Mitogenicity of a spread film of monophosphoryl lipid A

Experimental and Molecular Pathology 79 (2005) 161 – 167 www.elsevier.com/locate/yexmp

Mitogenicity of a spread film of monophosphoryl lipid A Gregory S. Retzinger a,*, Kuni Takayama b a

b

Department of Pathology and Laboratory Medicine, University of Cincinnati, Cincinnati, OH 45267-0529, USA Mycobacteriology Research Laboratory, William S. Middleton Memorial Veterans Hospital, Madison, WI 53705, USA Received 10 June 2005 Available online 27 July 2005

Abstract When spread at the air – water interface, monophosphoryl lipid A (MPLA) forms stable insoluble monolayers that collapse at ¨55 dyn/ ˚ 2, consistent with the cross-sectional area of the lipid’s 6 acyl chains. The cm. At collapse, the exclusion area of each molecule is ¨119 A ˚ , determined, presumably, by the length of the acyl chains. For biological modeling of MPLA films, nominal thickness of such films is ¨22 A a system was developed in which monolayers of the lipid are supported by monodisperse hydrophobic beads of microscopic dimensions. Beads coated with MPLA monolayers within which the nominal area of each molecule is approximately equivalent to the ‘‘take-off’’ area of ˚ 2, are mitogenic for spleen cells. Given the natural occurrence of lipid A in the bacterial cell wall as the lipid at the air – water interface, 280 A well as the inherent stability of lipid A films, it seems reasonable to assume that at least some of the biological activities attributed to the lipid derive from its presentation/operation at an interface, i.e., on a surface. We propose beads coated with adsorbed films of lipid A will prove useful tools for modeling the activities of the lipid both in vitro and in vivo, and for elucidating the surface dependency and structural requirements of those activities. D 2005 Elsevier Inc. All rights reserved. Keywords: Monophosphoryl lipid A; Lipid A; Lipopolysaccharide; Adsorbed film; Surface properties; Mitogenicity; Innate immunity

Introduction Many amphiphilic lipids derived from bacterial cell walls and coatings, including lipid A from Gram-negative bacteria, lipoteichoic acid from Gram-positive bacteria, and trehalose dimycolate (TDM) from Mycobacterium tuberculosis, are believed to be involved in bacterial pathogenicity. As amphiphilic molecules, such lipids are naturally suited to occupy the water-contacting outer surface of microbes. It is within the context of the microbial surface that these lipids first contact a host’s immune system. Thus, it seems reasonable to assume it is within that same context the lipids are pathogenic. Earlier, we used microscopic hydrophobic beads coated with monomolecular films of TDM to demonstrate unequivocally that such films are responsible for many, if not all, of the biological activities of the mycobacterial glycolipid, * Corresponding author. E-mail address: [email protected] (G.S. Retzinger). 0014-4800/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.yexmp.2005.06.004

including acute inflammation, granuloma formation, and immune enhancement, i.e., adjuvant activity (Retzinger, 1987; Retzinger et al., 1981, 1982). Because lipid A, the toxic and biologically active moiety of lipopolysaccharide (LPS), has physicochemical features of TDM, i.e., a relatively small hydrophilic ‘‘head’’ group and a relatively large hydrophobic ‘‘tail,’’ it occurred to us the bead system might be used advantageously to probe the role of surface in the biological activities of lipid A, including activation of the innate immune system. Indeed, others have already demonstrated lipid A forms insoluble monolayers at the air – water interface (Gidalevitz et al., 2003), a finding that, by itself, strongly suggests the lipid operates on surfaces by design. Inasmuch as the physical form that constitutes the biologically active species of lipid A is still controversial, i.e., monomer vs. aggregate (Takayama et al., 1994; Mueller et al., 2004), the bead system should be ideally suited to studies aimed at determining the lipid’s active form: because the only lipid A would be that spread as a monolayer on beads, the environment of every molecule would be the same.

