Advertisement
JBC

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


     

Originally published In Press as doi:10.1074/jbc.M507042200 on July 25, 2005

J. Biol. Chem., Vol. 280, Issue 40, 33960-33967, October 7, 2005
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
280/40/33960    most recent
M507042200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
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 arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Glukhov, E.
Right arrow Articles by Deber, C. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Glukhov, E.
Right arrow Articles by Deber, C. M.
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?

Basis for Selectivity of Cationic Antimicrobial Peptides for Bacterial Versus Mammalian Membranes*

Evgenia Glukhov{ddagger}1, Margareta Stark{ddagger}2, Lori L. Burrows§||, and Charles M. Deber{ddagger}3

From the Divisions of {ddagger}Structural Biology and Biochemistry, and §Infection, Immunity, Injury and Repair, Research Institute, Hospital for Sick Children, Toronto, Ontario M5S 1A8, and the Departments of Biochemistry and ||Surgery, University of Toronto, Toronto, Ontario M5G 1X8 Canada

Received for publication, June 28, 2005 , and in revised form, July 21, 2005.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Novel cationic antimicrobial peptides typified by structures such as KKKKKKAAXAAWAAXAA-NH2, where X = Phe/Trp, and several of their analogues display high activity against a variety of bacteria but exhibit no hemolytic activity even at high dose levels in mammalian erythrocytes. To elucidate their mechanism of action and source of selectivity for bacterial membranes, phospholipid mixtures mimicking the compositions of natural bacterial membranes (containing anionic lipids) and mammalian membranes (containing zwitterionic lipids + cholesterol) were challenged with the peptides. We found that peptides readily inserted into bacterial lipid mixtures, although no insertion was detected in model "mammalian" membranes. The depth of peptide insertion into model bacterial membranes was estimated by Trp fluorescence quenching using doxyl groups variably positioned along the phospholipid acyl chains. Peptide antimicrobial activity generally increased with increasing depth of peptide insertion. The overall results, in conjunction with molecular modeling, support an initial electrostatic interaction step in which bacterial membranes attract and bind peptide dimers onto the bacterial surface, followed by the "sinking" of the hydrophobic core segment to a peptide sequence-dependent depth of ~2.5–8 Å into the membrane, largely parallel to the membrane surface. Antimicrobial activity was likely enhanced by the fact that the peptide sequences contain AXXXA sequence motifs, which promote their dimerization, and possibly higher oligomerization, as assessed by SDS-polyacrylamide gel analysis and fluorescence resonance energy transfer experiments. The high selectivity of these peptides for nonmammalian membranes, combined with their activity toward a wide spectrum of Gram-negative and Gram-positive bacteria and yeast, while retaining water solubility, represent significant advantages of this class of peptides.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Natural antimicrobial peptides are part of the innate immunity of a wide range of species ranging from insects and amphibians to mammals, including humans, defending against infections from bacteria, fungi, parasites, and enveloped viruses, with some peptides also effective against tumor cells (1, 2). Currently, data bases report over 800 sequences for natural antimicrobial peptides and proteins from animals and plants (www.bbcm.univ.trieste.it), whereas several thousand others have been designed de novo and produced synthetically (3). Several classes of peptides have emerged including the following: (i) linear peptides free of cysteines and often with an amphipathic sequence (e.g. magainins); (ii) peptides with disulfide bonds that can produce a flat dimeric {beta}-sheet structure (e.g. HBD-2); and (iii) peptides with an unusual bias toward certain amino acids, such as proline, arginine, tryptophan, or histidine (e.g. indolicidin) (4–7). Many antimicrobial peptides are highly positively charged and exist predominantly as monomers with random coil structure in solution (8). Although they differ widely in sequence and structure, cationic antimicrobial peptides (CAPs)4 generally consist of 12–50 residues, ~50% of which are hydrophobic (9), and accordingly have the potential to form an amphipathic {alpha}-helical structure when bound to membranes.

The increasing prevalence of antibiotic resistance necessitates the development of new ways to combat bacterial infection. Although some antimicrobial peptides are already in clinical and commercial use, the future design of novel antimicrobial peptides will necessitate the optimization of multiple parameters, notably reduction of toxicity against eukaryotic cells and of susceptibility to proteolytic degradation. A key problem in clinical development of CAPs is the degree of selectivity between microbial and host cells. Peptides likely make this differentiation based on variations in the composition of each particular cell membrane (1). Thus, for eukaryotic cells, the primary membrane features are the presence of cholesterol (up to 25%), predominance of phosphatidylcholine lipids, and an essentially neutral (zwitterionic) outer leaflet (10). In contrast, bacterial cells lack cholesterol, have phosphatidylethanolamine as their most common zwitterionic lipid, but also contain 20–25% of negatively charged lipids in their outer membranes, including phosphatidylglycerol and cardiolipin. Cationic antimicrobial peptides are active in the low medium micromolar range and show little target or L-versus D-residue specificity (their D-enantiomers exhibit similar activity to their L-counterparts), indicating that they interact with achiral components of the cell membrane (11, 12) through a mechanism of physical disruption. Accordingly, bacteria may not easily develop resistance.

Many cationic peptides studied to date have some toxicity, as measured by their tendency to induce lysis of erythrocytes at higher concentrations (e.g. gramicidins (13), pardaxin (14), mellitins (100% at 10 µM) (15, 16), mastorpans (90% at 25 µg/ml) (17), tachyplesins II (18), protegrin I (19), indolicidin (7), and cathelicidins (20)). Examples of relatively nontoxic peptides include mammalian defensins (21), dermaseptins (22), spinegirin (no hemolysis at 100 µM) (23), and magainin (hemolysis only at the relatively high concentration of 100 µg/ml) (24).

As derived from an earlier series of 25-residue CAPs of prototypic sequence KKAAAXAAAAAXAAWAAXAAAKKKK-NH2, where X = each of the 20 commonly occurring amino acids (25), a class of novel synthetic 17-residue CAPs has been developed in our laboratory (26). The latter series of peptides contain "guest" X residues embedded in an Ala-rich sequence, generally containing a Trp residue as fluorescent probe (TABLE ONE); several of these peptides have been studied previously in MIC and hemolysis assays (26). Key features of the peptides are the consecutive nonamphipathic hydrophobic core of 11 residues, and the multipositively charged Lys/Arg in either segregated (all basic residues at the N or C terminus) or separated forms (basic residues at both termini). The Lys or Arg tags solubilize the hydrophobic peptides in aqueous media (25, 27). Several of these peptides have been found to be highly effective against a series of Escherichia coli strains (MICs in the range of 4–128 µM or 8–256 µg/ml) (26) and a wide variety of organisms, including Pseudomonas aeruginosa, an opportunistic pathogen in cystic fibrosis lung infections. In some instances, D-enantiomers of these sequences show somewhat higher activity versus their L-counterparts. Most of these same peptides display no hemolytic activity against rabbit or human red blood cells up to relatively high concentrations (325 µM or 650 µg/ml) (26). However, the detailed mechanism of their selective antimicrobial action has not been elucidated. Through studying this series of CAPs with a variety of biophysical techniques, including fluorescence and spin-labeling studies in phospholipid vesicles, we report here an assessment of the chemical/structural factors that are predominantly responsible for their high selectivity for nonmammalian membranes.


