JAC Advance Access originally published online on January 3, 2008
Journal of Antimicrobial Chemotherapy 2008 61(2):341-352; doi:10.1093/jac/dkm479
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Original research |
Analysis of in vitro activities and modes of action of synthetic antimicrobial peptides derived from an 
-helical ‘sequence template’
1 Institute for Medical Microbiology, Immunology and Parasitology—Pharmaceutical Microbiology Section, University of Bonn, 53105 Bonn, Germany 2 Department of Biological Chemistry, The Weizmann Institute of Science, 76100 Rehovot, Israel 3 Institute for Experimental Medicine of the Russian Academy of Medical Sciences, 197376 St Petersburg, Russian Federation 4 Department of Biochemistry, Biophysics and Macromolecular Chemistry, University of Trieste, 34127 Trieste, Italy
* Corresponding author. Tel: +49-228-73-5272; Fax: +49-228-73-5267; E-mail: sahl{at}mibi03.meb.uni-bonn.de
Received 16 August 2007; returned 10 October 2007; revised 16 November 2007; accepted 18 November 2007
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsSupplementary dataReferences |
|---|
Objectives: Cationic antimicrobial peptides (AMPs) are indispensable components of innate immune systems and promising candidates for novel anti-infective strategies. We rationally designed a series of peptides based on a template derived from known
-helical AMPs, which were then analysed regarding efficacy against clinical isolates and antibiotic mechanisms.
Methods: Efficacy tests included standard MIC and synergy assays. Whole cell assays with staphylococcal strains included killing kinetics, efflux experiments and determination of membrane depolarization. The transcriptional response of AMP-treated Staphylococcus aureus SG511 was analysed using a Scienion genomic microarray covering (
90% of) the S. aureus N315 genome and AMP P16(6|E).
Results: The AMPs showed remarkable broad-spectrum activity against bacteria and fungi regardless of any pre-existing antibiotic resistance mechanism. Whole cell assays indicated that the AMPs target the cytoplasmic membrane; however, significant membrane leakage and depolarization was only observed with a standard laboratory test strain. Transcriptional profiling identified up-regulation of putative efflux pumps and of aerobic energy generation mechanisms as major counter activities. Important components of the staphylococcal cell wall stress stimulon were up-regulated and the lipid metabolism was also affected.
Conclusions: The broad spectrum activity of amphiphilic helical AMPs is based on multiple stresses resulting from interactions with microbial membranes; however, rather than killing through formation of pores, the AMPs appear to interfere with the coordinated and highly dynamic functioning of membrane bound multienzyme complexes such as electron transport chains and cell wall or lipid biosynthesis machineries.
Keywords: membrane permeabilization , intracellular targets , transcriptional profiling , microarray
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsSupplementary dataReferences |
|---|
Cationic antimicrobial peptides (AMPs) represent a conserved and highly effective component of innate immunity. They are ubiquitous in nature and have been isolated from a wide variety of sources including bacteria, invertebrates, vertebrates and plants.1–4 AMPs exhibit a broad spectrum of killing activity in vitro against various targets, such as bacteria, fungi, enveloped viruses, parasites and even tumour cells.5,6 Moreover, taking into account the considerable potential of some AMPs to modulate the innate immune system,3,7 they clearly provide a new strategy for the development of novel anti-infective agents.
Although AMPs can show a marked variation in size, sequence and structure, they share two common and functionally important characteristics: they are polycationic, a feature that attracts them to anionic microbial surfaces, and they tend to spatially separate polar and hydrophobic residues, which favours the interaction with and insertion into microbial membranes.8,9 Naturally occurring AMPs are generally 12–50 amino acids long, and most fit into four broad groups: peptides forming (i) -helices, (ii) β-sheets, (iii) extended helices and (iv) loops.1 The largest group obviously comprises those peptides which fold into an amphipathic -helical conformation when interacting with the target membrane. Their activity depends on several parameters including the sequence, size, propensity for helix formation, cationicity, hydrophobicity and amphipathicity. In recent studies, a sequence template for -helical peptides was developed based on the analysis of numerous natural AMPs, in order to generate potent artificial lead AMPs.10 Sequences were then varied in a rational manner, using both natural and non-proteinogenic amino acids to systematically and individually vary the structural and physical parameters.11 These de novo designed AMPs had been evaluated with regard to their activity against standard laboratory strains, their haemolytic activity as well as their structural and physico-chemical characteristics, demonstrating their high potential for the development of new anti-infectives.
