, OCH2CH of Fmoc), 3.93–3.84 (m, 2H, H-6b, H-6’a), 3.81–3.62 (m,
6H, H-2’, H-4, H-4’, H-5’, H-6’b, H-3S), 3.45–3.39 (m, 2H, H-2, H-5), 2.62
(dd, 1H, J2Sa,2Sb =15.6, J2Sa,3S =6.3 Hz, H-2Sa), 2.41 (dd, 1H, J2Sa,2Sb =15.6,
J
2Sb,3S =5.4 Hz, H-2Sb), 1.73–1.64 [m, 1H, CH
H-4S), 1.28–1.09 [m, 14H, H-(5S-11S)], 0.96–0.83 [m, 15H, H-12S, SiC-
(CH3)2CH(CH3)2], 0.23 (s, 3H, SiCH3), 0.22 ppm (s, 3H, SiCH3);
13C NMR (75 MHz, CD3COCD3): d
171.80 (C=O), 156.48 (C=O),
G
A
ACHTREUNG
=
154.86 (C=O), 144.81–120.51 (m, OCH2CH=CH2, aromatic), 118.51
(OCH2CH=CH2), 102.01 (C-1’), 97.02 (C-1), 79.21 (C-3), 71.11 (C-5),
76.72, 76.56, 76.07 (C-3’, 4, 3S), 74.93 (CH2Ph), 74.66 (C-5’), 71.37
(CH2Ph), 69.66 (C-4’), 68.86 (OCH2CH=CH2), 68.46 (C-6), 67.20 (C-2),
67.02 (OCH2 of Fmoc), 62.40 (C-6’), 56.79 (C-2’), 47.59 (OCH2CH of
The resulting lactol (810 mg, 1.11 mmol) was dissolved in a mixture of tri-
chloroacetonitrile (2.0 mL) and CH2Cl2 (6 mL), and then Cs2CO3
(181 mg, 0.55 mmol) was added. The reaction mixture was stirred at
room temperature for 1 h, after which it was diluted with CH2Cl2
(20 mL), and then washed with saturated aqueous NaHCO3 (2”25 mL)
and brine (2”25 mL). The organic phase was dried (Na2SO4), filtered,
and the filtrate was concentrated in vacuo. The residue was purified by
silica gel column chromatography (hexane/ethyl acetate 2:1) to yield 6 as
a pale yellow foam (880 mg, 91%). A suspension of trichloroacetimidate
6 (880 mg, 1.01 mmol), acceptor 5 (480 mg, 0.92 mmol) and molecular
sieves (4) Š, 500 mg) in CH2Cl2 (10 mL) was stirred at room temperature
for 1 h. The mixture was cooled (ꢀ508C) and then TfOH (4.4 mL,
0.05 mmol) was added. After stirring the reaction mixture for 30 min, it
was allowed to warm up to ꢀ108C in 30 min and then quenched with
solid NaHCO3 (50 mg) and diluted with CH2Cl2 (20 mL). The solution
was washed with saturated aqueous NaHCO3 (2”25 mL) and brine (2”
25 mL). The organic phase was dried (MgSO4), filtered, and the filtrate
was concentrated in vacuo. The residue was purified by silica gel column
chromatography (hexane/ethyl acetate 7/1!4/1) to yield disaccharide 15
as a colorless solid (1.07 g, 94%). Rf =0.50 (hexane/ethyl acetate 4:1);
[a]2D4 =ꢀ9.68 (c=1.0, CHCl3); 1H NMR (600 MHz, CDCl3): d = 7.82–
7.19 (m, 18H, aromatic), 6.76 (d, 1H, JNH’,2 =9.6 Hz, NH’), 5.88–5.83 (m,
Fmoc), 39.98 (C-2S), 34.83–14.09 [m, SiC
N
N
ꢀ1.83 (SiCH3), ꢀ3.34 ppm (SiCH3); HR MS: m/z: calcd for
C65H88N4O15SiNa: 1215.5913; found: 1215.6797 [M+Na]+.
Dimethylthexylsilyl benzyl (7,8-di-O-benzyl-3-deoxy-4,5-O-isopropyli-
dene-a/b-d-manno-oct-2-ulopyranosyl)onate-(2!6)-3-O-[(R)-3-benzyl-
A
nosyl-(1!6)-3-O-allyloxycarbonyl-2-azido-4-O-benzyl-2-deoxy-b-d-gluco-
pyranoside (17): A suspension of 16 (610 mg, 0.51 mmol), 7 (495 mg,
0.91 mmol) and molecular sieves (4) Š, 400 mg) in CH2Cl2 (8 mL) was
stirred at room temperature for 1 h. The mixture was cooled (08C) and
then BF3·Et2O(77 mL, 0.61 mmol) was added dropwise. After stirring
the reaction mixture for 1 h, it was quenched with solid NaHCO3
(100 mg) and then diluted with CH2Cl2 (20 mL). The solids were re-
moved by filtration, and the filtrate was washed with saturated aqueous
NaHCO3 (2”25 mL) and brine (2”25 mL). The organic phase was dried
(MgSO4), filtered, and the filtrate was concentrated in vacuo. The residue
was purified by silica gel column chromatography (hexane/ethyl acetate
6:1!4:1) to afford 17 as an amorphous solid (590 mg, 67%). Rf =0.45
(hexane/ethyl acetate 3:1); [a]2D4 =ꢀ4.98 (c=1.0, CHCl3); 1H NMR of a-
anomer (600 MHz, CD3COCD3): d = 7.84–7.18 (m, 33H, aromatic), 6.62
(d, 1H, JNH’,2 =9.0 Hz, NH’), 5.89–5.82 (m, 1H, OCH2CH=CH2), 5.30–23