View this table:
[in this window]
[in a new window]
  		TABLE ONE
Sequences and designations of synthetic cationic antimicrobial peptides employed in the present work Lowercase letters in sequences indicate D-enantiomers of the amino acids. Polar residues (Lys, Arg, and Glu) are shown in boldface; aromatic residues are underlined. MICs in selected Gram-negative and Gram-positive bacteria and hemolytic activities in rabbit and human erythrocytes have been reported for peptides F17, F17-6K, All-D F17-6K, F17-6R, and W17-6K (26).

 

   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Materials—Reagents for peptide synthesis, cleavage, and purification include the following: 9-fluorenylmethoxy carbonyl (Fmoc)-protected amino acids (Novabiochem); dansyl and dabcyl chlorides (Molecular Probes, Inc. Eugene, OR); Fmoc-peptide amide linker-resin, piperidine (Applied Biosystems, Foster City, CA); N,N-dimethylformamide (DMF), methanol, diethyl ether, and acetonitrile (Caledon Laboratories Ltd., Georgetown, Ontario, Canada); N,N-diisopropylethylamine (DIEA; Aldrich); O-(7-azabenzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (GL Biochem Ltd., Shanghai, China); 1,2-dipalmitoyl-sn-3-[phospho-rac-(1-glycerol)], 1,2-dipalmitoyl-sn-3-[phospho-rac-(1-choline)], the nitric spin probes 1-palmitoyl-2-stearoyl (n-doxyl)-sn-glycero-3-phosphocholine (n-doxyl 1-palmitoyl-2-stearoyl (n-doxyl)-sn-glycero-3-phosphocholine (where n indicates the position of doxyl group in the stearoyl chain; in the present series, n = 5, 10, 12, and 16), 1,2-dimyristoyl-sn-glycero-3-(phospho-L-serine), 1-palmitoyl-2-oleyl-glycero-3-phosphoethanolamine, cholesterol, and cardiolipin (Avanti%20Polar%20Lipids">Avanti Polar Lipids, Alabaster, AL). All chemicals were used without further purification. Buffer was prepared using Tris sodium salt in doubly distilled water with adjustment to pH 7.0 by HCl.

Peptide Synthesis—Peptides were synthesized by standard solid phase protocols using Fmoc chemistry on a PerSeptive Biosystems Pioneer peptide synthesizer as described (25, 26). A low load (>0.15 mmol/g) of peptide amide linker resin was used to incorporate an amide function at the peptide C terminus. O-(7-Azabenzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate coupling reagent (0.45 M in DMF) and DIEA base (1.0 M in DMF) were used with a 4-fold excess of protected amino acids. Fluorescent probes (either dabcyl (4-dimethylaminophenylazobenzoyl) chloride (10 mg) or dansyl (N-(5-dimethylaminophthalene-1-sulfonyl) chloride (10 mg)) were coupled manually overnight after completion of solid phase synthesis using 100 mg of peptidyl-resin each in 1 ml of DMF with 50 µl of DIEA. Following a standard cleavage procedure (89% trifluoroacetic acid, 4.5% phenol, 4.5% triple distilled water, 2% triisopropylsilane), crude peptides were precipitated in cold ethyl ether. Deprotected peptides were purified on a reverse phase C18 high performance liquid chromatography column (Vydac, 250 x 21.5 mm) using a linear gradient of 20–30% B (Buffer A: 0.1% trifluoroacetic acid, 95% triple distilled water, 5% AcCN; Buffer B: 0.1% trifluoroacetic acid, 95% AcCN, 5% triple distilled water) for the first 10 min and then 30–50% B for another 30 min. Absorbance was monitored at 215 nm. Molecular masses were confirmed by matrix-assisted laser desorption ionization-mass spectrometry. Concentrations of peptides were determined in triplicate by standard BCA assay. Peptide solutions were stored at -20 °C.

Circular Dichroism—Circular dichroism spectra were collected at different temperatures using a Jasco J-720 spectropolarimeter. Spectral scans were performed from 250 to 195 nm, with step resolution of 0.1 nm and bandwidth of 1.0 nm at speed 50 nm/min (28). A 1-mm path length quartz cuvette was used for the measurements, and values from 7 scans were averaged per sample. Freshly prepared samples were measured at 20– 60 µM range. The aqueous buffer containing 10 mM buffer of Tris-HCl, 10 mM NaCl, pH 7.0, was used as a solvent for measurements, and its background (in the presence of appropriate detergents where needed) was subtracted for each sample. 25 mM SDS solutions were generally prepared by dissolving the desired amount of reagents into this buffer. The molar ellipticity (29) was calculated following Chen et al. (30) as shown in Equation 1,

(Eq. 1)
where n = number of residues.

SDS-PAGE—Peptide samples were subjected to SDS-PAGE, using NuPAGE pre-cast 12% BisTris gels (1.0 mm x 10 well) and buffers as follows: NuPAGE SDS sample buffer (NOVEX, San Diego, CA) and NuPAGE MES/SDS running buffer. Peptides were dissolved in different concentrations in sample buffer and heated at 85 °C for 10 min prior to electrophoretic separation at 125 mV. The values of Mr(exp):Mr(theor) ratios were calculated from each stained gel (either Coomassie Blue or Silver staining for peptides (10–140 µM) using See Blue and Mark12 markers and the NIH 1.62 Image Program, software available at www.cellbio.med.unc.edu/henson_mrm/pages/NIH.html) (31).

Preparation of Phospholipid Vesicles—The desired mixtures of phospholipids, cholesterol, and cardiolipin were dried in glass tubes first under nitrogen and then lyophilized overnight to obtain lipid films. The dry lipid films were then suspended for 1 h in a water bath at 40–50 °C in Tris-HCl buffer, pH 7.0 (10 mM Tris, 10 mM NaCl), and sealed with parafilm under nitrogen. The mixtures were vortexed occasionally to disperse the lipids. Vesicles were kept at 44 °C to maintain the lipids above the gel-to-liquid crystalline phase transition and used at the same day. The appropriate aliquots of the desired peptides were added.

Small unilamellar vesicles (SUVs) were prepared using standard procedure as described previously (28) but with no peptides present. Sonication of lipid dispersions was performed in 5-ml glass tubes in a bath-type sonicator (G112SP1G, Laboratory Supplies Co., Hicksville, NY) for 5 min in cold water (until clear). To avoid degradation of unsaturated lipids, sonication was performed at ~10 °C under a nitrogen atmosphere.

Large unilamellar vesicles (LUVs) were prepared as described (29) by freeze-thawing the desired lipid suspension five times under nitrogen to produce large multilamellar vesicles. The suspension was extruded 11 times through polycarbonate membranes with 0.1-mm diameter pores (Nuclepore Corp., Pleasanton, CA) on an Avanti mini-extruder apparatus.

