Biochimica et Biophysica Acta 1830 (2013) 3427–3436

Contents lists available at SciVerse ScienceDirect

Biochimica et Biophysica Acta

j ourna l homepage: www.e lsev ie r .com/ locate /bbagen

Enhanced antimicrobial activity of novel synthetic peptides derivedfrom vejovine and hadrurin☆

Lorenzo Sánchez-Vásquez a, Jesus Silva-Sanchez b, Juana Maria Jiménez-Vargas a, Adela Rodríguez-Romero c,Carlos Muñoz-Garay a, Maria C. Rodríguez b, Georgina B. Gurrola a, Lourival D. Possani a,⁎a Departamento de Medicina Molecular y Bioprocesos, Instituto de Biotecnología, Universidad Nacional Autonóma de México, Avenida Universidad 2001, Cuernavaca 62210, Méxicob Instituto Nacional de Salud Pública, Centro de Investigación sobre Enfermedades Infecciosas Avenida Universidad 655, Cuernavaca Morelos 62508, Méxicoc Instituto de Química, Universidad Nacional Autonóma de México, Circuito Exterior, Ciudad Universitaria, Coyoacán 04510, México D. F., México

Abbreviations: AMPs, antimicrobial peptides; ATCC,tion; CD, circular dichroism; CFU, colony forming unitsStandards Institute; DIPEA, N,N-diisopropylethylamineeagle's medium; DMF, N,N-dimethylformamide; FmocHBTU, 2-(1H-benzotriazole-1-yl)-1,1,3,3-tetramethylurHOBt, N-hydroxibenzotriazole hydrate; MIC, minimal inmulti-drug resistant; MTS, 3-(4,5-dimethylthiazol-2-yl)-5(4-sulfonyl)-2H- tetrazolium); POPC, L-α-phosphatidyloleoyl-sn-glycero-3-phospho-(1′-rac-glycerol); RP-HPLC, rliquid chromatography; SUVs, small unilamellar vesicles; T☆ A patent on Novel Hybrid Antibiotic Peptide and its Va(IMP Folio MX/E/2011/044744).⁎ Corresponding author at: Departamento de Medic

Universidad Nacional Autonóma de México, Avenida U62210, México. Tel.: +52 777 3171209; fax: +52 777

E-mail address: possani@ibt.unam.mx (L.D. Possani)

0304-4165/$ – see front matter © 2013 Elsevier B.V. Allhttp://dx.doi.org/10.1016/j.bbagen.2013.01.028

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 9 June 2012Received in revised form 15 January 2013Accepted 28 January 2013Available online 9 February 2013

Keywords:AntibioticAntimicrobial peptideCalcein releaseCell viabilityMDR clinical isolatePlasmodium berghei

Background: Microbial antibiotic resistance is a challenging medical problem nowadays. Two scorpion peptidesdisplaying antibiotic activity: hadrurin and vejovine were taken as models for the design of novel shorter pep-tides with similar activity.Methods: Using the standard Fmoc-based solid phase synthesis technique of Merrifield twelve peptides (18 to 29amino acids long) were synthesized, purified and assayed against a variety of multi-drug resistant Gram-negativebacteria from clinical isolates. Hemolytic and antiparasitic activities of the peptides and their possible interactionswith eukaryotic cells were verified. Release of the fluorophore calcein from liposomes treated with these peptideswas measured.Results:A peptide with sequence GILKTIKSIASKVANTVQKLKRKAKNAVA), and three analogs: Δ(Α29), Δ(K12-Q18;Ν26−Α29), and K4N Δ(K12-Q18; Ν26−Α29) were shown to inhibit the growth of Gram-negative (E. coliATCC25922) and Gram-positive bacteria (S. aureus), as well as multi-drug resistant (MDR) clinical isolated.The antibacterial and antiparasitic activities were found with peptides at 0.78 to 25 μM and 5 to 25 μM con-

centration, respectively. These peptides have low cytotoxic and hemolytic activities at concentrations sig-nificantly exceeding their minimum inhibitory concentrations (MICs), showing values between 40 and900 μM for their EC50, compared to the parent peptides vejovine and hadrurin that at the same concentrationof their MICs lysed more than 50% of human erythrocytes cells.Conclusions: These peptides promise to be good candidates to combat infections caused by Gram-negative bac-teria from nosocomial infections.General significance: Our results confirm that well designed synthetic peptides can be an alternative for solvingthe lack of effective antibiotics to control bacterial infections.

© 2013 Elsevier B.V. All rights reserved.

American type culture collec-; CLSI, Clinical and Laboratory; DMEM, Dulbecco's modified, 9-fluorenylmethoxicarbonyl;onium hexafluorophosphate;hibitory concentration; MDR,-(3-carboxymethoxyphenyl)-2-choline; POPG, 1-palmitoyl-2-everse-phase high-performanceFA, trifluoroacetic acidriants was deposited inMexico

ina Molecular y Bioprocesos,niversidad 2001, Cuernavaca3172388.

.

rights reserved.

1. Introduction

The phenomenon of antimicrobial resistance is one of the biggesthealth challenges faced today by themedical community. This problemhas created an urgent need for the development of novel classes of an-timicrobial agents [1,2]. Unfortunately, the number of new antibiotics inthe pipeline of the leading pharmaceutical companies has been declin-ing because they have shifted their attention towards more lucrativeareas of drug development [3,4]. A promising alternative for today'santibiotics is antimicrobial peptides (AMPs), since they have shown abroad spectrum of activity against pathogens (bacteria, fungi, parasitesand virus) and induce killing in a short contact time [5–7]. They areusually gene-encoded peptides that are expressed either constitutivelyor are inducible (through signal received from infectious or inflammatoryagents). All natural antimicrobial peptides sharemany common features,including small size (generally 12–50 amino acid residues long), cationiccharacter (with an overall net charge ranging from +2 to +9) and an

3428 L. Sánchez-Vásquez et al. / Biochimica et Biophysica Acta 1830 (2013) 3427–3436

amphipatic structure (containing ~50% hydrophobic residues) [8,9].Despite their sequence diversity, AMPs can be classified into four majorclasses: (a) α-helical peptides, (b) β-sheet stabilized by two or threedisulphide bridges, (c) extended helices with a predominance of one ormore amino acids (like histidine, arginine and proline) and (d) loopforming structures [10–12]. At present time, considerable effort hasbeen made to elucidate the mode of action of antimicrobial peptides.However, the molecular mechanism by which they exert activity againstmicroorganisms is not clear [13]. It is generally agreed that AMPs perturbmembranes by forming transient pores via one of the various modelsproposed to account for this effect, i.e. barrel-stave, carpet-like, toroidal(or wormhole) pore formation, detergent-type micellization, and in-duction of non-lamellar phases, leading to membrane permeabilizationand either leakage of cell content and osmotic instability, and/or pep-tide diffusion to intracellular targets [13–15]. The most abundant andwidely studied naturally-occurring types are the linear, amphipathicandα-helical peptides [16]. In a search for such new antibiotics, peptideswith antimicrobial activity from the venomof scorpionswere found. Thefirst molecule isolated was a defensin from the hemolymph of Leiurusquinquestriatus Hebraeus [17]. Subsequently, IsCTs 1 and 2were obtainedfrom Opisthacanthus madagascariensis [18]. Furthermore, opistoporines1 and 2 were isolated from Opistophtalmus carinatus [19]. In addition,scorpine was isolated from the scorpion Pandinus imperator, whichshowed antibacterial and antiparasitic activity, against Plamodiumberghei[20]. Recently, from Vejovis mexicanus and Hadrurus gertschi, two novelpeptides were identified: vejovine [21] and hadrurin [4]. Although theyshowed activity against multidrug resistant bacteria they also lyseeukaryotic cells (assayed by measuring hemolytic activity of humanerythrocytes). Therefore, one approach to generate highly active antimi-crobial peptides, which possess a low hemolytic activity, is to designnovel analogs based on the structure of natural peptides, but changingphysico-chemical properties that are known to influence the cytolyticactivity [8,20]. Among the properties taken into account are: helicity,hydrophobicity, hydrophobic moment, polar angle, net positive chargeand total number of amino acids [22]. This strategy was applied hereand produced peptides with better microbicide activity and less toxicitytowards host cells than vejovine and hadrurin, in view of potentialtherapeutic applications [5,21]. From twelve newly synthesized peptidesfollowing this strategy, four are shown to be effective in inhibiting bacte-rial growth, including bacteria strains isolated from a hospital clinicalsetting, and are promising leading compounds for the development ofnovel antibiotics.