(m, 2H, CO2CHHPh, OCH2CH=CHH), 5.20–5.12 (m, 3H, H-3’,
CO2CHHPh, OCH2CH=CHH), 4.84–4.49 (m, 12H, H-1, H-1’, H-3, 7”
CHH2Ph, OCH2CH=CH2), 4.43–4.39 (m, 3H, H-4’’, H-5’’, CHHPh), 4.23–
4.18 (m, 1H, OCHH of Fmoc), 4.17–4.10 (m, 3H, H-6a, OCHHCH of
Fmoc), 4.00–3.91 (m, 3H, H-6’’, H-7’’, H-8a’’), 3.84 (dd, 1H, J5’,6a’ =5.4,
1H, OCH CH=CH2), 5.29 (d, J=16.8 Hz, OCH2CH=CHH), 5.23 (t, 1H,
2
J
2,3 =J3,4 =9.6 Hz, H-3’), 5.16 (d, J=10.2 Hz, OCH2CH=CHH), 4.88 (d,
1H, J1’,2’ =8.4 Hz, H-1’), 4.80–4.69 (m, 2H, H-1, H-3), 4.69 (d, 1H, J=
10.8 Hz, CHHPh), 4.60–4.52 (m, 4H, CH2Ph, OCH2CH=CH2), 4.39 (d,
1H, J=11.4 Hz, CHHPh), 4.24–4.21 (m, 1H, OCHH of Fmoc), 4.12 (d,
1H, J6a,6b =10.8 Hz, H-6a), 4.08–4.01 (m, 2H, OCHHCH of Fmoc), 3.89–
3.84 (m, 2H, H-6’a, H-6b), 3.82–3.79 (m, 2H, H-4’, H-6’b), 3.78–3.72 (m,
3H, H-2’, H-4, H-3S), 3.67–3.65 (m, 1H, H-5), 3.40–3.33 (m, 2H, H-2, H-
5’), 2.57 (dd, 1H, J2Sa,2Sb =15.6, J2Sa,3S =6.0 Hz, H-2Sa), 2.33 (dd, 1H,
J
2Sa,2Sb =15.0, J2Sb,3S =6.0 Hz, H-2Sb), 1.71–1.66 [m, 1H, CH
1.39 (m, 2H, H-4S), 1.39–1.11 [m, 20H, CH3 of isopropylidene, H-(5S-
11S)], 0.91–0.83 [m, 15H, H-12S, SiC(CH3)2CH(CH3)2], 0.24 (s, 3H,
SiCH3), 0.23 ppm (s, 3H, SiCH3); 13C NMR (75 MHz, CDCl3): d
171.30 (C=O), 156.58 (C=O), 154.91 (C=O), 144.85–118.58 (m,
OCH2CH=CH2, aromatic), 102.42 (C-1’), 99.91 [C(CH3)2 of isopropyli-
G
J
6a’,6b’ =10.8 Hz, H-6a’), 3.78–3.77 (m, 2H, H-3S, H-8b’’), 3.73–3.70 (m,
3H, H-4, H-5, H-6b), 3.67–3.65 (m, 2H, H-2’, H-6b’), 3.48 (brs, 2H, H-4’,
H-5’), 3.81 (dd, 1H, J1,2 =8.4, J2,3 =9.6 Hz, H-2), 2.62–2.54 (m, 2H, H-3a’’,
H-2Sa), 2.39 (dd, 1H, J2Sa,2Sb =15.6, J2Sb,3S =5.4 Hz, H-2Sb), 1.97 (dd, 1H,
U
ACHTREUNG
J
3a’’,3b’’ =15.0, J3b’’,4’’ =2.4 Hz, H-3b’’), 1.68–1.62 [m, 1H, CH(CH3)2], 1.43–
A
AHCTREUNG
1.39 (m, 2H, H-4S), 1.31 (s, 3H, CH3 of isopropylidene), 1.27 (s, 3H, CH3
dene], 97.08 (C-1), 75.66 (C-3S), 79.23 (C-3), 76.67, 75.97 (C-4, 3S), 75.05
(CH2Ph), 74.51 (C-5), 72.96 (C-3’), 73.46 (C-4’), 71.29 (CH2Ph), 68.93
(OCH2CH=CH2), 68.51 (C-6), 67.86, 67.25 (C-2, 5’), 67.08 (OCH2 of
Fmoc), 62.40 (C-6’), 57.27 (C-2’), 47.58 (OCH2CH of Fmoc), 40.00 (C-2S),
of isopropylidene), 1.16–1.08 [m, 14H, H-(5S-11S)], 0.88–0.83 [m, 15H, H-
12S, SiC
13C NMR (75 MHz, CD3COCD3): d
156.46 (C=O), 154.85 (C=O), 144.81–120.48 (m, OCH2CH=CH2, aromat-
ic), 118.59 (OCH2CH=CH2), 109.11 (C-2’’), 102.02 (C-1’), 98.34 [C(CH3)2
A
ACHTREUNG
=
34.88–14.14 [m, SiC
A
G
ACHTREUNG
ꢀ1.81 (SiCH3), ꢀ3.26 ppm (SiCH3); HR MS: m/z: calcd for
of isopropylidene], 97.00 (C-1), 79.20 (C-3), 77.87 (C-7’’), 76.99 (C-4),
76.48 (C-3’), 76.07 (C-3S), 75.29 (C-5’), 74.99 (CH2Ph), 74.85 (C-5), 73.54
(CH2Ph), 73.47 (CH2Ph), 72.32 (C-4’’), 71.74 (C-6), 71.41 (C-8’’, CH2Ph),
70.71 (C-5’’), 70.21 (C-6’’), 69.72 (C-4’), 68.90 (OCH2CH=CH2), 68.73 (C-
6), 67.11 (C-2, CO2CH2Ph), 67.00 (OCH2 of Fmoc), 63.30 (C-6’), 56.74
(C-2’), 47.64 (OCH2CH of Fmoc), 39.96 (C-2S), 34.79–14.08 [m, 3’’, SiC-
C68H92N4O15SiNa: 1255.6221; found: 1255.6168 [M+Na]+.
Dimethylthexylsilyl 3-O-[(R)-3-benzyloxy-dodecanoyl]-2-deoxy-2-(9-fluo-
renylmethoxycarbonyl)-b-d-glucopyranosyl-(1!6)-3-O-allyloxycarbonyl-
2-azido-4-O-benzyl-2-deoxy-b-d-glucopyranoside (16): TFA/H2O(3:2,
250 mL) was added dropwise to
a stirred solution of 15 (960 mg,
A
(CH3)2, CH3 of isopropylidene, C-(4S-12S)], ꢀ1.72 (SiCH3),
0.78 mmol) in CH2Cl2 (15 mL). The reaction mixture was stirred at room
temperature for 30 min, after which it was diluted with ethyl acetate
(15 mL) and then washed with saturated aqueous NaHCO3 (2”15 mL)
ꢀ3.29 ppm (SiCH3); HR MS: m/z: calcd for C99H126N4O20SiNa:
1745.8218; found: 1745.9780 [M+Na]+.
Chem. Eur. J. 2008, 14, 558 – 569
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
565

G.-J. Boons et al.
