JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Vogt, T. C. B.
Right arrow Articles by Bechinger, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Vogt, T. C. B.
Right arrow Articles by Bechinger, B.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

J Biol Chem, Vol. 274, Issue 41, 29115-29121, October 8, 1999


The Interactions of Histidine-containing Amphipathic Helical Peptide Antibiotics with Lipid Bilayers
THE EFFECTS OF CHARGES AND pH*

Titus C. B. Vogt and Burkhard BechingerDagger

From the Max-Planck-Institute for Biochemistry, Am Klopferspitz 18A, 82152 Martinsried, Germany

    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

The alpha -helix of the designed amphipathic peptide antibiotic LAH4 (KKALLALALHHLAHLALHLALALKKA-NH2) strongly interacts with phospholipid membranes. The peptide is oriented parallel to the membrane surface under acidic conditions, but transmembrane at physiological pH (Bechinger, B. (1996) J. Mol. Biol. 263, 768-775). LAH4 exhibits antibiotic activities against Escherichia coli and Bacillus subtilis; the peptide does not, however, lyse human red blood cells at bacteriocidal concentrations. The antibiotic activities of LAH4 are 2 orders of magnitude more pronounced at pH 5 when compared with pH 7.5. Although peptide association at low pH is reduced when compared with pH 7.5, the release of the fluorophore calcein from large unilamellar 1-palmitoyl-2-oleoyl-sn-glycerol-3-phosphocholine or 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoglycerol vesicles is more pronounced at pH values where LAH4 adopts an orientation along the membrane surface. The calcein release experiments thereby parallel the results obtained in antibiotic assays. Despite a much higher degree of association, calcein release activity of LAH4 is significantly decreased for negatively charged membranes. Pronounced differences in the interactions of LAH4 with 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoglycerol or 1-palmitoyl-2-oleoyl-sn-glycerol-3-phosphocholine membranes also become apparent when the mechanisms of dye release are investigated. The results presented in this paper support models in which antibiotic activity is caused by detergent-like membrane destabilization, rather than pore formation by helical peptides in transmembrane alignments.

    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

As pathogenic bacteria and fungi turn resistant against many commonly used antibiotics, considerable efforts are undertaken to develop novel ways to fight infections. Host defensive polypeptides are an integral part of the innate immune system and have been discovered in a wide variety of species, including insects, vertebrates, and humans (1). Some of these peptides are stored in intracellular compartments and their release allows for an immediate response when infections occur. Amphipathic peptides exhibit a strong activity against a wide range of bacteria, fungi, and viruses (2). Examples include the naturally occurring linear polypeptides PGLa (2, 3), magainins (4, 5), cecropins (6), and defensins (7, 8), as well as derivatives thereof (5, 9-11). More recently interest in these peptides has further increased when their tumoricidal activity was demonstrated (12-16).

Linear peptide antibiotics, such as magainins and cecropins, are thought to express their biological activity by related mechanisms (17). Although they show no primary sequence homology, they are all positively charged and form amphipathic alpha -helices in the presence of lipid membranes. Experimental evidence suggests that direct peptide-lipid interactions are important for the expression of antibiotic activity of these substances (18), rather than specific association with a chiral receptor. In particular, the first recognition and membrane association is strongly governed by electrostatic interactions with the negative surface charges of the bacterial cell wall and/or the plasma membrane (19, 20).

There is good agreement that this class of polypeptides exerts its antibiotic activity by permeabilizing the membranes of sensitive cells, which results in decoupling of the transmembrane ionic gradients and consecutively cell death (21, 22). The mechanisms of membrane permeabilization are, however, unknown and several models have been proposed to explain the interaction of polypeptides with lipid bilayers (23). Electrophysiological single channel recordings are difficult to perform because amphipathic peptide antibiotics exhibit a strong propensity to destabilize lipid bilayers. In a few cases, however, stepwise fluctuations in conductivity have been observed, the step-size ranging over 3 orders of magnitude (24, 25). Based on these observations a transmembrane helical bundle, with a water-filled pore lined by basic and polar residues, has been suggested to be the active configuration (26). The formation of a similar peptide aggregate has been suggested to explain the pores formed by the dodecamer alamethicin (27, 28). These latter polypeptides lack the high charge density and the electrophysiological properties are defined much more reproducibly when compared with amphipathic peptides. The observed small cation specificity of the channels formed by positively charged peptides resulted in an extension of this first model, in which the basic transmembrane peptides together with negatively charged lipids form the pore (23).

Structural studies indicate that the large majority of charged amphipathic peptides are oriented along the bilayer surface at physiological peptide-to-lipid ratios (29). In addition, association of several peptides in transmembrane helical bundles with many basic lysine residues accumulating in the pore lumen seems energetically unfavorable (23). The possibility of a detergent-like membrane disruption that is based on the bilayer destabilizing properties of amphipathic peptides has, therefore, also been taken into consideration (23). Our present understanding of the energies involved in peptide-lipid and peptide-peptide interactions is insufficient to safely test for all possible macromolecular aggregation states that could be involved in pore formation by a merely theoretical analysis. Experiments are, therefore, required that allow one to test the validity of the suggested models.

In order to gain further insight into the mechanisms of antibiotic activity and channel formation we designed synthetic model peptides and investigated their functional and biophysical properties (30). One of them, LAH41 with the sequence KKALLALALHHLAHLALHLALALKKA-NH2, forms an amphipathic alpha -helix (Fig. 1). Whereas four terminal lysines serve as membrane anchors and improve the solubility of the peptide, four histidines in the central part of the peptide allow one to alter the charge of the peptide in a pH-dependent manner without altering its chemical composition. The presence of an amidated carboxyl group is a modification also found in nature and ensures that the COOH terminus is uncharged. Alanines and leucines were chosen to form a hydrophobic surface as they show a high propensity to form alpha -helical structures.

