Current Protein and Peptide Science, 2005, 6, 000-000
1
Membrane Interactions of Host-defense Peptides Studied in Model
Systems
Raz Jelinek* and Sofiya Kolusheva
Department of Chemistry and Staedler Minerva Center for Mesoscopic, Macromolecular Engineering, Ben Gurion
University of the Negev, Beersheva 84105, Israel
Abstract: Host-defense, antibiotic peptides are believed to generate their cytolytic effects by interacting with the
membranes of bacterial cells. Direct analyses of peptide interactions with real cellular membranes are difficult, however,
due to the high complexity of physiological membranes. This review summarizes experimental work aiming to understand
peptide-membrane interactions and their relationships with the peptides' biological actions using specific model systems.
Varied model assemblies have been constructed that generally aim to mimic the fundamental lipid bilayer organization of
the membrane. The model systems we will describe include multilamellar and unilamellar vesicles, planar lipid bilayers,
lipid monolayers and micelles, and colorimetric biomimetic membranes. The different artificial models have facilitated
examination of specific biological or chemical parameters affecting peptide action, for example the effect of membrane
lipid composition on peptide affinities and membrane penetration, the relationship between membrane fluidity and peptide
interactions, the conformations of active peptides, and other factors. We evaluate the strengths and limitations of the
various approaches, and point to future directions in the field.
1. INTRODUCTION
The potential of host-defense peptides (also referred to as
antimicrobial peptides, or AMPs) as promising therapeutic
agents for combating bacterial infections has been a
powerful driving force for conducting numerous studies
aiming to decipher their mode of action. Early observations
indicated a possible connection between membrane
permeation and bacterial killing [1]. Indeed, most known
AMPs are short amphipathic cationic sequences that interact
strongly with bacterial membranes [2]. These interactions,
often resulting in pore formation and/or cell lysis, are
believed to constitute primary factors of the bactericidal
capacity of AMPs. Thus, the study of peptide-membrane
interactions and elucidating the effects of AMPs on the cell
membrane have become increasingly important for
understanding the biological activities of such peptides.
Despite a significant amount of experimental data, many
of the central tenets of AMP action remain unresolved and
debated. Among the central issues pursued are the molecular
mechanisms and characteristics of peptide-mediated cell
lysis, degree of peptide insertion into the membranes, the
extent and significance of pore formation, membrane
micellization processes, and others. Such questions are
almost universally addressed through examination of model
membrane systems. Studying peptide-membrane complexes
in model systems is primarily due to the molecular and
structural complexity of the membrane of actual cells,
combined with the difficulties encountered in working with
complete cellular systems. The construction of artificial
*Address correspondence to this author at the Department of Chemistry and
Staedler Minerva Center for Mesoscopic, Macromolecular Engineering, Ben
Gurion University of the Negev, Beersheva 84105, Israel; Tel: +972-86461747; Fax: +972-8-6472943; E-mail: [email protected]
1389-2037/05 $50.00+.00
models further allows examination of the contribution of
specific parameters to membrane-peptide interactions and
peptide activities such as lipid type and composition,
peptide-lipid ratio, surface curvature, and others by diverse
bio-analytical and spectroscopic methods.
This Review summarizes the different model lipid
systems for studying the action of host-defense peptides,
including multilamellar and unilamellar vesicle assemblies,
supported lipid bilayers, micelles, lipid monolayers at the air
water interface, and recently developed chromatic vesicle
platforms. We discuss the important properties of each
approach and the type of information extracted by the
experiments. Due to the large body of work in the field and
limited space we only refer to representative publications. In
particular, we have tried to provide here a critical summary
of the subJect that would allow the reader to evaluate the
biological insights provided on AMP-membrane interactions
by using different model systems. Several excellent reviews
on related topics have appeared in recent years [3-10].
The use of model systems for elucidating AMP
biological actions has progressed hand in hand with a
simultaneous development of bio-analytical techniques. The
maJority of spectroscopic methodologies applied for
studying host defense peptides in model systems have
focused on the structural modifications of the peptides
following their interactions with the lipid assembly. Such
studies have attempted to correlate the peptide
conformational changes induced within the lipid
environments with their bactericidal properties. Accordingly,
we also summarize the analytical techniques applied in
conJunction with biomimetic membranes, including
fluorescence spectroscopy, nuclear magnetic resonance
(NMR), circular dichroism (CD), infrared spectroscopy, and
others.
© 2005 Bentham Science Publishers Ltd.
2
Current Protein and Peptide Science, 2005, Vol. 6, No. 1
2. LIPID VESICLES
2.1 Multilamellar Vesicles
Multilamellar vesicles (MLVs) have been used for
studying various aspects of membrane permeation by hostdefense peptides and their bactericidal mechanisms. MLVs
retain the lipid bilayer organization of cellular membranes
(Fig. 1) and thus could serve as a useful model for studying
peptide association and induced effects within the bilayer.
The primary advantages of MLVs for elucidating membrane
properties and interactions are often related to the technical
requirements of several bio-analytical techniques employed
for investigating peptide-membrane interactions, such as
solid state NMR, calorimetry, and X-ray diffraction.
Jelinek and Kolusheva
31
P solid-state NMR of DMPC/DMPG MLVs similarly
suggested that the action of the wasp venom peptide
mastoparan involves induction of non-lamellar phases in the
bilayer [12].
Solid-state NMR techniques have been applied to
investigate the modification of AMP molecular properties in
membranes. High-resolution solid-state NMR experiments
could provide information on molecular interactions as well
as structural and dynamical parameters pertaining to both the
lipid molecules as well as the inserted peptides. A solid-state
NMR study revealed, for example, that the potent AMP
protegrin-1 (PG-1) formed highly immobile large aggregates
within lipid bilayers [17]. Another NMR study utilizing
Mn2+ ions incorporated within the lipid bilayer examined the
depth of insertion of PG-1 through the induction of the
distance-dependent dipolar relaxation of the nuclear spins
[18].
Solid-state NMR methods have been applied in MLV
suspensions to address one of the most important question
related to the bactericidal mechanisms of AMPs, specifically
the significance of interactions of the peptides with the
phospholipid headgroups vs. insertion into the hydrophobic
interior of the bilayer. 31P and 2H investigations were carried
out to determine the extent of headgroup disruption by short
AMPs compared with their penetration into the hydrophobic
environment of the acyl chains [19]. That research
determined that short peptides preferably accumulated at the
lipid surface, while longer sequences tended to penetrate
deeper into the bilayer. Similar interpretation emerged from
solid-state NMR studies of oriented lipid systems (see
below) [14, 20]. Surface dimerization of the amphipathic
AMP magainin-2 was suggested as the primary mechanism
for the membrane permeabilization properties of the peptide,
using a newly developed NMR technique denoted transferred
nuclear Overhauser enhancement (TRNOE) spectroscopy
[21].
Fig. (1). Schematic picture of a multilamellar vesicle assembly.
The use of MLVs has been essential in varied solid-state
NMR studies of peptide-membrane assemblies. Lipid
multilamellae provide sufficient sample quantities and
acceptable signal-to-noise ratios, but also, critically, maintain
a favorable time-regime for the solid-state NMR experiments
[11-13]. Studies of structure, dynamics, and orientation of
antimicrobial peptides in membranes, and their structurefunction relationships as probed by solid-state NMR
spectroscopy were reviewed [14, 15]. Examples for specific
NMR experiments for deciphering AMP-membrane
interactions include the application of 31P NMR for studying
pore formation of equinatoxin II, a eukaryotic toxin, and its
specificity to phospholipid targets within bacterial
membranes [16]. The appearance of toroidal pores within the
bilayers was deduced from the observation of an isotropic
component in the 31P NMR spectra of the MLV suspensions.