Oligosaccharide 18: 1H-Tetrazole (3% wt, 10.0 mmol) in CH2Cl2
(2.5 mL) was added to a solution of compound 17 (480 mg, 0.28 mmol)
6a, H-6a’, H-8a’’), 3.63 (d, 1H, J6a’,6b’ =10.8 Hz, H-6b’), 3.59–3.54 (m, 2H,
H-5, H-6b), 3.80 (dd, 1H, J=4.8, 9.6 Hz, H-5’), 3.33 (t, 1H, J3,4 =J4,5
=
and
N,N-diethyl-1,5-dihydro-3H-2,4,3-benzodioxaphosphepin-3-amine
9.0 Hz, H-4), 3.26 (dd, 1H, J1,2 =7.8, J2,3 =10.2 Hz, H-2), 3.11 (dd, 1H, J=
7.8 Hz, H-2’), 2.69–2.63 (m, 2H, H-2S), 2.32–2.20 (m, 3H, H-2L, H-2L’),
(133 mg, 0.56 mmol) in CH2Cl2 (8 mL). After the reaction mixture was
stirred at room temperature for 40 min, it was cooled (ꢀ208C), stirred
for another 10 min and then 3-chloroperoxybenzoic acid (mCPBA)
(500 mg, 50–55% wt, 1.12 mmol) was added. The reaction mixture was
stirred at ꢀ208C for 20 min, and then quenched by the addition of satu-
rated aqueous NaHCO3 (20 mL) and diluted with CH2Cl2 (20 mL). The
solution was washed with saturated aqueous NaHCO3 (2”30 mL) and
brine (2”20 mL). The organic phase was dried (MgSO4), filtered, and
the filtrate was concentrated in vacuo. The residue was purified by silica
gel column chromatography (hexane/ethyl acetate 4:1 to give 18 as an
amorphous solid (470 mg, 88%). Rf =0.45 (hexane/ethyl acetate 3:1);
2.18 (dd, 1H, J3a’’,3b’’ =14.4, J3a’’,4’’ =6.6 Hz, H-3a’’), 2.10 (dd, 1H, J3a’’,3b’’
14.4, J3b’’,4’’ =5.4 Hz, H-3b’’), 2.05 (dd, J2La,2Lb =15.0, J2Lb,3L =5.4 Hz, H-2Lb),
1.66–1.46 [m, 7H, H-4S, H-4L, H-3L’, CH(CH3)2], 1.38 [s, 3H, CH3 of iso-
=
AHCTREUNG
propylidene], 1.31 [s, 3H, CH3 of isopropylidene], 1.22 [brs, 48H, H-(5S-
11S), H-(5L-13L), H-(4L’-11L’)], 0.86–0.84 [m, 21H, H-12S, H-14L, H-12L’,
SiC
A
G
N
CD3COCD3): d = 173.59 (C=O), 171.06 (C=O), 169.98 (C=O), 167.78
(C=O), 154.35 (C=O), 138.84–127.46 (m, OCH2CH=CH2, aromatic),
119.17 (OCH2CH=CH2), 108.66 (C-2’’), 99.89 (C-1’), 98.58 [C(CH3)2 of
U
isopropylidene], 96.48 (C-1), 78.60 (C-3), 77.00 (C-7’’), 76.87 (C-4), 75.54
(C-3S), 74.46 (CH2Ph), 74.12 (C-4’), 73.84 (C-5), 73.35 (CH2Ph), 73.19
(CH2Ph), 72.56 (C-5’), 71.72 (C-3’), 71.16 (CH2Ph), 70.83 (C-5’’), 70.74
(C-3L), 69.79 (C-4’’, 8’’), 68.83 (C-6, OCH2CH=CH2), 68.44 (C-6’’,
CH2Ph), 68.71 (CH2Ph), 66.97 (CO2CH2Ph), 66.62 (C-2), 61.81 (C-6’),
[a]2D4 =+6.08 (c=1.0, CHCl3); 1H NMR (600 MHz, CD3COCD3): d
=
7.85–7.20 (m, 35H, aromatic), 6.84–6.74 (m, 3H, aromatic”2, NH’), 6.62
(d, 1H, JNH’,2 =9.0 Hz, NH’), 5.91–5.85 (m, 1H, OCH2CH=CH2), 5.46 (t,
1H, J2’,3’ =J3’,4’ =9.6 Hz, H-3’), 5.32–5.23 (m, 3H, CO2CH2Ph, OCH2CH=
CHH), 5.18 (dd, 1H, J=1.2, 10.8 Hz, OCH2CH=CHH), 5.08–4.89 (m,
5H, H-1’, H-3, 3”CHHPh), 4.85–4.79 (m, 3H, H-1, 2”CHHPh), 4.71–
4.63 (m, 4H, H-4’, H-4’’, 2”CHHPh), 4.61–4.56 (m, 6H, 4”CHHPh,
OCH2CH=CH2), 4.46–4.43 (m, 2H, H-5’’, CHHPh), 4.20–4.16 (m, 3H,
OCH2CH of Fmoc), 4.10–4.07 (m, 2H, H-6a, H-6’’), 4.02–4.01 (m, 2H, H-
7’’, H-8’’), 3.93 (dd, 1H, J5’,6a’ =4.8, J6a’,6b’ =11.4 Hz, H-6a’), 3.85 (dd, 1H,
56.76 (C-2’), 41.62 (C-2L), 38.91 (C-2S), 34.47–14.09 [m, 3’’, SiC(CH3)2CH-
G
A
(SiCH3), ꢀ3.41 ppm (SiCH3); HR MS: m/z: calcd for C105H129N4O25SiNa:
2114.1273; found: 2114.2964 [M+Na]+.
Oligosaccharide 20: [Pd(PPh3)4] (32.5 mg, 0.028 mmol) was added to a so-
A
lution of 19 (295 mg, 0.141 mmol), nBuNH2 (28 mL, 0.28 mmol), and
HCOOH (11 mL, 0.28 mmol) in THF (5 mL). After stirring the reaction
mixture at room temperature for 20 min, it was diluted with CH2Cl2
(20 mL), and washed successively with water (20 mL), saturated aqueous
NaHCO3 (2”20 mL) and brine (2”20 mL). The organic phase was dried
(MgSO4), filtered, and the filtrate was concentrated in vacuo. The residue
was purified by silica gel column chromatography (hexane/ethyl acetate
4:1) to give an alcohol intermediate as a colorless syrup (268 mg, 95%).
Rf =0.45 (hexane/ethyl acetate 3:1); 1H NMR (500 MHz, CDCl3): d =
7.43–7.14 (m, 27H, aromatic), 6.75 (d, 1H, J=7.5 Hz, aromatic), 6.65 (d,
1H, J=7.0 Hz, aromatic), 5.68 (d, 1H, JNH’,2 =8.0 Hz, NH’), 5.53 (t, 1H,
J
7’’,8a’’ =4.8, J8a’’,8b’’ =10.8 Hz, H-8a’’), 3.81–3.77 (m, 4H, H-5, H-6a, H-6b’,
H-3S), 3.67–3.60 (m, 3H, H-2’, H-4, H-5’), 3.35 (dd, 1H, J1,2 =7.8, J2,3
10.2 Hz, H-2), 2.70 (dd, 1H, J2Sa,2Sb =16.8, J2Sb,3S =6.6 Hz, H-2Sb), 2.53 (dd,
1H, 2Sa,2Sb =16.8, 2Sb,3S =4.8 Hz, H-2Sb), 2.26 (dd, 1H, 3a’’,3b’’ =14.4,
3a’’,4’’ =7.2 Hz, H-3a’’), 2.12 (dd, 1H, J3a’’,3b’’ =14.4, J3b’’,4’’ =4.8 Hz, H-3b’’),
1.68–1.63 [m, 1H, CH(CH3)2], 1.47–1.43 (m, 2H, H-4S), 1.37 (s, 3H, CH3
of isopropylidene), 1.31 (s, 3H, CH3 of isopropylidene), 1.16–1.08 [m,
=
J
J
J
J
ACHTREUNG
14H, H-(5S-11S)], 0.89–0.83 [m, 15H, H-12S, SiC
A
N
(s, 3H, SiCH3), 0.23 ppm (s, 3H, SiCH3); 13C NMR (75 MHz,
CD3COCD3): d = 171.78 (C=O), 168.35 (C=O), 156.47 (C=O), 155.03
(C=O), 144.93–120.62 (m, OCH2CH=CH2, aromatic), 118.71 (OCH2CH=
J
2’,3’ =J3’,4’ =10.0 Hz, H-3’), 5.23 (d, 1H, J=12.5 Hz, CO2CHHPh), 5.18 (d,
CH2), 109.06 (C-2’’), 102.03 (C-1’), 99.13 [C(CH3)2 of isopropylidene],
N
1H, J=12.5, CO2CHHPh), 5.10–4.92 (m, 4H, H-1’, H-3L, 2”CHHPh),
4.85–4.50 (m, 13H, H-1, H-4’, H-4’’, 10”CHHPh), 4.41 (brs, 1H, H-5’’),
4.10–4.08 (m, 1H, H-7’’), 4.02 (d, 1H, J=9.5 Hz, H-6’’), 3.95–3.89 (m,
3H, H-6a’, H-8a’’, H-3S), 3.83–3.79 (m, 2H, H-6a, H-8b’’), 3.73 (d, 1H,
97.02 (C-1), 79.25 (C-3), 78.01 (C-7’’), 77.49 (C-4), 75.91 (C-5 or 3S),
74.96, 74.90 (C-5 or 3S, CH2Ph), 74.68 (C-4’), 73.97 (C-3’), 73.76 (CH2Ph),
73.42 (C-5’, CH2Ph), 71.80 (C-5’’), 71.65 (CH2Ph), 70.56 (C-4’’), 70.35 (C-
8’’), 69.54 (C-6), 69.19 (C-6’’), 69.03 (OCH2CH=CH2), 68.82 [2”
(OCH2)2Ph], 67.43 (CO2CH2Ph), 67.34 (C-2), 67.25 (OCH2 of Fmoc),
62.91 (C-6’), 57.32 (C-2’), 47.77 (OCH2CH of Fmoc), 39.78 (C-2S), 34.88–
J
6a,6b =10.5 Hz, H-6b), 3.67 (dd, 1H, J5’,6b’ =5.5, Ja’,b’ =11.0 Hz, H-6b’),
3.48–3.42 (m, 3H, H-3, H-5, H-5’), 3.33–3.24 (m, 2H, H-2’, H-4), 3.15
(dd, 1H, J1,2 =8.0, J2,3 =10.0 Hz, H-2), 2.76–2.67 (m, 2H, H-2S), 2.37–2.27
(m, 4H, H-3a’’, H-2La, H-2L’), 2.15–2.08 (2H, H-3b’’, H-2Lb), 1.68–1.53 [m,
14.20 [m, 3’’, SiC
N
N
ꢀ1.61 (SiCH3), ꢀ3.18 ppm (SiCH3); HR MS: m/z: calcd for
7H, H-4S, H-4L, H-3L’, CH(CH3)2], 1.42 (s, 3H, CH3 of isopropylidene),
1.36 (s, 3H, CH3 of isopropylidene), 1.28 [brs, 48H, H-(5S-11S), H-(5L-
A
C105H129N4O25Si: 1927.8350; found: 1927.8330 [M+Na]+.