Fluorescence Measurements—Fluorescence emission spectra of peptide Trp residues were recorded on a Hitachi F-400 Photon Technology International C-60 fluorescence spectrometer equipped with water bath circulator for regulating the temperature. Semimicro quartz cuvettes of 0.5 ml (10-mm excitation path length and 4-mm emission path length) and 2 ml (10 x 10 mm, in titrations) (Hellma, Concord, Ontario, Canada) were used. Excitation wavelength was 280 nm, and emission spectra were recorded from 305 to 365 nm. Spectra were collected with a step size of 1 nm with an average of three cycles. Excitation and emission slit widths were 2 and 6 nm (2 and 6 nm band pass, 1 and 3 turns of the slit micrometers), respectively. All measurements were corrected for light scattering effects of vesicles by subtraction of background and by the correction function of the manufacturer's software. Background samples (blank suspensions of buffer or lipids in buffer) were prepared using the same protocol, except pure water instead of peptide solution was added. Blue shifts were calculated as the difference in wavelength of the maxima in emission spectra of lipid-peptide and aqueous peptide samples. Peptides were used in 1 or 4 mM of detergent, such that lipid-to-peptide ratios of either 250 or 1000 were established at a constant peptide concentration of 4 µM in all experiments except titrations. In the latter case, data were corrected for dilution of the peptides during the course of the experiment.

For depth quenching measurements, 10% (mol %) of n-doxyl 1-palmitoyl-2-stearoyl (n-doxyl)-sn-glycero-3-phosphocholine into different types of vesicles was incorporated (32). The appropriate aliquots of the desired peptides were added to 1 mM of lipid vesicles in Tris-HCl buffer, pH 7.0 (10 mM Tris, 10 mM NaCl), 10 min prior to measurements. Each experiment was repeated at least twice, usually four times. The fluorescent intensities of the emission maxima were taken in the presence and in the absence of n-doxyl quenchers in vesicles. Parallax analysis was performed according to published procedures (33). The distance of the Trp residue from the bilayer center (ZCF) was calculated as shown in Equation 2,

(Eq. 2)
where LC5 represents the distance from the bilayer center to the shallow placed n-doxyl quencher (n = 5), which is 10.8 Å for LUV-RBC(outer) and LUV-bact vesicles, and 9.9 Å for SUV-I and SUV-II vesicles (TABLE TWO); C is the mole fraction of quencher (0.1 in all cases) divided by the lipid area (70 Å) (34); F5 and F16 are the relative fluorescence intensities of the shallower and deeper quenchers, respectively; and L5:16 is the difference in the depth of this two quenchers. For depth measurements, we used 0.9 Å per methylene group CH2. The indole ring insertion depth (d) was calculated as one-half the bilayer thickness (= 28.8 Å for SUV-I and SUV-II vesicles, or 30.6 Å for LUV-RBC(outer) and LUV-bact vesicles) minus ZCF.


View this table:
[in this window]
[in a new window]
  		TABLE TWO
Molar percentages of lipid composition and designations for vesicle preparations used in the present work Compositions for E. coli (minor components omitted) and erythrocyte membranes are as given by Gennis (54). SUV, small unilamellar vesicles; LUV, large unilamellar vesicles; bact, bacterial membrane composition; RBC, red blood cell membrane composition; RBC(outer), outer leaflet of RBC membrane composition; Zwit, zwitterionic.

 
FRET Measurements—Labeling of peptides with dansyl (donor) and dabcyl (acceptor) was achieved by coupling of dansyl or dabcyl chlorides to the N-terminal Lys (35). Steady-state fluorescent spectra were recorded on the Photon spectrometer. Samples were examined at constant stirring in a stoppered 10 x10 mm disposable cuvette by removing and adding new portions of mixtures. Peptide stock solutions were diluted in buffer (10 mM Tris-HCl, 10 mM NaCl, pH 7.0) with either 25 mM SDS detergent at room temperature or 1 mM of SUV-bact or SUV-RBC(outer) (TABLE TWO) at 44 °C. The concentration of dansyl-labeled peptides (donor) was kept constant at 1 µM, and the total concentration of dansyl-, dabcyl-, and unlabeled peptides was kept constant at 5 µM in SDS mixtures or at 0.5 and 2.5 µM in experiments with vesicles. Each mixture was allowed to equilibrate for 3 min. Emission spectra were collected from 450 to 650 nm at room temperature in SDS micelles and at 44 °C in SUV-bact. The {lambda}ex = 341 nm, 0.5 s/nm, and band pass was 2 nm for excitation and 4 nm for emission. Each spectrum is the average of two or three runs.

Computational Modeling—Energy-minimized models of the interaction between two idealized helices were produced using a global conformation search program CHI as described (36). The program CHI identifies structures with energetically favorable packed interfaces between paired helices, analyzes all possible interactions, and returns sets of probable structures.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
The disruption mechanism of bacterial membranes by the present family of synthetic CAPs was examined in cell membrane mimic conditions by circular dichroism, fluorescence, and electrophoretic methods. The majority of these peptides (TABLE ONE) have average core values of +1.4 to 1.5 on the Liu-Deber scale (25), which is above the threshold value of 0.4 for spontaneous cell membrane insertion. We also studied one significantly more hydrophobic peptide (All D W17-6K-(4L)), which possesses an average core hydropathy value of ~3.1.



View larger version (6K):
[in this window]
[in a new window]
  		FIGURE 1.
Secondary structures of the F17 antimicrobial peptide (a) and the natural CAP magainin II (b), as determined by circular dichroism spectroscopy. Spectra were recorded at room temperature in buffer containing 10 mM Tris-HCl, 10 mM NaCl, pH 7.0, in the presence or absence of 25 mM SDS detergent as indicated on the diagram. Peptide sequences are as given in TABLE ONE. Peptide concentration was 40µM. The curves reported are based on the average of three runs with background subtracted. aq, aqueous.

 
Peptide Secondary Structure—TM Finder data analysis (37) predicts that in membrane-mimetic environments, segments of these CAPs will adopt an {alpha}-helical secondary structure. In case of Lys-segregated peptides (such as F17-6K), this segment is predicted to consist of residues 4–10 of the hydrophobic core, whereas for Lys-separated peptides (such as F17), residues 3–9 from the corresponding core are likely to be involved. Across this series, no more than seven residues (about 40% of total peptide residues) are predicted to be involved in formation of {alpha}-helix during membrane insertion. Circular dichroism data are consistent with these predictions. As typified by F17, the present CAP peptides are largely in random coil conformations in aqueous buffer (Fig. 1a). In SDS micelles, transition of the CD spectrum to a helical type pattern is observed; however, as derived from the experimental molar ellipticity values, only 5–8 residues are involved in helix formation in agreement with TM Finder analysis. No influence of concentration (20–60 µM), temperature (25–75 °C), or pH (in the range of 7–11) was found on the spectra in the case of F17 (not shown). Qualitatively similar effects were observed for the natural CAP magainin II (TABLE ONE and Fig. 1b), with the increased helical content in SDS micelles versus F17 arising from its greater length and longer helical segment(s). Similar to certain other CAPs (16, 38), no correlation has generally been observed between degree of helicity and antimicrobial activity; high helicity often correlates with high hemolytic properties, albeit not with antimicrobial activity (16).