In this paper, we describe the in vitro activities of a number of such synthetic -helical peptides against a set of multiresistant clinical isolates, confirming their potential against both multiresistant Gram-positive and Gram-negative bacteria. In addition, we address the controversial issue of the molecular mechanisms by which such AMPs kill microbes using a series of whole cell assays as well as microarray techniques to analyse the transcriptional response to AMP stress. Since methicillin-resistant Staphylococcus aureus (MRSA) strains pose an ever increasing problem, we used clinical MRSA strains as test organisms in comparison with a laboratory strain, Staphylococcus simulans 22, which is well characterized for mode of action studies. By correlation of cell viability and with various cytoplasmic membrane functions under identical experimental conditions, we could demonstrate that membrane perturbations contribute most to the killing activity; however, transcriptional response patterns of cells treated with sublethal peptide concentrations indicate that cell death may result from interference with the dynamic organization of membrane bound pathways rather than just from membrane lesions or pore formation.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsSupplementary dataReferences |
|---|
Bacterial strains
Eighteen unrelated clinical isolates as well as two standard laboratory strains were included in this study: S. simulans 22 and Micrococcus luteus ATCC 4698 as well-characterized laboratory strains (e.g. Wiedemann et al.12); bacterial and fungal clinical isolates comprised Enterococcus faecium (two strains, I-11305b and I-11054); Staphylococcus haemolyticus, coagulase-negative (CoNS; I-10925); Staphylococcus epidermidis, methicillin-resistant (MRSE; LT1324); S. aureus, methicillin-susceptible (two strains, 5185 and I-11574); S. aureus, methicillin-resistant (MRSA; two strains, LT1338 and LT1334); Citrobacter freundii (I-11090); Klebsiella pneumoniae (I-10910); Escherichia coli (two strains, I-11276b and O-19592); Stenotrophomonas maltophilia (two strains, O-16451 and I-10717); Pseudomonas aeruginosa (two strains, 4991 and I-10968); and Candida albicans (two strains, I-11301 and I-11134). Furthermore, a Listeria monocytogenes strain (322) and its nisin-resistant variant (322N) were tested against selected peptides. In addition, the following strains were used for activity tests in the course of designing the peptides: E. coli ATCC 43827 (ML-35), P. aeruginosa ATCC 27853, Salmonella typhimurium ATCC 14028, S. aureus 710A and Bacillus megaterium BM11. All strains were maintained on Mueller–Hinton (MH) agar or on blood agar.
The strains used are multiresistant. See Table S1 [available as Supplementary data at JAC Online (http://jac.oxfordjournals.org/)] for the resistance pattern of the strains used in this study.
All strains were subcultured weekly on blood agar (Becton Dickinson, Erembodegem, Belgium) or MH agar (Oxoid, Basingstoke, UK). For the efflux experiments, PYG medium containing 0.2% Bacto-Peptone (Difco, MI, USA), 0.8% glucose (Merck, Darmstadt, Germany), 0.4% yeast extract (Oxoid) and 2 mM potassium phosphate (Merck) (pH 7) was used.
Peptide synthesis and characterization
Solid-phase peptide syntheses were performed on a PE Biosystems PioneerTM peptide synthesizer as described previously.13 The peptides were purified by preparative reversed phase-HPLC (RP-HPLC) (Waters Delta-PakTM C18, 15 µm, 300 Å, 25 x 100 mm) and eluted with a 0% to 60% gradient of acetonitrile in 0.05% trifluoroacetic acid in water. The correctness and purity of peptides were determined by analytical RP-HPLC (Waters Symmetry® C18, 3.5 µm, 100 Å, 4.6 x 50 mm), followed by mass determination of the eluate with an API I electrospray ionization mass spectrometer (PE Biosystems/Sciex, Foster City, CA, USA). Peptide concentrations were determined using tyrosine absorption (
280 = 1290 M–1 cm–1) and confirmed using the bicinchoninic acid assay (Pierce, Rockford, IL, USA).
CD spectra were obtained on a Jasco J-715 spectropolarimeter (Jasco, Tokyo, Japan) using 2 mm path length quartz cells and peptide concentrations of 40 µM, in 5 mM sodium phosphate buffer, pH 7, in the presence of 50% trifluoroethanol or 10 mM SDS. The percentage helicity was determined as ([
]meas – []rc)/([] – []rc), where []meas is the measured ellipticity at 222 nm, []rc the ellipticity for unstructured peptide in the absence of additives (generally close to zero) and [] the ellipticity of a fully structured helix of length n, calculated using the relation [] = 39 000 (1 – 4/n).14 The presence of achiral -branched amino acid Aib was taken into account in these calculations.15
Determination of MICs was performed in 96-well polypropylene microtitre plates (Life Technologies) throughout to reduce AMP binding.16 A series of 2-fold dilutions in MH broth was prepared from a stock solution of the respective peptide (16 µM). The indicator strains were grown to an optical density (600 nm) of 1.0 in half-concentrated MH broth and diluted 1:105 with the same medium. Then 100 µL of this suspension was mixed with 100 µL of the peptide dilution in the well of a microtitre plate. After incubation for 18 h at 37°C, the MIC was read as the lowest concentration of antimicrobial agent resulting in the complete inhibition of visible growth, and the results given are mean values of three independent determinations.
Lysis of erythrocyte membranes was determined by monitoring the release of haemoglobin at 415 nm from 0.5% human erythrocyte suspensions in PBS, in relation to complete (100%) haemolysis as determined by addition of 0.2% Triton X-100.
Synergistic activities were assayed by chequerboard titrations with half-concentrated MH broth. Fractional inhibitory concentration (FIC) indices were calculated as follows: [(A)/MICA] + [(B)/MICB] = FICA + FICB = FIC index, where MICA and MICB are the MICs of drugs A and B when used alone, and (A) and (B) are the MICs of drugs A and B when used in combination. The interaction was defined as synergistic if the FIC index was 
0.5 and antagonistic if the FIC index was >4.0; FIC values of >0.5–4.0 indicate that both compounds do not interact.17
All strains were grown overnight in half-concentrated MH broth and diluted in fresh medium to an optical density of 0.1. Peptides were added in concentrations corresponding to 4x and 10x the MIC as determined after 18 h. The viable count was monitored up to 22 h. Aliquots were taken at defined intervals, diluted in 10 mM potassium phosphate buffer, and 100 µL of the dilutions were plated in triplicate on MH agar. The plates were incubated at 37°C and the colony forming units (cfu) were counted after 24 h.
