| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Research Center Borstel, Center for Medicine and Biosciences, Department of Immunochemistry and Biochemical Microbiology, Borstel, Germany;
Martin-Luther-Universität Halle-Wittenberg, Kardiologische Intensivmedizin, Forschungslabor, Halle (Saale), Germany;
XOMA Corporation, Berkeley, CA 94710;
§
European Molecular Biology Laboratory c/o DESY, Hamburg, Germany; and
¶
Department of Chemistry, Graduate School of Science, Osaka University, Toyonaka, Osaka, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Endotoxins are amphiphilic molecules, and therefore tend to form multimeric aggregates above a critical aggregate concentration (CAC) that depends on their hydrophobicity. Above this concentration, there is an equilibrium between free monomers and aggregates. The CAC, which determines the number of free monomers, is supposed to be influenced by the amount of sugars linked to the lipid A portion of LPS as well as by the number and distribution of charges. Thus, the CAC should be highest for S-form LPS of wild-type strains, with high amounts of sugars represented by the O-specific chain, decreasing with the amount of sugars in LPS from rough mutant and deep-rough mutant strains, and be lowest for free lipid A.
The structure of the aggregates formed above the CAC can be either nonlamellar inverted (cubic, Q, or hexagonal, HII) or lamellar, depending on the conformation (shape) of the contributing molecules (6). From the aggregate structure, the conformation of the individual molecules can be deduced: it is conical or wedge-shaped, with the hydrophobic moiety adopting a higher cross-section than the hydrophilic in the case of nonlamellar inverted and cylindrical, with the respective cross-sections being nearly identical in the case of lamellar aggregate structures. We have shown previously that a particular endotoxic conformation of the lipid A portion of endotoxin is a prerequisite for the expression of endotoxic activity. Thus, only those endotoxins having a lipid A portion that adopts a nonlamellar HII or cubic aggregate structure in the isolated form, i.e., having a conical conformation, were biologically highly active (7, 8).
Several mechanisms of interaction of endotoxin with host cells have been proposed. There is experimental evidence that endotoxin molecules, similar to other amphiphilic molecules, interact unspecifically with target membranes via hydrophobic interaction. This process may cause a direct intercalation of endotoxin monomers or small aggregates into the phospholipid matrix of host cells (9), leading to either activation or neutralization. Another mechanism of interaction has been described to proceed via the specific binding of LPS to a membrane-bound receptor protein, the differentiation Ag CD14 of myelomonocytic cells, or via binding to the serum protein LPS-binding protein (LBP) and the endotoxin/LBP complex then binding to CD14 (10, 11, 12, 13, 14). In case of CD14-negative endothelial and epithelial cells, a soluble form of CD14 has been shown to mediate cellular activation by LPS (15, 16). There are also several investigations showing a CD14-independent cell activation, in particular at high endotoxin concentrations (17, 18, 19, 20), but the details of this activation pathway are largely unknown. A CD14-independent, LBP-mediated transport of endotoxin molecules from aggregates directly into phospholipid membranes could recently be elucidated (21). Thus, LBP might act as a shuttle, causing a disaggregation of endotoxin aggregates by transporting the molecules directly or indirectly to target cell membranes. We and others could show that the transport activity of LBP is not restricted to endotoxin, but is also valid for negatively charged phospholipids (21, 22). Nevertheless, the role of charges for this activity of LBP remains to be defined. Independent of the detailed mechanism, however, it may be expected that the unspecific intercalation as well as the specific binding of LPS should depend on physical parameters like CAC, size, and stability of endotoxin aggregates, and the molecular conformation of the molecules. Many investigations have been reported on the interaction of LPS with CD14 and its amplifiying role in host cell activation (23, 24, 25, 26), but only very limited data are available on the direct interaction of LPS with the target cell membrane (27, 28, 29).
In this work, we report on an approach to more closely characterize the interaction mechanism of endotoxin with target cell membranes, in particular the role of charges. The experiments are mainly focused on lipid A as the endotoxic principle and lipid A derivatives differing in the number, location, and nature of negative charges. LPS Re of E. coli and its dephosphorylated derivative were also included for a further variation of charges in the direct vicinity of lipid A. Except LPS Re of E. coli, none of the endotoxin derivatives used in these investigations are generated by intact, living bacteria. They are merely applied to obtain insight into the principles of endotoxin recognition, transport, intercalation, conformation, and biologic activity. The influence of the variations in molecular charge on the molecular conformation was derived from synchrotron small-angle x-ray diffraction data of the supramolecular structure of endotoxin aggregates in aqueous dispersion. Their influence on the LBP-mediated transport was investigated by determining the intercalation of endotoxin molecules into phospholipid liposomes resembling in their lipid composition the cytoplasmic membrane of macrophages by resonance energy transfer spectroscopy (RET). The biologic activity of the samples was assessed from the determination of the IL-6 production of endotoxin-stimulated PBMCs or ex vivo whole blood.