View larger version (14K):
[in this window]
[in a new window]
  		FIGURE 2.
Blue shifts in wavelength maxima emission ({Delta}{lambda}max) of Trp fluorescence upon exposure of selected peptides (4 µM) to freshly prepared vesicles (1 mM of lipids). Blue bars, LUV-bact (25% AL,75% zwitterionic lipids, similar to the composition of bacterial cell membranes). Red bars, LUV-RBC (outer) (0% AL, 75% zwitterionic lipids, 25% cholesterol, similar to the outer leaflet of erythrocyte cell membranes). Purple bars, LUV-Zwit (0% AL, 100% zwitterionic lipids, no cholesterol). See TABLE TWO for details of lipid components. Temperature = 44 °C. Experiments were performed in duplicate. Error bars are based on standard deviations derived from the measured wavelength positions.

 
Peptide Location within the Bilayer—The first indication of bacterial membrane insertion, and evidence of peptide selectivity for bacterial versus mammalian cell membranes, was obtained by exposure of the peptides to anionic and zwitterionic unilamellar lipid vesicles. In these experiments, lipid/peptide (L/P) ratios were maintained high enough (250 and above) at low peptide concentration (4 µM, below the lowest MIC values) in order to mimic the initial steps of peptide-mediated membrane disruption. Measurements of fluorescence emission intensity and shifts in the {lambda}max of the Trp probe incorporated into the hydrophobic core of peptides in the library (TABLE ONE) were particularly useful for analysis of the position of a given peptide within the membrane. Exposure of suitable peptides at 4 µM to freshly prepared anionic LUV-bact vesicles containing a lipid mixture corresponding to a typical bacterial membrane (TABLE TWO; Fig. 2) caused a blue shift in Trp {lambda}max values, accompanied by intensity enhancement. These results provide evidence of insertion of peptides into the bilayer (39) and effective removal of the peptides from the bulk aqueous environment. The significant blue shift of {Delta}{lambda}max = 17–23 nm indicates a membrane-buried Trp in negatively charged LUVs containing 25% anionic lipids (AL), corresponding to "bacterial vesicles" (LUV-bact). Bilayer insertion of peptides was similarly apparent for 10% AL (LUV-RBC) and 90% AL (SUV-I) as well (data not shown). In contrast, virtually no shifts in either maximum wavelength or fluorescent intensity were seen even for the most hydrophobic peptide (All-D W17-6K-(4L)) with LUVs formed from a mixture of pure zwitterionic lipids and cholesterol, mimicking the outer leaflet of red blood cell membranes (LUV-RBC(outer) (0% AL, 25% cholesterol)) (TABLE TWO and Fig. 2). In the absence of cholesterol, the zwitterionic vesicles (LUV-Zwit, see TABLE TWO) similarly show evidence of lack of peptide insertion; however, partial insertion was observed for All-D W17-6K-(4L) (Fig. 2).



View larger version (6K):
[in this window]
[in a new window]
  		FIGURE 3.
Trp depth values (d, Å) of selected peptides (4 µM) in LUV-bact (1 mM) at 44 °C, sorted by decreasing MIC values (= increasing antimicrobial activity) in various bacterial cells. A, E. coli C498; B, S. epidermis C621; C, P. aeruginosa ATCC 27853. Values of d given in the figure are the averages of 2–4 repetitions of each experiment. Error bars are based on standard deviations derived from the measured fluorescence intensities.

 
Depth of Membrane Penetration by Synthetic CAPs—To estimate the penetration depth of Trp and, correspondingly, of the peptides into model cell membranes, a dual quencher assay (parallax analysis) was employed (see "Experimental Procedures"). In our experiments, 10% of spin-labeled PC with a nitroxide (doxyl) group, covalently attached to the methylene carbon at either the 5 or 16 position of the acyl chains, was incorporated into phospholipid vesicles ({approx}30 nm (SUVs) and {approx}100 nm (LUVs) diameter). The resulting quenching by the doxyl group during Trp fluorescence measurements provided an accurate probe for estimating the penetration depth of this residue into the lipid bilayer.

Negatively charged LUV-bact vesicles (25% AL, no cholesterol, TABLE TWO) were employed in Trp parallax method experiments to estimate depth of insertion (d) as a function of peptide concentration from 1.5 to 6 µM for one of the more effective CAPs (F17-6K) and for one of the less active peptides (F17); above 6 µM, the mixture with F17-6K loses transparency. Although the depth of Trp penetration increased with increasing peptide concentration in the case of F17-6K to a plateau around 6.5 Å near 3 µM, no depth variation was found for F17 over the same concentration range (depth = 1.5–2 Å). Both peptides were shown to be closer to the bilayer surface then to its center, but the more active peptide was found to penetrate about 3 Å deeper into the bilayer at this range of concentrations.



View larger version (4K):
[in this window]
[in a new window]
  		FIGURE 4.
Relative fluorescence intensity of peptide All-D F17-6K-(2L) in negatively charged SUVs (SUV-I and SUV-II) in the absence (labeled "no dox") and in the presence of 10% doxyl-labeled groups at 5-, 10-, 12-, and 16-positions. Lipid preparations are as follows: SUV-I, no cholesterol present; and SUV-II, 25% cholesterol. Experiments were performed in duplicate. Error bars are based on standard deviations derived from the measured fluorescence intensities.

 
Membrane Penetration Depth Versus Antimicrobial Activity—Values of Trp indole depth after spontaneous insertion into freshly prepared LUV-bact vesicles were calculated in a similar manner for six selected peptides from TABLE ONE at 4 µM (Fig. 3). The amount of quenching varied for each peptide as a function of the doxyl moiety position from 35 to 80%. The depth of Trp penetration into "bacterial" vesicles was found to range between 2.4 and 7.9 Å. Peptides sorted in decreasing order of their Trp depth versus their MIC values, as shown for three bacterial strains Fig. 3, A–C. Overall, we found that depth of penetration and MIC values were highly correlated: as a given peptide penetrates deeper into the bilayer, it has lower MIC values (i.e. higher activity) against various bacterial strains in vivo. Similar experiments performed for three peptides (F17, F17-6K, and F17-6R) using LUV-RBC(outer) vesicles gave no quenching and hence no indication of peptide insertion (not shown).