The effect of AMP permeation on membranes has been
often described in terms of lipid disruption and decreased
order. Several techniques, primarily differential scanning
calorimetry (DSC) and x-ray diffraction, have been applied
to probe the ordering and molecular packing of lipid bilayers
within MLVs prior and following interactions with AMPs.
Several studies utilizing these methods in model lipid
systems have yielded quantitative information on the effects
of host defense peptides on membrane structure, as well as
on peptide localization within the lipid bilayer [22].
Specifically, DSC has been applied in multilamellar
phospholipid systems to probe the thermodynamic profiles of
the lipid assemblies and the consequences of peptide
interactions in term of ordering and cooperative properties of
the bilayers [22]. A detailed DSC study of membrane
interactions of gramicidin S (GS) indicated that the peptide
appeared to perturb packing in liquid crystalline lipid phases
inducing the formation of inverted cubic phases in certain
conditions [23]. That study suggested that this type of bilayer
destabilization is a primary constituent of the antimicrobial
mode of action of GS.
Similar to NMR, DSC analyses require relatively large
sample quantities for probing modifications of the lipid
phase transitions induced by the associated peptides [23-25].
Membrane Interactions of Host-defense Peptides Studied in Model Systems
Grasso et al, for example, employed DSC to elucidate the
effect of a scrambled hydrophobic/hydrophilic sequence on
the cooperative transitions of MLVs composed of DPPC
[26]. A similar microcalorimetric study analyzed the binding
properties of the peptide lactoferricin B to phosphatidylglycerol (PG) or phosphatidylcholine (PC) MLVs [27].
These studies pointed to the higher affinity of certain AMPs
to negatively-charged phospholipids, generally more
abundant within bacterial membranes. Particularly important,
an x-ray diffraction and DSC investigation indicated that
membrane curvature strain and hydrophobic mismatch
between the interacting peptides and lipids are prominent
factors in the mechanisms of membrane perturbation and cell
lysis [24]. This interpretation was supported by a DSC study
indicating that a short AMP containing 17 β-amino acids
promoted a negative curvature within phospholipid
assemblies [28].
Special MLV preparation methods have been introduced
for facilitating investigation of peptide-membrane
interactions using advanced structural techniques such as
neutron diffraction. By carefully modifying sample
hydration, neutron diffraction analyses could identify distinct
lipid crystallization processes occurring within the
multilamellae and relate this information to structural
transformations induced by bound AMPs. Yang et al showed
that crystallization of membrane pores occurred
spontaneously following association of magainin and
protegrin with MLVs [29].
Studies of peptide-membrane interactions have
frequently utilized different model assemblies rather than a
single system, lending better credence to the experimental
conclusions. An example of this multiprong approach has
been the application of Fourier transform IR spectroscopy to
characterize interactions of gramicidin S (GS) with MLVs,
lipid monolayers, and micelles (see subsections below) [30].
That systematic work examined the dependence of peptide
interactions on temperature and lipid phase. Interestingly, the
researchers found that GS was completely or partially
excluded from the gel states of all of the lipid bilayers
examined but strongly partitioned into micelles, lipid
monolayers, or bilayers in the liquid crystalline phase,
suggesting a possible mechanistic prerequisite for
bactericidal action of the peptide.
Gramicidin S represents an AMP that was extensively
studied in diverse model systems to elucidate its bactericidal
mechanisms. MLVs were mostly employed for deciphering
GS-membrane interactions using electron spin resonance
(ESR) [31], fluorescence spectroscopy using the
diphenylhexatriene (DPH) probe for lipid dynamics [32],
FTIR spectroscopy [33], and Raman spectroscopy [34]. X
ray studies carried out in total membrane lipid extracts from
microbial species indicated that GS induced the formation of
cubic lipid phases, an important observation that could shed
light on the underlying influence of the peptide on bacterial
membranes [35]. Experiments utilizing various analytical
techniques have indicated that cholesterol attenuates the
interaction of GS with lipid bilayers [36].
MLV suspensions have made possible morphological
characterization of membrane-peptide complexes using
visualization techniques such as electron microscopy. Cryo-
Current Protein and Peptide Science, 2005, Vol. 6, No. 1
3
electron microscopy, complemented by 31P and 2H NMR,
was applied to study the permeability and morphological
perturbations induced by the pore-forming peptide nisin in
phosphatidylcholine membranes [37]. The researchers
observed dramatic modifications of the morphologies of the
lipid assemblies following interactions with the peptide, and
could further detect an inhibitory effect of cholesterol. The
electron microscopy data suggested a distinct interplay
between nisin and lipid constituents of the bilayer resulting
in membrane permeability.
2.2 Unilamellar Vesicles
Unilamellar vesicles have been used widely for studying
membrane processes in general and peptide-membrane
interactions in particular. This is due in most part to the fact
that conceptually such vesicles mimic a cell assembly - in
which the lipid bilayer forms an enclosed volume separated
from the external solution. Traditionally, unilamellar vesicles
have been divided into three main groups depending on their
sizes: small unilamellar vesicles (SUVs), large unilamellar
vesicles (LUVs), and giant unilamellar vesicles (GUVs) (Fig.
2). The preparation of the different vesicle classes depends
upon the types of lipids used and experimental methods [3841]. Larger-size vesicles have been generally preferred as
model systems for studying membrane processes since their
curvature more closely resembles actual cells [42, 43].
Unilamellar vesicles have been used in combination with
spectroscopic techniques to decipher membrane association
and permeation by AMPs, and to correlate the biophysical
data with the antibacterial activities of the peptides. CD
spectroscopy has been an important tool for probing
secondary structures of vesicle-associated peptides. This
technique has been applied to probe the effect of negative
phospholipids compared to zwitterionic phospholipids on
AMP conformations [4, 44, 45]. Electron spin resonance
(ESR) has been an important tool for studying AMP
insertion into the bilayer and the resultant modification of
bilayer fluidity and lipid motion [46]. ESR experiments have
usually employed spin labels covalently attached to the lipid
acyl chains at various positions.
A large number of studies have employed fluorescence
leakage techniques to probe AMP effect on the cell
membrane. In such experiments the changes in the
fluorescence of probes entrapped within unilamellar vesicles
are monitored following interactions of the membrane-active
peptides with the vesicles. Specifically, pore formation or
complete bilayer destruction by the peptides result in leakage
of the encapsulated fluorescence dyes, thereby significantly
modifying [increasing or reducing] their fluorescence
emission. Vesicle-encapsulated dye release by Hagfish
intestinal antimicrobial peptides (HFIAPs) confirmed the
high specificity of these peptides to bacterial membranes
[47]. Leakage experiments were carried out in a study
comparing the membrane activity of a synthetic amphipathic
β-sheet cationic AMP to the action of known bactericidal
sequences [48]. Oren and Shai investigated membrane
permeation by cyclic peptides using a K+ - sensitive
fluorescent dye [49].