Oriented solid-state NMR and ATR-FTIR spectroscopies indicate that the topology of LAH4 depends on the pH of the external medium (30, 44). The protonation states of the histidines are pH-dependent, and, as a result LAH4 switches reversibly from a transmembrane alignment at neutral pH into an orientation along the bilayer surface when the surrounding medium is acidified. The midpoint of the transition between the two states is governed mainly by hydrophobic, electrostatic, and polar interactions. In case of models in which transmembrane helical aggregates form the active configuration one would, therefore, expect that this peptide is by 2 to 3 orders less active in functional assays at acidic pH when compared with neutral proton activity.

We will show that this designed sequence exhibits antibiotic activity against Gram-negative and Gram-positive bacteria in a similar concentration range, as do naturally occurring polypeptides. Under the same experimental conditions, however, this family of antibiotic peptides does not induce release of hemoglobin from red blood cells. The characterization of this peptide is, therefore, not only interesting because of its potential pharmaceutical use, but the analysis of its structural and functional properties as a function of pH also allows one to obtain a more general knowledge of the mechanisms of channel formation and antibiotic activity of amphipathic peptides. The release of the fluorescent dye calcein from POPC and POPG large unilamellar vesicles is used to quantitatively characterize the pore forming activities of LAH4 in well defined model systems. We will show that the LAH4-induced release of fluorescent dye exhibits considerable quantitative and qualitative differences when added to zwitterionic or acidic model membranes.

    MATERIALS AND METHODS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Phospholipids were purchased from Avanti Polar Lipids (Birmingham, AL), calcein and fluorescein labeled dextran (Mr 3,000 and 10,000) from Molecular Probes (Leiden, The Netherlands). The peptides LAH4 (Mr 2,777) and LAH4-Trp12 (Mr 2,849) with the sequence KKALLALALHHLAHLALHLALALKKA-NH2 and KKALLALALHHWAHLALHLALALKKA-NH2, respectively, were synthesized by solid-phase peptide synthesis on a Millipore 9060 automatic peptide synthesizer using Fmoc (9-fluorenylmethyloxycarbonyl) chemistry. The synthetic product was purified using reversed phase high performance liquid chromatography. The identity and purity was confirmed by mass spectrometry. Escherichia coli and Bacillus subtilis with catalogue numbers DSM-3948 and DSM-675, respectively, were from the Deutsche Sammlung für Mikroorganismen und Zellkulturen GmbH (Braunschweig, Germany).

Biological Activity

Growth Inhibition-- A bacterial pre-culture in Luria broth was started from a single colony and grown overnight at 37 °C. 100 µl of this culture was used to inoculate 10 ml and grown to mid-logarithmic phase. The cells were diluted with LB to OD600 = 0.05 (approximately 4 × 107 cells/ml). 30 µl of 2-fold peptide dilutions in LB medium were added to 200-µl aliquots of the bacterial suspension. This mixture was grown at 37 °C until the controls reached an OD600 of 5 (approximately 4 h).

Hemolytic Activity-- 4 ml of fresh human blood, collected in tubes containing 6 mg of EDTA, was washed three times with a buffer containing 100 mM NaCl, 2 mM EDTA, 10 mM Tris, pH 7.3, and resuspended in the same buffer to obtain a 2% suspension. 30 µl of a peptide solution, prepared as 2-fold dilutions, was added to 200 µl of the red blood cell suspension. After incubation at 37 °C for 4 h the solutions were centrifuged. The OD550 in the supernatant provides a reliable indicator of the release of hemoglobin from red blood cells.

Circular Dichroism Spectroscopy

CD spectra were recorded on an auto-dichrograph mark IV (Jibon-Yvon) in the range 190-260 nm using quartz cuvettes with a path length of 0.2 mm. 10 scans were averaged and corrected for the contributions of vesicles and buffer. Small unilamellar vesicles were prepared by sonication to reduce the influence of scattering at low wavelength. The peptide/lipid ratio was 1/150 (mol/mol) at a peptide concentration of 0.23 mg/ml. The molar ellipticity was calculated using d10-camphorsulfonic acid (Theta 290.5 = 7783 deg cm2 dmol-1, OD285 = 34.5 M-1 cm-1) as in Ref. 31.

Membrane Association and Partitioning

For quantitative studies of peptide membrane association the tryptophan analogue of LAH4 was investigated. The tryptophan in this peptide is located at position 12 (LAH4-Trp12), therefore, it resides on the hydrophobic face approximately in the middle of the amphipatic helix (Fig. 1). Tryptophan fluorescence emission spectra were recorded in the range 290-400 nm with an excitation wavelength of 280 nm using a Perkin-Elmer 50b spectrofluorometer. Phospholipid small unilamellar vesicles, prepared by extrusion through 50-nm filters, were injected into a solution of 6 µM LAH4-Trp12 at 37 °C. The measured intensities were corrected for the contribution of light scattering in the presence of vesicles. A membrane insertion model was used to analyze the experimental data, which results in the following sigmoidal function,
I=<FR><NU>L</NU><DE>K<SUB>d</SUB>+L</DE></FR>(I<SUB><UP>max</UP></SUB>−I<SUB>o</SUB>)+I<SUB>o</SUB> (Eq. 1)
where I0 and I are the fluorescent intensities at 330 nm before and after injection of the lipid vesicles, respectively, Imax is the maximum intensity, L the lipid concentration, and Kd the dissociation constant. Least square line fit analysis was used to extract the dissociation constants from experimental measurements of the fluorescence intensity.


View larger version (12K):
[in this window]
[in a new window]
  		Fig. 1.   Helical wheel representation of LAH4. For LAH4-Trp12 Leu-12 was replaced by tryptophan at the hydrophobic face of the helix. The four lysines, two at each end, are not shown.