Effect of Cholesterol on Depth of Peptide Insertion—We investigated a proposed specific role of cholesterol, which is present in mammalian cells but absent in bacterial membranes, as a "protectant" against penetration by the peptides. By using anionic vesicles as a model, we found that the depth of the Trp residue from All-D F17-6K-(2L) was essentially unaffected by the incorporation of 25 mol % cholesterol (SUV-I versus SUV-II) into the bilayer, as determined by parallax analysis with 10% of either 5-, 10-, 12-, or 16-doxyl-labeled 1,2-dipalmitoyl-sn-3-[phospho-rac-(1-choline)] (Fig. 4). These results are discussed further below.



View larger version (10K):
[in this window]
[in a new window]
  		FIGURE 5.
SDS-PAGE (silver staining) analysis of CAPs. The lanes correspond to the following: A, F17; B, F17-6K; C, All-D F17-6K(2L); D, F15-4K; and E, All-D F21-10K. Peptides were diluted in SDS-containing sample buffer at ~10µM (200 ng) and heated for 10 min at 85 °C prior to electrophoresis using 12% BisTris NuPAGE gel at 125 constant voltage. Mr(exp)/Mr(theor) (MWexp/MWtheor) is the ratio of experimentally determined to actual molecular weight of each CAP, where a value = 2 corresponds to dimer. Molecular mass standard (Mark 12) is given at the left of the diagram. Molecular weights of CAPs are given in TABLE ONE.

 
Peptide Oligomerization State(s)—Although the high content of Ala residues had originally been introduced into the peptide design process to serve as "background" or "template" residues of mid-range hydrophobicity (25), an emerging body of work on helix-helix interaction motifs in membrane-based peptides has suggested that two "small" residues separated by three residues (termed GXXXG or AXXXA motifs) are important mediators of helix-helix dimerization in membranes (40, 41). Consistent with this situation, all peptides from TABLE ONE studied at concentrations between 10 and 100 µM in SDS-PAGE assays on NuPAGE gels at pH 7.3 gave indication of formation of discrete SDS-resistant dimers (Fig. 5). We observed Mr(exp)/Mr(theor) values around 1.8–2.7 (indicative of dimers) for peptides with one (at the hydrophobic core center), two (central and N-terminal), and three (both ends and center of the core) sequential AXXXA motifs. However, under the same conditions, Mr(exp)/Mr(theor) values of 1.1–1.3, typical for monomers, were observed for the natural antimicrobial peptides magainin II and cecropin P2 (TABLE ONE) up to loading concentrations of 140 µM (not shown). No correlation was found between helicity levels in CD spectra and values of the Mr(exp)/Mr(theor) ratio.

FRET Measurements of Peptide Self-association—Given the observation of SDS-resistant peptide dimers, we used FRET to confirm further the dimerization of the peptides upon membrane insertion. Specific labeling of peptides was achieved during synthesis by attaching either dansyl or dabcyl chloride to the N-terminal Lys residue. Labeled peptides were cleaved from the resin and purified using the same protocols as for unlabeled peptides. CD spectra and SDS-PAGE analysis of dansyl-labeled and dabcyl-labeled All-D F17-6K-(2L) revealed that the helical content and relative mobility were the same for both labeled and unlabeled forms of the peptide (not shown). Quenching of donor (N-terminal dansyl-labeled peptide) fluorescence as a function of the acceptor (N-terminal dabcyl-labeled peptide) fraction can be used to determine the stoichiometry of association in a peptide oligomer (35, 42); in the case of monomers, no quenching should be observed. If oligomerization occurs, the relative quantum yield of the donor decreases linearly in the case of dimerization or according to a more complex function upon higher order oligomer formation (43, 44). FRET experiments performed at pH 7.0 in anionic bacterial membrane vesicles (SUV-bact with 25% AL (2.5 µM of total peptides)) (Fig. 6) confirmed the high tendency of the peptides toward dimerization. SUVs were used here for reduction of light scattering effects. As seen in Fig. 6, where quenching versus the mole fraction of dabcyl acceptor is plotted for the All-D F21-10K and All-D F17-6K peptides, essentially linear relationships are obtained. These two most active peptides exhibited a similar level of quenching in all membrane environments examined. Thus, these peptides form discrete dimers in the membrane-mimetic environment of SUV-bact bilayers. Similar results were obtained in comparable experiments performed in SDS micelles (not shown).



View larger version (9K):
[in this window]
[in a new window]
  		FIGURE 6.
FRET efficiencies in experiments using dansyl-labeled donor peptides in the presence of increasing mole fractions of the corresponding dabcyl-labeled acceptor peptides. Peptides are the labeled forms of All-D F21-10K and All-D F17-6K, as indicated on the diagram. Peptides are studied at 0.5 µM in anionic SUV-bact vesicles. Total peptide concentration is 2.5 µM. Solvent buffer is 10 mM Tris-HCl, 10 mM NaCl, pH 7.0. Curves are consistent with dimer formation in each case. Values shown are the average of at least three repetitions of each measurement. Error bars are based on standard deviations derived from the measured fluorescence intensities normalized to the value in the absence of acceptor.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
Because the majority of antimicrobial peptides are positively charged at physiological pH, a prevailing view is that selectivity stems fundamentally from electrostatic attraction of the cationic peptide to the anionic bacterial membranes (46). Although this view is consistent with the present findings, electrostatics cannot be the only contributing factor, because as mentioned in the Introduction, many cationic peptides also disrupt neutrally charged mammalian cells at higher concentrations. Therefore, more subtle properties of the peptide, including its hydrophobic moment (47), oligomerization state (48), and/or the specific type and orientation of larger residues as dictated by sequence, may also play a role in determining the extent of peptide insertion and disruption of membrane integrity. In one example, monomeric peptides were practically devoid of antimicrobial activity, whereas their pentameric covalently attached oligomers were highly active on human erythrocytes (49).

Peptide Selectivity for Bacterial Membranes—Whether as monomers or higher oligomers, it is widely believed that a general nonreceptor-mediated mechanism is responsible for peptide antimicrobial activity in most cases, which appears to involve permeabilization of phospholipid bilayer membranes via "barrel-stave," "toroidal pore," or "carpet detergent-like" formation (1, 3, 8). Microbial cell membrane disruption by the present family of CAPs can likely be ascribed, at least in large measure, to one of these nonstereospecific mechanisms. Because the CAPs studied here are relatively short, with a hydrophobic stretch of 11 residues, they would not be expected to form pores, which would require full membrane-spanning segments with {alpha}-helical structures in bilayers. Also, strong electrostatic interactions among several adjacent positively charged residues with negatively charged lipid head groups may inhibit peptide reorientation and thus the formation of well defined pores (48). In addition, the Trp depths of ~2–8 Å found in the present study are too shallow for transmembrane insertion. Thus, this class of peptides appears unlikely to operate via either the barrel-stave or toroidal pore mechanism of bacterial killing.