2. Materials and methods

2.1. Material

Amino acids protected by 9-fluorenylmethoxicarbonyl (Fmoc) andFmoc-peptide amide linker resin were obtained from Novabiochem(La Jolla, CA). Other materials were: 2-(1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HBTU), from ChemPepInc., (Wellington, FL); N-hydroxibenzotriazole hydrate (HOBt), fromLC Sciences (Houston, TX); N,N-diisopropylethylamine (DIPEA) fromSigma-Aldrich (St. Louis, MO); trifluoroacetic acid (TFA), fromAmerican Bioanalytical Inc. (Natick, MA); analytical grade N,N-dimethylformamide (DMF) fromAldrichChemical Co. Inc. (Milwaukee,WI); 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfonyl)-2H- tetrazolium), abbreviated MTS was from Promega(Madison, WI); DMEM was supplied by HyClone (Logan, UT);1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-(1′-rac-glycerol) sodi-um salt (POPG) was purchased from Sigma (St. Louis, MO); calceinfrom Molecular Probes (Grand Island, NY.); L-α-phosphatidylcholine(Egg POPC, Chicken-99%) and cholesterol were obtained from AvantiPolar Lipids Inc. (Alabaster, AL); and fetal bovine serum (FBS) fromGibco (Carisbad, CA). HEK 293 and COS 7 cell lines were purchasedfrom American Type Culture Collection (ATCC; Bethesda, MD). Analytical

C18 RP-HPLC column (5 mm i.d.×250 mm) was from Vydac (Hysperia,CA). All other reagents were analytical grade products. The buffers wereprepared with double quartz-distilled water.

2.2. Bacterial strains

Escherichia coli ATCC25922, S. pneumonaie ATCC49619 andStaphylococcus aureus ATCC25923 were purchased from the AmericanType Culture Collection. MDR clinical isolates, which are non-susceptibleto at least one agent in three or more antimicrobial categories [23] andcausing nosocomial infections, were obtained from the Center forResearch on Infectious Diseases collection of the National Institute ofPublic Health, Cuernavaca, Morelos, Mexico. The various strains includethe following isolates: E. coli 170 [7], 4530, 5580 and 09240; E. cloacae2524, 6780, 06266 and 06268; K. pneumoniae 913, 1625 [24], 068228and 14218, which are extended-spectrum ß-lactamase-producers, inconsequence they are resistant to all penicillins and cephalosporins;P. aeruginosa 3599, 4660 [25], 5106 and 6102 which are metallo-ß-lactamase-producers and are resistant to all ß-lactamantibiotics includ-ing carbapenems; A. baumannii 5821, 5825, 5838, 5852, 7804, 7839 and7847 (carbapenem resistant); S. pneumoniae 150, penicillin resistant;S. typhimurium 2205, 2211 and 2217 (cephalosporins resistant).

2.3. Design of AMPs based on structural determinants

AMPs were designed based on three major structural determinants,which include: hydrophobicity (H), hydrophobic moment (μH) andcharge (Q), taking care that all peptides used in this work conserved anamphipatic and helicoidal conformation [26]. Twelve synthetic AMPswere designedwith hydrophobicity from−0.203 to−0.292, hydropho-bic moment of 0.422 to 0.639 and net charges of +5 to +8. The aminoacid sequences, molecular mass and structural parameters determinedfor the 12 AMPs used in this study are summarized in Fig. 1B and Table 1.

2.4. Peptide synthesis and purification

Peptides listed in Fig. 1B were prepared manually using the standardFmoc-based solid phase synthesis technique (Merrifield, 1963) on Rinkamide MBHA resin (0.54 mol/g resin) [11]. HBTU and HOBt were usedas coupling reagents, and double fold excess of Fmoc amino acids wasadded during every coupling cycle. After cleavage and deprotectionwith a mixture of trifluoroacetic acid/phenol/thioanisole/dithiotreitol/H2O (84:5:5:1:5, v/v) for 2 h at room temperature, crude peptides wererepeatedly extracted with diethyl ether and purified by reverse-phasehigh-performance liquid chromatography (RP-HPLC) on a C18 analyticalcolumn using an appropriate water/acetonitrile gradient in the presenceof 0.1% TFA for 60 min (purity >98%). The molecular masses of purifiedpeptides were determined by electrospray ionizationmass spectrometry(ESI-MS) using Mass Spectrometer LCQDuo (Thermo electron/Finningan,San Jose, CA).

2.5. Antibacterial assays

Initially the antibacterial activity of the peptides was qualitativelymeasured following the Kirby–Bauer method (1996), according to theCLSI (Clinical and Laboratory Standards Institute) recommendations[27]. Briefly, Petri dishes containing Müeller-Hinton agar weresown with bacteria inoculums from 1 to 2×108 colony-forming units(CFU)/ml, and then 3 μl of peptide solution was placed over the agar. In-cubation time was from 16 to 19 h at 35±2 °C. A halo of growth inhibi-tion was observed as a positive result. Two reference strains were used:E. coli ATCC25922 and S. aureus ATCC29213. Then, minimal inhibitoryconcentration (MIC) was determined by using a broth micro-dilutionmethod as indicated by the CLSI recommendations. A 96-well platewas used for bacterial growth in the presence of Müeller-Hinton brothmedium (85 μl) and using a peptide final concentration from 0.38 to

Fig. 1. Alignment of the peptide sequences. A. Comparison of the primary structure of vejovine and hadrurin with other peptides of the same family. The alignment was performedusing Clustal W2 server (http://www.ebi.ac.uk/Tools/clustalw2/index.html). Amino acids conserved residues are highlighted, filled gaps were used to maintain the alignmentaccording to the consensus sequence. B. Alignment of the native sequences with the designed peptides. Natural sequences have a fragment containing positively charged aminoacids (KLKR), which was included in the new peptides proposed. In the consensus sequence, asterisk (*) means identical residues, point (.) means more than one different residueand dash (–) means deletion.

3429L. Sánchez-Vásquez et al. / Biochimica et Biophysica Acta 1830 (2013) 3427–3436

50 μM(depending on the amount of the peptide available). To eachwell,a 5 μl of bacterial culture with 5×105 CFU/ml was added.

Microbial growth inhibition was observed after incubationfor 16–19 h at 35±2 °C. Experiments were performed with theGram-negative bacteria reference and different clinical isolates causingnosocomial infections, such as: E. coli (5 isolates), E. cloacae (4 isolates),K. pneumoniae (4 isolates), P. aeruginosa (4 isolates), A. baumannii(7 isolates), S. pneumonia (2 isolates) and S. typhimurium (3 isolates).

Table 1Parameters that influence the activity of synthetic peptides.

Peptide #aaa MWb RTc Hd μHe pIf Qg

Peptide 29 3080 47.07 0.176 0.507 11.47 +8Δ(A29) 28 3008 46.32 0.171 0.535 11.47 +8Δ(G1-T5; A29) 23 2496 35.65 0.087 0.516 11.43 +7Δ(G1-T5) 24 2566 36.50 0.097 0.483 11.43 +7Δ(G1-S8; A29) 20 2166 28.56 0.062 0.486 11.39 +6Δ(G1-S8) 21 2238 29.60 0.074 0.450 11.39 +6Δ(G1-S11, A29) 17 1896 25.50 −0.049 0.495 11.39 +6Δ(G1-S11) 18 1968 26.46 −0.029 0.455 11.39 +6Δ(G1-Q18; A29) 10 1154 17.47 −0.203 0.491 11.33 +5Δ(G1-Q18) 11 1225 18.44 −0.156 0.422 11.33 +5Δ(K12-Q18; N26-A29) 18 1954 42.33 0.270 0.639 11.39 +6K4N Δ(K12-Q18; N26-A29) 18 1940 44.46 0.292 0.618 11.33 +5

nd = Not determined.a Number of amino acids.b Molecular weight.c Retention time was obtained by RP-HPLC using C18 analytical column with a linear

gradient from buffer A (0.12% (v/v) TFA in water) to buffer B (0.1% TFA acetonitrile).The gradient used was 0–60% buffer B, run for 60 min.

d Mean hydrophobicity per residue of the whole peptide sequence.e Mean hydrophobic moment per residue of the whole peptide sequence.f pI = Isoelectric point.g Charge, positive charge at neutral pH.