Efflux of radioactively labelled glutamate
The influence of peptides P19(5|B), P19(9) and P19(6|E) on the uptake and retention of glutamate was investigated as described previously with some modifications.18 Briefly, strains were cultured in PYG medium at 37°C to an absorbance of 1.0 at 600 nm. Centrifuged cells were resuspended 1:3 in fresh medium, which was supplemented with 100 µg of chloramphenicol per mL to prevent glutamate incorporation. After 15 min of pre-incubation, radiolabelled L-[3H]glutamate (42 Ci/mMol) was added (final concentration 10 µCi/mL), and the culture was immediately divided into two parts. One aliquot was transferred into a flask that already contained the respective peptide (10x the MIC for the respective strain) to test its effect on the uptake of glutamate; the other part was run as control. After 30 min of incubation, the control was further subdivided into two aliquots, one of which received the respective peptide (10x the MIC) to determine the AMP effect on pre-accumulated amino acids. Samples of 100 µL were filtered through cellulose acetate filters (pore size, 0.2 µm; Schleicher & Schüll, Dassel, Germany) and washed twice with 5 mL of 200 mM potassium phosphate buffer (pH 7) containing 100 µM unlabelled glutamate. Filters were dried and transferred to counting vials filled with scintillation fluid (Quickszint 100; Zinsser, Frankfurt, Germany). The radioactivity was measured in a β-counter (1900CA; Packard).
Estimation of the membrane potential
Cells were grown in PYG medium at 37°C to an absorbance of 1 at 600 nm, centrifuged and resuspended 1:3 in fresh medium. To monitor the membrane potential, 1 µCi/mL of [3H]tetraphenylphosphonium bromide (TPP+; 26 Ci/mMol) was added. TPP+ is a lipophilic cation which diffuses across the bacterial membrane in response to a trans-negative 
. The culture was treated with P19(5|B), P19(9) or P19(6|E), respectively (10x the MIC for the respective strain), and samples were filtered and washed as described above. Counts were corrected for unspecific binding of [3H]TPP+ by subtracting the radioactivity of 10% butanol-treated cell aliquots. For calculation of the membrane potential (), TPP+ concentrations were applied to the Nernst equation [ = (2.3 x R x T/F) x log (TPP+in/TPP+out)]. A mean was calculated from a minimum of two independent determinations.
Fluorimetric assay for membrane potential
Cells were grown in half-concentrated MH broth to an OD600 of 0.5 and incubated for 5 min with 1 µM of the membrane-potential-sensitive fluorescent probe bis-(1,3-dibutylbarbituric acid) trimethine oxonol [DiBAC4(3); Molecular Probes—Invitrogen, Karlsruhe, Germany]. P19(5|B), P19(9) or P19(6|E) were added at a concentration corresponding to 10x MIC. Fluorescence was measured at the excitation and emission wavelengths of 492 and 515 nm, respectively.
For analysis of long-term changes, S. aureus SG511 was grown in the presence of 0.5 µM P19(6|E) (0.25x MIC) to an optical density of 1.0 (22 h); short-term responses were recorded with cells grown to an optical density of 0.7, and then treated with 1 µM P19(6|E) (0.5x MIC) for 20 min. In order to ensure reliable gene expression, the RNA was immediately stabilized by addition of two volumes RNAprotect Bacteria reagent (Qiagen, Hilden, Germany) followed by incubation for 5 min at room temperatures. Cells were harvested by centrifugation, lysed in the presence of 400 mg/L lysostaphin (Dr Petry Genmedics GmbH, Reutlingen, Germany) and total RNA was isolated using the PrestoSpin R Kit (Molzym, Bremen, Germany) according to the instructions of the manufacturer. The RNA was concentrated by using the RNeasy MinElute Kit (Qiagen) and spectrophotometrically quantified.
Synthesis and fluorescent labelling of cDNA
Total RNA (9 µg) was transcribed into cDNA applying the BioScript reverse transcriptase (Bioline, Luckenwalde, Germany). For simultaneous labelling of the cDNA, the transcription reaction was performed in the presence of 0.1 mM cyanine-3'- or cyanine-5'-labelled dCTP (Perkin-Elmer Life Science, Mechelen, Belgium) and 0.2 mM dCTP, 0.5 mM dATP, dGTP and dTTP, 75 mg/L random hexamer primer (GE Healthcare, Munich, Germany) and 25 U/mL RNase-Out (Invitrogen). RNA was degraded by alkaline hydrolysis at 65°C and cDNA was purified using the MinElute PCR Purification Kit (Qiagen). The concentration of cDNA as well as the amount of incorporated dye was measured with a NanoDrop ND-1000 spectrophotometer (Peqlab, Erlangen, Germany).
