Synthesis, characterization, antimicrobial activity and LPS-interaction properties of SB041, a novel dendrimeric peptide with antimicrobial properties

Peptides 31 (2010) 1459–1467

Contents lists available at ScienceDirect

Peptides journal homepage: www.elsevier.com/locate/peptides

Synthesis, characterization, antimicrobial activity and LPS-interaction properties of SB041, a novel dendrimeric peptide with antimicrobial properties Michela Bruschi a,e , Giovanna Pirri a , Andrea Giuliani a , Silvia Fabiole Nicoletto a , Izabela Baster b , Mariano Andrea Scorciapino c,d , Mariano Casu c , Andrea C. Rinaldi e,∗ a

Research & Development Unit, Spider Biotech S.r.l., I-10010 Colleretto Giacosa (TO), Italy Department of Microbiology, Jagiellonian University, Cracow, Poland c Department of Chemical Sciences, University of Cagliari, I-09042 Monserrato (CA), Italy d CNR/INFM SLACS (Sardinian LAboratory for Computational materials Science), I-09042 Monserrato (CA), Italy e Department of Biomedical Sciences and Technologies, University of Cagliari, I-09042 Monserrato (CA), Italy b

a r t i c l e

i n f o

Article history: Received 1 March 2010 Received in revised form 26 April 2010 Accepted 26 April 2010 Available online 10 May 2010 Keywords: Peptidomimetics Synthetic peptides LPS Endotoxin Binding Neutralization Sepsis NMR

a b s t r a c t Multimeric peptides offer several advantages with respect to their monomeric counterparts, as increased activity and greater stability to peptidases and proteases. SB041 is a novel antimicrobial peptide with dendrimeric structure; it is a tetramer of pyrEKKIRVRLSA linked by a lysine core, with an amino valeric acid chain. Here, we report on its synthesis, NMR characterization, antimicrobial activity, and LPS-interaction properties. The peptide was especially active against Gram-negative strains, with a potency comparable (on molar basis) to that of lipopeptides colistin and polymixin B, but it also displayed some activity against selected Gram-positive strains. Following these indications, we investigated the efficacy of SB041 in binding Escherichia coli and Pseudomonas aeruginosa LPS in vitro and counteracting its biological effects in RAW-BlueTM cells, derived from RAW 264.7 macrophages. SB041 strongly bound purified LPS, especially that of E. coli, as proved by fluorescent displacement assay, and readily penetrated into LPS monolayers. However, the killing activity of SB041 against E. coli was not inhibited by increasing concentrations of LPS added to the medium. Checking the SB041 effect on LPS-induced activation of pattern recognition receptors (PRRs) in Raw-Blue cells revealed that while the peptide gave a statistically significant decrease in PRRs stimulation when RAW-Blue cells were challenged with P. aeruginosa LPS, the same was not seen when E. coli LPS was used to activate innate immune defense-like responses. Thus, as previously seen for other antimicrobial peptides, also for SB041 binding to LPS did not translate necessarily into LPS-neutralizing activity, suggesting that SB041–LPS interactions must be of complex nature. © 2010 Elsevier Inc. All rights reserved.

1. Introduction Antimicrobial peptides (AMPs; also known as host defense peptides) are widespread components of the innate immune systems of virtually all multicellular organisms, which they protect from invasion by pathogenic microorganisms [15,26]. Although highly diverse in sequence and structure, AMPs do share some basic features, such as an overall positive charge and an amphipathic fold. They are generally believed to kill microbes through binding to and permeabilization of cell membranes [20], although the potential for some AMPs to hit intracellular targets has been recently recognized [4]. Given their natural role and evolutionary success, AMPs have been long regarded as obvious candidates to be developed as

∗ Corresponding author. Tel.: +39 070 6754521; fax: +39 070 6754527. E-mail address: [email protected] (A.C. Rinaldi). 0196-9781/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.peptides.2010.04.022

the new, much-sought-after anti-infectives to contrast growing bacterial resistance to conventional antibiotics [14,34]. Indeed, AMPs offer an intriguing and maybe unique combination of widespectrum antimicrobial, immunostimulatory, immunomodulatory, and anti-inflammatory properties [8,16]. With this perspective, an extensive amount of effort has been devoted during at least two decades to understanding the mechanism of action of AMPs, their exact role in host defense (including interaction with adaptive immunity in humans and higher vertebrates), and the possible manners their therapeutic potential could be harnessed. This intense research activity, however, has not yet translated into any significant practical outcome, since no AMP has completed its journey from the lab to the pharmacy or hospital drug shelve so far. Notwithstanding the hurdles encountered, information coming from basic investigation on the features and properties of many AMPs continues to accumulate at an increasing pace, and several compounds are currently under clinical development as treatments for a variety of infection-related conditions.

1460

M. Bruschi et al. / Peptides 31 (2010) 1459–1467

Fig. 1. Primary sequence and structure of the dendrimeric (tetrameric) antimicrobial peptide SB041. 5-Ava = 5-amino valeric acid; pyrE = pyroglutamic acid. All amino acids have L-configuration. The peptide is amidated at the C-terminus (5-Ava).