2.6. Antiparasitic assays

The gametocyte/producing clone of Plasmodium berghei Anka strain2.34 (kindly provided by R. E. Sinden, Imperial College of Science Technol-ogy and Medicine, UK) was used. Gametocytes from infected in BALB/cwere obtained to produce ookinetes. Mice with gametocytemia around5% were bled by heart puncture using heparin (100 IU/ml blood).Ookinete cultures were carried out as described [28]. Leucocyte-depletedinfected-mouse blood was suspended in a proportion of 1:5 in culturemedium and tested in 100 μl aliquots in flat-bottom 96 well plates, andincubated at 20 °C during 24 h to allow ookinete development. Peptidesat concentrations of 5 μM and 25 μM were added to triplicate wells. Thenumber of ookinetes was assessed 24 h later in Giemsa-stained bloodsmears.

2.7. Hemolytic activity test

For hemolysis testing, 4 ml of freshly collected human blood wasadded to a tubewith EDTA. Red blood cells (RBCs) werewashed severaltimes using 40 ml of sterile phosphate buffered saline (PBS) andcentrifuged at 2000 g for 5 min, until the upper solution became clear.In the last stage, the RBCs were diluted in PBS adjusted to 2.6×108

cells/ml. The assay was performed in a final volume of 200 μl (195 μlof the solution of RBCs with 5 μl of selected peptide). Peptides at con-centrations of 1–1000 μM were incubated with RBCs at 37 °C for 1 h.PBS and 10% Triton X-100 were used as negative and positive controls,respectively. The samples were centrifuged at 2000 g for 5 min, andthe release of hemoglobin wasmonitored bymeasuring the absorbanceof the supernatant at 570 nm in a spectrophotometer NanoDrop. Eachassay was performed in triplicate, and the data were expressed asmean±SD. Percentage of hemolysis was calculated using the followingformula: % hemolysis=100 (Apeptide-APBS)/(Atriton-APBS). The peptides

3430 L. Sánchez-Vásquez et al. / Biochimica et Biophysica Acta 1830 (2013) 3427–3436

concentration that causes 50% hemolysis of the human erythrocytes(HC50) were obtained using a non-linear regression analysis, where thedata were fit with a logistical sigmoidal equation using the softwarepackage Origin 7 (Microcal Inc. U.S.A.).

2.8. Circular dichroism (CD) spectroscopy

CD spectra were recorded on a JASCO J-700 spectropolarimeter(Jasko, Tokyo, Japan) at a protein concentration of 0.25 mg/ml forall the peptides tested in both water and 60% of 2,2,2-trifluoroethanol(TFE) solution. Spectra were measured at 20 °C from 250 to 190 nm,using a 0.1 cm quartz cell thickness. For all samples, an average ofthree scans was used after subtracting average buffer spectra. Data areexpressed as mean residue molar ellipticity ([θ]), calculated as follows:[θ]=(θ×MRW)/(L×C), where θ is the ellipticity (in millidegrees), C isthe concentration (mg/ml), L is the pathlength cm), and MRW is themean residue weight (110 Da) [29].

2.9. Cell proliferation assays

The HEK 293 [30] and COS 7 cell lines [31], were used in MTS as-says [32] for the evaluation of cytotoxic effects of peptides. HEK 293and COS 7 cells were grown in Dulbecco Modified Eagle's Medium(DMEM) supplemented with 10% (v/v) fetal bovine serum (FBS) usingCorning cellBIND flasks and incubated in the presence of 5% CO2 at37 °C,while harvesting the cells for assay. For theMTS assay, the CellTiter96® AQneous ONE Solution Cell Proliferation Reagent (Promega,Madison, Wis, USA) was used following the manufacturer's instruc-tion. HEK 293 and COS 7 cells were seeded into 96-well plates at adensity of 2.5×104 and 1×104 cells per well, respectively. Peptideswere added to the wells at different concentrations (1, 12.5, 25, 50,100, 250, 500, and 1000 μM) and the medium with and withoutcells were used as controls. Following the addition of peptides, thecells were incubated for 48 h at 37 °C in an atmosphere of 5% CO2.After the time incubation, 20 μl of the reagent solution was added toeach well and incubated for 4 h in 5% CO2 atmosphere at 37 °C. In livingcells, mitochondrial reductases convert the MTT to formazan, whichproduces a purple color, allowing for monitoring of the reaction. Theabsorption at 492 nm is read using amicroplate reader (Bio-Rad Lab-oratories Inc.). The percentage of viable cells after treatment with thepeptides was calculated as AT/AC X100, where AT and AC are the ab-sorbance of treated and control cells (100% survival), respectively.The EC50, which refers to the peptide concentration that inhibitscell viability by 50% after 48 h exposure time, was calculated basedon non-linear regression analysis of the concentration–responsecurves obtained for MTT assays. All the experiments were performedin triplicate.

2.10. Preparation of phospholipid vesicles

The desired mixtures of phospholipids and cholesterol (PC:PG,1:3 mol:mol; PC:Chl, 7:3 mol:mol) were dissolved in chloroform andthen dried under nitrogen atmosphere, and subsequently kept over-night under high vacuum. Typically, 1 ml of buffer (10 mM HEPES,150 mMKCl, pH 7.4) was added to the lipid and the dispersion was ex-tensively vortexed. For preparation of small unilamellar vesicles (SUVs)the lipid dispersion was sonicated in a bath type sonicator until thesolution became translucent. SUVs for dye release experiments wereprepared as described above using calcein containing buffer (80 mMcalcein, 10 mMHEPES pH 7.4). After preparation, the untrapped calceinwas removed from the vesicles by gel filtration on a Sephadex G-50column eluted with buffer containing 10 mM HEPES in 150 mM KClat pH 7.4) [33].

2.11. Dye release experiments

The fluorescent dye experiments were performed essentially asdescribed by Belokoneva et al. [34]. Briefly, calcein leakage from vesicleswas fluorometrically monitored using a SLM-AMINCO spectrofluorome-ter by measuring the time-dependent increment of the fluorescence ofcalcein release. As calcein leaks from the vesicles due to the disrupting ac-tivity of cationic peptides, it becomes diluted in bath solution (150 mMNaCl, 2.5 mM CaCl2, 10 mM HEPES) and therefore dequenched;the increment in fluorescence is recorded (excitation=490 nm;emission=520 nm). Aliquots of a calcein-containing vesicle suspension(20–30 μl, depending on the lipid concentration)were injected into a cu-vette containing 0.9 ml of bath solution, stirred at constant temperatureand peptide solution of a defined concentrationwas injected into the cu-vette in volumes minor than 50 μl. The percentage of calcein releasewas evaluated by the equation: 100(F-F0)/(Ft-F0), where F is the fluo-rescence intensity achieved by the peptides, F0 is the fluorescenceintensity observed without the peptides, and Ft is the fluorescenceintensity corresponding to 100% calcein release determined by theaddition of 4.5 μl of 10% Triton X-100 solution.

3. Results

3.1. Peptide design and synthesis

Because of their antibacterial activity, the peptide sequences ofhadrurin and vejovine were chosen as templates for the developmentof new analogs. The first step for designing the AMPs consisted ofperforming an alignment of the natural sequences of these two peptidesand three others, isolated from scorpion venom, that show some se-quence similarities as indicated in Fig. 1A. They also are all effectiveantibacterial agents. These peptides have some sequence resemblanceat the N-terminal segment and part of the C-terminal tail sequence(highlighted in bold in Fig. 1). An additional observation from this com-parison is the presence of a stretch of basic amino acid residues KLKRK(see Fig. 1B). It is known that among important characteristics of AMPsis the presence of cationic amino acid residues. Another piece of decisiveinformation taken into consideration was the fact that removal of thefirst eight residues of vejovinemodifies its potency, rendering it inactive[21]. Although there is no clear rule for choosing the new analogs, therationale used here (see consensus stretches of sequences in Fig. 1B),was tomaintainmost of the N-terminal sequence, to include the segmentof basic residues and to use the consensus sequence for the remainingpart of the novel peptides. Furthermore, we have taken into account thechemical modifications earlier performed by our group, when workingwith hadrurin [4]. In that work a few amino acids were substituted intothe primary structure aimed at increasing the antibiotic activity and selec-tivity toward prokaryotic cells. Considering the above, the consensus se-quence as shown in Fig. 1 B was taken as a template for synthesizingpossible analogs with activity. The chosen peptides were submitted toanalysis of several parameters which include: charge, polar angle, hydro-phobicity and hydrophobicmoment (the values are shown in Table 1). Inaddition, another important point was to certify again the need for con-serving anN-terminal sequence similar to that of vejovine. This suggestedthe design of the two last peptides, shown in Fig. 1B, where theN-terminal sequence and thebasic stretch of amino acidswere conserved.These peptides: Δ(K12-Q18; N26-A29) and K4N Δ(K12-Q18;N26-A29)are also quite short, another desirable situation if eventually they couldbe applied in clinical trials, taken into consideration the cost–benefits ofsynthetic materials. According to this strategy, the peptides used hereshould assume an amphipatic, helical structure in the presence of ahydrophobic medium; conditions required for their interaction with thehydrocarbon chains of phospholipid membranes, as will be discussedlater (see Fig. 1, SupplementaryMaterial). The peptideswere synthesizedmanually using the solid phase methodology of Merrifield [11] and werepurified by HPLC (examples are included in Fig. 2 of Supplementary

Fig. 2. Circular dichroism spectra of synthetic peptides. For each spectrum the peptideconcentration was 0.25 mg/mL, in the presence of 60% TFE. Each reading was performedby triplicate.