In order to compare the transcription profiles of S. aureus SG511 in the presence and absence of P19(6|E), differentially labelled cDNAs of both samples were competitively hybridized to the custom PCR product Full Genome Chip sciTRACER (Scienion AG, Berlin, Germany) representing 2338 open reading frames of the S. aureus N315 genome. RNA samples from three independent cultures of each approach were prepared and each RNA sample was hybridized to two separate gene chips, including a dye swap. The gene chips were hybridized for up to 72 h at 42°C, washed for 5 min at room temperature in SSC buffer with decreasing concentrations of salt (1x SSC/0.3% SDS, 0.2x SSC, 0.06x SSC) and read out with a GenePix 4000B scanner (Axon Instruments/Distribution by Biozyme, Oldendorf, Germany). Image analysis and acquisition of relative data were performed by using GenePixPro 4.1 software (Axon Instruments). The single datasets were normalized by applying the LOWESS algorithm and subsequently merged using acuity 3.1 software (Axon Instruments). Finally, significant changes in gene expression were identified with SAM (significance analysis of microarrays) software using the one class response type and a false discovery rate of <1%.19 These data were filtered for genes having at least a 5- (short-term) or 3-fold change (long-term) in expression.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsSupplementary dataReferences |
|---|
Peptide synthesis
All peptides (Table 1) were designed using a ‘sequence template’ obtained by analysing over 150 naturally occurring -helical peptides,9–11 produced by several vertebrate (mammalian, fish and amphibian) and invertebrate animal (insect) species, and by initially ‘filling’ it with a limited set of amino acids (Gln, Glu, Orn and Nle), so as to obtain different cationicities [+5, +6 and +9 in the P19(5), P19(6) and P19(9) series, respectively]. Gly was invariant at position 1 (as present in over 70% of natural peptides) and also at position 18 to separate a Tyr (required for peptide quantification) from the helical domain, and thus reduce its contribution in CD spectra.20 Peptides were invariably amidated to increase cationicity and facilitate syntheses. The -branched amino acid Aib was introduced in some peptides as a helix-stabilizing agent.15 Subsequently, all residues in P19(9) were replaced with the most closely equivalent proteinogenic amino acid for bioproduction purposes. Gly substitutions were then effected at two key positions (7 and 14) to modulate selectivity.13
|
The propensity for helix formation for all the designed peptides was evaluated in terms of percentage helix content in the presence of 50% trifluorethanol (a helix promoting solvent) or SDS micelles and is reported in Table 2.
|
Antimicrobial activity
Throughout the design procedure, activity was tested against five type culture collection strains. All peptides except for P19(9|G14) and P19(9|G7,14) displayed potent broad-spectrum activity against both Gram-positive and Gram-negative bacteria with MIC values between 0.5 and 4 µM (Table 2). Since for most whole cell assays described below 108–109 cells/mL were needed, we performed MIC determinations with both the standard inoculum of 105 cfu/mL and with 108 cfu/mL; however, a significant inoculum effect was not observed (data not shown). P19(9|G14) and P19(9|G7,14) were less active demonstrating the importance of the amino acid in position 14 for the biological activity. Nevertheless, both peptides, which also have a considerably reduced haemolytic activity,13 continued to show antimicrobial activity against E. coli and P. aeruginosa.
For testing clinical isolates, we selected strains displaying multiresistance to antibiotics, including those of current clinical concern such as 4-quinolone-resistant Enterobacteriaceae, methicillin-resistant S. aureus, VanA-type vancomycin-resistant enterococci and genera of intrinsically high natural resistance such as Stenotrophomonas and Pseudomonas and, if possible, the respective susceptible counterpart strains. Regardless of any pre-existing resistance mechanism, most peptides showed very good activity with MIC values between 0.5 and 4 µM against these isolates (Table 3). Like with the laboratory strains, the peptide P19(9|G7,14) showed a generally reduced activity. For subsequent mode of action studies, we selected the globally most active peptides P19(5|B), P19(6|E) and P19(9), which show a good spread of cationicities.
|
Cytotoxic activity
The ability of AMPs to lyse blood cells can be taken as a preliminary indication of the cytotoxicity and thus the selectivity of AMPs. As shown in Table 2, most peptides have a basal haemolytic activity of 10%, which may partly derive from the relatively low erythrocyte concentrations (0.5%) we used to increase the assay’s sensitivity. For most peptides, the extent of haemolysis became significant at 100 µM, a concentration well above their MIC values, although the absence of helix-favouring Aib residues [P19(9)] or the presence of Gly residues in key positions [P19(9|Gn)] series, correlated with a decreased lysis of red blood cells.
Chequerboard titrations were carried out using P19(9) and P19(6|E) in combination with each other, with different conventional antibiotics and with another type of the de novo designed peptide, the diastereomeric AMP ‘amphipathic-2D’, which also has broad spectrum antimicrobial activity and acts synergistically with several conventional antibiotics.21 We did not observe synergistic effects between P19(9) and P19(6|E); however, remarkable synergy was observed for some combinations with amphipathic-2D (Table 4). Surprisingly, synergy patterns were not consistent for individual strains, making interpretations of the underlying mechanisms difficult. In line with the transcriptional data reported below, it appears that the antibiotic activity of AMPs is based on multiple stresses resulting from various low affinity interactions with membrane components. These may differ within each strain in some molecular detail for the all-D peptide and for the peptides of the series reported here and thus cause synergistic effects in some strains and no effects in others.