162

G.S. Retzinger, K. Takayama / Experimental and Molecular Pathology 79 (2005) 161 – 167

To date, much has been learned about the structural requirements necessary for the biological activities of lipid A. In this regard, monophosphoryl lipid A (MPLA, Fig. 1), a glucosamine disaccharide containing 6 fatty acyl groups and a single phosphate group, has been shown to be the minimal, nontoxic, structural unit with lipid A activity (Qureshi and Takayama, 1990; Qureshi et al., 1982; Rietschel et al., 1994). Given that MPLA is the structurally simplest, biologically active derivative of lipid A, it seemed appropriate to focus our investigations on it. Herein we document the surface activity of MPLA, and we describe the use of that information to develop a means by which to coat microscopic hydrophobic beads with a monomolecular film of the lipid. Using the mitogenic response of murine spleen cells as a representative indicator of MPLA activity, we then show MPLA coated onto beads is potently mitogenic. These very promising preliminary results suggest the bead system may prove generally useful, at both chemical and biological levels, for investigating the operation of lipid A and its derivatives.

Diagnostics, Indianapolis, IN. LPS prepared by phenolic extraction from Escherichia coli 0127:B8 was from Sigma, St. Louis, MO. All buffer materials and tissue culture reagents were of the highest quality available commercially. All solvents were of a grade suitable for HPLC. Eight- to 10-week-old female C57BL/6 mice were from Laboratory Supplies, Indianapolis, Indiana. The animal protocol used during this study was approved by the Institutional Animal Care and Use Committee. Monolayer studies A force – area curve of MPLA was generated using an automated Langmuir-type film balance. The monolayer was spread from a dilute solution of the lipid in benzene:ethanol, 9:1, v/v. The subphase was 0.05 M Tris – HCl, pH 7.40, containing 0.10 M NaCl. All buffer materials were of the highest quality available commercially. Water was deionized, distilled using an all-glass apparatus, and deaerated before use. All measurements were made at 21.5-C. Coating beads with lipids and/or protein

Materials and methods Lipids, reagents, chemicals, and mice MPLA was isolated from Salmonella typhimurium G30/ C21 as described elsewhere (Qureshi et al., 1982). [14C]MPLA was derived from Salmonella minnesota R595 as follows. Bacterial cells were grown in a 100-mL culture to an absorbance of 0.5 at 650 nm. Then, 1.0 mCi of sodium [1-14C]acetate (60 mCi/mmol, Amersham, Piscataway, NJ) was added to the culture, after which incubation was continued for 60 min. Subsequently, the cells were harvested by using centrifugation, and the [14C]LPS was isolated from them. Finally, [14C]MPLA was prepared from the [14C]LPS as described previously (Qureshi et al., 1985). TDM was extracted from Mycobacterium smegmatis and purified according to an established procedure (Takayama and Armstrong, 1976). Uniform poly(styrene-divinylbenzene) beads of diameter 5.7 T 1.5 Am were from Dow

Beads as delivered were first washed using an existing procedure (Retzinger et al., 1981) under sterile conditions. They were then either left lipid-free for subsequent dispersal in sterile, serum-supplemented culture medium, or they were coated with TDM or MPLA prior to dispersal in sterile, serum-supplemented culture medium. Beads coated with TDM to a nominal surface concentration of 0.15 Ag/ cm2 were prepared as described elsewhere (Retzinger et al., 1981). When coating beads with MPLA, the coating procedure was the same as that used to coat beads with TDM except that the solvent was benzene:ethanol, 9:1, v/v. Because the cross-linked beads are themselves aromatic, they swelled somewhat during the coating process. Beads, either lipid-free or lipid-coated, were then dispersed ultrasonically in 2.0 mL of RPMI 1640 medium containing 5% by volume of human AB serum. After sedimenting them at 1500g for 10 min and aspirating the supernatant, the beads were washed twice using RPMI 1640 medium. They were readied for mitogenicity assays by dispersal to a final concentration of 1.20  107/mL in serum-supplemented medium. Once adjusted to that concentration, the MPLAcoated beads only were further divided into 2 equal aliquots. After sedimenting the beads in one of the aliquots, the resulting bead-free supernatant was removed and used as diluent to disperse a sample of lipid-free beads to a final concentration of 1.20  107/mL. Binding of MPLA to beads

Fig. 1. Structure of MPLA.