3431L. Sánchez-Vásquez et al. / Biochimica et Biophysica Acta 1830 (2013) 3427–3436

Material). Purity of peptides were confirmed by analytical HPLC separa-tion (a symmetrical peak), and molecular weight determination bymass spectrometry, showing only one component after HPLC purification.Themolecular weights obtainedwere the expected ones, as shown inTable 1. The retention times of pure peptides are also indicated inTable 1.

Table 2Inhibition of bacterial growth by synthetic peptides.

Peptide E. coli ATCC 25922 S. aureus ATCC 29213

Peptide + +Δ(A29) + +Δ(G1-T5; A29) + −Δ(G1-T5) + −Δ(G1-S8; A29) + −Δ(G1-S8) + −Δ(G1-S11; A29) + −Δ(G1-S11) + −Δ(G1-Q18; A29) − −Δ(G1-Q18) − −Δ(K12-Q18; N26-A29) + +K4N Δ(K12-Q18; N26-A29) + +

The concentration of each peptide used for the biological activity assay was 10 μg/μL ina total volume of 3 μL. Plus (+) means inhibition of microbial growth. Minus (−)means no inhibition of microbial growth. The assay was realized by duplicate.

3.2. Characteristics of the synthetic peptides

Characteristics of the synthetic derivatives such as net positivecharge, hydrophobicity, and pI values are summarized in Table 1. Itshould be emphasized that the peptides known as Δ(K12-Q18;N26-A29) and K4N Δ(K12-Q18; N26-A29) have the highest degreeof hydrophobicity (0.27 and 0.292, respectively) and hydrophobicmoment (0.639 and 0.618, respectively), having a relatively high pos-itive charge (+6 and+5, respectively), whereas Δ(G1-S11; A29) andΔ(G1-S11) have the same charge value (+5) but a lower degree ofhydrophobicity. The peptide and the analog Δ(A29) have a balancebetween hydrophobicity (0.171 and 0.176, respectively) and hydro-phobic moment (0.535 and 0.507, respectively), having a largevalue of positive charge (+8 for both sequences). All the remainingproposed analogs (Δ(G1-T5; A29), Δ(G1-T5), Δ(G1-S8; A29) andΔ(G1-S8)) have intermediate values to those mentioned above fordifferent structural characteristics. On the other hand, one can observethat the sequences of the first ten synthetic peptides (Fig. 1B) differ byone amino acid (alanine) at the carboxyl terminal. This change falls inthe polar face of the helical structure, which has been reported to affectthe stability of the α helix [6,35]. Thus, as discussed previously, helicalstability is associated with hydrophobicity and, therefore, with the hy-drophobic moment, affecting the microbicidal activity and bacterialspecificity. All peptides maintain their amphipathic character (Fig. 1 ofSupplementary Material) and thus the variation of structural parame-ters is expected to help the understanding of how these changes mod-ulate the microbicidal/cytotoxic activity. Since helicity is one of themain properties of antimicrobial peptides, the circular dichroism (CD)spectra of all components were measured in trifluoroethanol/water(TFE/H2O, 60:40, v/v) mixtures, since TFE simulates a hydrophobicenvironment [36]. The CD spectra for all synthetic analogs showedthe expected characteristics of α-helical structure: a positive bandat ~194 nm and two negative bands at ~208 nm and ~222 nm, respec-tively (see Fig. 2) [37].

3.3. Antibiotic and hemolytic activity

As a first screening for antibacterial activity of the synthetic de-rivatives, a qualitative assay was performed with two referencestrains, which are susceptible to antibiotics: E. coli ATCC25922 andS. aureus ATCC29213. These strains correspond to Gram-negative andGram-positive bacteria, respectively. The peptide and analogs Δ(A29),Δ(K12-Q18; N26-A29) and K4N Δ(K12-Q18; N26-A29) inhibited thegrowthof both strains,whereasΔ(G1-Q18) andΔ(G1-Q18; A29) showedno antibacterial effects at concentrations up to over 50 μM (Table 2). It isimportant to highlight the conservation of the first five amino acids(GILKT) for preserving the biological activity in both bacterial types,since from the first fragment lacking these residues (Δ(G1-T5; A29);see Fig. 1B), the microbicidal effect was lost in the Gram-positive modelstrain (S. aureus). Once the analogswith antimicrobial activitywere iden-tified, the correspondingminimal inhibitory concentrationwas evaluatedagainst the E. coli ATCC25922 strain and several other MDR clinicalisolates causing nosocomial infections, among which are: K. pneumonia,E. cloacae, P. aeruginosa, and A. baumannii. The results in Table 3 showthat not all synthetic peptide analogs inhibit bacterial growth efficiently.In particular, the peptide and analogsΔ(A29),Δ(K12-Q18; N26-A29) andK4N Δ(K12-Q18; N26-A29) have the major antibiotic effect within therange from 0.78 μM to 25 μM, whereas the remaining peptides showedvalues above the highest concentration tested (>50 μM).

Among the characteristics shared by cationic amphipathic pep-tides are the ability to form pores in the membrane of both prokary-otes and eukaryotes. Lysis of human erythrocytes is a known test usedfor identification of possible cytolytic activity towards eukaryotic cells[34]. Consequently the synthetic peptides were assayed for determi-nation of their hemolytic activity, using peptide concentrations with-in the range shown to cause inhibition of bacterial growth (0.05 to50 μM). Experiments conducted with the highest peptide concentra-tion (50 μM) gave the following results, in which the values betweenparenthesis after the name of the peptides are the percentage ofhemolysis: the peptide (12.1); and the analogs Δ(A29) (11.7); Δ(G1-T5;A29) (3.27); Δ(G1-T5) (3.58); Δ(G1-S8; A29) (1.47); Δ(G1-S8) (1.54);Δ(G1-S11; A29) (0.88); Δ(G1-S11) (3.45), Δ(K12-Q18; N26-A29) (8.2)and K4N Δ(K12-Q18; N26-A29) (5.2). The effective concentration thatcaused 50% erythrocyte lyses (HC50) alone was determined for peptideswith better antimicrobial activity (the peptide, and analogs Δ(A29),Δ(K12-Q18; N26-A29) and K4N Δ(K12-Q18; N26-A29)). The concentra-tion of peptides causing hemolysis exceeded the concentration of MICby about one order of magnitude. The analogs Δ(A29) and Δ(K12-Q18;N26-A29) showed the lowest degree of hemolytic activity (824 and923 μM, respectively), whereas the peptide and the analog K4NΔ(K12-Q18; N26-A29) exhibited the highest activity (665 and 640 μM,respectively). It should be noted that the minimum effect determinedon erythrocytes is at a concentration below the MICs (Tables 3 and 4).

Table 3Minimum inhibitory concentrations of the peptides against E. coli control strain and multi-resistant bacteria.