|
Mode of action assays with whole cells of S. simulans 22
Mode of action assays with whole cells of S. simulans 22 revealed a relatively homogeneous picture for all three peptides consistent with membrane impairment as the primary mode of action (Figure 1). Killing was rapid; immediately after contact with the AMP, the cfu were reduced by several log. Complete killing, i.e. no survivors detectable after 22 h, was achieved with 10x the MIC of P19(5|B) and P19(6|E), whereas with P19(9) 4x MIC was sufficient (Figure 1). P19(9) is more cationic (+9) than the other two peptides (+5 and +6), which may be of particular relevance for the interaction with some bacteria such as S. simulans. The cell wall splitting system of this species is strongly activated by cationic peptides such as nisin, Pep5 and defensins, and the number of charges was found to correlate with the level of activation.22–24 It appears very likely that rapid cell wall hydrolysis significantly contributes to killing of S. simulans.
|
Effects on the integrity of the cytoplasmic membrane should impair energy-dependent transport processes. Indeed, when pre-treated with P19(5|B), P19(9) or P19(6|E), respectively, cells were unable to carry out active transport of glutamate (Figure 1b, e and h). Addition of the peptides to cells, which had accumulated but not incorporated radiolabelled glutamate, induced rapid efflux of the marker within a few minutes. Also, the potential across the cytoplasmic membrane (
), as estimated by means of radiolabelled tetraphosphonium bromide (TPP+), dropped from approximately –140 mV (negative inside) to –100 mV (Figure 1c, f and i) after addition of P19(5|B), P19(9) or P19(6|E).
In theory, such a level of energization should be sufficient for survival; however, since the method applied gives an average value for all cells present, it is likely, in agreement with kill kinetics (Figure 1a, d and g), that the decrease in reflects the presence of depolarized cells and survivors.
When the same series of experiments was performed with a methicillin-resistant S. aureus clinical isolate (strain LT1334), the picture was much less consistent. Generally, this strain was 2- to 8-fold less susceptible than S. simulans 22 to all three peptides (Table 3). At 4x MIC, the effect on the viability of exponentially growing cells was comparatively small and growth was rapidly resumed [e.g. P19(6|E), Figure 2a]. After 20 min treatment with 10x MIC, the viable count was still in the range of 105; however, complete killing could be achieved after 4 h. (Figure 2a). The impact of the peptides [e.g. P19(9), Figure 2b] on the accumulation of glutamate by pre-treated cells was as significant as with S. simulans; however, the ability to retain the amino acid in the cytosol was hardly impaired. Similarly, no direct impact of the peptides on the energization state of the membrane was observed [e.g. P19(5|B), Figure 2c]; both results do not argue for generalized membrane poration as a principal killing mechanism in this strain. The membrane potential of this clinical MRSA isolate was estimated to be about –120 mV, which is 20–30 mV lower than that usually found for staphylococci. This may be attributed to QacA-type export proteins which frequently occur in clinical staphylococcal isolates and which export lipophilic drugs including the TPP+ probe used here.25,26 Extrusion of the probe could account for the underestimation of ,21 which, however, should not disguise a direct impact of the peptides on the energization state of the membrane. To exclude any technical problems, we used the fluorescent potential probe bis-(1,3-dibutyl-barbituric acid) trimethine oxonol [DiBAC4(3)] in addition to monitor the effect of the peptides (data not shown). Some increase in fluorescence, indicative of membrane depolarization, was observed with this technique; however, the use of fluorescent probes can only visualize changes in, but not enable quantitative determination of, .
|
Analysis of the transcriptional response pattern to P19(6|E)
Treatment with antibiotics provokes inhibitor-characteristic response patterns on the transcriptome or proteome level which provide information on modes of action. For example, the cell wall stress stimulon of Gram-positive bacteria, which is activated by cell wall active antibiotics, is well characterized in bacilli and staphylococci.27–29 In a proteomics-based approach, clusters of marker proteins were identified that reflected antibiotic mechanisms and allowed the prediction of mechanisms of novel antibiotic compounds.30 When we analysed the immediate response to treatment of S. aureus SG511 with P19(6|E) (sublethal concentration, 20 min), we found a total of 638 genes involved in all aspects of cellular physiology to be differentially transcribed demonstrating that the peptide exerts multiple stresses at various cellular sites. Complete lists of the 355 up- and 283 down-regulated genes are given in Tables S2 and S3 [available as Supplementary data at JAC Online (http://jac.oxfordjournals.org/)]; Tables 5 and 6 list the most drastic changes with arbitrary cut-offs set at 5-fold changes for short-term responses and 3-fold for long-term adaptation, respectively.