Different strategies are available to transform AMPs in therapeutically valuable anti-infective agents. The first, and more obvious, is to directly use naturally occurring molecules, or sequence analogs [e.g., 7]. However, due to their peptidic essence, AMPs found in nature suffer from poor bioavailability and poor proteolytic stability, two features that have severely hampered clinical progress to date. An alternative, which is being selected in an expanding number of instances, is to develop semi-synthetic or synthetic analogs that mimic the main physico-chemical and membrane activity-related properties, including positive charge and amphiphilic nature, of AMPs [25,29]. The advantages offered by this avenue include – but are not limited to – the greater plasticity of designed peptides versus natural counterparts, with the possibility, for example, to insert non-standard, D, or even ␤-amino acids into the sequence. In the present study we report on the synthesis, nuclear magnetic resonance (NMR) characterization, antimicrobial activity and LPS interactions of a novel synthetic peptide characterized by a multimeric (tetrameric) scaffold, named SB041 (Fig. 1). As a distinctive feature, SB041 is endowed with a short lipophilic chain (C5) so as to increase its membrane affinity. 2. Materials and methods 2.1. Peptide synthesis Amino acids and a NovaPEG Rink Amide resin (0.67 mmol/g) were purchased from Sigma–Aldrich–Fluka (St. Louis, MO) and Novabiochem (Merck Chemicals Ltd., Nottingham, UK). Peptide synthesis grade N,N-dimethylformamide (DMF), N-methylpyrrolidone (NMP), trifluoroacetic acid (TFA), dichloromethane, diethyl ether and O-(Benzotriazol-1-yl)N,N,N ,N -tetramethyluronium hexafluorophosphate (HBTU) were purchased from ChemImpex (Wood Dale, IL) and Sigma–Aldrich. All other reagents and solvents were purchased from Sigma–Aldrich at the highest available purity and were used with no further purification The multimeric peptide SB041 was synthesized as tetrameric amide peptide by manual standard solid-phase peptide Fmoc (9fluorenylmethyoxy-carbonyl) strategy, working under nitrogen flow. Coupling reactions with Fmoc amino acids were activated in situ using HBTU, 1-hydroxybenzotriazole (HOBt) and diisopropylethylamine (DIPEA) at the following ratio: HOBt/DIPEA/HBTU 1/2/0.9. The tetrameric lysine core was synthesized on the resin by using (Fmoc)2 Lys–OH protected amino acid, and the first amino acid on the core was 5-amino valeric acid (5-Ava). A sixfold excess of Fmoc amino acid was employed during every coupling synthesis step and amino acid residues with reactive side

chains were used protected with the following acid-labile protecting groups: 2,2,4,6,7-pentamethyldihydro-benzofuran-5-sulfonyl for arginine; tert-butyl ether for serine; tert-butyloxycarbonyl for lysine. Fmoc group was removed by using 20% piperidine in NMP. The other protecting groups were removed during cleavage of the peptide from the solid support upon treatment with a TFA/triisopropylsilane/H2 O solution at a 95/2.5/2.5 ratio for 2 h. After cleavage, the solid support was removed by filtration, and the filtrate was concentrated under reduced pressure. The crude peptides were precipitated from diethyl ether, washed several times with diethyl ether and dried under reduced pressure. RP-HPLC peptide analysis was performed on a Jupiter Proteo analytical C12 column (4.6 mm × 250 mm) supplied by Phenomenex (Torrance, CA), using 0.1% TFA/H2 O as solvent A, and 0.1% TFA/MeCN as solvent B. Column was equilibrated with A/B ratio of 95/5 at a flow rate of 1.0 ml/min, and the concentration of B was raised to 95% (v/v) over 14 min using gradient mode conditions. Peptide was purified on a Jupiter Proteo semipreparative C12 column (10 mm × 250 mm) and the major peak in the chromatogram was collected by an automatic faction collector. Peptide dendrimer was obtained with final purity around 95%. The monoisotopic molecular mass of peptide dendrimer was determined by MALDI-TOF MS (Bruker Daltonik, Bremen, Germany), by using sinapinic acid as acidic matrix. The instruments were calibrated with peptides of known molecular mass in the 1000–6000 Da range. 2.2. NMR characterization SB041 was dissolved at 1 mM concentration either in D2 O (99.98% D – Sigma–Aldrich), or in 90% (v/v) deionized H2 O/10% (v/v) D2 O, or in 60% (v/v) deionized H2 O/40% (v/v) CF3 CD2 OD (99.5% D – Sigma–Aldrich). In all cases, final pH resulted between 6 and 7 at 298 K (pH-meter reading). NMR spectra were recorded at 298 K in 5 mm tube on a Varian (Palo Alto, CA) Unity-Inova spectrometer at a proton resonance frequency of 399.948 MHz. Chemical shifts were quoted relative to the water–protons resonance or relative to CH2 protons of TFE with respect to TMS. 1 H spectra were recorded using 6.5 ␮s pulse (90◦ ), 1 s delay time, 2 s acquisition time, a spectral width of 5 kHz, and 1024 scans. Magnitude correlation spectroscopy (COSY) [17] spectra were acquired over the same spectral window using 2048 complex points and sampling each of the 256 increments with 256 scans. Phase-sensitive total correlation spectroscopy (TOCSY) [17] spectra were collected using the same parameter as COSY, with 128 scans. The spin lock time was 50 ms using the MLEV-17 mixing scheme. Suppression of the intense water signal was always achieved by direct saturation during the relaxation delay. 2.3. Antimicrobial assays Amikacin sulfate, amphotericin B, ciprofloxacin, colistin sulfate, erythromycin, ethambutol, 5-fluorocytosine, gentamicin, polymyxin B sulfate, vancomycin HCl, used as control, were purchased from Sigma–Aldrich. All compounds were dissolved in DMSO (Becton Dickinson, Franklin Lakes, NJ), distilled water, or Na-phosphate buffer pH 6.0 according to CLSI guidelines (formerly NCCLS) [21], to obtain stock solutions of 10 mg/ml. All compounds were subsequently diluted in Müller Hinton Broth (MHB; Difco Laboratories, Sparks, MD), Cation-adjusted MHB, RPMI-1640 (Sigma–Aldrich), or 7H9 (Becton Dickinson) medium to obtain working solutions. All strains used in the present study belong to NeED Pharmaceuticals S.r.l. strain collection. All clinical isolates showed antibacterial resistance phenotypes and proved resistant to various antibacterial