[14C]MPLA was used as tracer to determine the amount of MPLA that bound to beads when using the standard coating and washing protocols. After redispersing [14C]MPLAcoated beads in serum-supplemented culture media, both

G.S. Retzinger, K. Takayama / Experimental and Molecular Pathology 79 (2005) 161 – 167

163

the radioactivity associated with the beads and that free in solution were determined in triplicate, using each time 3.9 mL of scintillation fluor (Ultima Gold XR, Perkin-Elmer, Boston, MA) and 0.1 mL of either the bead dispersion or the bead-free supernatant. From the specific activity of the [14C]MPLA, the number of beads in an aliquot, and the surface area per bead, the nominal surface concentration of MPLA on the beads was determined. Mitogenicity assay The incorporation of [3H]thymidine into the DNA of dividing spleen cells was used to assess mitogenicity (Humanson, 1967). Deeply anesthetized mice were sacrificed by using cervical dislocation. The spleens of the animals were removed under sterile conditions, and then homogenized using Hank’s balanced salt solution and a glass tissue homogenizer. After washing them twice, the cells were suspended to a final concentration of 3.0  106/ mL in RPMI 1640 medium containing human AB serum, 5% (v/v); penicillin G, 100 units/mL; streptomycin, 100 Ag/ mL; MOPS, 10 mM; 2-mercaptoethanol, 50 AM; and glutamine, 2 mM. Under sterile conditions, 200 AL of spleen cell suspension and 50 AL of medium containing either a test or a control material were added to individual wells of a 96-well microtiter tray (Lindbro, New Haven, CT). All assays were performed in triplicate. Trays were incubated for 72 h at 37-C and in the presence of a humidified atmosphere containing 5% CO2. Sixteen hours before harvesting the cells, 25 AL of RPMI 1640 medium containing 1.0 ACi of [3H]thymidine (New England Nuclear, Boston, MA) was added to each well that contained cells. Cells were concentrated onto glass fiber filters by using an automated sample harvester (Otto Hiller, Madison, WI). Radioactivity associated with the fiber filters was measured using for each 8.0 mL of scintillation fluor, and a liquid scintillation counter. Results are expressed as a stimulation index (SI), i.e., the radioactivity incorporated into stimulated cells divided by the radioactivity incorporated into unstimulated cells.

Results Force– area curve of MPLA The force – area curve of MPLA spread at the air –water interface is shown in Fig. 2. The surface isotherm is similar qualitatively to that of diphosphorylated lipid A (Gidalevitz et al., 2003). Like diphosphorylated lipid A, MPLA forms a stable monolayer that exhibits significant lateral compressibility at low surface pressures, transitions from an expanded to a more condensed state at ¨25 dyn/cm, and then collapses at ¨55 dyn/cm. For P  1.0 dyn/cm, the data of Fig. 2 fit well (R = 0.9999) the two-dimensional gas

Fig. 2. Force – area curve of MPLA spread at the air – water interface.

equation P = c/(A A o), where c (in erg/molecule) is a ˚ 2/molecule) is the nominal molecular constant and A o (in A exclusion area of MPLA at zero surface pressure. That ˚ 2, a value that indicates fitting yields A o = 260 T 1.5 A MPLA in the gaseous state is rather extended and flattened at the interface. Within collapsing films, the exclusion area ˚ 2, consistent with a crossof each molecule is ¨119 A sectional area of 6 acyl chains. Using the densities of its component parts (phosphoric acid, q = 1.83 g/cm3; glucosamine, q = 1.55 g/cm3; and myristic acid, q = 0.84 g/cm3), one calculates for MPLA a density of ¨1.10 g/cm3, which, given the molecular weight of the lipid, 1717, corresponds ˚ 3. Dividing its to a nominal molecular volume of ¨2600 A molecular volume by its molecular area at collapse yields ˚ , a value for MPLA a nominal molecular thickness of ¨22 A eminently consistent with the length of the lipid’s acyl chains. MPLA binds stably to microscopic poly(styrene-divinylbenzene) beads One of us has shown that other amphiphilic lipids that form insoluble monolayers at the air –water interface, e.g., TDM, phospholipids and poly(ethylene oxide)/poly(propylene oxide) block copolymers, bind avidly to hydrophobic surfaces, including that of poly(styrene-divinylbenzene) beads (Retzinger et al., 1981, 1985; O’Connor et al., 1999). Either monolayers or multilayers can be bound to beads, depending upon the amount of lipid used during the coating process (Retzinger et al., 1981). Our experience has taught us the most stable confluent monomolecular film is one in which the nominal surface concentration of the lipid approximates that which corresponds to the ‘‘take-off’’ area of the lipid at the air – water interface (Retzinger et al., 1981, 1985; O’Connor et al., 1999). For MPLA, the take-off area is ˚ 2/molecule, or a surface concentration of ¨0.10 Ag/ ¨280 A 2 cm . When coating beads with an amount of lipid just less than what is necessary to form a confluent monolayer, the efficiency of the coating process is very high, approaching 100% (Retzinger et al., 1981, 1985). As the amount of lipid