We found a strong modulating effect of the number, location, and nature of charges of endotoxin molecules on the molecular conformation, the interaction with LBP, and the biologic activity. According to our findings, the presence of a net negative charge is a necessary, but not a sufficient, prerequisite for the LBP-mediated transport and intercalation of endotoxin into phospholipid membranes. For the expression of biologic activity, besides a net negative charge, a particular chemical structure and endotoxic conformation are essential prerequisites.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Bacteria from the different strains were cultured under standard conditions usually at 37°C, for the phototropic strain Rhodospirillum fulvum (DSM117 obtained from the Deutsche Stammsammlung von Mikroorganismen, Braunschweig, Germany) in light (100 lx) at 30°C, and LPS was extracted from the phenol-killed bacteria. The LPS of the deep rough (Re) mutant of E. coli strain F515 were obtained by a modified PCP (30) procedure (PCP I, i.e., phenol/chloroform/petroleum ether 2:5:8 vol %), those from the smooth strain Salmonella enterica serovar Friedenau were obtained by the phenol/water extraction method (31). For the phototropic strain R. fulvum, phenol/water extraction was followed by ultracentrifugation at 105,000 x g for 4 h. The LPS in the water phase was subsequently enzyme treated with proteinase K, RNase, and DNase (Sigma, Deisenhofen, Germany; and Boehringer Mannheim, Mannheim, Germany), and further purified by PCP extraction (PCP II, i.e., phenol/chloroform/petroleum ether 5:5:8 vol %). The resulting LPS were lyophilized, and are thus present in their natural salt form.
For the preparation of dephosphorylated LPS (dephosphorylated LPS Re),
750 mg LPS Re from E. coli strain F515 was suspended in 20
ml aqueous hydrofluoric acid (48%) for 3 days at 4°C under stirring.
The reaction mixture was dialyzed extensively three times against 5 L
water and lyophilized, giving 600 mg crude dephosphorylated LPS. One
aliquot (390 mg) of this dephosphorylated LPS was purified on 10
preparative layer plates (2 mm, Kieselgel 60 from Merck, Darmstadt,
Germany) and developed with chloroform:methanol:water (100:87.5:22.5
v/v). The resulting fractions could be visualized by dipping the dried
plates into water and recovered by scraping off the white opalescent
bands. Fractions with an Rf value 
0.5 in the analytical
TLC (same solvent) were found to be the major product and collected (90
mg). For final purification, the dephosphorylated LPS was
rechromatographed in three successive runs on one preparative layer
plate using chloroform:methanol:water (100:55:7 v/v). Pure
dephosphorylated LPS was visualized as stated above, expressing a
single distinct band that, after elution from the plates, could be
obtained in a satisfactory yield (27 mg). The resulting purified
dephosphorylated LPS was analyzed with a MALDI-TOF mass spectrometer
(Reflex II; Bruker, Karlsruhe, Germany) as well as 1H and
13C NMR (600 MHz Avance DRX; Bruker) and gave spectra fully
compatible with phosphate-free LPS Re (molecular mass of
[M+Na]+ = 2098 Da).
Lipid A was obtained from LPS Re of E. coli F515 or from LPS
of R. fulvum DSM 117 either by sodium acetate buffer (0.1 M,
pH 4.4, 100°C for 1 h) or by HCl (0.1 M, 100°C for 0.5 h)
treatment, purified, and converted to the triethylamine salt. The
resulting preparations differ in their degree of 2-keto-3-deoxyoctonate
(Kdo) and phosphate cleavage; in the acetate buffer treatment, Kdo is
cleaved to 90–95%, whereas with HCl, Kdo is removed quantitatively,
as well as the 1-phosphate. Thus, by the latter procedure, it is
possible to obtain monophosphoryl lipid A. As shown by MALDI-TOF mass
spectrometry, both the bis- and the monophosphoryl lipid A were mainly
hexaacylated (85%) and contained only minor fractions of penta- and
tetraacylated structures (10 and 5%, respectively). Both
compounds were used without further purification. As phosphate-free
lipid A, the hexaacyl E. coli-lipid A analog (compound 503),
synthesized as described earlier (32), was used.
The LPS Re of E. coli was in the natural salt form, and all other preparations were converted to the triethylamine salt form, except compound 503, which is uncharged and does not contain counterions.
The deacylated backbone of lipid A was derived from Re LPS of S. enterica serovar Minnesota strain R595, as described (33). Briefly, lipid A was de-O-acylated with hydrazine and subsequently de-N-acylated with KOH (4 M, 120°C for 16 h), followed by neutralization with HCl (4 M, 0°C). The free fatty acids were extracted with CH2Cl2, and the water phase was purified by gel-permeation chromatography and lyophilized.