Preparation of Large Unilamellar Vesicles Loaded with Fluorescent Dye

Dried films of phospholipids (30 µmol) were hydrated with 1 ml of a solution containing 60 mM calcein or 10 mM fluorescein-labeled dextrans, respectively, 10 mM EDTA, 100 mM NaCl, and 50 mM buffer. The buffers used were: ethanolamine, pH 8.5; Tris-HCl, pH 7.5; and acetate, pH 5.5 or pH 4.5. The suspensions were exposed to 10 freeze-thaw cycles and extruded through 200-nm filters (LipoFast, Avestin, Inc., Ottawa, Canada) (32). Fluorescent dye not enclosed within the vesicles was removed by gel filtration (Sephacryl-S300HR, Amersham Pharmacia Biotech). The final concentration of phospholipid was determined by phosphorus analysis (33).

Calcein Release from LUVs

The release of calcein from vesicles was followed with fluorescence spectroscopy using an emission wavelength of 517 nm. The excitation wavelength was chosen in the range 430-490 nm to obtain a quantum yield well within the optimal work range of the photo detector. When entrapped within lipid vesicles, calcein self-quenches to approximately 80% at concentrations >20 mM. Under our experimental conditions at the highest lipid concentration (30 µM) the calcein concentration in the cuvette after addition of Triton X-100 is 5 µM, which is well within the lower linear region. The release of fluorescent dye was initiated by injecting 10 µl of peptide solution into 800 µl of buffer containing the appropriate amount of large unilamellar vesicles (<= 30 µM) loaded with fluorescent dye. The release of fluorescent dye was calculated according to,
R<SUB>f</SUB>=(F<SUB>t</SUB>−F<SUB>0</SUB>)/(F<SUB>T</SUB>−F<SUB>0</SUB>) (Eq. 2)
where Rf is the fraction of dye released, F0, Ft, and FT are the fluorescence intensities at times t = 0, t, or after addition of 10 µl 10% reduced Triton X-100, respectively. The ratio F0/FT ranged from 0.1 to 0.3 and is a function of the pH and the vesicular phospholipid composition. In the absence of peptide or detergent, POPC and POPG large unilamellar vesicles retained the enclosed dye for a time period of several weeks. The experiments were performed at 37 °C, well above the phase transition temperatures of POPC and POPG (both at -2 °C). Control experiments using proton-decoupled 31P solid-state NMR spectroscopy (50) show no evidence for major changes in the macroscopic phase properties of these phospholipid membranes due to the presence of LAH4.

    RESULTS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

In order to test for the antibiotic activity of LAH4 the growth of bacterial cells was determined as a function of peptide concentration (Fig. 2). At pH 5.5 and in the presence of >= 0.1 mg/ml (~36 µM) LAH4 or magainin 2 the growth rate of the Gram-negative bacterium E. coli is considerably reduced when compared with peptide-free media. At pH 7.5, however, the same LAH4 concentration barely influences the growth of E. coli cells. At this pH the histidines of LAH4 are uncharged and growth inhibition is only observed at much higher peptide concentrations (1 mg/ml). The antibiotic activity of magainin 2 is also shown in Fig. 2 and independent of pH. The minimal concentration at which magainin 2 inhibits growth of E. coli is found to be 42 µg/ml (17 µM) in agreement with previously published results (34).


View larger version (12K):
[in this window]
[in a new window]
  		Fig. 2.   The growth of E. coli as a function of peptide concentration (closed symbols) for LAH4 at pH 7.4 (circles) and pH 5.5 (triangles) as well as magainin 2 at pH 7.4 (squares), and the release of hemoglobin from red blood cells (open symbols) at pH 7.4 in the presence of LAH4 (circles) or magainin 2 (squares).

To determine whether LAH4 is bacteriotoxic or bacteriostatic the number of surviving cells at the lowest peptide concentration at which full growth inhibition is observed (28 µg/ml (10 µM) at pH 5.5) was measured by counting the number of surviving cells in a dilution series on agar plates. Out of a starting density of 4 × 107 cells/ml only 103 cells/ml survived. A similar result was obtained in the presence of 50 µM magainin 2. Together with the observation that the optical density at 600 nm decreases immediately after the addition of peptide, these results indicate that LAH4 is bacteriotoxic. The antibiotic activity of LAH4 is even more pronounced when added to cultures of the Gram-positive B. subtilis. Full growth inhibition is observed even at the lowest concentrations tested (0.2 µM at pH 7.5 and pH 5.5). The lowest concentration at which full growth inhibition is observed for magainin under the same experimental conditions is 26 µM. As a test for the cytotoxic activity of LAH4 against eukaryotic organisms we studied the release of hemoglobin from human red blood cells. The open symbols in Fig. 2 show the absence of hemolysis at concentrations where LAH4 or magainin exhibit antibiotic activities.

The secondary structure of LAH4 was investigated under the experimental conditions that are applied during the studies presented in this paper using circular dichroism. Fig. 3A shows CD spectra of 100 µM LAH4 as a function of pH and in the absence of membranes. At acidic pH, when the histidines are charged, the peptide exhibits random coil structures. At basic pH the increased ellipticity at 222 nm indicates that helical conformations are induced. The helical content at pH 4.5, 5.5, 6.5, 7.5, and 8.5 was calculated from the ellipticity at 222 nm (35) to be 13, 26, 30, 39, and 43%, respectively. The peptide LAH4 was designed to form amphipathic or hydrophobic alpha -helical structures in the presence of membranes at high or low proton activity. CD spectroscopy indicates that addition of phospholipid membranes (Fig. 3B) induces helical conformations in LAH4. The apparent helix content of LAH4 at pH 5.5 increases from 26% in the absence of membranes to 55% and 70% in the presence of POPC or POPG, respectively. This observation is also in accordance with previous results obtained by solution NMR spectroscopy in the presence of detergent micelles (30). At pH 5.5 the formation of helical conformations is more pronounced in the presence of the negatively charged phospholipid POPG when compared with the zwitterionic POPC (Fig. 3B). The formation of helical structures due to association with membrane interfaces and/or peptide aggregation has also been observed for magainins (36), melittin (37), and other amphipathic polypeptides (38).