164

G.S. Retzinger, K. Takayama / Experimental and Molecular Pathology 79 (2005) 161 – 167

used for the coating process is increased beyond that amount, however, the coating efficiency falls – even as multilayers form on the beads – but can still be relatively high, > 60%. To avoid multilayer formation, we used for our routine coating process an amount of MPLA that, if all bound to the beads, would yield a nominal packing density ˚ 2/molecule. After using a sonicating waterbath to of ¨240 A wash the beads 3 times, we found, as expected, most of the lipid, 94.7 T 2.0 %, to be associated with the particles. Thus, using our standard protocol, the nominal packing density of ˚ 2/molecule, a value roughly MPLA on beads is ¨250 A equivalent to the take-off area of the lipid at the air – water interface. Using a space-filling molecular model of MPLA, we determined that the composite head group of the lipid ˚ 2/molecule. This has, at most, a cross-sectional area of 150 A means that, even in confluent films, the hydrophobic domains of the lipid must be in contact with water (see below). Once coated to confluency onto beads, virtually all of the MPLA, as had been the case for TDM (Retzinger et al., 1981), remains bound to the particles for at least 1 week when they are stored at 4-C or room temperature in either buffer or serum-supplemented medium.

their corresponding SIs being no different than that of unstimulated cells. In contrast and as expected, the mitogenic response of spleen cells to LPS was, for all concentrations of LPS tested, significantly higher than that of unstimulated cells, appearing saturable above a concentration of ¨10 Ag/mL. As good as the mitogenic response to LPS was, the response to MPLA-coated beads was even better, being nearly 30% higher than the maximal response to LPS. This is particularly interesting, given that LPS as an aqueous dispersion is significantly more toxic than MPLA (Qureshi and Takayama, 1990; Rietschel et al., 1994; Schletter et al., 1995). Given the stability of the bead-bound MPLA film and assuming a critical micelle concentration of ¨10 8 M for lipid A (Mueller et al., 2004), it is exceedingly unlikely MPLA micelles contributed to the mitogenic response elicited by MPLA-coated beads. In support of this, lipid-free beads administered in medium from which MPLA-coated beads had been removed were inactive. We conclude MPLA presented as a monomolecular film on beads is potently mitogenic.

Discussion MPLA coated onto beads as a monomolecular film is mitogenic We next tested the mitogenicity of MPLA presented as a monolayer on beads. At the same time, we assessed the mitogenicity of beads coated with a well-characterized film of the mycobacterial glycolipid TDM. Such TDM-coated beads elicit acute inflammation and granuloma formation in vivo, and they are adjuvant-active (Retzinger, 1987; Retzinger et al., 1982). The ratio of beads to cells in the mitogenicity assays was 1:1. As positive controls, we used a commercially available LPS prepared to various concentrations, using as diluent serum-supplemented, RPMI 1640 medium. As shown in the Table 1, neither lipid-free beads nor beads coated with TDM were mitogenic for spleen cells, Table 1 Mitogenic response of murine spleen cells to various materials Material

Stimulation indexa

Medium alone Lipid-free beads TDM-coated beads (0.08 Ag)b LPS, 1.0 Ag/mL (0.25 Ag) LPS, 10.0 Ag/mL (2.5 Ag) LPS, 50.0 Ag/mL (12.5 Ag) LPS, 100.0 Ag/mL (25.0 Ag) MPLA-coated beads (0.07 Ag) Lipid-free beads + mediumc