All LPS and lipid A samples were analyzed chemically with respect to the amount of phosphate, Kdo, and GlcN with standard procedures (34), and also the content of fatty acid residues was monitored by MALDI-TOF mass spectrometry. Only those batches that were identical to the known chemical structures were used.
The chemical structures of the various LPS and lipid A samples are
schematically shown in Fig. 1
. Lipid A
from LPS of R. fulvum has, instead of phosphate groups, a
galacturonic acid in 1 position and a
D-manno-heptose in the 4' position (35). The
main variations of the chemical structures used in this study and shown
in Fig. 1 consist in the number and distribution of negative charges,
the former ranging from four for LPS Re to two for dephosphorylated LPS
Re and lipid A from LPS Re, one for monophosphoryl lipid A from LPS Re
and lipid A from LPS of R. fulvum to none for the
phosphate-free lipid A.
|
Phospholipids
Bovine brain 3-sn-phosphatidylserine, egg
3-sn-phosphatidylcholine, and sphingomyelin from bovine
brain were obtained from Sigma.
3-sn-phosphatidylethanolamine from E. coli was
from Avanti Polar Lipids (Birmingham, AL). For preparation of liposomes
from the phospholipid mixture resembling the composition of the cell
membrane of macrophages (PLM
),
3-sn-phosphatidylcholine,
3-sn-phosphatidylserine,
3-sn-phosphatidylethanolamine, and sphingomyelin in a molar
ratio of 1:0.4:0.7:0.5 (36), the lipids were solubilized in chloroform,
the solvent was evaporated under a stream of nitrogen, and the lipids
were resuspended in the appropriate volume of PBS and further treated
as described for LPS.
X-ray diffraction
X-ray diffraction measurements were performed as described (37)
at the European Molecular Biology Laboratory outstation at the
Deutsches Elektronen Synchrotron (DESY) in Hamburg using the
double-focusing monochromator-mirror camera X33 (38). In the
diffraction patterns presented in this work, the logarithm of the
diffraction intensity log I is plotted vs the scattering vector s
(s = 2sin
/n
; 2, scattering angle; = 0.15 nm,
wavelength), and the x-ray spectra were evaluated as described
previously (37, 39). Briefly, from the spacing ratios of the
diffraction maxima, an assignment to defined three-dimensional
aggregate structures is possible, i.e., to lamellar, nonlamellar cubic,
and inverted hexagonal II. From this, the conformation of the
individual molecules can be approximated (8, 40), which is cylindrical
in the case of lamellar structures (the cross-sections of the
hydrophilic and hydrophobic moieties are identical) and conical or
wedge-shaped in the case of nonlamellar cubic and inverted
HII structures (the cross-section of the hydrophobic
portion is higher than that of the hydrophilic one).
RET spectroscopy
The RET assay was performed as described earlier (21). Briefly,
phospholipid liposomes PLM were double labeled with the
fluorescent dyes
N-(7-nitrobenz-2-oxa-1,3-diazol-4yl)-phosphatidylethanolamine
(NBD-PE) and N-(lissamine rhodamine B
sulfonyl)-phosphatidylethanolamine (Rh-PE) (Molecular Probes, Eugene,
OR). Intercalation of unlabeled molecules into the double-labeled
liposomes leads to probe dilution, and with that to a decrease in the
efficiency of RET: the emission intensity of the donor increases and
that of the acceptor decreases (for the sake of clarity, only the donor
emission intensity is shown in this work). The double-labeled
PLM liposomes were preincubated with unlabeled LPS, and
human rLBP (41) was added. For a calibration of the magnitude of the
intercalation of the endotoxins into the phospholipid liposomes, common
preparations of PLM and endotoxins at given molar ratios
were prepared by dissolving them, together with constant amounts of the
dyes, in chloroform, evaporating the solvent under a stream of
nitrogen, resuspending the lipid mixture in buffer, and recording the
donor emission intensity.
IL-6 induction
For the isolation of PBMCs, blood was taken from healthy donors and heparinized (20 IE/ml). The heparinized blood was mixed with an equal volume of HBSS and centrifuged on Ficoll density gradient for 40 min (21°C, 500 x g). The interphase layer of PBMCs was collected and washed three times in serum-free RPMI 1640 containing 2 mM L-glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin. The cells were resuspended in serum-free medium and the cell number was equilibrated at 5 x 106 N/ml.