13L), H-(4L’-11L’)], 0.91–0.90 [m, 21H, H-12S, H-14L, H-12L’, SiC-
Oligosaccharide 19: DBU (100 mL) was added dropwise to a stirred solu-
tion of 18 (300 mg, 0.16 mmol) in CH2Cl2 (5 mL). The reaction mixture
was stirred at room temperature for 30 min, and then concentrated in
vacuo. The residue was purified by silica gel column chromatography
(hexane/ethyl acetate 5:2) to yield an amine intermediate as a pale
yellow oil (250 mg, 94%). Rf =0.25 (hexane/ethyl acetate 5:2). A reaction
mixture of 9 (95 mg, 0.22 mmol) and DCC (62 mg, 0.30 mmol) in CH2Cl2
(3 mL) was stirred at room temperature for 10 min, and then the above
obtained amine (250 mg, 0.15 mmol) was added, and stirring was contin-
ued for another 12 h. The insoluble materials were removed by filtration,
and the residue was washed with CH2Cl2 (2”0.5 mL). The combine fil-
trates were concentrated in vacuo, and the residue was purified by silica
gel column chromatography (hexane/ethyl acetate 5:1) to yield 19 as an
amorphous solid (280 mg, 89%). Rf =0.55 (hexane/ethyl acetate 5:2);
[a]2D5 =+7.18 (c=1.0, CHCl3); 1H NMR (600 MHz, CDCl3): d = 7.36–
7.07 (m, 27H, aromatic), 6.69 (d, 1H, J=7.2 Hz, aromatic), 6.51 (d, 1H,
J=7.2 Hz, aromatic), 5.90–5.84 (m, 1H, OCH2CH=CH2), 5.62 (d, 1H,
A
G
A
C112H163N4O24SiNa: 2030.1062; found: 2030.4662 [M+Na]+.
A solution of (R)-8 (72 mg, 0.234 mmol) and DCC (72 mg, 0.351 mmol)
in CH2Cl2 (5 mL) was stirred at room temperature for 10 min, and then
the above obtained intermediate (235 mg, 0.117 mmol) and DMAP
(7 mg, 0.06 mmol) were added. The reaction mixture was stirred for an-
other 10 h, after which the solids were removed by filtration and washed
with CH2Cl2 (2”2 mL). The combined filtrates were concentrated in
vacuo. The residue was purified by silica gel column chromatography
(hexane/ethyl acetate 5:1!4:1) to afford 20 as a white solid (228 mg,
84%). Rf =0.60 (hexane/ethyl acetate 3:1); [a]2D5 =+7.98 (c=1.0, CHCl3);
1H NMR (500 MHz, CDCl3): d = 7.42–7.12 (m, 32H, aromatic), 6.74 (d,
1H, J=7.5 Hz, aromatic), 6.56 (d, 1H, J=7.0 Hz, aromatic), 5.65 (d, 1H,
J
NH’,2 =7.5 Hz, NH’), 5.58 (t, 1H, J2’,3’ =J3’,4’ =9.5 Hz, H-3’), 5.25 (s, 2H,
CO2CH2Ph), 5.16 (t, 1H, J2,3 =J3,4 =9.5 Hz, H-3), 5.10–00 (m, 2H, H-3L,
CHHPh), 4.98 (d, 1H, J1’,2’ =8.5 Hz, H-1’), 4.92 (dd, 1H, J=11.0, 14.0 Hz,
CHHPh), 4.81–4.47 (m, 15H, H-1, H-4’, H-4’’, 12”CHHPh), 4.44 (brs,
1H, H-5’’), 4.12–4.10 (m, 1H, H-7’’), 4.05 (d, 1H, J=9.5 Hz, H-6’’), 3.96–
3.89 (m, 3H, H-8a’’, 2”H-3S), 3.85–3.79 (m, 3H, H-6a, H-6a’, H-8b’’),
3.68 (d, 1H, J6a’,6b’ =11.0 Hz, H-6b’), 3.64–3.57 (m, 2H, H-5, H-6b), 3.41
(dd, 1H, J=3.0, 10.0 Hz, H-5’), 3.33 (t, 1H, J3,4 =J4,5 =9.0 Hz, H-4), 3.22
(dd, 1H, J1,2 =8.0, J2,3 =10.5 Hz, H-2), 3.18 (dd, 1H, J=7.5 Hz, H-2’),
2.76–2.68 (m, 2H, H-2S), 2.63 (dd, 1H, J2Sa,2Sb =16.0, J2Sa,3S =7.0 Hz, H-
J
NH’,2 =7.2 Hz, NH’), 5.53 (t, 1H, J2’,3’ =J3’,4’ =9.6 Hz, H-3’), 5.20 (d, 1H,
J=17.4 Hz, OCH2CH=CHH), 5.23–5.16 (m, 3H, CO2CH2Ph, OCH2CH=
CHH), 5.03–5.16 (m, 3H, H-1’, H-3L, CHHPh), 4.89–4.81 (m, 2H, H-3,
CHHPh), 4.75–4.71 (m, 2H, H-3, CHHPh), 4.68–4.64 (m, 4H, H-4’,
CHHPh”3), 4.61–4.57 (m, 4H, H-1, H-4’’, O CH2CH=CH2), 4.57–4.47 (m,
6H, 6”CHHPh), 4.38 (brs, 1H, H-5’’), 4.04 (brs, 1H, H-7’’), 3.99 (d, 1H,
J=10.2 Hz, H-6’’), 3.89–3.84 (m, 2H, H-8a’’, H-3S), 3.80–3.74 (m, 3H, H-
566
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 558 – 569

Synthetic Lipid A Derivatives
FULL PAPER
2Sa), 2.50 (dd, 1H, J2Sa,2Sb =16.0, J2Sb,3S =6.0 Hz, H-2Sb), 2.38–2.26 (m, 3H,
H-2La, H-2L’), 2.22 (dd, 1H, J3a’’,3b’’ =14.0, J3a’’,4’’ =7.5 Hz, H-3a’’), 2.16 (dd,
2’), 56.20 (C-2), 41.56 (C-2L), 41.47 (C-2L), 39.60 (C-2S), 38.96 (C-2S),
34.47–14.09 [m, 3’’, SiC
A
E
1H, J3a’’,3b’’ =14.0, J3b’’,4’’ =5.5 Hz, H-3b’’), 2.11 (dd, J2La,2Lb =14.5, J2Lb,3L
=
(4S-12S), 2”C-(4L-14L), 2”C-(2L’-12L’)], ꢀ1.56 (SiCH3), ꢀ3.31 ppm
(SiCH3); HR MS: m/z: calcd for C157H241N2O29SiNa: 2700.6850; found:
2700.6572 [M+Na]+.