1.0 1.0 1.1 5.7 14.9 16.9 16.6 21.6 1.0

a

T T T T T T T T T

0.1 0.0 0.3 1.0 1.2 0.3 1.4 1.5 0.3

Results are reported as the mean T SD of triplicate testing. Number in parentheses is the absolute amount of test lipid in each well of the microtiter plate used for the assay. c Lipid-free beads were dispersed and added to wells in bead-free medium recovered from a preparation of MPLA-coated beads. See Materials and methods for details. b

We are interested in the mechanism(s) of pathogenicity of lipids found in microbial cell walls, membranes, and coatings. Because such lipids occupy the interface that constitutes the boundary between a microbe and its external, often watery, environment, one might assume it is within the context of an interface the lipids are pathogenic. Indeed, a unifying property of many microbial lipid pathogens is their surface activity; for some in particular, their ability to form insoluble monolayers at the air – water interface. We believe this feature is intimately involved in the pathogenicity of lipids of the sort typified by LPS and TDM. At the very least, this feature hints of the lipids’ site of action: surfaces. In this report, we demonstrated, first, that MPLA forms a stable monolayer at the air – water interface. We then showed that microscopic hydrophobic beads coated with an amount of MPLA equivalent to that of a confluent monolayer at the air – water interface are mitogenic for cultured murine spleen cells. What follows is a discussion of the relevance – both practical and theoretical – of our findings. Presently, most aqueous preparations of LPS or LPSrelated materials, including lipid A and MPLA, are heterogeneous in nature, containing both monomeric species and multimeric species, i.e., aggregates, of varying size and shape. Indeed, that heterogeneity is responsible for the controversy that exists regarding the identity of the biologically active form of lipid A. Microscopic beads coated with a monomolecular film of lipid circumvent the issue of heterogeneity because they provide a means by which to present the lipid uniformly, i.e., the environment of every lipid molecule is the same. In part because of our previous experience, and in part because we wanted to

G.S. Retzinger, K. Takayama / Experimental and Molecular Pathology 79 (2005) 161 – 167

model bacteria, we used solid microscopic beads as supports for MPLA films. However, if one assumes that radius of curvature of the coated surface plays no role, then there is no reason a priori why, for example, the surface of the plastic microtiter plates used for the mitogenicity assays could not have been covered with MPLA directly. We can imagine that a whole range of hydrophobic materials – particulates and otherwise – could be coated with biologically active lipid films for various uses in vitro and/or in vivo. The physical form of lipid A that constitutes the biologically active species of the lipid, i.e., monomer vs. aggregate, is still controversial, and data that either support or refute the activity of one or the other of those two forms exist (Takayama et al., 1994; Mueller et al., 2004). Despite the controversy, there is general agreement on the involvement of certain proteins in the operation of lipid A. The prevailing notion has lipid A, in some form, binding to LPSbinding protein (LBP) either in plasma (Schletter et al., 1995) or membrane bound (Gutsmann et al., 2001) for delivery to a receptor complex on phagocytic cells. That receptor complex includes CD14, Toll-like receptor 4 (TLR4), and myeloid differentiation protein-2 (MD-2) (Fujihara et al., 2003). Lipid A also binds in LBPindependent fashion in a process involving CD11b/CD18 (Troelstra et al., 1999; Wright and Jong, 1986). Importantly, there is as yet no convincing evidence lipid A has to be transported across the cytoplasmic membrane of a phagocytic cell to signal its biological effects (Latz et al., 2002). Although the work presented here does not specifically address either interaction/complexation of MPLA with proteins or transduction of the lipid’s mitogenic ‘‘signal,’’ it may help reconcile existing theories on the physical form of the biologically active species of lipid A. Namely, MPLA presented as a monomolecular film is mitogenic. Taken at face value, such a finding suggests the lipid surface itself is biologically active. Because aggregates are equipped with a surface of macromolecular dimensions, they seem reasonable, even obvious, as active species. On the other hand, monomers, which can only contribute to the formation of a macromolecular surface, seem much less reasonable, certainly not obvious, as active species. The latter conjecture becomes more reasonable, however, if one recognizes that monomers of MPLA, even those in equilibrium with a film of the lipid, will readily adsorb from an aqueous phase to any hydrophobic/amphiphilic material in contact with that phase, e.g., a cytoplasmic membrane, lipid-rich particles (Brandenburg et al., 2002), or even the walls of plastic microtiter wells. Thus, it is entirely possible that monomers of MPLA and, by inference, monomers of diphosphorylated lipid A are not recognized as discrete units by plasma and cellular effectors, but as constituents of a biologically active film adsorbed to some third party. Others have proposed the ratio of the cross-sectional areas of the hydrophilic and the hydrophobic domains of lipid A is an important determinant of the lipid’s pathoge-