Among the most common experiments in that regard is
the construction of vesicles encapsulating self-quenching
4
Current Protein and Peptide Science, 2005, Vol. 6, No. 1
Jelinek and Kolusheva
Fig. (2). Schematic pictures of unilamellar vesicles: small unilamellar vesicles (SUVs, left), large unilamellar vesicles (LUVs, middle), giant
unilamellar vesicles (GUVs, right).
fluorescent probes such as calcein and fluorescein. When
such labels are entrapped within the vesicle interior a
considerable self-quenching occurs. However, bilayer
destruction or formation of pores within the enclosing lipid
surface gives rise to an increase in fluorescence due to
leakage of the probe molecules into the external solution
[50]. Mangoni et al recorded the leakage induced in vesicles
containing different phospholipid compositions by
temporins, short AMPs secreted by an amphibian species,
and examined the link between the biophysical data and the
bactericidal activities of the peptides [51]. A systematic
study of the effect of different cationic AMP families,
including α-helical, β-sheet, extended, or cyclic structures
examined lipid disruption and calcein leakage from LUVs
comprised mostly of negatively-charged phospholipids [52].
The experiments assessed peptide translocation across the
lipid bilayers and the induction of lipid “flip-flop” processes
at a range of peptide/lipid ratios, however no common
structural peptide feature was identified as a central
determinant for bactericidal activity.
Beside of monitoring dye leakage, numerous experiments
have recorded the fluorescence of trypthophane residues
within the AMP sequences, either Trp residues within the
native sequences or inserted intentionally. The method is
based on the high sensitivity of the position and intensity of
the fluorescence peak to the hydrophobicity of the Trp
environment, thus providing a measure of the depth of
bilayer insertion of the peptide [53, 54]. Other fluorescence
techniques have utilized dyes embedded within the lipid
bilayer, either physically incorporated or covalently bound to
the lipid moieties. Such experiments were similarly designed
to evaluate the penetration and bilayer localization of the
peptide. 8-anilino-1-naphthalene sulfonic acid (ANS) has
been a common fluorescence probe for studying the effect of
AMPs [55]. Another example is the use of DPH- and pyrenelabeled phospholipids [56]. These fluorescence dyes were
used for studying temporin-peptide interactions and revealed
that dynamical modifications were induced by the peptides at
specific parts within the lipid bilayer [56]. The researchers
detected that peptide penetration was enhanced by
negatively-charged bacterial phospholipids, while an
opposite effect was observed when the vesicles contained the
eukaryotic lipid cholesterol.
Indeed, one of the most important questions pertaining to
the activities of host-defense peptides, studied extensively in
unilamellar vesicle systems, concerns their specificity to
bacterial rather than host-cell membranes. Several
hypotheses emphasize AMP affinity to membranes
containing an abundance of negatively-charged phospholipids [52, 57, 58], while other models focus on specificity
towards particular components and domains on bacterial
walls [59], or binding to molecular constituents on outer
bacterial surfaces, such as lipopolysaccharides (LPS) [60]. A
comparative study of magainin 2 penetration into LPScontaining SUVs vs. its insertion into LPS-free vesicles
determined that LPS is most likely the primary molecular
determinant for binding and bacterial lysis by this peptide
[60]. Experiments utilizing LUVs formed exclusively from
the anionic lipid palmitoyloleoylphosphatidylglycerol
(POPG) suggested that multimeric pore formation is the
predominant bactericidal mechanism of human defensins
[61].
Unilamellar vesicles have been prepared not Just from
specific phospholipid units but also from total lipid extracts
from varied microorganisms. Hristova et al discovered that
rabbit neutrophil defensins permeate LUVs composed of E.
coli lipid extracts to a significantly higher degree compared
to vesicles constructed from mixtures of distinct components
[62]. The researchers proposed that the overall composition
of the bacterial membrane, in particular the presence and
ratios between phosphatidylethanolamine, phosphatidylglycerol, and cardiolipin, play critical roles in affecting the
bactericidal profiles of the peptides.
Membrane Interactions of Host-defense Peptides Studied in Model Systems
Valuable insights into the biological profiles and modes
of action of AMPs could be gained from analysis of
differences between the peptide activities as measured in
model membranes (such as unilamellar vesicles) and their
effects on intact bacterial cells. For example, significant
differences in activities were observed between the
interactions of nisin with unilamellar vesicles and its effects
on varied microorganisms [63, 64]. Intriguingly, the
differences were traced to the presence of lipid II on the
bacterial surface - a putative precursor for high affinity
binding of nisin and its increased activity towards bacterial
cells.
Many investigations tailored the choice of vesicle type to
the technique applied and the information desired. Epand et
al, for example, characterized cytolytic peptides composed of
leucines and lysines through application of CD spectroscopy
on SUVs, while LUVs were prepared for determination of
induced leakage [65]. While most spectroscopic studies of
peptide insertion into unilamellar vesicles focused on SUVs
or LUVs as the preferred model systems, giant unilamellar
vesicles (GUVs) have been useful for actual visualization of
the topological and morphological effects of AMPs. Optical
microscopy was employed for studying the effects of
magainin II and indolicidin on GUVs [66]. The experiments
demonstrated that magainin 2 had essentially no influence on
GUVs that contained either zwitterionic phospholipids or
zwitterionic/cholesterol mixtures – echoing the minimal
hemolytic capabilities of this peptide. Contraction and
fractionation were observed, however, when acidic
phospholipids were present in the giant vesicles. Similar
shrinkage effects were detected when indolicidin was added
to GUVs composed of zwitterionic phospholipids and
cholesterol, pointing to the cytotoxic properties of this
particular peptide [66].
3. COLORIMETRIC BIO-MIMETIC MEMBRANE
ASSEMBLIES
A recent development in the field of biomimetic
membrane systems has been the introduction of mixed
vesicles composed of physiological lipids and
Current Protein and Peptide Science, 2005, Vol. 6, No. 1
5
polydiacetylene (PDA) that exhibit unique chromatic
properties. This novel platform could be used for studying
and rapid screening of membrane processes in general, and
for analysis of AMP-membrane interactions in particular.
The PDA polymer is formed through 1,4 addition of the
diacetylenic monomers, initiated by UV irradiation, and is
intensely blue to the naked eye due to absorption of light at
around 650 nm in the visible region of the electromagnetic
spectrum [66]. ConJugated PDA vesicles have attracted
increasing attention due to their versatile electronic and
optical properties [67]. Specifically, rapid blue-red
transformations of the PDA can be induced by varied
external perturbations, such as temperature, surface pressure,
ion concentration, and others [68-70].
Relevant to biological membrane applications, it was
recently demonstrated that mixed vesicles can be prepared
that contain both PDA as well as varied lipid species.
Biophysical analyses confirmed that such assemblies
comprise of phospholipid bilayers interspersed within the
framework of the PDA without affecting the unique
chromatic properties of the polymer [71]. Fig. 3 depicts a
schematic description of the lipid/PDA vesicle assemblies,
showing the mixed domains of phospholipids and the
polymer. Several studies have shown that interactions of
membrane-active peptides with the mixed vesicles induce
dramatic colorimetric transitions [72-77]. The visible
transformations of the vesicles could both report upon the
occurrence of peptide-membrane binding as well as provide
information on the mechanisms of peptide interaction with
the lipid bilayer.