For stimulation, 200 µl/well heparinized whole blood or PBMCs (5 x 106 N/ml) were filled into 96-well culture plates. Whole blood or PBMCs in serum-free medium were stimulated with the endotoxins. The stimuli were serially diluted in RPMI 1640 and added to the cultures at 20 µl/well. The cultures were incubated for 24 h at 37°C and 5% CO2. Supernatants were collected after centrifugation of the culture plates for 10 min at 400 x g and stored at -20°C until determination of cytokine content.
For IL-6 determination, the murine cell line 7TD1 was applied (42). The culture supernatants or human rIL-6 as standard (10 ng/ml) were diluted serially in 1:4 steps in microtiter plates in 50 µl 7TD1 medium (DMEM plus 4.5 g/L glucose, 10% FCS, 2 mM glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin). The cells were washed twice (10 min, 300 x g) in 7TD1 medium, and adjusted to 8 x 104 cells/ml. An aliquot (50 µl) of this cell suspension was incubated with the supernatants or rIL-6 as standard, respectively, and incubated 72 h at 37°C in 5% CO2-containing atmosphere. For the determination of the cell proliferation, the mitochondrial reduction of 3-(4, 5-dimethyltiazol-2yl)-2,5-diphenyltetrazoliumbromide (MTT) to formazane was measured. Briefly, 10 µl MTT (5 mg/ml in PBS) was added to the cells that were incubated for 4 h with subsequent addition of 100 µl SDS solution (5% in 50% dimethylformamide). After 2-h incubation at room temperature, the absorbance at 550 nm was measured in an ELISA plate photometer. The absolute IL-6 content was determined with respect to the standards.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The structure of endotoxin aggregates in aqueous dispersions is
primarily determined by the conformation of the monomers. Therefore,
the conformation of the monomers can be derived from the supramolecular
aggregate structure determined by x-ray diffraction. The x-ray
diffraction patterns of LPS Re (Fig. 2A) and bisphosphoryl lipid A
from LPS Re (Fig. 2C) are characteristic of nonlamellar
cubic structures with different symmetries that are frequently observed
for enterobacterial lipid A and for LPS Re at high water content and in
the absence of divalent cations. This can be deduced from the positions
of the diffraction maxima that are grouped in spacing ratios
corresponding to cubic structures of space group Q212 (LPS
Re) and Q224 (lipid A from LPS Re). The pattern of
monophosphoryl lipid A from LPS Re (Fig. 2D) indicates a
superposition of a multilamellar with a nonlamellar structure. However,
the latter is not resolvable in a straightforward way. Dephosphorylated
LPS Re (Fig. 2B) as well as lipid A from R.
fulvum (Fig. 2E) unequivocally adopt (multi)lamellar
structures characterized by sharp reflections at equidistant spacing
ratios. The diffraction patterns of phosphate-free lipid A (synthetic
compound 503, Fig. 2F) indicate the existence of a normal
and an interdigitated bilayer structure.
|
LBP-mediated transport of endotoxins into phospholipid membranes
RET spectroscopy was used to investigate the ability of LBP to
transport the various endotoxins into a phospholipid membrane
corresponding to the composition of the macrophage cell membrane
(PLM). The results are shown in Fig. 3. It can clearly be seen that addition
of LBP induces an increase of the NBD fluorescence intensity in
PLM membranes for all phosphate- or carboxylate-bearing
endotoxins. The main change in NBD-fluorescence intensity occurs within
about 10 s, followed by a longer period (some minutes) with an
only gradually increasing intensity. These observations can be
interpreted to result from an intercalation of endotoxin molecules into
the liposomal membrane. In contrast, the intercalation of lipid A from
LPS of R. fulvum carrying one net negative charge and of
phosphate-free and thus neutral lipid A into the liposomal membrane is
negligible.
|
membranes was also examined. In this case, no increase in fluorescence
intensity, but a small decrease due to dilution of the suspension, was
observed.
From a comparison of the intercalation behavior of the different
compounds, it may be concluded that there is no direct correlation of
the efficiency of LBP-mediated transport into the liposomal membrane to
the number of negative charges. Interestingly, lipid A from LPS of
R. fulvum, in which the 1 and 4' phosphates are substituted
by a heptose and a galacturonic acid, respectively, is intercalated
into the PLM membrane only to a very low degree despite
its effective negative charge. The effective charge of the carboxyl
group of this lipid A was determined by comparing the infrared spectrum
in the 1500–1400 cm-1 range comprising the scissoring
mode of the methylene groups (at 1467 cm-1) and the
stretching vibration of the 
s(COO-) group
(at 1415–1427 cm-1) with that of LPS Re (two negatively
charged carboxyl groups) and of lipid A from LPS Re (no carboxyl
groups). The spectra were very similar for lipid A from LPS of R.
fulvum and for LPS Re, exhibiting an intense absorption band
corresponding to s(COO-) that was
completely absent in the lipid A isolated from LPS Re (data not shown).