View larger version (19K):
[in this window]
[in a new window]
  		Fig. 3.   Circular dichroism spectra of 100 µM LAH4. A, in the absence of phospholipid at pH 4.5 (continuous line), pH 6.5 (dashed line), and pH 8.5 (dotted line). B, in the absence of phospholipid (continuous line) and in the presence of 15 mM POPC (dashed line) or 15 mM POPG SUVs (dotted line) at pH 5.5.

The association of LAH4 was quantitatively studied with the help of a tryptophan-labeled analogue and fluorescence spectroscopy. Fig. 4 indicates that the addition of phospholipid model membranes to a solution of LAH4-Trp12 results in a blue shift of the emission maximum wavelength (from 350 to 330 nm) and a concomitant increase in intensity at 330 nm. This result suggests that the tryptophan residue is located in the hydrophobic interior of the phospholipid membrane. Association of LAH4-Trp12 reaches equilibrium within 30 s after injection of the vesicle dispersion into the peptide solution. The spectral changes were used to determine the amount of bound peptide as a function of lipid concentration as well as the apparent dissociation constant of LAH4-Trp12 (Table I). The positively charged peptide exhibits a 2 orders of magnitude lower apparent dissociation constant in the presence of negatively charged POPG when compared with the zwitterionic phospholipid POPC. Dissociation constants with liquid crystalline phospholipid membranes in the 10-6 M range are also observed for other amphipathic peptides such as magainin (39), melittin (40), or synthetic model polypeptides (41). The POPC dissociation constant of LAH4 at pH 7.5 has been determined to be 239 µM, and corresponds to an association energy of -20 to -30 kJ/mol. The higher dissociation constant (356 µM) observed at acidic pH can be explained by the increased positive charge density of LAH4 due to histidines which augments the peptide solubility in aqueous environments.


View larger version (11K):
[in this window]
[in a new window]
  		Fig. 4.   Fluorescence emission spectra of 6 µM LAH4-Trp12 at pH 5.5 (continuous line) and 7.5 (dotted line), and in the presence of 30 µM POPG SUVs at pH 5.5 (dashed line).

                              
View this table:
[in this window]
[in a new window]
 
Table I
Dissociation (Kd) and activity constants (KA) as well as the relative activity of LAH4-Trp12 at pH 5.5 and 7.5, respectively
The dissociation constants were determined experimentally in the presence of 6 µM peptide, the activity constants in the presence of 0.14 µM LAH4-Trp12. Kd and KA are used to calculate the relative activities of membrane-associated peptide (normalized to the value obtained in the presence of POPG at pH 7.5). The estimated errors from two to four experiments are within 10%.

Whereas association of LAH4 with zwitterionic membranes is well described by models where the peptide is in exchange between the aqueous and the membrane phase, such a mechanistic model is insufficient when its association to POPG membranes is analyzed. The enhanced association of this basic peptide with this negatively charged membrane can, however, be accounted for by its accumulation within the Helmholtz double layer along the bilayer surface due to electrostatic attraction and subsequent intercalation into the membrane.

It is noteworthy that in the absence of lipid the fluorescence maximum of 6 µM LAH4 solutions shifts to a lower wavelength when the pH values of the LAH4 solutions are increased. At the same time the fluorescence intensity measured at 330 nm is 3-fold higher at pH 7.5 as compared with pH 5.5 (Fig. 4). This shift of the fluorescence maximum is also present when the peptide is diluted 10-fold. In the presence of phospholipids the maximal fluorescence intensity and the fluorescence emission maximum are independent of the pH value. These observations suggest that at basic pH small peptide aggregates form in which the hydrophobic residues are buried. This result is in accordance with the pH-dependent increase in helicity observed during CD measurements (Fig. 3A). The reversible pH and salt-dependent formation of aggregates (probably tetramers) has also been observed for melittin in aqueous environments (37).

To characterize the functional mechanisms of membrane permeabilization in quantitative detail, the release of calcein from model membrane vesicles was investigated. Fig. 5A shows a recording of the fluorescence intensity of large unilamellar vesicles loaded with calcein. At the beginning of the recording the calcein fluorescence is significantly reduced due to self-quenching of the concentrated fluorophore (60 mM). A significant increase in fluorescence is observed, however, when the dye is diluted into the surrounding medium. This effect has been used to monitor the formation of pores by polypeptides and other membrane-active compounds (42, 43). Addition of LAH4 to large unilamellar vesicles of POPC causes the release of calcein and results in a bifunctional increase in fluorescence intensity. The initial fast release of fluorescent dye is followed by a slower process that leads to the release of all calcein within 24 h. In contrast, the release of fluorophore from large unilamellar vesicles composed of POPG shows only a fast one-step mechanism. Thereafter the fraction of released calcein remains unaltered within a period of 24 h.


View larger version (12K):
[in this window]
[in a new window]
  		Fig. 5.   Calcein release experiments: A, time dependent release of calcein followed by the relative fluorescence at 517 nm. The arrows indicate the addition of calcein-loaded phospholipid LUVs, LAH4, and Triton X-100, respectively. The concentration of LAH4 is 0.25 µM in the presence of 30 µM POPC (continuous line), or 2.0 µM in the presence of 30 µM POPG (dotted line). B, release of calcein from POPC (closed symbols) and POPG LUVs (open symbols) after the addition of LAH4 as a function of peptide concentration at pH 8.5 (circles), pH 7.5 (triangles), pH 6.5 (inverted triangles), pH 5.5 (squares), or pH 4.5 (diamonds). The phospholipid concentration is 30 µM.