5.0 Hz, H-2Lb), 1.69–1.54 [m, 9H, 2”H-4S, H-L, H-3L’, CH(CH3)2], 1.44 (s,
AHCTREUNG
3H, CH3 of isopropylidene), 1.37 (s, 3H, CH3 of isopropylidene), 1.22
[brs, 62H, 2”H-(5S-11S), H-(5L-13L), H-(4L’-11L’)], 0.92–0.90 [m, 24H, 2”
Oligosaccharide 22: TFA/H2O3:2 (100 mL) was added dropwise to a
stirred solution of 21 (30 mg, 11 mmol) in CH2Cl2 (2 mL). The reaction
mixture was stirred at room temperature for 6 h, after which it was dilut-
ed with ethyl acetate (10 mL) and then washed with saturated aqueous
NaHCO3 (2”10 mL) and brine (2”10 mL). The organic phase was dried
(MgSO4), filtered, and the filtrate was concentrated in vacuo. The residue
was purified by preparative silica gel TLC (hexane/ethyl acetate 3:1) to
afford the lactol as a pale yellow syrup (25 mg, 89%). Rf =0.25 (hexane/
acetone 1:1); 1H NMR (600 MHz, CDCl3): d = 7.42–7.19 (m, 32H, aro-
matic), 6.82 (d, 1H, J=7.8 Hz, aromatic), 6. 64 (d, 1H, J=6.0 Hz, aro-
matic), 5.91 (d, 1H, J2,NH =9.6 Hz, NH), 5.81 (d, 1H, J2’,NH’ =7.8 Hz,
NH’), 5.44 (t, 1H, J2’,3’ =J3’,4’ =9.6 Hz, H-3’), 5.38 (d, 1H, J1’,2’ =7.8 Hz, H-
1’), 5.32 (dd, 1H, J=9.6, 10.2 Hz, H-3), 5.19 (d, 1H, J=12.6 Hz,
CO2CHHPh), 5.14 (d, 1H, J=12.6 Hz, CO2CHHPh), 5.11–5.07 (m, 2H,
H-1, H-3L), 5.01–4.88 (m, 3H, H-3L, 2”CHHPh), 4.83 (dd, 1H, J=12.2,
13.8 Hz, CHHPh), 4.74–4.67 (m, 3H, 3”CHHPh), 4.64–4.59 (m, 1H, H-
4’), 4.58–4.41 (m, 10H, 7”CHHPh), 4.35–4.31 (m, 2H, H-4’’, CHHPh),
4.12–3.94 (m, 5H, H-2, H-5, H-5’’, H-6’’, H-7’’), 3.86–3.72 (m, 7H, H-6a,
H-12S, H-14L, H-12L’, SiC
C
G
G
(C-2’’), 99.90 (C-1’), 98.60 (C(CH3)2 of isopropylidene), 96.40 (C-1),
E
77.43–76.58 (C-3, 7’’), 76.58 (C-4), 75.75 (C-3S), 75.51 (C-3S), 74.17 (C-5),
73.94 (C-4’), 73.87 (CH2Ph), 73.36 (CH2Ph), 73.19 (CH2Ph), 72.52 (C-5’),
71.74 (C-3’), 71.47 (CH2Ph), 71.15 (CH2Ph), 70.83 (C-5’’), 70.75 (C-3L),
69.80 (C-4’’, 8’’), 68.95 (C-6), 68.40 (C-6’’, CH2Ph), 68.09 (CH2Ph), 66.97
(CO2CH2Ph), 66.90 (C-2), 61.79 (C-6’), 56.70 (C-2’), 41.64 (C-2L), 39.72
(C-2S), 38.92 (C-2S), 34.48–14.09 [m, 3’’, SiC
N
N
propylidene, 2”C-(4S-12S), C-(4L-14L), C-(2L’-12L’)], ꢀ1.84 (SiCH3),
ꢀ3.42 ppm (SiCH3); HR MS: m/z: calcd for C131H191N4O26SiNa:
2318.3151; found: 2318.5374 [M+Na]+.
Oligosaccharide 21: A suspension of compound 20 (120 mg, 52 mmol) and
zinc (< 10 micron, 210 mg, 3.2 mmol) in
a mixture of acetic acid
(100 mL) and CH2Cl2 (5 mL) was stirred at room temperature for 1 h,
after which it was diluted with ethyl acetate (20 mL). The solids were re-
moved by filtration, and the residue was washed with ethyl acetate (2”
3 mL). The combined filtrates were washed with saturated aqueous
NaHCO3 (2”20 mL) and brine (2”20 mL). The organic phase was dried
(MgSO4), filtered, and the filtrate was concentrated in vacuo. The residue
was purified by silica gel column chromatography (hexane/ethyl acetate
4:1) to afford the amine as a pale yellow syrup (108 mg, 92%). Rf =0.45
(hexane/ethyl acetate 5:2).
H-6a’, H-6b’, H-8a’’, H-8b’’, 2”H-3S), 3.66 (d, 1H, J5’,6b’ =7.2, J6a’,6b’
=
12.0 Hz, H-6b’), 3.46 (brs, 1H, H-5’), 3.24 (t, 1H, J3,4 =J4,5 =9.6 Hz, H-4),
3.02 (dd, 1H, J=7.8 Hz, H-2’), 2.68 (dd, 1H, J2Sa,2Sb =16.2, J2Sa,3S =4.8 Hz,
H-2Sa), 2.61 (dd, 1H, J2Sa,2Sb =16.2, J2Sb,3S =6.6 Hz, H-2Sb), 2.55 (dd, 1H,
J
2Sa,2Sb =15.6, J2Sa,3S =7.8 Hz, H-2Sa), 2.41 (dd, 1H, J2Sa,2Sb =16.2, J2Sb,3S
4.8 Hz, H-2Sb), 2.34–2.22 (m, 7H, 2”H-2La, H-2Lb, H-2L’ ”4), 2.18 (dd,
1H, J3a’’,3b’’ =12.6, J3a’’,4’’ =4.6 Hz, H-3a’’), 2.11 (dd, J2La,2Lb =15.6, J2Lb,3L
=
=
A solution of (R)-3-dodecanoyloxy-dodecanoic acid (9; 43 mg, 100 mmol)
and DCC (31 mg, 150 mmol) in CH2Cl2 (5 mL) was stirred at room tem-
perature for 10 min, and then the above obtained intermediate (120 mg,
53 mmol) was added. The reaction mixture was stirred for another 10 h,
after which the solids were removed by filtration and washed with
CH2Cl2 (2”1 mL). The combined filtrates were concentrated in vacuo.