Thus, lipid A from LPS of R. fulvum can be assumed to be
present in the fully dissociated form, i.e., the carboxyl group carries
one net negative charge.
IL-6 induction in whole blood and PBMCs
As a measure for the biologic activity of the lipid A and LPS
samples, their ability to induce IL-6 in ex vivo whole human blood was
determined. IL-6 is, besides IL-1 and TNF-
, one of the important
mediators induced by endotoxin. The results are shown in Fig. 4A for endotoxin
concentrations in the range 0.001–1000 ng/ml and are compared with the
data obtained for LPS from Salmonella friedenau
(S-form LPS) used as a calibration standard. Clearly, LPS Re has the
strongest IL-6-inducing capacity, which is nearly identical to that of
LPS S-form from S. friedenau. Lipid A from LPS Re is, in the
medium concentration range 0.1–10 ng/ml, one order of magnituide less
active, followed by the monophosphoryl lipid A and dephosphorylated LPS
Re, which are nearly three orders of magnitude less active. Lipid A
from LPS of R. fulvum displays very weak activity. The
synthetic phosphate-free lipid A is completely inactive.
|
B, show
that the reference LPS S-form from S. friedenau and LPS Re
behaves in a way similar to that in the whole blood assay regarding the
amount of induced IL-6 as well as the lowest endotoxin concentration
required to provoke a measurable effect (0.01 ng/ml). However, in
the serum-free system a steep increase in IL-6 production at endotoxin
concentrations in the range 0.01–0.1 ng/ml and a subsequent plateau up
to the highest concentrations (saturation) are observed. In contrast,
in the serum-containing system there is a gradual increase over a wide
range of endotoxin concentrations.
In Fig. 5, the various physicochemical
parameters (molecular charge, molecular conformation, and LBP-mediated
intercalation) and the biologic data (amount of endotoxin necessary to
induce 1 ng/ml IL-6 in whole blood or PBMCs) are summarized to allow a
direct comparison and correlation.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). In these investigations,
particular attention has been paid to a consequent variation of the
chemical structure of the endotoxins. Thus, all derivatives differing
in the net charge have been obtained from the same LPS batch (LPS Re of
E. coli F515), and the synthetic compound 503 corresponds to
the dephosphorylated lipid A component of this LPS. Lipid A from LPS of
R. fulvum was used, because its only negative charge is
represented by the galacturonic acid. The physical measurements
comprised the determination of the supramolecular structure of
endotoxin aggregates and the conformation of the molecules derived
therefrom and the LBP-mediated transport of endotoxin into liposomal
phospholipid membranes corresponding to the composition of the
macrophage cytoplasmic membrane. For the following discussion of the influence of the molecular charge on physicochemical properties of endotoxins and their IL-6-inducing capacity, it should be considered that the judgment of the separate influence of one of the parameters, molecular charge, molecular conformation, and LPB-mediated intercalation, is hampered by the interdependence between them.
There is increasing evidence that the biologic activity of endotoxin is mediated by oligomeric endotoxin units down to monomers (43). Therefore, it may be assumed that the structure and the stability of endotoxin aggregates influence not only the equilibrium between aggregates and monomers, but also the interaction of endotoxin with LBP, which is supposed to shuttle endotoxin to CD14 or soluble CD14, lipoprotein particles (14, 44, 45), or phospholipid membranes (21, 22, 46). Endotoxin monomers, regardless whether existing per se or being liberated from aggregates by the action of LBP, may interact with the various endotoxin-binding proteins or may intercalate into host cell membranes, thus causing a further release of monomers from the aggregates to maintain the equilibrium, and thereby keeping the process in progress.
It was found previously that the molecular conformation adopted by the lipid A component of endotoxin molecules influences the biologic activity. Thus, highest activity was found for lipid A (and the corresponding LPS) with a conical shape, whereas samples adopting a cylindrical conformation were largely inactive (6). In some cases, the data from physical measurements were correlated with those from biologic experiments performed in other laboratories with other batches of endotoxins, the biologic systems comprising a variety of different activities such as pyrogenicity in rabbits, lethal toxicity in mice, and cytokine induction in vitro and in vivo using monocytes and endothelial cells (47, 48, 49). In the present investigations, we correlated physicochemical parameters of the same samples of clearly defined lipid A and LPS Re directly with their ability to induce IL-6 production in human ex vivo whole blood or PBMCs.
According to Fig. 2, lipid A from LPS Re of E. coli F515 in
bound or free form adopts a nonlamellar inverted structure. Therefore,
the single constituting molecules should have a conical geometry, and
should thus be biologically active. Phosphate-free lipid A (compound
503) and lipid A from LPS of R. fulvum adopt lamellar
supramolecular structures due to the cylindrical geometry of the
monomers, and they should, therefore, be largely biologically inactive.