M. Bruschi et al. / Peptides 31 (2010) 1459–1467

agents of common use in nosocomial institutions. Stock cultures of Gram-positive and Gram-negative bacteria, Candida spp. and Mycobacterium smegmatis mc2 155 were prepared from isolated colonies selected on Müller Hinton Agar (MHA), Sabouraud (both from Difco Laboratories) or 7H11 (Becton Dickinson) agar plates, respectively, and diluted into MHB/Sabouraud/7H9 medium to 0.2 O.D.625 , rapidly frozen, and stored at −80 ◦ C. MIC assays were performed by broth microdilution methodology in sterile 96-well microtiter plates (Greiner Bio-One, Monroe, NC), according to CLSI procedures [6,22,23]. Microorganisms were added at final concentration of 1 × 104 CFU/ml for Candida spp., 1–5 × 106 CFU/ml for M. smegmatis mc2 155, and 1–5 × 105 CFU/ml for Gram-negative and Gram-positive bacteria, respectively. Plates were incubated at 37 ◦ C and read after 20–24 h for Gram-negative and Gram-positive bacteria, 48 h for Candida spp., and 72 h for M. smegmatis mc2 155, respectively. MIC was defined as the lowest drug concentration causing complete suppression of visible bacterial growth. 2.4. Measurement of penetration into LPS monolayers and of LPS-binding activity Insertion of SB041 into monolayers formed of LPS from Escherichia coli or Pseudomonas aeruginosa was evaluated as an indication of the peptide’s ability to bind endotoxin and penetrate the Gram-negative’s outer membrane. To this end, LPS from either E. coli 0111:B4 and P. aeruginosa 10 (both Sigma–Aldrich) was spread at an air/buffer (5 mM Hepes, pH 7) interface, and penetration was monitored by measuring surface pressure () with a Wilhelmy wire attached to a microbalance (DeltaPi, Kibron Inc., Helsinki) connected to a PC and using circular glass wells (subphase volume 0.5 ml). After evaporation of LPS solvent (chloroform/methanol/water 17/7/1) and stabilization of monolayers at different initial surface pressures (0 ), the peptide (0.1–2 ␮M) was injected into the subphase, and the increment in surface pressure of the LPS film upon intercalation of the peptide dissolved in the subphase was followed for the next 35 min. The difference between the initial surface pressure and the value observed after the penetration of SB041 into the film was taken as . Measurement of the SB041 ability to bind LPS was also performed by a fluorescent displacement assay using the probe BODIPY TR cadaverine (BC; Molecular Probes, Eugene, OR) as described elsewhere [32]. All measurements were performed at room temperature, using colistin sulfate as benchmark reference compound. 2.5. Time-kill assay in the presence of LPS To evaluate whether endotoxin could influence the peptide’s antimicrobial activity, the antimicrobial activity of SB041 against E. coli ATCC 25922 has been measured also in the presence of LPS, following a procedure reported elsewhere [11]. A stock solution of SB041 (at a final concentration of 16 ␮g/ml, four fold the MIC value for this strain) was pre-incubated with a solution of LPS from E. coli O111:B4 (at a final concentration of 0.5 or 1.5 ␮g/ml) for 30 min at 37 ◦ C. These solutions were subsequently added to exponentially growing cultures of E. coli (∼6.4 × 107 CFU/ml) in MHB medium and incubated at 37 ◦ C. SB041 alone at the same final concentration was used as positive control. After 0, 10, 30 and 50 min, 0.1 ml of cultures were withdrawn, appropriately diluted in MHB medium and spread on MHA plates. Colonies were counted after 24 h of incubation at 37 ◦ C. Colistin sulfate was used as reference compound at a final concentration of 2 ␮g/ml (fourfold the MIC for this strain), either alone or incubated with a solution of LPS at a final concentration of 1 ␮g/ml.

1461

2.6. Neutralization of LPS-mediated biological activities: RAW-Blue cells assay RAW-BlueTM cells, derived from RAW 264.7 macrophages, were from InvivoGen (San Diego, CA). They stably express a secreted embryonic alkaline phosphatase (SEAP) gene inducible by NF-␬B and AP-1 transcription factors. DMEM, 4.5 g/l glucose, 10% heatinactivated fetal bovine serum was used as cell culture medium, with ZeocinTM (InvitroGen, Carlsbad, CA) as selectable marker. LPS from P. aeruginosa 10 and from E. coli 0111:B4 were used in the assay, performed as follows. Briefly, a cell suspension of 550000 cells/ml in fresh growth medium was prepared and 180 ␮l (∼100,000 cells) were added to each well of a Falcon flat-bottom 96-well plate (Becton Dickinson). After different incubation times (see below the different conditions tested), 10 ␮l of a 20 ␮g/ml LPS solution (at a final concentration of 1 ␮g/ml) was added to each well together with 10 ␮l of a SB041 solution (at a final concentration of 0.1, 1 and 10 ␮g/ml). Three different conditions were tested. Pre-treatment: SB041 was pre-incubated with RAW-Blue cells 30 min before LPS addition; post-treatment: LPS was preincubated with RAW-Blue cells 30 min before peptide addition; co-incubation: SB041 and LPS were co-treatment with RAW-Blue cells. SB041 alone was also added at the same final concentrations to evaluate its effect on the basal level of induction; the positive (LPS) and the negative (endotoxin-free water) controls were also included. Finally, colistin sulfate (at a final concentration of 0.1 and 1 ␮g/ml), co-incubated with LPS (1 ␮g/ml), was used as reference compound. The plates were incubated at 37 ◦ C in a 5% CO2 incubator for 18–24 h. Thereafter, a QUANTI-BlueTM (InvivoGen) solution – a detection medium developed to determine the activity of any alkaline phosphatase present in a biological sample – was prepared following the manufacturer instructions: 180 ␮l of resuspended QUANTI-Blue were added to each well of a flat-bottom 96-well plate, followed by 20 ␮l of induced RAW-Blue cells supernatant. The plate was incubated for 30 min to 3 h at 37 ◦ C and the SEAP levels were determined using a spectrophotometer (Biorad, Benchmark Plus, UK) at 620–655 nm. Each experiment was performed in three replicates and data were analyzed by one-way analysis of variance, ANOVA, using GraphPadPrism (GraphPad Software, La Jolla, CA).