Peptide/bacteria Ec Ec Ec Eclo Eclo Kpn Kpn Pa Pa Ab Ab

25922 170 4530 2524 6780 913 1625 4677 5106 5838 5852

Δ(A29) 1.56 0.78 3.12 3.12 3.12 3.12 3.12 6.25 12.5 3.12 0.78Peptide 1.56 3.12 3.12 6.25 3.12 3.12 1.56 25 12.5 1.56 1.56Δ(G1-T5; A29) 12.5 3.12 25 >50 25 >50 3.12 12.5 >50 12.5 6.25Δ(G1-T5) 12.5 1.56 >50 3.12 >50 >50 6.25 >50 >50 6.25 3.12Δ(G1-S8; A29) 25 >50 >50 >50 >50 >50 >50 >50 >50 >50 >50Δ(G1-S8) >50 12.5 >50 6.25 >50 25 >50 >50 >50 12.5 6.25Δ(G1-S11; A29) >50 >50 >50 25 >50 >50 >50 >50 >50 >50 >50Δ(G1-S11) >50 >50 >50 12.5 >50 >50 >50 >50 >50 >50 >50Δ(K12-Q18; N26-A29) 1.56 1.56 25 >50 3.12 3.12 1.56 25 25 3.12 1.56K4N Δ(K12-Q18; N26-A29) 1.56 1.56 >50 >50 3.12 6.25 3.12 >50 >50 3.12 1.56

Ec = Escherichia coli, Eclo = Enterobacter cloacae, Kpn = Klebsiella pneumoniae, Pa = Pseudomonas aeruginosa, Ab = Acinetobacter baumannii. The values shown in the table are theconcentration (μM) at which no growth is observed. Each test was performed twice by duplicate.

3432 L. Sánchez-Vásquez et al. / Biochimica et Biophysica Acta 1830 (2013) 3427–3436

In contrast, hadrurin [4] and vejovine [21] have a HC50 (50% of hemolyticactivity) at a concentration of 100 μM and 5 μM, respectively. As a result,the therapeutic indexes of the new peptides described here are signifi-cantly improved. The therapeutic index is a measurement calculated bydividing the HC50 by the MIC. High values of therapeutic index indicatebetter possibility of using the antibacterial peptide for control of infectionsin experimental animals [21]. Subsequently, the peptides with enhancedtherapeutic index (the peptide, analogs Δ(A29), Δ(K12-Q18; N26-A29)and K4N Δ(K12-Q18; N26-A29)) were tested against more than 30different clinical isolates from distinct hospitals of Mexico. Table 4shows the variations in antibiotic susceptibility of the different strainsof Gram-negative bacteria to the synthetic analogs. The best resultsdetermined in MIC assays were observed for strains: E. coli (0.78–6.25 μM), E. cloacae (3.12–12.5 μM), K. pneumoniae (1.56–6.25 μM),A. baumannii (0.78–3.12 μM) and S. typhimurium (1.56–6.25 μM).

Table 4MIC comparison values of parental peptides with synthetic peptides.

Bacteria/peptide Hadrurin (μM) Vejovine (μM) Δ(A29) (μM)

A. baumannii 7804 nd 20 3.1A. baumannii 7839 nd 20 3.1A. baumannii 7847 nd 5 3.1E. coli ATCC25922 nd 4.4 1.6E. coli ATCC25926 b10 nd ndE. coli 170 nd 4.4 0.8E. coli 4530 nd 4.4 3.1E. coli 5580 50 10 1.6E. coli 9240 nd 20 1.6E. cloacae 129 b10 nd ndE. cloacae 2524 nd 4.4 3.1E.cloacae 6266 nd >20 3.1E.cloacae 6268 nd >20 3.1E.cloacae 6780 nd nd 3.1E. feacalis 51 50 nd ndK. pneumoniae 9 b10 nd ndK. pneumoniae 913 nd 17.7 3.1K. pneumoniae 1625 nd 17.7 3.1K. pneumoniae 68228 nd 50 3.1K. pneumoniae 14218 nd 50 3.1P. aeruginnosa 4677 nd 17.7 6.2P. aeruginnosa 5106 nd 17.7 12P. aeruginnosa 3599 nd 50 50P. aeruginnosa 4660 nd 50 50P. aeruginnosa 6102 nd 100 50P. aeruginosa ATCC9027 40 nd ndP. aeruginosa PG201 40 nd ndS. pneumoniae ATCC49619 nd nd 50S. typhimurium 50 nd ndS. typhimurium 2205 nd nd 3.1S. typhimurium 2211 nd nd 1.6S. typhimurium 2217 nd nd 3.1S. marscencens ATCC13880 50 nd nd

nd =Not determined. The values shown in the table are the concentrations (μM) at which

On the other hand, for P. aeruginosa and S. pneumoniae the microbi-cidal activity obtained varied from 12.5 to 25 μM, while the growthinhibition for the remaining analogs was achieved at concentrations inthe range of 25–50 μM. For the strains of S. pneumoniae ATCC49619and clinical isolates the MIC values obtained were from 1.56 to 50 μM.

3.4. Antiparasitic activity of the synthetic peptides

Earlier work performed by our group with scorpine, a scorpionantimicrobial peptide, showed that the same peptide was capable ofinhibiting the development of ookinetes (the invasive phase of P. berghei,a parasite that causes malaria in rodents) [20]. This discovery was theinspiration for trials aiming at using this kind of peptide for control ofmalaria [38]. For this reason it was decided to test the effect of the pep-tides described here for their effective action against the same parasite.

Peptide (μM) Δ(K12-Q18; N26-A29) (μM) K4N Δ(K12-Q18; N26-A29)

3.1 1.6 1.63.1 1.6 1.63.1 1.6 1.61.6 1.6 1.6nd nd nd3.1 1.6 1.63.1 25 501.6 1.6 6.21.6 3.1 6.2nd nd nd6.2 50 503.1 3.1 6.6.2 3.1 253.1 3.1 3.1nd nd ndnd nd nd3.1 3.1 6.21.6 1.6 3.16.2 3.1 3.13.1 3.1 3.125 25 5012 25 5050 25 5050 12 2550 12 25nd nd ndnd nd nd25 50 50nd nd nd1.6 1.6 3.11.6 1.6 3.16.2 1.6 6.2nd nd nd

no growth is observed. Each test was performed twice by duplicate.

Fig. 3. Antiparasitic activity of synthetic peptides. The columns show the inhibitionpercentage of ookinetes formation for the four more active peptides.

3433L. Sánchez-Vásquez et al. / Biochimica et Biophysica Acta 1830 (2013) 3427–3436

The results obtained in growth inhibition assays of ookinetes fromP. berghei (Fig. 3) showed that the peptide, and the analogs Δ(A29),Δ(K12-Q18; N26-A29) and K4N Δ(K12-Q18; N26-A29) inhibit the for-mation of this phase of the parasite by 40% at 5 μM, doubling this per-centage if the concentration is increased five folds. These results showthat some of our synthetic peptides are in fact capable of inhibitingthe development of parasitic organisms, similar to the effect of scorpine.

Fig. 4. Cytotoxic activity of the synthetic peptides on HEK 293, COS 7 and erythrocytes cells. CThe cells were treated with peptides at different concentrations and incubated at 37 °C forhemolytic activities of peptides (C) were measured by using the hemoglobin release assay.

3.5. Viability of eukaryotic cells in culture

In addition to the cytolytic effects conducted with human erythro-cytes (see results listed above) it was decided to assay the effect ofthese synthetic antimicrobial peptides on the viability of two differentlines of eukaryotic cells in culture: HEK 293 and COS 7. The cytotoxiceffect of the peptide, analogs Δ(A29), Δ(K12-Q18; N26-A29) and K4NΔ(K12-Q18; N26-A29) were tested in these two cell lines in culture.The results show that concentrations up to 50 μM of the peptides,applied for the period of 48 h in both HEK 293 as COS 7 cells presentno significant changes with respect to the controls. However, the per-centage of cellular viability for a broader range of peptide concentrationis increased, varying from 40 to 83% (see Fig. 4). The values of EC50

for HEK 293 cells were 46, 47, 80 and 139 μM (Peptide, Δ(A29),Δ(K12-Q18; N26-A29) and K4N Δ(K12-Q18; N26-A29, respectively),whereas for COS 7 cells the EC50 were: 44, 52, 132 and 123 μM for thesame peptides, respectively. These results suggest that application ofpeptide at much higher concentrations than their MIC values againstmicroorganisms can cause mild disruption to these eucakryotic cells.

3.6. Interactions of synthetic peptides with phospholipid membranes

In viewof the above results obtainedwith erythrocytes and eukaryoticcells in culture it was decided to verify if these peptides were capableof damaging artificially prepared membrane bilayers. Cationic AMPswith antibacterial activity usually show amembrane-permeabilizingactivity [39]. To confirm whether AMPs used in the current study cause

ytotoxic effect of the peptides on HEK 293 (A) and COS 7 cells (B), using the MTT assay.48 h. The percentages of viability cells were calculated as described in Section 2. TheThe results are shown as mean±SD of three independent experiments.