Despite the striking difference in the topological arrangement of the peptide above or below pH 6.1 (30, 44) the release of calcein is virtually independent of pH (Fig. 5B). Only at pH values as high as 8.5 is a modest decrease of calcein release from POPC vesicles observed. Moreover, our experiments indicate that the release of calcein from 30 µM POPC LUV suspensions requires an order of magnitude lower peptide concentration when compared with the release from POPG large unilamellar vesicles. This is unexpected as the dissociation constants indicate that the amount of peptide associated with POPG is much higher when compared with POPC membranes at the given lipid concentrations (Table I). When the amount of associated LAH4 (instead of the total amount added) is taken into consideration, LAH4 is 5-6 times more effective in pore formation at acidic pH when compared with pH 7.5; and 3 orders of magnitude more effective against PC as compared with PG membranes.

The binding characteristics of LAH4 were investigated in a second experiment, which was designed to allow approximately 50% of the dye to be released after a first addition of peptide. Under these experimental conditions all of the added peptide is associated with lipid membranes (not shown). Once equilibrium has been reached a second batch of calcein-loaded LUVs is added. When POPC LUVs are investigated an equivalent amount of calcein is also released from this second batch indicating that the peptide exchanges between the "old" and the "new" vesicle populations. In contrast, no additional release is observed with POPG LUVs, indicating that the peptide remains unavailable for the freshly added lipid probably due to its tight association with the first population of vesicles. Once bound, the exchange kinetics with the aqueous environment is sufficiently slow that its binding appears to be irreversible on the time scale of our experiments.

This difference in membrane binding and exchange kinetics is also reflected in the dependence of fluorescence dye release which was investigated as a function of peptide concentration and in the presence of different amounts of lipid (Fig. 6). In the presence of increasing peptide concentrations within the vesicular suspension the relative amount of dye release from POPC vesicles is independent of lipid concentration (Fig. 6A). In contrast, a functional dependence on POPG concentration is observed (not shown). On the other hand, release from POPG vesicles follows a uniform functional dependence when the relative increase in calcein fluorescence is analyzed in terms of peptide-to-lipid ratios (Fig. 6B).


View larger version (13K):
[in this window]
[in a new window]
  		Fig. 6.   Release of calcein from POPC (A) and POPG LUVs (B) at pH 5.5 and lipid concentrations of 30 µM (circles), 10 µM (squares), 3 µM (diamonds), or 1 µM (triangles).

To estimate the size of the membrane defects caused by LAH4 the release of fluorescein isothiocyanate-dextrans from large unilamellar vesicles was investigated. Fig. 7 shows the fluorescence increase of fluorescein isothiocyanate-3000 and calcein as a function of peptide concentrations. Whereas a 10-fold higher peptide concentration is required to allow passage from POPC vesicles of the Mr 3000 fluorophore when compared with calcein release (Mr 623), further increase in the size of the attached dextran moiety (Mr 10,000) does not reduce the fluorophore permeability (not shown). In contrast, all fluorophores investigated (Mr 600-40,000) exhibit similar permeability from POPG vesicles in the presence of LAH4. It should be noted that the ratio of bound peptide-to-lipid required for calcein release to be efficient is approximately 10-4 for POPC but 10-2 for POPG.


View larger version (9K):
[in this window]
[in a new window]
  		Fig. 7.   Release of fluorescein isothiocyanate-dextran 3,000 (triangles) or calcein (circles) from POPC (closed) and POPG LUVs (open) at pH 5.5. The lipid concentrations are 30 µM.


    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Similar to its naturally occurring templates magainin or cecropin, the designed peptide LAH4 shows strong antibiotic activity, and at the same time lacks hemolytic action. Little is known about the mechanisms of membrane permeabilization and antibiotic activities by these polypeptides. The lack of primary sequence homology within this family of peptides, the broad range of species sensitive to these antibiotics, and a similar degree of bacteriocidal activity measured for their all-D-analogues suggest that they interact with lipid membranes in a direct manner, rather than through specific interactions with a chiral receptor (45).

Magainins and related peptides have been shown to decouple the ionic gradients across the cell membranes of sensitive organisms (22), cytochrome oxidase liposomes (21), as well as model membranes (34). Similarly, 1-anilinonaphtalene-8-sulfonic acid fluorescence spectroscopy indicates that LAH4 destroys the transmembrane electric potential of living E. coli cells.2 The detailed mechanism of membrane permeabilization remains unknown, but it is commonly believed that the antibiotic activity of these peptides is a direct result of their pore forming activities. Several models have been proposed to explain the increase in ion permeability across phospholipid membranes in the presence of amphipathic polypeptides. These include the formation of transmembrane helical bundles similar to those that have been suggested for alamethicin (23, 27), wormhole structures that are formed by a complex of transmembrane helical peptides and phospholipid (46, 47), and a detergent-like action of amphipathic helical peptides that disrupts the bilayer structure locally or on a larger scale (23). Whereas the first two models assume a transmembrane orientation of the peptide, this is not a requirement if the activity is mediated through a destabilization of the lipid packing of the bilayer.

LAH4 and related sequences have been designed to investigate the functional dependence between bilayer topologies of alpha -helical peptides and electrostatic, hydrophobic as well as polar interactions (30, 48). Solution NMR, CD, and FTIR spectroscopies confirm that LAH4 adopts helical conformations in the presence of membranes (30, 44). Furthermore, oriented solid-state NMR and ATR-FTIR spectroscopies show that the alignment of LAH4 with respect to the bilayer normal is a function of the pH of the aqueous environment (30, 44). At pH values < 6 LAH4 assumes an orientation parallel to the membrane surface, while at pH > 7 a transmembrane alignment is observed. The secondary structure and the in-plane topology of naturally occurring peptide antibiotics including magainins resemble closely to those of LAH4 at low pH.