|
|
In particular, genes coding for proteins involved in energy generation and in carbohydrate transport and metabolism such as msmX (multiple sugar-binding transport ATP-binding protein), SA0223 (acetyl-CoA acetyltransferase homologue), aldA (aldehyde dehydrogenase homologue), SA0209 (maltose/maltodextrin transport permease homologue) and acsA (acetyl-CoA synthetase) were strongly up-regulated; this suggests that the activity of P19(6|E) and related peptides may primarily hit membrane-bound energy generation pathways and that cells struggle to produce energy via tricarboxylic acid cycle and the aerobic respiratory chain. Consistent with this is the up-regulation of NADH-dependent dehydrogenases (SA0210 and SA0211) and various other dehydrogenases that could generate reduction equivalents, as well the strong down-regulation of lactate dehydrogenase LctE (SA0232), of a hypothetical lactate permease (SA2156) and a hypothetical protein similar to NirR (SA2189) which, in S. carnosus, is involved in nitrite reduction and anaerobic respiration.31 Interestingly, a decrease in the expression level of NirR and of lactate-dehydrogenase was also observed in response to bacitracin, D-cycloserine and oxacillin32 as well as a human endogenous AMP, the β-defensin hBD3 (V. Sass, U. Pag, G. Bierbaum and H.-G. Sahl, unpublished data), indicating that these responses may be part of a general cell wall stress sensing and repair system. Indeed, further strongly up-regulated genes such as vraRS (two-component regulatory system), SA1702 (conserved hypothetical protein), htrA (heat-shock protein) and prsA (peptidyl-prolyl cis/trans isomerase homologue) and the hypothetical proteins SA1703, SA1712 and SA2221 were also identified upon vancomycin treatment28,29 and are considered important parts of the staphylococcal cell wall stress stimulon.
Further drastic changes were observed with several genes involved in the lipid metabolism supporting the notion that physical membrane perturbation is a major contributor to the killing activity of this peptide. This is also true for the strong up-regulation of several transporters which may be involved in detoxification of the cell. The most prominent examples were SA0192 (hypothetical protein similar to ABC transporter ATP-binding protein) and in particular the ABC transporter VraDE, which we also found highly up-regulated upon defensin stress (V. Sass, U. Pag, G. Bierbaum and H.-G. Sahl, unpublished data).
Another interesting aspect of the transcriptional response is that, under AMP stress, S. aureus SG511 strongly down-regulates the holin proteins CidAB and the autolysin Atl, and correspondingly up-regulates the antiholin LrgA to avoid cellular lysis, which on the phenotypical level seems quite the opposite with S. simulans 22, in which the autolytic enzymes appeared to be activated by the AMPs.23,24
We also compared the transcription profile of cells which had been growing for 22 h (OD600 of 1) in the presence of P19(6|E) aiming at identification of long-term adaption processes and a better understanding of the lifestyle of staphylococci when permanently confronted with amphiphilic AMPs in the host. The overall number of differentially expressed genes was greatly reduced; only a total of 99 genes differed, with 37 genes up-regulated and 62 genes down-regulated. Apparently, the cells reached a new equilibrium in which putative sugar uptake systems such as msmX, SA0209 and others, the NADH dehydrogenase SA0210 and possibly corresponding regulatory elements (SA1339 and SA1805) have considerably higher transcription levels. Obviously, AMP stress causes elevated dissimilatory carbon metabolism in compensation for reduced efficiency of the respiratory chain.
Among the long-term down-regulated genes, only bioAB were above the 3-fold change threshold; BioAB are involved in biotin synthesis, which is an essential cofactor for the acetyl CoA carboxylase-catalysed malonyl CoA formation, the rate-limiting step for fatty acid biosynthesis. Apparently, lipid synthesis and membrane turnover need to be differently controlled under AMP stress, and regulation occurs on the coenzyme level rather than on the carbon substrate level, which may be elevated as indicated above.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsSupplementary dataReferences |
|---|
The series of helical, cationic and amphiphilic peptides described here can be considered as prototypical for natural AMPs of this type, which are important in the host defence of both vertebrate and invertebrate animals, including mammals. They show remarkable broad-spectrum activity against both widely used laboratory strains and multiresistant clinical isolates. Importantly, any pre-existing resistance mechanism does not seem to provide cross-resistance against these AMPs, indicating that on the molecular level their activity differs completely from those of conventional antibiotics. It is generally assumed that cationic amphiphiles physically impair membranes and form pores; several models have been proposed to describe such activities (for review, see Shai33). Our results clearly show that the cytoplasmic membrane is the primary target of the AMPs studied here; however, we were not able to demonstrate convincingly that pore formation takes place and that membrane lesions represent the primary cause of cell death. Rapid efflux of metabolites and ions and concomitant membrane depolarization, as described e.g. for the pore-forming lantibiotic nisin,18,34 was only observed to some extent with the standard laboratory strain S. simulans 22 in which cell wall hydrolysis clearly adds to rapid killing. In clinical isolates, non-selective inhibition of all biosynthesis pathways argues for membrane impairment, but solute efflux and membrane depolarization patterns were not consistent with pore formation.