Table 5Percentages of calcein release in small unilamellar vesicles (SUVs).

Peptide PC:Chl (7:3 mol:mol) PC:PG (1:3 mol:mol)

Peptide 32 100Δ(A29) 31 100Δ(K12-Q18; N26-A29) 18 66K4N Δ(K12-Q18; N26-A29) 33 100

The values shown in the table are the maximum percent release at 25 μM registered.Each reading was performed by triplicate.

3434 L. Sánchez-Vásquez et al. / Biochimica et Biophysica Acta 1830 (2013) 3427–3436

dye-leakage through pore formation and/or membrane perturbation,we employed the liposomemembranemodels with a composition sim-ilar to eukaryotic (PC:Chl, 7:3 molar ratio) or prokaryotic (PC:PG, 1:3molar ratio) cells. The results of the experiment are shown in Fig. 5and Table 5. As seen in panel A the highest percentage of release isreached at a concentration of 20 μM and, except for the peptide, anapparent saturation of the system is reached in the range of concentra-tions assayed. The most active sequence toward the eukaryotic mem-brane was the analog Δ(A29), although for highest concentrationtested (25 μM), the percentage of calcein release was not greater than40%. In contrast, for the analog K4N Δ(K12-Q18; N26-A29) it barelyexceeded 15% of calcein release (Table 5). Fig. 5B shows data obtainedwith the vesicles similar to prokaryotic cell membrane, in which thepercentage of calcein release determined for all peptides is >60%, 65%for Δ(K12-Q18; N26-A29), and nearly 100% for the other syntheticderivatives (Table 5). These results indicate that membranes similar toprokaryotic cells are more susceptible to the deleterious action of theAMPs reported here. It is surmised that the pore formation mechanism,responsible for releasing the fluorophore is probably the molecularmechanism taking place here. It was not possible, though, to distinguishwhich exactly of the proposed mechanisms [barrel-stave, carpet-like,toroidal (or wormhole) pore formation or others] is responsible forthe dye leakage.

4. Discussion

Comparative analysis of the amino acid sequence of several anti-microbial peptides found in scorpion venoms allowed us to place in

Fig. 5. Calcein dye-leakage experiments. The graphics shows percentage of calcein releaseas a function of peptide concentration. In (A) the composition of the liposomeswas PC:Chl(7:3 mol:mol). In (B) the composition of the liposomes was PC:PG (1:3 mol:mol).

evidence certain sequence features putatively important for retentionof the microbicidal activity. From the analysis conducted it was evidentthat the highest percentage of similarity was found in the amino- andcarboxy-terminal regions of the peptides (Fig. 1A). Based on theknownmodels, alternative AMPs could be designed bymeans of specificsubstitutions, deletions and hybridization of sequences of the naturalpeptides [40,41]. As shown in Fig. 1B, parental peptides have a positivelycharged residue fragment (KLKR) resulting in the electrostatic attrac-tion between cationic peptides andmicrobial membranes. Several stud-ies suggest that hydrophobicity is one of the main parameters thatinfluence the antimicrobial activity, since it is an important factor inpromoting hydrophobic interaction with the lipid bilayer [42]. Further-more, it has been shown that the reduction of the total number of aminoacids in the sequence of some antimicrobial peptides can lead to greaterefficiency of the microbicidal activity, when compared to natural tem-plates [43].

From the twelve peptides designed, at least four of them (the pep-tide, analogs Δ(A29), Δ(K12-Q18; N26-A29) and K4N Δ(K12-Q18;N26-A29)) were shown to be efficient antibacterial agents against sev-eral MDR clinical isolates causing nosocomial infections. These peptideshave high positive charge, amphipaticity, hydrophobicity (Table 1 andFig. 1 of Supplementary Material) and preserve the same sequence ofthe first five amino acids of the amino region, which are important formaintaining the activity against Gram-positive bacteria (Table 2). One ex-planation for this phenomenon could be the different membrane com-position and the thickness of the layer of peptidoglycans of this typeof bacteria [44].The variability obtained in the MIC assays (Tables 3and 4) could be explained by taking into consideration the differencesin the composition of the outer membrane of the bacterial strains, con-ferring a barrier against different antibacterial agents. Additional litera-ture discussing this point can be found in [45–48]. The properties ofthe peptides such as amphipathicity, hydrophobicity and charge enablethem to act selectively toward prokaryotic membranes. Hence, theypresent a better antimicrobial activitywith less hemolytic effect (see re-sults above) than the parental peptides. Vejovine has a MIC of 4.4 μMagainst E. coli ATCC25922 and a HC50 of 100 μM, whereas hadrurin hasa 7.5 μM MIC value in the same strain with a HC80 of 20 μM. In this re-gard, the peptide, and analogs Δ(A29), Δs(K12-Q18; N26-A29) andK4N Δ(K12-Q18; N26-A29) improved the microbicidal activity 2.9-foldin the E. coli ATCC25922 strain. It should be noted that in the generaEnterobacter and Klebsiella the biological activity was enhanced 2.5and 3-fold respectively. According to the CD spectra, all the peptidesassayed adopt α-helical conformation in the presence of TFE, whichwas used as a membrane mimetic media (Fig. 2). The antimicrobial ac-tivity observed with the peptide, analogs Δ(A29), Δ(K12-Q18;N26-A29) and K4N Δ(K12-Q18; N26-A29) is parallel to that describedby Sunkyu et al. [49]. These authors reported better antimicrobial activ-ity for peptides which usually are retained for a longer time on the C18columns of the HPLC, meaning that they contain more hydrophobicamino acid sequences. This is confirmed here. Our four peptides (pep-tide, Δ(A29), Δ(K12-Q18; N26-A29) and K4N Δ(K12-Q18; N26-A29)have higher retention times on the HPLC column (Table 1), and are ef-fective over a wide range of bacterial strains (Fig. 2 of SupplementaryMaterial).

3435L. Sánchez-Vásquez et al. / Biochimica et Biophysica Acta 1830 (2013) 3427–3436

As mentioned earlier, due to the broad spectrum and potent anti-microbial activity of the peptide, Δ(A29), Δ(K12-Q18; N26-A29) andK4N Δ(K12-Q18; N26-A29), they were assayed concerning inhibitionof ookinete formation by the malaria parasite (P. berghei), using sim-ilar assays as reported by others [50,51]. As seen in Fig. 3, all of thepeptides inhibit by 40% the development of ookinetes, when assayedat a concentration of 5 μM; higher concentrations (25 μM) inhibitedabove 80%. The peptide scorpine was shown to possess antiparasiticactivity against ookinetes and gametocytes of P. berghei with ED50

0.7 μM and 10 μM, respectively [20]. Additional results related to para-site inhibition by similar peptides are described in [52,53]. Concerningthe cytolitic effect, as reported in Section 3, our peptides showed a lowhemolytic activity towards human erythrocytes, which contrasts withthe same effect caused by the parent peptides vejovine and hadrurin(more than 50%), used at lower concentrations. However, the possibleeventual application of the peptides for topical uses prompted us totest whether other cell lines would be resistant to the extent demon-strated by the erythrocytes. As shown in Fig. 4, the viability both ofHEK 293 and COS 7 cells in culture is quite significant. The inhibitionfor COS 7 cells was in the range of 27%when treatedwith a high concen-tration (50 μM) of peptide Δ(K12-Q18; N26-A29). This contrasts withresults obtained with strains of cells 3T3, where the effects of syntheticpolymers of three amino acids (KLA) with a variable length showedcytotoxicity at relatively low concentrations. Since, peptides containingfourteen amino acids caused cytotoxic activity at 3–8 μM, whereasthose with twenty one have a lethal activity at concentrations rangingfrom 6 to 22 μM [54].

Finally, the experiments conducted with liposomes, which are lipidbilayers artificially prepared,were performed to see howmuch the pep-tides reported heremight damage the bilayer. The results show that thelipid compositions more sensitive to the disruptive effect of the pep-tides were those composed of charged lipids (PC:PG, 1:3 molar ratio)compared to lipid bilayers composed of zwiterionic lipids (PC:Chl, 7:3molar ratio). The former liposomes released 100%of calceinwhen treat-ed with 25 μM peptides (Fig. 5, panel B), whereas the latter reachedonly ~33% (Fig. 5 panel A) with an apparent saturation of the response(Table 5). Hence, the conclusion is that these peptides preferentiallyrecognize liposomes containing negatively charged lipids, such asthose present in prokaryotic cells. The literature reports that antimicro-bial peptides like cecropins, melittin and magainins exert their effectson the lipid bilayer via a pore formation mechanism, which leads tothe disruption of the membrane and finally cell death [55]. In addition,it was determined that dermaseptin, cecropin A, B and P dissipate theionic gradient at low concentrations, while higher concentrationswere required to achieve complete release of calcein. These resultsindicate that insertion of the peptide into the lipid bilayer is dependenton the concentration of the AMPs [27].