A comparison of the antibiotic and channel forming activities of LAH4 at pH values above and below the in-plane -to-transmembrane transition, therefore, allows one to differentiate between some of the suggested models. If transmembrane helical bundles are the functional configurations, LAH4 should be 2-3 orders of magnitude more active at neutral pH when compared with acidic conditions where spectroscopic techniques indicate an alignment of the large majority of peptide along the bilayer surface (30). In contrast to models of biological activity which are based on transmembrane configurations, LAH4 exhibits 2 orders of magnitude higher antibiotic activity at low pH. Furthermore, calcein release occurs at five times lower LAH4 to POPC ratios at acidic pH when compared with pH 7.5. Mechanistic models that explain these activities by equilibria of the kind: in-plane intercalated monomer left-right-arrow  transmembrane monomer left-right-arrow  transmembrane multimer left-right-arrow  open pore can, therefore, be excluded. Alternatively we suggest that in-plane intercalated amphipathic peptide helices exhibit detergent-like activities that result in transient increases in membrane permeability as well as membrane disruptions, thereby explaining the antibiotic activities of these peptides. Possible mechanisms include local phase transitions such as the separation of lipid-peptide micelles from the bilayers, or local changes of the membrane curvature due to intercalating peptides that transiently cause the collapse of intact bilayer structures. Models based on detergent-like properties of amphipathic peptides are in good agreement with energetic considerations as well as experimental data which indicate a preferential alignment of amphipathic peptides along the bilayer surface (23, 29, 30, 36). Furthermore, amphipathic peptide antibiotics have been shown to cause membrane thinning (49) as well as changes in the phospholipid phase preferences similar to those observed in the presence of detergents or short chain phospholipids (50).

The experimental results of this work cannot, however, for certain exclude that the stepwise increase in single-channel conductivity which is observed in electrophysiological experiments is a result of a different mechanism based on, for example, the formation of transmembrane helical bundles. On the other hand the detailed characteristics of "single-channel" measurements (14, 24, 25) as well as energetic considerations (23) seem to indicate that the detergent-like properties of amphipathic peptides provide a more probable explanation for the rare observation of the stepwise increase in conductivity. Whereas the electrophysiological properties of amphipathic helical peptides are significantly different from those of "classical hydrophobic channel peptides" such as alamethicin (23, 27), they resemble those observed in the presence of detergents (51-53), pure lipid membranes (54-56), or small unilamellar phospholipid vesicles when added to planar lipid bilayers (57). The conductivity increases measured in the presence of detergents or magainin antibiotic peptides vary over several orders of magnitude, show large variations also within each conductance level, and exhibit a low degree of ion selectivity.

When the amount of membrane-associated peptide is taken into consideration a 10 to 100 times higher surface concentration is required to permeabilize POPG membranes when compared with POPC LUVs. Large differences not only exist in the peptide concentrations necessary for calcein release from large unilamellar vesicles but also in the apparent properties of the release mechanisms (Figs. 5-7). It is possible that negative surface charges arrest the peptide in an inactivated topology, or favor peptide inactivation by aggregation as has been observed for ameloid peptides (58). Furthermore, it is possible that the smaller size of the 1,2-diacyl-sn-glycero-3-phosphoglycerol head group accommodates the peptide more easily within a bilayer packing, therefore, higher concentrations of intercalated LAH4 helices are required to disrupt the POPG membrane. Nevertheless, negatively charged phospholipids enhance the association of basic amphipathic peptides with the membrane surface (19, 20) (this study). In biological membranes of mixed composition the increase in peptide concentration along the membrane surface due to electrostatic attraction enhances the interactions of peptides with all lipids and, therefore, result in an increase in cytotoxic activities.

One might speculate that the difference in composition of eukaryotic and prokaryotic membranes is responsible for the antibiotic selectivity against bacterial when compared with human red blood cells. In particular, the presence of cholesterol has been shown to reduce the activity of peptides in model membranes (59). A detergent-like action of amphipathic peptides makes their antibiotic activity susceptible to the many detailed characteristics of biological and model membranes, including surface charge density, fatty acyl chain, and head group packing, local curvature strain as well as lipid phase preferences and protein-lipid interactions. A multitude of factors that affect the lipid properties and the lipid-peptide interactions, therefore, allow for the modulation of antibiotic activities of detergent-like peptides in a very subtle manner.

    ACKNOWLEDGEMENTS

We acknowledge Prof. Dr. Luis Moroder for making available the fluorescence and circular dichroism spectrometers. Furthermore, the help of Susan Schinzel during the synthesis and purification of peptides is gratefully acknowledged.

    FOOTNOTES

* The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Dagger To whom correspondence should be addressed: Max-Planck-Institute for Biochemistry, Am Klopferspitz 18A, 82152 Martinsried, Germany. Tel.: 49-89-8578-2466; Fax: 49-89-8578-2876; E-mail: bechinge@ biochem.mpg.de.

2 T. C. B. Vogt and B. Bechinger, unpublished results.

    ABBREVIATIONS

The abbreviations used are: LAH4, KKALLALALHHLAHLALHLALALKKA-NH2; ATR-FTIR, attenuated total reflection Fourier transform infrared spectroscopy; LUV, large unilamellar vesicle; POPC, 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine; POPG, 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoglycerol.