The new lipid/PDA platform exhibits several important
advantages for use as an assay for studying peptidemembrane interactions and processes, the most significant
being the presence of organized lipid bilayers – the
biomimetic membrane target - incorporated within the PDA
matrix which acts as a colorimetric reporter. Importantly, it
was shown that the lipid domains within the mixed vesicles
can be composed of a variety of synthetic and natural
phospholipids, glycolipids, lipopolysaccharides, cholesterol,
and other molecules, and that these mixtures functionally
Fig. (3). Schematic picture of a lipid/polydiacetylene chromatic vesicle. The polymer domains are shown in blue, while lipid domains in
green.
6
Current Protein and Peptide Science, 2005, Vol. 6, No. 1
mimic biological membrane platforms [71]. Moreover,
diverse biological processes that occur exclusively at the
lipid bilayer moieties induce the visible colorimetric
transformations within the adJacent PDA.
The phospholipid/PDA assay has been applied for
studying varied aspects of AMP-membrane interactions and
AMP biological action. Underlying the color change
mechanism is the disruption of the lipid interface by
membrane-interacting compounds and the depth of
penetration into the lipid layer [72-74]. In particular,
correlation was observed between the relative perturbation of
the lipid bilayer surface by membrane-active compounds and
the degree of color changes [72-74, 76, 77]. Peptides that
preferably disrupt the lipid head-group region were shown to
induce more pronounced color transitions, while deeper
penetration into the hydrophobic bilayer core generally gave
rise to more moderate blue-red transformations [73]. These
structural relationships were confirmed through applications
of analytical and spectroscopic techniques [72-77].
Specifically, the colorimetric transitions observed in the
PDA assemblies have been ascribed to the perturbation of
the pendant side-chains and modification of molecular
packing of the PDA framework induced by the interfacial
effects [71, 74]. Thus, greater lipid surface interactions
would result in more pronounced perturbations of the
adJacent polymer domains, giving rise to higher colorimetric
transitions.
The new colorimetric assay has been applied to
characterize membrane interactions and relate those
interactions to the biological activities of several AMP
families. Several studies elucidated parameters such as
bilayer localization and the relationship with the
antimicrobial and cytolytic properties of melittin and its
derivatives [72, 73], polymyxin B analogues [74], and
human defensins [76, 77]. A colorimetric study,
complemented with advanced fluorescence techniques,
recently determined the extent of interface localization of
model AMPs within the biomimetic vesicles [76, 77].
Phospholipid/PDA vesicles have also exhibited sensitivity to
synergy effects occurring in certain antimicrobial peptide
families [74]. The colorimetric membrane assay can also
provide important insight into the cellular pathways affecting
the bactericidal activity of host defense peptides,
experiments utilizing the phospholipid/PDA platform
demonstrated that proteolytic cleavage of the prosegment of
human defensin cryprdin-4 is essential for conferring
membrane interaction to the mature biologically-active
peptide [76].
4. PLANAR LIPID SYSTEMS
Planar lipid systems have been employed in varied
applications for studying peptide association with
membranes. Among the assemblies constructed were
Langmuir monolayers [78], Langmuir-Blodgett films [79,
80], supported planar bilayers [81, 82], self-assembled
monolayers [83, 84], polymer-tethered membranes [85, 86],
and black lipid membranes [87, 88]. Planar systems
generally model the phospholipid ordering within cellular
membranes, and can be also designed to mimic the lateral
organization of cell surfaces [89]. The use of planar lipid
Jelinek and Kolusheva
models for studying peptide-membrane interactions often
stems from the availability of varied experimental techniques
for surface characterization, including microscopy methods
[90], scattering [85], thermodynamic analysis [91, 92], and
others.
4.1 Langmuir Monolayers
Phospholipid monolayers deposited at the air/water
interface of aqueous solutions (Langmuir monolayers, Fig. 4)
have been used as model membrane surfaces for studying
peptide interactions [92]. In general, investigations using
such systems have yielded information both on the
cooperative properties of the lipid-interacting peptides, as
well as the organization and disruption of the lipid layers
following AMP interactions. In many instances, physical and
morphological analyses of Langmuir film / AMPs assemblies
have been conducted by thermodynamic methods (pressurearea isotherms) and microscopy techniques (fluorescence
microscopy, Brewster angle microscopy). A systematic study
of a series of bombolitins, bumblebee-venom peptides, and
their synthetic analogs employed Langmuir monolayers
coupled with fluorescence microscopy to investigate the
structural disruption and phase separation induced by the
incorporated peptides within the phospholipid monolayers
[93]. Zhang et al investigated the relationships between the
antibacterial activities and membrane interactions of several
variants of the horseshoe crab antimicrobial peptide
polyphemusin I, in which some of the analogues displayed
entirely different amphipathic properties [94]. Langmuir
monolayer analysis showed that the degree of negative
charge on the membrane surface was closely related to
bacterial membrane permeabilization. Importantly, binding
of the peptides to monolayers composed solely of LPS
pointed to preferential peptide binding to the saccharide
headgroups, and re-emphasized the significance of
electrostatic interactions for promoting peptide insertion into
the external leaflet of the bacterial membrane.
Other experimental approaches for studying AMP
association with Langmuir monolayers have been described.
A synchrotron grazing incidence diffraction and X-ray
reflectivity study of frog skin antimicrobial peptides added to
Langmuir monolayers was reported [95]. The study
concluded that the peptides were fully interspersed within
negatively-charged phospholipids but formed distinct
domains in zwitterionic monolayers, suggesting that AMPs
can discriminate among the maJor phospholipid components
of bacterial and mammalian membranes. Recent
developments have enabled the application of surfacesensitive X-ray and neutron scattering techniques for
characterization of molecularly thin peptide crystal sheets
constructed via Langmuir monolayers [85].
4.2 Solid-supported Lipid Systems
Solid-supported lipid monolayers and bilayers (Fig. 5)
have been popular biomimetic membrane systems since they
allow examining peptide interactions at the lipid/water
interface while at the same time limiting lipid motion
through surface immobilization. Constricting lipid mobility
in supported lipid layers might make such systems a more
realistic model compared to Langmuir films since this better
Membrane Interactions of Host-defense Peptides Studied in Model Systems
Fig. (4). Schematic picture of a Langmuir monolayer.
Fig. (5). Schematic picture of a hybrid bilayer system.
Current Protein and Peptide Science, 2005, Vol. 6, No. 1
7
8
Current Protein and Peptide Science, 2005, Vol. 6, No. 1
resembles lipid dynamics within real membranes. In
addition, supported bilayers on solid substrates have become
particularly useful for measuring secondary structures and
orientations of peptides in lipid environments by using
spectroscopic techniques such as attenuated total reflection
Fourier transform IR (ATR-FTIR) [78, 96].
Solid-supported lipid bilayers served as a platform for
studying the effect of AMPs on the cooperative properties of
the lipids. As an example, planar lipid monolayers with
different compositions were constructed for investigation of
the bactericidal mechanism of protegrin-1 (PG-1) [58]. The
researchers applied epifluorescence microscopy and surface
x-ray scattering in order to examine whether the strong
surface interactions of PG-1 indeed depend on the lipid
composition of the membrane, and whether such interactions
affect the lipid order. The analysis showed a pronounced
monolayer destabilization by the peptide when the lipid
assembly consisted mainly of lipid A - the maJor constituent
of the outer membrane of Gram negative bacteria [97, 98].