The residue was purified by silica gel column chromatography (hexane/
ethyl acetate 5:1!4:1) to afford 21 as an amorphous solid (126 mg,
89%). Rf =0.65 (hexane/ethyl acetate 5:2); [a]2D5 =+4.38 (c=1.0, CHCl3);
1H NMR (600 MHz, CDCl3): d = 7.47–7.17 (m, 32H, aromatic), 6.76 (d,
1H, J=7.8 Hz, aromatic), 6.66 (d, 1H, J=6.0 Hz, aromatic), 5.71–5.69
(m, 2H, NH, NH’), 5.64 (t, 1H, J2’,3’ =J3’,4’ =9.6 Hz, H-3’), 5.30 (d, 1H, J=
13.2 Hz, CO2CHHPh), 5.25 (d, 1H, J=13.2 Hz, CO2CHHPh), 5.16 (t,
1H, J2,3 =J3,4 =9.5 Hz, H-3), 5.15 (t, 1H, J2,3 =J3,4 =9.6 Hz, H-3), 5.13–5.05
(m, 3H, 2”H-3L, CHHPh), 5.04 (d, 1H, J1’,2’ =8.4 Hz, H-1’), 4.97 (dd,
1H, J=11.4, 13.8 Hz, CHHPh), 4.87–4.75 (m, 4H, 4”CHHPh), 4.73–4.69
(m, 1H, H-4’), 4.68 (d, 1H, J1,2 =7.8 Hz, H-1), 4.65–4.46 (m, 10H, H-4’’,
H-5’’, 8”CHHPh), 4.15–4.13 (m, 1H, H-7’’), 4.06 (d, 1H, J=9.6 Hz, H-
6’’), 3.99–3.93 (m, 3H, H-2, H-8a’’, H-3S), 3.89–3.84 (m, 4H, H-6a, H-6a’,
H-8b’’, H-3S), 3.66 (d, 1H, J6a’,6b’ =10.8 Hz, H-6b’), 3.62–3.56 (m, 2H, H-5,
H-6b), 3.49 (dd, 1H, J=4.2, 10.2 Hz, H-5’), 3.38 (t, 1H, J3,4 =J4,5 =9.0 Hz,
6.0 Hz, H-2Lb), 1.96 (t, 1H, J3a’’,3b’’ =J3b’’,4’’ =12.6 Hz, H-3b’’), 1.61–1.48 (m,
14H, 2”H-4S, 2”H-4L, 2”3L’), 1.22 [brs, 96H, 2”H-(5S-11S), 2”H-(5L-
13L), 2”H-(4L’-11L’)], 0.96–0.87 ppm (m, 18H, 2”H-12S, 2”H-14L, 2”H-
12L’); HR MS: m/z: calcd for C146H219N2O29SiNa: 2518.5359; found:
2518.2606 [M+Na]+.
To
a
cooled (ꢀ788C) solution of the lactol intermediate (23 mg,
9.2 mmol) and tetrabenzyl diphosphate (10 mg, 18.4 mmol) in THF
(2 mL) was added dropwise lithium bis(trimethylsilyl)amide in THF
(1.0m, 15 mL, 15 mmol). The reaction mixture was stirred for 1 h, and
then allowed to warm up to ꢀ208C. After stirring the reaction mixture
for 1 h at ꢀ208C, it was quenched with saturated aqueous NaHCO3
(10 mL), and diluted with ethyl acetate (15 mL). The organic phase was
washed with brine (2”15 mL), dried (MgSO4), filtered, and the filtrate
was concentrated in vacuo. The residue was purified by Iatrobeads
column chromatography (hexane/ethyl acetate 3:1 ! 2:1 ! 1:1 ! 3:4)
to give 22 as a colorless syrup (19 mg, 75%). Rf =0.50 (hexane/acetone
1:1); 1H NMR (500 MHz, CDCl3): d = 7.40–7.14 (m, 42H, aromatic),
6.82 (d, 1H, J=8.0 Hz, aromatic),.69 (d, 1H, J=7.5 Hz, aromatic), 6.41
(d, 1H, J2’,NH’ =8.5 Hz, NH’), 6.04 (d, 1H, J2,NH =8.5 Hz, NH), 5.65 (brs,
1H, H-1), 5.35 (t, 1H, J2’,3’ =J3’,4’ =9.5 Hz, H-3’), 5.24–4.73 (m, 12H, H-1’,
H-3, 5”CHHPh), 4.61–4.36 (m, 11H, H-4’, 5”CHHPh), 4.38 (d, 1H, J=
11.5 Hz, CHHPh), 4.27–4.23 (m, 1H, H-4’’), 4.19–4.14 (m, 1H, H-2),
4.09–3.96 (m, 4H, H-5, H-5’’, H-6’’, H-7’’), 3.91–3.74 (m, 8H, H-6a, H-6a’,
H-6b, H-6b’, H-8a’’, H-8b’’, 2”H-3S), 3.53–3.44 (m, 3H, H-2’, H-4, H-5’),
H-4), 3.19 (dd, 1H, J=7.8 Hz, H-2’), 2.77 (dd, 1H, J2Sa,2Sb =16.2, J2Sa,3S
7.2 Hz, H-2Sa), 2.72 (dd, 1H, J2Sa,2Sb =16.2, J2Sb,3S =4.8 Hz, H-2Sb), 2.61 (dd,
1H, 2Sa,2Sb =16.2, 2Sa,3S =7.2 Hz, H-2Sa), 2.49 (dd, 1H, 2Sa,2Sb =16.2,
2Sb,3S =5.4 Hz, H-2Sb), 2.43–2.24 (m, 8H, H-3a’’, 2”H-2La, H-2Lb, 2”H-
2L’), 2.19 (dd, 1H, J3a’’,3b’’ =15.0, J3b’’,4’’ =5.4 Hz, H-3b’’), 2.16 (dd, 1H,
2La,2Lb =15.0, J2Lb,3L =6.0 Hz, H-2Lb), 1.68–1.58 [m, 15H, 2”H-4S, 2”H-4L,
2”H-3L’, CH(CH3)2], 1.46 (s, 3H, CH3 of isopropylidene), 1.39 (s, 3H,
=
J
J
J
J
2.70 (dd, 1H, J2Sa,2Sb =16.5, J2Sa,3S =7.0 Hz, H-2Sa), 2.65 (dd, 1H, J2Sa,2Sb
16.5, J2Sb,3S =5.0 Hz, H-2Sb), 2.56 (dd, 1H, J2Sa,2Sb =16.0, J2Sa,3S =7.5 Hz, H-
Sa), 2.48 (dd, 1H, J2Sa,2Sb =16.0, J2Sb,3S =5.0 Hz, H-2Sb), 2.31–2.18 (m, 7H,
2”H-2La, H-2Lb, 2”H-2L’), 2.12 (dd, 1H, J3a’’,3b’’ =11.0, J3a’’,4’’ =5.0 Hz, H-
3a’’), 2.09 (dd, J2La,2Lb =16.5, J2Lb,3L =6.0 Hz, H-2Lb), 2.04 (t, 1H, J3a’’,3b’’
3b’’,4’’ =12.5 Hz, H-3b’’), 1.61–1.49 (m, 14H, 2”H-4S, 2”H-4L, 2”3L’), 1.26
=
J
2
ACHTREUNG
CH3 of isopropylidene), 1.22 [brs, 96H, 2”H-(5S-11S), 2”H-(5L-13L), 2”
H-(4L’-11L’)], 0.96–0.87 [m, 30H, 2”H-12S, 2”H-14L, 2”H-12L’, SiC-
=
J
A
N
13C NMR (75 MHz, CDCl3): d = 173.60 (C=O), 173.54 (C=O), 171.62
(C=O), 171.06 (C=O), 169.95 (C=O), 169.05 (C=O), 168.04 (C=O),
[brs, 96H, 2”H-(5S-11S), 2”H-(5L-13L), 2”H-(4L’-11L’)], 0.90–0.88 ppm
(m, 18H, 2”H-12S, 2”H-14L, 2”H-12L’); HR MS: m/z: calcd for
C
160H232N2O32SiNa: 2778.5961; found: 2778.9185 [M+Na]+.