    REFERENCES
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

1. Boman, H. G. (1995) Annu. Rev. Immunol. 13, 61-92[CrossRef][Medline] [Order article via Infotrieve]
2. Soravia, E., Martini, G., and Zasloff, M. (1988) FEBS Lett. 228, 337-340[CrossRef][Medline] [Order article via Infotrieve]
3. Hoffmann, W., Richter, K., and Kreil, G. (1983) EMBO J. 2, 711-714[Medline] [Order article via Infotrieve]
4. Zasloff, M. (1987) Proc. Natl. Acad. Sci. U. S. A. 84, 5449-5453[Abstract/Free Full Text]
5. Maloy, W., and Kari, U. (1995) Biopolymers 37, 105-122[CrossRef][Medline] [Order article via Infotrieve]
6. Steiner, H., Hultmark, D., Engstrom, A., Bennich, H., and Boman, H. G. (1981) Nature 292, 246-248[CrossRef][Medline] [Order article via Infotrieve]
7. Lehrer, R. I., Lichtenstein, A. K., and Ganz, T. (1993) Annu. Rev. Immunol. 11, 105-128[CrossRef][Medline] [Order article via Infotrieve]
8. Martin, E., Ganz, T., and Lehrer, R. I. (1995) J. Leukocyte Biol. 58, 128-136[Abstract]
9. Wade, D., Andreu, D., Mitchell, S. A., Silveira, A. M., Boman, A., Boman, H. G., and Merrifield, R. B. (1992) Int. J. Pept. Protein Res. 40, 429-436[Medline] [Order article via Infotrieve]
10. Wieprecht, T., Dathe, M., Schumann, M., Krause, E., Beyermann, M., and Bienert, M. (1996) Biochemistry 35, 10844-10853[CrossRef][Medline] [Order article via Infotrieve]
11. Akiyo, I., Yukiko, H., Ruriko, S., Sanae, M., and Naganori, N. (1998) Biol. Pharm. Bull. 20, 267-270
12. Cruciani, R. A., Barker, J. L., Zasloff, M., Chen, H. C., and Colamonici, O. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 3792-3796[Abstract/Free Full Text]
13. Jacob, L., and Zasloff, M. (1994) Ciba Found. Symp. 186, 197-216[Medline] [Order article via Infotrieve]
14. Haimovich, B., and Tanaka, J. C. (1995) Biochim. Biophys. Acta 1240, 149-158[Medline] [Order article via Infotrieve]
15. Ohsaki, Y., Gazdar, A. F., Chen, H. C., and Johnson, B. E. (1992) Cancer Res. 52, 3534-3538[Abstract/Free Full Text]
16. Soballe, P. W., Maloy, W. L., Myrga, M. L., Jacob, L. S., and Herlyn, M. (1995) Int. J. Cancer 60, 280-284[Medline] [Order article via Infotrieve]
17. Segrest, J. P., De, L., Dohlman, J. G., Brouillette, C. G., and Anantharamaiah, G. M. (1990) Proteins 8, 103-117[CrossRef][Medline] [Order article via Infotrieve]
18. Wade, D., Boman, A., Wahlin, B., Drain, C. M., Andreu, D., Boman, H. G., and Merrifield, R. B. (1990) Proc. Natl. Acad. Sci. U. S. A. 87, 4761-4765[Abstract/Free Full Text]
19. Matsuzaki, K., Harada, M., Funakoshi, S., Fujii, N., and Miyajima, K. (1991) Biochim. Biophys. Acta 1063, 162-170[Medline] [Order article via Infotrieve]
20. Wenk, M. R., and Seelig, J. (1998) Biochemistry 37, 3909-3916[CrossRef][Medline] [Order article via Infotrieve]
21. Juretic, D., Hendler, R. W., Kamp, F., Caughey, W. S., Zasloff, M., and Westerhoff, H. V. (1994) Biochemistry 33, 4562-4570[CrossRef][Medline] [Order article via Infotrieve]
22. Westerhoff, H. V., Juretic, D., Hendler, R. W., and Zasloff, M. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, 6597-6601[Abstract/Free Full Text]
23. Bechinger, B. (1997) J. Membr. Biol. 156, 197-211[CrossRef][Medline] [Order article via Infotrieve]
24. Cruciani, R. A., Barker, J. L., Durell, S. R., Raghunathan, G., Guy, H. R., Zasloff, M., and Stanley, E. F. (1992) Eur. J. Pharmacol. 226, 287-296[CrossRef][Medline] [Order article via Infotrieve]
25. Duclohier, H., Molle, G., and Spach, G. (1989) Biophys. J. 56, 1017-1021[Abstract/Free Full Text]
26. Sansom, M. S. (1993) Eur. Biophys. J. 22, 105-124[Medline] [Order article via Infotrieve]
27. Boheim, G. (1974) J. Membr. Biol. 19, 277-303[CrossRef][Medline] [Order article via Infotrieve]
28. Rizzo, V., Stankowski, S., and Schwarz, G. (1987) Biochemistry 26, 2751-2759[CrossRef][Medline] [Order article via Infotrieve]
29. Bechinger, B., Zasloff, M., and Opella, S. J. (1993) Protein Sci. 2, 2077-2084[Abstract]
30. Bechinger, B. (1996) J. Mol. Biol. 263, 768-775[CrossRef][Medline] [Order article via Infotrieve]
31. Chen, G. C., and Yang, J. T. (1977) Anal. Lett. 10, 1195-1207
32. MacDonald, R. C., MacDonald, R. I., Menco, B., Takeshita, K., Subbarao, N. K., and Hu, L. R. (1991) Biochim. Biophys. Acta 1061, 297-303[Medline] [Order article via Infotrieve]
33. Rouser, G., Fleischer, S., and Yamamoto, A. (1970) Lipids 5, 494-496[Medline] [Order article via Infotrieve]