Within the category of the "carpet model," one can consider three possible modes of peptide insertion into membranes: parallel (0°), perpendicular (90°), and diagonal (at some intermediate angle) to the membrane surface. The experimentally determined distances of 2.4–7.9 Å from the bilayer surface for Trp residues in a 28–30 Å model membrane demonstrate relatively shallow peptide penetration that most likely corresponds to side chain insertion with the peptide positioned at least partially parallel to, and within or slightly below, the effective membrane surface region. In addition, perpendicular and/or diagonal insertion is probably not the case for Lys-separated peptides such as F17 (d = 2.4 ± 0.9 Å), as three sequential Lys residues are unlikely to be buried deeply inside the bilayer. Nevertheless, in the case of the segregated peptides, except one with positive charge on both ends (F17-6K C-term, which also has an N-terminal free amino group (d = 2.8 ± 1.4 Å)), all three geometric modes of insertion remain possible, because the depth of Trp insertion for the perpendicular mode of membrane insertion is calculated from models to be between 5.6 and 8.4 Å from the lipid-water interface, consistent with our experimentally determined values of 5.0– 7.9 Å. Calculation of magainin depth in bilayers gave similar values of 8–10 Å (50). Overall, our data regarding depth of Trp insertion suggest that the peptide backbone per se probably resides partially toward the increasingly hydrophobic region of the bilayer, with the precise depth of its penetration a complex function of both the percent/type of anionic lipids present in membrane composition, as well as the peptide sequence itself. Because all peptides must remain tethered to the membrane surface via electrostatic bonds between Lys/Arg residues and membrane phosphate groups, this sinking motion may concomitantly place torsional stress on the bound phospholipid molecules, further contributing to bilayer disruption. The inverse correlation observed between values of Trp depth in LUV-bact and peptide MIC values (Fig. 3) suggests that bacterial membranes could become increasingly susceptible to lysis as a given peptide penetrates deeply enough. Perhaps the source of the most potent antimicrobial activity of these CAPs resides in their capacity to not only insert but to do so in a manner parallel to the bilayer surface, with resulting disruption of bilayer packing.

Peptides Penetrate Bacterial Lipid Membranes as Dimers—Several peptides in the present CAP series contain AXXXA motifs in their sequences. These motifs, analogous to the well known GXXXG motif in which the two "small" residues promote close approach of helices in transmembrane dimers (40, 41), suggest that these CAPs have the innate capacity to form dimers or higher oligomers in membrane environments. Indeed, SDS-PAGE and FRET experiments confirmed the strong tendency of these peptides toward dimerization even at very low concentrations (below MIC values) (Figs. 5 and 6). With the present category of peptides, such strong dimer formation might serve as a key determinant of the high antimicrobial activity, i.e. oligomerization within the membrane surface region is expected to create substantially larger disturbances in the bilayer once insertion occurs. It should be noted that in CAPs which act through barrel-stave and toroidal pore mechanisms, peptides oligomerize before or during bilayer insertion (3). Also, synthetically produced dimers of antimicrobial peptides, e.g. by formation of intermolecular disulfid bridges, show increased activity (51).

Role of Cholesterol in CAP/Lipid Interactions—Our overall experimental data suggest that the presence or absence of cholesterol per se generally plays a secondary role in CAP interactions with various lipid mixtures. The present findings demonstrate that cholesterol does not influence insertion of peptides in a significant manner when studied in anionic bilayers without or with added cholesterol (Fig. 4). With respect to mammalian membranes, we found that this series of water-soluble peptides, containing highly charged Lys or Arg tags, ultimately cannot leave the bulk water for attachment/insertion into erythrocyte-like bilayers until the average hydrophobicity of the peptide core sequence begins to approach sufficiently high levels (as in the case where the sequence contains four Leu residues), and we found that the peptides can use this latter property as a mode of insertion only when cholesterol is absent (Fig. 2). Because a given peptide becomes antimicrobially active once the average hydropathy of its core sequence exceeds the minimal "hydrophobicity threshold" for insertion into zwitterionic micellar membranes (26), it appears that there may additionally exist an upper hydrophobicity limit, in effect creating a window of hydrophobicity that, if exceeded, becomes deleterious to peptide bioactivity. This latter phenomenon may also be associated with a tendency toward onset of hemolytic character for increasingly hydrophobic peptides. Conceptually related observations in systematic studies of the cyclic peptide antibiotic gramicidin S indicate that a defined range of peptide hydrophobicity and sequential amphiphilicity are key requirements for therapeutic effectiveness (52, 53). It should be emphasized that certain of the lipid mixtures created for study here (a bacterial lipid mixture with cholesterol (SUV-II) and a mammalian lipid mixture without cholesterol (LUV-Zwit)) would in any case not be encountered by antimicrobial peptides in vivo. Therefore, the presence of negative charge on the outer cell membrane surface of bacterial cells remains the principal reason generally for peptide selectivity.



View larger version (19K):
[in this window]
[in a new window]
  		FIGURE 7.
Energy-minimized dimer models showing the interaction surface between antiparallel helical segments from the hydrophobic core of the cationic antimicrobial peptide F17-6K. a, side view; b, end view. The model was generated using the global conformational search software CHI (see "Experimental Procedures") with idealized helices corresponding to peptide residues 9-17 (FAAWAAFAA). A dielectric constant of 1 was used in the calculation. See text for a further discussion.

 
Mechanism of Bacterial Membrane Disruption by CAPs—Data reported here reinforce the essentials of a "grip and dip" process (26), now embellished by the present experimental measurements of peptide penetration depth and peptide dimerization. Thus, antimicrobial activity appears to devolve from peptides acting broadly analogously to a carpet model mechanism by the following: (i) attachment of anti-parallel peptide dimer species to anionic (but not zwitterionic) membrane surfaces via electrostatic attractions through Lys/Arg side chains; (ii) insertion of the peptide hydrophobic core segment by several Å in a sequence-dependent manner, largely parallel to the membrane surface; and (iii) consequential membrane disruption/lysis, where the parallel-inserted peptides act as a "submarine-like" species to force the chains of the bacterial bilayer apart, in essence "unzipping" the membrane. Paradoxically, because native membrane proteins are seen to reside in bilayers with transmembrane helices aligned to the major lipid axes (i.e. perpendicular to the membrane surface), ostensibly a manner least disruptive to lipid packing, one can speculate that CAPs may potentially inflict the greatest harm to microbial cells when inserted parallel to the membrane surface, as found in the present work. Peptide dimerization would be expected to catalyze such destruction of the bacterial membrane by, in effect, forming large hydrophobic particles of damaging dimensions, likely enhanced by clustering of large aromatic residues.

Molecular modeling supports the involvement of AXXXA motifs in peptide/peptide packing interactions. An energy-minimized dimer of the putative helical regions F17-6K (Fig. 7a) indicates key Ala residues at the interface between the two peptide molecules. The model displays the anti-parallel dimer of F17-6K, as this arrangement maximally separates the polar Lys tag regions. If depicted against a membrane background, the peptide dimer in Fig. 7b projects its large Phe and Trp side chains downward toward the bilayer interior, largely within one sector of the dimer circumference, a situation that could potentially act as a locus of local disruption of lipid packing. These latter considerations represent a manifestation of where the peptide sequence is likely to play a subtle role in CAP activity. Qualitative structural analysis suggests that placement of the helical backbone of the model in Fig. 7b near a membrane surface region would position the Trp side chain in the region 5–8 Å below the surface, consistent with the values deduced from doxyl-labeled lipid experiments.