139.00–127.43 (m, aromatic), 108.69 (C-2’’), 99.60 (C-1’), 98.43 [C(CH3)2
A
(3-Deoxy-a-d-manno-oct-2-ulopyranosyl)onate-(2!6)-3-O-[(R)-3-hy-
droxy-dodecanoyl]-2-deoxy-2-[(R)-3-dodecanoyloxy-tetradecanoyl]-b-d-
glucopyranosyl-(1!6)-3-O-[(R)-3-hydroxy-dodecanoyl]-2-deoxy-2-[(R)-
3-dodecanoyloxy-tetradecanoyl]-a-d-glucopyranose 1,4’-biphosphate (3):
A mixture of 22 (9 mg, 3.3 mmol) and Pd-black (15.0 mg) in anhydrous
of isopropylidene], 85.85 (C-1), 77.26 (C-7’’), 76.90 (C-4), 75.50–75.41
(3C, C-3, C-3S ”2), 74.22–74.08 (4C, C-4’, 5, CH2Ph”2), 73.37 (CH2Ph),
72.52 (C-5’), 71.75 (C-3’), 71.27 (CH2Ph), 71.21 (CH2Ph), 70.98 (C-5’’),
70.71 (C-3L ”2), 70.20 (C-8’’), 69.88 (C-4’’), 68.85 (C-6), 68.62 (C-6’’),
68.41 (CH2Ph), 68.13 (CH2Ph), 66.98 (CO2CH2Ph), 61.90 (C-6’), 56.74 (C-
Chem. Eur. J. 2008, 14, 558 – 569
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
567

G.-J. Boons et al.
THF (3 mL) was shaken under an atmosphere of H2 (65 psi) at room
temperature for 30 h, after which it was neutralized with triethylamine
(10 mL), and the catalyst was removed by filteration and the residue
washed with THF (2”1 mL). The combined filtrates were concentrated
[3] P. Brandtzaeg, R. Ovsteboo, P. Kierulf, J. Infect. Dis. 1992, 166, 650–
652.
[4] M. van Deuren, P. Brandtzaeg, J. W. van der Meer, Clin. Microbiol.
Rev. 2000, 13, 144–166.
[5] S. Akira, K. Takeda, T. Kaisho, Nat. Immunol. 2001, 2, 675–680.
[6] C. Pasare, R. Medzhitov, Semin. Immunol. 2004, 16, 23–26.
[7] M. Karaghiosoff, R. Steinborn, P. Kovarik, G. Kriegshauser, M. Bac-
carini, B. Donabauer, U. Reichart, T. Kolbe, C. Bogdan, T. Leander-
son, D. Levy, T. Decker, M. Muller, Nat. Immunol. 2003, 4, 471–
477.
in vacuo to afford
3 as a
colorless film (5.0 mg, 78%). 1H NMR
(600 MHz, CDCl3): d = 5.24 (brs, 1H, H-1), 4.91–4.83 (m, 4H, H-3, H-
3’, 2”H-3L), 3.86 (m, 1H, H-2), 3.82–3.45 (m, 11H, H-5, H-5’, H-5’’, H-
6a, H-6b, H-6’a, H-6’b, H-6’’, H-7’’, H-8’’a, H-8’’b), 3.80 (m, 2H, H-4’, H-
4’’), 3.76 (m, 1H, H-3S), 3.72 (m, 1H, H-3S), 3.60 (m, 1H, H-2’), 3.29 (m,
1H, H-4), 2.23–2.07 (m, 8H, 2”H-2S, 2”H-2L), 2.04–1.98 (m, 4H, 2”H-
2L’), 1.76 (dd, 1H,
J3a’’,3b’’ =15.0, J3a’’,4’’ =5.4 Hz, H-3a’’), 1.61 (t, 1H,
[8] H. J. Jennings, A. K. Bhattacharjee, L. Kenne, C. P. Kenny, G.
Calver, Can. J. Biochem. 1980, 58, 128–136.
J
3a’’,3b’’ =J3b’’,4’’ =15.0 Hz, H-3b’’), 1.41–1.19 (m, 14H, 2”H-4S, 2”H-4L, 2”
H-3L’), 1.12 [brs, 96H, 2”H-(5S-11S), 2”H-(5L-13L), 2”H-(4L’-11L’)],
0.91–0.88 ppm (m, 18H, 2”H-12S, 2”H-14L, 2”H-12L’); HR MS: m/z:
(negative) calcd for C160H232N2O32Si: 1933.1838; found: 1932.6287 [M]+,
1954.5441 [M+NaꢀH]+, 1976.4611 [M+2Naꢀ2H]+.
Cell maintenance: RAW 264.7 gNO(ꢀ) cells, derived from the RAW
264.7 mouse monocyte/macrophage cell line, were obtained from ATCC.
The cells were maintained in RPMI 1640 medium (ATCC) with l-gluta-
mine (2 mm), adjusted to contain sodium bicarbonate (1.5 gLꢀ1), glucose
(4.5 gLꢀ1), HEPES (10 mm), and sodium pyruvate (1.0 mm) and supple-
mented with penicillin (100 umLꢀ1)/streptomycin (100 mgmLꢀ1; Medi-
atech) and fetal bovine serum (FBS, 10%; Hyclone). Cells were main-
tained in a humid 5% CO2 atmosphere at 378C.
[9] A. Gamian, M. Beurret, F. Michon, J. R. Brisson, H. J. Jennings, J.
Biol. Chem. 1992, 267, 922–925.
[10] R. P. Darveau, Curr. Opin. Microbiol. 1998, 1, 36–42.
[11] V. A. Kulshin, U. Zahringer, B. Lindner, C. E. Frasch, C. M. Tsai,
B. A. Dmitriev, E. T. Rietschel, J. Bacteriol. 1992, 174, 1793–1800.
[12] C. Alexander, U. Zahringer, Trends Glycosci. Glycotechnol. 2002,
14, 69–86.