3. Results 3.1. SB041 synthesis and NMR characterization SB041 peptide dendrimer (Fig. 1) was obtained with a final purity around 95%, as checked by reverse phase HPLC. SB041 molecular mass found by MALDI-TOF analysis was (M+H) 5155.2 m/z, while calculated value was 5154.5 Da. Fig. 2 shows 1D 1 H NMR spectra (0.5–3.5 ppm) of the SB041 peptide dissolved either in H2 O/D2 O (90/10% (v/v); Fig. 2A) or in TFE/H2 O (40/60% (v/v); Fig. 2B). Peptide backbone 1 HN signals were not observed in water. This was not really surprising since labile protons were expected to undergo rapid exchange with solvent under our experimental conditions. This is typical for small and unfolded peptides since they are characterized by a relatively fast proton exchange (∼103 min−1 at pH 6.5) between the backbone amide groups and water solvent [5]. The resonances of peptide were assigned on the basis of the corresponding 2D COSY and TOCSY spectra (Table 1S). The integrated areas of all the signals in water spectra resulted to be in agreement with the expected peptide composition. Experimental chemical shifts were found to be very close to the literature ones for the single amino acids dissolved in aqueous media [2,33]. The lack of chemical shift dispersion and the considerable chemical-shift degeneracy for the same residue type, suggest that SB041 is unstructured (random coil) in water. As stated above, even if an accurate structural

1462

M. Bruschi et al. / Peptides 31 (2010) 1459–1467

istin was not active against all Gram-positive strains tested, while polymyxin B showed marginal activity only against S. epidermidis (32 ␮g/ml). Limited activity of SB041 against Mycobacterium smegmatis mc2 155 (64 ␮g/ml) and no activity against Candida spp. were finally observed (data not shown). 3.3. Endotoxin-binding properties of SB041

Fig. 2. 1 H NMR spectra of SB041 dissolved in (A) H2 O/D2 O (90/10%, v/v) and (B) TFE/H2 O (40/60%, v/v) at 298 K in the 0.5–3.5 ppm range.

study was beyond the scope of the present paper, peptide spectra were acquired also in TFE/water mixture and some interesting evidence could be derived. By moving to this less polar environment, which is known to mimic the conformational changes induced by biomembranes on small amphiphilic peptides [18,30], we were able to observe and assign 1 HN resonances, indicating a lower protons exchange rate, typically associated with their shielding from the solvent and commonly resulting from hydrogen bonding interactions [5]. However, the spectrum of the peptide dissolved in TFE/H2 O (Fig. 2B) is of poor quality compared to that of the peptide dissolved in H2 O (Fig. 2A). The signals are broader and less resolved, which is indicative of peptide aggregation. Indeed, in TFE/H2 O solutions, the peptide precipitated in the test tube within ∼3 days, confirming that it has a strong tendency to aggregate in such a biomembrane-mimicking environment. 3.2. MIC determination The in vitro activity of SB041 and reference compounds against ATCC strains and an expanded panel of clinical isolates recently collected has been determined (Table 1). All tested strains showed resistance to several currently used antibacterial agents. SB041 showed to be active against six Gram-negative bacteria, namely Acinetobacter baumannii, Enterobacter cloacae, E. coli, Klebsiella pneumoniae, P. aeruginosa and Stenotrophomonas maltophilia, with MICs ranging between 4 and 16 ␮g/ml. Polymyxin B showed antibacterial activity similar to that of colistin, and an overall potency higher than SB041 (Table 1). The activity of SB041 against the Gram-positive Enterococcus faecalis, E. faecium, Staphylococcus epidermidis, and S. aureus is reported in Table 2. The peptide showed a limited activity against enterococci, with MICs in the range 16-128 ␮g/ml, whereas the MIC was 64 ␮g/ml and 8 ␮g/ml against S. aureus and S. epidermidis, respectively. Col-

Monomolecular lipid and LPS films have been widely used as suitable model systems to investigate the interactions of a wide range of peptides and proteins with biological membranes and with components of the bacterial outer membrane. Here, we took advantage of the monolayer technique to get insights into the ability of SB041 to bind LPS and to mimic its interaction with E. coli and P. aeruginosa outer membranes. SB041 efficiently penetrated into E. coli LPS monolayers, as demonstrated by the increase in film surface pressure, whereas the affinity for P. aeruginosa LPS monolayers was significantly lower (Fig. 3A). Analyzing measurements in terms of  versus 0 , the critical surface pressure corresponding to the LPS lateral packing density preventing the intercalation of SB041 into E. coli or P. aeruginosa LPS films could be derived by extrapolating the –0 slope to  = 0, giving a value ≈45 and 35 mN/m, respectively (Fig. 3A). Under experimental conditions, in both cases monolayer penetration was dependent on peptide concentration, although while in the case of P. aeruginosa LPS a substantial stability was reached around 1.0 ␮M SB041, this plateau was not approached when interaction with E. coli LPS was tested (Fig. 3B). Using colistin as a reference peptide, it was confirmed that also this peptide did penetrate E. coli LPS monolayers more efficiently than those formed by P. aeruginosa LPS (Fig. 3C and D). The kinetics of the insertion of SB041 into E. coli and P. aeruginosa LPS monolayers were comparable, characterized by a prolonged lag phase (up to 1000 s in P. aeruginosa LPS, somewhat shorter with E. coli) subsequent to the injection of SB041 into the subphase, followed by a slow but constant increase in surface pressure, which approached (but not always reached) a plateau within 35 min from peptide insertion into the subphase (Fig. 4). This general kinetics pattern was apparently independent from initial surface pressure and from peptide concentration. To further determine the SB041–LPS binding affinity, we used a fluorescent probe displacement method developed by Wood et al. [32]. The fluorescent probe BC binds LPS, interacting specifically with its toxic center lipid A, probably via salt-bridges with its glycosidic phosphate group, and the binding results in a progressive quenching of fluorescence. BC can then be competitively displaced by compounds displaying an affinity for lipid A, with a proportional dequencing of fluorescence. As shown in Fig. 5, SB041 bound to both purified E. coli and P. aeruginosa LPS and induced a displacement of BC with a quantitative effective displacement (ED50 ) corresponding approximately to 1 and 2.4 ␮M, respectively. Thus, data coming from monolayers and fluorescence displacement experiments are