Many studies have applied electrophysiological
techniques for characterizing AMP-membrane interactions
using reconstituted lipid bilayers on solid surfaces. A recent
study revealed that several α-helical CAP18-derived
peptides induced lesion formation within different types of
artificial membranes at clamp voltages below the
transmembrane potential of the natural membrane [99]. That
study suggested that the interplay between the
physicochemical properties of both the peptides as well as
the target membranes is an important factor for
determination of the antibacterial activity. A similar
electrophysiological study of the action of Hagfish intestinal
antimicrobial peptides reported disordered current
fluctuations following addition of the peptides to supported
lipid bilayers, which was interpreted as an evidence for pore
formation induced by the peptides [47]. A lipid multibilayer
assembly deposited on a polyion/alkylthiol solid-stabilized
layer was fabricated for studying the topological and
electrochemical consequences of the insertion of protegrin-1
(PG-1) into membranes [83]. This intriguing approach
explored the relationship between peptide binding, pore
formation, and modification of surface roughness by the
adsorbed peptide using cyclic voltametry, surface plasmon
resonance (SPR, see below) and atomic force microscopy
(AFM).
Oriented lipid layers have been useful models for
providing structural information on peptide conformations in
membrane and their insertion processes. Such assemblies
have facilitates the use of spectroscopic techniques primarily
solid-state NMR for extraction of structural parameters.
Specifically, application of solid-state NMR on oriented
samples yields varied structural parameters that cannot be
extracted in conventional solid samples [14, 100]. The
orientation of the peptide molecules within the organized
lipid assemblies reduces anisotropic broadening, thereby
simplifying analysis of the NMR spectra while additionally
allows extraction of various distance and angle parameters
pertaining to the peptide. Solid-state NMR applied to
oriented bilayers provided structural analyses of membraneincorporated AMPs that included transmembrane helical
bundles, wormholes, carpets, as well as detergent-like effects
Jelinek and Kolusheva
or the in-plane diffusion of peptide-induced bilayer
instabilities [14].
Oriented planar bilayers have been also used in other
experimental approaches. CD experiments on alamethicin
and magainin-1 in oriented lipid layers yielded interesting
insights into transitions of the peptides from surface-bound
into transmembrane structures [101]. Oriented lipid
multibilayers were also employed in neutron reflectometry
experiments. Neutron reflectometry indicated that magainin
II significantly modified the thermal fluctuations of the lipid
layers [102].
5.3 Other Planar Lipid Models
Various other planar models have been developed for
studying AMP-membrane interactions. Black lipid
membranes (BLMs) constitute of thin lipid films placed
across small apertures separating two chambers containing
ionic solutions (Fig. 6) [103]. The BLM technique relies
upon conductance measurements to study the formation of
pores and channels within the model membrane and their
relative permeability. The use of BLMs also provides
information on the ionic selectivity of the pores/channels
formed by peptides. BLMs were instrumental in exposing the
formation of permanent and transient channels by salmon
and human calcitonins, host defense peptides that exhibit
remarkable therapeutic capabilities [104]. The experiments
also indicated that the peptide-induced channels were equally
permeable to cations and anions.
Tethered lipid bilayer membranes conceptually resemble
BLMs as a planar biomimetic model system [105]. The
artificial membrane in this arrangement is created by
covalently binding a lipid monolayer onto gold electrodes,
followed by deposition of another lipid film resulting in the
formation of a tethered bilayer [106]. Incorporation of
channel-forming AMPs into tethered bilayers has been
demonstrated, and electrical impedance spectroscopy was
applied for studying channel formation induced by the
peptides and screening of blocking inhibitors [105].
A planar asymmetric bilayer designed for electrical
application has been developed [107]. The assembly consists
of one leaflet composed solely of LPS and the other layer
containing phospholipids of bacterial plasma membranes.
The goal of this design was to examine the role of LPS in
targeting of antibacterial peptides to invading bacterial cells.
The researchers mostly applied electrical measurements to
study the association of the bactericidal/permeabilityincreasing protein (BPI) with the reconstituted bilayer, as a
representative model for bacterial membrane interactions of
common AMPs such as cathelicidins, defensins, and
polymyxin B (PMB). One of the main conclusions drawn
from these studies was that the negative charges of LPS are
most likely the primary determinants for peptide-membrane
interactions. However, the detailed mechanisms of bacterial
killing may involve further steps, which could differ for
specific AMPs.
Hybrid bilayer systems have been the main experimental
arrangement for studying AMP-membrane interactions using
surface plasmon resonance (SPR), a technique that is
increasingly applied for characterizing peptide binding and
Membrane Interactions of Host-defense Peptides Studied in Model Systems
Current Protein and Peptide Science, 2005, Vol. 6, No. 1
9
Fig. (6). Schematic picture of a black lipid membrane (BLM) system.
association with lipid bilayers [108]. Sample organization for
SPR experiments usually consist of covalently-attached selfassembled hydrophobic films (in most instances alkane-thiol
monolayers), onto which lipid SUVs are adsorbed, putatively
forming homogeneous monolayers [4]. The preformed lipid
monolayers could also include receptor molecules and
proteins [109]. SPR analysis applied to hybrid bilayers
containing mucopeptides detected a correlation between the
binding strengths of several glycopeptides and their
antimicrobial activities [109]. A detailed SPR investigation
explored the differences between pore formation and
detergent-like membrane micellization (the “carpet model”)
as the two membrane permeation mechanisms of cytotoxic
peptides [110]. That study characterized and compared the
binding profiles of native melittin on the one hand and
diastereomeric melittin (D-melittin) on the other hand. These
two peptides represented the two models for membrane
interactions with zwitterionic vs negatively-charged
phospholipid assemblies, and indeed the SPR experiments
demonstrated distinct behavior of the two peptides in the
membrane models constructed. The dependence of binding
affinity upon membrane composition of the hemolytic
peptide lysenin was similarly uncovered by a kinetic SPR
analysis [111].
SPR experiments have been carried out not only on
phospholipid monolayer arrangements but also on supported
lipid bilayers [110]. Such comparisons are important since
they allow exploration of the affinity of AMPs to the outer
membrane surface vs. the inner bilayer leaflet, since the lipid
compositions of these monolayers is generally different in
many cells. In particular, the quantitative ratio between the
SPR-derived binding constants of specific peptides to the
lipid bilayer and the monolayer, respectively, could give
indications on the peptide's binding process and the depth of
penetration into the membrane core [110].
5. Micelles
Micellar assemblies are typically formed in aqueous
solutions by organic co-solvents, lysophospholipids, and
detergents below their critical micelle concentrations
(CMC's). Among the most popular micelle-forming
compounds
have
been
trifluoroethanol
(TFE),
dodecylphosphocholine (DPC), and sodium dodecyl sulfate
(SDS). Despite the fact that micelles constitute of single lipid
or lipid-like layers rather than bilayers, and their surface
curvatures are significantly higher than those encountered in
real cellular membranes, micelles have been extensively
used as biomimetic systems for membrane environments and
for deciphering structural features of host-defense peptides.
One of the advantages of micelles for studying peptidemembrane association has been the observation that such
assemblies often stabilize a single peptide conformation,
rather than an ensemble of structures [112]. Furthermore,
rapid molecular tumbling of micelles in aqueous solutions
has facilitated structural characterization of the incorporated
peptides using important bio-analytical techniques such as
high-resolution NMR, Fourier transform infrared
spectroscopy (FTIR), and circular dichroism (CD).