[13] C. Erridge, E. Bennett-Guerrero, I. R. Poxton, Microbes Infect.
2002, 4, 837–851.
[14] S. M. Zughaier, S. M. Zimmer, A. Datta, R. W. Carlson, D. S. Ste-
phens, Infect. Immun. 2005, 73, 2940–2950.
[15] S. Kusumoto, in Molecular Biochemistry and Cellular Biology, Vol.
1, Bacterial Endotoxin Lipopolysaccharides, Chemical Synthesis of
Lipid A, CRC Press, Boca Raton, 1992, pp. 81–105, Chapter 9.
[16] S. Kusumoto, K. Fukase, M. Oikawa, in Endotoxin in Health and
Disease (Ed.: H. Brade), Marcel Dekker, New York, 1999, pp. 243–
256.
[17] S. M. Zughaier, Y. L. Tzeng, S. M. Zimmer, A. Datta, R. W. Carlson,
D. S. Stephens, Infect. Immun. 2004, 72, 371–380.
[18] S. Zughaier, S. Agrawal, D. S. Stephens, B. Pulendran, Vaccine 2006,
24, 1291–1297.
[19] M. Muroi, K. Tanamoto, Infect. Immun. 2002, 70, 6043–6047.
[20] H. Yoshizaki, N. Fukuda, K. Sato, M. Oikawa, K. Fukase, Y. Suda,
S. Kusumoto, Angew. Chem. 2001, 113, 1523–1528; Angew. Chem.
Int. Ed. 2001, 40, 1475–1480.
[21] A. V. Demchenko, M. A. Wolfert, B. Santhanam, J. N. Moore, G. J.
Boons, J. Am. Chem. Soc. 2003, 125, 6103–6112.
[22] B. Santhanam, M. A. Wolfert, J. N. Moore, G. J. Boons, Chem. Eur.
J. 2004, 10, 4798–4807.
[23] H. S. Lee, M. A. Wolfert, Y. H. Zhang, G. J. Boons, Chembiochem
2006, 7, 140–148.
[24] Y. Zhang, M. A. Wolfert, G. J. Boons, Bioorg. Med. Chem. 2007, 15,
4800–4812.
[25] M. Vasan, M. A. Wolfert, G. J. Boons, Org. Biomol. Chem. 2007, 5,
2087–2097.
[26] P. A. M. van der Klein, G. J. P. H. Boons, G. H. Veeneman, G. A.
van der Marel, J. H. van Boom, Tetrahedron Lett. 1989, 30, 5477–
5480.
[27] K. Fukase, Y. Fukase, M. Oikawa, W. C. Liu, Y. Suda, S. Kusumoto,
Tetrahedron 1998, 54, 4033–4050.
Cytokine induction and ELISA tests: RAW 264.7 gNO(ꢀ) cells were
plated on the day of the exposure assay as 2”105 cells/well in 96-well
tissue culture plates (Nunc). Cells were incubated with different stimuli
for 5.5 h in replicates of five. Culture supernatants were then collected,
pooled, and stored frozen (ꢀ808C) until assayed for cytokine production.
Cytokine ELISA tests were performed in 96-well MaxiSorp plates
(Nunc). A DuoSet ELISA Development Kit (R&D Systems) was used
for the quantification of mouse TNF-a according to the manufacturerꢁs
instructions. The absorbance was measured at 450 nm with wavelength
correction set to 540 nm using a microplate reader (BMG Labtech). Con-
centrations of IFN-b in culture supernatants were determined as follows.
ELISA MaxiSorp plates were coated with rabbit polyclonal antibody
against mouse IFN-b (PBL Biomedical Laboratories). IFN-b in standards
and samples was allowed to bind to the immobilized antibody. Rat anti-
mouse IFN-b antibody (USBiological) was then added, producing an an-
tibody-antigen-antibody “sandwich”. Next, horseradish peroxidase
(HRP) conjugated goat anti-rat IgG (H+L) antibody (Pierce) and a
chromogenic substrate for HRP 3,3’,5,5’-tetramethylbenzidine (TMB;
Pierce) were added. After the reaction was stopped, the absorbance was
measured at 450 nm with wavelength correction set to 540 nm. All cyto-
kine values are presented as the means ꢂ SD of triplicate measurements,
with each experiment being repeated three times.
Data analysis: Concentration-response data were analyzed using nonlin-
ear least-squares curve fitting in Prism (GraphPad Software, Inc.). These
data were fit with the following four parameter logistic equation: Y=
Emax/(1+
(EC50/X)Hillslope), where Y is the cytokine response, X is logarithm
G
of the concentration of the stimulus, Emax is the maximum response, and
EC50 is the concentration of the stimulus producing 50% stimulation.
The Hill slope was set at 1 to be able to compare the EC50 values of the
different inducers.
[28] T. Eisele, H. Ishida, G. Hummel, R. R. Schmidt, Liebigs Ann. 1995,
2113–2121.
[29] N. Kubasch, R. R. Schmidt, Eur. J. Org. Chem. 2002, 2710–2726.
[30] M. F. Loewe, R. J. Cvetovich, L. M. DiMichele, R. F. Shuman, E. J. J.
Grabowski, J. Org. Chem. 1994, 59, 7870–7875.
[31] L. Jiang, T. H. Chan, Tetrahedron Lett. 1998, 39, 355–358.
[32] M. Sakagami, H. Hamana, Tetrahedron Lett. 2000, 41, 5547–5551.
[33] R. R. Schmidt, M. Stumpp, Liebigs Ann. Chem. 1983, 1249–1256.
[34] R. R. Schmidt, Angew. Chem. 1986, 99, 213–236; Angew. Chem. Int.
Ed. Engl. 1986, 25, 212–235.
Acknowledgements
This research was supported by the Institute of General Medicine of the
National Institutes of Health (GM061761). N. meningitides LOS was pro-
vided by Dr. Russell W. Carlson.
[35] M. Imoto, N. Kusunose, Y. Matsuura, S. Kusumoto, T. Shiba, Tetra-
hedron Lett. 1987, 28, 6277–6280.
[36] F. M. Unger, D. Stix, G. Schulz, Carbohydr. Res. 1980, 80, 191–195.
[37] Y. Watanabe, Y. Komoda, K. Ebisuya, S. Ozaki, Tetrahedron Lett.
1990, 31, 255–256.
[38] Y. Tsukamoto, H. Nakagawa, H. Kajino, K. Sato, K. Tanaka, T.
Yanai, Biosci. Biotechnol. Biochem. 1997, 61, 1650–1657.
[1] A. Robbins, P. Freeman, Sci. Am. 1988, 259, 126–133.
[2] P. Brandtzaeg, P. Kierulf, P. Gaustad, A. Skulberg, J. N. Bruun, S.
Halvorsen, E. Sorensen, J. Infect. Dis. 1989, 159, 195–204.
568
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 558 – 569

Synthetic Lipid A Derivatives
FULL PAPER
[39] M. Oikawa, T. Shintaku, H. Sekljic, K. Fukase, S. Kusumoto, Bull.
Chem. Soc. Jpn. 1999, 72, 1857–1867.
[40] Y. Zhang, J. Gaekwad, M. A. Wolfert, G. J. Boons, J. Am. Chem.
Soc. 2007, 129, 5200–5216.
Received: July 27, 2007
Published online: October 17, 2007
Chem. Eur. J. 2008, 14, 558 – 569
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
569