For the case reported here, the analysis performed and the selection ofparameters, like hydrophobicity, helicity, hydrophobic moment, positivenet charge and selection of amino acid sequence similar to knownantimi-crobial peptides (vejovine and hadrurin) were important for obtainingrandom coil peptides that could adopt an α-helical structure when incontact with the prokaryotic cells, causing disruption of the membraneorganization, ultimately resulting in cell lysis.

5. Conclusion

In conclusion taking into account the low hemolytic effect, thesignificant antimicrobial activity against pathogenic clinical isolatesand the low affinity for eukaryotic cell membranes, confirmed bythe high percentage of cell viability of cell cultures, the synthetic pep-tides presented in this work have the potential to be developed as anti-microbial agents.

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.bbagen.2013.01.028.

Acknowledgments

The authors acknowledge Mr. Fernando Reyna from the NationalInstitute of Public Health, M. Sci. Rocío Patiño of the Chemistry Insti-tute and Dr. Fernando Z. Zamudio from the Institute of Biotechnologyfor their technical assistance. In addition MBA Mario Trejo Loyo andM. Sci. Martin Patiño from the Institute of Biotechnology are recog-nized for their technical assistance in preparing the patent applicationpresented at the Mexican Office of Patent. This work was supportedby grants from DGAPA-UNAM number IN200113 and the scholarshipCONACyT 290564 to Lorenzo Sánchez-Vásquez.

References

[1] P.M. Hawkey, A.M. Jones, The changing epidemiology of resistance, J. Antimicrob.Chemother. 64 (Suppl. 1) (2009) i3–i10.

[2] L.B. Rice, The clinical consequences of antimicrobial resistance, Curr. Opin.Microbiol.12 (2009) 476–481.

[3] B. Spellberg, R. Guidos, D. Gilbert, J. Bradley, H.W. Boucher, W.M. Scheld, J.G.Bartlett, J. Edwards Jr., The epidemic of antibiotic-resistant infections: a call to actionfor the medical community from the Infectious Diseases Society of America, Clin.Infect. Dis. 46 (2008) 155–164.

[4] A. Torres-Larios, G.B. Gurrola, F.Z. Zamudio, L.D. Possani, Hadrurin, a new antimi-crobial peptide from the venom of the scorpion Hadrurus aztecus, Eur. J. Biochem.16 (2000) 5023–5031.

[5] H. Powers, Bad Bugs, No Drugs: As Antibiotic Discovery Stagnates… A PublicHealth Crisis Brews, Infectious Disease Society of America, Arlington, VA, 2004,pp. 1–37.

[6] Y. Chen, A.I. Vasil, L. Rehaume, C.T. Mant, J.L. Burns, M.L. Vasil, R.E.W. Hancock, R.S.Hodges, Comparison of biophysical and biologic properties of alpha-helical enan-tiomeric antimicrobial peptides, Chem. Biol. Drug Des. 67 (2006) 162–173.

[7] J. Silva, C. Aguilar, G. Ayala, M.A. Estrada, U. Garza-Ramos, R. Lara-Lemus, L.Ledezma, TLA-1: a new plasmid-mediated extended-spectrum beta-lactamasefrom Escherichia coli, Antimicrob. Agents Chemother. 44 (2000) 997–1003.

[8] B.M. Peters, M.E. Shirtliff, M.A. Jabra-Rizk, Antimicrobial peptides: primeval mole-cules or future drugs? PLoS Pathog. 6 (10) (2010) e1001067.

[9] J.D.F. Hale, R.E.W. Hancock, Alternative mechanisms of action of cationic antimi-crobial peptides on bacteria, Expert Rev. Anti Infect. Ther. 5 (2007) 951–959.

[10] M. Dathe, T. Wieprecht, H. Nikolenko, L. Handel, W.L. Maloy, D.L. MacDonald, M.Beyermann,M. Bienert, Hydrophobicity, hydrophobicmoment and angle subtendedby charged residues modulate antibacterial and haemolytic activity of amphipathichelical peptides, FEBS Lett. 403 (1997) 208–212.

[11] R.B. Merrifield, Solid phase peptide synthesis. I. The synthesis of a tetrapeptide,J. Am. Chem. Soc. 85 (1963) 2149–2154.

[12] A.R. Koczulla, R. Bals, Antimicrobial peptides: current status and therapeutic po-tential, Drugs 63 (2003) 389–406.

[13] K.A. Brogden, Nat. Rev. Microbiol. 3 (2005) 238–250.[14] Y. Huang, J. Huang, Y. Chen, Antimicrobial peptides: pore formers or metabolic

inhibitors in bacteria? Protein Cell. 1 (2010) 143–152.[15] A.K. Marr, W.J. Gooderham, R.E. Hancock, Antibacterial peptides for therapeutic

use: obstacles and realistic outlook, Curr. Opin. Pharmacol. 6 (2006) 468–472.[16] M. Stark, L.P. Liu, C.M. Deber, Cationic hydrophobic peptides with antimicrobial

activity, Antimicrob. Agents Chemother. 46 (2002) 3585–3590.[17] P. Díaz, G. D'Suze, V. Salazar, C. Sevcik, J.D. Shannon, N.E. Sherman, J.W. Fox,

Antibacterial activity of six novel peptides from Tityus discrepans scorpion venom. Afluorescent probe study of microbial membrane Na+ permeability changes, Toxicon54 (2009) 802–817.

[18] G. Corzo, P. Escoubas, E. Villegas, K.J. Barnham, W. He, R.S. Norton, T. Nakajima,Characterization of unique amphipathic antimicrobial peptides from venom ofthe scorpion Pandinus imperator, Biochem. J. 359 (Pt 1) (2001) 35–45.

[19] L. Dai, G. Corzo, H. Naoki, M. Andriantsiferana, T. Nakajima, Purification,structure-function analysis, and molecular characterization of novel linear peptidesfrom scorpion Opisthacanthus madagascariensis, Biochem. Biophys. Res. Commun.293 (2002) 1514–1522.

[20] R. Conde, F.Z. Zamudio, M.H. Rodríguez, L.D. Possani, Scorpine, an anti-malaria andanti-bacterial agent purified from scorpion venom, FEBS Lett. 471 (2001) 165–168.

[21] C.A. Hernández-Aponte, J. Silva-Sanchez, V. Quintero-Hernández, A.Rodríguez-Romero, C. Balderas, L.D. Possani, G.B. Gurrola, Vejovine, a new antibioticfrom the scorpion venom of Vaejovis mexicanus, Toxicon 57 (2011) 84–92.

[22] M.R. Yeaman, N.Y. Yount, Mechanisms of antimicrobial peptide action and resistance,Pharmacol. Rev. 55 (2003) 27–55.

[23] A.P. Magiorakos, A. Srinivasan, R.B. Carey, Y. Carmeli, M.E. Falagas, C.G. Giske, S.Harbarth, J.F. Hindler, G. Kahlmeter, B. Olsson-Liljequist, D.L. Paterson, L.B. Rice,J. Stelling, M.J. Struelens, A. Vatopoulos, J.T.Weber, D.L. Monnet, Multidrug-resistant,extensively drug-resistant and pandrug-resistant bacteria: an international expertproposal for interim standard definitions for acquired resistance, Clin. Microbiol.Infect. 18 (3) (2012) 268–281.

[24] J. Silva, R. Gatica, C. Aguilar, Z. Becerra, U. Garza-Ramos, M. Velázquez, G. Miranda,B. Leaños, F. Solórzano, G. Echániz, Outbreak of infection with extended-spectrumbeta-lactamase-producing Klebsiella pneumoniae in a Mexican hospital, J. Clin.Microbiol. 39 (2001) 3193–3196.