High-resolution solution NMR spectroscopy, in
particular, has been applied to determine the conformations
of AMPs in micelle environments [15, 113]. NMR studies
have generally suggested that peptide conformations induced
through micelle binding resemble the physiological
membrane-bound structures. Even though this claim is
probably reliable only for simple secondary structural
elements, in many cases the NMR structural information was
important in providing information on the bactericidal and
hemolytic activities of the peptides, aiding efforts for rational
design of improved analogs for Clin.ical use.
Two-dimensional 1H NMR was applied to determine the
structure of micelle-associated alamethicin [114], arginine-
10
Current Protein and Peptide Science, 2005, Vol. 6, No. 1
tryptophane sequence [25], and others. Circular dichroism
(CD) spectroscopy was also applied for studying peptide
conformations in micelles, in particular exposing the
differences between micelle-bound vs. aqueous solutions [4].
In that regard, CD analyses have mostly confirmed the
prominence of helical conformations of AMPs in micelle
environments.
CD
experiments
carried
out
in
lysophospholipid micelles probed the effect of helix
destabilization in a series of doubly D-residue labeled
sequences [115].
6. Miscellaneous Model Systems
A variety of additional membrane-mimicking systems
designed to elucidate structure-function relationships of hostdefense peptides was described in the literature. Similar to
the techniques described above, the goals for introducing
new model systems have been both to reconstitute
lipid/protein assemblies that would mimic a cell membrane,
as well as facilitating application of specific bio-analytical
techniques for probing peptide-membrane interactions. A
recent development, inspired partly by NMR studies of
micelles, was the introduction of oriented phospholipid
aggragates, or “bicelles”. These magnetically-oriented
phospholipid/detergent micelles facilitate high-resolution
structure determination of membrane-associated peptides and
proteins using advanced NMR techniques [116].
Other model systems consisting of membrane lipids and
proteins have been tailored for specific bio-analytical
techniques. Poly(ethylene glycol)-lipid conJugates were
found to promote bilayer formation in various lipid systems
[117].
Reversed-phase
high
performance
liquid
chromatography (RP-HPLC) was applied to evaluate peptide
conformations at hydrophobic-hydrophilic interfaces [4, 118,
119]. In such experiments, the hydrocarbon moieties within
the retention column mimicked the lipid assemblies,
inducing conformational changes of the peptides in the
mobile phase [4, 119]. RP-HPLC analyses could further
probe in detail the relative affinities of membrane-active
peptides to different phospholipid compositions [4].
The obvious goal of most studies utilizing model systems
has been to best mimic the “real cellular world”.
Accordingly, there has been an ongoing effort to evaluate
AMP activities in whole cell systems. It could be argued that
such analyses constitute less of “model systems” more fitting
the description of microbiological assays, however several
elegant studies thread nicely the line between the model and
the actual biological assembly (see further the Review by A.
Tossi in this volume). The laboratory of H.-G. Sahl, for
example, has introduced several whole-cell approaches for
determination of bacterial permeation kinetics and
mechanisms of varied AMPs [120,121].
CONCLUSIONS
Biomimetic membrane models have laid the groundwork
for deciphering many biophysical and mechanistic aspects of
the antibacterial properties of host defense peptides. Various
proposed mechanisms for AMP action have originated from
model system studies, including the “barrel stave” and
“carpet” mechanisms, the significance of electrostatic
attraction between the negative membrane-surface charge
Jelinek and Kolusheva
and cationic AMPs, the induction of transient holes and
permanent pores in lipid bilayers by AMPs, and other
models.
Diverse bio-analytical, spectroscopic, and microscopic
techniques were applied for studying host defense peptides
in model membranes, aiming to elucidate the characteristics
of peptide-mediated cell lysis. The experimental approaches
have provided a wealth of information upon the extent and
significance of pore formation, the contributions of
headgroup interactions vs. degree of peptide insertion into
the membranes, the structure and orientation of AMPs in the
membrane environment, and other factors believed to be
central to the biological actions of AMPs. It is important to
emphasize, however, that while in many cases important
correlations between peptide-membrane interactions and
peptides’ biological activities were detected in model
systems, other works have unearthed the absence of such
relationships – pointing to putative molecular targets of
AMPs either on the membrane surface of the intact bacteria,
or inside the bacterial cells. Furthermore, it has been often
demonstrated that high peptide activity observed in model
membranes is not necessarily predictive of its actual
biological activity and vice versa.
Overall, the availability of diverse biomimetic membrane
models, each illuminating different aspects of peptidemembrane interactions, and the considerable amount of
experimental data accumulated using artificial membrane
models have provided us with a comprehensive insight into
the properties and bactericidal activities of AMPs. The
molecular and functional information acquired could further
help design new therapeutic substances with improved
antimicrobial properties and reduced toxicity.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
Lehrer, R.I., Barton, A., Daher, K.A., Harwig, S.S., Ganz, T.,
Selsted, M.E. (1989) J. Clin. Invest., 84, 553.
Papo, N., Shai, Y. (2003) Peptides, 24, 1693.
Sansom, M.S.P. (1998) Current Opinion in Colloid & Interface
Science, 3, 518.
Blondelle, S.E., Lohner, K., Aguilar, M.-I. (1999) Biochimica et
Biophysica Acta, 1462, 89.
Shai, Y. (1999) Biochimica et Biophysica Acta, 1462, 55.
Dathe, M., Wieprecht, T. (1999) Biochimica et Biophysica Acta,
1462, 71.
Tossi, A., Sandri, L. , iangaspero, A. (2000) Biopolymers, 55, 4.
Shai, Y. (2002) Biopolymers, 66, 236.
Hancock, R.E.W., Rozek, A. (2002) FEMS Microbiology Letters,
206, 143.
Epand, R.M., Epand, R. F. (2003) Methods in Enzymology, 372,
124.
Lee, D.K., Santos, J. S., Ramamoorthy, A. (1999) Journal of
Physical Chemistry B, 103, 8383.
Hori, Y., Demura, M., Niidome, T., Aoyagi, H., Asakura, T. ( 1999)
FEBS Letters, 455.
Picard, F., Paquet, M.-J., Dufourc, E. J., Auger, M. (1998)
Biophysical Journal, 74, 857.
Bechinger, B. (1999) Biochimica et Biophysica Acta, 1462, 157.
Hwang, P.M., Vogel, H. J. (1998) Biochemistry and Cell Biology,
76, 235.
Anderluh, G., Serra, M. D., Viero, G., Guella, G., Macek, P.,
Menestrina, G. (2003) Journal of Biological Chemistry, 278,
45216.
Buffy, J.J., Waring, A. J., Lehrer, R. I., Hong, M. (2003)
Biochemistry, 42, 13725.
Buffy, J.J., Hong, T., Yamaguchi, S., Waring, A. J., Lehrer, R. I.,
Hong, M. (2003) Biophysical Journal, 85, 2363.
Membrane Interactions of Host-defense Peptides Studied in Model Systems
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
Marcotte, I., Wegener, K. L., Lam, Y.-H., Chia, B. C. S., De
Planque, M. R. R., Bowie, J. H., Auger, M., Separovic, F. (2003)
Chemistry and Physics of Lipids, 122, 107.
Bechinger, B. (1997) J. Membrane Biol., 156, 197.