165

nicity (Schromm et al., 2000; Seydel et al., 2000). According to them, that ratio determines the supramolecular structure of lipid A aggregates and, consequently, the biological activity of those aggregates. They also believe the ratio is responsible for a presumed lateral stress imposed within the cytoplasmic membrane of responding cells by lipid A molecules that become intercalated there. Although we share the view that ‘‘hydrophile– lipophile balance’’ (HLB) plays a role in the pathogenicity of lipid A, we propose that role has more to do with the topography/ configuration of the surface formed by the lipid than with either formation of three-dimensional aggregates or lateral diffusion within membranes. Like TDM, MPLA and lipid A form insoluble monolayers. A distinguishing feature of the monolayers formed by these particular lipids – one that relates to HLB – is the rather large contribution the lipids’ hydrophobic domains make to the total cross-sectional areas of the films, even at high surface pressures. Conceptually, this means the area covered by molecules of these lipids is populated even in confluent films by a regular array, or ‘‘molecular pattern,’’ of water-exposed hydrophobic patches and hydrophilic nodes of the sort already visualized for lipid A (Wang and Hollingsworth, 1996). In an aqueous environment, the hydrophobic patches are ideally suited to binding proteins (Retzinger et al., 1981, 1985), whether those proteins are free in solution or fixed to an interface, e.g., a cell surface. Leucine-rich, amphiphilic secondary structural elements, i.e., a-helices and h-strands, bind especially well to hydrophobic domains in lipid films (Retzinger et al., 1985). Given current thinking, it is reasonable to suppose that extracellular leucine-rich repeats of TLR-4 (Bell et al., 2003) recognize and bind to a specific configuration of hydrophobic domains expressed on MPLA/ lipid A surfaces while clusters of basic amino acids of MD-2 (Gruber et al., 2004) recognize and bind in complementary fashion to the phosphate-rich hydrophilic nodes expressed on those same surfaces. We do not mean to imply that all insoluble monolayers composed of MPLA-like lipids will be similarly recognized and treated by responding cells and other biologic effectors. After all, TDM forms films of the same sort as MPLA, yet films of the former lipid, although biologically active in vivo, are not mitogenic in vitro. Thus, there must be something that distinguishes the two surfaces. Because the hydrophobic domains of pathogenic lipids that form insoluble monolayers are all rather large and chemically monotonous, head group heterogeneity likely most distinguishes one such monolayer from another. If that is the case, then the modularity of receptors involved with innate immunity (Barton and Medzhitov, 2003) may be an evolutionary consequence of the immune system having paired global recognition of hydrophobic domains (Seong and Matzinger, 2004) of suitable dimensions with variably specific recognition of hydrophilic ones. Although we are confident the innate immune system recognizes and processes an MPLA surface of the sort

166

G.S. Retzinger, K. Takayama / Experimental and Molecular Pathology 79 (2005) 161 – 167

described here, the information of this report does not allow identification of the surface feature(s) that is (are) ultimately involved in transduction of MPLA-elicited biological signals. While we favor the hypothesis topography per se is primarily important, i.e., it is the surface in its totality that is recognized and processed, we cannot yet exclude the possibility surface presentation facilitates recognition and processing of individual lipid molecules by, for example, LBP. Polymerized films (Fendler, 1984) of bead-bound lipid A analogues might help reconcile which of these two hypotheses is correct. We had two goals for this research. Firstly, we wanted to demonstrate that MPLA presented as a monomolecular film to responsive cells is biologically active. Secondly, we wanted to show that the bead system is a useful tool for exploring the operation of MPLA films. Although we achieved our goals, much still needs to be done. As an example, it will be important to compare directly MPLA on beads to aqueous dispersions of the lipid. It will also be important to explore the area and surface concentration dependencies of biological responses attributed to MPLA, including not only mitogenicity, but also production of cytokines and activation of nuclear regulatory factors. The bead system is especially suited to such investigations. Because all of the lipid is presented uniformly on a surface, the role of the interfacial material in the elicitation of biological phenomena can be more easily explored and ascertained.