3436 L. Sánchez-Vásquez et al. / Biochimica et Biophysica Acta 1830 (2013) 3427–3436

[25] U. Garza-Ramos, R. Morfin-Otero, H.S. Sader, R.N. Jones, E. Hernández, E.Rodriguez-Noriega, A. Sanchez, B. Carrillo, S. Esparza-Ahumada, J. Silva-Sanchez,Metallo-beta-lactamasegenebla(IMP-15) in a class 1 integron, In95, fromPseudomonasaeruginosa clinical isolates from a hospital in Mexico, Antimicrob. Agents Chemother.52 (2008) 2943–2946.

[26] M. Schiffer, A.B. Edmundson, Use of helical wheels to represent the structures of pro-teins and to identify segments with helical potential, Biophys. J. 7 (1967) 121–135.

[27] T.D. Wikins, L.V. Holdeman, I.J. Abramson, W.E. Moore, Standardized single-discmethod for antibiotic susceptibility testing of anaerobic bacteria, Antimicrob.Agents Chemother. 1 (1972) 451–459.

[28] M.C. Rodríguez, G. Margos, H. Compton, M. Ku, H. Lanz, M.H. Rodríguez, R.E.Sinden, Plasmodium berghei: routine production of pure gametocytes, extracellulargametes, zygotes, and ookinetes, Exp. Parasitol. 101 (2002) 73–76.

[29] C. Perez-Iratxeta, M.A. Andrade-Navarro, K2D2: estimation of protein secondarystructure from circular dichroism spectra, BMC Struct. Biol. 13 (2008) 8–25.

[30] F.L. Graham, J. Smiley, W.C. Russell, R. Nairn, Characteristics of a human cell linetransformed by DNA from human adenovirus type 5, J. Gen. Virol. 36 (1977) 59–74.

[31] F.C. Jensen, A.J. Girardi, R.V. Gilden, H. Koprowski, Infection of human and simiantissue cultureswith Rous sarcoma virus, Proc. Natl. Acad. Sci. U. S. A. 52 (1964) 53–59.

[32] J.A. Plumb, Cell sensitivity assays: the MTT assay, Methods Mol. Med. 88 (2004)165–169.

[33] T. Wieprecht, O. Apostolov, J. Seelig, Binding of the antibacterial peptide magainin 2amide to small and large unilamellar vesicles, Biophys. Chem. 85 (2000) 187–198.

[34] O.S. Belokoneva, E. Villegas, G. Corzo, L. Dai, T. Nakajima, The hemolytic activity of sixarachnid cationic peptides is affected by the phosphatidylcholine-to-sphingomyelinratio in lipid bilayers, Biochim. Biophys. Acta 1617 (2003) 22–30.

[35] Y. Chen, M.T. Guarnieri, A.I. Vasil, M.L. Vasil, C.T. Mant, R.S. Hodges, Role of pep-tide hydrophobicity in the mechanism of action of α-helical antimicrobial pep-tides, Antimicrob. Agents Chemother 5 (2007) 1398–1406.

[36] E.F. Haney, H.N. Hunter, K.Matsuzaki, H.J. Vogel, Solution NMR studies of amphibianantimicrobial peptides: linking structure to function? Biochim. Biophys. Acta 1788(8) (2009) 1639–1655.

[37] N.J. Greenfield, Using circular dichroism spectra to estimate protein secondarystructure, Nat. Protoc. 1 (2006) 2876–2890.

[38] W. Fang, J. Vega-Rodríguez, A.K. Ghosh, M. Jacobs-Lorena, A. Kang, St.R.J. Leger,Development of transgenic fungi that kill human malaria parasites in mosquitoes,Science 331 (2011) 1074–1077.

[39] H.T. Chou, H.W. Wen, T.Y. Kuo, C.C. Lin, W.J. Chen, Interaction of cationic antimicro-bial peptides with phospholipid vesicles and their antibacterial activity, Peptides 31(2010) 1811–1820.

[40] Z.G. Tian, T.T. Dong, D. Teng, Y.L. Yang, J.H. Wang, Design and characterization ofnovel hybrid peptides from LFB15(W4,10), HP(2–20), and cecropin A based onstructure parameters by computer-aided method, Appl. Microbiol. Biotechnol.82 (2009) 1097–1103.

[41] A.A. Langham, H. Khandelia, B. Schuster, A.J. Waring, R.I. Lehrer, Y.N. Kaznessis,Correlation between simulated physicochemical properties and hemolycity of

protegrin-like antimicrobial peptides: predicting experimental toxicity, Peptides29 (2008) 1085–1093.

[42] T. Tachi, R.F. Epand, R.M. Epand, K. Matsuzaki, Position-dependent hydrophobic-ity of the antimicrobial magainin peptide affects the mode of peptide-lipid inter-actions and selective toxicity, Biochemistry 41 (2002) 10723–10731.

[43] S.H. Lee, S.J. Kim, Y.S. Lee, M.D. Song, I.H. Kim, H.S. Won, De novo generation ofshort antimicrobial peptides with simple amino acid composition, Regul. Pept. 166(2011) 36–41.

[44] R.M. Epand, R.F. Epand, Lipid domains in bacterial membranes and the action ofantimicrobial agents, Biochim. Biophys. Acta 1788 (2009) 289–294.

[45] A.A. Polyansky, A.A. Vassilevski, P.E. Volynsky, O.V. Vorontsova, O.V. Samsonova,N.S. Egorova, N.A. Krylov, A.V. Feofanov, A.S. Arseniev, E.V. Grishin, R.G. Efremov,N-terminal amphipathic helix as a trigger of hemolytic activity in antimicrobialpeptides: a case study in latarcins, FEBS Lett. 583 (2009) 2425–2428.

[46] W.G. Bessler, M. Huber, W. Baier, Bacterial cell wall components asimmunomodulators-II. The bacterial cell wall extract OM-85 BV as unspecificactivator, immunogen and adjuvant in mice, Int. J. Immunopharmacol. 19 (1997)551–558.

[47] Y. Rosenfeld, Y. Shai, Lipopolysaccharide (Endotoxin)-host defense antibacterialpeptides interactions: role in bacterial resistance and prevention of sepsis,Biochim. Biophys. Acta 1758 (2006) 1513–1522.

[48] S. Rieg, A. Huth, H. Kalbacher, W.V. Kern, Resistance against antimicrobial pep-tides is independent of Escherichia coli AcrAB, Pseudomonas aeruginosa MexABand Staphylococcus aureus NorA efflux pumps, Int. J. Antimicrob. Agents 33(2009) 174–176.

[49] S. Kim, S.S. Kim, B.J. Lee, Correlation between the activities of α-helical antimicrobialpeptides and hydrophobicities represented as RP HPLC retention times, Peptides 26(2005) 2050–2056.

[50] A. Dagan, L. Efron, L. Gaidukov, A. Mor, H. Ginsburg, In vitro antiplasmodiumeffects ofdermaseptin S4 derivatives, Antimicrob. Agents Chemother. 46 (2002) 1059–1066.

[51] I. Radzishevsky, M. Krugliak, H. Ginsburg, A. Mor, Antiplasmodial activity oflauryl–lysine oligomers, Antimicrob. Agents Chemother. 51 (2007) 1753–1759.

[52] B. Gao, J. Xu, M. del C. Rodriguez, H. Lanz-Mendoza, R. Hernández-Rivas, W. Du, S.Zhu, Characterization of two linear cationic antimalarial peptides in the scorpionMesobuthus eupeus, Biochimie. 92 (2010) 350–359.

[53] S. Zhu, B. Gao, A. Aumelas, M. del C. Rodríguez, H. Lanz-Mendoza, S. Peigneur, E.Diego-Garcia,M.F. Martin-Eauclaire, J. Tytgat, L.D. Possani, MeuTXKbeta1, a scorpionvenom-derived two-domain potassium channel toxin-like peptide with cytolyticactivity, Biochim. Biophys. Acta. 1804 (2010) 872–883.

[54] M.M. Javadpour, M.M. Juban, W.C. Lo, S.M. Bishop, J.B. Alberty, S.M. Cowell, C.L.Becker, M.L. McLaughlin, De novo antimicrobial peptides with low mammaliancell toxicity, J. Med. Chem. 39 (1996) 3107–3113.

[55] D. Venugopal, D. Klapper, A.H. Srouji, J.B. Bhonsle, R. Borschel, A. Mueller, A.L.Russell, B.C. Williams, R.P. Hicks, Novel antimicrobial peptides that exhibit activityagainst select agents and other drug resistant bacteria, Bioorg. Med. Chem. 18 (2010)5137–5147.