We found the analysis of transcriptional responses of sublethally treated cells rather informative for a better understanding of the molecular mechanisms underlying the activities of cationic AMPs. The overall profile obtained with peptide P19(6|E) was not comparable with profiles published for other classes of antibiotic30,32,35 with control antibiotics in our own data file (V. Sass, U. Pag, G. Bierbaum and H.-G. Sahl, unpublished data) or with the human defence peptide hBD3 (V. Sass, U. Pag, G. Bierbaum and H.-G. Sahl, unpublished data). The sheer number of genes affected and the prominent changes in some pathways suggest that the AMPs studied here simultaneously produce multiple stresses at the cytoplasmic membrane. Particularly, aerobic energy generation seemed to be affected and it remains to be studied whether physical impairment of the membrane provokes the dysfunctions or whether there is a more specific action on a defined component of the respiratory chain as was described for the activity of the cationic peptide P2 against Salmonella.36 However, since cell wall biosynthesis and lipid metabolism are strongly affected also, it appears more likely that the peptides, rather than forming defined pores, insert into the membrane or the membrane interface and interfere with the dynamic and coordinated function of multienzyme complexes. For example, electron transport chains, complex nutrient and ion transport systems and the cell wall biosynthesis machinery require highly organized interactions of multiple proteins, coenzymes and substrates which need to be coordinated over time and space. The physical presence of cationic amphiphiles in high concentrations at crucial sites in or at the membrane and low-affinity interactions with various proteins, and other membrane components could disorganize functional processes and produce the complex cellular response pattern observed; in more descriptive terms, amphiphilic AMPs may act like ‘sand in a gearbox’ rather than through pore formation. Such activities could also explain the considerable variation of the AMP effects in various test organisms. It is conceivable that structural differences between the membrane-associated multienzyme machineries in various organisms as well as structural differences in the AMPs itself could cause diverse pathways to be affected to a different extent, yielding somewhat differing molecular activity patterns without changing too much the overall picture of the antibiotic activity. Such rather ‘unspecific and untargeted’ antibiotic activities could also provide an explanation for the fact that cationic AMPs do not select for high-level resistance and thus have remained effective for billions of years.37
|
Funding |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsSupplementary dataReferences |
|---|
Funding was provided by: German Research Foundation (DFG Sa 292/10-2); the BONFOR programme of the Medical Faculty, University of Bonn; the Friuli-Venezia-Giulia Region grant 200502027001; and the Commission of the European Communities (grant QLK2-CT-2000-00411 within the Fifth Framework Programme and INTAS grant 03-51-4984).
|
Transparency declarations |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsSupplementary dataReferences |
|---|
None to declare.
|
Supplementary data |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsSupplementary dataReferences |
|---|
Tables S1, S2 and S3 are available as Supplementary data at JAC Online (http://jac.oxfordjournals.org/).
|
Acknowledgements |
|---|
Members of the research group of H.-G. S. gratefully acknowledge support by the German Research Foundation (DFG Sa 292/10-2) and by the BONFOR programme of the Medical Faculty, University of Bonn. Members of the A. T. group gratefully acknowledge support by the Friuli-Venezia-Giulia Region grant 200502027001. We would also like to acknowledge that this work was also supported by the Commission of the European Communities (grant QLK2-CT-2000-00411 within the Fifth Framework Programme to H.-G. S., A. T. and Y. S. and INTAS grant 03-51-4984 to H.-G. S. and O. S.).
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsSupplementary dataReferences |
|---|
1 Hancock RE, Diamond G. The role of cationic antimicrobial peptides in innate host defences. Trends Microbiol (2000) 8:402–10.[CrossRef][Web of Science][Medline]
2 Lehrer RI, Ganz T. Cathelicidins: a family of endogenous antimicrobial peptides. Curr Opin Hematol (2002) 9:18–22.[CrossRef][Web of Science][Medline]
3 Hancock RE, Sahl HG. Antimicrobial and host-defense peptides as new anti-infective therapeutic strategies. Nat Biotechnol (2006) 24:1551–7.[CrossRef][Web of Science][Medline]
4
Zasloff M. Magainins, a class of antimicrobial peptides from Xenopus skin: isolation, characterization of two active forms, and partial cDNA sequence of a precursor. Proc Natl Acad Sci USA (1987) 84:5449–53.
5 Hancock RE, Lehrer R. Cationic peptides: a new source of antibiotics. Trends Biotechnol (1998) 16:82–8.[CrossRef][Web of Science][Medline]
6 Lehrer RI, Ganz T. Antimicrobial peptides in mammalian and insect host defence. Curr Opin Immunol (1999) 11:23–7.[CrossRef][Web of Science][Medline]
7 Scott MG, Dullaghan E, Mookherjee N, et al. An anti-infective peptide that selectively modulates the innate immune response. Nat Biotechnol (2007) 25:465–72.[CrossRef][Web of Science][Medline]
8 Boman HG. Peptide antibiotics and their role in innate immunity. Annu Rev Immunol (1995) 13:61–92.[CrossRef][Web of Science][Medline]
9
Tossi A, Sandri L, Giangaspero A. Amphipathic, -helical antimicrobial peptides. Biopolymers (2000) 55:4–30.[CrossRef][Web of Science][Medline]
10
Giangaspero A, Sandri L, Tossi A. Amphipathic helical antimicrobial peptides. Eur J Biochem (2001) 268:5589–600.[Web of Science][Medline]
11
Zelezetsky I, Tossi A. -Helical antimicrobial peptides—using a sequence template to guide structure–activity relationship studies. Biochim Biophys Acta (2006) 1758:1436–49.[Medline]
12
Wiedemann I, Breukink E, van Kraaij C, et al. 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 (2001) 276:1772–9.
13
Zelezetsky I, Pag U, Sahl HG, et al. Tuning the biological properties of amphipathic -helical antimicrobial peptides: rational use of minimal amino acid substitutions. Peptides (2005) 26:2368–76.[CrossRef][Web of Science][Medline]
14 Chen YH, Yang JT, Chau KH. Determination of the helix and beta form of proteins in aqueous solution by circular dichroism. Biochemistry (1974) 13:3350–9.[CrossRef][Web of Science][Medline]
15
Gobbo M, Biondi L, Filira F, et al. Helix induction potential of N-terminal -methyl, -amino acids. Lett Pept Sci (1998) 5:105–7.