Acknowledgments This work was supported by grants to GSR from Johnson & Johnson and the NIH (R21CA106257), and by a Merit Review grant to KT from the Research Service of the Department of Veterans Affairs. GSR thanks Ruth Mary Retzinger for inspiration.

References Barton, G.M., Medzhitov, R., 2003. Toll-like receptor signaling pathways. Science 300, 1524 – 1525. Bell, J.K., Mullen, G.E., Leifer, C.A., Mazzoni, A., Davies, D.R., Segal, D.M., 2003. Leucine-rich repeats and pathogen recognition in Toll-like receptors. Trends Immunol. 24, 528 – 533. Brandenburg, K., Jurgens, G., Andra, J., Lindner, B., Koch, M.H., Blume, A., Garidel, P., 2002. Biophysical characterization of the interaction of high-density lipoprotein (HDL) with endotoxins. Eur. J. Biochem. 269, 5972 – 5981. Fendler, J.H., 1984. Polymerized surfactant vesicles: novel membrane mimetic systems. Science 223, 888 – 894. Fujihara, M., Muroi, M., Tanamoto, K., Suzuki, T., Azuma, H., Ikeda, H., 2003. Molecular mechanisms of macrophage activation and deactivation by lipopolysaccharide: roles of the receptor complex. Pharmacol. Ther. 100, 171 – 194. Gidalevitz, D., Ishitsuka, Y., Muresan, A.S., Konovalov, O., Waring, A.J., Lehrer, R.I., Lee, K.Y.C., 2003. Interaction of antimicrobial peptide protegrin with biomembranes. Proc. Natl. Acad. Sci. 100, 6302 – 6307.