Wakamatsu, K., Takeda, A., Tachi, T., Matsuzaki, K. (2002)
Biopolymers, 64, 314.
Lohner, K., Staudegger, E., Prenner, E. J., Lewis, R. N. A. H.,
Kriechbaum, M., Degovics, G., McElhaney, R. N. (1999)
Biochemistry, 38, 16514.
Prenner, E.J., Lewis, R. N. A. H., KondeJewski, L. H., Hodges, R.
S., McElhaney, R. N. (1999) Biochimica et Biophysica Acta , 1417,
211.
Lohner, K., Prenner, E. J. (1999) Biochimica et Biophysica Acta,
1462, 141.
Jing, W., Hunter, H. N., Hagel, J., Vogel, H. J. (2003) Journal of
Peptide Research, 61, 219.
Grasso, D., Milardi, D., La Rosa, C., Impellizzeri, G., Pappalardo,
G. (2002) Thermochimica Acta, 390, 73.
Jing, W., Strom, M. B., Svendsen, J. S., Vogel, H. J. (2002)
Biochemistry and Cell Biology, 80, 161.
Epand, R.F., Umezawa, N., Porter, E. A., Gellman, S.H., Epand,
R.M. (2003) European Journal of Biochemistry, 270, 1240.
Yang, L., Weiss, T. M., Lehrer, R. I., Huang, H. W. (2000)
Biophysical Journal, 79, 2002.
Lewis, R.N.A.H., Prenner, E.J., KondeJewski, L. H., Flach, C. R.,
Mendelsohn, R., Hodges, R. S., McElhaney, R. N. (1999)
Biochemistry, 38, 15193.
Mihailescu, D., Horvath, L.I. (1998) Eur. Biophys., 28, 216.
Katsu, T., Kobayashi, H., FuJita, Y. (1986) Biochimica et
Biophysica Acta, 860, 608.
Prenner, E.J., Lewis, R. N.A.H., McElhaney, R. N. (1999)
Biochimica et Biophysica Acta, 1462, 201.
Susi, H., Sampugna, J., Hampson, J.W., Ard, J.S. (1979)
Biochemistry, 18, 297.
Staudegger, E., Prenner, E. J., Kriechbaum, M., Degovics, G.,
Lewis, R. N. A. H., McElhaney, R. N., Lohner, K. (2000)
Biochimica et Biophysica Acta, 1468, 213.
Prenner, E.J., Lewis, R. N. A. H., Jelokhani-Niaraki, M., Hodges,
R. S., McElhaney, R. N. (2001) Biochimica et Biophysica Acta,
1510, 83.
El Jastimi, R., Edwards, K., Lafleur, M. (1999) Biophysical
Journal, 77, 842.
Vanlerberghe, G. (1994) Oleagineux, Corps Gras, Lipides, 1, 174.
De Kroon, A.I.P.M., de Gier, J., de KruiJff, B. (1993) New
Comprehensive Biochemistry, 25, 107.
Hope, M.J., Nayar, R., Mayer, L. D., Cullis, P. R. (1993) Liposome
Technol., 1, 123.
Lichtenberg, D., Barenholz, Y. (1988) Methods of Biochemical
Analysis, 33, 337.
Kahya, N., Pecheur, E. I., de BoeiJ, W. P., Wiersma, D. A.,
Hoekstra, D. (2000) Biophysical Journal, 81, 1464.
Matsuda, R., Kaneko, N., Horikawa, Y. (1997) Biochemical and
biophysical research communications, 237, 499.
Jelokhani-Niaraki, M., KondeJewski, L. H., Chen, Y., Hancock, R.
E. W., Kay, C. M., Hodges, R. S. (1999) Peptide Science, 35, 81.
Mangoni, M.L., Papo, N., Mignogna, G., Andreu, D., Shai, Y.,
Barra, D., Simmaco, M. (2003) Biochemistry, 42, 14023.
Marsh, D. (2002) Biochemical Society transactions, 29, 582.
Basanez, G., Shinnar, A. E., Zimmerberg, J. (2002) FEBS Letters,
532, 115.
Blazyk, J., Wiegand, R., Klein, J., Hammer, J., Epand, R. M.,
Epand, R.l F., Maloy, W. L., Kari, U. P.. (2001) Journal of
Biological Chemistry, 276, 27899.
Oren, Z., Shai, Y. (2000) Biochemistry, 39, 6103.
Ladokhin, A.S., Wimley, W. C., Hristova, K., White, S. H. (1997)
Methods in Enzymology, 278, 474.
Mangoni, M.L., Rinaldi, A. C., Di Giulio, A., Mignogna, G., Bozzi,
A., Barra, D., Simmaco, M. (2000) Eur. J. Biochem., 267, 1447.
Zhang, L., Rozek, A., Hancock, R. E. (2001) Journal of biological
chemistry, 276, 349.
Zhao, H., Kinnunen, P. K. J. (2002) Journal of Biological
Chemistry, 277, 25170.
Matsuzaki, K., Fukui, M., FuJii, N., MiyaJima, K. (1991)
Biochimica et Biophysica Acta, 1070, 259.
Yagi, Y., Kimura, S., Imanishi, Y. (1990) Int. J. Pept. Prot. Res.,
36, 18.
Current Protein and Peptide Science, 2005, Vol. 6, No. 1
[56]
[57]
[58]
[59]
[60]
[61]
[62]
[63]
[64]
[65]
[66]
[67]
[68]
[69]
[70]
[71]
[72]
[73]
[74]
[75]
[76]
[77]
[78]
[79]
[80]
[81]
[82]
[83]
[84]
[85]
[86]
[87]
[88]
[89]
[90]
[91]
[92]
[93]
[94]
[95]
11
Zhao, H., Rinaldi, A. C., Di Giulio, A., Simmaco, M., Kinnunen, P.
K. J. (2002) Biochemistry, 41, 4425.
Peschel, A. (2002) Trends in Microbiology, 10, 179.
Gidalevitz, D., Ishitsuka, Y., Muresan, A. S., Konovalov, O.,
Waring, A. J. , Lehrer, R. I., Lee, K. Y. C. (2003) PNAS, 100,
6302.
Matsuzaki, K., Sugishita, K.-i., Harada, M., FuJii, N., MiyaJima, K.
(1997) Biochimica et Biophysica Acta, 1327, 119.
Matsuzaki, K., Sugishita, K.-i., MiyaJima, K. (1999) FEBS Letters,
449, 221.
Wimley, W.C., Selsted, M.E., White, S. H.. (1994) Protein Science,
3, 1362.
Hristova, K., Selsted, M. E., White, S. H. (1997) Journal of
Biological Chemistry, 272, 24224.
Breukink, E., de KruiJff, B. (1999) Biochimica et Biophysica Acta ,
1462, 223.
van Heusden, H.E., de KruiJff, B., Breukink, E. (2002)
Biochemistry, 41, 12171.
Epand, R.F., Lehrer, R. I., Waring, A., Wang, W., Maget-Dana, R.,
Lelievre, D., Epand, R. M.. (2003) Biopolymers, 71, 2.
Zhao, H., Mattila, J.-P., Holopainen, J. M., Kinnunen, P. K. J.
(2001) Biophysical Journal, 81, 2979.
Ringsdorf, H., Schlarb, B., Venzmer, J. (1988) Angew. Chem., Int,
27, 113.