16
Giacometti A, Cirioni O, Barchiesi F, et al. In vitro susceptibility tests for cationic peptides: comparison of broth microdilution methods for bacteria that grow aerobically. Antimicrob Agents Chemother (2000) 44:1694–6.
17
Odds FC. Synergy, antagonism, and what the chequerboard puts between them. J Antimicrob Chemother (2003) 52:1.
18
Ruhr E, Sahl HG. 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 (1985) 27:841–5.
19
Tusher VG, Tibshirani R, Chu G. Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci USA (2001) 98:5116–21.
20 Chakrabartty A, Kortemme T, Padmanabhan S, et al. Aromatic side-chain contribution to far-ultraviolet circular dichroism of helical peptides and its effect on measurement of helix propensities. Biochemistry (1993) 32:5560–5.[CrossRef][Web of Science][Medline]
21
Pag U, Oedenkoven M, Papo N, et al. In vitro activity and mode of action of diastereomeric antimicrobial peptides against bacterial clinical isolates. J Antimicrob Chemother (2004) 53:230–9.
22 Bierbaum G, Sahl HG. Induction of autolysis of staphylococci by the basic peptide antibiotics Pep 5 and nisin and their influence on the activity of autolytic enzymes. Arch Microbiol (1985) 141:249–54.[CrossRef][Web of Science][Medline]
23
Bierbaum G, Sahl HG. Autolytic system of Staphylococcus simulans 22: influence of cationic peptides on activity of N-acetylmuramoyl-L-alanine amidase. J Bacteriol (1987) 169:5452–8.
24
Sahl HG, Pag U, Bonness S, et al. Mammalian defensins: structures and mechanism of antibiotic activity. J Leukoc Biol (2005) 77:466–75.
25 Brown MH, Skurray RA. Staphylococcal multidrug efflux protein QacA. J Mol Microbiol Biotechnol (2001) 3:163–70.[Web of Science][Medline]
26
Paulsen IT, Brown MH, Littlejohn TG, et al. Multidrug resistance proteins QacA and QacB from Staphylococcus aureus: membrane topology and identification of residues involved in substrate specificity. Proc Natl Acad Sci USA (1996) 93:3630–5.
27 Mascher T, Margulis NG, Wang T, et al. Cell wall stress responses in Bacillus subtilis: the regulatory network of the bacitracin stimulon. Mol Microbiol (2003) 50:1591–604.[CrossRef][Web of Science][Medline]
28
McAleese F, Wu SW, Sieradzki K, et al. Overexpression of genes of the cell wall stimulon in clinical isolates of Staphylococcus aureus exhibiting vancomycin-intermediate-S. aureus-type resistance to vancomycin. J Bacteriol (2006) 188:1120–33.
29 Kuroda M, Kuroda H, Oshima T, et al. Two-component system VraSR positively modulates the regulation of cell-wall biosynthesis pathway in Staphylococcus aureus. Mol Microbiol (2003) 49:807–21.[CrossRef][Web of Science][Medline]
30
Bandow JE, Brotz H, Leichert LI, et al. Proteomic approach to understanding antibiotic action. Antimicrob Agents Chemother (2003) 47:948–55.
31
Neubauer H, Pantel I, Gotz F. Molecular characterization of the nitrite-reducing system of Staphylococcus carnosus. J Bacteriol (1999) 181:1481–8.
32
Utaida S, Dunman PM, Macapagal D, et al. Genome-wide transcriptional profiling of the response of Staphylococcus aureus to cell-wall-active antibiotics reveals a cell-wall-stress stimulon. Microbiology (2003) 149:2719–32.
33 Shai Y. Mode of action of membrane active antimicrobial peptides. Biopolymers (2002) 66:236–48.[CrossRef][Web of Science][Medline]
34 Brotz H, Josten M, Wiedemann I, et al. Role of lipid-bound peptidoglycan precursors in the formation of pores by nisin, epidermin and other lantibiotics. Mol Microbiol (1998) 30:317–27.[CrossRef][Web of Science][Medline]
35
Cirz RT, Jones MB, Gingles NA, et al. Complete and SOS-mediated response of Staphylococcus aureus to the antibiotic ciprofloxacin. J Bacteriol (2007) 189:531–9.
36 Barker HC, Kinsella N, Jaspe A, et al. Formate protects stationary-phase Escherichia coli and Salmonella cells from killing by a cationic antimicrobial peptide. Mol Microbiol (2000) 35:1518–29.[CrossRef][Web of Science][Medline]
37 Peschel A, Sahl HG. The co-evolution of host cationic antimicrobial peptides and microbial resistance. Nat Rev Microbiol (2006) 4:529–36.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. Q. Do, S. Moshkani, P. Castillo, S. Anunta, A. Pogosyan, A. Cheung, B. Marbois, K. F. Faull, W. Ernst, S. M. Chiang, et al. Lipids Including Cholesteryl Linoleate and Cholesteryl Arachidonate Contribute to the Inherent Antibacterial Activity of Human Nasal Fluid J. Immunol., September 15, 2008; 181(6): 4177 - 4187. [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 Appl. Envir. Microbiol., June 15, 2008; 74(12): 3764 - 3773. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||