The same should be expected for the lipid A component of
dephosphorylated LPS Re. These predictions are confirmed by biologic
assays (Fig. 4, A and B). The remaining
bioactivity of dephosphorylated LPS Re can be explained by its tendency
to adopt nonlamellar structures that are in fact observed at
temperatures >40°C and that give evidence for the influence of the
charges of the Kdo on the conformation of the lipid A moiety. From
these results, it may be concluded that a conical conformation of the
lipid A component with a higher cross-section of the hydrophobic than
of the hydrophilic part is an important prerequisite for the induction
of high IL-6 amounts.
Another interesting aspect of the present results arises from the
observation that the removal of the glycosidic phosphate group of lipid
A from LPS Re leads to a transition from a pure cubic into a mixed
lamellar/cubic, and the removal of both phosphates (corresponding to
synthetic compound 503) to a complete lamellarization with two
coexisting systems of different periodicity (Fig. 2). Because the
hydrophobic moiety is the same for all three samples, the formation of
lamellar systems is connected with an increase of the space requirement
of the hydrophilic moiety. This implies that the simple geometric
consideration, that the space requirement of the hydrophilic moiety
should be reduced by the removal of the phosphates, is inappropriate.
Rather, the space-reducing effect of counterions bridging phosphate
groups of neighboring molecules, leading to a reduction of the backbone
cross-section, is in dephosphorylated or phosphate-free compounds
ineffective.
The molecular charge also exerts a modifying effect on the LBP-induced
transport and intercalation into liposomal membranes
(PLM). This is apparent from the evaluation of the RET
data (Fig. 3) that clearly indicate strong differences for endotoxins
with different chemical structures. Because control experiments with
the deacylated lipid A backbone consisting of a bisphosphorylated
diglucosamine displayed no increase of the fluorescence signal at all,
a pure adsorption or association of the endotoxin molecules to the
liposome surface can be excluded. From Fig. 3, it becomes obvious that
the LBP-mediated transport and intercalation into PLM
liposomes are comparable for those endotoxin structures that carry two
or more negative charges (curves 1–3). From a comparison of the
behavior of dephosphorylated LPS Re (curve 3) with two negatively
charged carboxylate groups and free lipid A form LPS Re (curve 1)
carrying two negatively charged phosphates, it could be concluded that
phosphate groups are not required for the interaction of LPS with LBP.
However, this simple picture is complicated by a comparison of the
transport efficacy of the two endotoxins carrying one negative charge,
monophosphoryl lipid A from LPS Re (curve 4), and lipid A from LPS of
R. fulvum (curve 6), the former being intercalated into the
liposomal membrane at a significantly higher rate than the latter. This
implies that at least in those cases in which the only effective charge
is located at the lipid A backbone, the nature of the functional groups
representing the negative charge is critical for LBP-mediated
transport. Furthermore, these examples demonstrate that the nature of
the functional group representing the negative charge influences the
molecular conformation. Monophosphoryl lipid A from LPS Re with the
only phosphate group present in 4' position assumes a weakly developed
conical geometry, whereas the lipid A from LPS of R. fulvum,
which carries two voluminous sugars, in 1 position a galacturonic acid
and in 4' position a heptose, assumes a pure cylindrical conformation.
These considerations are backed by first unpublished results on the
IL-6-inducing capacity of a derivative of the synthetic E.
coli lipid A (compound 506) in which the 1 phosphate is replaced
by a carboxymethyl group. This derivative displayed activity similar to
that of the corresponding bisphosphorylated lipid A in a whole blood
assay. These data are, furthermore, indicative of a conical
conformation of this lipid A analogue.
At least in the absence of serum proteins, the spontaneous
intercalation of monomers into the macrophage membrane is likely to be
part of the activation process. The extent of this intercalation should
be proportional to the number of monomers available, which increases
with increasing length of the hydrophilic LPS region (higher CAC). Such
a dependence is impressively apparent from the data depicted in Fig. 4B, which show increasing IL-6-inducing capacity in the
sequence phosphate-free lipid A < dephosphorylated LPS Re and
lipid A from LPS Re < LPS Re < LPS S-form. In particular,
the almost identical biologic activities of dephosphorylated LPS Re and
lipid A from LPS Re are in agreement with this interpretation, because
it seems reasonable to roughly assume a similar value for their CAC,
both preparations having two negative charges in the backbone and an
identical lipid component. In addition, the observation that the
molecular conformation of LPS Re is identical to that of lipid A from
LPS Re, and also that of dephosphorylated LPS Re is identical to that
of phosphate-free lipid A, but the respective biologic activities
differ, can be explained by the influence of the number of available
monomers. Furthermore, the results for lipid A from LPS Re indicate
that under serum-free conditions a conical conformation is a necessary
but not sufficient condition for the expression of high biologic
activity because cell activation is only possible with a sufficient
number of monomers.