Table 1 Activity of SB041 vs Gram-negative bacteria. Microorganism

Acinetobacter baumannii Enterobacter cloacae Escherichia coli Klebsiella pneumoniae Klebsiella pneumoniae Pseudomonas aeruginosa Pseudomonas aeruginosa Pseudomonas aeruginosa Stenotrophomonas maltophilia a

Source

Clinical isolate Clinical isolate Clinical isolate Clinical isolate Clinical isolate ATCC 27853 Clinical isolate Clinical isolate Clinical isolate

SB041

8 8 4 8 16 8 16 4 8

MIC (␮g/ml) CLa

PB

AMI

CIP

ERY

GEN

1 0.5 0.5 2 1 1 2 0.5 1

1 1 0.5 1 4 1 2 0.5 1

16 1 2 16 >16 2 >16 4 4

>16 >16 ≤0.125 >16 >16 0.25 16 4 2

32 >128 32 >128 >128 128 >128 128 >128

>16 16 1 >16 >16 1 >16 2 2

Abbreviations: AMI, amikacin; CIP, ciprofloxacin; CL, colistin; ERY, erythromycin; GEN, gentamicin; PB, polimixin B.

M. Bruschi et al. / Peptides 31 (2010) 1459–1467

1463

Table 2 Activity of SB041 vs Gram-positive bacteria. Microorganism

Enterococcus faecalis Enterococcus faecalis Enterococcus faecalis Enterococcus faecium Staphylococcus aureus Staphylococcus epidermidis a

Source

ATCC29212 Clinical isolate Clinical isolate Clinical isolate Clinical isolate Clinical isolate

MIC (␮g/ml) SB041

CLa

PB

GEN

VAN

32 64 128 16 64 8

>128 >128 >128 >128 >128 128

>128 >128 >128 >128 128 32

16 >128 8 >128 0.5 32

2 0.5 0.5 0.5 0.25 2

Abbreviations: CL, colistin; GEN, gentamicin; PB, polimixin B; VAN, vancomycin.

mutually reinforcing, indicating a significantly greater affinity of SB041 for E. coli endotoxin.

3.4. Bactericidal activity of SB041 in the presence of LPS To evaluate if LPS could influence in a negative way the bactericidal activity of SB041 – by binding to it and sequestering the peptide – a time-kill assay was performed incubating a solution of SB041 at a concentration four fold the MIC value with LPS from E. coli at two different concentrations (0.5 and 1.5 ␮g/ml). Colistin sulfate was also included as a reference compound. Fig. 6 shows the time-kill curves of SB041 and colistin sulfate against E. coli ATCC 25922 in the presence of LPS. The results obtained seem to demonstrate that LPS from E. coli does not influence the bactericidal activity of SB041 against E. coli ATCC 25922 in a negative way. When incubated with both the concentrations of LPS, the killing activity of SB041 is even higher than the activity of the peptide alone; it seems that the pres-

ence of LPS can accelerate the reduction of the initial inoculum size. On the contrary, the killing activity of colistin sulfate at four fold its MIC value under the conditions tested appears to be reduced in the presence of LPS when compared to the bactericidal activity of colistin alone.

3.5. Raw-Blue cells assay Macrophages are major players in the innate immune defense. They express a large repertoire of different classes of pattern recognition receptors (PRRs), such as the Toll-like receptors (TLRs), RIG-I-like receptors (RLRs) and NOD-like receptors (NLRs). RAWBlue cells are murine macrophages designed for the study of these PRRs. RAW-Blue cells are derived from RAW 264.7 macrophages; they stably express a SEAP gene inducible by NF-␬B and AP-1 transcription factors. RAW-Blue cells are resistant to the selectable marker Zeocin. Upon TLR, RLR or NOD stimulation, RAW-Blue cells

Fig. 3. Insertion of SB041 and colistin into E. coli and P. aeruginosa LPS monolayers. Increments of surface pressure of LPS monolayers due to the addition of 1.0 ␮M SB041 (A) or colistin (C) into the subphase are illustrated as a function of initial surface pressure, or at increasing peptide concentration, with an initial surface pressure varying between 17.5 and 20.0 mN/m (B, SB041; D, colistin).

1464

M. Bruschi et al. / Peptides 31 (2010) 1459–1467

Fig. 4. Typical kinetics of surface pressure increase related to SB041 penetration into E. coli (A; 0 = 15.6, with 1.0 ␮M peptide) and P. aeruginosa LPS monolayers (B; 0 = 10.5, with 1.0 ␮M peptide). Arrow indicates peptide injection into the subphase (at ≈200 s).

Fig. 6. Bactericidal activity of SB041 in the presence of LPS. SB041 (at a final concentration of 16 ␮g/ml) was pre-incubated for 30 min at 37 ◦ C with LPS from E. coli O111:B4 (at a final concentration of 0.5 or 1.5 ␮g/ml), subsequently added to exponentially growing cultures of E. coli ATCC 25922 (∼6.4 × 107 CFU/ml) in MHB medium, incubated at 37 ◦ C for up to 50 min, and colonies counted. SB041 alone at the same final concentration was also included; colistin sulfate was included as reference compound at a final concentration of 2 ␮g/ml, either alone or incubated with a solution of LPS at a final concentration of 1 ␮g/ml. Values are means of three independent replicates; error bars were omitted for clarity. All data are percentages relative to the control.