Conclusion—The exceptional advantages in activity and selectivity of our model peptides can thus be explained by their favorable combination of geometric and physical properties. The fact that the peptides studied here show no affinity for zwitterionic lipid vesicles in the presence of biologically relevant contents of cholesterol is consistent with their lack of hemolytic activity in erythrocytes (26). When evaluated in the context of magainin and other natural CAPs, which do show some hemolysis, we suspect that the grouping of sequentially consecutive positive charges on one or both termini of the present CAPs may constitute an untenable locus for adsorption to zwitterionic membranes (versus the wider distribution of polar residues in the amphipathic natural CAPs), and thus represent the main source of the extremely high level of selectivity for bacterial membranes and the low toxicity of these peptides in mammalian membranes. These properties in combination with low MIC values toward a wide spectrum of Gram-negative and Gram-positive bacteria, while retaining water solubility, represent significant advantages of this class of peptides.


   FOOTNOTES
 
* This work was supported in part by grants from the Canadian Infectious Diseases Society (to L. L. B.), the Canadian Institutes of Health Research (to C. M. D.), and the Natural and Engineering Research Council of Canada (to C. M. D.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

1 Recipient of a postdoctoral award from the Canadian Institutes of Health Research Strategic Training Program in Structural Biology of Membrane Proteins Linked to Disease. Back

2 Recipient of a Sweden-America Foundation award in 2001-2002. Present address: Dept. of Molecular Biosciences, Swedish University of Agricultural Sciences, Biomedical Centre, Box 575, S-75123 Uppsala, Sweden. Back

3 To whom correspondence should be addressed. Fax: 416-813-5005; E-mail: deber{at}sickkids.ca.

4 The abbreviations used are: CAPs, cationic antimicrobial peptides; MIC, minimum inhibitory concentration; RBC, red blood cell; LUVs, large unilamellar vesicles; SUVs, small unilamellar vesicles; FRET, fluorescence resonance energy transfer; BisTris, 2-[bis(2-hydroxyethyl)amino]-2-(hydroxymethyl)propane-1,3-diol; MES, 4-morpholineethanesulfonic acid; dansyl, N-(5-dimethylaminophthalene-1-sulfonyl) chloride; dabcyl, (4-dimethylaminophenylazobenzoyl) chloride; DIEA, N,N-diisopropylethylamine; DMF, N,N-dimethylformamide; Fmoc, 9-fluorenylmethoxy carbonyl; AL, anionic lipids; Zwit, zwitterionic; bact, bacterial. Back