In this context, it might be interesting to discuss briefly the background of the missing biologic activity of some bisphosphorylated lipid A and lipid A part structures, in particular the lipid A from LPS of Rhodobacter capsulatus and lipid A precursor IVa (synthetic compound 406). For the former compound, we have determined a cylindrical conformation (40), which readily explains the lacking biologic activity, irrespective of the higher CAC value due to a reduced hydrophobicity (only five fatty acid residues at the lipid A backbone). This argumentation should hold even more for compound 406, which carries only four fatty acids.
In conclusion, our data show that in addition to a particular chemical structure such as that found in enterobacterial strains and a unique endotoxic conformation, the presence of negative charges in endotoxin molecules is a further prerequisite for biologic activity. Even though, at present, it cannot be excluded that further parameters influence endotoxin biologic activity, the influence of the various physicochemical parameters investigated in this study and their interdependences allows us to understand the sequence of the biologic activities of the various endotoxins and endotoxin derivatives.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Ulrich Seydel, Department of Immunochemistry and Biochemical Microbiology, Research Center Borstel, Center for Medicine and Biosciences, Parkallee 10, D-23845 Borstel, Germany. E-mail address:
3 Abbreviations used in this paper: GlcN, D-glucosamine; CAC, critical aggregate concentration; HII, hexagonal II; Kdo, 2-keto-3-deoxyoctonate; LBP, LPS-binding protein; MALDI-TOF, matrix-assisted laser desorptional ionization time of flight; MTT, 3-(4, 5-dimethyltiazol-2yl)-2,5-diphenyltetrazoliumbromide; NBD, N-(7-nitrobenz-2-oxa-1, 3-diazol-4yl); PLM, phospholipid mixture resembling the lipid composition of a macrophage membrane; Q, cubic; RET, resonance energy transfer.
Received for publication April 13, 1998. Accepted for publication July 13, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, and acute lethality of mice by toxic and non-toxic forms of lipid A. J. Immunol. 141:870.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  C. M. John, M. Liu, and G. A. Jarvis Natural Phosphoryl and Acyl Variants of Lipid A from Neisseria meningitidis Strain 89I Differentially Induce Tumor Necrosis Factor-{alpha} in Human Monocytes J. Biol. Chem., August 7, 2009; 284(32): 21515 - 21525. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. E. McConnell, J. N. Higginbotham, D. A. Shifrin Jr., D. L. Tabb, R. J. Coffey, and M. J. Tyska The enterocyte microvillus is a vesicle-generating organelle J. Cell Biol., June 29, 2009; 185(7): 1285 - 1298. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Tuin, K Poelstra, A de Jager-Krikken, L Bok, W Raaben, M P Velders, and G Dijkstra Role of alkaline phosphatase in colitis in man and rats Gut, March 1, 2009; 58(3): 379 - 387. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Zughaier, B. Lindner, J. Howe, P. Garidel, M. H.J. Koch, K. Brandenburg, and D. S. Stephens Physicochemical characterization and biological activity of lipooligosaccharides and lipid A from Neisseria meningitidis Innate Immunity, December 1, 2007; 13(6): 343 - 357. [Abstract] [PDF]
|
|
|
|
|
|
M.-A. Newman, J. M. Dow, A. Molinaro, and M. Parrilli Invited review: Priming, induction and modulation of plant defence responses by bacterial lipopolysaccharides Innate Immunity, April 1, 2007; 13(2): 69 - 84. [Abstract] [PDF]
|
|
|
|
|
|
J. Andra, T. Gutsmann, P. Garidel, and K. Brandenburg Invited review: Mechanisms of endotoxin neutralization by synthetic cationic compounds Innate Immunity, October 1, 2006; 12(5): 261 - 277. [Abstract] [PDF]
|
|
|
|
|
|
