Structural transition of lipopolysaccharide and reduction in the biological activity by amphiphilic lipid with cationic amino acid

Article abstract of DOI:10.1246/bcsj.20120353

Lipopolysaccharide (LPS), the endotoxin of Gram-negative bacteria, is a strong elicitor in the immune system by interacting with lipopolysaccharide- binding protein and CD14 with high specificity. The removal of LPS contamination in protein drug products expressed by bacteria is essential in pharmaceutical products for human use. Although polymyxin B (PMB)-immobilized columns are mainly used for removal of LPS, there are some problems, such as high production cost, and the toxicity of ligands. We synthesized aromatic lipids bearing lysine or arginine at the headgroup. These lipids form a complex with LPS through electrostatic interaction between cationic amino acids and phosphate groups in the lipid A backbone. The resultant complexes induce the structural transition of LPS from a cylindrical structure to a vesicle. Addition of amino-lipid/LPS complexes to RAW264.7 cells, a macrophage-like cell line, decrease the LPS activity. The efficiencies are higher than commonly used cationic compounds, such as dioleoyltrimethylammoniumpropane (DOTAP) and PMB. These results show that amphiphilic lipids with cationic amino acids can be used for deactivation of LPS.

Full text of DOI:10.1246/bcsj.20120353

© 2013 The Chemical Society of Japan
Bull. Chem. Soc. Jpn. Vol. 86, No. 5, 651-656 (2013)
651
Structural Transition of Lipopolysaccharide and Reduction in the
Biological Activity by Amphiphilic Lipid with Cationic Amino Acid
Wenjing Li,1 Shinichi Mochizuki,1 and Kazuo Sakurai*1,2
1Department of Chemistry and Biochemistry, The University of Kitakyushu,
1-1 Hibikino, Wakamatsu-ku, Kitakyushu, Fukuoka 808-0135
2CREST, Japan Science and Technology Agency, 4-1-8 Honcho, Kawaguchi, Saitama 332-0012
Received December 24, 2012; E-mail: sakurai@kitakyu-u.ac.jp
Lipopolysaccharide (LPS), the endotoxin of Gram-negative bacteria, is a strong elicitor in the immune system by
interacting with lipopolysaccharide-binding protein and CD14 with high specificity. The removal of LPS contamina-
tion in protein drug products expressed by bacteria is essential in pharmaceutical products for human use. Although
polymyxin B (PMB)-immobilized columns are mainly used for removal of LPS, there are some problems, such as high
production cost, and the toxicity of ligands. We synthesized aromatic lipids bearing lysine or arginine at the headgroup.
These lipids form a complex with LPS through electrostatic interaction between cationic amino acids and phosphate
groups in the lipid A backbone. The resultant complexes induce the structural transition of LPS from a cylindrical
structure to a vesicle. Addition of amino-lipid/LPS complexes to RAW264.7 cells, a macrophage-like cell line, decrease
the LPS activity. The efficiencies are higher than commonly used cationic compounds, such as dioleoyltrimethylammo-
niumpropane (DOTAP) and PMB. These results show that amphiphilic lipids with cationic amino acids can be used for
deactivation of LPS.
Innate immunity is the front line of biological defense
against microbial infections. It is initiated by recognition of
microbial components with a variety of pathogen sensors,
such as Toll-like receptors (TLRs)1,2 and CD14.3 Lipopoly-
saccharide (LPS) is the major structural component of Gram-
negative bacteria comprising of three distinct domains: O-
antigen polysaccharide chains, core oligosaccharides, and a
lipid (called lipid-A, Figure 1) decorated with phosphorylated
glucosamine disaccharides.4,5 The lipid-A hydrophobic chains
anchor LPS into the bacterial membrane and bacteriolysis or
external stimuli can cause the fragments containing the lipids to
be released. In our biological system, these fragments induce
furious immune response and thus lead to fever, diarrhea, and
fatal shock. This is the historical origin that such fragments are
called endotoxin and the related response is called endotoxic
shock. According to recent immunology, the immune response
to LPS is advocated as follows;6-10 the lipopolysaccharide
binding protein (LBP) binds to LPS and induces the transfer
of LPS to CD14. LPS/CD14 complexes are recognized by
TLR4/myeloid differentiation protein (MD2) heterodimer. The
resulting complexes activate signal transduction and induce
the secretion of proinflammatory cytokines. LPS can show a
strong immune response even at very low concentrations. For
instance, the intravenous administration of LPS at less than
kg body weight per hour for intravenous administration, where
1 EU is approximately 100 pg bacterial endotoxin.12 There-
fore the development of techniques to remove endotoxin from
products is crucial for therapeutic usage.