Tanaka, H., Gomez, M.A., Tonelli, A.E., Thakur, M. (1989)
Macromolecules, 22, 1208.
Wenzel, M., Atkinson, G.H. (1989) J. Am. Chem. Soc., 111, 4123.
Chance, R.R. (1980) Macromolecules, 13, 396.
Kolusheva, S., Wachtel, E., Jelinek, R. (2003) Journal of Lipid
Research, 44, 65.
Kolusheva, S., Boyer, L., Jelinek, R. (2000) Nature Biotechnology,
18, 225.
Kolusheva, S., Shahal, T., Jelinek, R. (2000) Biochemistry, 39,
15851.
Katz, M., Tsubery, H., Kolusheva, S., Shames, A., Fridkin, M.,
Jelinek, R. (2003) Biochemical Journal, 375, 405.
Rozner, S., Kolusheva, S., Cohen, Z., Dowhan, W., Eichler, J.,
Jelinek, R. (2003) Analytical Biochemistry, 319, 96.
Satchell, D.P., Sheynis, T., ShirafuJi, Y., Kolusheva, S., Ouellette,
A. J., Jelinek, R. (2003) Journal of Biological Chemistry, 278,
13838.
Sheynis, T., Sykora, J, Benda, A., Kolusheva, S., Hof, M., Jelinek,
R. (2003) European Journal of Biochemistry, 270, 4478.
Tamm, L.K. (2002) Current Topics in Membranes, 52, 191.
Fragneto-Cusani, G. (2001) Journal of Physics: Condensed Matter,
13, 4973.
Gabrielli, G. (1991) Advances in colloid and interface science, 34,
31.
Baumgart, T., Offenhaeusser, A. (2003) Langmuir, 19, 1730.
Frey, S., Tamm, L. K. (1991) Biophysical Journal, 60, 922.
Zhang, L., Vidu, R., Waring, A. J., Lehrer, R. I., Longo, M. L.,
Stroeve, P. (2002) Langmuir, 18, 1318.
Noinville, S., Bruston, F., El Amri, C., Baron, D., Nicolas, P.
(2003) Biophysical Journal, 85, 1196.
Losche, M. (2002) Current Topics in Membranes, 52, 117.
Sinner, E.-K., Knoll, W. (2001) Current Opinion in Chemical
Biology, 5, 705.
Gabrielli, G., Caminati, G., Puggelli, M., Gilardoni, A., Margheri,
E. (1993) Journal de Physique IV: Proceedings, 3, 279.
Tien, H.T., Diana, A. Louise. (1968) Chemistry and Physics of
Lipids, 2, 55.
Greenhall, M.H., Yarwood, J., Brown, R., Swart, R. M. (1998)
Langmuir, 14, 2619.
Chen, H.M., Leung, K. W., Thakur, N. N., Tan, A., Jack, R. W.
(2003) European Journal of Biochemistry, 270, 911.
Thomas, C.J., Surolia, N., Surolia, A. (2001) Journal of Biological
Chemistry, 276, 35701.
Maget-Dana, R. (1999) Biochimica et Biophysica Acta, 1462, 109.
Peggion, E., Mammi, S., Schievano, E. (1998) Biopolymers, 43,
419.
Zhang, L., Scott, M. G., Yan, H., Mayer, L. D., Hancock, R. E. W.
(2000) Biochemistry, 39, 14504.
Konovalov, O., Myagkov, I., Struth, B., Lohner, K. (2002)
Biophysics Journal, 31, 428.
12
[96]
[97]
[98]
[99]
[100]
[101]
[102]
[103]
[104]
[105]
[106]
[107]
Current Protein and Peptide Science, 2005, Vol. 6, No. 1
Axelsen, P.H., Braddock, W. D., Brockman, H. L., Jones, C. M.,
Dluhy, R. A., Kaufman, B. K., Puga, F. J., II. (1995) Applied
Spectroscopy, 49, 526.
Alexander, C., Rietschel, E. Th. (2001) Journal of Endotoxin
Research, 7, 167.
Yethon, J.A., Whitfield, C. (2001) Current Drug Targets:
Infectious Disorders, 1, 91.
Gutsmann, T., Hagge, S. O., Larrick, J. W., Seydel, U., Wiese, A.
(2001) Biophysical Journal, 80, 2935.
Marassi, F.M., Opella, S. J., Juvvadi, P., Merrifield, R. B. (1999)
Biophysical Journal, 77, 3152.
Ludtke, S.J., He, K., Heller, W. T., Harroun, T. A., Yang, L.,
Huang, H. W. (1996) Biochemistry, 35, 13723.
Salditt, T., Muenster, C., Mennicke, U., Ollinger, C., Fragneto, G.
(2003) Langmuir, 19, 7703.
Winterhalter, M. (2000) Current Opinion in Colloid & Interface
Science, 5, 250.
Stipani, V., Gallucci, E., Micelli, S., Picciarelli, V. , Benz, R.
(2001) Biophysical Journal, 81, 3332.
Yin, P., Burns, C. J., Osman, P. D.J., Cornell, B. A. (2003)
Biosensors and Bioelectronics, 18, 389.
Stora, T., Lakey, J.H., Vogel, H. (1999) Angew. Chem. Intern., 38,
389.
Wiese, A., Gutsmann, T., Seydel, U. (2003) Journal of endotoxin
research, 9, 67.
Jelinek and Kolusheva
[108]
[109]
[110]
[111]
[112]
[113]
[114]
[115]
[116]
[117]
[118]
[119]
[120]
[121]
Mozsolits, H., Aguilar, M.-I. (2002) Biopolymers, 66, 3.
Cooper, M.A., Williams, D. H., Cho, Y. R. (1997) Chemical
Communications, 17, 1625.
Papo, N., Shai,Y. (2003) Biochemistry, 42, 458.
YamaJi, A., Sekizawa, Y., Emoto, K., Sakuraba, H., Inoue, K.,
Kobayashi, H., Umeda, M. (1998) Journal of Biological Chemistry,
273, 5300.
Ikura, T., Go, N., Inagaki, F. (1991) Proteins: Struct. Funct.
Genet.,, 81.
Henry, G.D., Sykes, B.D. (1992) Methods Enzymology, 239, 515.
Franklin, J.C., Ellema, J.F., Jayasinghe, S., Kelsh, L.P., Cafiso,
D.S. (1994) Biochemistry, 33, 4036.
Maeda, M., Melnyk, R. A., Partridge, A. W., Liu, L.-P., Deber, C.
M. (2003) Biopolymers, 71, 77.
Sanders, C.R., Hare, B.J., Howard, K.P., Prestegard, J.H. (1994)
Progress NMR Spectros., 26, 421.
Holland, J.W., Cullis, P.R., Madden, T.D. (1996) Biochemistry, 35,
2610.
Regnier, F.E. (1983) Analytical Chemistry, 55, 1298A.
Blondelle, S.E., Forood, B., Perez-Paya, E., Houghten, R. A.
(1996) International Journal of Bio-Chromotography, 2, 133.
Ruhr, E. & Sahl, H.-G. (1985). Antimicrobial Agents and
Chemotherapy, 27, 841.
Pag, U., Oedenkoven, M., Papo, N., Oren, Z., Shai, Y., and Sahl,
H.-G. (2004). J. Antimicrobial Chemotherapy, 53, 203.