34. Matsuzaki, K., Sugishita, K., Harada, M., Fujii, N., and Miyajima, K. (1997) Biochim. Biophys. Acta 1327, 119-130[Medline] [Order article via Infotrieve]
35. Wu, C. S., Ikeda, K., and Yang, J. T. (1981) Biochemistry 20, 566-570[CrossRef][Medline] [Order article via Infotrieve]
36. Matsuzaki, K., Murase, O., Tokuda, H., Funakoshi, S., Fujii, N., and Miyajima, K. (1994) Biochemistry 33, 3342-3349[CrossRef][Medline] [Order article via Infotrieve]
37. Bello, J., Bello, H. R., and Granados, E. (1982) Biochemistry 21, 461-465[CrossRef][Medline] [Order article via Infotrieve]
38. Sansom, M. S. P. (1991) Progr. Biophys. Mol. Biol. 55, 139-235[CrossRef][Medline] [Order article via Infotrieve]
39. Matsuzaki, K., Harada, M., Handa, T., Funakoshi, S., Fujii, N., Yajima, H., and Miyajima, K. (1989) Biochim. Biophys. Acta 981, 130-134[Medline] [Order article via Infotrieve]
40. Beschiaschvili, G., and Seelig, J. (1990) Biochemistry 29, 52-58[CrossRef][Medline] [Order article via Infotrieve]
41. Polozov, I. V., Polozova, A. I., Mishra, V. K., Anantharamaiah, G. M., Segrest, J. P., and Epand, R. M. (1998) Biochim. Biophys. Acta 1368, 343-354[Medline] [Order article via Infotrieve]
42. Matsuzaki, K., Murase, O., and Miyajima, K. (1995) Biochemistry 34, 12553-12559[CrossRef][Medline] [Order article via Infotrieve]
43. Degeyter, C., Vogt, B., Benjellountouimi, Z., Sansonetti, P. J., Ruysschaert, J. M., Parsot, C., and Cabiaux, V. (1997) FEBS Lett. 400, 149-154[CrossRef][Medline] [Order article via Infotrieve]
44. Bechinger, B., Ruysschaert, J. M., and Goormaghtigh, E. (1999) Biophys. J. 76, 552-563[Abstract/Free Full Text]
45. Vunnam, S., Juvvadi, P., Rotondi, K. S., and Merrifield, R. B. (1998) J. Pept. Res. 51, 38-44[Medline] [Order article via Infotrieve]
46. He, K., Ludtke, S. J., Huang, H. W., and Worcester, D. L. (1995) Biochemistry 34, 15614-15618[CrossRef][Medline] [Order article via Infotrieve]
47. Matsuzaki, K., Murase, O., Fujii, N., and Miyajima, K. (1995) Biochemistry 34, 6521-6526[CrossRef][Medline] [Order article via Infotrieve]
48. Kinder, R., and Bechinger, B. (1999) Biophys. J. 76, A443
49. Ludtke, S., He, K., and Huang, H. (1995) Biochemistry 34, 16764-16769[CrossRef][Medline] [Order article via Infotrieve]
50. Bechinger, B., Kinder, R., Helmle, M., Vogt, T. C., Harzer, U., and Schinzel, S. (1999) Biopolymers 51, in press
51. Jones, R. V., Rousseau, E., and Meissner, G. (1990) Membr. Biochem. 9, 171-178[Medline] [Order article via Infotrieve]
52. Schlieper, P., and De Robertis, E. (1977) Arch. Biochem. Biophys. 184, 204-208[CrossRef][Medline] [Order article via Infotrieve]
53. Alder, G. M., Arnold, W. M., Bashford, C. L., Drake, A. F., Pasternak, C. A., and Zimmermann, U. (1991) Biochim. Biophys. Acta 1061, 111-120[Medline] [Order article via Infotrieve]
54. Antonov, V. F., Petrov, V. V., Molnar, A. A., Predvoditelev, D. A., and Ivanov, A. S. (1980) Nature 283, 585-586[CrossRef][Medline] [Order article via Infotrieve]
55. Kaufmann, K., and Silman, I. (1983) Biophys. Chem. 18, 89-99[CrossRef][Medline] [Order article via Infotrieve]
56. Yoshikawa, K., Fujimoto, T., Shimooka, T., Terada, H., Kumazawa, N., and Ishii, T. (1988) Biophys. Chem. 29, 293-299[CrossRef][Medline] [Order article via Infotrieve]
57. Woodbury, D. J. (1989) J. Membr. Biol. 109, 145-150[CrossRef][Medline] [Order article via Infotrieve]
58. Terzi, E., Hoelzemann, G., and Seelig, J. (1995) J. Mol. Biol. 252, 633-642[CrossRef][Medline] [Order article via Infotrieve]
59. Christensen, B., Fink, J., Merrifield, R. B., and Mauzerall, D. (1988) Proc. Natl. Acad. Sci. U. S. A. 85, 5072-5076[Abstract/Free Full Text]

Copyright © 1999 by The American Society for Biochemistry and Molecular Biology, Inc.
Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 Biophys. JHome page
A. J. Mason, A. Marquette, and B. Bechinger
Zwitterionic Phospholipids and Sterols Modulate Antimicrobial Peptide-Induced Membrane Destabilization
Biophys. J., December 15, 2007; 93(12): 4289 - 4299.
[Abstract] [Full Text] [PDF]

Home page
 Antimicrob. Agents Chemother.Home page
A. J. Mason, C. Gasnier, A. Kichler, G. Prevost, D. Aunis, M.-H. Metz-Boutigue, and B. Bechinger
Enhanced Membrane Disruption and Antibiotic Action against Pathogenic Bacteria by Designed Histidine-Rich Peptides at Acidic pH.
Antimicrob. Agents Chemother., October 1, 2006; 50(10): 3305 - 3311.
[Abstract] [Full Text] [PDF]