One of the most common methods to remove endotoxin
is affinity chromatography where ligands with a high biding
affinity for LPS are attached on a solid phase. Polymyxin B
(PMB, Figure 1) is a cyclic lipopeptide produced by Bacillus
polymyxa and can be used for such purpose.13 Since PMB
has four primary amino groups and the lipid-A backbone has
two phosphate groups (Figure 1), it is thought that the binding
between LPS and PMB is driven by electrostatic and hydro-
phobic interactions.5,14 It has been reported that the aggregated
structures are essential for the expression of bioactivity.15-18
When PMB binds to LPS, PMB leads to a strong fluidization
of the acyl chains of LPS, resulting in the reduction in the
endotoxin activity.17 However, there are several drawbacks
in use of PMB including high production cost, complexity
to immobilize ligands on supports, and the toxicity of PMB
itself.19 Furthermore, since PMB is difficult to synthesize, it
has to be exclusively prepared through biosynthesis. Therefore
there is a strong desire for alternatives. The most representa-
tive examples are synthetic compounds and oligopeptides with
cationic amino acids such as lysine and arginine.20,21
¹1
1 ng kg in humans induces endotoxin shock or sepsis.11 Pro-
We have reported that a series of lipids that have an aromatic
linker connected with primary amine, quaternary ammonium
salt, or ethylenediamine can be used as a transfection reagent
for DNA with better efficiency.22,23 It is expected that these
cationic compounds can form electrostatic complexes with
the negatively charged LPS as well as plasmid DNA. In this
study, we synthesized two lipids that have an aromatic linker
teins expressed by Gram-negative bacteria are often contami-
nated with endotoxin. This contamination must be removed
completely before the proteins are used for in vivo studies with
animal models or sold as pharmaceutical products. According
to FDA recommendation, the general threshold level of endo-
toxin is defined as no more than 5.0 endotoxin units (EU) per
Published on the web March 23, 2013; doi:10.1246/bcsj.20120353

652
Bull. Chem. Soc. Jpn. Vol. 86, No. 5 (2013)
LPS Deactivation by Lipid with Amino Acid
OH
O
O
O
O
O
-
O
P
O
-
O
P
O
HO
NH2
OH
O
O
H
HO
NH
NH
N
O
H2N
O
O
O
O
O
O
O
O
OH
OH
O
O
Arg-L
O
NH
2
H
N
H
N
HN
O
O
NH
2
Lipid A
Lys-L
NH2
NH2
O
O
O
O
O
O
H
H
H
H
N
N
N
N
N
O
O
N
H
N
H
H
N
O
HO
HN
O
O
O
O
HN
OH
H
N
N
H
O
O
NH2
NH2
DOTAP
Polymyxin B
Figure 1. Chemical structures of lipid A and the cationic compounds used in this work.
connected with lysine or arginine and evaluated the LPS deacti-
vation performance of them.
the fluorescence intensities with a fluorescence spectrophotom-
eter (Hitachi F-4500; Hitachi, Ltd., Tokyo, Japan). Excitation
was at 385 nm and the emission range was set between 400-
Material and Methods
700 nm.
¦-Potential Measurement.
¹1
Synthesis of the Amino-Lipid. Two types of a new lipid
bearing arginine (Arg-L) or lysine (Lys-L) were synthesized as
shown in Scheme 1 in a similar manner to a histidine bearing
lipid described elsewhere.24 The synthetic details and NMR
characterization are described in Supporting Information. The
chemical structures and their notations are shown in Figure 1.
Preparation of the Amino-Lipid/LPS Complexes. The
animo-lipid micelles were prepared with a common method
described as follows. Each lipid was dissolved in dichloro-
methane and vacuum dried. An appropriate amount of 50
mM Tris buffer (pH 8.2) was added to hydrate the lipid. We
obtained the lipid dispersion liquid with sonication for 10
min with a UH-50 (SMT Co., Tokyo, Japan). LPS (from
Escherichia coli 055:B5; Sigma-Aldrich, St. Louis, MO) was
We prepared 0.25 mg mL
LPS solution containing the amino-lipid or dioleoyltrimethyl-
ammoniumpropane (DOTAP; Sigma) at indicated concentra-
tions. The ¦-potentials for amino-lipid/LPS complexes were
measured with a Malvern Zetasizer nano (Malvern Instruments,
Malvern, U.K.) at room temperature.
Small-Angle X-ray Scattering (SAXS) Measurement.
SAXS measurements from the amino-lipid/LPS, DOTAP/
LPS, and PMB (Sigma)/LPS solution were carried out at 40B2
SPring-8 with a 1.8 m camera using a Rigaku imaging plate
(30 © 30 cm, 3000 © 3000 pixels) as a detector. The wave-
length of the beam was 1 ¡, and the exposure time was 30 s.
The obtained two-dimensional image was circularly averaged
to give an intensity I(q) vs. q plots, where q is the magnitude of
the scattering vector defined by q = 4³ sinª/- with the scat-
tering angle of 2ª. The concentration of cationic compounds
and LPS were fixed at 4.5 mM and 0.5 mg mL¹1, respectively.
Cytokine Assay. RAW264.7 cells were plated at 1.0 © 105
cells/well in 96-well plates, and incubated for 24 h. To the
cells were added amino-lipid/LPS, DOTAP/LPS, or PMB/
LPS complexes (LPS; 100 ng mL¹1) followed by incubation for
12 h at 37 °C. The concentration of cationic compounds was
fixed at 9.0 ¯M. Tumor necrosis factor (TNF-¡) released into
¹1
dissolved in 50 mM Tris buffer (pH 8.2) at 1 mg mL and
added to the amino-lipid solution at various compositions.
After mixing, the mixture was stored at room temperature
overnight.
Critical Micelle Concentration (CMC) Measurement.
We used 8-anilino-1-naphthalenesulfonic acid (ANS; Wako
Pure Chemical Industries, Ltd., Osaka, Japan) as a fluorescence
probe. After we prepared 2.0 © 10¹5 M ANS solution contain-
ing the amino-lipid at indicated concentrations, we measured

W. Li et al.
Bull. Chem. Soc. Jpn. Vol. 86, No. 5 (2013)
653
OH
O
1-Bromododecane
K2CO3
N
N
N
OH
DMF, 80 ºC
O
1
O
O
LiAlH4
r.t.
Et2O, 0 ºC
H2N
O
O
2
O
O
Boc-Lysine (Boc)-Osu
TEA
NHBoc
H
N
O
Lys-L
H
N
Dichloromethane, r.t.
2
O
BocHN
3A
O
O
O
TFA
NH
2
H
NHBoc
N
H
Dichloromethane, r.t.
N
H N
2
O
BocHN
O
Lys-L
3B
O
O
O
Boc-Arginine (Boc)2-OH
DMT-MM
O
NHBoc
H
H
Arg-L
MeOH, r.t.
BocN
N
N
H2
N
O
O
O
NHBoc
O
O
TFA
NHBoc
NH2
H
N
H
H
N
H
BocN
HN
N
Dichloromethane, r.t.
N
O
O
O
O
Arg-L
NHBoc
NH2
Scheme 1. Synthetic routes of Arg-L and Lys-L.
the supernatants was measured with a Murine TNF-¡ ELISA
Development Kit (R&D System, Inc., Minneapolis, MN).
(Figure 3a). With increase of the cationic lipid, the ¦-potentials
increased for all of the samples and eventually crossed the
zero line at the same concentrations, and further addition of
the cations caused positive charge owing to off-stoichiometry.
Essentially, there was no difference for all samples. These
results indicate that the lipids are interacting with LPS and can
conceal the negative charge of LPS in the same manner for
the quaternary, primary, and guanidine amines.
Results
Preparing Cationic-Lipid/LPS Complexes and the ¦-
Potential. We have been studying several types of aromatic
cationic lipids bearing two alkyl chains and showed that CMC
for these lipids is almost one or two orders of magnitude lower
than conventional cationic lipids such as DOTAP.25 This is
because of the synergetic effects of the hydrophobicity and ³-
stacking of the aromatic ring. Based on our previous studies,
we can suppose that when the chain length is longer than
C12, the aromatic lipids were expected to show a low CMC.
We presume that it is essential for the lipids to maintain
micellar structures to disturb the LPS activity. Figure 2 shows
the lipid concentration dependence of the ANS fluorescence
intensity. Since the fluorescence intensity of ANS sensitively
reflects the polarity of its environment, ANS is suitable for
a probe to determine CMC. The CMC was determined from
the breakpoint on each plot. The obtained CMCs for Arg-L and
Lys-L were almost the same values (2-3 ¯M) and are much
lower than DOTAP (CMC; 70 ¯M).26
LPS has a relatively large molecular weight and contains
many different lengths of polysaccharide chains. To study mix-
tures containing LPS, it may be inappropriate to determine the
molecular weight and then carry out discussion based on the
molar composition. Alternatively, we determined the electro-
statically stoichiometric composition (isoelectric point) at
which the negative charge of LPS is canceled by added cations.
We measured the ¦-potential and the hydrodynamic radius (Rh)
for the mixture for different compositions. LPS alone shows
¹18 mV because of the presence of negative phosphate groups
When carefully examined the lipid concentration depend-
ence, the ¦-potential was changed in a stepwise manner; rapid
increase up-to 0.2 mM and slow increase in 0.2-0.6 mM. The
¦-potential is related to the electrical charge at the interfacial
double layer and does not necessary reflect the total amount of
change of the entire particle. The rapid increase at 0.2 mM may
be ascribed to the added small amount of cations binding to the
surface of the LPS micelle and then cancelling the surface
charge. It can be related to the neutralization due to the ion-pair
formation between 0.2 and 0.6 mM. Therefore, morphological
transition of the micelles may occur in the second step. In
contrast to the zeta potential, Rh did not change (Figure 3b)
and we did not observe the formation of large aggregates at
the isoelectric point, which is normally observed for polyion
complexes. These features can be interpreted in that the solu-
bility of LPS is mainly ascribed to its saccharide portions and
thus elimination of its negative charge does not induce aggre-
gation. Further addition of the cationic lipids above 0.8 mM
showed that Arg-L gave larger positive charge than the others,
simply, relating the difference in the electrophilicity between
the cations.
Structural Changes of LPS Micelle by Addition of the
Cationic Lipids.
Figures 4 and 5b show the angular
dependence of DLS and SAXS for our LPS micelle solution,

654
Bull. Chem. Soc. Jpn. Vol. 86, No. 5 (2013)
LPS Deactivation by Lipid with Amino Acid
a
b
Lys-L
Arg-L
60
50
40
30
20
10
0
50
40
30
20
10
0
3 μM
3 μM
0
0
5
10
15
5
10
15
Cationic Lipid concentration/μM
Cationic Lipid concentration/μM
Figure 2. Arg-L (a) and Lys-L (b) concentration dependency of the ANS fluorescence intensity. The concentration of ANS was fixed
at 2.0 © 10¹5 M.
0
a
40
30
20
10
0
-1
-2
-3
-4
a
30
Arg-L/LPS
Lys-L/LPS
DOTAP/LPS
40
50
60
70
80
90
100
110
120
130
ζ
100
200
300
400x10-6
τ/s
-10
-20
10000
b
y = 2E+07x + 107.7
R2 = 0.9951
0.0
0.2
0.4
0.6
0.8
1.0
1.2
8000
6000
4000
2000
0
Cation lipid concentration/mM
100
80
60
40
20
0
b
Γ
Arg-L/LPS
Lys-L/LPS
DOTAP/LPS
0.E+00
1.E-04
2.E-04
3.E-04
4.E-04
5.E-04
q2/nm-2
Figure 4. The first-order autocorrelation function (a) and
the decay constant (Γ) against q2 (b) for LPS measured
with DLS.
0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
Cation lipid concentration/mM
respectively. As presented in Figure 4a, the first-order auto-
correlation function can be expressed by a single exponential
decay, suggesting monodisperse population. When we plotted
the decay constant (Γ) against q2 (Figure 4b), the data points
Figure 3. ¦-Potential (a) and hydrodynamic radius (b) for
mixtures of LPS (0.25 mg mL¹1) and the indicated cationic
lipid concentrations at 25 °C.

W. Li et al.
Bull. Chem. Soc. Jpn. Vol. 86, No. 5 (2013)
655
103
102
102
101
a
b
-2
Arg-L
Arg-L
101
100
Lys-L
100
10-1
10-2
10-3
10-4
10-5
Lys-L
10-1
10-2
10-3
10-4
-1
DOTAP
PolymyxinB
LPS
DOTAP
5
2
3
10-1
100
5
2
3
10-1
100
q/nm-1
q/nm-1
Figure 5. SAXS profiles of the Arg-L, Lys-L, and DOTAP at 4.5 mM (a) and the complexes composed of LPS (0.5 mg mL¹1) and
the cationic compounds (b). The straight lines show the slopes of ¹2 and ¹1, representing the expected values for vesicle- and
cylinder-like scattering objects, respectively.
can be fitted to a straight line with a nonzero intercept, indi-
cating ¹1 dependence in SAXS observed at low-q, indicating
the presence of cylinder-like structures, being consistent with
the DLS. The present result agrees with previous morpholog-
ical observations by transmission electron microscopy.20,27 The
L. The low-q slope became much steeper and it was almost 3,
suggesting formation of random aggregates. A diffraction peak
appeared at q = 1.1 nm¹1 for both addition of Arg-L and Lys-L,
although Lys-L produced a shaper peak than Arg-L. Because
of the absence of the second diffraction, we were not able to
assign the structure. These SAXS data indicated that amino
acid lipids potently bind to LPS and drastically disturb the
original rod structure of LPS.
¹1
second maximum at q = 1.0 nm in SAXS can be ascribed to
cross-sectional structure of the LPS micelles.
Figure 5a shows SAXS profiles for Arg-L, Lys-L, and
DOTAP micelles, respectively. All of the samples showed
Inhibition of the LPS-Induced TNF-¡ in Macrophages.
It is well-known that RAW264.7 cells secrete a large amount
of TNF-¡ when stimulated by LPS.31 In cells treated with
100 ng mL¹1 of LPS, the TNF-¡ level in supernatant increased
¹2
I(q) µ q at the low-q region. This q-dependence indicates
vesicle formation.28,29 The SAXS profiles after mixing LPS and
each cationic lipid is compared with LPS itself in Figure 5b.
The mixture with PMB showed a very similar profile with
that of LPS itself. There was subtle change observed for the
second maximum; the mixing caused the peak to broaden.
This is probably ascribed to disruption of the uniform cross-
sectional structure due to PMB attaching to the micellar
surface. Brandenburg et al. advocate the relation between the
type of aggregate structure of LPS and the biological activity
by use of SAXS.15-18,30 They measured SAXS from LPS/PMB
mixtures and reported that the mixing led to formation of multi-
lamellae. They measured considerably concentrated solutions,
i.e., 10%, probably owing to very week scattering intensity.
We presume that the difference in the obtained morphology
is due to the concentration effect. Although the data are not
shown, when we added arginine or lysine to LPS, there was no
major structural change.
¹1
to 4.0 ng mL (Figure 6). The addition of the cationic com-
pounds led to a decreased secretion of TNF-¡, but showed
completely different efficiencies. The cells to which PMB/
LPS or DOTAP/LPS complexes were added reduced TNF-¡
secretion less than one-third of that secreted from the cells
treated with LPS alone. Further reductions in TNF-¡ secretion
were observed in the cells to which Arg-L/LPS and Lys-L/LPS
complexes were added. The values were almost tenth-one of
that for LPS treatment. Although we expect higher suppres-
sion by Arg-L than Lys-L because of the high interaction with
phosphate groups of LPS, in this study, we did not observe
significant difference between Arg-L and Lys-L. When we
prepared the complexes, where the concentration for Arg-L
and Lys-L is lower than the CMC (Figure 2), the reductions
in LPS activities were not observed (data not shown). These
results indicate that cationic micelles or compounds binding
LPS cannot fully activate cells and the addition of micellar
form is essential to induce structural change in LPS. Among
these cationic compounds, amino-lipids have higher suppres-
sion ability than commonly used compound PMB.
DOTAP eliminated the second maximum of the original
LPS, suggesting that considerable disruption of the local struc-
ture was induced. On the other hand, the low-q slope was
almost maintained, indicating that rod-like shape still existed.
There were rather drastic changes observed for Arg-L and Lys-

656
Bull. Chem. Soc. Jpn. Vol. 86, No. 5 (2013)
LPS Deactivation by Lipid with Amino Acid
6
5
C. R. H. Raetz, C. Whitfield, Annu. Rev. Biochem. 2002,
71, 635.
6
S. C. Gangloff, N. Hijiya, A. Haziot, S. M. Goyert, Clin.
5
4
3
2
1
0
Infect. Dis. 1999, 28, 491.
7
W. A. Lynn, Y. Liu, D. T. Golenbock, Infect. Immun. 1993,
61, 4452.
8
Z. Jiang, P. Georgel, X. Du, L. Shamel, S. Sovath, S. Mudd,
M. Huber, C. Kalis, S. Keck, C. Galanos, M. Freudenberg, B.
Beutler, Nat. Immunol. 2005, 6, 565.
9
A. Teghanemt, R. L. Widstrom, T. L. Gioannini, J. P. Weiss,
J. Biol. Chem. 2008, 283, 21881.
10 A. Teghanemt, F. Re, P. Prohinar, R. Widstrom, T. L.
Gioannini, J. P. Weiss, J. Biol. Chem. 2008, 283, 1257.
11 S. Liu, R. Tobias, S. McClure, G. Styba, Q. Shi, G.
Jackowski, Clin. Biochem. 1997, 30, 455.
12 C. Hirayama, M. Sakata, J. Chromatogr., B 2002, 781, 419.
13 M. Gough, R. E. Hancock, N. M. Kelly, Infect. Immun.
1996, 64, 4922.
14 S. Bhattacharjya, Curr. Med. Chem. 2010, 17, 3080.
15 K. Brandenburg, H. Mayer, M. H. J. Koch, J. Weckesser,
E. T. Rietschel, U. Seydel, Eur. J. Biochem. 1993, 218, 555.
16 K. Brandenburg, A. Wiese, Curr. Top. Med. Chem. 2004, 4,
1127.
17 K. Brandenburg, I. Moriyon, M. D. Arraiza, G. Lewark-
Yvetot, M. H. J. Koch, U. Seydel, Thermochim. Acta 2002, 382,
189.
18 A. B. Schromm, K. Brandenburg, E. T. Rietschel, H.-D.
Flad, S. F. Carroll, U. Seydel, FEBS Lett. 1996, 399, 267.
19 Y. Zhang, H. Yang, K. Zhou, Z. Ping, React. Funct. Polym.
2007, 67, 728.
+ LPS
Figure 6. Reduction in LPS activity by cationic compounds
on the production of TNF-¡. RAW264.7 cells were stimu-
lated with LPS with or without complexation for 12 h
at 37 °C. Results represent the mean « S.D. of triplicate
wells.
Discussion and Concluding Remarks
When the micelles of Arg-L and Lys-L were mixed with
LPS, drastic structural changes occurred and antagonistic cell
activation by LPS was reduced. According to previous studies,
the role of PMB or other LPS binding peptides in deactivating
LPS is to block the signal transduction through TLRs or ion
channels owing to the structural transition of LPS.15 Accepting
this interpretation, Arg-L and Lys-L more radically disturbed
the original LPS structures than PMB and thus showed more
efficient deactivation of LPS.
20 Y. Kaconis, I. Kowalski, J. Howe, A. Brauser, W. Richter, I.
Razquin-Olazarán, M. Iñigo-Pestaña, P. Garidel, M. Rossle, G. M.
de Tejada, T. Gutsmann, K. Brandenburg, Biophys. J. 2011, 100,
2652.
21 A. Shanmugam, S. Rajoria, A. L. George, A. Mittelman, R.
Suriano, R. K. Tiwari, PLoS ONE 2012, 7, e30839.
22 S. Mochizuki, Y. Kamikawa, K. Nishina, S. Fujii, E.
Hamada, S. Kusuki, T. Matsuo, K. Sakurai, Bull. Chem. Soc. Jpn.
2012, 85, 354.
23 K. Koiwai, K. Tokuhisa, R. Karinaga, Y. Kudo, S. Kusuki,
Y. Takeda, K. Sakurai, Bioconjugate Chem. 2005, 16, 1349.
24 T. Nishimura, T. Matsuo, K. Sakurai, Phys. Chem. Chem.
Phys. 2011, 13, 15899.
25 T. Matsuo, S. Fujii, Y. Kamikawa, T. Nishimura, Y. Sanada,
S. Mochizuki, K. Sakurai, Chem. Lett. 2011, 40, 1340.
26 N. J. Zuidam, Y. Barenholz, Int. J. Pharm. 1999, 183, 43.
27 Y. Rosenfeld, H.-G. Sahl, Y. Shai, Biochemistry 2008, 47,
6468.
This work is financially supported by JST CREST program
and SAXS measurements were carried out at SPring-8 40B2.
Supporting Information
1
The synthetic details and H spectra of Lys-L and Arg-L.
This material is available free of charge on the Web at: http://
www.csj.jp/journals/bcsj/.
28 R.-J. Roe, Methods of X-Ray and Neutron Scattering in
Polymer Science, Oxford University Press, 2000.
29 O. Glatter, O. Kratky, Small Angle X-ray Scattering,
Academic Press, London, 1982.
References
1
R. Medzhitov, P. Preston-Hurlburt, C. A. Janeway, Jr.,
Nature 1997, 388, 394.
2
R. Shimazu, S. Akashi, H. Ogata, Y. Nagai, K. Fukudome,
30 K. Brandenburg, M. H. J. Koch, U. Seydel, Eur. J.
Biochem. 1998, 258, 686.
K. Miyake, M. Kimoto, J. Exp. Med. 1999, 189, 1777.
3
4
K. Miyake, J. Endotoxin Res. 2006, 12, 195.
C. R. H. Raetz, Annu. Rev. Biochem. 1990, 59, 129.
31 B. Beutler, N. Krochin, I. W. Milsark, C. Luedke, A.
Cerami, Science 1986, 232, 977.