Fig. 5. BC fluorescent displacement assay. The fluorescent probe BC binds LPS, and the binding results in a progressive quenching of fluorescence [32]. SB041 did bind to E. coli or P. aeruginosa LPS, displacing BC from it and causing a proportional dequencing of BC fluorescence (A). [LPS]: 5 ␮M; [BC]: 5 ␮M. Buffer: 50 mM Tris, pH 7.4. Aliquots of SB041 (at 0.1 ␮M final concentration) were successively added to the cuvette containing the BC:LPS complex, and the fluorescence recorded. Excitation: 580 nm; emission: 620 nm. Colistin was used as a reference compound, following the same experimental protocol (B).

activate NF-␬B and/or AP-1, leading to the secretion of SEAP which is easily detectable and measurable when using QUANTI-Blue. To evaluate the SB041 effect on LPS-induced immune response, the SB041 action on the innate immune defense system was investigated, alone or in combination with LPS from the two different bacterial species (E. coli and P. aeruginosa), using the Raw-Blue cells assay. In the first experiment, SB041 was tested alone and in combination with LPS from P. aeruginosa 10 under three different conditions – pre-treatment, co-treatment and post-treatment – and at three final concentrations (0.1, 1, and 10 ␮g/ml). The results, reported in Fig. 7, show that SB041 in pre-treatment and post-treatment with P. aeruginosa LPS gave a statistically

Fig. 7. PRR stimulation in RAW-Blue macrophages by P. aeruginosa LPS in the presence of SB041 and colistin. SB041 was tested either alone or in the presence of 1 ␮g/ml LPS under three different conditions – pre-treatment, co-treatment and post-treatment – and at three final concentrations (0.1, 1, and 10 ␮g/ml). Colistin sulfate (0.1 or 1 ␮g/ml) was used in co-treatment with LPS. See Section 2 for further details. The reference value (PRR stimulation 100%) was that obtained with LPS alone (at 1 ␮g/ml). Asterisks indicate statistical significance. For clarity, error bars are not reported for peptide when tested alone.

M. Bruschi et al. / Peptides 31 (2010) 1459–1467

Fig. 8. PRR stimulation in RAW-Blue macrophages by E. coli LPS in the presence of SB041 and colistin. SB041 was tested either alone or in the presence of 1 ␮g/ml LPS under three different conditions – pre-treatment, co-treatment and post-treatment – and at three final concentrations (0.1, 1, and 10 ␮g/ml). Colistin sulfate (0.1 or 1 ␮g/ml) was used in co-treatment with LPS. See Section 2 for further details. The reference value (PRR stimulation 100%) was that obtained with LPS alone (at 1 ␮g/ml). Asterisks indicate statistical significance. For clarity, error bars are not reported for peptide when tested alone.

significant decrease in the percentage of TLR, RLR, NOD stimulation; the amount of the decrease approached 30% with SB041 1 ␮g/ml in pre-treatment. Moreover, the inhibition seems not to be concentration-dependent. On the contrary, the reduction in PRRs activity was not significant when SB041 was used in co-treatment with LPS, the reason for which is not clear at this stage (Fig. 7). In the second experiment, SB041was tested alone and in combination with LPS from E. coli O111:B4 under the same different conditions as above: pre-treatment, co-treatment and post-treatment, and at the same final concentrations (Fig. 8). Unexpectedly, in this case SB041 inhibition activity towards PRRs stimulation was less evident than in the previous experiment with LPS from P. aeruginosa, being the SEAP production under all the conditions tested very similar to that obtained by LPS alone, which was considered as positive control (PRR stimulation 100%). This finding might at first be surprising when compared to the results obtained from the competitive binding assay (see above), where the affinity of SB041 for LPS from E. coli was higher than the affinity for LPS from P. aeruginosa. However, being the stimulation of PRRs generally elevated with LPS from both species and under all tested conditions, we can speculate that SB041 is only able to bind LPS with different affinity, but not to efficiently neutralize its biological activity on macrophages. This can be due to the fact that SB041 cannot disaggregate the micellar structures adopted by LPS in solution [14]. As expected, in both experiments SB041 when used alone at all the three concentrations tested did not significantly induce PRRs stimulation (Figs. 7 and 8). The cytotoxicity of SB041 against RAW cells was also investigated using the same cell density and the same peptide concentrations, which permitted to exclude any toxicity of the peptide (data not shown). As expected, colistin sulfate, when used at 1 ␮g/ml in co-treatment with LPS from both bacterial species, significantly reduced the PRRs stimulation (Figs. 7 and 8). This lipopeptide is well known for its LPS-binding and neutralizing activity [14,19]. 4. Discussion SB041 is a novel dendrimeric peptide with a tetra-branched structure, four identical peptides (pyrEKKIRVRLSA) linked by a lysine core, with an amino valeric acid chain. SB041s peptide sequence has been derived through rational modification and optimization of an antimicrobial peptide (QEKIRVRLSA) originally

1465

identified by selecting a random phage library against whole E. coli cells [24]. Generally speaking, dendrimeric peptides display increased activity compared to their monomeric counterparts – a fact which is probably related to the higher local concentration of bioactive units for multimeric peptides – as well as greater stability to peptidases and proteases, possibly due to the steric hindrance of the branching core that would limit the cleavage rates of plasma peptidases, thus increasing their pharmacokinetics properties [9,12,13]. As in the case of other peptides built in dendrimeric form (Multiple Antigen Peptide system) [3,31], also for SB041 a Lys core has been used as synthetic scaffold for the attachment of several copies (four, in the case of SB041) of the monomeric sequence using standard solid-phase chemistry. As innovative characteristics, SB041 incorporates both a lipophilic amino valeric acid chain, aimed to enhance the peptide’s membrane affinity, and a pyroglutamic acid at N-terminal end, aimed to confer more stability, avoiding the well known cyclization process involving a Gln residue at N-terminus. NMR characterization of SB041, besides confirming its composition, has revealed some details of its behavior. In water, the peptide was found to be basically unfolded, while moving to a less polar H2 O/TFE mixture – which can be assumed to roughly mimic a water/membrane interface – the peptide displayed a clear tendency to aggregate. This latter evidence suggests that the same aggregation could indeed occur when the peptide binds to the target’s cell membrane, possibly a preliminary (and maybe mandatory) step before exerting its antimicrobial action. Assessing the in vitro activity of SB041 and reference compounds against an expanded panel of recently isolated clinical strains has revealed that the peptide is mainly active against Gram-negative bacteria, while it retains a limited activity against Gram-positive bacteria and it is virtually inactive against M. smegmatis and Candida spp. Overall, the antimicrobial activity of SB041 for Gram-negative strains is comparable (on a molar basis), although lower, to those of the naturally occurring cyclic lipopeptides polymyxin B and colistin, the latter taken as a reference, benchmark antibiotic compound throughout our study. When compared to the antimicrobial properties and stability of the parent sequence, QEKIRVRLSA, either in its monomeric or tetra-branched form (named L1 and M1, respectively), and to one of its modified tetra-branched analogs with sequence QKKIRVRLSA (M6) [see 24], SB041 displayed both a stronger activity against Gram-negative strains and a greater stability over time once resuspended in solution due to the introduction of pyrE1 (data not shown). Considering the fact that the same trend was recorded before when passing from L1 to M1 and then to M6 [24], this confirms the validity of the approach based on the optimization of the monomeric sequence coupled to the multimeric peptide structure. Given the antimicrobial spectrum displayed by SB041, we considered worthwhile to explore further the interaction of the peptide with LPS, the endotoxin, a major component of the Gram-negative outer membrane which is responsible for the systemic inflammatory response syndrome and related, often fatal disorders that can follow Gram-negative infections. In our study, the direct binding of SB041 to LPS derived either from E. coli or P. aeruginosa has been assayed using two different, yet complementary, approaches. These included the insertion of the peptide into LPS monolayers formed at the air/buffer interface, and a fluorescent displacement assay, in which the peptide competed with BC for binding to LPS. Both these experimental settings revealed that SB041 has a good general affinity for LPS, in some instances comparable or even superior to that of colistin. Moreover, the results coincided in showing a significantly greater affinity of SB041 for the E. coli LPS with respect to the endotoxin extracted from P. aeruginosa. It can be speculated that the lag phase recorded in the monolayer penetration experiments can be linked to the peptide’s tendency for

1466

M. Bruschi et al. / Peptides 31 (2010) 1459–1467

aggregation, as also suggested by NMR observations, which could be preliminary to membrane-perturbing activity. It is to be noted that the insertion kinetics of linear monomeric peptides such as temporin L when their interaction with LPS monolayers was investigated using a similar approach was found to be significantly different [11], suggesting that this technique can offer glimpses into the various behaviors displayed by AMPs belonging to different structural classes when interacting with Gram-negative’s outer membranes. The results obtained with the direct binding assays fit well with data coming from microbiological studies, which clearly showed a stronger activity of SB041 against E. coli with respect to P. aeruginosa. However, when the bactericidal activity of SB041 was tested in the presence of E. coli LPS, SB041 binding to endotoxin did not translate into a peptide’s sequestration, at least not in a manner that prevented the exertion of its antimicrobial activity, contrarily to what seen with colistin. This finding suggests that the SB041–LPS interaction is a complex one. LPS-sensing by macrophages and neutrophils lay at the root of the intricate signaling pathway that drives the early innate and subsequent adaptive antibacterial defenses, but also the inflammatory over reaction that might culminate with septic shock [1,10]. Choosing Raw-Blue cells as innate immune response mimics, we have assayed the ability of SB041 to counteract LPS biological effects. The findings were mixed. Indeed, while on one side the peptide gave a statistically significant decrease in PRRs stimulation when RAW-Blue cells were challenged with P. aeruginosa LPS, the same was not seen when E. coli LPS was used to activate innate immune defense-like responses. Thus, for SB041 strongly binding to LPS (as that from E. coli) did not translate directly into LPS-neutralizing activity. This latter finding is not necessarily surprising. Indeed, previous studies have demonstrated that strong binding of a peptide to LPS is not sufficient to block LPS biological activity, and also that there is no direct correlation between the antimicrobial activity of AMPs and their ability to neutralize LPS [27]. As for the way AMPs can prevent LPS from causing immune over-response, it is generally believed that this may take place by a number of mechanisms. These include direct binding to LPS, in a form that dissociates LPS aggregates, making endotoxin unavailable to LBP (LPS-binding protein); competing with LPS for binding to the TLR signaling complex; inhibiting NF␬B translocation into the nucleus; direct killing of microbes either via disruption of their membranes or by reacting with internal molecules [14,28]. In summary, the combination of results from direct LPS binding, time-kill assay and macrophage stimulation, shows that the binding of SB041 to endotoxin does not necessarily result in either sequestration of LPS for purposes of macrophage activation or of the peptide for purposes of antimicrobial action. The anti-inflammatory activity of cationic peptides may provide a therapeutic advantage for the control of diseases with both bacterial and inflammatory components. Any anti-inflammatory activity of SB041 on host cells could represent an advantage in its therapeutic use for the control of infectious diseases that are accompanied by inflammation, as is the case of several topical skin infections and in wound healing. At the same time, its ability to kill multidrug resistant Gram-negative bacteria (or selected Gram-positive strains) could be exploited to combat life-threatening infections which frequently occur in the hospital setting or in chronic lung infections.

Acknowledgement Microbiological studies were carried out in collaboration with NeED Pharmaceuticals S.r.l. (Milan, Italy).

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.peptides.2010.04.022. References [1] Beutler B, Rietschel ETh. Innate immune sensing and its roots: the story of endotoxin. Nat Rev Immunol 2003;3:169–76. [2] Biomolecules Magnetic Resonance Data Bank. http://www.bmrb.wisc.edu/ 2009. [3] Bracci L, Falciani C, Lelli B, Lozzi L, Runci Y, Pini A, et al. Synthetic peptides in the form of dendrimers become resistant to protease activity. J Biol Chem 2003;278:46590–5. [4] Brogden KA. Antimicrobial peptides: pore formers or metabolic inhibitors in bacteria? Nat Rev Microbiol 2005;3:238–50. [5] Cavanagh J, Fairbrother WJ, Palmer III AG, Rance M, Skelton NJ. Protein NMR spectroscopy. Principles and practice. 2nd ed. Oxford: Elsevier Academic Press; 2007. [6] Committee for Clinical Laboratory Standards. Susceptibility testing of mycobacteria, nocardiae, and other aerobic actinomycetes. Approved Standard. NCCLS document M24-A. Wayne: NCCLS; 2006. [7] Conlon JM, Al-Ghaferi N, Abraham B, Leprince J. Strategies for transformation of naturally-occurring amphibian antimicrobial peptides into therapeutically valuable anti-infective agents. Methods 2007;42:349–57. [8] Easton DM, Nijnik A, Mayer ML, Hancock RE. Potential of immunomodulatory host defense peptides as novel anti-infectives. Trends Biotechnol 2009;27:582–90. [9] Falciani C, Lozzi L, Pini A, Corti F, Fabbrini M, Bernini A, et al. Molecular basis of branched peptides resistance to enzyme proteolysis. Chem Biol Drug Des 2007;69:216–21. [10] Freudenberg MA, Tchaptchet S, Keck S, Fejer G, Huber M, Schültze N, et al. Lipopolysaccharide sensing an important factor in the innate immune response to Gram-negative bacterial infections: benefits and hazards of LPS hypersensitivity. Immunobiology 2008;213:193–203. [11] Giacometti A, Cirioni O, Ghiselli R, Mocchegiani F, Orlando F, Silvestri C, et al. Interaction of temporin L with lipopolysaccharide in vitro and in experimental rat models of septic shock caused by gram-negative bacteria. Antimicrob Agents Chemother 2006;50:2478–86. [12] Giuliani A, Pirri G, Bozzi A, Di Giulio A, Aschi M, Rinaldi AC. Antimicrobial peptides: natural templates for synthetic membrane-active compounds. Cell Mol Life Sci 2008;65:2450–60. [13] Giuliani A, Pirri G, Nicoletto SF. Antimicrobial peptides: an overview of a promising class of therapeutics. Centr Eur J Biol 2007;2:1–33. [14] Giuliani A, Pirri G, Rinaldi AC. Antimicrobial peptides: the LPS connection. Methods Mol Biol 2010;618:137–54. [15] Giuliani A, Rinaldi AC. Antimicrobial peptides. Methods and protocols. Methods in molecular biology, vol. 618. New York: Humana Press; 2010. [16] Hancock REW, Sahl H-G. Antimicrobial and host-defense peptides as new antiinfective therapeutic strategies. Nat Biotechnol 2006;24:1551–7. [17] Keeler J. Understanding NMR spectroscopy. Chichester: Wiley & Sons; 2005. [18] Lancelin JM, Gans P, Bouchayer E, Bally I, Arlaud GJ, Jacquot JP. NMR structures of a mitochondrial transit peptide from the green alga Chlamydomonas reinhardtii. FEBS Lett 1996;391:203–8. [19] Li J, Nation RL, Milne RW, Turnidge JD, Coulthard K. Evaluation of colistin as an agent against multi-resistant Gram-negative bacteria. Int J Antimicrob Agents 2005;25:11–25. [20] Melo MN, Ferre R, Castanho MA. Antimicrobial peptides: linking partition, activity and high membrane-bound concentration. Nat Rev Microbiol 2009;7:245–50. [21] National Committee for Clinical Laboratory Standards. Performance standards for antimicrobial susceptibility testing. Approved standard. 16th Informational Supplement. NCCLS document M100-S16. Wayne: NCCLS; 2006. [22] National Committee for Clinical Laboratory Standards. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. Approved standard. 7th ed. NCCLS document M7-A7. Wayne: NCCLS; 2006. [23] National Committee for Clinical Laboratory Standards. Reference methods for broth dilution antifungal susceptibility testing of yeasts. Approved standard. 2nd ed. NCCLS document M27-A2. Wayne: NCCLS; 2006. [24] Pini A, Giuliani A, Falciani C, Runci Y, Ricci C, Lelli B, et al. Antimicrobial activity of novel dendrimeric peptides obtained by phage display selection and rational modification. Antimicrob Agents Chemother 2005;49:2665–72. [25] Pirri G, Giuliani A, Nicoletto SF, Pizzuto L, Rinaldi AC. Lipopeptides as antiinfectives: a practical perspective. Centr Eur J Biol 2009;4:258–73. [26] Radek K, Gallo G. Antimicrobial peptides: natural effectors of the innate immune system. Semin Immunopathol 2007;29:27–43. [27] Rosenfeld Y, Papo N, Shai Y. Endotoxin (lipopolysaccharide) neutralization by innate immunity host-defense peptides: peptide properties and plausible modes of action. J Biol Chem 2006;281:1636–43. [28] Rosenfeld Y, Shai Y. Lipopolysaccharide (endotoxin)-host defense antibacterial peptides interactions: role in bacterial resistance and prevention of sepsis. Biochim Biophys Acta 2006;1758:1513–22. [29] Scott RW, DeGrado WF, Tew GN. De novo designed synthetic mimics of antimicrobial peptides. Curr Opin Biotechnol 2008;19:620–7.

M. Bruschi et al. / Peptides 31 (2010) 1459–1467 [30] Sonnichsen FD, Van Eyk JE, Hodges RS, Sykes BD. Effect of trifluoroethanol on protein secondary structure: an NMR and CD study using a synthetic actin peptide. Biochemistry 1992;31:8790–8. [31] Tam JP, Lu YA, Yang JL. Antimicrobial dendrimeric peptides. Eur J Biochem 2002;269:923–32. [32] Wood SJ, Miller KA, David SA. Anti-endotoxin agents. 1. Development of a fluorescent probe displacement method optimized for the rapid identification

1467

of lipopolysaccharide-binding agents. Combin Chem High Throughput Screen 2004;7:239–49. [33] Wüthrich K. NMR of proteins and nucleic acids. Chichester: Wiley & Sons; 1986. [34] Zhang L, Falla TJ. Host defense peptides for use as potential therapeutics. Curr Opin Investig Drugs 2009;10:164–71.