Gruber, A., Mancˇek, M., Wagner, H., Kirschning, C.J., Jerala, R., 2004. Structural model of MD-2 and functional role of its basic amino acid clusters involved in cellular lipopolysaccharide recognition. J. Biol. Chem. 279, 28475 – 28482. Gutsmann, T., Mu¨ller, M., Carroll, S.F., MacKenzie, R.C., Wiese, A., Seydel, U., 2001. Dual role of lipopolysaccharide (LPS)-binding protein in neutralization of LPS and enhancement of LPS-induced activation of mononuclear cells. Infect. Immun. 69, 6942 – 6950. Humanson, G.L., 1967. Animal Tissue Techniques, 2nd edR W.H. Freeman and Company, San Francisco, pp. 483 – 484. Latz, E., Visintin, A., Lien, E., Fitzgerald, K.A., Monks, B.G., Kurt-Jones, E.A., Golenbock, D.T., Espevik, T., 2002. Lipopolysaccharide rapidly traffics to and from the golgi apparatus with the Toll-like receptor 4MD-2-CD14 complex in a process that is distinct from the initiation of signal transduction. J. Biol. Chem. 277, 47834 – 47843. Mueller, M., Lindner, B., Kusumoto, S., Fukase, K., Schromm, A.B., Seydel, U., 2004. Aggregates are the biologically active units of endotoxin. J. Biol. Chem. 279, 26307 – 26313. O’Connor, S.M., Gehrke, S.H., Retzinger, G.S., 1999. Ordering of poly (ethylene oxide)/poly(propylene oxide) triblock copolymers in condensed films. Langmuir 15, 2580 – 2585. Qureshi, N., Takayama, K., 1990. Structure and function of lipid A. In: Iglewski, B.H., Clark, V.L. (Eds.), The Bacteria, vol. XI. Academic Press, San Diego, pp. 319 – 338. Qureshi, N., Takayama, K., Ribi, E., 1982. Purification and structural determination of nontoxic lipid A obtained from the lipopolysaccharide of Salmonella typhimurium. J. Biol. Chem. 257, 11808 – 11815. Qureshi, N., Mascagni, P., Ribi, E., Takayama, K., 1985. Monophosphoryl lipid A obtained from lipopolysaccharides of Salmonella minnesota R595. Purification of dimethyl derivative by high performance liquid chromatography and complete structural determination. J. Biol. Chem. 260, 5271 – 5278. Retzinger, G.S., 1987. Dissemination of beads coated with trehalose 6,6Vdimycolate. A possible role for coagulation in the dissemination process. Exp. Mol. Pathol. 46, 190 – 198. Retzinger, G.S., Meredith, S.C., Takayama, K., Hunter, R.L., Ke´zdy, F.J., 1981. The role of surface in the biological activities of trehalose 6,6V-dimycolate. Surface properties and development of a model system. J. Biol. Chem. 256, 8208 – 8216. Retzinger, G.S., Meredith, S.C., Hunter, R.L., Takayama, K., Ke´zdy, F.J., 1982. Identification of the physiologically active state of the mycobacterial glycolipid trehalose 6,6V-dimycolate and the role of fibrinogen in the biologic activities of trehalose 6,6V-dimycolate monolayers. J. Immunol. 129, 735 – 744. Retzinger, G.S., Meredith, S.C., Lau, S.H., Kaiser, E.T., Ke´zdy, F.J., 1985. A method for probing the affinity of peptides for amphiphilic surfaces. Anal. Biochem. 150, 131 – 140. Rietschel, E.T., Kirikae, T., Schade, F.U., Mamat, U., Schmidt, G., Loppnow, H., Ulmer, A.J., Zahringer, U., Seydel, U., DiPadova, F., Schreier, M., Brade, H., 1994. Bacterial endotoxin: molecular relationships of structure to activity and function. FASEB J. 8, 217 – 225. Schletter, J., Heine, H., Ulmer, A.J., Rietschel, E.T., 1995. Molecular mechanisms of endotoxin activity. Arch. Microbiol. 164, 383 – 389. Schromm, A.B., Brandenburg, K., Loppnow, H., Moran, A.P., Koch, M.H., Rietschel, E.T., Seydel, U., 2000. Biological activities of lipopolysaccharides are determined by the shape of their lipid A portion. Eur. J. Biochem. 267, 2008 – 2013. Seong, S.-Y., Matzinger, P., 2004. Hydrophobicity: an ancient damageassociated molecular pattern that initiates innate immune responses. Nat. Rev., Immunol. 4, 469 – 478. Seydel, U., Oikawa, M., Fukase, K., Kusumoto, S., Brandenburg, K., 2000. Intrinsic conformation of lipid A is responsible for agonistic and antagonistic activity. Eur. J. Biochem. 267, 3032 – 3039. Takayama, K., Armstrong, E., 1976. Isolation, characterization, and function of 6-mycolyl-6V-acetyltrehalose in the H37Ra strain of Mycobacterium tuberculosis. Biochemistry 15, 441 – 447.

G.S. Retzinger, K. Takayama / Experimental and Molecular Pathology 79 (2005) 161 – 167 Takayama, K., Mitchell, D.H., Din, Z.Z., Mukerjee, P., Li, C., Coleman, D.L., 1994. Monomeric Re lipopolysaccharide from Escherichia coli is more active than the aggregated from in the Limulus amebocyte lysate assay and in inducing Egr-1 mRNA in murine peritoneal macrophages. J. Biol. Chem. 269, 2241 – 2244. Troelstra, A., de Graaf-Miltenburg, L.A.M., van Bommel, T., Verhoef, J., Van Kessel, K.P.M., Van Strijp, J.A.G., 1999. Lipopolysaccharidecoated erythrocytes activate human neutrophils via CD14 while

167

subsequent binding is through CD11b/CD18. J. Immunol. 162, 4220 – 4225. Wang, Y., Hollingsworth, R.I., 1996. An NMR spectroscopy and molecular mechanics study of the molecular basis for the supramolecular structure of lipopolysaccharide. Biochemistry 35, 5647 – 5654. Wright, S.D., Jong, M.T.C., 1986. Adhesion-promoting receptors on human macrophages recognize Escherichia coli by binding to lipopolysaccharide. J. Exp. Med. 164, 1876 – 1888.