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Zasloff, M. (2002) Nature 415, 389-395[CrossRef][Medline] [Order article via Infotrieve]
  2. Papo, N., and Shai, Y. (2004) Biochemistry 43, 6393-6403[CrossRef][Medline] [Order article via Infotrieve]
  3. Shai, Y. (2002) Biopolymers 66, 236-248[CrossRef][Medline] [Order article via Infotrieve]
  4. Boman, H. G. (1995) Annu. Rev. Immunol. 13, 61-92[CrossRef][Medline] [Order article via Infotrieve]
  5. Zasloff, M. (1987) Proc. Natl. Acad. Sci. U. S. A. 84, 5449-5453[Abstract/Free Full Text]
  6. Sawai, M. V., Jia, H. P., Liu, L., Aseyev, V., Wiencek, J. M., McCray, P. B., Jr., Ganz, T., Kearney, W. R., and Tack, B. F. (2001) Biochemistry 40, 3810-3816[CrossRef][Medline] [Order article via Infotrieve]
  7. Selsted, M. E., Novotny, M. J., Morris, W. L., Tang, Y. Q., Smith, W., and Cullor, J. S. (1992) J. Biol. Chem. 267, 4292-4295[Abstract/Free Full Text]
  8. Oren, Z., and Shai, Y. (1998) Biopolymers 47, 451-463[CrossRef][Medline] [Order article via Infotrieve]
  9. Hancock, R. E. (2001) Lancet Infect. Dis. 1, 156-164[CrossRef][Medline] [Order article via Infotrieve]
  10. Verkleij, A. J., Zwaal, R. F., Roelofsen, B., Comfurius, P., Kastelijn, D., and van Deenen, L. L. (1973) Biochim. Biophys. Acta 323, 178-193[Medline] [Order article via Infotrieve]
  11. Lee, W. J., and Brey, P. T. (1994) Anal. Biochem. 217, 231-235[Medline] [Order article via Infotrieve]
  12. Bessalle, R., Kapitkovsky, A., Gorea, A., Shalit, I., and Fridkin, M. (1990) FEBS Lett. 274, 151-155[CrossRef][Medline] [Order article via Infotrieve]
  13. Prenner, E. J., Lewis, R. N., Neuman, K. C., Gruner, S. M., Kondejewski, L. H., Hodges, R. S., and McElhaney, R. N. (1997) Biochemistry 36, 7906-7916[CrossRef][Medline] [Order article via Infotrieve]
  14. Shai, Y., Fox, J., Caratsch, C., Shih, Y. L., Edwards, C., and Lazarovici, P. (1988) FEBS Lett. 242, 161-166[CrossRef][Medline] [Order article via Infotrieve]
  15. Habermann, E., and Jentsch, J. (1967) Hoppe-Seylers Z. Physiol. Chem. 348, 37-50[Medline] [Order article via Infotrieve]
  16. Oren, Z., and Shai, Y. (1997) Biochemistry 36, 1826-1835[CrossRef][Medline] [Order article via Infotrieve]
  17. Argiolas, A., and Pisano, J. J. (1984) J. Biol. Chem. 259, 10106-10111[Abstract/Free Full Text]
  18. Dimarcq, J. L., Bulet, P., Hetru, C., and Hoffmann, J. (1998) Biopolymers 47, 465-477[CrossRef][Medline] [Order article via Infotrieve]
  19. Storici, P., and Zanetti, M. (1993) Biochem. Biophys. Res. Commun. 196, 1363-1368[CrossRef][Medline] [Order article via Infotrieve]
  20. Johansson, J., Gudmundsson, G. H., Rottenberg, M. E., Berndt, K. D., and Agerberth, B. (1998) J. Biol. Chem. 273, 3718-3724[Abstract/Free Full Text]
  21. Johnstone, S. A., Gelmon, K., Mayer, L. D., Hancock, R. E., and Bally, M. B. (2000) Anti-Cancer Drug Des. 15, 151-160[Medline] [Order article via Infotrieve]
  22. Oren, Z., and Shai, Y. (1996) Eur. J. Biochem. 237, 303-310[Medline] [Order article via Infotrieve]
  23. Lee, K. H., Shin, S. Y., Hong, J. E., Yang, S. T., Kim, J. I., Hahm, K. S., and Kim, Y. (2003) Biochem. Biophys. Res. Commun. 309, 591-597[CrossRef][Medline] [Order article via Infotrieve]
  24. Porter, E. A., Weisblum, B., and Gellman, S. H. (2002) J. Am. Chem. Soc. 124, 7324-7330[CrossRef][Medline] [Order article via Infotrieve]
  25. Liu, L. P., and Deber, C. M. (1998) J. Biol. Chem. 273, 23645-23648[Abstract/Free Full Text]
  26. Stark, M., Liu, L. P., and Deber, C. M. (2002) Antimicrob. Agents Chemother. 46, 3585-3590[Abstract/Free Full Text]
  27. Melnyk, R. A., Partridge, A. W., and Deber, C. M. (2001) Biochemistry 40, 11106-11113[CrossRef][Medline] [Order article via Infotrieve]
  28. Liu, L. P., and Deber, C. M. (1997) Biochemistry 36, 5476-5482[CrossRef][Medline] [Order article via Infotrieve]
  29. Mayer, L. D., Hope, M. J., and Cullis, P. R. (1986) Biochim. Biophys. Acta 858, 161-168[Medline] [Order article via Infotrieve]
  30. Chen, Y. H., Yang, J. T., and Chau, K. H. (1974) Biochemistry 13, 3350-3359[CrossRef][Medline] [Order article via Infotrieve]
  31. Choi, M. Y., Cardarelli, L., Therien, A. G., and Deber, C. M. (2004) Biochemistry 43, 8077-8083[Medline] [Order article via Infotrieve]
  32. London, E. (1982) Mol. Cell. Biochem. 45, 181-188[Medline] [Order article via Infotrieve]
  33. Chattopadhyay, A. L. E. (1987) Biochemistry 26, 39-45[CrossRef][Medline] [Order article via Infotrieve]
  34. Voglino, L., Simon, S. A., and McIntosh, T. J. (1999) Biochemistry 38, 7509-7516[CrossRef][Medline] [Order article via Infotrieve]
  35. Melnyk, R. A., Partridge, A. W., and Deber, C. M. (2002) J. Mol. Biol. 315, 63-72[CrossRef][Medline] [Order article via Infotrieve]
  36. Adams, P. D., Engelman, D. M., and Brunger, A. T. (1996) Proteins 26, 257-261[CrossRef][Medline] [Order article via Infotrieve]
  37. Deber, C. M., Wang, C., Liu, L. P., Prior, A. S., Agrawal, S., Muskat, B. L., and Cuticchia, A. J. (2001) Protein Sci. 10, 212-219[CrossRef][Medline] [Order article via Infotrieve]
  38. Wieprecht, T., Dathe, M., Schumann, M., Krause, E., Beyermann, M., and Bienert, M. (1996) Biochemistry 35, 10844-10853[CrossRef][Medline] [Order article via Infotrieve]
  39. Ladokhin, A. S., Jayasinghe, S., and White, S. H. (2000) Anal. Biochem. 285, 235-245[CrossRef][Medline] [Order article via Infotrieve]
  40. Lear, J. D., Stouffer, A. L., Gratkowski, H., Nanda, V., and Degrado, W. F. (2004) Biophys. J. 87, 3421-3429[CrossRef][Medline] [Order article via Infotrieve]
  41. Schneider, D., and Engelman, D. M. (2004) J. Mol. Biol. 343, 799-804[CrossRef][Medline] [Order article via Infotrieve]
  42. Adair, B. D., and Engelman, D. M. (1994) Biochemistry 33, 5539-5544[CrossRef][Medline] [Order article via Infotrieve]
  43. Veatch, W., and Stryer, L. (1977) J. Mol. Biol. 113, 89-102[CrossRef][Medline] [Order article via Infotrieve]
  44. Li, M., Reddy, L. G., Bennett, R., Silva, N. D., Jr., Jones, L. R., and Thomas, D. D. (1999) Biophys. J. 76, 2587-2599[Medline] [Order article via Infotrieve]
  45. Steiner, H., Andreu, D., and Merrifield, R. B. (1988) Biochim. Biophys. Acta 939, 260-266[Medline] [Order article via Infotrieve]
  46. Pouny, Y., Rapaport, D., Mor, A., Nicolas, P., and Shai, Y. (1992) Biochemistry 31, 12416-12423[CrossRef][Medline] [Order article via Infotrieve]
  47. Dathe, M., and Wieprecht, T. (1999) Biochim. Biophys. Acta 1462, 71-87[Medline] [Order article via Infotrieve]
  48. Papo, N., and Shai, Y. (2003) Peptides (Elmsford) 24, 1693-1703
  49. Sal-Man, N., Oren, Z., and Shai, Y. (2002) Biochemistry 41, 11921-11930[CrossRef][Medline] [Order article via Infotrieve]
  50. Ludtke, S. J., He, K., Heller, W. T., Harroun, T. A., Yang, L., and Huang, H. W. (1996) Biochemistry 35, 13723-13728[CrossRef][Medline] [Order article via Infotrieve]
  51. Hara, T., Mitani, Y., Tanaka, K., Uematsu, N., Takakura, A., Tachi, T., Kodama, H., Kondo, M., Mori, H., Otaka, A., Nobutaka, F., and Matsuzaki, K. (2001) Biochemistry 40, 12395-12399[CrossRef][Medline] [Order article via Infotrieve]
  52. Kondejewski, L. H., Lee, D. L., Jelokhana-Niaraki, M., Farmer, S. W., Hancock, R. E., and Hodges, R. S. (2002) J. Biol. Chem. 277, 67-74[Abstract/Free Full Text]
  53. Prenner, E. J., Kiricsi, M., Jelokhana-Niaraki, M., Lewis, R. N., Hodges, R. S., and McElhaney, R. N. (2005) J. Biol. Chem. 280, 2002-2011[Abstract/Free Full Text]
  54. Gennis, R. B. (1989) Biomembranes: Molecular Structure and Function (Cantor, C. R., ed) pp. 1-35, Springer-Verlag Inc., New York

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
 J. Biol. Chem.Home page
A. Schmidtchen, M. Pasupuleti, M. Morgelin, M. Davoudi, J. Alenfall, A. Chalupka, and M. Malmsten
Boosting Antimicrobial Peptides by Hydrophobic Oligopeptide End Tags
J. Biol. Chem., June 26, 2009; 284(26): 17584 - 17594.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
A. Ahmad, S. P. Yadav, N. Asthana, K. Mitra, S. P. Srivastava, and J. K. Ghosh
Utilization of an Amphipathic Leucine Zipper Sequence to Design Antibacterial Peptides with Simultaneous Modulation of Toxic Activity against Human Red Blood Cells
J. Biol. Chem., August 4, 2006; 281(31): 22029 - 22038.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
280/40/33960    most recent
M507042200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
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 arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Glukhov, E.
Right arrow Articles by Deber, C. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Glukhov, E.
Right arrow Articles by Deber, C. M.
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?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2005 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement