Originally published online as doi:10.1189/jlb.0804452 on December 6, 2004
Published online before print December 6, 2004
(Journal of Leukocyte Biology. 2005;77:466-475.)
© 2005
by Society for Leukocyte Biology
Mammalian defensins: structures and mechanism of antibiotic activity
Hans-Georg Sahl*,1,
Ulrike Pag*,
Sonja Bonness*,
Sandra Wagner*,
Nikolinka Antcheva
and
Alessandro Tossi
* Institute for Medical Microbiology and Immunology, University of Bonn, Germany; and
Department of Biochemistry, Biophysics and Macromolecular Chemistry, University of Trieste, Italy
1 Correspondence: Institute for Medical Microbiology and Immunology, University of Bonn, Sigmund-Freud-Str. 25, 53105, Bonn, Germany. E-mail: sahl{at}mibi03.meb.uni-bonn.de
 |
ABSTRACT
|
|---|
Antibiotic peptides are important effector molecules in host-parasite interactions throughout the living world. In vertebrates, they function in first-line host defense by antagonizing a wide range of microbes including bacteria, fungi, and enveloped viruses. The antibiotic activity is thought to be based on their cationic, amphipathic nature, which enables the peptides to impair vital membrane functions. Molecular details for such activities have been elaborated with model membranes; however, there is increasing evidence that these models may not reflect the complex processes involved in the killing of microbes. For example, the overall killing activity of the bacterial peptide antibiotic nisin is composed of independent activities such as the formation of target-mediated pores, inhibition of cell-wall biosynthesis, formation of nontargeted pores, and induction of autolysis. We studied the molecular modes of action of human defense peptides and tried to determine whether they impair membrane functions primarily and whether additional antibiotic activities may be found. We compared killing kinetics, solute efflux kinetics, membrane-depolarization assays, and macromolecular biosynthesis assays and used several strains of Gram-positive cocci as test strains. We found that membrane depolarization contributes to rapid killing of a significant fraction of target cells within a bacterial culture. However, substantial subpopulations appear to survive the primary effects on the membrane. Depending on individual strains and species and peptide concentrations, such subpopulations may resume growth or be killed through additional activities of the peptides. Such activities can include the activation of cell-wall lytic enzymes, which appears of particular importance for killing of staphylococcal strains.
Key Words: human ß-defensin 3 • membrane depolarization • staphylococci
|
INTRODUCTION
|
|---|
Antimicrobial peptides (AMPs) and host defense
AMPs are important effector molecules in innate immunity [1
2
3
]. A plethora of such molecules has been described over the past decade, and it has become obvious that they are used throughout the living world. Microorganisms themselves use them to antagonize competitors; in plants and insects, they are major effector molecules to prevent and combat microbial infections. In vertebrates, the AMP templates found in the more primitive organisms are conserved and provide a sophisticated first line of host defense, having the capacities to kill microbes directly and stimulate innate and adaptive defense systems (e.g., refs. [4
5
6
7
]). When considering the activities of AMPs, one must interpret the enormous body of work reporting their direct antimicrobial activity, often tested in vitro under nonphysiological conditions, in light of their role in vivo, which obviously reflects the sum of direct and indirect antimicrobial activities. For these reasons, some authors now prefer the term host defense peptides (HDPs) to AMPs [8
].
The body of in vivo evidence for a substantial role of these peptides in host defense is growing rapidly; Mentioned is a brief selection of recent, prominent examples.
Paneth cells in mouse small intestinal crypts secrete granules rich in microbicidal cryptdins when exposed to bacteria or bacterial antigens. The dose-dependent secretion occurs within minutes, and these 
-defensins account for 
70% of the released bactericidal peptide activity. Gram-negative bacteria, Gram-positive bacteria, lipopolysaccharide (LPS), lipoteichoic acid (LTA), lipid A, and muramyl dipeptide elicit cryptdin secretion [9
].
There is direct evidence of a correlation between defensin expression and incidence of infection in man. Human ß-defensins (hBDs) 1 and 2 are widely expressed in oral-inflamed tissue samples and primary oral keratinocytes [10
]. In patients affected by psoriasis, epithelial BDs are overexpressed, and skin lesions are relatively free from infection, and in atopic dermatitis, their expression is suppressed, and lesions are infection-prone [11
, 12
].
Inhibition of the proteolytic activation of cathelicidins in neutrophils had a drastic effect on the clearance of bacteria from wounds in an experimental porcine skin-wound model [13
], and in mice, cathelicidins were shown to protect against necrotic skin infection caused by Group A streptococci [14
]. An absence of cathelicidin, as well as defensins, has been demonstrated in patients suffering from Morbus Kostmann [15
].
The relevance of AMPs to host defense is also illustrated by counter-strategies developed by the microorganisms to escape their activities.
During Shigella infections, expression of the antibacterial peptides LL-37 and hBD-1 is reduced or turned off. The down-regulation was detected in patients with bacillary dysenteries and in Shigella-infected cell cultures of epithelial and monocyte origin. The down-regulation of immediate defense effectors might promote bacterial adherence and invasion into host epithelium and could be an important virulence parameter. Shigella plasmid DNA was identified as a putative mediator causing the down-regulation [16
].
The reduction of the overall negative charge of surface components appears to be the most effective strategy of Gram-positive and Gram-negative microbes to reduce the accumulation of AMPs in the cell envelope and thus, to escape from the direct killing activity [17
, 18
]. For example, MprF is a membrane protein of Staphylococcus aureus, which is involved in modification of phosphatidylglycerol with L-lysine; the modified phospholipid resulting from its activity is a major constituent of the staphylococcal membrane, and the negative surface charge of such a membrane is significantly reduced [19
]. An mprF– mutant strain was killed considerably faster by human neutrophils and exhibited attenuated virulence in mice, indicating a key role for defensin resistance in the pathogenicity of S. aureus. MprF homologues were identified in various other human pathogens [17
]. Similarly, the alanylation of teichoic acid and LTA reduces the surface charge [20
] but also interferes with Toll-like receptor 2-mediated defense responses [21
].
As mentioned above, the role of HDPs in innate immunity is more complex than just a capacity to antagonize microbes directly, and an increasing number of other functions, in addition to their antibiotic activity, including chemotactic or regulatory functions, are being discovered (e.g., refs. [22
23
24
]). In fact, the peculiar sensitivity of the antibiotic activity of some HDPs, in particular, most mammalian defensins and some cathelicidins, to environmental factors such as salt concentration and serum components, may cast some doubt on this being their principal role. However, although it is reasonable to suppose that the defensive role of HDPs derives from a combination of direct and defense-stimulatory activities, there is sufficient support for the significance of the direct antibiotic effect (e.g., refs. [17
, 19
]).
|
STRUCTURAL AND FUNCTIONAL CHARACTERISTICS OF HDPs
|
|---|
HDPs are, by definition, gene-encoded and derive from precursor peptides through one or two proteolytic activation steps. Structurally, they form a heterogeneous group [6
] but with some common features: They are generally small in size (10–50 amino acids), cationic, and with an amphiphilic residue distribution. Among the various molecular architectures, some structural types are predominant and ubiquitous and include ß-sheet peptides containing three or four disulfide bridges (plant, insect, and vertebrate defensins) [6
]; ß-hairpin peptides with one or two disulfide bridges (mammalian protegrins, arthropod tachyplesin, and some plant peptides) [6
]; linear peptides, which assume an -helical structure (e.g., cecropins, magainins, and some cathelicidins) [25
]; and linear-extended peptides, which are Gly-, Pro-, or Trp-rich.
The crystal structures of several defensins have been determined and indicate surprising similarity, based on a central, three-stranded ß-sheet core, despite different sizes, residue distributions, and cysteine-pairing schemes. Regarding the mammalian peptides, the structures of various -defensins and BDs have been determined, and it appears that each group represents a conserved scaffold, where the folding is apparently determined principally by the formation of the disulfides. As the sequences vary considerably, this conserved, tertiary-structure scaffold allows the molecules to display considerably different surfaces and a remarkable variation of cationicity, which ranges from almost neutral in mouse BD-2 to 11 in human BD-3. This reflects on their quaternary structure; some defensins form dimeric or oligomeric structures; others apparently remain monomeric. This likely represents evolved characteristics used to modulate the mode of action [26
]. In contrast, the helical peptides are usually unstructured in solution and adopt the typical helical, amphipathic structures in appropriate media (e.g., solvents such as trifluoroethanol or detergent micelles) and in physiological settings, in contact with biological membranes.
Although our knowledge about a number of aspects regarding the defense peptides (including biological roles, genetics of production, and structures) is currently, rapidly expanding, our understanding of the bactericidal modes of action is based mainly on pioneering studies of the last decade [27
] but is now rather stagnant. The fact that small, cationic, amphiphilic peptides are so widespread led to the postulation of a general principle of antibiotic activity, which has supposedly been conserved throughout the evolution of these molecules; i.e., these antibiotics should have a common molecular target, which should not be subject to substantial, evolutionary changes. From the defensive point of view, it is preferable to have a moderate activity against a broad range of target organisms with a common target rather than potent activities but restricted to a particular group of microorganisms having a specific target. In fact, AMPs often tend to act on several groups of micoorganisms but have relatively moderate potency [minimal inhibitory concentrations (MICs) ranging from 100 nM to 10 µM and higher], and there is considerable evidence that in most cases, the microbial membrane is involved as a target in their antibiotic action. The destructive effects on membranes with a negative surface charge by amphiphilic compounds have been known for a long time, and there is a wealth of information about the biophysical behavior of synthetic and natural defense peptides in and on model membranes (reviews are in, e.g., two special issues of Biopolymers [28
, 29
]).
The best-studied amongst AMPs, regarding the mode of action, are the -helical peptides, and various models have been elaborated to explain their membrane-compromising activity. The most prominent models are summarized in Figure 1
. These range from discrete, "barrel-stave" pore formation, which, however, seems limited to peptides such as alamethicin, to a more dynamic type of pore or channel suggested for several peptides (e.g., including magainin and melittin), diversely described as "toroidal," "wormhole," or dynamic peptide-lipid supramolecular pores [30
, 31
] or from less well-defined apertures, termed "aggregate channels" [32
]. A more generalized carpeting and subsequent disruption of the membrane also seem possible, especially at high peptide concentrations [33
].
However, shortcomings of such models are obvious, and it appears that biophysical studies go only part of the way in explaining how the actual killing of the microbes occurs. Pore or channel formation is likely to depend markedly on the precise nature of the membrane, as the entry of peptides into the bilayer results in a stretching and curvature of the outer leaflet, which will depend on the nature and abundance of specific phospholipids, affecting the concentration of peptide required to induce pore or channel formation [30
]. Furthermore, it will be accompanied by a series of other events, which may contribute, at different extents, to cellular inactivation, such as loss of compositional specificity in the membrane as a result of an increased rate of phospholipid flip-flopping and translocation of the peptides to the cytoplasmic side of the membrane [34
]. Data have been presented suggesting intracellular targets for AMPs (e.g., for the buforin peptides; ref. [35
]), and it was proposed that membrane interactions may serve primarily the permeation of the peptides into the cell and that the concomitant membrane damage may only be transient (for buforins and magainin; e.g., ref. [36
]). It seems reasonable to assume that once highly charged, amphiphilic molecules enter a cell to which they may bind unspecifically to many biomolecules, including nucleic acids or proteins, and thus, produce multiple effects in many in vitro assays; whether these effects represent the actual killing mechanism or at least contribute to the killing of microbes needs to be controlled carefully.
A more complex treatment of the interaction of AMPs with the microbial surface may much better explain the observed biological phenomena, involving components other than the membrane itself. For example, the Gram-positive bacterial cell wall has been implicated as a potential target involved in the action of BDs [10
]. Lehrer suggests [37
] for the porcine protegrin peptides a mechanism based on hydro-osmotic transtesseral extrusion and rupture mechanism; i.e., the peptides affect membrane integrity, which promotes K+ efflux and water influx, thus increasing the tugor and inducing cell rupture, at particular, yet-unidentified sites in the cell wall, which are for yet-unknown reasons, less osmotically stable. An important aspect of these considerations, however, is that although the membrane remains a principal factor, most likely the modes of action of peptides from the five structural groups mentioned above may differ and possibly even within these groups. It would thus be necessary to study different HDPs case-by-case.
A further complication for the elucidation of the modes of action is that the killing activities may not only vary or even differ for the individual groups of peptides but also for individual groups of target organisms. For example, when passing through the cell wall of a Gram-positive bacterium, defensins accumulate on anionic polymers and can activate autolysins (e.g., ref. [19
]; see below and ref. [38
]), which by dismantling the cell wall in an uncontrolled manner, contribute to the overall killing activity. In Gram-negative bacteria, they must instead pass through the LPS layer, in which its physical destabilization is an antibiotic activity by itself. In this respect, it cannot escape notice that ever since AMPs were discovered, much effort has been dedicated to developing such molecules as a new generation of antibiotics [28
, 29
] but so far, with little success. One reason may be that rational design strategies were based too narrowly on optimization of membrane interactions; i.e., they did not include all the relevant targets. These strategies were effective in increasing the membranolytic efficiency but eventually resulted in an increased, general toxicity, which reduced their applicability considerably.
|
THE LANTIBIOTICS AS A MODEL FOR SPECIFIC MEMBRANE-DIRECTED ACTIVITIES OF AMPs
|
|---|
The antibiotic activities lantibiotics, a unique class of cationic peptides of bacterial origin, provide an effective demonstration of the complexity of the mode of action of such amphipathic peptides. Lantibiotics contain the thioether amino acid lanthionine, which is introduced via post-translational modification of the prepeptide [39
]. This amino acid serves the same function as the disulfide bridges in defensins but is chemically much more stable. As early as 1985, we postulated a model similar to the barrel stave model for some lantibiotics, e.g., nisin [40
], epidermin, and Pep5 (summarized in ref. [41
]), which was later modified to a wedge model [42
] similar to the toroidal pore model in Figure 1
. These models, however, could not explain why it was necessary to apply micromolar quantities of nisin to see effects on liposomes and planar bilayers, and in vivo, it is bactericidal for a number of bacterial strains in the low nanomolar range. In the course of studying the lantibiotic mersacidin, which solely inhibits the biosynthesis of the Gram-positive cell wall [43
], we isolated the bactoprenol-bound precursor of the bacterial cell wall, lipid II. Using liposomes supplemented with lipid II, we could demonstrate that nisin was 1000-fold more active (i.e., at nM concentrations) in such a model system [44
, 45
]. Thus, the peptides obviously use lipid II as a docking molecule for specific binding to the bacterial membrane, and the subsequent pore formation could proceed at much lower concentrations than in its absence (Fig. 2
).
View larger version (31K):
[in this window]
[in a new window]
|
Figure 2. Lipid II-mediated activities of lantibiotics. Lipid II carries the completed subunit of the bacterial cell wall, which is shuttled across the cytoplasmic membrane and polymerized on the outside into the murein network. Many lantibiotics can form a complex with lipid II and thereby block the polymerization reactions (A). After binding to lipid II, some lantibiotics, e.g., nisin, can insert into the membrane and form a well-defined pore (B).
|
|
Genetically engineered nisin variants allowed us to identify the structural requirements for the interaction of the peptide with lipid II and for its capacity for forming pores in the bacterial membrane. The conformation of the N-terminal part comprising rings A–C was important for high-affinity binding of the lipid II target and consequently, the peptide concentration necessary for pore formation. Mutation of the flexible central hinge region reduced the capacity for pore formation greatly but only reduced the in vivo activity partly, as described by the MIC of the mutant peptides. The residual activity was based on the unaltered ability to bind lipid II and block it from incorporation into the cell wall similar to the activity of the glycopeptide antibiotic vancomycin. Therefore, through the interaction with the membrane-bound cell-wall precursor lipid II, nisin inhibits peptidoglycan synthesis and forms highly specific pores (Fig. 2)
, and the combination of these two killing mechanisms in one molecule potentiates the antibiotic activity and results in nanomolar MIC values toward some microorganisms [46
], and in the absence of lipid II, the peptide can display a residual activity in the micromolar range on a broader range of microorganisms, possibly based on a more generalized membrane-carpeting mechanism. Thus, the susceptibility of a given strain depends largely on the availability of lipid II as a docking molecule. Lipid II is highly conserved among eubacteria; however, its total number per cell may vary from 103 to 105 [44
] for different species. Certainly, resting cells expose fewer lipid II on the outside of the membrane than growing cells, which modulates their susceptibility. Moreover, in a more general sense, the accessibility of lipid II may be restricted permanently or transiently through species or genus-specific variation in the cell wall and membrane composition and by the architecture of the multienzyme cell-wall synthesis machinery.
However, this is not the complete picture. We had previously shown that nisin and other cationic peptides interfere with the separation of daughter cells, which normally occurs through site-specific hydrolysis of wall material after completion of the septum cell wall. Cationic peptides can displace the cell-wall lytic enzymes involved in these processes from the teichoic acid and LTA, where the enzymes are compartmentalized and thus, kept inactive when not required. This results in premature and uncontrolled lytic activity and subsequent cell lysis (Fig. 3
; ref. [47
]). Such an activity of nisin and other lantibiotics is independent from lipid II binding and contributes significantly to the overall bactericidal activity. There is every reason to suppose that such an activity could be a common feature of amphiphilic, cationic AMPs [48
].
View larger version (144K):
[in this window]
[in a new window]
|
Figure 3. Induction of autolysis in Staphylococcus simulans 22 by cationic peptides. Cells had been treated for 10 min with the lantibiotic Pep5 [38
]. Premature separation of the daughter cells as a result of activation of cell-separating enzymes is visible.
|
|
|
EFFECT OF SEQUENCE VARIATIONS IN BDs ON THE IN VITRO ANTIMICROBIAL ACTIVITY
|
|---|
Mammalian defensin sequences are highly variable, but structural studies indicate that the tertiary structures can be quite conserved. For example, in mammalian BDs, only the six invariant cysteines and a glycine are strictly conserved, but all the determined tertiary structures (hBD-1, hBD-2, and hBD-3, mouse BD-7 and -8, and bovine BD-12) consist of a triple-stranded, antiparallel ß-sheet with an N-terminal, -helical segment of differing length and stability present in all but the bovine peptide [49
50
51
52
53
]. As this conserved structural scaffold supports a considerable sequence variation, these molecules present quite different surfaces and dimerization or oligomerization patterns, which are likely important for their function.
In the hBD-1, dimerization is observed in the crystal structure but not in solution and occurs longitudinally, involving a symmetric pair of intermolecular salt bridges so that the dimer may form a sickle-shaped structure with a hydrophobic convex surface, which could lead to string-like oligomers. It has been suggested but not yet demonstrated that dimerization could have a biological significance [52
]. In contrast, hBD-2 dimerizes side-on via the formation of an extended, six-stranded ß-sheet in the crystal structure [52
]. It is surprising that the highly cationic hBD-3 forms stable dimers or oligomers in solution [53
, 54
], which appear to involve symmetric pairs of intermolecular and intramolecular salt bridges. It had been suggested that these structural characteristics could explain why of all the tested BDs, only hBD-3 shows a measurable antimicrobial activity in vitro at physiological salt concentration [10
].
It is reasonable to suppose that the variability in charge, surface properties and dimerization capacities and patterns, could play an important role in modulating the antimicrobial activity of these molecules. During the course of studies aimed at determining the effect of evolutionary variations in primate BDs, the human and macaque (Macaca fascicularis, mfa), congeners of BD-2, and human and gibbon (Hylobates concolor, hc), congeners of BD-3 (Fig. 4
), have been chemically synthesized, and structural characteristics and in vitro antimicrobial activity were determined [54
, 55
]. A completely artificial defensin, tBD, was used as a reference molecule. Its sequence was designed using a template obtained by the positional frequency analysis of 60 mammalian and avian defensins (N. Antcheva et al., unpublished results) in a manner similar to that used successfully with -helical peptides [25
].
View larger version (24K):
[in this window]
[in a new window]
|
Figure 4. Sequences of described BDs. Natural peptides are BD-2 variants from man and mfa and BD-3 variants from man and from the gibbon hc. Template BD (tBD) is an artificial defensin designed on the basis of a sequence template derived from the positional frequency analysis of 60 BDs from various mammalian and avian species. *E, Pyroglutamic acid.
|
|
hBD-2 and mfaBD-2 show a considerable sequence variation, and native polyacrylamide gel electrophoresis analyses indicated that they do not dimerize stably in solution. This does not, however, preclude that some form of dimerization or oligomerization might not occur at the microbial membrane surface, where the peptide is likely to concentrate. hBD-3, instead, forms quite a stable dimeric or more probably, multimeric complex in solution, but a single residue variation, which prevents formation of the intramolecular salt bridging, converts the gibbon congener into a monomer [54
].
Circular dichroism spectroscopy of the synthesized and folded peptides indicated that hBD-2 is well-structured in aqueous buffer [55
] and is consistent with the presence of the stable -helical, N-terminal stretch seen in the crystal structure [50
]. Spectra are largely unaffected by interaction with membrane-mimicking sodium dodecyl sulfate (SDS) micelles. The macaque congener shows a looser structure, in particular, with respect to -helical content. This was also the case of hBD-3 [54
] with the only ß-sheet core evident in aqueous buffer, and some -helical content became evident in the presence of SDS micelles. Furthermore, the structural stability appeared to decrease in the gibbon congener so that the intramolecular salt-bridging may also be relevant in stabilizing the tertiary structure. The synthetic tBD showed a stable ß-sheet structure in the presence of buffer and SDS with no evidence of helical content, likely a result of the N-terminal stretch being too short. In this respect, it could resemble the structure of the bovine neutrophil BD-12 [49
].
The peptides were tested for in vitro antimicrobial activity using different conditions, including low (5% v/v) and high (50% v/v) medium concentrations, showing somewhat differing potency patterns. The human and macaque BD-2 congeners were active only in special, low-concentration media [5% v/v Mueller-Hinton (MH) broth or Sabouraud (SAB) medium, see Table 1
], in line with the reported salt-sensitivity of defensins in general. hBD-2 was more active toward Gram-negative species and mfaBD-2 toward Gram-positive bacteria and the yeast C. albicans [55
]. Sequence and structure variations in these congeners thus seem to have a significant, functional effect with respect to antimicrobial activity. This is evidenced also by the behavior of the synthetic tBD, which has a similar cationicity but lacks the N-terminal stretch and has a somewhat different residue arrangement (Fig. 4) ; tBD showed a broad spectrum activity on all tested microorganisms, at least in terms of MIC values.
hBD-3 and hcBD-3 showed comparably potent, broad spectrum, and salt-insensitive activity, despite their different capacities to dimerize (Table 1)
[54
]. This was also evident from killing experiments carried out on representative Gram-positive and Gram-negative bacteria at increasing NaCl concentrations (Fig. 5c
). The supposition that the potent and salt-insensitive activity of hBD-3 might derive from stable dimerization in solution is thus not supported by in vitro assays, but rather, these characteristics may be imputed to the high-intrinsic cationicity of these defensins. It is, however, quite feasible that oligomerization, which results in an even higher charge-density, could play an important role in physiological settings (see also permeabilization data below).
View larger version (22K):
[in this window]
[in a new window]
|
Figure 5. Bacterial inactivation and membrane permeabilization kinetics. (a) Time-killing plots for E. coli ML-35 (107 CFU/ml, 8 µM peptide in 10 mM SPB); (b) time-killing plots for S. aureus 710A (107 CFU/ml, 8 µM peptide in 10 mM SPB); hBD-2 ( ), mfaBD-2 (•), hBD-3 ( ), and hcBD-3 ( ); (c) bacterial killing in terms of percent survival with respect to untreated bacteria, after exposure of 105 CFU/ml S. aureus (squares), E. coli (circles), and P. aeruginosa (diamonds) to 8 µM human (open symbols) and gibbon (filled symbols) BD-3 at increasing concentrations of NaCl; (d, e) Membrane permeabilization kinetics determined spectrophotometrically by following the unmasking of cytoplasmic ß-galactosidase or 6-phospho-ß-galactosidase, which were constitutively produced in specific strains of E. coli (ML-35 pYC, d) and S. aureus (710A, e), to the extracellular chromogenic substrates o-nitrophenyl-ß-D-galactopyranoside (ONPG) or ONPG-6P, respectively, as described previously (107 CFU/ml bacteria and 5 µM peptide in 10 mM SPB for both) [56
]; hBD-2 (—), mfaBD-2 (– –), hBD-3 (••••), hcBD-3 (····), and tBD (_.._; N. B., hBD-3, hcBD-3, and tBD transiently bind ONPG-6P, resulting in a signal that prevents an accurate assessment of S. aureus permeabilization kinetics. It was, however, estimated to be slow).
|
|
Time-killing studies were also revealing (Fig. 5 a and b)
. In low salt conditions, E. coli was eradicated rapidly and completely by hBD-2 and by hBD-3 and hcBD-3 [54
, 55
], and the mfaBD-2 was significantly slower. With respect to the Gram-positive S. aureus, mfaBD-2 was instead the most efficient, and hBD-2 and hBD-3 and hcBD-3 were slower. Their reasonable MIC values would thus seem to indicate that these more cationic peptides act with a slow but effective mechanism. tBD, with a minimized structure, is slower in killing than all the natural molecules (not shown).
Membrane permeabilization studies were carried out by following the kinetics of hydrolysis of extracellular chromogenic substrate to cytoplasmic galactosidases constitutively produced by special strains of E. coli and S. aureus (Fig. 5 d and e)
[54
, 55
]. These experiments confirmed a membrane-poration capacity for the BDs in general, although they displayed different permeabilization kinetics (Fig. 5 d and e)
[54
, 55
]. In line with the killing experiments, hBD-2 was the most efficient at permeabilizing the E. coli cytoplasmic membrane after an initial lag period, and mfaBD-2 was by far the most active on the S. aureus membrane. hBD-3 and hcBD-3 showed slower and quite different permeabilization curves so that their permeabilization mechanisms may differ in a manner reflecting their dimerization capacities, although the end effect in terms of MIC values and bacterial killing is similar. It was not possible to accurately assess their behavior with respect to the S. aureus membrane as a result of an artifactual signal deriving from their binding with the chromogenic substrate, but it was assessed to be slow. The synthetic tBD, with a minimized structure, was also less effective than the natural molecules in permeabilizing the bacterial membrane.
|
STRAIN VARIABILITY IN THE RESPONSE TO DEFENSINS
|
|---|
In another study, we compared the effects of hcBD-3 and tBD on three different staphylococcal strains including a well-characterized laboratory strain, S. simulans 22, and two clinical isolates (Table 2
). The aim was to detect variations within closely related strains in their response to the defense peptides. Throughout a series of experiments, which included MIC determinations, killing kinetics, membrane potential determination, leakage assays, and precursor incorporation assays, we used the same setup, i.e., the same medium (25% v/v MH broth) and the same culture volumes in identical vessels to ensure comparable growth conditions as to, for example, aeration, temperature, and a cell density of 106–107 CFU per ml. The experimental details have been described in a study with diastereomeric peptides [57
]. Moreover, the individual strains were treated with multiples of the MIC, generally 5x MIC, rather than with the same peptide concentrations for all strains.
The coagulase-negative strains S. haemolyticus I-10925 and S. simulans 22 were somewhat more susceptible to both peptides than the methicillin-resistant strain S. aureus LT1334 (Table 2) ; however, when treated with 5x MIC, comparable killing kinetics were obtained. During the first 10 min of treatment, there was a rapid decrease of the viable count by 0.5–1 log. With hcBD-3, no further decrease was observed for the next 3 h. In spite of its lower cationicity, tBD was slighty more active, as demonstrated for the S. aureus LT1334 (Fig. 6
), and caused an additional 1- to 2-log reduction of the CFU within 3 h. When cultures were further incubated overnight, regrowth of survivors was observed (not shown). The biphasic kill curve with a rapid, initial decrease and a subsequent, slow decrease may indicate two separate killing mechanisms. With S. simulans, the second phase could be accelerated strongly with elevated concentrations of tBD (Fig. 6A)
. At 10x MIC, the initial phase was followed by a rapid drop of the colony count to almost zero. Upon microscopic inspection of Gram-stained culture aliquots, hardly any intact cells could be observed, indicating that the cell wall had been actively hydrolyzed. The same phenomenon had been observed when S. simulans was treated with cationic lantibiotics such as nisin and Pep5 (Fig. 3)
, and we presume that the activation of the autolytic amidase and glucosaminidase of S. simulans, via displacement from their intrinsic inhibitors (teichoic acid and LTA), is responsible for this effect [38
, 39
].
Both defensins depolarized the cells of all three strains at 5x MIC; however, as compared with nisin [40
], the effect was rather small (10–20 mV, Fig. 7
). Technically, the inside-outside distribution of the lipophilic cation TPP+ was measured [57
]; i.e., the reduction of the potential could reflect a decrease of 10–20 mV in all cells or a stronger depolarization of a subpopulation of cells with a substantial number of cells remaining unaffected. In light of the killing kinetics, we prefer the latter explanation, as the depolarization could be representative of the cell fraction, which was killed rapidly. As the TPP+ method requires unrestricted diffusion of the probe through the cell envelope, which may be restricted in nonstandard laboratory strains, we controlled the experiment using the anionic potential-sensitive fluorescence dye bis-(1,3-dibutylbarbituric acid)trimethine oxonol, which also indicated rapid but partial depolarization (not shown), although quantitation of the membrane potential is not possible with this dye; for comparative measurements, we used nisin as a control for complete depolarization. At this point, it should be recalled that colony counts of survivors do not give information precisely about when a bacterial cell is actually killed. It can only be safely concluded that after, e.g., 10 min of treatment, a number of cells are unable to form a colony. Membrane depolarization and permeabilization may well be partial at this point and may continue, provided that the peptides do not readily diffuse from the cells after plating; considering the strong, amphipathic nature of the peptides and the high binding capacity of microbial cell envelopes for these peptides, such a scenario does not seem unlikely.
We also tested the impact of both defensins on the incorporation of precursors into biopolymers (Fig. 8
). Such experiments are primarily designed to indicate specific inhibitory activities of antibiotics but can also identify the inability of cells to actively take up precursors or perform macromolecular biosynthesis. As tBD and similarly, hcDB3 simultaneously affected DNA, RNA, protein, and peptidoglycan biosynthesis, it is likely that the majority of the cells is de-energized, in agreement with the killing kinetics and the membrane potential measurement. The experiment given in Figure 9
further supports this interpretation. Cells pretreated with the defensins were unable to actively take up glutamic acid, and uptake ceased as soon as the peptides were added. Conversely, and in contrast to the lantibiotics [41
], the labeled marker did not significantly leak out of treated cells. We are currently studying whether this is a result of the molecular radius of the marker by performing K+-electrode experiments.
|
CONCLUSIONS
|
|---|
Taken together, our studies indicate that the amino acid sequence variations observed in primate and synthetic BDs do, in effect, reflect on the antimicrobial activity but in different manners. hBD-2 and mfaBD-2 have different selectivities in all assays used. This would seem to indicate differing modes of action, which depend on the structural characteristics of the peptides, such as the stable N-terminal stretch in the human congener. The minimized synthetic defensin, lacking the N-terminal stretch, has an apparently better and broader activity in terms of MIC but is, in fact, less efficient than the natural peptides in terms of killing, particularly Gram-negative bacteria. For hBD-3 and hcBD-3, the potent and broad spectrum of activity seems to be related more to the charge than to the capacity to dimerize. Also, bacterial inactivation by hBD-3, although quite effective in the long run, is fast with the reference Gram-negative microorganism but quite slow with the Gram-positive one, possibly indicating different mechanisms of action.
The comparative experiments conducted with the staphylococcal strain also demonstrate that defensins have an impact on microbial membranes. However, the effects observed were smaller than anticipated when compared with agents forming defined pores. Complete disruption of the barrier function did not occur, at least in a major proportion of the treated cells and under the conditions tested, i.e., at 5x MIC. Possibly, membrane disruption may be more dramatic at higher peptide concentrations [58
]; however, to elucidate the complexity of defensin activities, it seems more relevant to avoid concentrations that overwhelm subtle effects that clearly contribute to killing or growth inhibition. Indeed, it will be interesting to study the physiology of cells surviving such conditions.
The staphylococcal strains tested here also differed to some extent in their response to the peptides. Particularly S. simulans 22 showed biphasic killing kinetics, indicative of two possibly independent mechanisms, which may partially overlap in time, e.g., membrane pertubation and cell-wall lysis. The lytic processes occurring with this strain could have several molecular reasons; e.g., premature activity of cell wall-degrading enzymes involved in separation of daughther cells produces rapid lysis, which agrees well with the killing kinetics observed here; however, inhibition of the polymerization steps in cell-wall biosynthesis cannot be excluded as a mechanism.
Our results indicate that although perturbation of membrane-barrier functions contributes significantly to killing, HDPs exhibit additional antibiotic activities. Their amphiphilic and cationic nature provides high reactivity and enables a maximum of interactions in biological systems. Thus, HDPs may produce, through interaction with membranes and a number of other low-affinity targets, inhibitory effects, which may result in additive or synergistic combinations during the killing process. Moreover, which of the inhibitory effects contributes most to killing may largely depend on the structure of the individual peptide, on its concentration, as well as on the target microbe species and physiological conditions of the target cells.
|
ACKNOWLEDGEMENTS
|
|---|
Work in our laboratories is supported by the Deutsche Forschungsgemeinschaft (Grant Sa 292/10-1 to H-G. S.) and by MIUR PRIN Italian Ministry of the Universities and Scientific Research (PRIN 2003057187). N. A. was supported by grants by the region Friuli-VG and the Centre of Excellence for Biocrystallography of the University of Trieste. Further support has been received from the European Commission, Contract QLK2-CT-2000-00411 within the Fifth Framework.
Received August 13, 2004;
revised October 29, 2004;
accepted November 3, 2004.
|
REFERENCES
|
|---|
- Boman, H. G. (1998) Gene-encoded peptide antibiotics and the concept of innate immunity: an update review Scand. J. Immunol. 48,15-25[CrossRef][Medline]
- Hoffmann, J. A., Kafatos, F. C., Janeway, C. A., Ezekowitz, R. A. (1999) Phylogenetic perspectives in innate immunity Science 284,1313-1318[Abstract/Free Full Text]
- Boman, H.G. (2000) Innate immunity and the normal microflora Immunol. Rev. 173,5-16[CrossRef][Medline]
- Ganz, T., Lehrer, R. I. (1998) Antimicrobial peptides of vertebrates Curr. Opin. Immunol. 10,41-44[CrossRef][Medline]
- Hancock, R. E., Diamond, G. (2000) The role of cationic antimicrobial peptides in innate host defenses Trends Microbiol 8,402-410[CrossRef][Medline]
- Tossi, A., Sandri, L. (2002) Molecular diversity in gene-encoded, cationic antimicrobial polypeptides Curr. Pharm. Des. 8,743-761[CrossRef][Medline]
- Zasloff, M. (2002) Antimicrobial peptides of multicellular organisms Nature 415,389-395[CrossRef][Medline]
- Hancock, R. E., Devine, D. R. (2004) Mammalian Host Defense Peptides ,1-4 Cambridge University Press Cambridge, UK.
- Ayabe, T., Satchell, D. P., Wilson, C. L., Parks, W. C., Selsted, M. E., Ouellette, A. J. (2000) Secretion of microbicidal -defensins by intestinal Paneth cells in response to bacteria Nat. Immunol. 1,113-118[CrossRef][Medline]
- Harder, J., Bartels, J., Christophers, E., Schroder, J. M. (2001) Isolation and characterization of human ß-defensin-3, a novel human inducible peptide antibiotic J. Biol. Chem. 276,5707-5713[Abstract/Free Full Text]
- Nomura, I., Goleva, E., Howell, M. D., Hamid, Q. A., Ong, P. Y., Hall, C. F., Darst, M. A., Gao, B., Boguniewicz, M., Travers, J. B., Leung, D. Y. (2003) Cytokine milieu of atopic dermatitis, as compared to psoriasis, skin prevents induction of innate immune response genes J. Immunol. 171,3262-3269[Abstract/Free Full Text]
- Ong, P. Y., Ohtake, T., Brandt, C., Strickland, I., Boguniewicz, M., Ganz, T., Gallo, R. L., Leung, D. Y. (2002) Endogenous antimicrobial peptides and skin infections in atopic dermatitis N. Engl. J. Med. 347,1151-1160[Abstract/Free Full Text]
- Cole, A. M., Shi, J., Ceccarelli, A., Kim, Y. H., Park, A., Ganz, T. (2001) Inhibition of neutrophil elastase prevents cathelicidin activation and impairs clearance of bacteria from wounds Blood 97,297-304[Abstract/Free Full Text]
- Nizet, V., Ohtake, T., Lauth, X., Trowbridge, J., Rudisill, J., Dorschner, R. A., Pestonjamasp, V., Piraino, J., Huttner, K., Gallo, R. L. (2001) Innate antimicrobial peptide protects the skin from invasive bacterial infection Nature 414,454-457[CrossRef][Medline]
- Putsep, K., Carlsson, G., Boman, H. G., Andersson, M. (2002) Deficiency of antibacterial peptides in patients with morbus Kostmann: an observation study Lancet 360,1144-1149[CrossRef][Medline]
- Islam, D., Bandholtz, L., Nilsson, J., Wigzell, H., Christensson, B., Agerberth, B., Gudmundsson, G. (2001) Downregulation of bactericidal peptides in enteric infections: a novel immune escape mechanism with bacterial DNA as a potential regulator Nat. Med. 7,180-185[CrossRef][Medline]
- Peschel, A. (2002) How do bacteria resist human antimicrobial peptides? Trends Microbiol 10,179-186[CrossRef][Medline]
- Gunn, J. S., Ryan, S. S., Van Velkinburgh, J. C., Ernst, R. K., Miller, S. I. (2000) Genetic and functional analysis of a PmrA-PmrB-regulated locus necessary for lipopolysaccharide modification, antimicrobial peptide resistance, and oral virulence of Salmonella enterica serovar typhimurium Infect. Immun. 68,6139-6146[Abstract/Free Full Text]
- Peschel, A., Jack, R. W., Otto, M., Collins, L. V., Staubitz, P., Nicholson, G., Kalbacher, H., Nieuwenhuizen, W. F., Jung, G., Tarkowski, A., Van Kessel, K. P., Van Strijp, J. A. (2001) Staphylococcus aureus resistance to human defensins and evasion of neutrophil killing via the novel virulence factor MprF is based on modification of membrane lipids with L-lysine J. Exp. Med. 193,1067-1076[Abstract/Free Full Text]
- Collins, L. V., Kristian, S. A., Weidenmaier, C., Faigle, M., Van Kessel, K. P., Van Strijp, J. A., Gotz, F., Neumeister, B., Peschel, A. (2002) Staphylococcus aureus strains lacking D-alanine modifications of teichoic acids are highly susceptible to human neutrophil killing and are virulence attenuated in mice J. Infect. Dis. 186,214-219[CrossRef][Medline]
- Kristian, S. A., Durr, M., Van Strijp, J. A., Neumeister, B., Peschel, A. (2003) MprF-mediated lysinylation of phospholipids in Staphylococcus aureus leads to protection against oxygen-independent neutrophil killing Infect. Immun. 71,546-549[Abstract/Free Full Text]
- Yang, D., Chertov, O., Bykovskaia, S. N., Chen, Q., Buffo, M. J., Shogan, J., Anderson, M., Schroder, J. M., Wang, J. M., Howard, O. M., Oppenheim, J. J. (1999) ß-Defensins: linking innate and adaptive immunity through dendritic and T cell CCR6 Science 286,525-528[Abstract/Free Full Text]
- Yang, D., Biragyn, A., Kwak, L. W., Oppenheim, J. J. (2002) Mammalian defensins in immunity: more than just microbicidal Trends Immunol 23,291-296[CrossRef][Medline]
- Yang, D., Biragyn, A., Hoover, D. M., Lubkowski, J., Oppenheim, J. J. (2004) Multiple roles of antimicrobial defensins, cathelicidins, and eosinophil-derived neurotoxin in host defense Annu. Rev. Immunol. 22,181-215[CrossRef][Medline]
- Tossi, A., Sandri, L., Giangaspero, A. (2000) Amphipathic, -helical antimicrobial peptides Biopolymers 55,4-30[CrossRef][Medline]
- Andreu, D., Rivas, L. (1998) Animal antimicrobial peptides: an overview Biopolymers 47,415-433[CrossRef][Medline]
- Oren, Z., Shai, Y. (1998) Mode of action of linear amphipathic -helical antimicrobial peptides Biopolymers 47,451-463[CrossRef][Medline]
- . (2000) Tossi, A. eds. Biopolymers 55,1-98 Wiley and Sons New York. [CrossRef][Medline]
- Andreu, D. eds. Biopolymers 1998;47,413-497 Wiley and Sons New York. [CrossRef]
- Huang, H. W. (2000) Action of antimicrobial peptides: two-state model Biochemistry 39,8347-8352[CrossRef][Medline]
- Matsuzaki, K., Sugishita, K., Ishibe, N., Ueha, M., Nakata, S., Miyajima, K., Epand, R. M. (1998) Relationship of membrane curvature to the formation of pores by magainin 2 Biochemistry 37,11856-11863[CrossRef][Medline]
- Hancock, R. E., Rozek, A. (2002) Role of membranes in the activities of antimicrobial cationic peptides FEMS Microbiol. Lett. 206,143-149[CrossRef][Medline]
- Shai, Y. (2002) Mode of action of membrane active antimicrobial peptides Biopolymers 66,236-248[CrossRef][Medline]
- Matsuzaki, K., Murase, O., Fujii, N., Miyajima, K. (1996) An antimicrobial peptide, magainin 2, induced rapid flip-flop of phospholipids coupled with pore formation and peptide translocation Biochemistry 35,11361-11368[CrossRef][Medline]
- Park, C. B., Kim, H. S., Kim, S. C. (1998) Mechanism of action of the antimicrobial peptide buforin II: buforin II kills microorganisms by penetrating the cell membrane and inhibiting cellular functions Biochem. Biophys. Res. Commun. 244,253-257[CrossRef][Medline]
- Takeshima, K., Chikushi, A., Lee, K. K., Yonehara, S., Matsuzaki, K. (2003) Translocation of analogues of the antimicrobial peptides magainin and buforin across human cell membranes J. Biol. Chem. 278,1310-1315[Abstract/Free Full Text]
- Lehrer, R. (2004) Overview: antimicrobial peptides, as seen from a rearview mirror Hancock, R. E. Devine, D. R. eds. Mammalian Host Defense Peptides Cambridge UK.
- Bierbaum, G., Sahl, H. (1991) Induction of autolysis of Staphylococcus simulans 22 by Pep5 and nisin and influence of the cationic peptides on the activity of the autolytic enzymes Jung, G. Sahl, H. eds. Nisin and Novel Lantibiotics Leiden The Netherlands.
- Sahl, H. G., Bierbaum, G. (1998) Lantibiotics: biosynthesis and biological activities of uniquely modified peptides from gram-positive bacteria Annu. Rev. Microbiol. 52,41-79[CrossRef][Medline]
- Ruhr, E., Sahl, H. G. (1985) Mode of action of the peptide antibiotic nisin and influence on the membrane potential of whole cells and on cytoplasmic and artificial membrane vesicles Antimicrob. Agents Chemother. 27,841-845[Abstract/Free Full Text]
- Sahl, H. (1991) Pore formation in bacterial membranes by cationic lantibiotics Jung, G. Sahl, H. eds. Nisin and Novel Lantibiotics Leiden The Netherlands.
- Driessen, A. J., van den Hooven, H. W., Kuiper, W., van de Kamp, M., Sahl, H. G., Konings, R. N., Konings, W. N. (1995) Mechanistic studies of lantibiotic-induced permeabilization of phospholipid vesicles Biochemistry 34,1606-1614[CrossRef][Medline]
- Brotz, H., Bierbaum, G., Reynolds, P. E., Sahl, H. G. (1997) The lantibiotic mersacidin inhibits peptidoglycan biosynthesis at the level of transglycosylation Eur. J. Biochem. 246,193-199[Medline]
- Brotz, H., Josten, M., Wiedemann, I., Schneider, U., Gotz, F., Bierbaum, G., Sahl, H. G. (1998) Role of lipid-bound peptidoglycan precursors in the formation of pores by nisin, epidermin and other lantibiotics Mol. Microbiol. 30,317-327[CrossRef][Medline]
- Breukink, E., Wiedemann, I., van Kraaij, C., Kuipers, O. P., Sahl, H., de Kruijff, B. (1999) Use of the cell wall precursor lipid II by a pore-forming peptide antibiotic Science 286,2361-2364[Abstract/Free Full Text]
- Wiedemann, I., Breukink, E., van Kraaij, C., Kuipers, O. P., Bierbaum, G., de Kruijff, B., Sahl, H. G. (2001) Specific binding of nisin to the peptidoglycan precursor lipid II combines pore formation and inhibition of cell wall biosynthesis for potent antibiotic activity J. Biol. Chem. 276,1772-1779[Abstract/Free Full Text]
- Bierbaum, G., Sahl, H. G. (1987) Autolytic system of Staphylococcus simulans 22: influence of cationic peptides on activity of N-acetylmuramoyl-L-alanine amidase J. Bacteriol. 169,5452-5458[Abstract/Free Full Text]
- Ginsburg, I. (2001) Cationic peptides from leukocytes might kill bacteria by activating their autolytic enzymes causing bacteriolysis: why are publications proposing this concept never acknowledged? Blood 97,2530-2531[Free Full Text]
- Zimmermann, G. R., Legault, P., Selsted, M. E., Pardi, A. (1995) Solution structure of bovine neutrophil ß-defensin-12: the peptide fold of the ß-defensins is identical to that of the classical defensins Biochemistry 34,13663-13671[CrossRef][Medline]
- Hoover, D. M., Rajashankar, K. R., Blumenthal, R., Puri, A., Oppenheim, J. J., Chertov, O., Lubkowski, J. (2000) The structure of human ß-defensin-2 shows evidence of higher order oligomerization J. Biol. Chem. 275,32911-32918[Abstract/Free Full Text]
- Bauer, F., Schweimer, K., Kluver, E., Conejo-Garcia, J. R., Forssmann, W. G., Rosch, P., Adermann, K., Sticht, H. (2001) Structure determination of human and murine ß-defensins reveals structural conservation in the absence of significant sequence similarity Protein Sci 10,2470-2479[Abstract/Free Full Text]
- Hoover, D. M., Chertov, O., Lubkowski, J. (2001) The structure of human ß-defensin-1: new insights into structural properties of ß-defensins J. Biol. Chem. 276,39021-39026[Abstract/Free Full Text]
- Schibli, D. J., Hunter, H. N., Aseyev, V., Starner, T. D., Wiencek, J. M., McCray, P. B., Tack, B. F., Vogel, H. J. (2002) The solution structures of the human ß-defensins lead to a better understanding of the potent bactericidal activity of hBD3 against Staphylococcus aureus J. Biol. Chem. 277,8279-8289[Abstract/Free Full Text]
- Boniotto, M., Antcheva, N., Zelezetsky, I., Tossi, A., Palumbo, V., Verga Falzacappa, M. V., Sgubin, S., Braida, L., Amoroso, A., Crovella, S. (2003) A study of host defense peptide ß-defensin 3 in primates Biochem. J. 374,707-714[CrossRef][Medline]
- Antcheva, N., Boniotto, M., Zelezetsky, I., Pacor, S., Falzacappa, M. V., Crovella, S., Tossi, A. (2004) Effects of positively selected sequence variations in human and Macaca fascicularis ß-defensins 2 on antimicrobial activity Antimicrob. Agents Chemother. 48,685-688[Abstract/Free Full Text]
- Tossi, A., Tarantino, C., Romeo, D. (1997) Design of synthetic antimicrobial peptides based on sequence analogy and amphipathicity Eur. J. Biochem. 250,549-558[Medline]
- Pag, U., Oedenkoven, M., Papo, N., Oren, Z., Shai, Y., Sahl, H. G. (2004) In vitro activity and mode of action of diastereomeric antimicrobial peptides against bacterial clinical isolates J. Antimicrob. Chemother. 53,230-239[Abstract/Free Full Text]
- Patrzykat, A., Friedrich, C. L., Zhang, L., Mendoza, V., Hancock, R. E. (2002) Sublethal concentrations of pleurocidin-derived antimicrobial peptides inhibit macromolecular synthesis in Escherichia coli Antimicrob. Agents Chemother. 46,605-614[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
S. Rotem, I. S. Radzishevsky, D. Bourdetsky, S. Navon-Venezia, Y. Carmeli, and A. Mor
Analogous oligo-acyl-lysines with distinct antibacterial mechanisms
FASEB J,
August 1, 2008;
22(8):
2652 - 2661.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. Abtin, L. Eckhart, M. Mildner, F. Gruber, J.-M. Schroder, and E. Tschachler
Flagellin is the principal inducer of the antimicrobial peptide S100A7c (psoriasin) in human epidermal keratinocytes exposed to Escherichia coli
FASEB J,
July 1, 2008;
22(7):
2168 - 2176.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. Raafat, K. von Bargen, A. Haas, and H.-G. Sahl
Insights into the Mode of Action of Chitosan as an Antibacterial Compound |
TOP
ABSTRACT
INTRODUCTION
Outer membrane in Gram-positive bacteria, thick peptidoglycan in Gram-negative bacteria; 
electrostatic attraction to anionic bacterial membrane surface; 
formation of an amphipathic structure capable of membrane insertion; 
possible permeabilization mechanisms, which depend on peptide sequence/concentration and membrane characteristics; 
translocation and binding to internal molecular targets, including proteins and nucleic acids: 
e.g., release of autolytic enzymes.
) without addition of peptide. (a, inset) Optical density at 600 nm (OD600) taken in parallel with the colony count.
); hcBD-3 (