A. Silipo, A. Molinaro, L. Sturiale, J. M. Dow, G. Erbs, R. Lanzetta, M.-A. Newman, and M. Parrilli The Elicitation of Plant Innate Immunity by Lipooligosaccharide of Xanthomonas campestris J. Biol. Chem., September 30, 2005; 280(39): 33660 - 33668. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. X. Tran, M. J. Karbarz, X. Wang, C. R. H. Raetz, S. C. McGrath, R. J. Cotter, and M. S. Trent Periplasmic Cleavage and Modification of the 1-Phosphate Group of Helicobacter pylori Lipid A J. Biol. Chem., December 31, 2004; 279(53): 55780 - 55791. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Tsukushi, N. Kido, K. Saeki, T. Sugiyama, N. Koide, I. Mori, T. Yoshida, and T. Yokochi Characteristic biological activities of lipopolysaccharides from Sinorhizobium and Mesorhizobium Innate Immunity, February 1, 2004; 10(1): 25 - 31. [Abstract] [PDF]
|
|
|
|
 |
|
 S. M. Zughaier, Y.-L. Tzeng, S. M. Zimmer, A. Datta, R. W. Carlson, and D. S. Stephens Neisseria meningitidis Lipooligosaccharide Structure-Dependent Activation of the Macrophage CD14/Toll-Like Receptor 4 Pathway Infect. Immun., January 1, 2004; 72(1): 371 - 380. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Brandenburg, P. Garidel, J. Andra, G. Jurgens, M. Muller, A. Blume, M. H. J. Koch, and J. Levin Cross-linked Hemoglobin Converts Endotoxically Inactive Pentaacyl Endotoxins into a Physiologically Active Conformation J. Biol. Chem., November 28, 2003; 278(48): 47660 - 47669. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Schwudke, M. Linscheid, E. Strauch, B. Appel, U. Zahringer, H. Moll, M. Muller, L. Brecker, S. Gronow, and B. Lindner The Obligate Predatory Bdellovibrio bacteriovorus Possesses a Neutral Lipid A Containing {alpha}-D-Mannoses That Replace Phosphate Residues: SIMILARITIES AND DIFFERENCES BETWEEN THE LIPID As AND THE LIPOPOLYSACCHARIDES OF THE WILD TYPE STRAIN B. BACTERIOVORUS HD100 AND ITS HOST-INDEPENDENT DERIVATIVE HI100 J. Biol. Chem., July 18, 2003; 278(30): 27502 - 27512. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. S. Van Amersfoort, T. J. C. Van Berkel, and J. Kuiper Receptors, Mediators, and Mechanisms Involved in Bacterial Sepsis and Septic Shock Clin. Microbiol. Rev., July 1, 2003; 16(3): 379 - 414. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Wiese, T. Gutsmann, and U. Seydel Review: Towards antibacterial strategies: studies on the mechanisms of interaction between antibacterial peptides and model membranes Innate Immunity, April 1, 2003; 9(2): 67 - 84. [Abstract] [PDF]
|
|
|
|
|
|
G. Jurgens, M. Muller, P. Garidel, M. H.J. Koch, H. Nakakubo, A. Blume, and K. Brandenburg Investigation into the interaction of recombinant human serum albumin with Re-lipopolysaccharide and lipid A Innate Immunity, April 1, 2002; 8(2): 115 - 126. [Abstract] [PDF]
|
|
|
|
|
|
T. Gutsmann, M. Muller, S. F. Carroll, R. C. MacKenzie, A. Wiese, and U. Seydel Dual Role of Lipopolysaccharide (LPS)-Binding Protein in Neutralization of LPS and Enhancement of LPS-Induced Activation of Mononuclear Cells Infect. Immun., November 1, 2001; 69(11): 6942 - 6950. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Seydel, O. Scheel, M. Muller, K. Brandenburg, and R. Blunck A K+ channel is involved in LPS signaling Innate Immunity, June 1, 2001; 7(3): 243 - 247. [Abstract] [PDF]
|
|
|
|
 |
|
 R. Blunck, O. Scheel, M. Muller, K. Brandenburg, U. Seitzer, and U. Seydel New Insights Into Endotoxin-Induced Activation of Macrophages: Involvement of a K+ Channel in Transmembrane Signaling J. Immunol., January 15, 2001; 166(2): 1009 - 1015. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Velasco, J. A. Bengoechea, K. Brandenburg, B. Lindner, U. Seydel, D. Gonzalez, U. Zahringer, E. Moreno, and I. Moriyon Brucella abortus and Its Closest Phylogenetic Relative, Ochrobactrum spp., Differ in Outer Membrane Permeability and Cationic Peptide Resistance Infect. Immun., June 1, 2000; 68(6): 3210 - 3218. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Forestier, E. Moreno, J. Pizarro-Cerda, and J.-P. Gorvel Lysosomal Accumulation and Recycling of Lipopolysaccharide to the Cell Surface of Murine Macrophages, an In Vitro and In Vivo Study J. Immunol., June 1, 1999; 162(11): 6784 - 6791. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A.B. Schromm, K. Brandenburg, R. Blunck, K. Fukase, S. Kusumoto, E.Th. Rietschel, and U. Seydel A biophysical approach towards an understanding of endotoxin-induced signal transduction Innate Immunity, February 1, 1999; 5(1-2): 41 - 45. [Abstract] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
